DEVELOPING AN INSTRUMENT TO MEASURE EDUCATOR PERCEPTIONS OF AFRICAN AMERICAN MALE STUDENTS PREK - 12

Delbert Christopher Eugene Scott

A Dissertation

Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements for the degree of

DOCTOR OF EDUCATION

December 2019

Committee:

Patrick D. Pauken, Advisor

Alicia Mrachko, Graduate Faculty Representative

Philip T.K. Daniel

Paul A. Johnson

Matthew Lavery © 2019

Delbert Christopher Eugene Scott

All Rights Reserved iii ABSTRACT

Patrick D. Pauken, Advisor

Educators are important in the academic and social development of students. Educator perceptions carry significant weight when interpreting behaviors, skills, and abilities of students

(Beckford, 2016; Simson, 2013). Research that investigates the possible consequences of educator perceptions of African American males and the relationship of those perceptions to student outcomes is scant.

This exploratory sequential research study reported psychometric properties of an instrument developed to examine educator perceptions of African American males held by public educators in PreK12. Extant research suggests that educator perceptions of Black males are more negative than those of noneducators (see Foster, 1995; Quinn, 2017). Specifically, overall perceptions of educators regarding African American males are negative (Fitzgerald, 2009;

Foster, 1995; Jackson & Crawley, 2003). The instrument created for the present study will guide future research that will enable researchers to examine the relationships between educator perceptions and outcomes for African American male students (e.g., eligibility in special education for EBDs).

Examining validity evidence for the public educator perceptions of African American males (PEPAAMS) PreK12 revealed significant relationships between educators (1) answering on behalf of the average person and (2) self-reporting personal perceptions. This study also found that the adapted brief social desirability scale did not function as intended. The

ABSDS was not a reliable measure to differentiate which dependent variable is best to use when there were different scores for personal and average perceptions of public educators using a iv paired samples t-Test and MANOVA. Due to the inadequacy of the ABSDS, findings revealed that personal value statements were a better indicator for determining which perceptions scores were more reliable to use. Finally, this study concluded that educators who were truly low prejudice (see Hing et al., 2008) were least likely to hold negative personal perceptions of

African American male students. v

This dissertation is dedicated to my beautiful children

Christopher Emmanuel,

Christion Elijah,

Annalyse Gabrielle, and

Alanna Grace Scott,

Without all of you, I would not have embarked upon such a profound journey. I love each of you

more than worldly wealth; I love you more than life itself.

The staff at Windsor, who I called the “dream team.” We transformed that school

despite union opposition, who vowed to get rid of me because we were

getting results. And then we went to Highland and did it again.

You all taught me how to be an educational leader.

To my pops, Dr. Patrick D. Pauken, your encouragement and support are immeasurable.

In loving memory of my mother in the gospel, Pastor Ethel L’Tanya Lemon, and my Godmother,

Sybil Truphelia Edwards-McNabb, without the two of you, I would not have been an educator.

I love you more than words can express, RIG. Finally, in loving memory of my pastor, Apostle

Cecil Alexander Pratt, Jr., pastor of Church of Jesus Family Worship Center in Springfield, Ohio

for over 50 years. “We may never pass this way again, but I remember how good it’s been.

I love you, RIG. vi ACKNOWLEDGMENTS

To Yahshua the Mashiach, the only true and living God. You kept your word through

David, King of Israel, when he declared in Psalms 27:10, “When my father and mother [forsook] me,” you, “the Lord [took] me up.” To Lynda Dickerson, my “Mother Dear.” You prayed me through. Thanks to you too auntie Dr. Tona Dickerson. To three African American women educators who profoundly impacted my life. First, Mrs. Stewart, my Kindergarten teacher at

South Mifflin Elementary school. Mrs. Donna Matthews, my 6th grade Reading and Language

Arts teacher at Woodward Park Middle School, you loved me without lowering your expectations. You treated me as if I were your child when I tried to run away from home. Mrs.

Emory Hill (now deceased), at Mifflin High School, a classically refined, beautiful Black woman who shaped the way I viewed words and language. My boss, Mr. Wilbert Jones, I will never forget you. You gave me my first opportunity at Windsor, despite being advised otherwise.

There were three scholars that changed my life, and to them I say thank you. At Central

State University in Wilberforce, Ohio, Dr. Hassan, my English professor from India. You taught me to write. In 1999, Dr. Eugene T.W. Saunders brought me into the master’s program at BGSU.

Here at BGSU I met Dr. Patrick David “Pops” Pauken. Dr. Pauken, you, as a woke scholar and father mentored me. You taught me more about how to understand my misunderstood male

Blackness. You taught me how to write efficiently and exposed me to opportunities with ELA.

Thank you for the conversations, mentorship, and support. To my committee, especially, grandpops, Dr. T.K. Daniel, you agreed to be on my committee because you believed in me. To

Leslie Peek, thank you for being my loving big sister! Finally, thanks to all my professors in the school of Educational Foundations Leadership & Policy. I love you all. vii

TABLE OF CONTENTS

Page

CHAPTER 1. INTRODUCTION ...... 1

Establishing the Definition for Perceptions of Black Males ...... 4

Statement of the Problem ...... 6

Purpose of the Study ...... 7

Research Hypotheses ...... 8

Importance of the Study ...... 9

Researcher Positionality...... 10

Overview of Theoretical Frameworks ...... 16

Limitations ...... 18

Assumptions ...... 19

Definition of Terms...... 20

Organization of the Study ...... 22

CHAPTER 2. LITERATURE REVIEW ...... 23

Process of Searching for Key Terms ...... 25

Educators and Noneducators’ Perceptions of Young Black Men ...... 26

Educator/Noneducator Perceptions of African American Males...... 29

Racial Attitudes of PreK – Postsecondary Educators and Noneducators ...... 35

Racial Attitudes, Principles, and Stereotypes ...... 38

Young Black Men Perceived as Bigger, Stronger, and More Capable of Doing

Harm ...... 42

Using Special Education to Control Young African American Males in Public viii

Schools ...... 53

Brief Overview of Special Education Statute for FAPE and Emotional and

Behavioral Disorders ...... 54

Evaluating Students Suspected of a Disability ...... 56

Safeguards to Deter Discrimination for Racial and Cultural ...... 58

Discretionary Decision Making ...... 59

Critical Race Theory ...... 64

Separation from Critical Legal Studies: No Room for Critical Race Theory. 68

Critical Race Theory in Education ...... 69

Intersection of Property and Race ...... 73

Phenomenon in Public Education ...... 76

Implicit and African Americans in the Public Education System ...... 78

Bias in Public Education and Curricula ...... 78

Six Steps of the Social Construction of Race in the Context of School Discipline ... 82

Social Meaning of Race in School Discipline ...... 84

Stigmatization and Biases ...... 86

Theory of Stigmatization ...... 86

Factors of Stigmatization (Haghighat, 2001) ...... 87

Levels of Stigmatization ...... 89

Implicit and Explicit Attitudes that Characterize Prejudice ...... 92

Truly Low Prejudice ...... 93

Aversive Racists...... 93

Principled Conservatives ...... 94 ix

Modern Racists ...... 95

Ideological Endorsements ...... 95

Testing the Two-Dimensional Model of Prejudice: Social Conservatism and

Political/Economic Conservatism ...... 103

CHAPTER 3. LITERATURE REVIEW FOR INSTRUMENT DEVELOPMENT ...... 109

Process for Searching for Key Terms Related to Instrument Development ...... 109

Four Phases of Instrument Development ...... 110

Phase One of Instrument Development ...... 114

Evaluating Existing Measures...... 114

Defining Content for an Instrument ...... 116

Phase Two: Planning and Construction of a Proposed Measure ...... 118

Test Construction ...... 119

Instrument Design ...... 120

Types of Items...... 120

Developing Quality Test Items ...... 121

Pilot Testing in Instrument Development ...... 122

Phase Three: Continuing Validity Examination ...... 123

Types of Validity Evidence for Reliability ...... 124

Factor Analysis ...... 125

Construct Validity in Instrument Development ...... 126

Scales and Scoring ...... 129

Scale Development ...... 130

Examining Validity Evidence Based on Reliability of Scales ...... 130 x

Interpretation and Use Argument ...... 130

Examining Evidence of Validity Using an Interpretation and Use

Argument ...... 131

Phase Four of Instrument Development ...... 134

Examining Evidence of Validity ...... 135

Validity Evidence for Reliability in Quantitative Research ...... 139

CHAPTER 4. METHODOLOGY ...... 141

Definition of Perceptions of African American Males ...... 141

Validity Evidence for Content: Developing the Public Educator Perceptions of

African American Males’ PreK-12 Survey ...... 142

Exploratory Sequential Research Design...... 143

Testing Research Hypotheses ...... 144

Instrumentation ...... 144

Validity Exploration...... 145

Assumptions ...... 146

Qualifying Questions ...... 146

Examining Validity Evidence for the Adapted Brief Social Desirability Scale ...... 147

Examining Evidence of Validity for Reliability Using the Adapted Brief Social

Desirability Scale (ABSDS) ...... 148

Part II - Answering for the Average Person and Self Reporting Personal Beliefs

Regarding African American Males ...... 149

Public Educators’ Reporting Personal Perceptions of African American

Males ...... 150 xi

Parts III and IV - Measuring Educators’ Perceptions of African American Male

Students PK12 Scales and Scoring ...... 150

Interpretation and Use Argument for Scales on the PEPAAMS PK12 ...... 151

Part V - Perceptions of African American Males Survey: Demographic Data ...... 152

Procedures ...... 153

Informed Consent...... 154

Data Collection for Each Phase of the Instrument Development Process ...... 155

Phase I ...... 156

Phase II...... 156

Phase III ...... 157

Phase IV: Ongoing Examination of Validity ...... 157

Data Analyses ...... 157

Descriptive Statistics ...... 158

Inferential Statistics ...... 158

Conclusion ...... 162

CHAPTER 5. RESULTS ...... 163

Cognitive Interviews for the Adapted Brief Social Desirability Scale (ABSDS) ..... 163

Examining Validity of the Adapted Brief Social Desirability Scale (ABSDS) ...... 164

Analysis of Cognitive Interviews...... 166

Expert Panel Review of PEPAAMS PK12 ...... 168

Expert Panel Feedback ...... 169

First Round of Revisions ...... 171

Changes to Wording for Specific Items ...... 172 xii

Second Round of Revisions ...... 173

Final Revisions to the PEPAAMS PK12 ...... 175

First Distribution of the PEPAAMS PK12 ...... 176

Second Distribution of the PEPAAMS PK12 ...... 179

Third and Final Distribution of the PEPAAMS PK12...... 180

Overall Response Rate and Useable Data ...... 180

Demographics of the Population ...... 181

Role in Education and Type of School District Represented ...... 181

Domestic and Political Affiliations ...... 182

Other Attributes of Study Participants ...... 182

Exploratory Factor Analyses...... 182

Factor Loadings for Perceptions for the Average Person ...... 185

Factor Loadings for Personal Perceptions Held by Public Educators ...... 188

Pattern Matrix for Factor Loadings for Personal Perceptions of Public

Educators...... 191

Research Question 1: Relationship Between (a) Perceptions for the Average

Person, and (b) Self-Report Scores ...... 192

Descriptive Statistics Results: Means and Frequencies ...... 192

Inferential Statistics Results ...... 194

Research Question 2: Differences Between (a) Perceptions Scores for the Average

Person and (b) Self-Report Perceptions Scores for Scores on the ABSDS ...... 196

Testing Assumptions for MANOVA ...... 196

Results from MANOVA Testing ...... 198 xiii

Descriptive Statistics for Results on the ABSDS ...... 199

Alternate Analysis Using Personal Value Statements: Ideological Values ...... 202

Ideological Values ...... 202

Alternate Analyses to Examine the Relationship and Differences Between Both

Dependent Variables and the Independent Variable ...... 207

Summary ...... 213

CHAPTER 6. DISCUSSION, CONCLUSIONS, AND RECOMMENDATIONS ...... 214

Differences in Personal Perceptions and the Average Person ...... 214

Negative Perceptions of African American Male Students: Aggressive and

Threatening ...... 216

Negative Perceptions of African American Males: Personal Characteristics ...... 219

Perceptions of Criminality ...... 220

Implications from Exploratory Factor Analysis on Personal Perceptions ...... 222

The Adapted Brief Social Desirability Scale ...... 223

The Relationship Between Perceptions and Political Ideologies ...... 225

Personal Value Statements and Political Affiliation ...... 227

Negative Perceptions of African American Male Students and Ideological Views .. 228

Recommendations for Future Research ...... 231

Exploring Age in Future Distributions of the PEPAAMS PK12 ...... 233

Response Rates ...... 234

Diverse Population and Sample ...... 235

Confirmatory Factor Analyses in Future Research ...... 235

Future Social Science Research and Policy ...... 236 xiv

Recommendations for Leadership and Practice ...... 240

Leadership in Relation to Policy, and Practice ...... 241

Organizational Repairs in Public School Districts...... 243

Leadership and Social Contracts in Organizations ...... 244

Perceptions of African American Males and School Leadership ...... 246

REFERENCES ...... 250

APPENDIX A. INFORMED CONSENT LETTER ...... 263

APPENDIX B. PEPAAM SURVEY PK12 ...... 265

APPENDIX C. IRB APPROVAL LETTER ...... 271

APPENDIX D. INITIAL EXPERT PANEL INVITATION ...... 273

APPENDIX E. INVITATION AND INFORMED CONSENT ...... 274

APPENDIX G. MARLOWE-CROWNE SOCIAL DESIRABILITY SCALE ...... 276

APPENDIX G. REVISED EXPERT PANEL INVITATION LETTER ...... 278

APPENDIX H. BRIEF SOCIAL DESIRABILITY SCALE ...... 279

APPENDIX I. COGNITIVE SCRIPT ...... 280

xv

LIST OF FIGURES

Figure Page

2.1 Social construction of race in the context of school discipline ...... 81

2.2 How self-interests influence attitudes related to stigmatization ...... 88

2.3 Relationships between prejudice profiles and implicit/explicit biases ...... 99

5.1 An example of the complex item before it was simplified to increase response

rates ...... 177

5.2 Simplified item information regarding participants’ role in education ...... 178

5.3 Scree plot of exploratory factor analysis for perceptions of the average

person ...... 184

5.4 Scree plot of exploratory factor analysis for personal perceptions ...... 189

xvi

LIST OF TABLES

Table Page

2.1 Studies Summarizing Perceptions of African American Males ...... 28

2.2 Overall Perceptions of Black Males ...... 31

2.3 Differences in Racist Principles Held by Educators and Noneducators ...... 38

2.4 Segment Summaries of Racial Bias in Judgment of Physical Size and Threat ...... 44

2.5 Summary of Race and Gender of Special Education Students and Psychotropic

Drug Use ...... 61

2.6 Percentage of Students with Psychotropic Medication Use Before and Between the

3-year Reevaluation ...... 62

2.7 Ideological Endorsements ...... 95

2.8 Scales Used to Define Prejudice Profiles...... 97

2.9 Summaries of Studies Regarding Explicit and Implicit Biases that Characterize

Prejudice ...... 102

3.1 Summary of Instrument Development Studies Presented in Chapter 3 ...... 111

3.2 Summary of Instruments Appraised by Pohan and Aguilar ...... 115

3.3 Statistical Tests Used in Instrument Development ...... 127

4.1 Public Educators’ Perceptions of African American Males Survey -

Qualifying Questions ...... 147

4.2 Adapted Brief Social Desirability Scale – Part II of the Public Educator

Perceptions of African American Males PK12 Survey ...... 148

4.3 Perceptions of Black Males Categories and Items: Foster ...... 149

4.4 Test Statistics Used in Inferential Analyses...... 159 xvii

5.1 ABSDS Results from Cognitive Interviews ...... 168

5.2 Expert Review Panel Questions ...... 171

5.3 Summary of Comments from Expert Review Panel Members ...... 174

5.4 Means, Standard Deviations, Pattern Coefficients, and Communalities for the

Average Person on the PEPAAMS PK12 Items ...... 186

5.5 Means and Standard Deviations for each Factor ...... 187

5.6 Means, Standard Deviations, Pattern Coefficients, and Communalities of Personal

Perceptions for the PEPAAMS PK12 ...... 191

5.7 Descriptive Statistics for Variables ...... 193

5.8 Overall Perceptions of African American Male Students by Percent and

Frequency ...... 194

5.9 Mean and Standard Deviations for Items on the ABSDS ...... 199

5.10 Mean Differences, Standard Error, and Significance ...... 201

5.11 Ideological Items from the PEPAAMS PK12 ...... 203

5.12 Definitions for Ideological Views for the PEPAAMS PK12 ...... 204

5.13 Personal Value Selections, Ideological Categories, Percentages, and

Frequencies ...... 206

5.14 Descriptive Statistics, Means and Standard Deviations/Error for Ideological

Values ...... 212

6.1 Corresponding Factors for Perceptions for the Average Person and Personal

Perceptions ...... 218

6.2 Social Desirability Item Communalities Before and After Extraction ...... 224

1

CHAPTER 1. INTRODUCTION

Discussion of public perceptions of Black men are taboo (Foster, 1995). When researching adverse outcomes for them in the United States (e.g., shooting deaths of unarmed

Black men by law enforcement and private citizens), research regarding typical experiences for

African American men are often challenged (Berliner & Calfee, 1996; Fitzgerald, 2009; Wilson,

Hugenberg, & Rule, 2017). Research suggests unfavorable revelations related to gender and race are the primary reason for dismissing research results (Berliner & Calfee, 1996). Post-racial sentimentalists in the United States do not want to acknowledge that schools are ineffective and inept (Berliner & Calfee, 1996).. Furthermore, post-racialism sensibilities are thwarted when it is revealed that schools are structured to be unequal, mainly when they are comprised of racial minorities (Berliner & Calfee, 1996; Crenshaw 1995; 2010).

Negative experiences and systemic inequities for Black men are rooted in the history of slavery (Fitzgerald, 2009; Gates, 1997; Jackson & Moore, 2006). The value of Black men is steeped in American traditions; they are not valued beyond their ability to provide muscle and might as the primary source of labor to keep the engines of free labor producing mass wealth for the slave economy (Jacobs, 2009). Henry Louis Gates (1997) asked a question: “what was

America to do with Black male slaves after emancipation?” After slavery, Black men were considered leftovers (Gates, 1997). Some scholars believe that there is fear that Black men would one day retaliate for the inhumane treatment throughout American history (Gates, 1997; Jacobs,

2009). Other scholars believed that the aforementioned fear was the result of guilt held by those who acted as oppressors. The fear of retribution of Black men provided a rationale for marginalization and stigmatization (Simson, 2013). For example, intermittent eruptions of 2

violence from the mid-to-late 1800s led by slaves are reasons why Black men were to be feared

(Gates, 1997; Haghighat, 2001; Jacobs, 2009).

Perceptions of inferiority regarding Black men encapsulate feelings that they lack utility in society. According to Gates (1997), Black men were “expired [commodities] of muscle and might which has lost value in the marketplace” (p. 16). Research results proposed that Black men be perceived as a threat to society (Wilson, Hugenberg, & Rule, 2017). The specific threat is that they are perceived as "bigger, stronger, and more capable of harm (Wilson, Hugenberg, & Rule,

2017, p. 1). Black men are not only perceived as a threat, but that they are more likely than any other racial minority to experience adverse effects associated with race and gender (Fitzgerald,

2009; Jackson & Crawley, 2003; Kunjufu, 1985).

Kunjufu (1985) believed that the racial majority’s perceptions of threat begin early in the life of young Black males. Understanding how Black males are perceived is a pivotal element in understanding their experiences in public schools. More specifically, perceptions held by public educators have significant power to influence outcomes of young Black males (e.g., making decisions regarding educational placements; Fitzgerald, 2009; Foster, 1995, Jackson, & Crawley,

2003; Simson, 2013). Race is a primary determinant that guides experiences of African

American students in public schools (Simson, 2013). For example, Foster (1995) provided an anecdote about a 17-year-old young Black male placed in a class for students with severe emotional and behavioral disorders. According to Foster (1995), the only documentation used to justify his placement was an account from his teacher, a White female, who viewed his way of walking as sexually provocative. 3

Foster (1995) reported findings from his qualitative study of more than 3000 educators and noneducators. He reported that perceptions of Black males had made little improvement over time. Furthermore, he asserted that educators’ perceptions of young Black males were more negative than those of noneducators in specific subcategories. The findings reported by Foster

(1995) conceptualized lived experiences that I and countless Black males have experienced in school. I use the results from Foster’s study as the inspiration to transform them into a quantitative, exploratory sequential research study. The present study of public educator perceptions of African American males in public schools adds a quantitative context to address issues for Black males in society, typically studied employing qualitative methods. Simson’s

(2013) social construction of race in the context of school discipline is the best theoretical construct that complements Foster’s (1995) inquiry. Combining the two, perceptions of Black males (Foster, 1995) and educator decision-making based on race (Simson, 2013), provides avenues for additional research studies to explore the effects of perceptions of young Black males in a socially constructed context (e.g., mental health, educational attainment, and experiences at predominantly White institutions of higher learning; Watkins, Walker, & Griffith,

2010).

Researching public educators’ perceptions of African American males adds quantitative content to scant social science research regarding outcomes for African American males in PreK

– 12 public schools in the United States of America. Furthermore, outcomes related to the present study provide a platform for public policy around restructuring teacher educator programs at publicly funded institutions, as well as prioritizing professional development agendas for public school districts. In Chapter 6, study results will correlate to racial and gender 4

of public school teachers in the U.S. (see Fairchild, Tobias, Corcoran, Djukic,

Kovner, & Noguera, 2012; Woodson & Harris, 2019).

Establishing the Definition for Perceptions of Black Males

Perceptions of Black males are as stereotypical beliefs, feelings, expectations, and fantasies held by the average person (Foster, 1995. The definition of perceptions includes implicit racial bias (Simson, 2013. Implicit biases determine how educators interpret behaviors of African American students. Implicit biases are internal workings that trigger subtle or overt physical actions (e.g., blinking, avoiding eye contact, rapid or repetitive body movements;

Simson, 2013; Haghighat, 2001; Hing, Chung-Yan, Hamilton, & Zenna, 2008. Hing et al.

(2008 and Haghighat (2001; 2007 provided useful information to expand the definition of perceptions for the present study. Explicit biases are more conspicuous and result in discriminatory behavior evidenced by isolation, and rejection (Haghighat, 2001; 2007; Hing et al. 2008. Consequently, when perceptions of young Black males are distorted due to implicit and explicit biases, hostile environments materialize and create emotional stress. Negative behaviors (e.g., aggression, violence, defiance, disrespect, socially unacceptable behaviors are the result of behavioral leakage that induces a hostile environment (Beckford, 2016.

Biases affect decision making (Hing, Chung-Yan, Hamilton, & Zenna, 2008; Simson, 2013.

Hing et al., (2008 concluded that high/low implicit and high/low explicit biases reflect varying degrees of prejudice impacting behaviors toward racial minorities. For example, individuals with specific prejudice profiles can diminish the effectiveness of culturally diverse teams (Dovidio,

Gaertner, Kawakami, & Hodson, 2002. 5

I conclude that perceptions related to African Americans include personal and societal beliefs that are stereotypical, which include implicit and explicit biases. Perceptions also include feelings, expectations, fantasies and values (Foster, 1995; Haghighat, 2001; 2005; 2007; Pohan

& Aguilar, 2001; Quinn, 2017; Sears & Henry, 2003; Simson, 2013; Washington, 1981; Wilson,

Hugenberg, & Rule, 2017; Wood, 1994).

The inspiration for the present study came from Foster (1995), who reported results from a one-question survey regarding perceptions of Black men. Participants were asked to list perceptions of African American males held by the average person, which were reported as the personal perceptions of both educators and noneducators. I argue that allowing participants to answer for the average person may not be the best way to draw conclusions about personal perceptions. There is no way to tell without asking participants to self-report their perceptions of

African American males.

Self-reporting personal perceptions may expose a research study to measurement error due to social desirability bias. There are potential threats to validity in qualitative research like the study of educators and noneducators’ perceptions of Black males, which include researcher bias and reactivity (Maxwell, 2013; Milner, 2007; Peshkin, 1988). The present study will examine educator perceptions of African Americans by asking them to answer for the average person, and for themselves personally, while controlling for social desirability bias (Crowne &

Marlowe, 1960; Foster, 1995; Haghighat, 2001; 2007; Strahan & Gerbasi, 1972). Social desirability, according to researchers like Crowne and Marlow (1960), Strahan and Gerbasi

(1972), and Van de Mortel (2008), is the tendency for participants to answer in a way that makes them appear to be good and socially acceptable. 6

An example of researcher bias is when data interpretation is manipulated to fit the researcher's existing theories. Reactivity, on the other hand, is the researcher’s influence on the setting or individuals' studies (p. 124. According to Foster (1995, his existing theory about perceptions of Black males was evident in how he defined the term. He used a one-question survey that included words like, “stereotypical beliefs, feelings, expectations, and fantasies” for which he asked participants to answer for the average person regarding perceptions of Black men. Over a four-year period, Foster (1995 reported giving surveys at the conclusions of his seminars in various parts of the country, whereby he noted some participants expressed “anger” and “hostility” for researching the topic. Reading research results, I contacted Foster in 2016 regarding this study. Foster gave some clues that criticism from his study was immense. He ultimately let me know that he no longer had the notes and other related materials from the study.

Statement of the Problem

This present study explores negative perceptions of African American males held by

PreK – 12 public educators. African American males are likely to experience academic and social difficulties in school (Horsford, 2011; Simson, 2013. Relevant research studies have examined barriers that impede success for African American males. Impediments to success may result from biased interpretation of behavior, as well as biased application of school policies

(Simson, 2008. Oelrich (2012 refers to them as cultural misunderstandings attached to biases.

Furthermore, relevant research suggests that educator biases have adverse effects for African

American males PreK -12 (e.g., racial, implicit, explicit; Hing, Chung-Yan, Hamilton, & Zenna,

2008; Simson, 2013; Wilson, Hugenberg, & Rule, 2017. Additionally, racial bias is suspected to play a prominent role in disproportionate school discipline (e.g., school suspensions and 7

expulsions; Simson, 2013. Biases, coupled with cultural misunderstandings might influence drop-out rates and special education membership (Fairchild, Tobias, Corcoran, Djukic, Kovner,

& Noguera, 2012; Fitzgerald, 2009; Foster, 1995; Kunjufu, 2005; Oelrich, 2012. Finally, biases and cultural misunderstandings are suspected to be a contributing factor to the lack of academic success for Black males in public schools (Foster, 1995; Oelrich, 2012; Simson, 2012.

Currently, no quantitative research directly links biases to educator decision making in educational placements (e.g., general eligibility and referrals to special education; Kunjufu,

1995. However, Oelrich (2012 used published data that noted African American males are six times more likely to be diagnosed with emotional disorders. Furthermore, African American comprised 17% of the student population but made up 27% of the special education referrals

(Oelrich, 2012. Finally, there is ample evidence to assert that African American males have disproportionate representation in a phenomenon known as the school to prison pipeline

(Fairchild, Tobias, Corcoran, Djukic, Kovner, & Noguera, 2012; Simson, 2013; Wald & Losen,

2003).

Purpose of the Study

With no quantitative measure to explore public educators’ perceptions of African

American males in PreK – 12 public schools, correlating perceptions with professional judgment in decision-making processes is not feasible. Going forward, this research study will provide information to (1 highlight the importance of examining public educators’ perceptions of

African American males in PreK – 12 public schools; (2 use relevant research to build an argument that creating a measure to explore public educators perceptions of African American 8

males is groundbreaking for more detailed research studies; and (3 develop a quantitative instrument to measures public educator perceptions of African American males.

Overall perceptions of African American males will be reported using scores from five subcategories: (1 crime, (2 education, (3 attitude, (4 personality, and (5 sexual prowess. In addition to developing an instrument to measure PreK -12 educators’ perceptions of African

American males, the instrument was constructed to control for social desirability bias and consideration for implicit and explicit biases (Haghighat, 2007; Hing, Chung-Yan, Hamilton, &

Zenna, 2008; Simson, 2013, The study concludes by providing preliminary data regarding psychometric properties. The proposed measure, public educator perceptions of African

American males’ survey (PEPAAMS, PK12 examined (1 the independent variable, social desirability bias; and, (2 two dependent variables, reporting perceptions for the average person; and self-reporting personal perceptions of African American Males.

Research Hypotheses

Foster (1995) has been cited over 3500 times, and he is widely respected for other publications examining urban codes and language expression (see Foster, 1974; 1975; 1986;

1990a; 1990b. Allowing participants to answer for the average person could have been an attempt to control for . In quantitative methods, controlling for social desirability bias using advanced multivariate statistical tests (e.g., hierarchal regression analyses, or including a social desirability (SD scale are two examples available to researchers to mitigate measurement error from SD bias (Crowne & Marlowe, 1960; Haghighat, 2007; Strahan &

Gerbasi, 1972; Van de Mortel, 2008. The research hypotheses are intended to guide me as I 9

attempt to understand the complexities of researching race and gender-specific topics like perceptions of African American males:

(1) Is there a relationship between (a) perceptions of the average person and (b) self-

report scores among public educators of African American male students?

(2) Is there a difference in the relationship between (a) perceptions scores for the average

person and (b) self-report perceptions scores for participants with different scores on (c)

the adapted brief social desirability scales)?

Importance of the Study

Research studies that examine gender and race-specific outcomes (e.g., Black men) concerning public institutions are strongly discouraged research emphases (Berliner & Calfee,

1996. Some studies suggest that researching sensitive topics, specifically related to perceptions of Black men, reveal intentional efforts to ignore them in quantitative research agendas (Jackson

& Moore, 2006; Katz, 2002. Oelrich (2012 affirmed this assertion citing and cultural misunderstandings often leave issues relevant to Black men unexamined. The present study fills a void that currently exists by the lack quantitative research regarding experiences of

African American males in public schools. Future research studies will use the instrument developed to measure perceptions of African American males in various institutions like public education. For example, the present study inspires additional quantitative studies to explore relationships between educators’ perceptions of African American males and decision-making

(Castillo, March, Stockslager & Hines, 2016; Oelrich, 2012; Simson, 2013.

Researchers like Haghighat (2007 and Castillo, et al. (2016 understood developing instruments to examine sensitive topics is not only necessary but helpful to address issues 10

thathave been ignored in research. Finally, the present study reported psychometric properties of the instrument used to measure educator perceptions of African American males, which will lead to a larger-scale research study that could prove or disprove a hypothesis that explains African

American males' disproportionate eligibility in special education for EBDs.

Researcher Positionality

The present study has personal significance for two reasons. First, it reconciles the need for healing for past traumas related to my experiences as a Black male in three contexts: in-group trauma (e.g., experienced within African American communities, outgroup trauma as a student impacted by systems (e.g., growing up post-desegregation as a Black male in public schools, and outgroup trauma as both a post-secondary student in higher education and a professional working in PreK -12 public education as a teacher and principal.

Researcher epistemologies, or ways of knowing, are a consideration when researching a topic within a racial minority group wherein the researcher has a membership or displays a particular bias relating to the topic and group. The term epistemology refers to what is known, how it came to be known, and how others impact the process of knowing (Bess and Dee, 2008;

Milner, 2007; Peshkin, 1988. Milner (2007 provided a researcher positionality framework that they should evaluate to decrease the impact of bias or researcher reactivity may have in a research study. Milner’s positionality is relevant because of the complex nature of my membership in the group at the center of study. The present study examines perceptions of

African American males held by public educators. The following are Milner’s four researcher positionality frames: (1 researching the self, (2 researching self in relation to others, (3 engaged reflection and representation, and (4 shifting from self to systems. 11

Researching the self as a researcher permits me to examine what I know about being a

Black male, how I know it, and how it affects my research emphasis. In 1978, I was the first class to experience the full implementation of desegregation in one of Ohio’s largest public school districts. Mrs. Stewart was my Kindergarten teacher. She was a beautiful, young Black woman. All the students in my class were Black like me. I lived in a Black community. Doctors and nurses at the clinic for low-income families looked like me.

I never had any formal interaction with a member of the racial majority until first grade.

My experiences attending a school in an all-White community led me to discover the difference between Black and White people. First through third grades, I would embark upon a seven-mile trip across town on the school bus to a middle-class community where I attended a nicely manicured elementary school in a housing development comprised of the racial majority. The formula for desegregation for the school district I attended required all Kindergarteners to attend schools in their communities. First through third-grade students were bussed to outlying communities. All students returned to their community schools for fourth and fifth grades. There was a clear memory of being separated from the White students in the outlying communities.

Researching educator perceptions of African American males is researching myself as one who has experienced the effects of perceptions. West (1993) discussed self-liberation for Black men.

Milner (2007) included self-liberation in the first positionality frame, researching the self. West

(1993) believed that working for the emancipation of others to ameliorate their problems lacked effectuality if one has not realized self-emancipation. Examining perceptions of African

American males provides the key to a locked door behind which lies the answers for other

African American males’ experiences in public schools. 12

Researching the self in relation to others requires me to examine my epistemologies or ways of knowing about the world by considering racial experiences of internal others (other

Black males) and external others (those who are not Black and male). Black males experience the world in different ways transcending gender and racial. Extant research suggests that these experiences have remained constant in public school systems since desegregation (Beckford,

2016; Oelrich, 2012; Simson, 2013). The effects of stigmatization shaped the way I, along with other African American males, have experienced the world. If the reader were to visit my classrooms in the elementary school, they would see my desk situated in the outside margins of the room. In cases where my teacher thought that my behavior was more peculiar than usual, my desk might have been positioned in the hallway directly outside the classroom.

Researching experiences related to Black males exposes how unexamined practices forecast how Black males experience the world outside of school with their place in the outer margins or outside of society. Examining educator perceptions of myself and others opens my understanding to not only how they are perceived, but also why they are perceived in consistently negative patterns that have not changed over time. The present study acts the foundation to examine perceptions and to discover factors that might contribute to outcomes for them in school and society.

So far, I discussed two of the four researcher positionality frames proposed by Milner

(2007): (1) researching the self and (2) researching the self in relation to others. Engaged reflection and representation is the third researcher positionality frame (Milner, 2007). Engaged reflection and representation encourage the processes of self-examination to stretch beyond personal experiences and way of knowing to converge with epistemological beliefs of internal 13

and external others. The alternate narrative is developed when researchers take deliberate methodological steps to look outside their individual belief paradigms to consider other possibilities. For example, developing an instrument to examine educator perceptions using questions that point to racial bias as the sole source of negative attitudes is limiting, and also reflective of researcher reactivity (Maxwell, 2010). Engaged reflection and representation in instrument development, specifically item construction, involves broadening the scope of the literature review used to inform content. As a result, engaged reflection and representation does not cast away racial bias as a source that informs perceptions of African American males, but stretches beyond the limits of the obvious to make room for epistemologies that have not been considered. Summarily, the literature review discussed in Chapter 2 includes implicit and explicit biases, ideological/political beliefs, and theories related to stigmatization and marginalization of racial outgroups to explore all possibilities that influence perceptions (see Haghighat, 2001,

2005, 2007; Hing, Chung-Yan, Hamilton, & Zenna, 2008; Quinn, 2017; Simson, 2013; Wilson,

Hugenberg, & Rule, 2017).

Milner’s (2007) fourth positionality frame suggested the need for shifting from epistemic values of self and the internal. Additionally, the fourth frame shifts from external others to systems (e.g., schools), wherein the alternative narrative of external others exposes vulnerabilities in systems. Shifting from self to systems is most troublesome for me as a researcher because it involves bringing awareness to how my behaviors conformed to systemic values and practices that contributed to adverse outcomes that marginalized African American males. My professional ethics as a Black man were in some ways complicit in the very things that were used against me as a student and professional within the system. For example, I was 14

guilty of administering discipline with innocent African American males so teachers would not file a grievance with the teachers’ union for my failure to follow the contract related to a school wide discipline plan. Every building principal was required to work with teachers in the building to create a school wide discipline to be kept on file at union headquarters. Milner (2007) used an example of shifting from self to system to reveal how the researcher can work to uncover policies that protect students from being abused by systems that blame them “solely” for what happens to them in the school and community (p. 393). I knew my silence favored teachers who triggered Black males in order to gather data to help qualify them for special education. Also, my silence was acting in agency within a system to perpetuate harm against myself by allowing damage for other African American males. My behaviors did not stop there; I was complicit in promoting system values by suspending students for infractions that were brought on by adult behaviors. Through the process of researching the self, it was fear of the teachers’ union that impaired my ethical responsibility to protect marginalized students. In my experiences, racial minorities serving as principals were often the targets of teachers’ unions when they used tools within the system to hold teachers accountable for stigmatizing behaviors that were morally and ethically wrong. In order to preserve my livelihood, I complied when pressured by teachers’ union leadership to operate under neutral values that did not disrupt system hegemony.

Teachers’ unions are powerful systems that preserve policy and practices that perpetuate harm for African American students, particularly, males (Quinn, 2017). Teachers’ unions serve the interests of their constituencies. Any principal who provides corrective feedback or unsatisfactory ratings in performance appraisals for teachers that perpetuate stigmatizing 15

behaviors against African American males is in danger of becoming a target of marginalization and stigmatization.

First, the fear of losing my livelihood helped me to rationalize my inaction, which resulted in strengthening and preserving the system. Additionally, I made excuses regarding the lack of support from school officials made it unprofitable for me to fight for the liberation of

Black males when no-win situations were erected against them. Finally, my captivity from personal and professional traumas was relived each time I saw myself in the students who were victims of the same system that I was preserving to their harm.

In many respects, this research study is evidence of self-liberation. Some might argue that my position as an African American male and author of the present study, who also has a vested interest in the research topic, should be regarded as a limitation, prone to bias. Despite criticism, I argue that devoting this section with information regarding my positionality informs readers about the need to offer this quantitative study as a contribution to understanding African

American male experiences in public schools. Maxwell (2013) suggested that the researcher disclose biases that may influence the interpretation of results. Milner (2007), on the other hand, encouraged self-examination to disable research bias that could lead to unforeseen pitfalls in researching topics wherein the primary researcher has personal attachments. Including my research positionality using Milner (2007) implicates that overarching assumptions with the propensity to spark reactivity in the present study have been examined and exposed beyond traditional delimitation statements.

In Chapter 6, I discuss Milner’s (2007) framework using systems theory to provide missing links that research findings cannot answer. Systems theory can offer remedies to explain 16

processes within systems that shift between tight and loose coupling interposing dominate hegemony over racial minority groups, which results in controlled narratives for how issues related to race, and gender are interpreted. The present study using quantitative research scholarship by an African American male adds statistical value to the subject matter, which may result in the liberation of other Black researchers. Furthermore, using critical, and social construction of race theories to expose how negative perceptions of African American males, will no longer be dependent on interest convergence that is advantageous to hegemonic values

(Milner, 2007. Discussing trends in macro systems (e.g., historical, political, social, economic, racial, gender, and cultural impact on research epistemologies or ways of knowing about Black males, broadens the scope for quantitative research to become a distinguished subcategory within critical race theory.

Overview of Theoretical Frameworks

This section briefly describes theoretical constructs that aids in building the argument regarding the importance of educator perceptions relating to African American males. More detail about the theoretical frameworks is in Chapter 2's literature review. Two theoretical frameworks are essential to this research study. Critical Race Theory and the Social

Construction of Race in the Context of School Discipline serves as complementing frameworks that guide the research study’s focus on examining educator perceptions of African American males.

Critical Race Theory (CRT is an area of scholarship that energized legal scholars and students from Harvard and other American universities who formerly operated within the critical legal studies movement (Bell, 1988; Davis, 2012; Harris, 1983. According to Davis (2012, 17

critical legal studies was the incubator wherein in emerged. Critical legal scholars interrogated the legitimacy of policies and jurisprudence that marginalized minority groups. Racial hegemony was deemed to be the perpetrator of all injustices. Critical legal scholars became discomfited by

CRT’s critique of White male dominance, which according to some, was a source for most inequities amongst African Americans. Scholars within critical legal studies withdrew support from critical race scholars because it was believed that the attack on racial dominance as divisive. Extrication of scholars from critical legal studies allowed the movement to flourish as a separate discipline devoted to critical race studies (Bell, 1988; Davis, 2012; Harris, 1983; 2012).

Since then, Critical Race Theory has moved to other disciplines, particularly education (Davis,

2012; Harris et al., 2012; Ladson-Billings & Tate, 1995).

The present study used Critical Race Theory to (1) interrogate institutional and structural racism that is endemic in public schools, (2) expose paradoxes in government and school district policies that reify racism, (3) redirect the narrative of racism and oppression by using quantitative research to speak truths of oppressed groups, and (4) challenge racial hegemony and pervasive interdisciplinary ahistoricism that reifies negative stereotypes and biases. Critical Race

Theory alone cannot support the development of an instrument to measure educator perceptions.

Simson (2013) uses CRT to interrogate institutional and structural racism in schools using the

Social Construction of Race Model in the Context of School Discipline (SCRMCSD).

According to Simson (2013), race determines (1) how African Americans are perceived, (2) how they experience school, (3) how their behaviors are interpreted, and (4) how decisions are made based on implicit biases. In Chapter 2, I will describe how Simson (2013) used the social construction of race model in the context of school discipline (SCRMSD) to describe six levels 18

of interpreting African American students’ behavior. Once a student has been racially categorized, implicit biases are used to interpret behaviors for Black students differently than those in the racial majority. Biases contribute to experiences that stigmatize racial minorities

(Haghighat, 2001, 2005; Simson, 2013. The social construction of race model in the context of school discipline incorporates critical race theory to interrogate public school system decision- making that adversely affects African American students. Chapter 2 provides an in-depth discussion of CRT and SCRMCSD, which bolsters the argument supporting examination of educator perceptions.

Limitations

The present research study proposing to develop an instrument to measure educator perceptions of African American males is a controversial topic. Despite the drawbacks, examining educator perceptions is relevant to researchers who want to address issues related to certain racial minorities. Research suggests that studying African American males is taboo because of the social stigma attached to their racial experiences in the United States (Haghighat,

2005; Wilson, Hugenberg, & Rule, 2017. It is imperative that the present study provide an adequate examination of validity related to content to develop a valid and reliable instrument

(Atwater, Lau, Bass, Avolio, Camobreco, & Whitmore, 1994; Guion, 1978; Kane, 2013. The majority of African American male students attend large urban districts (Beckford, 2016,

Noguera, 2012. Lack of resources to access a broader population is a concern. Data collected in field testing, piloting, and validation might not be relatable to the targeted sample intended for use with the instrument developed in the present study. To remedy the current limited sample scope, I applied for a research grant through the American Education Research Association to 19

fund a subsequent, larger-scale study indicative of the intended population. Once funding is secured, a subsequent study will ensue to sample public educators in the largest urban school districts in the United States (e.g., New York City, Los Angeles, and Chicago).

Assumptions

A significant assumption associated with the present study is that a broad, flexible interpretation and use argument is needed in the process of collecting and interpreting data in the field testing, piloting, and validation stages of instrument development. The interpretation and use argument (IUA is a statement that describes how conclusions about data are drawn after exploring evidence of validity (e.g., scale reliabilities. The researcher used the IUA to disclose how data derived from scores informed results and conclusions. The present study measures educator perceptions. The IUA is summarized by a clear concise statement that clarifies decisions used to draw conclusions when reporting results. Researchers are cautioned that the

IUA should be sensible and practical (Kane, 2013. However, there are no set rules for researchers when determining the criteria for interpretation, and use of scores, other than the criteria must be clearly stated and reasonable (Kane, 2013. The present study is susceptible to response and social desirability biases that could affect usable data. I assumed that public educators with specific prejudice profiles are more likely to demonstrate social desirability bias.

Alternatively, I also assume that instead of engaging social desirability bias, some participants may elect to withdraw from the study altogether because they hold immense feelings of antipathy for African American males.

To mitigate the effects of these biases, the PEPAAMS PK12 includes an adapted brief social desirability scale (ABSDS, which is essential for constructing the interpretation and use 20

argument that will determine how scores are used to draw conclusions about the sample population (Kane, 2013. Participants will self-report perceptions in two contexts: (1 for the average person and (2 their personal perceptions of African American males. This leads me to my final assumption, with two possibilities. First, participants that have high scores on the

ABSDS indicate social desirability bias. Secondly, participants with ABSDS scores that fall within an acceptable range are more likely to be honest when reporting personal perceptions of

African American males.

Definition of Terms

Attributional describes justification for discriminating against people of color. Individuals with negative racial attitudes make negative responses, decisions, or ratings when conditions are unclear and cannot be directly linked to race, which does not expose their racism, thus preventing exposure as a racist (Hing, Chung-Yan, Hamilton, & Zenna, 2008)

Construct Validity Scale. Creswell (2012) describes construct validity scale as the result of statistical procedures that correlate items on an instrument in the factor analysis process (p. 399)

Emotional Disturbance/Behavior Disorders (EBDs) - Any student qualifying for special education for the following reasons: (1) interpersonal problems with peers and teachers, (2) inappropriate types of behavior or feelings under normal circumstances, (3) pervasive mood of unhappiness or depression, and (4) a tendency to develop physical symptoms or fear associated with personal or school problems. Overall, Foster (1995) describes ED/BD in practical terms as severe acting-out, aggression, or threatening behavior in the school setting (p. 38). Schizophrenia is included among the conditions for a child to qualify for protections under Individuals with

Disabilities Act IDEA §300.8 (c)(4)(i) & (ii) in the 2006 Amendment of the law. 21

Explicit Bias is consciously held negative evaluation of an outgroup retrieved activated from memory and can be self-reported (Hing, Chung-Yan, Hamilton, & Zenna, 2008, pp. 971 – 972).

Hypotheses are assumptions, beliefs, and expectations used to gather evidence to support said beliefs used in Theory (Nickerson, 1988; Foster, 1995)

Implicit Bias is prejudice that is an automatically activated negative association with an outgroup (see, e.g., Cunningham, Nezlek, & Banaji, 2004, Hing, Chung-Yan, Hamilton, &

Zenna, 2008).

Perceptions are "generally held stereotypes, beliefs, feelings, and fantasies" (Foster, 1995, p.

52).

Racial Attitudes refer to beliefs, stereotypes, and affective domains that people have of other racial groups (Quinn, 2017)

Racial Politics refer to the applied practices with enumerable occurrences that show patterns of behavior towards racial minorities in American institutions (Bell, 1988).

Racial Power - Interests of the privileged and the mechanisms that serve them (Crenshaw, 2011;

Haghighat, 2001)

Racial Principles - A sub-dimension of racial beliefs that a member of the dominant races may have regarding racial minorities (Quinn, 2017)

Racial Stereotypes - An exaggerated belief associated with a [racial] category to justify [and] rationalizing [individual and group] conduct toward [African Americans] (G. Allport, 1979, as cited by Solórzano, 1997; Crenshaw, 2011; Haghighat, 2001) 22

Structural Racism - Confluence of a system and structural conditions whereby one way of being or characteristic produce a negative consequence for cultural and racial minorities

(Ladson-Billings & Tate, 1995).

Validity Tests are used to measure how desired attributes systemically and precisely (Kane,

2013)

Organization of the Study

This research study is composed of six chapters. Typically, a dissertation will comprise of 5 chapters (Roberts, 2010); however, developing an instrument involves a detailed review of the literature. Consequently, this dissertation's Chapter 2, which is the review of the literature is divided into two chapters, Chapter 2 and 3. More information regarding Chapters 2 and 3 is presented later in this section.

In this Chapter, I provided background that builds the foundation for the study of educator perceptions of African American males in public schools; an overview of the problem, and the importance of the study. Additionally, Chapter 1 provides an expanded definition of perceptions of African American males, the research hypothesis, researcher positionality, delimitations of the study, and definitions of terms. The remaining chapters include the literature review presented in two parts. Chapter 2 contains a review of the literature that aids in building an argument related to perceptions of African American males, Critical Race and social construction of race in school discipline theoretical frameworks, as well as measuring educator perceptions. Chapter 3 will provide a review of the literature regarding instrument development, including related topics like, psychometric properties of instrument development, stages of 23

instrument construction, reliability, validity, and establishing a valid interpretation and use argument.

Chapter 4 contains the methodology and procedures used to develop the PEPAAMS

PK12. Subsequently, Chapter 5 provides the results from reliability and validity tests and other psychometric property evaluations of the PEPAAMS PK12, an instrument that measures educator perceptions when self-reporting responses for the average person and personal perceptions. Finally, Chapter 6 includes conclusions, implications for school leaders, as well as policy and practice, recommendations for future research, and final thoughts. 24

CHAPTER 2. LITERATURE REVIEW

The present study has significant meaning because the results reported by Foster (1995) provided a narrative to start an inquiry. Throughout this chapter, Foster’s (1995) study introduced in Chapter 1 and detailed in this chapter will be referenced as educator noneducator perceptions of Black males (ENEPBM) study. Foster’s study inspired me to examine how racial biases, negative perceptions, and stereotypes might play a crucial role in decision-making for public educators. Foster linked perceptions of young Black men to special education eligibility for emotional and behavioral disorders (EBD). Similarly, Oelrich (2012) believed that a probable link to a disproportionate representation of African American males for EBDs in special education could result from cultural biases and misunderstandings. The results from the

ENEPBM study will serve as the primary source to create a valid and reliable instrument for use to examine educator perceptions. Chapter 2 will construct the argument that examining educator perceptions of African American males in decision making is a relevant topic of research. In this

Chapter, a summary of four studies will raise the argument that (1) educator perceptions of

African Americans are more negative than those of non-educators (Foster, 1995; Quinn, 2017),

(2) African American males are negatively perceived as being a bigger and stronger, with an increased capacity for threat and harm (Wilson, Hugenberg, and Rule, 2017), and (3) negative perceptions effect decision-making in special education eligibility, not favoring Black males

(Fitzgerald, 2009).

An instrument that measures public educator perceptions of African American males in

PreK – 12 schools will allow for further investigations linking decision making to negative perceptions. Researching young Black men is taboo (Foster, 1995, p. 52). Quantitative studies 25

that focus on the negative effects of racial bias, negative stereotypes and misperceptions against

Black men appear to be non-existent. For example, Katz (2012, studying minority women in school superintendent roles, implied that scant research regarding Black men in viable school leadership roles made their inclusion in the research study futile. Quantitative studies of young

Black men in significant societal roles is scant, even non-existent. Most research regarding

African American males employed qualitative methods (e.g., Beckford, 2016. The lack of quantitative research of African American males is not seen as a challenge, yet an opportunity to add relevancy for a subject routinely ignored in research.

Process of Searching for Key Terms

Key terms used in the search for the review of literature was conducted by recording recurring themes that presented during the course of reading research articles. For example,

Foster (1995 provided the framework that led me to Quinn (2017 who focused on educators’ perceptions and racial attitudes and bias. Additionally, Quinn’s (2017 study led me to examine racial attitudes and explicit/implicit biases. Prior to settling on two theoretical frameworks,

Simson (2013 provided context to focus on implicit biases that determine how behaviors of

African American students are interpreted. Furthermore, Critical Race Theory (Davis, 2011;

Delgado, 1995; Harris, et al., 2012; 1983; Solorzano and Yosso, 1995 informed Simson’s query to determine that racial bias is the driving force behind stigmatization and marginalization. In its purest form racism emanates from social meanings that result from biological and ecological factors, which determine how members of certain outgroups are viewed (Beckford, 2016;

Haghighat, 2001, 2005. Researching implicit and explicit biases led me to Hing et al. (2008 and

Dovidio et al. (2002, which factored in prejudice profiles and political ideologies. Summarily, 26

studying the dynamics that influence racial attitudes and the complexity of its topic is inexhaustible. Notwithstanding, in order to understand how African American males are perceived, it was important to include likely factors that are existential and metaphysical in addition to those that are concrete (e.g., physical aggression and marginalization of Black males;

Fitzgerald, 2009, Jackson & Crawley, 2003; Wilson, et al. 2017. Relating to the present study the most recurring terms include, but are not limited to confirmatory and racial biases, racial attitudes, marginalization, stigmatization, truly low prejudices, social/political conservatism, aversive racist, modern racists, attributional ambiguity effect, social desirability, outgroup, properties of Whiteness, critical race theory, social construction of race, decision-making, educators’ perceptions, stereotypes, and beliefs.

Educators and Noneducators’ Perceptions of Young Black Men

The study of educator and noneducator perceptions of young Black men’ survey

(ENEPBM; Foster, 1995 was a qualitative study that concluded that perceptions of African

American males are profoundly negative (Foster, 1995. This study served as the inspiration for the present study. Foster’s (1995 study was used to develop questions for the present study’s

Public Educator Perceptions of African American Males Survey (PEPAAMS PK12. Five categories - crime, education, personal characteristics, attitudes, and sexual prowess - provide sub-variables that will be used to understand the nature of perceptions. At the conclusion of this study, an instrument to examine public educators’ perceptions of African American males will be available to further examine educator perceptions of African American males in public schools.

The five categories listed previously will serve as sub-variables related to perceptions of

African American males. The study of educator and noneducator perceptions reported five 27

categories by name. Table 2.1 provides a brief description of major research studies included in this chapter. Additionally, information from the research studies provides evidence for content validity relating to perceptions of African American males. 28

Table 2.1

Studies Summarizing Perceptions of African American Males

Significance of the Variables related to the Research Author(s) Relevance study study Design

Reported perceptions of (1) crime, (2) education, (3)

**Foster, Young Black men held byattitude, (4) negative validity evidence Qualitative Survey 1995 educators and personality, (5) sexual for content noneducators prowess

Quinn, Reported findings three attitudinal Quantitative validity evidence 2017 regarding racial dimensions (1) perceived Correlational for content; attitude trends of beliefs, (2) stereotypes, (3) Research supports educators PreK-12 – affective orientations. Design theoretical secondary and post- frameworks secondary educators using GSS data from 1972 – 2014 published by the National Center for Educational Statistics Wilson, Reported findings from Stigmatization; Black men Quantitative validity evidence Hugenberg a series of 7 related are perceived as (1) Experimental for content; & Rule, segments that indicate stronger, (2) larger, (3) Research supports 2017 Black men are viewed capable of harm, (4) Design theoretical as larger, stronger, and phenotypical features frameworks capable of harm distort perceptions

Fitzgerald, Reported findings that Stigmatization; educator Experimental provides validity 2009 examined special decision making: (1) Research evidence for education eligibility special education Design content; supports and psychotropic drugs eligibility, (2) emotional theoretical prescriptions used to and behavior disorders frameworks control young Black men in a midwestern public school district 29

Educator/Noneducator Perceptions of African American Males

Young Black men experience disaffection (M.L. King, 1965) in public schools in the

United States. Furthermore, young Black men are considered to be one of the most disliked of all minority groups (see Hugenberg, Wilson, & Rule, 2017. Foster (1995 reported findings from a qualitative study of educator and noneducator perceptions of young Black men (ENEPBM. A total of 3,130 educators and noneducators; (1627 educators and 1503 noneducators were surveyed to understand perceptions of young Black men by comparing the two groups. The study drew unverified conclusions that indicated young Black men are more likely to be suspended or expelled for subjective infractions. Researchers like Beckford (2016 and Simson (2013 had similar beliefs. Harsher punishments for young Black men are believed to be based for race and gender (Beckford, 2016; Simson, 2013. Most disconcerting, African American males are likely to have disproportionate representation in special education (Foster, 1995; Kunjufu, 2005;

Noguera, 2008; Oelrich, 2012. Foster (1995 and Oelrich (2012 asserted that young Black men’s eligibility in special emotional and behavior disorders (EBDs are the primary areas of disproportionality. Foster (1995 recounted a story from his colleague regarding a 17-year-old

African American male in a program for students with severe behavioral and emotional disorders. According to Foster (1995 the documentation used to justify his eligibility in special education for emotional and behavioral disorders was the way he was observed walking by his teacher, a White female. The style of walking, “ditty-boppin,” was viewed as sexually provocative. Foster concluded that perceptions of young Black men are based on race. Simson

(2013 provided another example of how perceptions lead to subjective decisions based on race are likely to affect African American males and females. Simson (2013 along with Beckford 30

(2016) asserted similar beliefs that racial bias via phenotypical characteristics like dress, accent, language, religion, walking, and socializing patterns determines how behaviors of African

American students are interpreted.

Foster hypothesized that educators’ conscious or unconscious racist or “ethnocentric” stereotypes of young Black men may lead them to interpret these students’ language and behavior in ways that influences their decisions (e.g., referrals to special education; Foster, 1995, p. 40). Foster’s participants were asked to list the stereotypical beliefs, feelings, expectations, or fantasies that people have about young Black men. The ENEPBM study concluded that views of

African American males are pervasively negative. He disclaimed that researching young Black men “was a taboo subject” (p. 52). He also acknowledged that 50 to 75 individuals refused to participate. In some cases, he was called a racist for conducting the study (Foster, 1995). For some, Foster described the negative emotional display of anger, followed by refusal to participate in the study as typical reactions. Foster reported that some participants simply wrote that they did not know what the average person felt (Foster, 1995, p. 61). Results from the

ENEPBM study indicated that non-educator perceptions of African American males were more negative regarding neurobiological attributes, a term used to describe behaviors connected with learning (Beckford, 2017). Surprisingly, educator perceptions of young Black men were more negative for attitude, negative personality, and sexual Prowess

The study of ENEPBM spanned a period of four years as Foster traveled the country lecturing, conducting workshops, and teaching graduate courses (Foster, 1995). Surveys were also administered to students taking courses taught by others. Participants were asked to list all the stereotypical beliefs, feelings, expectations, and fantasies that the average person has about 31

Black men. Six subcategories – attitude, crime, education, negative personality, sexual prowess, and special education for emotional and behavioral disorders were representative of responses of that best fit the concerns of the research study. Themes that emerged were “gender, social class, expulsions, discipline, suspension, special education, race and ethnic stereotyping” (p. 53.

Results from the study indirectly point to negative perceptions of educators regarding African

American males and might be a factor for eligibility in special education, specifically, programs for EBDs (Foster, 1995. Results were reported in percentages, with a total of 19 items for each of the six categories are used narrow the focus of the present study. Table 2.2 lists all the items under each category used to describe overall perceptions of Black males.

Table 2.2

Overall Perceptions of Black Males (Foster, 1995)

Attitude Crime Education Personality Sexual Prowess belligerent dumb larger penis lazy criminal mean ignorant strong sex drive shiftless dishonest vain abusive low intelligence good lovers irresponsible steals aggressive uneducated rapes women drinks sells drugs threatening unmotivated plays around on welfare uses drugs violent apathetic abuses women

Overall, 47.1% of educators believed that the average person viewed young Black men as lazy, shiftless, and irresponsible, compared to 38.1% of noneducators. Furthermore, of White male educators, 49.1% reported that young Black men are perceived as lazy, shiftless, and irresponsible, compared to 45.7% of White female educators. Despite the difference between

White male and female educators, it could be argued that White female educators, comprising the majority of educators in the United States, might have a greater effect (Beckford, 2016). For 32

example, in 2012, the latest data available from the National Center for Education Statistics

(NCES), 76.3% of all American PreK-12 educators were female; and of them, 81.9% were

White.

The study found that young Black men are perceived as more likely to abuse alcohol and other substances. Researchers like Watkins, Walker, and Griffith (2010), Beckford (2016),

Shakoor and Chalmers (1991), and Massie (2015) seem to support the results reported by Foster

(1995). More educators than noneducators reported that the average person views young Black men negatively related to sexual prowess, particularly having “stronger sex drives” and being

“good lovers” (p. 60). Thirty-one and six-tenths percent (31.6%) of educators believed that young Black men are perceived to have stronger sex drive and acute abilities in performing sexual acts compared to 21.4% of noneducators. Most concerning, White women educators and noneducators ranked higher than White males on perceptions of sexual prowess.

Perceptions that young Black men rape, pimp, and mistreat women were sub-categories in the category of sexual prowess. More educators, 12.2%, believed that young Black men are perceived to rape, pimp, and mistreat women compared to 10.6% of noneducators. White female educators, 12.4%, reported higher negative perceptions in the sexual prowess category compared to 7.8% of White males. Interestingly, Black male and female educators reported higher percentages than all subgroups in each sub-category related to sexual prowess, except for rapes pimps, and mistreats women. Foster explained that Black educators are keenly aware of how

Black males are perceived but may not subscribe to the same perceptions. Overall, the results indicated that Black educators have differing views on perceptions favoring young Black men. 33

Foster’s study divided race into three categories: White, Black, and Non-Black People of

Color (NBPOC). Non-Black persons of Color included Asian, Pacific American, Hispanic,

Native American, and Other (Foster, 1995, p. 52). Results indicated that male educators identifying as NBPOC ranked higher in reporting negative perceptions of young Black men when compared to White men within each category. For example, there was more than a 20 percentage point difference between NBPOC male educators and White educators (male and female) who believed that young Black men were irresponsible, lazy, and shiftless. Female and male educators identifying as NBPOC reported higher perception percentages in all perception categories, except sexual prowess, than all other subgroups. Overall, male NBPOC educators showed the greatest difference in negative perceptions when compared to noneducators in most categories.

There may be an explanation for the differences in NBPOC educator perceptions because of their small numbers within the teaching population compared White and Black educators.

According to the latest data available from the National Center for Education Statistics in 2012,

Asian, Pacific American, Native Americans, and Other comprised 3.4% of the U.S. teaching force compared to Black and White educators (6.8% and 81.9% respectively). Hispanic educators, according to latest teacher demographic data available from the NCES make up 7.8% of educators in 2012, more than African American educators. Foster (1995) included Hispanics and Latinos as NBPOC.

To test for reliability for the defined categories and sub-categories, Foster randomly selected 50 completed open-ended responses. Responses were checked by classifying the open- ended responses into the 31 sub-categories. From the 50 randomly selected forms, 25 were 34

randomly selected and sorted by the defined sub-categories. Results were compared to data entered in the computer. It was not clear at which point data were entered into the computer and at what phase of the process; however, Foster asserted that the data were found to be 95 percent accurate, which was considered to be acceptable because of the large sample size (p. 67).

The study’s large sample size provided confidence that the emerging themes gathered were indicative of generally accepted beliefs held by the average person of young Black males in society. Conversely, the large sample size also poses a challenge because Foster did not provide specific details about how data were entered and categorized. According to Maxwell (2013), reading data collected in a qualitative study is a first step in analyzing. Notwithstanding, understanding the researcher’s process of data analyses and categorization strategies that led to coding and thematic analyses, was not altogether clear. I used Creswell (2013) and Maxwell

(2013) to make certain assumptions regarding data collection and analyses associated with this study: (1) Foster read responses from all 3130 participants, (2) He made sense of the data throughout the process to understand the overall patterns trends associated with participants in the study using grounded theory to determine how data were categorized, (3) The large sample size afforded opportunities to identify patterns, problems, and repeated themes along the way as data were collected and categorized over the four-year period, and (4) Coding and categorizing was on-going and consistent throughout the study.

Foster (1995) included “other” categories for two areas related to perceptions of young

Black men: attitude and negative personality. The other categories for attitude and personality indicated that educators’ perceptions were higher compared to noneducators, the largest differences were among White educators. To better understand “other” categories identified the 35

study, I contacted Dr. Foster in September 2016. Specific to my inquiry were the types of words or word phrases used to justify his decision to exclude specific details about the “other” categories in reported results. Dr. Foster responded to my email within three days. We spoke briefly on the phone. He promised to try to locate the information from the study, “if he still had it,” he said. There was no further response after several attempts were made to follow up with him by e-mail and phone. Subsequently, the decision was made to focus the present study on five of the six categories reported by Foster. I excluded “good at sports” because it was not relevant for the present study. Again, the five categories from Foster’s study are attitude, crime, education, personality, and sexual prowess. Perceptions related to athletics and “other” categories were excluded because they fell outside the scope of the present study.

Racial Attitudes of PreK – Postsecondary Educators and Noneducators

Quinn’s (2017) findings seemed to support results reported by Foster (1995) regarding negative perceptions of African Americans. Quinn (2017 was guided by two research questions:

(1 what are the racial attitude trends over time, and do the trends differ for educators and noneducators and (2 to what extent do demographics explain educator and noneducators differences in attitudes or attitude trends (p. 400. Quinn (2017 reported findings from three groups: (1 PreK-12 educators, (2 postsecondary educators, and (3 noneducators. He used

General Social Survey (GSS data from 1972 through 2014 to examine racial attitudes toward

African Americans over time. Results indicated that some racial attitudes of PreK– 12 and postsecondary educators were more negative than noneducators. Similar results were reported in the ENEPBM survey conducted by Foster (1995. 36

Quinn (2017) wanted to understand educator differences in racial attitudes in relationship to noneducators. In addition to using GSS data to document PreK– 12 and postsecondary educator racial attitudes, comparing them to noneducators, his study investigated how educator racial attitude differences can be explained by demographic variables (p. 397). According to

Quinn (2017), racial attitudes refer to beliefs, stereotypes, and affective domains that people have of other racial groups (Quinn, 2017). Affective domains include social emotional, social distance, and collective resentments (Quinn, 2017). Accordingly, educators who subscribe to racist principles are likely to discriminate and not put forth quality effort in educating students in racially minoritized groups (Quinn, 2017). Conscious or subconscious, educators who believe that certain racial groups, specifically African American males, lack the neurobiological behaviors to learn, will not expend the effort to provide high-quality instruction (Beckford, 2016;

Kunjufu, 2005; Quinn, 2017;). Social emotional orientations are the feelings that regulate how educators interact with students across races. Social distance characterizes how close educators feel to students in other racial groups. Collective resentments are acrimonious feelings toward racially minoritized students who are viewed as taking advantage of government funded social accommodations to advance themselves using different rules to gain advantage over Whites

(Quinn, 2017).

Quinn used the results from his study to conclude that new teacher candidates are ill prepared to teach students from different backgrounds. More importantly, some teacher candidates believed that racism is not a factor and that minoritized racial groups may overreact to racial statements. Quinn’s assertions were indicative of findings from relevant researcher studies like Crenshaw (2011), Davis (2012), and Hing, Chung-Yan, Hamilton, and Zanna (2008). 37

Furthermore, Quinn (2017) asserted that teachers’ unions have more power than noneducators to shape educational policy. Indirectly, he asserted that negative racial attitudes held by educators are indirectly correlated to the influence and protection of teachers’ unions who have the power to shape policy.

The study found that educators preferred less government involvement in race relations, including programs to equalize access for African Americans (Quinn, 2017). The study also reported that educators believed that people who hold racist attitudes should be allowed to teach college and are “slightly” less likely to support a law that allows homeowners to refuse to sell their homes to African Americans. The author’s use of the term “slightly less likely” may reflect a concern that data suggest a small margin supporting a law to prohibit African Americans from purchasing homes in certain neighborhoods might be an issue when controlling for social desirability responses. Additionally, there was no information regarding how self-report bias was considered to interpret GSS data. When participants in a research study answer questions to appear to be good or socially accepted, examination of self-report bias is recommended to prevent measurement error (Crowne & Marlowe, 1960; Haghighat; 2007; Pohan & Aguilar,

2001; Strahan & Gerbasi, 1972; Van de Mortel, 2008).

Educators were “slightly less likely” to oppose preferential hiring for African Americans during years 1994 to 2014; and from 1990 to 2014, they were “slightly less likely” to believe that a White person would be turned down for a job or promotions in favor of an equally or less qualified Black person (p. 401). From 1972 to 2014, GSS data indicated that racial beliefs toward racially minoritized groups were negative for all groups. However, racial beliefs held by educators were less negative in comparison to noneducators. Quinn (2017) raised concern 38

regarding some racial beliefs held by educators. He gave a specific example of troubling racial beliefs of educators regarding racist speech of students from the racial majority against racially minoritized groups and the first amendment right of free speech. Data indicated that racist speech was weighted against racially minoritized students’ civil rights. Educators believed that deference should be paid to free speech. Researchers like Karpinski (2006 and Lewis (2008 asserted that racially charged speech and aggressive and violent actions, post-desegregation, resulted in retaliation from Black students, resulting increased suspensions and expulsions of

African American students.

Racial Attitudes, Principles, and Stereotypes

Racial principles are a sub-dimension of racial beliefs. Racial beliefs are indicative of prominent racist ideologies used justify the impracticality of civil rights demands of African

Americans and other racialized minorities (Crenshaw, 2011; Quinn, 2017. Quinn reported that racial beliefs held by PreK– 12 and postsecondary educators reflected beliefs that the government should be less involved in race relations. As mentioned previously, Quinn reported that educators were more likely to believe that racists should be permitted to teach at the college level. Furthermore, educators were less likely to believe that a racist book should be removed from the library. See Table 2.3 below.

Table 2.3

Differences in Racist Principles Held by Educators and Noneducators (Quinn, 2017)

PreK-12 Postsecondary Noneducators

Racists should be allowed to teach 54% 67% 49% college Racist books should be removed from 27% 15% 35% school libraries 39

The table above depicts how educator racial attitudes compared to noneducators are more troubling and should be considered as potentially harmful to racially minoritized students in public schools. Quinn’s study also affirmed what other researchers have concluded: ahistoricity, or unwillingness to value the experiences of racialized minorities, is a roadblock to supportive learning environments where students’ lived experiences are valued and celebrated (Lintner,

2004; Solórzano & Yosso, 2001). The implications of racists teaching college might suggest a harmful trajectory that could lead to a decrease in African American males completing college,

(Giuffrida, 2005). Educators who hold racist ideologies should be a serious concern, particularly when decisions are made that have any attribute of ambiguity (e.g., grading, academic advising, and educational placements (Carr & Klassen, 1997; Fleming, 2005; Giuffrida, 2005; Harber,

Gorman, Gengaro, Butisingh, Tsang, & Oulette, 2012). Critical race scholars proposed that despite denunciation of racist views, racism is a protected fixture in American institutions

(Harris, 2012). As a result of educators possessing racist principles in protected spaces, negative attitudes they hold could lead to microaggressions or subtle racial attitudes that influence educator behaviors and decisions (Linter, 2004; Quinn, 2017; Simson, 2013).

Quinn (2017) reported the dichotomy in racial stereotypes of educators and noneducators related to African-Americans and other racial minorities (e.g., Asian, Hispanic/Latino), consisting of (1) hardworking vs. lazy, (2) intelligent vs. unintelligent, and (3) violence-prone vs. not violence-prone, educators rated African Americans neutrally. Conversely, educators rated

Asians and Hispanics more favorably in the same areas. Educators rated Whites as being more intelligent, not violence prone, and harder working. Racial stereotyping reflects how individuals view certain racial groups in the United States using them to justify behaviors toward them 40

(Haghighat, 2001; Nickerson, 1988; Solórzano, 1997). Furthermore, race as a social construct originates from racial superiority bestowed upon the racial majority (Bell, 1998; Harris, 1993;

Johnson-Bailey & Baumgartner, 2008; Solórzano & Yosso, 2001), as evidenced by how White educators rated themselves (Quinn, 2014). The racial stereotypes published in this study are contextualized by comparing how educators rated Hispanics, Asians, and Whites, more favorably than African Americans. Educators rated African American neutrally in racial stereotypes.

Neutrality may be understood to be a negative phenomenon because participants’ attempts to conceal racist views (Haghighat, 2005; Van de Mortel, 2008; Quinn, 2017).

Quinn’s reported findings indicated that negative racial attitudes, beliefs, and stereotypes held by educators exceeded noneducators in some areas. However, Quinn (2017) posited that students are less likely to experience negative racial attitudes inside the school than outside.

Negative racial attitudes can harm student development (Beckford, 2016; Quinn, 2017).

Solórzano and Yosso (2001) suggested that harmful racial attitudes can be ameliorated. Kunjufu

(2005) posited racial microaggressions can be eradicated in schools. Solórzano and Yosso (2001) and Kunjufu (2005) agree that culturally appropriate curriculum, acknowledgment and acceptance of negative racial views, and redeveloping school policies that had been designed to marginalize African Americans and other students of color are fundamental keys to success for

African Americans.

Quinn admitted that some educators held racial attitudes that were detrimental to students’ learning and development (p. 409). For example, data from 2014 indicated that an estimated 4% of all PreK-12, and 3% of postsecondary educators reported believing that African

Americans have less “inborn” abilities to learn. It is also concerning that in 2014, 31% of PreK– 41

12, and 20% of postsecondary educators reported that African Americans lack the motivation or will power to pull themselves up out of poverty (p. 209). Whether educators believe African

Americans have lower “inborn abilities” to learn or are devoid of motivation and “will power” to overcome poverty, negative racial attitudes held by educators may be tied to affective orientations that are expressed as “collective resentments” by shifting the blame from educators to racially minoritized students and parents (Kunjufu, 2003, 2005: Noguera, 2012). The results from this study and others like it provide a fundamental imperative that multiple dimensions of racial attitudes of educators should be examined and trends recorded through quantitative research.

Results from this study warrant further examination due to some limitations. First, the author uses the total data set of the (GSS) from years 1972 through 2014, which may be a concern because certain items are taken out of circulation over time.

Consequently, replicating this study may be very difficult. Next, decision making for coding and creating variables may vary when considering individual research objectives. Any attempt to replicate a study involving missing or incomplete data may have some inconsistencies due to varying decisions for how to code, interpret, and test data assumptions (Mertler & Vannatta,

2013). Furthermore, the study has some limitations regarding validity evidence for reliability because items that require participants to self-report personal views might result in measurement error due to self-report and social desirability biases. Researchers like Van de Mortel (2008) were concerned that participant responses on topics that attempt to present themselves more favorably or to hide true feelings should include a scale to control for bias. This is known as the 42

bias of social desirability (SD. Using social desirability scales can "detect, minimize, and correct" when bias is present (Van de Mortel, p. 40).

Young Black Men Perceived as Bigger, Stronger, and More Capable of Doing Harm

Racial biases against Black males are realized by misperceiving them in in size, formidability, and threat. Inaccurate size, formidability, and threat perceptions can have life impacting consequences for Black males in American society. Wilson, Hugenberg and Rule

(2017) provides a third leg of the four studies that will be used to bolster the argument that racial bias and negative stereotypes of African American males warrants examination of PreK-12 educator perceptions, which will inform future studies regarding their impacts on educator decision making (Rivard, Missiuna, Hanna, and Wishart, 2007; Simson, 2013; Solórzano, 1997).

Additionally, they provide evidence used to critique a system that is seen by some researchers like Kunjufu (1995), Fitzgerald (2009), Crenshaw (2012), and Beckford (2016) as institutions that perpetuate psychological harm to many African American males.

Crenshaw (2011) suggested that institutional and structural racism within schools result in racial harm to students of color, particularly African American males. Another study conducted by Fairchild, Tobias, Corcoran, Djukic, Kovner, and Noguera (2012) examined relational demography and job satisfaction between White and Black teachers. One finding of this study revealed that White teachers who teach Black students are less satisfied with their jobs when compared to Black teachers. Fairchild’s (2012) findings, along with other studies presented in this chapter will implicate biases related to race could negatively impact emotional development and produce psychological harm for students in racially minoritized groups. This 43

assertion cannot be empirically proffered because current measures are inadequate and do not test educators perceptions of Black males.

Wilson et al. (2017) examined racial bias in judgment of physical size and threat. Racial biases against Young Black men were realized by misperceiving them in in size, formidability, and threat. Inaccurate size and threat perceptions have been shown to have life-impacting consequences for young Black men in the United States. Wilson et al. (2017) studied threat perceptions in his study with eleven parts.

Misperception of size and threat is a compounded concern when combined with stereotypes of African American males. For example, young Black males are perceived as prone to commit sexual violence against women (Foster, 1995; Gates, 1997). According to Wilson, et al. (2017), White males are prone to be racially biased in their perceptions of young Black men.

Wilson et al. also asserted that White females also view African American males as bigger, stronger and more capable of harm; however, unlike their male counterparts, misperceptions in size is motivated by fear (Wilson, et al., 2017). Foster (1995) reported that White female educators had higher perceptions of violence toward women (e.g., rapes, pimps, and abusive toward women). Table 2.4 lists each part of the study conducted by Wilson et al. (2017), the purpose, sample size, and an overview of findings for each part. 44

Table 2.4

Segment Summaries of Racial Bias in Judgment of Physical Size and Threat (Wilson,

Hugenberg, & Rule, 2017)

Purpose Sample Findings Study 1A tested height and weight nn = 60 significant difference found in height and estimates of Black and weight estimations, with Black targets White male faces (height perceived taller and heavier than White targets and weight not available)

Study 1B replicated Study 1A height n = 47 significant differences in participants’ and weight data available perceptions rating Black target taller and from high school football heavier than White targets, controlling for players recruited to play targets actual height and weight. college football with young men 16–19 years old Study 1C n = 60 Participants selected more muscular bodies for tested race-based the Black than for the White targets and were misperceptions to paired with larger, muscular physiques when perceive formidability matching target faces to bodies Study 1D investigated perceptions n = 60 using mixed model analysis, participants of physical formidability rated Black targets stronger than White by estimating targets’ targets physical strength.

Study 1E tested participants significant evidence for racial bias in n = 60 estimations of strength and judgments of strength, controlling for bench fighting ability using photos press totals, target height, and weight of NFL bodies and face photos and draftees. other data from 2015 NFL Draft Combine Study 2 investigated impact of n=168 misperceptions on judgments Participants judged the Black targets as of capability for physical more capable of harm than White targets. harm.

Study 3 investigated differences in n=120 Muscularity differences predicted harm racially biased estimates of differences. Prejudice differences significantly muscularity to predict predicted harm differences as well differences in racially biased estimates of harm 45

Study 4 Tested relationship of n= 60 Participants judged Black targets (more misperceptions of size and muscular than White targets but the harm capability for Black and participant’s race qualified this difference. White targets differing by race. Study 5 examined perceptions of n = 80 Participants rated the use of force against Black appropriate use of force men more justified than the use of force against by police White using 90 faces from 1B

Study 6 examined (a) Black facial, n = 60 Face perceptions of racial prototypicality skin one influences on predicted judgments of physical formidability. perceptions of fighting ability (b) Afrocentric looking faces. (c) how facial features relate to physical size that signal aggression Study 7 investigated imposing n =120 significance found for bias judgments in race and beliefs about the race of physical size of identical bodies when believed to racially ambiguous targets be Black or White. Racially ambiguous bodies affect judgments of perceived taller and heavier when labeled as physical size Black.

Wilson et al. (2017) referenced the deaths of Dontre Hamilton and Tamir Rice as a pretext to confirm why examining racial biases of perceivers from the racially dominant group have broad consequences. Dontre Hamilton was shot 14 times by a White police officer in

Milwaukee, Wisconsin; and Tamir Rice, a 12-year-old Black male, was shot and killed by a police officer in Cleveland, Ohio. Both Dontre and Tamir were killed within months of each other in 2014. In 2012, Trayvon Martin, an unarmed young Black male, was killed by a community patrolman in Florida because a he was viewed as a potential robber in a White 46

community (Wilson, et al., 2017). Researchers tested the hypothesis that race distorts perceptions of physical size and threat.

All participants from the study were U.S. citizens. African American participants were excluded from the majority of stud segments because researchers wanted to test hypotheses using the racial majority representative in the United States. For this study, researchers used the terms

“White perceivers” and “participants” to describe the sample population in reported findings.

Hereafter, they will be called “perceivers” or “participants”, used interchangeably. Additionally,

Wilson, et al. (2017) used “target” or “subjects” referring to images of young Black and White men used to examine perceptions held by perceivers. Targets and subjects will also be used to describe the images as objects of the study.

In studies 1A through 1E, Wilson, et al. (2017) examined how racial bias plays an integral role in judging size and threats by perceivers. For example, in Study 1A, perceivers were asked to estimate the height and weight of young Black and White men using photos of their faces without having actual data for the images used. In study 1B, researchers used actual height and weight data for subjects to test the hypothesis further that racial bias in height and weight perceptions of Black and White subjects was an integral factor in making judgments. According to Wilson, et al. (2017), White subjects were “marginally taller”, yet significantly heavier than

Black male subjects in their study. Study 1C, used 90 athletes’ faces along with arrays of seven male body forms that varied in overall body size and muscularity (see Frederick and Peplau,

2007), but proportional to height and muscularity. Perceivers viewed randomly selected faces to judge the body that was the best match for each subjects’ face. For example, researchers used a 47

7-point scale, beginning with 1 (shorter less muscular) to 7 (tallest, most muscular). (Wilson, et al., 2017, p. 5).

Researchers reported that participants distorted the images of young Black men. The perceived to them larger than their actual sizes, and in 1C, perceivers assigned more muscular builds to Black subjects (Wilson, et al., 2017, p. 6). Building on segments 1A –1 C, researchers tested the degree to which distortion in size occurred against young Black men in segments 1D and 1E. The sample size for segments 1A through 1E ranged from n = 47 to n = 60. Variation in sample size was the result of excluding African American participants or people who chose not to define their racial identity as they were not central to the study’s focus. Segment 1D, perceivers rated overall physical strength of 90 athletes from segment 1C from 1 (not at all strong) to 7 (very strong), with faces appearing in random order. Wilson, et al. (2017) believed that upper body strength was indicative of an individual’s fighting ability. Study 1E focused on the threat factor as perceived by participants related to upper body strength of male subjects.

Results indicated that participants’ with Black faces were perceived to be (1) taller than subjects with White faces when height and weight were unknown by researchers, (2) taller and heavier than White faces when the values for height and weight were known by researchers, (3) Black targets were assigned more muscular bodies over White males, (4) perceivers rated Black subjects stronger in overall strength, and (5) Black subjects were perceived to have stronger upper body strength over White subjects with similar strength capabilities. Data from this segment were based on bench press data from the 2015 NFL Combine. Segments 1C – 1D used the same athlete subjects. Summarily, segment 1D and 1E concluded that perceivers rated 48

subjects with Black faces as having greater upper body strength, which researchers believed to be indicative of fighting capability (see Arthur & Thanzami, 2007).

Race plays a crucial role in how subjects are perceived. In segment 2, Wilson, et al.

(2017) used the same subject stimuli from 1B for n = 168 participants (92 male and 72 females).

They asked perceivers to imagine that they were in a verbal altercation with the displayed targets, with the situation escalating resulting in threatening behavior by the target. Participants rated the targets’ fighting ability from 1 to 8. A rating of 1 indicated that there was no potential for harm, with 8 indicating significant potential for harm. Researchers used cross-classified regression models to conclude that participants judged Black targets as more capable of harm compared to White targets (p. 7).

Segment 3 tested the hypothesis to determine if the participants judged Black targets to be more muscular, they are also likely to be misperceived as more capable of harm than White men. Segment 4, however, explored whether misperceptions of size and harm capability among

Black and White targets differed according to a participants’ race. African American perceivers were not excluded from this segment of the study because they were central to testing the hypothesis of race, size, and threat perceptions. Racial differences in judgment using both size and threat perceptions for Black and White targets were also evaluated in Segment 4. Results from segments 3 and 4 concluded that both White and Black perceivers perceived young Black men to be more muscular. Notwithstanding, Black perceivers misjudged size to a lesser degree than did White perceivers (Wilson, et al., 2017, p. 10). More significant to this study, African

Americans viewed Black targets as more muscular, but less capable of harm. White perceivers’ perception of harm, on the other hand, was significantly biased against young Black men. 49

Researchers concluded that perceptions of harm for White subjects is directly correlated to their perception of the target’s size. The larger a Black target was perceived to be, the more they were viewed as capable of harm. Furthermore, Wilson et al. (2017) believed that racial bias influenced misperceptions for White men, while White women, were influenced by fear. Overall, racial bias was believed to motivate misperceptions of Black targets. Central to racial bias in size and threat perceptions, Wilson, Hugenburg and Rule (2017) referenced police shootings of Black men as a key motivation for studying racial bias in perceptions of size, formidability, and threat. It could also be reasonable to infer that young Black men are more likely to be shot by a White male police officer than any other racial and gender demography. According to published on the Data

USA website in 2016 there were approximately 645,936 police officers in the United States, comprised of about 86.7% males. Seventy-nine percent of male police officers are White. I argue that race and gender play a significant role in perceptions that do not favor young African

American men.

Foster (1995) and Wilson et al., (2017) are used to support the claim of race and gender’s role on perceptions of African American males. I would also argue that young African American males are more likely to be adversely affected by perceptions from White female educators that are motivated by racial bias and fear (Hugenberg, Wilson & Rule, 2017). Using racial bias as a pretext, Wilson et al., (2017) predicted that perceivers would justify force against Black targets disproportionately in favor of White targets. They also anticipated perceptions of Black targets’ physical size and formidability would predict racial bias in force justification. Racially biased judgments coupled with perceptions of physical size, and threat poses what researchers called

“downstream consequences” (Wilson, Hugenberg, & Rule, 2017, pp. 8 - 9, 14, 18), which results 50

in justification of force. In Segment 5, researchers Wilson et al., (2017) explored the role that justification of use of force against young Black men was significantly justified when comparing

White men to other demographic groups in this study. Also, consistent with their hypothesis, there was a significant indirect effect through formidability and harm composites indicating that the effect of race on use of force justification remained significant. Summarily, race alone may have accounted for variance in force justification, but perception of harm was the primary source of variance (Wilson et al., 2017, p. 12).

Throughout the series of studies, the researchers concluded the impact that negative perceptions, racial bias, and fear have on decisions. Alarmingly, researchers warn of dire consequences, citing, “such perceptions can have dangerous consequences [for young Black males in the United States]” (Wilson et al. 2017, p. 15). Support for the conclusions of Wilson et al. (2017) can be found in Ma and Correll (2011) who reported participants in their study, using simulators, mistakenly shot unarmed White targets who looked less White. Furthermore,

Eberhardt, Davies, Purdie-Vaughns and Johnson (2006) agreed that racial bias in capital sentencing is evident when considering Black appearance (p. 383).

Racial bias in individuals who hold certain beliefs about a particular race can be triggered by mere suggestion a target could be Black as shown in the results in Segment 7 (Wilson et al.,

2017). In the last segment of their study, researchers reported the results when participants were given color inverted images of male bodies to conceal their race. In case one, color-inverted images were presented to participants along with pictures of Black or White faces. Conditions were altered in case two. Participants were given stereotypical names of Black or White races, to give a suggestion of the target’s racial identity (Wilson et al., 2017, p. 15). For example, 51

DeShawn is a typical name of a young Black man (Vitaliano, 2006). For between-subjects in case one and two, researchers measured participant estimates of the target’s height and weight, to prove the hypothesis that participants would rate the bodies as larger when they believed the targets were Black versus White (Wilson, Hugenberg & Rule, 2017). They reported strong associations between races were strong enough to bias their judgments of physical size of identical bodies simply because they were led to believe images were Black. Moreover, when provided the image of a Black face or a Black sounding name, participants rated the color inverted bodies, which were race concealed as both taller and heavier. Wilson et al. (2017) concluded giving a suggestion of a person’s race, biases perceivers perceptions of physical size.

Findings from the series of studies conducted by Wilson et al. (2017) have implications for policy makers and school districts. Racial bias is a factor when estimating the size and threat of young African American males. As reported in this study, justification of force was also sanctioned more for Black targets (Wilson et al., 2017). The killings of unarmed Black men by police officers in the United States should warrant concern regarding educators who hold negative racial attitudes (Quinn, 2017), stereotypes and perceptions (Foster, 1995), and faulty estimations of threat (Foster, 1995; Wilson et al., 2017) could add additional concerns of racial harm for young African American males in public schools if teachers are permitted to be armed

(e.g., from discussion resulting from the February 14, 2018 school shooting in Parkland,

Florida). Summarily, the findings by Wilson et al., (2017) are central to the argument how racial bias could affect decision-making for educators.

Briefly, not all racial bias is transacted with physical evidence (Nickerson, 1988). Despite programs meant to diminish effects of racial prejudice, racial bias in judgments can be initiated 52

by suggesting an individual may be Black (Wilson, Hugenburg & Rule, 2017). Relevant research suggests that semantic information about an individuals’ race could result in biased consequences from hiring decisions to socially distancing behavior (Hing, Chung-Yan, Zanna, &

Hamilton, 2008; Wilson, Hugenburg, & Rule; Haghighat, 2001; Quinn, 2017). When members of the racially dominant group have access to positions of power to make decisions; they, as actors, may be triggered to (1) overreact, (2) overestimate, (3) overcompensate (Wilson,

Hugenberg, & Rule), and overstate (Ladson-Billings & Tate, 1995; Hing, Chung-Yan, Hamilton,

& Zanna, 2008) and are convinced their actions are justified (Fitzgerald, 2009). The research suggests members of the racially dominant group might have genetic predisposition to act violently towards African American males. The justification of physical aggression as noted in

Segment 7 by Wilson et al., 2017, is the result of the psychologically rationalization, which commences at the point of contact with Black males (Haghighat, 2001; Hing et al., 2008).

I argue that White male body images with the image of a Black face introduced in the study were sufficient to prove racially dominant group members can be triggered to act on racial bias and harm at the subconscious level, thus confirming what researchers believe to be a biological predisposition to initiate life threatening harm when the target is believed to be Black

(Doll, Hutchinson, & Frank, 2011; Haghighat, 2001; Pedro in Berliner, & Calfee, 1996). If biological and psychological phenomena can be linked to increased racial bias, harm, and violence toward African American males, then the argument of educator perceptions of African

American males negatively affecting decision making will have more credibility.

In a study conducted by Doll, Hutchinson, and Frank (2011), researchers concluded dopaminergic genes may predict an individuals’ susceptibility to stigmatize as a result of bias. 53

Researchers believed [stigmatizers] and people prone act upon personal biases are overly influenced by prior learning. Researchers further elaborated that the predisposition to rely upon prior learning, particularly erroneous information, poses a disadvantage to learning new or different information. Specifically, they investigated activities in the prefrontal cortex of the brain provides psychological rewards for errors in predictions in favor of faulty information.

Researchers found that the brain gathered, sorted, and applied information to efficiently make sense of large amounts of data (Doll, Hutchinson & Frank, 2011. Biases that feed stigmatizing behaviors are most likely to be reinforced by the brain in the prefrontal cortex. Researchers concluded that genetic influences on the brain suggested a permanence of striatal learning processes. Summarily, permanence of striatal learning processes, according to Doll, Hutchinson and Frank (2011, are influenced by the associative weights of prior expectations, and cannot be easily undone. The study conducted by Doll et al. (2011 supports Simson (2013 regarding educator decision-making. The brain reinforces negative data about African Americans because the brain’s biological disposition will only reinforce what it believes it already knows. The research studies presented thus far suggest racial bias may result in irrational compulsions and maladaptive behaviors that are difficult to unlearn. Later in this chapter, two studies by

Haghighat (2001 and Hing, Yan, Hamilton, and Zanna (2008 will be used to support the theory regarding stigmatizing behaviors derived from negative stereotypes and biases could be related to racist ideologies that affect educator decision making against young Black males.

Using Special Education to Control Young African American Males in Public Schools

Fitzgerald (2009) examined how control might play out for young Black males in public

schools. Previous literature implicated use of deadly force to control young African American 54

males is a serious concern. Fitzgerald (2009 believed that psychotropic drugs might be a mechanism to control young African American males in public schools. He investigated whether psychotropic medication prescriptions (e.g., Ritalin were motivated by bias; and to control academic and social behavior of Black boys (p. 225. Fitzgerald implied that federal policies

(e.g., the Individuals with Disabilities Education Act of 1990 and Section 504 of the

Rehabilitation Act of 1973 make it easy for educators to misapply special education labels on

African American males. He developed a hypothesis that proffered vague and accommodating language of the law allows for disproportionate prescribing of behavioral-stimulants, along with special education eligibility determinations, are used to control African American males.

Brief Overview of Special Education Statute for FAPE and Emotional and Behavioral

Disorders

Free and appropriate education (FAPE) requirements for students include instructional, behavioral, and placement plans for meeting short-term goals; requirements for reevaluation of the IEP; and, due process procedures when student discipline must be administered [20 U.S.C. §

1412; 1414(d; 1414(c(4; 1415(k; 1415(k(G].

According to the IDEA of 1990 and the revision in 2004, a student may qualify for protections under the law if there is a condition affecting educational performance over a long period of time and to a “marked degree” (34 C.F.R. § 300.8c4i. Any condition which cannot be explained by intellectual sensory, or health factors qualifies a student to be considered for the disability category of emotional disturbance (ED. Students who qualify for the ED distinction may according to the law have the following characteristics: (1 interpersonal problems with peers and teachers, (2 inappropriate types of behavior or feelings under normal circumstances, 55

(3) pervasive mood of unhappiness or depression, and (4) a tendency to develop physical symptoms or fear associated with personal or school problems. Schizophrenia is included among the conditions for a child to qualify for protections under IDEA §300.8 (c)(4)(i) & (ii) in the

2006 Amendment of the law. According to IDEA, socially maladjusted students are excluded from consideration for ED. Social maladjustment is not factored into the qualifying criteria for

ED; however, LEAs and school districts have the latitude to qualify students for ED using language taken from 34 C.F.R. §300.8(c)(4)(i), which qualifies a student if learning is impaired

“over a long period of time” and to a “marked degree”, affecting school performance.

Emotional and Behavioral Disorders from the standpoint of some researchers can be an affective result of health problems related to impoverished conditions, physical and sexual abuse, disrupted families, (e.g. divorce or death of a family member, and lack of education; Abubakar-

Abdullateef, Adedokun, & Omigbodun, 2017). Maladjustment as a factor qualifying a student for

ED, is defined according to IDEA, as a condition existing over an extended period of time to a significant degree. Maladjustment includes homelessness, exposure to violence, physical abuse, sexual abuse, parental conflict, psychiatric disorders of parents, substance abuse, lack of family support and community disorganization (Abubakar-Abdullateef, 2017; Shakoor & Chalmers,

1991). Fundamental outcomes for students with emotional and behavior disorders are a diminished capacity for learning resulting in poor academic performance, school truancy, increased rates of suspensions and expulsions, alcohol and drug abuse, as well as increased risk of not completing high school (Abubakur-Abdullateef, 2017; Foster, 1995; Shakoor & Chalmers,

1991; Horsford, 2011; Simson 2008). 56

Local Education Agencies (LEAs and school districts have the latitude to qualify students for ED using “social maladaptation” taken from IDEA’s §300.8(c(4(i, which qualifies a student if learning is impaired “over a long period of time” and to a “marked degree”, affecting school performance. I am using relevant research to make inferences regarding how; (1 racially biased attitudes (Quinn, 2017, and (2 negative stereotypes and perceptions of young African

American males (Foster, 1995; Beckford, 2016; Kunjufu, 1995; Kunjufu, 2002; Haghighat,

2001; Wilson, Hugenberg & Rule, 2017 could be used to create conditions that results in special education eligibility for emotional and behavioral disorders [§300.8(c(4(i]. Adult behaviors that impact the social and emotional well-being of students over an extended period of time may be the cause that many African American males are eligible for special education.

Evaluating Students Suspected of a Disability

The individuals with disabilities act (IDEA) requires each state and local education agency (LEA to conduct a full and individual evaluation before qualifying any child for services related to special education (Center for Education and Employment Law, 2011, p. 15. The law also requires that an evaluation for eligibility for special education shall consist of a variety of assessment tools and strategies in order to gather relevant functional, developmental and academic information about the child, including information provided by parents [34 C.F.R. §

300.304(1].

Under the same provision, the law specifies what is acceptable to determine eligibility for special education services. First, educators should not use any single assessment to determine whether a child has a disability. Second, the law requires that schools and LEAs use technically sound instruments that assess contributing cognitive, behavioral, physical, and developmental 57

factors. Furthermore, the law specifies that assessments should be administered by knowledgeable, trained professionals, following the directions by the producer of the measure being used.

There are some diagnostic measures that have been tested in various countries to examine validity evidence for internal consistency. The Diagnostic and Statistical Manual (DSM is a common measure that uses scales to assess affective, anxiety, hyperactivity, as well as oppositional, conduct, and defiance disorders (Rescorla et al., 2007. The American Psychiatric

Association’s, Diagnostic and Statistical Manual of Mental Disorders (DSM-V; APA, 2015 has two subscales that assess for attention deficit and hyperactive impulsive disorders. Researchers reported that the two most commonly used measures to assess boys are believed to be problematic when attempting to describe externalizing behaviors (Rescorla et al., 2007.

Diagnostic measures such as the DSM-V have areas of concern because they rely on factor analyses to determine problem behaviors, which may impact racialized and cultural minorities

(Oelrich, 2012).

Safeguards to Deter Discrimination for Racial and

The law specifies that evaluation materials used to assess a child suspected of a disability should be selected so as not to be racially or culturally biased [34 C.F.R. § 300.304(3(1(i].

Research suggests that interviews, rating scales, and may be flawed due to implicit biases held by those that administer them (Gable, Park, & Scott, 2014. Indirect approaches, such as interviews and questionnaires, are susceptible to poor memory of the person conducting them, as well as bias and other distortions (Gable, Park, & Scott, 2014, p. 114.

Accordingly, the previously described limitations may result in drawing incorrect conclusions 58

regarding the function of the students’ behavior and ineffective, or even harmful interventions

(e.g., eligibility for special education services; Fitzgerald, 2009; Gable, Park, & Scott, 2014;

Woodson & Harris 2019). FBAs identify functional relationships between environmental events and when the target behaviors occur. If done with fidelity, FBAs can assist in reliably predicting problem behaviors (Gable et al., 2014).

Oelrich (2012) believed that African American males are disproportionately represented in special education because of racial and cultural bias. More specifically, Oelrich (2012) inferred that racial and cultural biases are the result of cultural misunderstandings. Fairchild, et al. (2012) reported that White teachers’ dislike for students in racial minority groups contributed to negative outcomes for both teachers and students. White teachers expressed that their job satisfaction was diminished when required to teach racial minorities. Students in racial minority groups subjected to racial or cultural bias could be labeled with irrevocable disabilities (both emotional and behavioral) that could precipitate other negative consequences (see, Beckford,

2016). Negative perceptions can distort judgment when determining educational placements

(Beckford, 2016; Carr & Klassen, 1997; Foster, 1995; Gable et al., 2014; Kunjufu, 1985;

Noguera, 2012, Oelrich, 2012; Simson, 2013; Terry, 2007). Functional Behavior Assessments have been proven to be effective under tightly controlled clinical conditions.

The information presented in this section covered assessment protocols that are meant to assist educators in addressing the needs for students that are in need of support. A common concern regarding the DSM-V is its reliance on teacher reporting forms. Some educators are not adequately trained to administer assessments (Gable et al., 2014). Additionally, students are negatively affected when directions provided by the producers of the assessments are not 59

followed, educators do not have adequate time to conduct a full and thorough assessment of students and are alleged to have racial or cultural bias (Fitzgerald, 2009; Gable, et al., 2014;

Oelrich, 2012).

Discretionary Decision Making

Fitzgerald (2009) examined the Individual Education Plans (IEPs) of Black boys within a racially homogeneous, moderately sized public school system near a university community in

Illinois. According to Fitzgerald (2009, the socio-economic makeup of the school district ranged from upper middle class to low socioeconomic status. The population of the school district was approximately 67,552 people, with about 9,400 students. The school district had 11 elementary schools, three middle schools, and two high schools. There were nearly 456 special education cases within the district. To investigate his research hypothesis, Fitzgerald (2009 focused on young Black males eligible and referred to special education. He examined the number of young

Black males on behavioral stimulants in comparison to White males with an IEP.

Fitzgerald’s research study spanned two academic years, 2000 - 2001 and 2001- 2002.

He used referral data from building support teams tasked with processing referrals for special education. Fitzgerald (2009 examined thirty-six cases, excluding students with more than one referral, during the research study period. According to Fitzgerald (2009, all those eligible for special education having emotional and behavioral disorders (EBDs were Black males, n = 2 (p.

237. Two of four White males eligible for special education were not reported to have academic difficulties. Fitzgerald (2009 asserted that young Black men were disproportionately referred more than once for behavioral and academic challenges 60

Fitzgerald (2009) used 456 special education cases divided into three different categories,

(1) academic, (2) behavior, and (3) physical disabilities. Fitzgerald defended his methodological decisions to separate special education cases academics, behavior, and physical disabilities. He believed that they are in step with normative practices used in most education determination conferences (EDCs). The educator determination conference (EDC) is the formal process which officially defines and designates educational disabilities, based on collective agreements of

Building Support Teams (BSTs). Categorizing special education cases into the defined categories is a practice most common in a majority of public schools (Fitzgerald, 2009). For example, a disability impeding a student’s academic achievement, such as mental impairment (MI) and specific learning disability (SLD), is most often categorized as an academic disability (p. 237). In addition to MI and SLD, if educators feel a student’s diagnosis of ADD or ADHD negatively affects his or her academic achievement, they were qualified for special education as Other

Health Impairment (OHI) students with behavior and academic issues may qualify for an emotional and behavioral disability. Fitzgerald believed the rationale for including a behavioral disability in combination with a determination of SLD stems from educators having issues determining which variable to use (e.g., academic or behavior). Fitzgerald (2009) reported that students in the academic only group were not regarded as having negative behaviors. Students with physical impairments were classified as simply having one disability (e.g., physical disability). In order to investigate his hypothesis for psychotropic drug prescriptions used to control Black males, Fitzgerald (2009) created two dichotomous groups: (1) cases having psychotropic prescriptions, and (2) cases that did not. See Table 2.5. 61

Table 2.5

Summary of Race and Gender of Special Education Students and Psychotropic Drug Use

(Fitzgerald, 2009)

Total Special Academic Behavioral Disability Psychotropic Drug Education Use Black White Black White Black White Black White

Male 37% 32% 31% 35% 67% 14% 43% 34%

Female 17% 9% 18% 10% 14% 5% 10% 12%

**Controlling the Black School-age Male: results published in Urban Education (p. 238). Using estimates based on data published in this study, a total of n=456 cases were reviewed. Twenty- three cases were excluded from this study. Estimated cases included in this study include approximately 169 Young Black men, 146 White males, 77 Black females, and 41 White females.

Table 2.6 shows African American males are more likely to be eligible for special education eligibility. Additionally, Black males are more likely than White males to qualify for special education for both academics and behavior (p. 238). Finally, the table shows that African

American males in his study were more likely to be prescribed psychotropic drugs, such as

Adderall or Ritalin (Fitzgerald, 2009). 62

Table 2.6

Percentage of Students with Psychotropic Medication Use Before and Between the 3-year

Reevaluation (Fitzgerald, 2009)

No Psychotropic Drug Use Before Psychotropic Drug Use Before Reevaluation Reevaluation Black White Black White Male 43% 34% 43% 36% Female 10% 12% 9% 10%

**Controlling the Black School-age Male: results published in Urban Education (p. 238)

The table above indicates that young Black males are slightly more likely than White

males to begin psychotropic medication between the 3-year reevaluation cycles. Furthermore,

African American males are much more likely than Black and White females to start

psychotropic medications between the 3-year reevaluation cycles. The difference between Black

and White female’s psychotropic medication use before and between reevaluations is small,

which indicated that gender played a role in psychotropic prescriptions. Further analyses of data

summarized in Tables 2.5 and 2.6, suggested that young Black males in this study were more

likely to qualify for special education for academic and behavior disabilities compared to White

males. The study concluded that psychotropic prescriptions occur more for African American

males regardless of special education disability eligibility compared to their male and female

counterparts (Fitzgerald, 2009).

Fitzgerald (2009), Foster (1995), Quinn (2017), and Wilson, Hugenberg and Rule (2017)

all purported that adverse racial attitudes and biases, as well as negative stereotypes and

perceptions might be a factor that influences decision making for African American males. 63

Based on the literature presented so far implied that Black males are (1) perceived as threats, (2) are not accurately depicted when reporting facts about academic and behavior characteristics, and (3) may be victims of biased decision-making.

Fitzgerald (2009) reviewed forms used in the process to refer African American males to special education. He found that subjective statements were often included in reports, such as (a)

“should be able to interact with others in a positive manner,” (b) “not staying focused/poor attention,” (c) “difficulty following the simplest direction,” and (d) “not respectful to adults” (p.

237). He further reported that similar commentary was not present for White and Hispanic students in the referral process. Fitzgerald concluded forms submitted to BSTs, written in IEPs, and medical documents for students with psychotropic drug prescriptions used specific phrases previously referenced for students’ educators felt exhibited characteristics consistent with attention deficit disorders (ADD), attention deficit hyperactivity disorders (ADHD), and behavioral disorders (EBDs). Moreover, Fitzgerald (2009) posited that when a Black male was judged not to “interact in a positive manner” or was not “respectful to adults” or was perceived to have other behavioral characteristics that conflicted with teachers or administration, they were referred for special education evaluation or were diagnosed with a behavioral disorder (p. 237).

There are research studies whose results conflict with Fitzgerald (2009). For example,

Abidin and Robinson’s (2002) study, which examined the factors that influence referrals of students with behavioral challenges contradicted the findings of Fitzgerald (2009). Abidin and

Robinson (2002) concluded that race, gender, and social economic status did not influence teacher perceptions (p. 208). Notwithstanding, Fitzgerald (2009) postulated that his original hypothesis regarding the process to refer and qualify young Black males perpetuates social 64

control had veracity. Social control based on race is a preserved technique used in public schools

(Fitzgerald, 2009; Ladson-Billings & Tate, 1995; Noguera, 2012). However, researchers like

Noguera (2012) and Beckford (2016) offered an alternate hypothesis that supported the idea of social control materializes in the over-policing (e.g., armed police officers) in public schools with higher African American and other racialized minority populations. Both Beckford (2016) and Noguera (2012) linked the idea of social control by over policing in public schools, to feeding the criminal justice system with African American males criminalized in schools.

Overall, Fitzgerald’s (2009) study suggested that African American male membership in special education for specific disabilities are disproportionate when compared to their White male counterparts. It could be argued that Fitzgerald’s (2009) study is a reflection of national trends in school districts across the United States who educate large African American male populations.

The strength of Fitzgerald’s (2009) study offers a frame on which my argument in favor of examining educators’ perceptions that may affect decisions to refer and qualify young Black males for special education. Research is lacking regarding how perceptions influence decision- making processes of public educators. Relevant research has not linked a direct relationship of how educator biases against Black males relate to negative perceptions and stereotypes of

African American males because no valid instrument tool of measurement has been developed to date. Fitzgerald’s (2009) study affirmed the need for researchers to have access to a measure to examine educator perceptions of African American males.

Critical Race Theory

Critical Race Theory (CRT) was birthed out of Critical Legal Studies scholarship (CLS).

The purpose of CLS was to ardently refute claims that application of the law is race neutral 65

(Crenshaw, 2011). Prominent critical race scholars like Simson (2013) and Crenshaw (2011) urge current researchers to engage in scholarship using critical race theory that does not disconnect from America’s dark past that began with the systemic enslavement of Africans from the early 1600s to the late 1800s. In order to justify the treatment of enslaved Africans, the power structure of the racially dominant group, the legitimizers of slavery’s existence, had to provide a narrative to rationalize the maltreatment of African slaves (Hing, Chung-Yan, Hamilton &

Zanna, 2008; Simson, 2013). Inhumane treatment of Africans would evolve into the widely accepted phenomenon of racial dominance, also referred to as White supremacy (Crenshaw,

2011). Maltreatment of Blacks was justified by detaching human value from the lives of the

Black slaves. Detaching humanlike qualities from Blacks justified isolation and marginalization in American institutions. (Crenshaw, 2011; Hing, Chung-Yan, Hamilton & Zanna, 2008;

Simson, 2013).

After the Civil War, several movements led by Black leaders and some members of the racially dominant group sympathetic to their cause were compelled to help improve life for

Blacks and emancipated slaves (Simson, 2013). This period was also known as the Black

Reconstruction (Crenshaw, 2011; Simson, 2013). The Black Reconstruction was marked by economic and educational advancements (Anderson, 1988; W.E.B. Dubois, 1935; 1998).

Powerful members of the racially dominant group viewed the fast pace at which free Blacks and emancipated slaves were advancing as a threat to their interests, survival, and dominance

(Haghighat, 2001; Fitzgerald, 2009; Simson, 2013). Strategic laws were passed to deter and retract economic and educational advances of Blacks antebellum (Anderson, 1988; Simson,

2013). Proponents of critical race theory urged scholars to interrogate laws, jurisprudence, as 66

well as state and local policies enacted to impede economic and educational progress of Blacks that preserved racial hegemony (Anderson, 1988; Crenshaw, 2011; Harris, 2012; Simson, 2013).

In the beginning, the frameworks for Critical Race thought were incubated within Critical

Legal Studies (CLS) spaces in the late 1980s (Crenshaw, 2011; Harris, 2012). As a compartment within CLS, Critical Race tenets were formed using CLS’s challenge of illegitimate power which caused the racially dominant group to thrive at the expense of racial minorities (Crenshaw, 2011;

Davis, 2012). In the spirit of W.E.B. Dubois, who called racism America’s problem, and echoed by Shirley Chisolm, who is said to have proclaimed racism “is so engrained in the fabric of

America’s identity that it is natural and seamless” (Crenshaw, 2011), racism in America, is as natural as American patriotism and as invisible as the air (Bell, 2001; Steeh, 1995).

Critical Race Theory exposes the paradox in “colorblind” doctrine adopted by the Supreme

Court, the criminal justice systems, and education (Crenshaw, 2011; Davis, 2012; Ladson-

Billings & Tate, 1995). The “colorblind” doctrine is one of the symbols whereby hegemony and epistemologies are developed to perpetuate patterns of marginalization, stigmatization, and racism in American institutions (Crenshaw, 2011; Davis, 2012; Ladson-Billings & Tate, 1995;

Ledesma & Calderon, 2015; Simson, 2013; Solórzano & Yosso, 2001; Solórzano, 1997).

Critical Race Scholars like Bell (1992), Davis (2012), and Crenshaw (2011) were seminal actors in the development of the CRT movement. They, along with other early contributors attacked the jurisprudential traditions of the Supreme Court which were used to mask hegemony through (1) neutrality of law (Crenshaw, 2011; Davis, 2012; Ladson-Billings & Tate, 1995); (2) objectivity (Ladson-Billings & Tate, 1995); (3) colorblindness of the supreme court; and, (4) meritocracy as adopted legal doctrines. Critical race scholars believed that the ideal of 67

“colorblindness” was a strategy deployed in legal traditions intimating the constitution and subsequently the court should not see color. To focus on color as a means of interpreting the law and Constitution would be the epitome of racial bias (Davis, 2012, Crenshaw, 2011; Simson,

2013; Ladson-Billings & Tate, 1995).

The Supreme Court’s inability to be neutral was a major contention within Critical Race scholarship. CRT challenges the Supreme Court’s practice of deferring to state and local governments as well as other duplicitous actors whose roles advance discrimination and marginalization of persons of color, particularly Black people (Davis, 2012). Angela Davis described the actors associated with local governments as criminal prosecutors and juries who are known to favor biased policies against African Americans and other people of color (Simson,

2013; Davis, 2012). Also included in the Supreme Court’s deference doctrine is deference to

“privatized providers and free marketers tasked with resource allocation and [distribution]”

(Davis, 2012). Critical race scholars believed that the Supreme Court’s jurisprudence evoked illusory doctrinal standards of meritocracy when legal challenges of racial discrimination and bias are presented to expose discriminatory practices (Crenshaw, 2011).

Critical race theory purported that jurisprudential standards regard race as a moot point when considering merit and a person’s qualifications, irrespective of race (Crenshaw, 2011;

Davis, 2012). Scholars within Critical Race Theory, also known as RaceCrits, argued that meritocracy initiated a subjective system of moving targets and benchmarks that work against people of color (Crenshaw, 2011; Davis, 2012). Overall, CRT examines all legal actions through the lens of race. 68

Separation from Critical Legal Studies: No Room for Critical Race Theory

Both Critical Legal Studies and Critical Race Theory interrogated the legitimacy of power associated to applying the law. To interrogate the law is to question the legitimacy of racially dominant social power inherent in the interpretation of the law (Crenshaw, 2011. An example of one such interrogation was encapsulated in Brown v. Board of Education (1954; 55, which initiated the end of desegregation in the United States. The U.S. Supreme Court ruled that the long-held “separate but equal” doctrine carved into American psychology was fundamentally and morally wrong. The U.S. Supreme court struck down the practice of separate but equal schools, ruling that it was illegal.

Early founders of the Critical Race movement focused their attention on the illegitimacy of racial power (Crenshaw, 2011; Davis, 2012. Challenging racial power associated with men in the racially dominant group, manifested as social power in American institutions (Davis, 2012.

CRT’s challenge of White men as the source of power in racial hegemony caused the relationship between Critical Legal and Critical Race Theories to end because it was viewed as discursive discourse. Also, it created dissension within the CLS movement and was viewed as

“Mau-Maui-ing” (Crenshaw, 2011; Davis, 2012. Mau-Mauig was a term that described the open, unapologetic, unfiltered critique of racial power and White male hegemony. Like Blacks, the dominant racial group believed it to be “aggressive, intimidating, or harassing” (Crenshaw,

2011, p. 1291. The present study uses the tenets of critical race theory to challenge racial and social power in public education by examining educators’ perceptions. Critical race theorists, also known as RaceCrits, framed the critique of racial power around challenging White male hegemony (Crenshaw, 2011. Instead of challenging White male hegemony, the present study 69

challenges racial attitudes, assumptions, beliefs, biases, expectations, fantasies, perceptions, and stereotypes that educators hold regarding African American males (Fitzgerald, 2009; Foster,

1995; Hing et al., 2008; Quinn, 2017).

Critical Race Theory in Education

Public education provides a lens to extend Critical Race Theory’s critique of institutional and structural racism (Ladson-Billings & Tate, Ledesma & Calderon, 2015; 1995; Linter, 2004;

Lynn, 2002. African Americans comprise only 12% of the U.S. population (Ladson-Billings &

Tate, 1995; Simson, 2013; however, they are the majority in America’s largest urban school districts (Ladson-Billings & Tate, 1995. Relevant research suggests African American students perform significantly lower academically than students in the racially dominant group and other groups comprised of non-Black persons of color (Ladson-Billings & Tate; Simson, 2013. One reason African American students lag in academic performance is poverty in conjunction with the condition of schools, as wells as their academic experiences (Ladson-Billings & Tate, 1995, p. 55.

Davis (2012 described structural racism as implicitly biased actions which preserve economic, racial, and cultural hegemony in opposition to people of color. An example of preserving racial hegemony in public schools can reflect in decisions made by educators that perpetuate deficit discourses and stereotyping, such as referring African American males to special education for EBDs due to cultural misunderstanding (Oelrich, 2012. Simson (2013 and

Oelrich (2012 believed that deficit decision making preserves racial hegemony.

Racially biased practices may be reflected in low academic expectations, disproportionate eligibility in special education, increased suspension and expulsions, and low graduation rates for 70

Black and Brown students (Ladson-Billings & Tate, 1995; Noguera, 2012; Simson, 2013).

Patterns of structural racism in public schools include reifying practices and decision making reflective of past attitudes (Bell, 2008; Crenshaw, 2011). An example of decision-making that reflects past attitudes is the belief that African American students lack intelligence, resulting in teachers not exposing them to academically enriching opportunities (e.g., gifted programs) or offering access to advanced college preparatory courses (Fleming, 2005).

Organizations can change but remain committed to traditions of the past that result in negative experiences for racially minoritized groups. According to Burke (2009), structural remaking within organizations is called “autopoiesis.” Autopoiesis, a term used in biological sciences, describes the molecular regeneration of cells when other cells that define its identity die

(Capra, 1996). This complex, yet “deep ecological” process is marked by patterns of inseparable relationships (as cited in Burke, 2011, p. 39). Racism in American institutions, like education, are indicative of its preservation through transformation (Bell, 2008; Davis, 2012). Implicit biases are the molecular structures that preserve racial hierarchies in organizations (Hing et. al,

2008; Simson, 2013).

Institutional racism, like structural racism, involves complex relationships (Davis, 2012).

However, institutional racism is defined by systemic practices favoring members of the racially dominant group over African Americans and non-Black people of color (Simson, 2013). Irvine and Irvine (1983) reported that institutional racism is characterized by domination and control exerted upon African Americans and other non-Black people of color, which results in predetermined outcomes that cannot be prevented by external [controls]”(p. 416). 71

In the traditions of Critical Race Theory, scholars like Bell (2008), Crenshaw (2011), and

Ladson-Billings and Tate (1995), argued that Brown v. Board of Education failed to deliver the hope of equal access and basic civil rights in education for Black students. Instead, Bell (2008) and Ladson-Billings and Tate (1995) asserted that an unintended consequence of Brown v. Board of Education decision transformed segregation. Bell (2008) referred to Brown v. Board of

Education decision as a historical artifact without contemporary significance. Race and racism in the educational system is another example of racisms preservation through transformation (Bell,

2008). In 1962, Dr. Martin Luther King referred to the effects of desegregation vis-à-vis Brown v. Board of Education as “physical proximity without spiritual affinity” and “physically desegregated and spiritually segregated”, describing how the law of the highest court in the land could not abridge racism in education (Horsford, 2001).

Ladson-Billings and Tate (1995) and Bell (2008) both had significant criticisms regarding the Brown v. Board of Education decision. They concluded that urban schools comprising of mostly Black and Brown students are still segregated (Bell, 2008; Ladson-Billings

& Tate, 1995). While Solórzano and Yosso (2002) applauded the litigatory effort of Thurgood

Marshall and the Legal Defense team of the NAACP, they criticized the limitations of their work that omitted important racial nuances. According to Solórzano and Yosso (2002), critical elements of White privilege were not unearthed and disabled, which was the intent of desegregation (p. 161). Bell (1997), Horsford (2008) and Ladson-Billings and Tate (1995) proposed that integrating Black students into desegregated White schools did not solve the problem of equal access and better instruction. Decades after Brown v. Board of Education, scholars expressed remorse highlighting what students of color lost (Bell, 1997; Crenshaw, 2011; 72

Davis, 2012; Solórzano & Yosso, 2002). In fact, some argue that African American students lost more than they gained through desegregation. First, the ruling did not hold the racial majority responsible for the system that created and directly benefitted from educational inequality.

African American students lost highly educated Black teachers, who were denied employment in predominantly White schools. Additionally, the court’s decision did not demand equality in distribution of resources within and between schools. Finally, school districts did not hire Black administrators to reflect the racial composition of desegregated schools (Solórzano & Yosso,

2002). The Brown ruling denounced segregation, calling for integration; however, a similar

“pushing out” became synonymous with desegregation as evidenced by loss of Black educators, and schools. Similarly, African American students experienced exclusion by inequitable disciplinary practices that resulted in disproportionate suspensions and expulsions (Horsford,

2011; Simson, 2013; Solórzano & Yosso, 2002). The losses associated with desegregation had significant impact on Black students through the property function of Whiteness – the right to exclude. Horsford (2011) documented the historical measures of state and local governments to avoid desegregation.

Harris (1993) and Horsford (2011) provided an example of the right to exclude as a function (e.g., property) of Whiteness. Epistemologies, or ways of knowing and understanding, are accompanied by the right to determine value and legitimacy (Linter, 2004). Relevant research suggests that racially dominant epistemologies have never accepted African Americans as social equals (Crenshaw, 2011; Ladson-Billings & Tate, 1995; Simson, 2013). To avoid school integration, local school boards implemented legislation authorizing districts to close public schools and provide tuition vouchers to Whites only to attend State-supported all-White private 73

schools (Horsford, 2011. The Virginia General Assembly is an example to pictorialize the right to exclude as legislatures convened special sessions to pass legislation to illegally close schools and cut off funding for integrated schools. Additionally, the Virginia Assembly, attempted to authorize state funds to provide retirement benefits for White teachers at newly established all-

White private schools (Horsford, 2011.

The most telling sign of exclusion was epitomized in the Virginia Pupil Placement Act of

1964. In theory, students could transfer into White schools with the following exceptions: they could not affect (1 the orderly administration of the public schools, (2 the competent instruction of the pupils already enrolled, and (3 the health, safety, education, and general welfare of the pupils. Horsford (2011 concluded that it was impossible for Black students to attend the schools without violating at least one, if not all, of the indicated criteria. Horsford (2011 purported that the Virginia Assembly repealed compulsory attendance laws. School attendance was a matter of local choice and preference.

Intersection of Property and Race

Ladson-Billings and Tate (1995) proffered that institutional racism in education prevails because education is viewed as a property right are only belonging to the racial majority (p. 52.

They, along with Crenshaw (2011 and Davis (2012, furthered their claims that protection of property in America is reflected in legal doctrines (e.g., jurisprudence giving deference to capitalism as a fundamental right embedded in American democracy. Critical race scholars like

Bell (1996 and Ladson-Billings and Tate (1995 believed that American democracy and capitalism are in opposition to the other. Furthermore, citing Bell (1987, Ladson-Billings and

Tate proffered existing tension between property rights and human rights. The jurisprudence of 74

the Supreme Court protects the property of the rich. The slave status of African Americans in

America “objectified them as property” (p.53). The following are the implicit and explicit characteristics of property rights in American education proposed by Ladson-Billings & Tate

(1995): (1) real estate, (2) Whiteness, (3) reputation and status, (4) the right to or not to legitimize (e.g., determine value or importance), and (5) intellectual property reflected in the right to define (e.g., giving or interpreting meaning; pp. 52- 62).

Some laws (e.g. fugitive slave laws) enacted by Congress were designed to control racially minoritized groups, such as Native Americans, Japanese, and Chinese-Americans. These laws also defined racial minorities as property in varying forms (Ladson-Billings & Tate, 1995;

Guion, 1978; Simson, 2013). For example, real estate property determines the quality of schools.

Also, taxing real property to fund public education strikes a chord of resentment by some who believe funding public schools made up of most racially minoritized, economically disadvantaged students is an infringement upon the property rights of the racial majority.

Property rights of the rich work in favor of their children, entitling them to a superior education

(Ladson-Billings & Tate, 1995; Quinn, 2017).

As slaves, Black people were the property of Whites (Ladson-Billings & Tate, 1995;

Simson, 2013). Extant research suggested that racially dominant hegemony capitalizes on the history of African Americans which deemed their culture, norms, and social meanings as inferior. Ladson-Billings and Tate (1995), Simson (2013), and Solórzano (1997) suggested property rights associated with Whiteness are (1) preservation of self-interests, (2) power, and

(3) privilege. The benefits of membership in the racially dominant group is characterized by power to act upon or carry out behaviors indicative of racial superiority (Solórzano 1997). 75

Harris (1993; 2012) described how Whiteness functions as property. First, it can function as the right of disposition. Membership in the racial majority places its members at the top of social hierarchy. It also functions with the right to use and enjoy the privileges of being at the top of social strata. According to Harris (1983), the fundamental right to obtain reputation and status is afforded to members of the racial majority. It is bestowed at birth and remains a right or privilege used to disadvantage racially minoritized groups. Finally, critical race scholars, suggested that

Whiteness functions with the right to exclude (e.g., housing and job discrimination; Harris, 1993;

2012; Linter, 2004; Simson, 2013).

Research suggests that racially minoritized students who conform to “Whiteness” as a property, are rewarded when they performatively adapt (Ladson-Billings & Tate, 1995, p. 59) to socially acceptable norms. Non-White people of color might seek to be rewarded when they dress, speak, and internalize accepted racial and historical epistemologies (Ladson-Billings &

Tate, 1995; Linter, 2004; Simson, 2013). Whiteness has value, just as real estate has value

(Harris, 1993). Accordingly, Whiteness functions as property. A person with property has rights.

Schools in wealthier areas enjoy robust curricula that includes critical thinking, reasoning, and logic skills (p. 59). The opposite is true in economically disadvantaged areas where schools are heavily attended by Black, Hispanic, Latino, and other non-White people of color; these schools are plagued with low performing teachers and inadequate facilities (Ladson-Billings & Tate,

1995).

Both Ladson-Billings and Tate (1995) and Solórzano and Yosso (2002) implied that social capital afforded to White identity allows members of the racial majority to enjoy privileges such as racial, intellectual, and social legitimacy. Legitimacy is a benefit associated 76

with membership in the racially dominant group and a property of Whiteness (Harris, 1983;

Harris, 2012; Ladson-Billings & Tate, 1995; Solórzano & Yosso, 2002. Social capital that legitimizes is another benefit bestowed upon the racial majority (Solórzano & Yosso, 2002.

Legitimacy associated with racial dominance is conspicuous, whether it be shaping social programs, allocating government resources, performing various functions of leadership, providing stewardship of government funds, and educating. Also, membership in the racially dominant group, legitimizes defining and articulating narratives for public discourse on a variety of topics, including race (Crenshaw, 2011).

Properties of Whiteness and the Right to Exclude and the Push-out Phenomenon in Public

Education

Intellectual legitimacy as a property of Whiteness is affirmed in public schools, as well as colleges and universities in America (Baumgartner & Johnson-Bailey, 2008; Ladson-Billings &

Tate, 1995; Ledesma & Calderon, 2015; Linter, 2004; Quinn, 2017; Solórzano & Yosso, 2002.

In addition to affirmation and legitimacy, Ladson-Billings and Tate (1995 implied that property function of Whiteness extends to narrative discourses regarding the experiences of members of outgroups.

Critical race and the social construction of race in the context of school discipline theories (Simson, 2013 is the perfect complement to construct a theory regarding negative perceptions of African American males can precipitate a pushing out from mainstream school environments into isolation. An example of the “push out” phenomenon was the case of Isaac

Lora et al. vs. The New York City Board of Education (1975; 1978; 1980. Plaintiffs argued

Black and Hispanic children’s disproportionate membership in special education for severe 77

behavior or emotional impediments was evident in that they were assigned to special day schools in New York. Integrating critical race and Simson’s (2013) social construction of race in the context of school discipline theories provides the foundation to support the argument that measuring educator perceptions in a quantitative study paves the way to link perceptions to decision making.

The social construction of face model in the context of school discipline supports the theory that implicit biases activate in a way that flaws logical decision-making capacity resulting in stigmatization and marginalization (Simson, 2013). The social construction of race in school discipline as proposed by Simson, supports the argument that African American males, as the most marginalized of all racially minoritized groups (Wilson et al., , 2017) are systemically targeted for membership in specific areas of special education (e.g., EBDs). This practice begins the cycle of exclusion from society (Fitzgerald, 2009; Foster, 1995; Haghighat, 2001; Serafini et al., 2011; Simson, 2013; Wilson, Hugenberg & Rule, 2017).

Horsford (2011) described how rejection and isolation affected Black students integrating into White schools, but Lewis (2008) simplified it at the micro level as he postulated the failure of Black males began when White teachers refused to teach them post-Brown (Lewis, 2008).

The literature used to inform content for the present study described specific examples of rejection and isolation of Black students during desegregation. Rejection and isolation relate to current trends in public schools through suspensions, expulsions, and special education membership of African American students (Fitzgerald, 2009; Ladson-Billings & Tate, 1995;

Noguera, 2012; Simson, 2013). 78

Implicit Biases and African Americans in the Public Education System

Horsford (2011) provided a critical element to the idea that exclusion is a function of ownership reserved for members of the racial majority. Furthermore, she provides a narrative that is relative today for how African American males experience the “push out” phenomenon

(Ladson-Billings & Tate, 1995; Simson, 2013). If it can be established that African American males pose a threat to administering effective education in schools, compromise competent instruction, or are a threat to the health, safety, and general welfare, it further bolsters the theory that perceptions of African American males can be used to determine their fate. Solórzano &

Yosso (2002), interrogated illogical assertions that Black students were better off by having the opportunity to sit next to White students in school, resulting in a better education. Instead of a better education by integration, some of the losses for communities of color were: (1) the values of Whites were legitimized and valued at the expense of Black students; (2) Black students were culturally and educationally devalued; and (3) communities of color bore the burden of trying to

“change the nation’s approach to race” (p. 162).

Explicit biases that lead to rejection and isolation for African Americans are easier to identify than implicit examples that reflect attitudes which influence decisions. Implicit biases are not easily detectable because they materialize in ambiguous situations (e.g., deciding which students receive access to programs for gifted learners; Beckford, 2016; Fleming, 2005;

Hamilton, & Zanna, 2008; Hing, Chung-Yan, Simson, 2013).

Bias in Public Education and Curricula

According to Hing, et al. (2008), implicit biases automatically activate by negative associations with a member of a racially minoritized group. Both Wilson et al. (2017) and 79

Simson (2013) suggested that physical attributes of a member of a racially minoritized group may activate implicit biases (e.g. dress, dark skin, full lips, and wide nose). Simson (2013) further asserted implicit biases might create “leakage” activating instinctive behaviors to discriminate or marginalize. Accordingly, when instinctive behaviors manifest in educators, they influence educator decision-making (Simson, 2013, p. 545).

Implicit biases in school curricula are speculated to have harmful effects for African

Americans (Linter, 2004). Overt racism is not politically correct. Bigoted and harmful acts that were hallmarks of America’s dislike of Blacks and people of color (Hing, Chung-Yan, Hamilton,

& Zanna, 2008; Linter, 2004; Quinn, 2017) are no longer legal. Covert racist ideologies might pose a significant threat to success for racially minoritized groups (Dovidio et al., 2002). Hing et al. (2008) agreed with Dovidio et al. (2002) regarding the role prejudice profiles attached to certain ideological beliefs influenced decision making when there are situations of ambiguity that provides a cover for racist beliefs.

Educators act on personal beliefs and perceptions by the way they interpret the actions of

African Americans and other racial minorities (Lanier & Wittmer, 1977; Simson, 2013).

Interpreting actions of racial minorities originates from racial stereotypes that influence behavior towards racially minoritized groups, particularly Black males (Linter, 2004). Quinn (2017) and

Linter (2004) theorized that racial stereotypes result in psychological harm. When African

American students are exposed to negative racial attitudes and behaviors, their behavior may change to cope with the effect these negative experiences have on them psychologically (Simson,

2013). Dovidio et al. (2002) suggested implicit biases are not easy to see, but they may be easier to detect by members of racially minoritized groups. In fact, Dovidio et al. (2002) reported that 80

Blacks were more concerned when implicit biases were concealed by facades of niceness and feigned acceptance. Implicit attitudes and behaviors are covert, functioning as microaggressions, but overtly as macroaggressions (Hing, et al., 2008; Quinn, 2017; Solórzano & Yosso, 2002).

Microaggressions are implicit (Hing, et al., 2008; Quinn, 2017; Solórzano & Yosso, 2002).

Macroaggressions are overt (Hing, et al., 2008; Quinn, 2017; Solórzano & Yosso, 2002). Micro and macro aggressions will be linked to the right to exclude, as a property function of the racial majority, relating to school curricula (Ladson-Billings & Tate, 1995; Linter, 2004).

Textbooks and negative media depictions of African Americans and other non-White people of color are examples of implicit, isolating images which describe a micro aggressive

“push out” phenomenon in school curricula (Linter, 2004; Simson, 2013). Micro aggression in curricula is characterized by subtle insults, such as excluded or devalued perspectives in classroom discussions, textbooks, and instructional processes (Linter, 2004; Solórzano & Yosso,

2001). The “push out” phenomenon is also characterized by depictions and images which amplify insignificance or lack of belonging of African Americans (Simson, 2013). Referencing history as an academic discipline taught in schools, Linter (2004) suggested the teacher and the text are the factors used to designate significance of what is historically valuable. Furthermore, when teaching American history, personal biases and perceptions of teachers and textbook authors inform their views of non-White people of color, particularly African Americans (Linter,

2004).

Racial hegemony is confirmed through images of triumph and victory, which implicitly bolster the image of the racially dominant group. Conversely, repeated images of hardships, oppression, showing African Americans in perpetual subjugation communicates a message of 81

inferiority of African Americans in textbooks (Linter, 2004). The Social Construction of Race

Model in the Context of Educator Decision Making Race is a social construct (Crenshaw, 2011;

Davis, 2012; Simson, 2013). Furthermore, race differentiates people for the purposes of discriminating against them (Solórzano & Yosso, 2002). Historically, race in the United States has been socially constructed to serve as a basis for systemic discrimination” (Solórzano &

Yosso, 2002, p. 158). Figure 2.1 shows how raced-based decision-making filters through individual biases and results in negative outcomes for African American students in public schools (Simson, 2013).

Figure 2.1. Social construction of race in the context of school discipline. 82

Simson (2013) summarized race as a social construct but concluded that it is also legally constructed in which the legal system interprets laws along racial contexts. Simson (2013) proffered that the impact of social and legal constructions of race of males in the racial majority were not neutral. Instead, according to Simson (2013), race is a powerfully coercive and ideological tool to privilege the racial majority and subordinate people of color (p. 527).

Six Steps of the Social Construction of Race in the Context of School Discipline

Figure 2.1 explains each step in the social construct of race as a theoretical construct

(Simson, 2013). Step 1 provided context for race and how the racially dominant group determines it. Steps 2 and 3 described the sorting process used to categorize race. Simson (2013) believed some steps may overlap. Step 4 highlighted social meaning attached to race. More specifically, social meaning esteems superiority and dominance to being White (Harris, 1993;

Ladson-Billings & Tate, 1995; Simson, 2013). To be Black is to exist in perpetual inferiority and subjugation (Harris, 1993; Linter, 2004; Solórzano & Yosso). Step 5 described how social meaning determines racial experiences expressed through racial hierarchy. For example, a young

White male standing up for his rights expresses his assertiveness as a leader; He is encouraged and supported. Conversely, a young Black male standing up for his rights is viewed as angry and aggressive. His disposition is viewed as threatening and violent; he is feared and closely monitored or controlled (Fitzgerald, 2009; Simson, 2013).

The final step in the model, described in step 6, depicted how laws and policies work in tandem to secure the racial hierarchy wherein the racially dominant group maintains their status.

Step 6 also describes outcomes related to decision making. White students remain exclusive benefactors of liberty and freedom. However, with slavery as a pretext, Black students retain 83

their status as inferior and legally positioned for subjugation (Simson, 2013, pp. 529-532).

Perceptions of White educators are often distorted when comparing social behaviors of White students and Black A students (Beckford, 2016; Simson, 2013).

Simson (2013) extended the social construction of race to discipline practices against

African American students in schools. In the context of school discipline, Simson proffered that:

Differential perception and evaluation prompt disciplinary decision-makers to evaluate the behavior of students within the Social Construction framework, in turn giving social meaning in accordance to performative norms ascribed to racial categories. (p. 533)

According to Simson’s model, genealogical and phenotypical characteristics are the primary indicators for how race is determined. Race, according to Simson (2013) is the basis for how behavior of African American students is interpreted. Phenotypical characteristics determine how social meaning is ascribed to behavior. Interpreting behavior from social meaning is reflected through perceptions (Simson, 2013). Examples of phenotypical characteristics are dress, accent/language, demeanor and culture (Ladson-Billings & Tate, 1995;

Simson, 2013). Beckford (2016) referred to phenotypical characteristics described by Simson

(2013) as bioecological behaviors. She also included ways of walking and speaking to behavior that provides social meaning to behaviors of African American males. Black males are perceived negatively when race combined with phenotypical characteristics are used to inform social meaning (Simson, 2013).

The social construction of race in the context of school discipline is an example of race informing social meaning (see Figure 2.1). The model suggested to be Black is to be defiant, disrespectful, and dangerous; to have lower potential; to not a value to society; and to not truly 84

belong (Foster, 1995; Quinn, 2017; Simson, 2013). Solórzano and Yosso (2002) seemed to concur that social meaning of behaviors are associated with being Black. They believed that overtly derogatory terms have evolved to more subtle and private words to describe racial attitudes. It is no longer acceptable to refer to African American students as being “dumb, dirty, or lazy”; however, Solórzano and Yosso (2002) offered different terminologies used to express racially stereotypical attitudes. Examples of derogatory terminologies include words like, “lack hygiene” or “lack motivation” used to imply dirty, as well as dumb or lazy (p. 11).

Social Meaning of Race in School Discipline

The six steps of Simson’s model reveal three dimensions: (1) race, (2) social meaning, and (3) confirmation of social meaning in society. The first dimension describes how race, racial categories, and racial assignments are synonymous processes that activate perceptions of student behaviors derived from social meanings and racial categories. For example, Hing, et al. (2008) proffered that implicit bias is “automatically activated as negative associations with an outgroup

(p. 972). The degree to which biased perceptions and decisions are activated might vary by the intensity of individual racist ideologies (Hing, et al., 2008). The more racist ideologies an individual has may predict their likelihood of discriminating against a member of an outgroup

(Hing, et al., 2008). Based on Simson’s (2013) construct, it could be argued that racial ideologies of educators may predict how decisions are made for African American males. Once social meaning has been constructed, negative stereotypes, biases, and misperceptions influence decision making for public educators (Simson, 2013). Additionally, the more racist individual educator ideologies are, the more likely Black males will have adverse racial experiences in school (Hing, Chung-Yan & Hamilton, 2008; Quinn, 2017; Simson, 2013). 85

The second dimension indicates that the way an educator applies social meaning to student behavior may provide a strong indication of how African American males will be treated in school and subsequently in society (Beckford, 2016). Steeh (1995) in her review of Sniderman and Piazza’s, “the scar of race”, suggested that America’s thinking about race is not likely to change. Haghighat (2001) and Solórzano and Yosso (1997) seemed to find agreement that justification for the ill treatment of African Americans and other marginalized outgroups are reified through media’s perpetuation of popular stereotypes of Black men.

Educator decision-making in interpreting the behaviors of young Black males is believed to be the primary reason, they are suspended, expelled, or inducted in the criminal justice system

(Simson, 2013). Simson (2013) raised several points regarding decision making of public educators. First, he suggested that educators evaluate behaviors of Black and White students differently, resulting in disproportionality in discipline against African Americans. Additionally,

Simson (2013) asserted that implicit racial biases are most likely to affect decision making when the situation is ambiguous, thus providing a way to justify it on non-racial grounds (p. 545).

Hing, et al. (2008) concluded that ambiguous situations allow individuals who do not want to be exposed as racist to act upon their racist sensibilities by a phenomenon known as the attributional-ambiguity effect (p. 977). Additionally, Simson (2013) proposed that America’s long history of stigmatization and stereotyping is more likely to create an excuse for educators justify exclusion from society through biased decision making. Finally, decision making based on racial bias is likely to result in behavioral leakage (Simson, 2013). Behavioral leakage occurs when an environment becomes hostile for the person who is the target stigmatization. Leakage is 86

ideal when there is a confluence of racial bias from negative stereotypes and marginalization that functions simultaneously resulting in reactive behaviors from the target (Simson, 2013).

Stigmatization and Biases

Stigmatization is an accumulation of negative attributes imputed upon an individual or group (Simson, 2013). Also, stigma transcends the true identity of an individual or group

(Simson, 2013, p. 535). In relation to the present study, stigma that is associated with African

Americans in this country is derived from their historical status (Ladson-Billings & Tate, 1995;

Simson, 2013). Furthermore, stigmatizing African Americans is the preferred way to justify their maltreatment (Haghighat, 2001; Hing, et al., 2008). Psychological factors associated with stigmatization is not limited to justifying maltreatment. Justifying maltreatment allays the guilt associated with stigmatizing (Hing, et al., 2008). For example, when justifying the abuse, rape, lynching, and other terrors of slavery, Simson (2008) and Ladson-Billings and Tate (1995) proffered that the lack of guilt of oppressors was the result of dehumanizing Blacks. In doing so,

Ladson-Billings and Tate (1995) inferred that stigmatizing Blacks gave credence to their inferiority and bolstered the ideal of racial dominance. Stigmatization reinforces firmly held arguments that depict African American males as inferior, dangerous, violent, sexually deprived, intellectually bereft, and prone to criminality (Fitzgerald, 2009; Foster, 1995; Haghighat, 2001;

Nickerson, 1988; Quinn, 2017; Simson, 2013; Wilson, et al., 2017).

Theory of Stigmatization

The unitary theory of stigmatization is an embedded theory within Simson’s (2013) social construction of race in the context of school discipline. It is also supported by Capra’s (1996) deep ecology of firmly held beliefs that operationalize against outgroups to preserve the interests 87

of the racial majority, who benefit from social dominance. Nickerson (1988) believed that stigmatization is a psychological process. However, Haghighat (2001) deemed stigmatization to be a biological process that transmits genetically (p. 213). Haghighat (2001) defined stigmatization as rejecting and isolating behaviors. Furthermore, according to Haghighat, stigmatization is one aspect of genetic behavioral programming that fights for survival. Simply put, he implied that isolating and rejecting the stigmatized is a natural evolutionary process”

(Haghighat, 2001, p. 213). An example of the unitary theory of stigmatization embedded in the

Social construction of race in the context of school discipline explains how behavioral leakage activated by racial biases creates an environment that is marginalizing for African American males (Haghighat, 2001; Nickerson, 1988; Simson, 2013).

Factors of Stigmatization (Haghighat, 2001)

Figure 2.2 shows the origins of stigmatization, which includes constitutional factors of stigmatization comprised of economic, psychological, and evolutionary factors. Stigmatization is not one dimensional. Stigmatization is bilateral and multidimensional (Haghighat, 2001).

Stigmatization consists of self-interests and stigmatizing attitudes. Examining educator perceptions will aid the researcher in understanding what is gained (e.g., self-interests) and the process of how stigmatization presents to students (e.g., attitudes). Factors of stigmatization consist of the ways in which the brain interferes with “proper social perception” and

“information processing” (Haghighat, 2001). The brain gives preference to negative evaluations at the expense of positive ones. When overloaded with data, the brain attempts to operate efficiently treating repeated examples of behavior from an individual to the entire outgroup

(Haghighat, 2001; Nickerson, 1988; Wilson, Hugenberg & Rule, 2017). The outgroup is any 88

group outside one’s own racial identity (Wilson, Hugenberg, & Rule, 2017). Wilson, et al (2017) also used the term outgroup to describe any group not a part of the dominant racial group.

Figure 2.2. How self-interests influence attitudes related to stigmatization.

Constitutional factors associated with stigmatization work to the disadvantage of any outgroup, particularly racial minorities (Foster, 1995; Garver & Noguera, 2012; Kunjufu, 1990;

2003; Simson, 2013; Swerdlick & Fullilove, 1991; Terry, 2007). Haghighat (2001) concluded that constitutional factors serve to preserve self-interests and likely to lead to inappropriate judgments (p. 207).

A single group enjoying liberty, freedom, and privilege at the expense of an oppressed group that experiences rejection and isolation is an example of stigmatization (Harris, 1983; 89

Ladson-Billings & Tate, 1995; Haghighat, 2001; 2005; Simson, 2013). Stigmatizers gain emotionally, psychologically, and financially (Haghighat, 2001). Constitutional structures of stigmatization are also known as biological structures describing how the brain genetically inherits, interprets, and operationalizes the process of rejecting and isolating behaviors toward the stigmatized (Haghighat, 2001; 2005). Stigmatizing attitudes, vis-à-vis self-preservation, denies the rights of others at the: (a) affective level, (b) cognitive level, and (c) discriminatory level (Haghighat, 2001).

Levels of Stigmatization

Anger, fear, and resentment are characteristic of the affective level of stigmatization

(Haghighat, 2001). Quinn (2017) believed that at the affective level motivate racial attitudes.

The cognitive level of stigmatization relates to the effect emotions have on an individual that cause them to reject and isolate a target. Simson (2013) alluded to stigmatization at the cognitive level in step 5 of the SCRMCSD. He called them racial experiences. An example of cognitive rejection that leads to isolation is the perception that young Black males are prone to violence and criminality. Racial experiences are created by stigmatizers to confirm their beliefs

(Fitzgerald, 2009; Foster, 1995; Nickerson, 1988; Simson, 2013; Quinn, 2017; Wilson,

Hugenburg & Rule, 2017). Stigmatization at the discriminatory level starts with feelings (e.g., affective level), followed by the brain’s disposition to process information in order to avoid a threat (e.g., cognitive), ending with refusal, denial of access, or failure to make accommodations for a target (Haghighat, 2001; Ladson-Billings & Tate, 1995; Linter, 2004; Simson, 2013).

Stigmatization can work simultaneously in favor of the racially dominant group (Haghighat,

2001; Simson, 2013). Summarily, stigmatization results in isolation and rejection in multiple 90

institutions of society for young African American males (e.g., the work place and higher education; Beckford, 2017; Garver & Noguera, 2010; Simson, 2013; Swerdlick & Fullilove,

1991; Terry, 2007).

Stigmatization emanates from an individual’s pursuit to preserve self-interests

(Haghighat, 2001, p. 208). Preservation of self-interests as seen in figure 2.2 links Simson’s

(2013) the social construction of race in the context of school discipline to stigmatization because it maps out how race defines social meaning and interpretation of behaviors of educators define predictive outcomes for Black students. Figure 2.2 outlines the structures of stigmatization. Constitutional or biological factors are the basis for stigmatizing behaviors that preserves self-interests. Haghighat (2001) believed that preserving self-interests are motivated by economic, psychological, and evolutionary interests. He believed that internal processes result in stigmatizing attitudes.

It may appear that too much emphasis is placed on specific examples of stigmatization at the economic origin; however, my argument asserts that African American males might be the most marginalized group in America’s public-school system (Fitzgerald, 2009; Simson, 2013;

Foster, 1995). Recalling Step 6 of Simson’s Social Construction of Race in the Context of

School Discipline, social meaning describes the forecast for how Black males will be treated in society (e.g., incarceration). Simson’s theory and espousal to the social meaning of race in school discipline is a powerful tool for school educators to determine the outcome of racial experiences in school (2013). Foner (1990) confirms the power of social meaning. Over 2000 Blacks were murdered in 1865 in or around Shreveport, Louisiana. In 1866, after a dispute with a local freedman, a Black settlement near Pine Bluff, Arkansas was set on fire. After which, the bodies 91

of 24 men, women and children were found hanging to trees around the settlement (Foner, 1990, p. 53). Context for Simson’s social meaning and Harris’ (1993) race and property rights. Foner

(1990) encapsulates the meaning of the social meaning and race as property in the following excerpt:

Pervasiveness of violence reflected [whites’ determination to define their own meaning

of freedom] and to resist black efforts to establish their own, whether in matters of

family, church, labor, or personal demeanor. (p. 53)

Social meaning was applied to behaviors of Black men between 1865-1868. Behaviors not deemed to be socially acceptable were met with violence. For example, a Black man was viewed as “insolent” or “insubordinate” for (1) claiming the right to discipline their child instead of allowing his employer and former master to do so; (2) bowing to a white man, saying, “good morning”; because a Black man was not permitted to speak to a White man unless spoken to”

(Foner, 1990). Foner (1990) supported the argument of social meanings applied to behaviors.

Stigmatizers gain advantage by stigmatizing of others (Haghighat, 2001; Simson, 2013;

Wark & Galliher, 2007). The original forms of stigmatization are: (1) economic, (2) psychological, and (3) evolutionary (Haghighat, 2001). The various forms of self-interest according to Haghighat (2001) “intermingle, overlap, work together, and interrelate” (p. 289).

Stigmatizers compete for economic dominance. An example of economic interests focuses on the competition for resources (e.g., control and allocation; Haghighat, 2001). Evolutionary self- interests are relevant when environmental, political, or social conditions change (Crenshaw,

2011; Davis, 2012; Harris, 1993; Haghighat, 2001; Hing, et al., 2008). Stigmatizers need to evolve to maintain the superior position, which may be realized through institutional policies and 92

legislation in favor of their positions (Crenshaw, 2011; Davis, 2012; Harris, 1993). Having a need to be remain in the dominant position, stigmatizers, will create environments that deprive the stigmatized from thriving (e.g. economically, educationally, and socially; Ladson-Billings &

Tate, 1995; Solórzano & Yosso, 2002). All facets of stigmatization work in tandem to effect decision-making to perpetuate accepted norms in favor of hegemonic values in school and society (Crenshaw, 2011; Davis, 2012; Simson, 2013).

Implicit and Explicit Attitudes that Characterize Prejudice

Simson (2013) and Haghighat (2001) attribute biases, stereotypes, negative perceptions and behaviors to stigmatization. Simson (2013) implied that implicit biases are a key factor in stigmatization influencing educator decision-making. Hing et al. (2008) examined four profiles related to prejudice: (1) truly low prejudice (TLP), (2) aversive racist (AR), (3) principled conservatives (PC), and (4) modern racists (MR). Researchers Hing et al. (2008) conducted a study using a two-dimensional model of prejudice, which included measures for explicit and implicit attitudes to create four prejudice profiles. The profiles were used to predict a persons’ inclination to stigmatize and discriminate against a target (Hing, et al., 2008). Hing, et al. (2008) predicted differences amongst prejudice profiles ranging from least to most likely to discriminate. Hing et al. (2008) framed prejudice profiles along three ideological endorsements: egalitarianism, social conservatism, and economic/political conservatism. Explicit and implicit biases are factors which comprise four profiles related to prejudice. 93

Truly Low Prejudice

Truly low prejudice (TLP) describes a person with double lows (e.g., low implicit and low explicit biases; Hing, Chung-Yan, Hamilton, & Zenna, 2008). In studies 1 through 3, results reported TLP people did not hold prejudiced ideologies and did not discriminate when situations of high ambiguity were present, which could have justified disqualifying a job applicant. When considered alongside low implicit and explicit biases, TLP people are higher in egalitarianism/humanism, but low in social and economic/political conservatism (Hing, et al.,

2008). Simply put, truly low prejudice people have double-lows; low implicit and explicit biases.

Aversive Racists

Aversive racists differ from truly low prejudice people. Aversive racists self-report as low in prejudice (Dovidio, Gaertner, Kawakami, & Hodson, 2002; Hing, et al., 2008). Aversive racists also are said to have some evidence of egalitarianism as expressed in deliberate efforts to moderate their observable behaviors which might be viewed as racist (Dovidio, Gaertner,

Kawakami, & Hodson, 2002). Aversive racists hold liberal political ideologies (Dovidio,

Gaertner, Kawakami, & Hodson, 2002, p. 972). Quinn (2017) validated the notion of people how hold liberal ideologies when understanding racial attitudes of postsecondary educators. Dovidio, et al. (2002) alluded to aversive racists are likely to hold liberal ideologies. Findings reported by

Quinn (2017) showed some racial attitudes of postsecondary educators were cause for concern

(e.g., racists should be allowed to teach college). Also, troubling attitudes, ascribed to aversive racists are that they believe that first amendment protections from hate speech should be 94

protected above racial minorities’ right to exist in a learning environment free of discrimination

(Quinn, 2017).

Aversive racists consider themselves to be non-prejudiced because they subscribe to egalitarian values (Dovidio, Gaertner, Kawakami, & Hodson, 2002; Hing, et al., 2008; Quinn,

2017); however, because they have negative implicit feelings toward racial minority groups, they were predicted to score higher in implicit biases, but lower in explicit biases. When comparing aversive racists and their ideological endorsements, researchers reported that they tended to score higher in egalitarianism/humanism, and lower in social conservatism as well as economic/political conservatism (p. 976). Albeit, researchers reported that aversive racists are higher in egalitarianism/humanism, but not as high when compared to truly low prejudice people

(Hing, et al., 2008).

Principled Conservatives

Implicit biases are automatic. They are activated by negative associations with an outgroup (Hing, et al., 2008). Explicit biases are consciously held, negative evaluations of an outgroup (pp. 971 – 972). Principled conservative ideologies are compatible with low implicit biases related to egalitarianism/humanism and social conservatism, but high explicit biases related to values for economic/political conservatism. Findings by Hing, et al. (2008) are consistent with Quinn (2017) who found that postsecondary educators showed liberal sensibilities but failed to support programs that promote equality for all people (e.g., government programs that level the playing field for minority outgroup members). 95

Modern Racists

Hing, et al. derived the term modern racism (MR) from the Modern Racism Scale (MRS; see Kinder & Sears, 1981; McConahay, 1986), they believed levels of egalitarianism/humanism were fundamentally linked to prejudice (p. 972). In this study researchers assessed egalitarianism, which is described by fairness and consideration of all people as a value to humankind, by reversing the Social Dominance Orientation (Hing, Chung-Yan, Hamilton, &

Zanna, 2008).

Ideological Endorsements

Hing et al. (2008) concluded individuals with certain prejudice profiles were more likely to discriminate against a target. Furthermore, they concluded that participants with certain political ideologies are likely to hide their prejudice out of fear of being exposed as a racist.

Alternatively, they believed that racists will act on prejudice or racist beliefs in a situation where they can justify their actions in an ambiguous situation. Table 2.7 summarizes three ideological endorsements used to derived prejudice profiles outlined in the study by Hing, et al. (2008).

Table 2.7

Ideological Endorsements (Hing, Chung-Yan, Hamilton, & Zanna (2008)

Egalitarian/Humanism Social Conservatism Economic/Political Conservatism supports equality, values conventionalism, endorse race neutral policies, do not recognizes worth, feels traditionalism, desire for support the government support of empathy for all groups social control programs that level the playing

Egalitarianism/humanism is characterized by individuals who are supportive of equality, recognize worth of all, expressing empathy, and are able to connect with all individuals in racial 96

minority groups (Hing, et al., 2008). People who are truly low prejudiced are most likely to identify with egalitarian and humanistic views (Hing, et al., 2008). Social conservatism, according to the authors of this study, is characterized by conventionalism, traditionalism, and desire for social control. In a social context, social conservatives support adherence to conventional American values (Hing, et al., 2008; Piazza, 1988), deference to authority, and aggressive acts toward social outcasts by authorities (Hing, et al., 2008). Economic/political conservatives are more likely to oppose government sponsored equal opportunity programs

(Hing, et al., 2008; Quinn, 2017). Individuals who are more prejudiced and strongly endorse stereotypes are often associated with economic and political conservatives are of particular interest to the present study (see Quinn, 2017). Modern racists are characterized as a blend of intense racial dislike and abstract racial values. Table 2.8 lists the scales that Hing et al. (2008) used to derive prejudice profiles. The present study will utilize prejudice profiles to differentiate participants who score high on adapted brief social desirability scale to determine if prejudice profiles can provide additional information about participants not reflected in the adapted brief social desirability scale. Chapter 4 will provide more detail how prejudice profiles are incorporated in the instrument used to examine educator perceptions. 97

Table 2.8

Scales Used to Define Prejudice Profiles (Hing, Chung-Yan, Hamilton, & Zanna, 2008)

Prejudice Scale Purpose Attributes Profile

16-item social dominance assessed desire for reverse scoring; low scores orientation (SDO) scale; hierarchy; predicts are indicative of Truly low Pratto, Sidanius, prejudice egalitarianism Prejudice Stallworth, & Malle, 1994

implicit associations test tested automatic computer-based assessment; Principled (IAT), Greenwald, associations for categories given at the top conservative McGhee, & Scwartz, 1998 various racial right and left, words sorted by categories participants; 7-minute completions time.

30 – item right-wing assessed 8-point Likert Scale Aversive authoritarianism (RWA) conventional values; racist/social scale; Altmeyer, 1996 aggression towards conservative social deviants, threats to moral order

modern racism scale measures racial scenario based assessment to Modern (MRS), Hing et al., 2002 antipathy, views make hiring decisions racist Blacks as threat to justice, order, conservatism

Hing et al. (2008) connected ideological endorsements to prejudice profiles to describe how perceptions relate to racism. Prejudice profiles are used in this present study to defend the 98

argument that racial attitudes against African American males play out in decision making. Hing et al. (2008) concluded that individuals within certain prejudice profiles are more likely to negatively evaluate or refuse to hire an Asian target when attributional ambiguity is high, leading to discrimination. Accordingly, Hing, et al. (2008) concluded that certain prejudice profiles (e.g. aversive racists) will discriminate without violating their self-image as nonprejudiced because of the desire to appear to be good (Haghighat, 2001; Hing, et al., 2008; Van de Mortel, T. F., 2008).

Conversely, others falling within differing prejudice-related ideologies did not discriminate when attributional ambiguity was high. Attributional ambiguity uses non-race-related justifications for giving negative responses or when appropriate action is unclear (Hing et al., 2008). Furthermore, aversive and modern racists care about their image and do not desire to be exposed as racist; therefore, they would only act upon their biases when attributional ambiguity can preserve their image (Hing, et al., 2008).Explicit and implicit biases have weak associations, representing independent dimensions of racism (e.g. see Gawronski, Gschwendner, Le, & Schmitt, 2005;

Rudman, 2004; Hing et al., 2008).. In four studies explicit and implicit biases were loaded as separate factors. Explicit bias is consciously held negative evaluation of an outgroup retrieved from memory and can be self-reported (Hing, Chung-Yan, Hamilton, & Zenna, 2008, pp. 971 –

972). Implicit bias is prejudice is automatically activated from negative association with an outgroup (Hing, Chung-Yan, Hamilton, & Zenna, 2008). Interestingly, the historical understanding of prejudice has been assessed as an explicit, consciously held negative evaluation of an outgroup (Hing, et al., 2008; Wilson, Hugenberg, & Rule, 2017; Simson, 2013). However,

Simson (2013) referenced implicit bias as the key factor motivating stigmatization of African

American students when educators make disciplinary decisions (See Figure 2.1, p. 85 ). Implicit 99

bias, according to the social construction of race in the context of school discipline is simultaneously activated from steps 1 through 3; identification, categorization, and assignment of race, to steps 4 through 6; social meaning, racial experiences, and decision-making based on race (Simson, 2013, p. 532). Figure 2.3 shows where each prejudice profile falls in relationship to the high and low dichotomies of implicit and explicit bias (Hing, et al., 2008).

Figure 2.3. Relationships between prejudice profiles and implicit/explicit biases. 100

Hing, et al. (2008) connected ideological endorsements to prejudice profiles to describe how perceptions relate to racism. Prejudice profiles are used in this present study to defend the argument that racial attitudes against African American males play out in decision making. Hing et al. (2008) concluded that individuals within certain prejudice profiles are more likely to negatively evaluate or refuse to hire an Asian target when attributional ambiguity is high, leading to discrimination. Accordingly, Hing, et al. (2008) concluded that certain prejudice profiles (e.g. aversive racists) will discriminate without violating their self-image as nonprejudiced because of the desire to appear to be good (Haghighat, 2001; Hing, et al., 2008; Van de Mortel, T. F., 2008).

Conversely, others falling within differing prejudice-related ideologies did not discriminate when attributional ambiguity was high.

According to Hing, et al. (2008), explicit and implicit biases have weak associations, representing independent dimensions of racism (e.g., see Gawronski, Gschwendner, Le, &

Schmitt, 2005; Rudman, 2004). In four studies explicit and implicit biases were loaded as separate factors. Explicit bias is consciously held negative evaluation of an outgroup retrieved from memory and can be self-reported (Hing, Chung-Yan, Hamilton, & Zenna, 2008, pp. 971 –

972). Implicit bias is prejudice is automatically activated from negative association with an outgroup (Hing, Chung-Yan, Hamilton, & Zenna, 2008). Interestingly, the historical understanding of prejudice has been assessed as an explicit, consciously held negative evaluation of an outgroup (Hing, et al., 2008; Simson, 2013; Wilson, Hugenberg, & Rule, 2017). However,

Simson (2013) referenced implicit bias as the key factor motivating stigmatization of African

American students when educators make disciplinary decisions (See Figure 2.1). Implicit bias, according to the social construction of race in the context of school discipline is simultaneously 101

activated from steps 1 through 3; identification, categorization, and assignment of race, to steps 4 through 6; social meaning, racial experiences, and decision-making based on race (Simson, 2013, p. 532). Figure 2.3 shows where each prejudice profile falls in relationship to the high and low dichotomies of implicit and explicit bias (Hing et al, 2008).

Hing et al. (2008) tested a two-dimensional model to examine if truly low prejudiced

(TLP) people, aversive racists (AR), principled conservatives (PC), and modern (MR) racists differed in their political ideologies. Four groups were created using the median scores from the measures administered in their study (see Table 2.9). Groups were separated by median scores

(e.g., high and low scores). Participant scores below the median were grouped in the low category, while scores above the median were designated to be high. Table 2.9 lists each study conducted by Hing et al. (2008), which show how low/high implicit biases impact decision making. 102

Table 2.9

Summaries of Studies Regarding Explicit and Implicit Biases that Characterize Prejudice (Hing et al., 2008)

Purpose Sample Findings

Study 1 Tested two-dimensional n = 1623 Modern racists were lower in egalitarianism/humanism model by examining if truly n = 132 than were Princip principled conservatives. Aversive low prejudiced people, racists tended to be lower in egalitarianism/humanism aversive racists, principled than were truly low prejudiced people. Modern racists conservatives, and modern were lower in egalitarianism/humanism than were racists differ in their aversive racists. Principled conservatives were lower in sociopolitical ideologies egalitarianism/humanism than were truly low prejudiced done in 2 phases. people.

(1) participants in the no-excuse condition had higher implicit prejudice scores than those in the excuse Study Explored the differences n = 305 condition. (2) participants rated the target’s intellectual 2 between principled n = 69 skills higher than his social skills. (3) job requiring conservatives and modern stronger social skills is believed to provide a subtle, racists in situations with non-race-related excuse to negatively evaluate the target high attributional ambiguity done in 2 phases. (1) Asian condition participants correctly identified the target as Asian. (2) White condition correctly identified Study tested two-dimensional n =184 the target as Caucasian (3) Asian candidate was 3 model by probing the n = 146 significantly less likely to be recommended for hire by discriminatory behavior of aversive racists in the excuse condition (4) aversive truly low prejudiced racists were less likely to recommend the Asian participants and aversive candidate for hire in the excuse condition than their racists with improvement to counterparts in the no-excuse condition (5) aversive study design by including a racists demonstrated attributional ambiguity effect (6) target ethnicity as a factor aversive racists in the excuse condition were less likely to recommend the Asian candidate for hire compared to their counterparts who evaluated the White candidate (7) when a candidate is not clearly qualified for a position, aversive racists demonstrated ethnic discrimination

Study 2/3 (1) combined studies 2 and n = 139 (1) participants higher in implicit prejudice gave lower 3 to explore the effects of hiring recommendations to the Asian candidate (2) no- attributional ambiguity for excuse condition had no effect of implicit prejudice (3) participants with a full range in the no excuse condition, participants lower in implicit of explicit modern racism prejudice was not significant (4) in the excuse 103

scores (2) conducted participants higher in implicit prejudice gave lower supplementary analyses in hiring recommendations than did those in the no-excuse which participants assigned condition (5) modern racists evaluated the Asian job to the Asian condition in candidates more negatively than did aversive racists. (6) Studies 2 and 3 were modern racists gave lower hiring recommendations to combined to test by factorial the Asian candidate for the human relations position design (3) the independent than did aversive racists and interactive effects associated with excuse condition, explicit modern racism, and implicit prejudice

Testing the Two-Dimensional Model of Prejudice: Social Conservatism and

Political/Economic Conservatism

The study conducted by Hing, et al. (2008 found that there were theoretical differences between modern and aversive racists. In study 1, results indicated that modern racists are more conservative socially and economically/politically than were aversive racists. Additionally, findings were consistent with their hypothesis; aversive racists were more egalitarian/humanist than were modern racists. When compared to truly low prejudice people, aversive racists subscribe to prejudice-related ideologies more than TLP people.

Additionally, Study 1 concluded the following: (1 there were similarities and differences in each prejudice profile (e.g., social conservatism was related to prejudice depending on the social context, (2 modern racists endorsed prejudice-related ideologies more strongly than principled conservatives, (3 aversive racists showed higher values related to egalitarianism/humanism than modern racists did, (4 modern racists, had higher values in social conservatism than aversive racists, and (5 aversive racists were lower in economic/political conservatism because they are theorized to be liberal, whereas modern racists are thought to be economically and politically conservative (Hing, et al., 2008, p. 973. 104

Hing, et al. (2008) investigated the effect of attributional ambiguity on evaluations using a fictitious Asian job candidate (e.g., an Asian sounding name like Chad Wong) in study 2. They tested and identified the nature of their racial attitudes by manipulating the non-race-related excuse to negatively evaluate the target using subtly constructed ambiguous job qualifications using two test conditions: (1) no excuse versus excuse, and (2) principled conservatives versus modern racists. Hing, et al (2008) predicted that not all participants would give lower hiring recommendations in the excuse versus no-excuse condition. They concluded, that participants who, according to the two-dimensional model were categorized as modern racists were likely to not recommend hiring when a non-race-related excuse was present. Hing, et al. (2008) also supported that attributional-ambiguity effect aligned with the assertions proffered by Simson

(2013) as described in his model that showed decision making for educators is biased by racial attitudes. Both Simson (2013) and Hing, et al. (2008) provided strong evidence that modern racism could operationalize in the context of school discipline and educator decision making. In light of this discovery, prejudice related ideologies will be incorporated in the measure to examine educator perceptions of African American males.

In study 3, Hing, et al. found that in low ambiguity situations, modern racists rated targets the same as principled conservatives using interview summaries in front of them. More specifically, modern racists could not justify discrimination when there was clear evidence that would expose them as racist (Hing, et al., 2008). This suggests that modern racists work hard to maintain a nonracist self-image or appearance. Similar to study 2, Hing, et al. (2008) used study

3 to evaluate 23 modern racists and 21 principled conservatives using a White target named Gary

Walsh for the same position as the Asian candidate from Study 2. Results indicated that modern 105

racists gave significantly higher recommendations for hiring the White target (M=4.50,

SD=1.13) than did modern racists in study 2, M=4.00), t(22)SD=2.13, p>.045. In this same study, encouraging results showed that principled conservatives gave equivalent hiring recommendations for the White target (M=4.86, SD=1.13) and the Asian target in Study 2 (M =

4.88), t(20) = 0.07, p=.94). The findings suggested that modern racists will always discriminate when given a situation of high attributional ambiguity (Hing, et al., 2008). Conversely, principled conservatives are significantly less likely to discriminate against an Asian job candidate in favor of a White job candidate. In support of examining educators’ perceptions of

Black males, educators with certain ideological leanings, like modern racists, pose a significant threat to Black males. Developing an instrument to truly understand educator perceptions of

African American males is all-the-more relevant to current bodies of research.

McConahay (1986) supported Hing et al (2008) regarding the usefulness of the attributional-ambiguity effect theory to describe modern racists desire to conceal prejudice from others. Results from Hing et al. (2008) provided further evidence that modern racists were the only ones to distinguish the target’s suitability for the two positions. This according to Hing, et al. (2008) might provide further proof that motivation to find an excuse to negatively appraise the target was activated early (e.g., Stage one of Simson’s social construction of race model).

Researchers who study bias, prejudice, and racism could pose an argument that attributional ambiguity realized in school discipline policies and laws for special education provides a safe space for modern racists to discriminate against Black males, given the high ambiguity in criteria for qualifications and eligibility (Beckford, 2016; Foster, 1995; Oelrich,

2012; Simson, 2013). 106

One drawback of the study conducted by Hing et al. (2008) is that it did not consider other racial minorities (e.g., African Americans) because stereotypes, prejudice, and discrimination differ between racial minority groups (Dovidio, Gaertner, Kawakami, & Hodson,

2002; Hing, et al., 2008; Quinn, 2017; Wilson, Hugenberg, & Rule, 2018), Generalizing the findings reported by Hing et al. (2008) for the present study is important because African

American males are considered the most disliked of all racialized minority groups (Fitzgerald,

2009; Foster, 1995; Wilson, Hugenberg, & Rule, 2017). Hing, et al., (2008) purported that people who are prejudiced toward Asians, who are considered to be the ideal minority group, might be more prejudiced toward other groups (e.g. African American males) when considering explicit and implicit measures. Summarily, Hing’s, et al. (2008) findings indicated that varying forms of prejudice when correlated to political ideologies intervenes in decision-making. The present study will link prejudice and political ideologies to educator decision making based on implicit and explicit biases.

Overall results of the study conducted by Hing et al. (2008) concluded that modern racists endorsed prejudiced related ideologies most strongly. Additionally, researchers concluded that modern racists were conservative and most likely to demonstrate attributional-ambiguity effect. Principled conservatives, on the other hand, were economically/politically conservative, but did not endorse prejudice-related ideologies and did not discriminate against Asian targets when there was an excuse to do so. Aversive racists, however, were low in conservatism, but displayed ethnic discrimination only when they had an excuse. Finally, truly low prejudiced people were most egalitarian, least socially conservative, and not likely to discriminate. 107

The large sample size and use of participant data and procedures across studies increases my confidence in Hing’s, et al. (2008) overall findings, correlating them to results found by

Quinn (2017), which concluded that PreK -12 and postsecondary educators held more negative racial attitudes than noneducators. Fitzgerald (2009), Foster (1995), and Quinn (2017) all provided strong evidence to support the findings of Hing, et al. (2008), which suggested that prejudice profiles of public educators might be able to explain decision making (e.g., attributional ambiguity effect) toward African American males in public schools when racial bias is a factor. Hing et al. (2008) disclosed criticisms of their study by providing an alternate narrative that explained perceived weaknesses related to median cut off scores that led to unintentional dichotomies between high/low relationships for implicit and explicit bias variables.

For example, explicit and implicit prejudice were reported as continuous variables, and according to Hing, et al. (2008), some might object to categorizing participants as lows or highs.

In this chapter, the literature was used to evaluate the impact of racial attitudes of PreK – 12 educators by making a loose coupling to decision making (Fitzgerald, 2009; Foster, 1995; Hing et al., 2008; Quinn, 2017; Simson, 1995; Wieck, 1976). Furthermore, the argument is raised that perceptions, when related to implicit and explicit biases may predict discrimination when there are conditions with high attributional ambiguity (Hing et al., 2008). Conversely, the argument was challenged because implicit biases may not be a good predictor of discrimination where there is low-attributional ambiguity. Hing, et al. (2008) concluded that their model revealed higher explicit bias from modern racists might not lead to discriminatory behavior using low implicit prejudice as a mediating factor. However, aversive racisms’ duplicitous nature makes it potentially more harmful than modern racism. Dovidio, et al. (2002) suggested that aversive 108

racists posed a greater threat to positive interracial actions because of their implicit behaviors, which are not easily controlled. Examples of negative implicit behaviors are excessive blinking, denoting uneasiness. Avoiding eye contact indicates dislike for members of racial minorities and indicates rejection or lack of respect (Dovidio, et al., 2002; Haghighat, 2001; Simson, 2013).

Critical race theory and the social construction of race in the context of school discipline supported the notions that stigmatization and marginalization of African American males are related to behavioral and academic efficacy in public schools (Crenshaw, 2011; Davis, 2012,

Dovidio, et al., 2002; Simson, 2013). Dovidio, et al. (2002) reported that task oriented performance is affected by the presence of certain prejudice profiles (e.g. aversive racists).

Findings from Foster (1995) and Quinn (2017) indirectly support the findings of Dovidio, et al.,

(2002). PreK-12 and postsecondary educators have more negative racial attitudes than noneducators relating to sexual proclivities of Black males (Foster, 1995), desiring to live in close proximity with African Americans, as well as belief that African Americans’ lack the motivation to gain social, economic, and political equality (Quinn, 2017). 109

CHAPTER 3. LITERATURE REVIEW FOR INSTRUMENT DEVELOPMENT

Perceptions of young African American males in the United States are negative. The extent to which negative perceptions of young Black men influences specific outcomes in public schools is not known. Extant research indicates that negative perceptions of young African

American males are a phenomenon whose consequences extend far beyond school (Beckford,

2016; Fitzgerald, 2009; Foster 1995; Kunjufu, 1985; Oelrich, 2012; Simson, 2013. A dearth in quantitative research to examine the impact of public educator perceptions of African American males poses a unique challenge in social science research because there are no instruments available to measure such perceptions. This chapter will review the literature for developing a quantitative instrument that is valid and reliable. According to Benson and Clark (1983, there are four phases in instrument development: Creswell (2010 simplified Benson and Clark's

(1983 four-phase model into four stages that included specific steps for researchers to observe to develop a quantitative measure.

Process for Searching for Key Terms Related to Instrument Development

Searching key terms for instrument development and educator perceptions was done through Google Scholar, EBSCO Research Databases, and Summons power search of library resources. Whenever possible, I was able to acquire electronic or hard copy versions of literature for the present study. After reviewing research articles for instrument development and educator perceptions, common statistical tests that were referenced in studies, resulted in examining specific methods procedures using texts from doctoral coursework, and other related materials.

Recurring terms from the review of literature for instrument development include, but not limited to perceptions, validity examination, content, construct, and criterion validity, factor analysis, 110

test construction, instrument design, item analyses, analysis of variance (ANOVA, data screening, piloting, cross-validation, expert panels, and collinearity.

Four Phases of Instrument Development

Creswell (2010) simplified Benson and Clark's (1983) four-phase model into four stages when developing a quantitative measure. First, the researcher must identify the purpose of the instrument (Atwater, Lau, Bass, Avolio, Camobreco, & Whitmore 1994; Castillo, March,

Stockslager, & Hines, 2016; Creswell, 2010; Haghighat, 2001; 2005; Kane, 2013. Second, the literature review informs the content of a proposed measure (Castillo, March, Stockslager, &

Hines, 2016; Creswell, 2010. Next, an effective instrument should contain well-written items

(Creswell, 2010. Finally, testing items with individuals that reflect the population intended in the proposed study is essential to developing an instrument that reflects evidence of validity and reliability (Castillo, March, Stockslager, & Hines, 2016; Creswell, 2010.

At the onset of the process of instrument development, extant research suggests that the researcher engages in a process that allows them to think about the intended use of a proposed instrument as a part of the planning phase. Lavery, Jong, Krupa, and Bostic (proposed book

Chapter, 2018 include seven questions that assist the researcher in examining evidence to support interpretation and use of scores. Three questions in particular serve the interests of ongoing processes of examination of validity: (1 what construct is measured by the instrument,

(2 how the construct is being measured, (3 how and by whom are scores interpreted and used

(Lavery et al., 2018. Considering the aforementioned questions, the emphasis is on examining the interpretation and use argument over the instrument itself and will be answered in the piloting phases of instrument development (AERA et al., 2014. 111

Table 3.1 shows a list of studies used in this chapter to explore instrument development.

Castillo et al. (2016) and Pohan and Aguilar (2001) developed instruments that are related to the

present study that include educator perceptions. Moreover, Pohan and Agular’s (2001) study

examined psychometric properties of the instrument they developed to examine educator’s

personal and professional perceptions of multicultural beliefs. Haghighat (2001; 2007)

developed two measures (1) a brief social desirability scale, and (2) a measure to assess

stigmatization regarding patients with schizophrenia. The studies referenced in this chapter will

guide the present study to develop an instrument that has evidence of validity for reliability and

addresses threats to validity due to bias.

Table 3.1

Summary of Instrument Development Studies Presented in Chapter 3

Author(s) Purpose of the instrument Instrument Research Design

Castillo, 50-items skills and March, Measuring educators’ Quantitative data Stockslager, perceptions of skills relating to Correlational Research implementation & Hines, response to intervention RtI Design survey 2016 Haghighat, Development of a brief social 4-item brief social Quantitative Survey 2001 desirability scale desirability scales Research Design Mixed Methods - Haghighat, Development of an instrument 20-item Exploratory Sequential 2007 to measure stigmatization of Design patients with schizophrenia 22-item personal Pohan & Developing two scales to beliefs scale & 30- Experimental Research Aguilar, measure educators’ personal and item professional Design 2001 professional beliefs of diversity beliefs scale 112

The four phases of instrument construction are (1) planning, (2) construction, (3) quantitative evaluation, and (4) validation. In the planning phase, the researcher states the purpose of the test and identifies the target group (Benson & Clark, 1983; Castillo et al., 2016;

Creswell, 2010; Haghighat, 2001, 2005, 2007; Pohan & Aguilar, 2001). An instrument must state a clearly defined purpose that includes the intended population sample. Furthermore, the instrument should contain relevant content needed to clearly define variables (e.g., public school educators’ perceptions of African American males in PreK -12; Atwater, Lau, Bass, Avolio,

Camobreco, & Whitmore 1994; Castillo, March, Stockslager, & Hines, 2016; Creswell, 2010).

The review of literature serves several purposes. First, the literature provides information for content related to the study and helps the researcher clearly define variables of interest (Atwater et al., 1994; Castillo et al., 2016; Creswell, 2010; Pohan & Aguilar, 2001). Additionally, the literature review provides information to evaluate the effectiveness of existing measures for use in a research study (Castillo et al., 2016; Creswell, 2010; Pohan & Aguilar, 2001). Creswell

(2010; 2012) provided guidelines for researchers when evaluating existing measures. He proposed that an existing measure can used in a study once necessary permissions are obtained, or it can be modified to fit the scope of a study. When review of existing measures does not yield one that is adequate for a study, a researcher has the option of developing one (see Castillo et al.,

2016; Creswell, 2010; Haghighat, 2001, 2005, 2007; Pohan & Aguilar, 2001).

Once the purpose of the instrument has been clearly defined, items are constructed to develop the instrument. Developing items for a proposed measure is the next step following the literature review that provides information for content. Once items are constructed, the measure is disseminated to a sample group to examine evidence of validity for content and construct 113

(Creswell, 2010). During this process, the researcher solicits comments using open-ended items within the measure to determine the quality of items (e.g., well written with clear objectives, and efficient selection of item format (Creswell, 2010; 2012).

The four phases of instrument development are not fixed. The researcher remains attentive to processes related to refining the measure. Phase two of instrument construction includes examining content (Benson & Clark, 1983; Creswell, 2010, 2012; Castillo et al., 2016).

This phase includes selecting judges to evaluate the items for representativeness (Castillo et al.,

2016; Creswell, 2010). Developing and revising items for a proposed measure follows the examination by the expert panel (Castillo, March, Stockslager, & Hines, 2016; Creswell, 2010;

Haghighat, 2001). An important part of instrument development is preparing the instrument for piloting. According to Benson and Clark (1983), phase three includes quantitative evaluation of test items along with: (1) administration of the first pilot test; (2) debriefing subjects; (3) calculating validity evidence based on reliability; (4) running item analyses; (5) revising the instrument; and, (6) making revisions of instrument to prepare for the subsequent pilots.

In phase three and four of instrument development, quantitative methods are used to determine validity evidence based on reliability (Benson & Clark, 1983; Castillo et al., 2016;

Creswell, 2010). Phases three and four are concurrent, cyclical steps using a variety of statistical methods to explore evidence of validity (Creswell, 2010). Examining evidence of the validity of an instrument includes (1) completing item analyses after administration of second piloting of the instrument; (2) repeating revisions, pilot administrations, and item analyses; and, (3) beginning the iterative process for validation (p. 161). 114

Phase One of Instrument Development

Identifying the purpose of the instrument and reviewing literature is the primary focus of phase one in instrument development (Creswell, 2010; 212). After reviewing the literature, examination of existing instruments for use in a quantitative study is the next step (Creswell,

2010; 2012). A researcher who deems available instruments inadequate for their study may modify an existing instrument or develop an instrument for a proposed study. For example,

Castillo et al. (2016), Haghighat (2001; 2005), and Pohan and Aguilar (2001), developed instruments because existing measures were not representative of the topic of study, not represented in content gathered from literature reviews, and did not report psychometric properties of validation.

Evaluating Existing Measures

Pohan and Aguilar (2001) believed that existing measures lacked a comprehensive definition of multicultural diversity, thus did not coalesce with their purpose in examining personal and professional beliefs of multicultural diversity among educators. They summarized empirical studies to examine the adequacy of existing measures. Pohan and Aguilar (2001) found that existing measures were lacking in (1) a defined scope of multicultural diversity; (2) examination of validity evidence for reliability; and, (3) reported data regarding examining the evidence for validity, which included reliability testing. For example, in one study examined by

Pohan and Aguilar (2001), preservice teacher perceptions of academic achievement among diverse students were investigated (e.g., race, social class). They also examined a study regarding stereotypical attitudes held by educators; however, it did not include information regarding examining validity. Therefore, Pohan and Aguilar (2001), deemed its use impractical. 115

Pohan and Aguilar’s definition for multiculturalism was not represented in a single measure. Extant research indicates that studies with broad definitions relating to the topic of study may have limitations when applying results to specific demographics (Atwater et al.,

1994). However, Pohan and Aguilar (2001) needed broaden the definition of diversity to reflect current sociocultural dimensions of multiculturalism (e.g., sexual orientation). Their review of literature found definitions of multiculturalism too constricting, and not reflective of inclusiveness of other marginalized groups (e.g., English language learners). Moreover, Pohan and Aguilar (2001) believed that historically marginalized groups could not be limited to race and ethnicity. Table 3.2 lists an overview of existing measures examined by Pohan and Aguilar

(2001) to measure personal and professional beliefs of cultural diversity. Additionally, Table 3.2 gives a brief overview of each measure, the targeted group, number of items, and the limitations of each measure as determined by Pohan and Aguilar (2001).

Table 3.2

Summary of Instruments Appraised by Pohan and Aguilar (2001)

Target Group Topic Length Limitations did not report data for evidence of Preservice teachers diversity attitudes 18 items validity cultural diversity the narrow scope of diversity; no Teachers 28 items awareness scale data teacher-student did not report data for psychometric Teachers 47 items relationships properties of the instrument attitudes toward First-year university no information on scales and cultural diversity 81 items students subscales using **CDAI stereotypic attitudes no data reported related to Teachers 44 items racial minorities examination of validity Note: ** Cultural Diversity Awareness Inventory 116

Each measure, except one, examined teacher attitudes, with items ranging from 18 – 81.

Common limitations for existing measures were the limited definition of multiculturalism and lack of evidence of validation examination (e.g., limited data reporting psychometric properties), which Pohan and Aguilar determined were not adequate for use in their study. The present study, wherein no adequate tool exists to measure educator perceptions of African American males, will result in developing an instrument to measure perceptions that focuses on evidence of validity related to content.

Haghighat (2005) derived content for the instrument he developed to measure stigmatization of patients with mental illness (e.g., schizophrenia). Content for the first draft of his measure came from more than 120 patients' accounts from groups in which the author acted as a therapist in four hospitals in London. Haghighat derived additional content from the ideas of

358 patients, relatives, and social workers during community assessments conducted in homes, hostels, prisons, police stations, streets, day centers, and homeless shelters. Finally, Haghighat

(2005) used a collection of views from patients and relatives with whom the author worked in

Birmingham, Tehran, London, and Paris were used to obtain content for the measure analyze stigmatization of patients. For the present study, the proposed measure to examine educator perceptions of African American males derived its content from five categories (e.g., criminality, education, attitudes, personality, sexuality; Foster, 1995).

Defining Content for an Instrument

I used Pohan and Aguilar (2001) as an example for expanding the definition of perceptions. For example, Pohan and Aguilar (2001) expanded the definition of sociocultural diversity to include characteristics that were broad and inclusive. Key terms used to broaden the 117

definition included: (1) race, (2) ethnicity and culture, (3) social class, (4) gender, (5) sexual orientation, (6) exceptionality (e.g., disabilities and other impairments), (7) language diversity

(e.g., non-English speakers), and (8) religion. The broadened definition of multicultural diversity informed content considered for the measure to investigate personal and professional beliefs about diversity. The present study requires an expanded definition of perceptions beyond racial bias. Consequently, I expanded the definition of perceptions to include implicit and explicit biases (Hing et al. 2008), stereotypes, beliefs, fantasies, and perceptions (Foster, 1995), and racial attitudes (Wilson, 2017).

Castillo, March, Stockslager, & Hines (2016) measured educators’ perceptions and skills related to response to intervention and implementation. They found that important information for their study was not centralized in one body of content. Castillo, March, Stockslager, & Hines

(2016) developed content for the Perceptions of Response to Intervention Skills Survey (PRSS), by expanding their content parameters to include five broad areas: (a) data-based decision making, (b) tiered service delivery, (c) the problem-solving process in RtI, (d) data collection, and (e) data analysis and technology use (p. 96) The proposed use of the PRSS instrument served a dual purpose: (1) to develop a research tool to analyze educators’ perceptions (Guion, 1978;

Kane, 2013) of their individual skills (see Guion, 1978); and (2) a tool to formally evaluate skill development of elementary educators (Castillo, et al. 2016).

Haghighat (2005) assessed stigmatization relating to patients with Schizophrenia.

Stigmatization is likely to worsen the prognosis of mental illness. Haghighat (2005) specifically looked at stigmatization relating to schizophrenia. He conducted a literature review by examining existing measures. Haghighat (2005) observed that existing measures often (1) 118

confirmed misconceptions, and therefore perpetuated stigmatization; and (2) lacked a social desirability scale or the recommendation to use one. Perceptions of Black males are negative

(Fitzgerald, 2009; Foster, 1995; Jackson & Crawley, 2003; Wilson et al., 2017). It may appear that it is contradistinctive use Haghighat’s concern for reifying misconceptions. The preset study will take into consideration how questions are worded so as not to “lead” participants to answer in a particular way. The intended measure used to examine educator perceptions of African

American males will include an adapted brief social desirable scale to control for measurement error. Haghighat acquired content for standardized stigmatization questionnaire, version 1

(SSQ1). He acquired 2000 possible items regarding the experiences of patients and information from patients’ relatives and social workers who provided support for patients with Schizophrenia and related mental disorders. The studies referenced in phase one of the instrument development process substantiated recommendations to examine validity for content that aligns with Creswell

(2010) and Guion (1978). This is the planning stage of instrument development, which lead to phase two, the construction phase.

Phase Two: Planning and Construction of a Proposed Measure

Validity evidence for instrument content requires gathering information in the planning stage. Once content has been established that is directly linked to a clearly stated purpose, constructing items for an instrument come next. Items that are included for a measure should represent the universal acceptability of all possible outcomes relating to the proposed topic

(Creswell, 2010). Sufficient evidence of validity describes how well scores on an instrument relate to reported outcomes. Additionally, evidence of validity that satisfies inquiries related to the meaning and significance of scores is reflective of construct validity (Creswell, 2010; Kane, 119

2013). Evidence of construct validity is when scores are examined for their satisfactory representation of the topic and purpose of the instrument (Creswell, 2010). Phase two involves constructing items used to measure something specific. The present study measures educator perceptions of African American males. Most importantly, phase two focuses on writing well written items, testing the items, and making necessary changes to increase reliability in scores

(Creswell, 2010; 2012)

Test Construction

Content for a measure has a clear definition of what is to be measured (Atwater et al.,

1994). Questions should reflect the relevance and representativeness of the topic (Castillo et al.,

2016; Creswell, 2012; Dillman, Smyth, & Christian, 2009; Pohan & Aguilar, 2001).

Additionally, responses to test items should fit the question (Creswell, 2012). Questions should be precise and efficiently worded in order to prevent a type of reactivity called bias of unequal proportions (Dillman et al., 2009). Biases of unequal comparisons pertain to the wording of items that creates ambiguity in how a question is interpreted. According to Dillman et al. (2009) eliminating neutral response categories and providing efficiently worded details can mitigate this type of bias (p. 129).

Creswell (2012) cautioned researchers to pay attention to the quality of the questions.

Effective questions, according to Creswell (2012) effectuates an adequate understanding of the question, which participants to provide meaningful answers (p. 387). Additionally, Creswell

(2012) proffered that well-written items are: (1) clear and unambiguous, (2) do not confuse the participants, and (3) show respect for the participant by using non-biased language (e.g., sensitive to gender, class, and cultural needs). For example, Pohan and Aguilar (2001) modified 120

items that showed respect for gender sensitivity and response neutrality by modifying terms like

"mankind" to "humankind." Creswell summarized that the benefit of having effective items encourages participants to complete the instrument fully. The present study uses cognitive interviews to check for clarity and conciseness of the adapted brief social desirability scale.

Additionally, interviewing expert panel members regarding the content contained in the

PEPAAMS PK12 addresses the desire to have clear and concise language that will not yield any unpredictable (e.g, special cases; AERA et al., 2014).

Instrument Design

Creswell (2010; 2012) provided characteristics of a quality instrument design. First, there are different types of items: (1) personal, (2) attitudinal, (3) behavioral, (4) sensitive, and (5) closed -and open-ended (Creswell, 2012, p. 385). Next, instrument design includes strategies for constructing effective items as discussed in the preceding paragraph. The last step in instrument design is piloting test items. Piloting involves administering the instrument using a small sample size similar to the population for the proposed research study. Changes are made to the instrument once participant feedback is evaluated from the pilot (Creswell 2010; 2012).

Types of Items

The public educator perceptions of African American males survey (PEPAAMS) PreK

-12 will consist of each type of question in varying degrees. According to Creswell (2012), personal items, also known as demographic items are targeted to get information about participants (e.g., age, teaching experience, education). Attitudinal or opinion items appraise participants’ feelings about a topic. A question asking participants to rate their agreement regarding negative perceptions of African American males is one example of an attitudinal 121

question (Creswell, 2012) Attitude items can be administered using a Likert type scale (e.g., strongly agree, agree, disagree, strongly disagree; Creswell, 2010; 2012). Behavioral items reflect present or past actions of participants (e.g., as a classroom teacher having input on student placements).

Sensitive items, according to Creswell (2012) address topics that require careful wording.

For example, Haghighat (2001; 2007) addressed topic sensitivity regarding stigmatization of patients with schizophrenia by constructing items after gathering specific information from patients with schizophrenia, caregivers, and relatives of patients. Open-ended items express the intent of an instrument to extrapolate information from participants by not providing choices.

Items that include preset responses are examples of closed-ended responses. An example of a closed-ended response is the brief social desirability scale created by Haghighat (2001). His brief social desirability scale required participants to respond to preset questions with yes or no answers.

Developing Quality Test Items

Creswell (2012) stressed the importance of developing high quality test items. Vague and imprecise items should be reworded to be precise, using concise language. Pohan and Aguilar

(2001) and Castillo, et al. (2016) used simplified questions for the instrument they created during examining validity evidence. They combined multiple items into one question to eliminate redundancy. Negatively worded items and items that contain jargon may detract from the quality of test items. One example of negatively worded items are items that use phrases like “should not,” and “pro-life.” Creswell (2010; 2012) recommends rewording items that remove negative connotations. 122

The use of jargon or phrases that are not familiar to everyone could lead to misinterpretation. For example, the proposed measure developed in present study includes content from Foster’s (1995) study of educator, noneducator perceptions of Black males. In the sexual prowess category, Foster reported findings that Black men are perceived to “play around,” which is an example of wording that may lead to confusion or misinterpretation. The term was eliminated because its meaning is not consisted in a culturally relevant context.

Overlapping responses (e.g., items that do not contain distinct options), and unbalanced response items, where responses do not match the intent of the question are examples of unbalanced response. Additionally, questions that ask for the participant to rate importance of something should have consistent and logical wording reflected in answer choices. For example, a responses choice on a Likert type scale may start with “strong agreement” and end with “little importance.” Creswell (2012) and Dillman, et al. (2009) assert that question imbalances could result in measurement error. Measurement error from question construction may also occur when there is a mismatch or lack of congruence between the items and the answers. Creswell (2012) cites the use of overly technical language might confuse or isolate respondents that lack knowledge of a particular topic. Simplifying language increases question efficiency. Finally,

Creswell (2012) addresses items that may not apply to all participants, also known as

“branching” or “contingency” items (e.g., if so, then scenarios; p. 390).

Pilot Testing in Instrument Development

The purpose of piloting an instrument is to determine whether participants can easily complete it with directions that are easy to understand. Piloting also allows the researcher to gain feedback from sample participants regarding item clarity and conciseness (Creswell, 2010; 123

2012). During the piloting phase, the researcher uses feedback to make the instrument more effective. Changes to the instrument may occur after examining feedback from a small sample of participants in the pilot (p. 390). Written feedback from the initial piloting of an instrument is integral to instrument development. According to Creswell (2012), piloting data is not included in the final sample for the study (p. 390).

The present study will use Pohan and Aguilar (2001) and Haghighat (2005) as models for the public educator perceptions of African American males’ survey (PEPAAMS) PK12. Pohan and Aguilar (2001) conducted initial pilot testing to examine item clarity, scale reliabilities, and to address any procedural issues in administering their personal and professional beliefs of diversity survey. Specifically, they used data for two samples in the initial piloting to focus on completion time and clarity of directions.

Haghighat (2005) tested the 20-item questionnaire SSQ1 with 30 users recruited in a random draw to comment on the clarity of the questionnaire. Minor changes were made by the author in the wording to make the language clearer and more straightforward. Summarily, in pilot or field-testing stage, participants might offer comments for any problems encountered with a questionnaire, such as (1) poorly worded items, (2) responses that do not make sense, or (3) provide recommendations if the instrument takes an excessive amount of time to complete.

Phase Three: Continuing Validity Examination

Construct validity is essential for any instrument intended to measure something specific.

Kane (2013) indicated that the relationship between the measure and interpretation of the results derived from it are fundamental characteristics of construct validity. Creswell (2012) indicated that validity evidence based on construct examines the inferences made about the variables 124

within a study. Evidence of validity for construct will reflect internal consistency of scores. This research study examines PreK – 12 educators’ perceptions of African American males.

Types of Validity Evidence for Reliability

Creswell (2010) outlines five plausible procedures used to examine validity evidence for reliability of an instrument. The five procedures used to establish validity evidence for reliability are (1) test-retest, (2) alternate forms reliability, (3) alternate forms and test-retest reliability, (4) inter-rater reliability, and (5) internal consistency reliability (p. 163). Administering a measure twice at two different time periods, using one version of the instrument is called the test-retest method to examine validity evidence. At this stage, participants complete the instrument twice.

Alternate forms validity evidence for reliability requires that each participant complete each version of an instrument. The measures are administered once using two different versions of the instrument with same concept or variable. Alternate forms and test-retest validity is a combination of test-retest validity evidence for reliability and alternate forms validity. In this stage each measure is administered twice, at two different intervals, using two different versions of the same instrument.

The PEPAAMS PK12 has two measures requiring examination of internal validity evidence: (1) the adapted brief social desirability scale (ABSDS), and (2) the actual scales used to measure educator perceptions of African American males. The ABSDS will employ qualitative methods to examine validity evidence to evaluate response or cognitive processes. To complete cognitive interviews, researchers administer a questionnaire in order to understand participant experiences or thinking processes. Data from cognitive interviews reveal if participants responded to stimuli the same or in similar fashion (Creswell, 2012). Cognitive 125

interviews (CIs) play a critical role in substantive aspects of construct validity. For example, understanding how educators might respond on an instrument adapted from another study can be bolstered by using CIs to understand how respondents share interpretations of social desirability and perceptions of African American males (Castillo-Diaz & Padilla, 2013). Examining validity evidence for perceptions scales will utilize the test-retest process for internal reliability (AERA et al., 2014; Creswell, 2012; Lavery et al., 2019).

Factor Analysis

Confirmatory factor analysis is a statistical procedure that explains overlap in test items

(Mertler & Vannatta, 2013). In short, factor analysis describes the underlying structure that explains a set of variables. Running statistical procedures that examines item overlap is important to eliminant collinearity wherein multiple items are measuring the same thing.

Perceptions of African American males are negative, but the degree to which negative characteristics overlap is important to monitor. Castillo et al. (2016) used a process called factor loading in which a Pearson correlation coefficient was used to determine the degree to which items overlapped within a specified variable. The Pearson correlation coefficient can range from

-1.00 (a perfect negative correlation with a factor) to +1.00 (a perfect positive association), which could indicate item redundancy or measuring the same thing, therefore, likely to increase variance due to overlap (Castillo, et al., 2016; Mertler & Vannatta, 2013). Testing educator perceptions of African Americans males using an overall perceptions score is not a concern.

Notwithstanding, when examining specific perceptions like personality and attitude, if the 11 items that comprise both categories show evidence of overlap, those areas must be examined further to distinguish the nature of overlap between the two variables. 126

Castillo, et al. (2016; PRSS) used confirmatory factor analyses determine if the instrument they developed supported a correlated, three-factor structure of perceptions of RtI skills. They examined internal structures by extracting factors using the principal factor extraction method that included Eigenvalues and a scree plot. Castillo et al. (2016) made decisions for factor retention using this method. When evidence examination indicated overlap, they used oblique rotation methods to interpret the nature or degree of overlap (p. 97). They concluded that high internal consistency estimates provided evidence of the three latent constructs related to perceptions of skills applied to academic content (p. 99). However, using the principal extraction method that included examining factors using Eigenvalues, scree plot, and oblique rotation, Castillo et al. (2016) reduced items to eliminate redundancies (p. 97).

Castillo et al. (2016).

Construct Validity in Instrument Development

The process of establishing construct validity includes: (1) predicting a defined criterion

(e.g., scale scores), (2) the criterion reasonably reflects its intended purpose, and (3) articulating a logical rationale for decisions made when evaluating variables of study (Kane, 2013. More specifically, scores will be meaningful if they are related to other variables (e.g., perceptions of education completion and learning abilities related to African American males; Plano-Clark &

Creswell, 2015). If scores match researcher expectations related to educators’ perceptions of academic ability and high school completion, this would be one example of construct validity of the PEPAAMS PK12. Researching topics related to attitudes or beliefs regarding race and gender requires additional measures to ensure validity (e.g., condensed SD Scale, Haghighat, 2001,

2007; Pohan & Aguilar, 2001). Kane (2013) proposed that construct validity can be attained with 127

multiple kinds of evidence as well as integrating different types of evidence. Table 3.3 lists provides a list of variables and statistical methods used when examining evidence of validity for a quantitative measure.

Table 3.3

Statistical Tests Used in Instrument Development

Type of Test Purpose Variables Reference

Exploratory Factor extraction to test independent and Castillo, et al, 2016; Field, factor analysis for redundancies using dependent 2009; Pohan & Aguilar, Eigenvalues, oblique variables 2001 factor rotation, and scree plot Analysis of Testing assumptions, independent and Castillo, et al, 2016; Field, Variance variance, independence, dependent 2009; Pohan & Aguilar, ANOVA normality variables 2001 t-Test for Determining significant dependent Pohan & Aguilar, 2001, independent differences between variables Field, 2009 samples groups, comparing means, testing hypotheses Cross Validation Uses a regression model independent Field, 2009; Haghighat, (e.g., Chi Square, Fisher’s variable 2007 exact test) which tests score probability and predicts outcomes of scores

Exploratory factor analyses allowed Castillo et al (2016) to examine internal structures by extracting factors using the principal factor extraction method, which uses Eigenvalues and scree plot to make informed decisions about which factors to retain. Also included in the process used to explore factors, Castillo et al (2016) used oblique rotation methods for factor interpretation (p.

97). They determined that high internal consistency estimates provided evidence for the three 128

latent constructs related to perceptions of skills applied to academic content, which was their intent (p. 99).

Eigenvalues and a scree plot are used to identify the number of factors considered for retention. The eigenvalue is a term derived from the criterion developed in 1960 by Kaiser (cited in Mertler & Reinhart, 2013). Eigenvalue informs how factors researchers retain factors in an instrument. Eigenvalue is the total amount of variance explained by each factor in relation to the total amount of variability in factor analyses. For example, the PEPAAMS PK12 has six sub- categories, with a total of 27 items that comprise overall perceptions of African American males.

If all 27 items were plotted on a graph, any factors to the left of the sharp descent would be retained because they have high eigenvalues to extract to examine the underlying structures behind them. Ford, Gambino, Lee, Mayo and Ferguson (2004) discovered a two-factor structure when they considered six characteristics that explained a large portion of the total variance in their examination of manager's pre-interview bias against African American applicants. Finally,

Creswell (2012) indicated that eigenvalues are used to consider retention or rejection of items.

Oblique rotation uses the tilting of the x-axis to expose any underlying factors that are not discovered by Eigenvalues. When performing oblique rotations, a structure matrix of correlations between factors and variables are plotted. Additionally, a pattern matrix display unique relationships using plots with no overlap among factors and variables (Mertler &

Vannatta, 2013). If used, oblique rotation allows the researcher to explore the structure of an instrument in order to make decisions with greater confidence before removing or retaining factors. If the present reveals possible overlap of factors, oblique rotation will be used if

Eigenvalues are high in more than two variables. 129

The scree plot is a graphical method for determining the number of factors considered for retention. It shows magnitude of each eigenvalue (Mertler & Vannatta, 2013). When observing the scree plot, the magnitude of each eigenvalue is plotted along the vertical axis against their ordinal numbers on the horizontal axis. Mertler and Vannatta (2013) suggest that all factors after the bend, but before the leveling off could remain. Plotted points after the bend indicates that the eigenvalues are relatively low, thus likely equal in size. Castillo et al. (2016) examined internal consistency as validity evidence to interpret results from factor loadings using Oblique rotation.

Mertler and Vannatta (2013) cautioned that making decisions by interpreting results from oblique rotation is highly subjective. Castillo et al. (2016) addressed decision subjectivity by reporting that reducing items and eliminating redundancies were due in part to stringent factor loading criteria they set for the correlation coefficient using .50 as the cut off. For the present study, if factors load high on each rotation, decisions for factor retention or deletion will be clearly described and defined by parameters not yet determined.

Scales and Scoring

Examining validity evidence considers future outcomes and correlate scores of a measure

(Creswell, 2010, p. 165). Statistical procedures conducted to examine relationships among items on an instrument or to test a theory against scores characterize evidence of validity. Creswell

(2010) offered alternative paths to reporting validity evidence relating to internal construct by (1) use of non-statistical procedures (e.g., examining values in the interpretation of scores), (2) making assessments regarding the relatability of scores and the purpose of the study; and (3) exploring the likely social consequences (Creswell, 2010, p. 165). 130

Scale Development

Frequency distributions are used to graph how many times a score occurs (Field, 2009).

Scale development for the present study will be determined by analyzing frequency distribution tables or histograms to evaluate distribution normality. Pohan & Aguilar’s (2001) explained their decision to use scale scores to establish a range of views toward multicultural diversity instead of evaluating individuals. Pohan and Aguilar (2001) reported that Personal Beliefs About Diversity

Scales ranged from 15 to 75. The possible range of scores for the Professional Beliefs Diversity

Scale was reported to have been 25 to 125. Reliability can be determined by collecting pre and post-test data from the same samples to test scale reliabilities (Pohan & Aguilar, 2001).

Examining Validity Evidence Based on Reliability of Scales

Pohan and Aguilar (2001) examined scale validity evidence based on reliability in two phases of field testing. Samples were pooled from five states. They monitored scale reliabilities by investigating response set bias due to the sequence of scale administration, as well as response bias due to social desirability. After they examined validity based on internal consistency, a significantly revised, 16-item personal beliefs scale resulted.

Interpretation and Use Argument

Alternate evidence of validity relating to internal construct using non-statistical approaches is the primary way in which scales and scoring will be determined for the present study. According to Guion (1978) and Kane (2013), an argument is used to examine validity evidence to support an intended claim. The use of relevant evidence to support a plausible claim can be supported by the interpretation and use statement. Guion (1978) also asserted that validity refers to quantitative evaluations regarding broad categories. He continued by adding, validity 131

refers to evaluation, not fact (p. 499). Developing an instrument to examine perceptions requires making judgments or evaluation of quantities and numbers that are not precise (Guion, 1978).

Scores from a measure provide an opportunity to make inferences regarding data that can be used to make assumptions (e.g., negative perceptions, decision-making, likelihood for referring for special education eligibility; Guion, 1978).

Kane (2013) proposed that validation is more manageable when there is a clear and concise statement to support intended claims. The statement also indicates how claims were evaluated. Castillo et al. (2016) wanted to analyze the internal structure of the PRSS to make assertions regarding educator perceptions of their skills and actual behavioral components of skills related to response to intervention. They reported that scores revealed strong correlations between the perceptions of RtI skills applied to academic content and perceptions of RtI skills applied to behavior content factors. Castillo et al. (2016) concluded strong correlations were due to a high degree of similarity across academic and behavioral approaches. As a result, unique variations regarding differences in specific processes and procedures such as data sources used, and differences in instructional targets were a result of item redundancies. Their interpretation and use statement included an explanation for how alpha coefficients were used to make decisions regarding item retention.

Examining Evidence of Validity Using an Interpretation and Use Argument

Researchers have broad latitude in how values are interpreted. The interpretation and use argument (IUA) describe the way in which decisions are made. Additionally, the IUA is a critical aspect of examining validity evidence (Atwater et al., 1994; Guion, 1978; Kane, 2013). An instrument that may yield hypothetical variables (Guion, 1978). Hypothetical variables or 132

inferences are made based on individual responses. For example, participants in the present study will express their perceptions of African American related to scholastic aptitude, inferences can be drawn about perceptions regarding academic ability. Inferences are also referred to as assumptions (Guion, 1978). The IUA can guide the researcher in determining which statistical tests can be employed to test assumptions (Kane, 2013). For example, measuring perceptions proposes using analysis of variance (ANOVA) to test assumptions, variance, variable dependency, and normality.

Summarily, IUA statement for present study will be succinct and clear. The interpretation and use argument will focus on data collected that examines PreK -12 Educators’ beliefs, biases, perceptions, and stereotypes of African American males in PreK -12 public schools relating to attitude, crime, education, personal characteristics, sexual prowess, and probable eligibility for special education (Fitzgerald, 2009; Foster, 1995; Jackson & J.L. Moore, 2006; Quinn, 2017;

Wilson, Hugenberg, & Rule, 2017). When considering all of the categories listed above as an aggregation of implicit and explicit biases, perceptions of Black males will inform assumptions and inferences regarding African American males. Specific assumptions that negative perceptions of African American males are based on race and therefore influences educator decision-making will be tested by comparing overall perceptions to combined variables related to education and negative characteristics (Fairchild, Tobias, Corcoran, Djukic, Kovner, & Noguera,

2012; Rivard, Missiuna, Hanna, & Wishart, 2007; Simson, 2013; Washington, 1981).

Pohan and Aguilar (2001) briefly discussed the limitations affecting IUA statements. The main limitation was the impossibility to exhaust every known psychometric test to ensure validity. This assertion was corroborated by Atwater et al. (1994); Cronbach (1946); and Guion 133

(1978). They concluded that both beliefs measures were refined throughout the developmental process by the following processes: (1) examining item-total correlations, (2) appropriate and use using the range of scores, (3) examination of frequency distributions, and (4) monitoring variable contributions to scale validity using reliability. As an aggregation of the process, the number and wording of items varied throughout the developmental stages of test construction (Pohan &

Aguilar, 2001, pp. 175-176).

Analysis of Variance (ANOVA) tests the significance of group differences between two or more means, e.g., overall perceptions of African American males, academic ability, and criminality between elementary) between high school teachers (Mertler & Vannatta, 2013).

ANOVAs are also used to observe the differences between observed and predicted scores in a correlational analysis (Creswell, 2013, p. 339). The statistical symbol for ANOVA is F that tests differences between two or more groups related to one dependent variable (e.g., educators’ perceptions of criminality).

Examining group differences may assist in determining scale reliabilities. If there is a detection of group differences in a given measure may reflect evidence of construct validity

(Anastasi, 1976; Cronbach & Meehl, 1955; Pohan & Aguilar, 2001). Pohan and Aguilar (2001) argued that individuals who had a strong ethnic identity might be more accepting of other ethnic and cultural outgroups. Pohan and Aguilar (2001) conducted ANOVAs to determine whether ethnic identity was related to belief scores. They concluded that strong ethnic identity reflected favorability in perceptions and belief scores.

Means may vary by a small or large margin (Field, 2009). Means are expected to be similar or equal. Large mean differences are a cause for concern (Field, 2009). The present study 134

examines if using perception scores for the average person is more reliable than using self-report scores from the same population. If the hypothesis is null, then the means will be similar.

Standard error scores are the gauge that determines if the difference between means. When standard error is small, small differences in means is likely. The opposite is true when standard error is large. When examining evidence of validity for the present study, a t-Test of dependent samples will indicate if there are large mean differences between groups (e.g., race, age, gender;

Field, 2009).

Cross validation assesses consistency across different samples. This is an important step when making generalizations. However, this is not possible at the present stage of this study because true values are not known. Cross validation uses a regression model with two main methods of cross validation. The first model entails randomly splitting data in two halves. Once split, the second model compares both halves using the R2 test statistic (Field, 2009). This process enables me to check for consistency of scores across groups (e.g., scores for overall perceptions between males, female, or age categories).

Phase Four of Instrument Development

According to Benson and Clark (1983), phase three includes quantitative evaluation of test items along with: (1) administration of the first pilot test; (2) debriefing subjects; (3) calculating validity evidence based on reliability; (4) running item analyses; (5) revising the instrument; and, (6) making revisions of instrument to prepare for the subsequent pilots.

In phases three and four of instrument development, quantitative methods are used to determine validity evidence based on reliability (Benson & Clark, 1983; Castillo et al., 2016;

Creswell, 2010). Phases three and four are concurrent, cyclical steps using a variety of statistical 135

methods to explore evidence of validity (Creswell, 2010). Examining evidence of validity of an instrument includes (1) completing item analyses after administration of a second pilot of an instrument; (2) repeating revisions, pilot administrations, and item analyses; and, (3) beginning the iterative process for validation (p. 161). Although reliability (e.g., precision) was discussed here as an independent characteristic of test scores, it should be recognized that the level of reliability of scores has implications for validity (AERA, APA, & NCME, 2014, p. 34).

Furthermore, validity and reliability should be a priority for the interpretation and use of scores, and not solely on the instrument itself. The instrument is only as reliable as the claims made from the information it yields (Lavery, Jong, Krupa & Bostic, 2019).

Examining Evidence of Validity

Standards for Educational and Psychological Testing assert five sources of validity evidence (1) content based, (2) response processes, (3) internal structure (e.g., construct), (4) evidenced based in relation to other variables, and (5) outcomes of testing (AERA et al., 2014).

Both Castillo, et al. (2016) and Pohan and Aguilar (2001) examined evidence of validity by addressing biases (testing outcomes; AERA, et al., 2014). Creswell (2010) listed the following ways in which threats to validity can affect an instrument. Evidence of validity is compromised when (1) test items are unclear or ambiguous, (2) standardization in testing procedures is not consistent, and (3) participants are fatigued by the length or a combination of all three.

The present study requires examination of additional threats to validity that include: (1) non-response bias, (2) response set bias, (3) social desirability and self-report biases, and (4) researcher bias relating to improper or illogical interpretation and use of scores. Response set bias, according to Cronbach (1946) is the tendency for a respondent to consistently give different 136

responses in a test that would typically be reported if the content was presented in an alternate form (p. 476). Pohan and Aguilar (2001) found evidence of response set bias because items were negatively worded.

Pohan and Aguilar addressed response set bias by reversing the sequence of personal and professional beliefs scales. Pohan and Aguilar (2001) reported that the order of administration did not influence scores on both measures using t-Tests for independent samples to determine if significant differences were the result of scale sequence. Pohan and Aguilar (2001) determined that neither personal nor professional beliefs scores varied significantly for two sample groups.

Lengthy instruments require a significant time commitment and may expose a research study to measurement error due to non-response bias. The present study is at risk for non- response bias because of (a) large number of items needed to test dependent variables, (b) inclusion of a measure to control for social desirability bias, and (c) the sensitive nature of the topic (e.g., perceptions of African American males). Researchers like Haghighat (2007) and

Strahan & Gerbasi (1972) both agreed that including the full Marlowe-Crowne inventory (1960) for social desirability disadvantaged their research instruments because it made them too long.

Lengthy instruments might lead to a higher risk of non-response bias (Nicholls, Chapman, Shaw,

Perkins, Sullivan, Crutchfield, & Reed, 2011; Olson, 2006). Non-response bias is more likely when the subject matter is sensitive, or a combination of the former.

Haghighat (2007) asserted that the 33– item Marlowe Crowne Social Desirability Scale

(1960) did not represent the overall measure of social desirability. Additionally, he asserted that the MCSDS scale increased the chance for measurement error because of its length. Haghighat addressed the issue by creating the BSDS. Strahan and Gerbasi (1972) revised the MCSDS from 137

33 to 10 items. Haghighat (2007) condensed Strahan and Gerbasi’s 10 - item scale to 4 - items.

Haghighat (2007) advised that the brief social desirability scale BSDS cut off score could be set from > 1 (more than one socially desirable answer) to > 2 (more than two socially desirable answers) in order to exclude scores with a high tendency towards social desirability. Haghighat concluded that cut-off levels should depend on the importance of interpreting use for scores from participants who answer more transparently about their personal attitudes. Simply put, the sensitivity of the topic should inform the decision to set cut off scores. For example, public educators might be more likely to answer in a socially desirable way regarding perceptions of

African American males.

According to van de Mortal (2008), faking it (p.41), by appearing to be good might be reflected by a score 1.5 standard deviations above the mean and is a crucial factor when considering socially desirable cut off scores. When examining social desirability using Crowne

& Marlowe (1960), high scores indicate a high need to appear to be socially desirable. Crowne &

Marlow (1960) reported Mean scores related to the MCSDS (p. 352) as M=13.72 with a

Standard Deviation of SD=5.78. The aforementioned information is necessary when constructing an IUA statement. The IUA will address scores for social desirability by not factoring low scores, which indicate no SD bias. High scores on SD scales are the lone criteria for examining evidence of validity.

Haghighat (2007) used cross validation data to evaluate his five-item brief social desirability scale (BSDS) from 61 subjects. After cross validation, a brief 4- item social desirability scale was adopted for use in his study to examine stigmatization of patients with schizophrenia. Haghighat (2007) reported validity evidence using the Cronbach Alpha 138

coefficient. The Cronbach alpha coefficient for the 4-item BSDS was 0.6 for all participants, which he deemed to be acceptable as they fell within the moderate range.

The Cronbach alpha coefficient measures scale reliability by splitting the data in half to compute the correlation coefficient for each split. According to Field (2009), the average value of the two splits yields an alpha (a) score. The a score calculates variance within and between two items (e.g., all participants and gender). Alpha values below .7 are acceptable when dealing with psychological constructs like SD scales because of the diversity of constructs being measured (Field, 2009, p. 675).

Self-report bias differs slightly from social desirability biases. Self-report bias is significant in organizational behavior research. Self-report bias, like social desirability biases, can distort or yield false inferences about correlational and causal relationships (Donald &

Grant-Vallone, 2002). When participants believe that there is a risk of management gaining access to their responses, participants are more likely to answer in socially desirable ways.

Donald & Grant-Vallone (2002) posited that misinterpretation of data could occur when studies rely on self-reporting, due to shared variance.

There are few exceptions wherein excluding social desirability scales did not impact internal validity evidence based on reliability. For example, van de Mortel (2008), examined two studies related to biases against obese people, and attitudes toward bullying, where excluding socially desirable scales did not impair variables or study outcomes.

Evidence points to the fact that studies that evoke heightened emotions such as resentment, prejudice, or fear are subject to scrutiny when electing not to use social desirability scales, such as the Marlowe-Crowne Social Desirability Scale (MCSDS). Social desirability 139

scales allow researchers to examine correlational effects on variables or study outcomes

(Haghighat, 2001; 2005; Pohan & Aguilar, 2001; Tejeda, 2004). For example, a study correlating hate ideation against gay men and lesbians did not control for social desirability bias with a scale like the Marlowe Crowe Social Desirability Scale (Tejeda, 2004). Consequently,

Tejeda (2004) used stepwise hierarchal methods before examining variable relationships (cited by van de Mortel, 2008, p. 43).

In addition to using a MCSDS, Pohan and Aguilar (2001) added a dogmatism scale to provide additional evidence of validity. According to Pohan and Aguilar (2001), higher scores on the dogmatism scale (see, Rokeach, 1973) might indicate increased intolerance and closedness in an individual’s personal belief system. They used the Rokeach Dogmatism Scale and reported that the dogmatism scale score served as evidence of validity that dogmatism scores were significant but negatively correlated with scores on the personal beliefs scale (r = −.24; p <.008), with low to moderate correlation.

Validity Evidence for Reliability in Quantitative Research

Developing a quality measure requires that scores from a particular measure are stable and consistent. Validity evidence based on reliability can also be described in terms of internal homogeneity (Kane, 2013; Mertler & Vannatta, 2013). Examination of reliability is one way to describe the internal consistency of a measure using the coefficient alpha (Kane, 2013, p. 29).

Kane (2013) asserted that validity evidence based on reliability are different and cannot predict the assurance evidence of the other examinations of validity. Evidence of validity based on reliability relates to the items within an instrument and their overall representation of a universe of all possible items that exist regarding a topic (Atwater, Lau, Bass, Avolio, Camobreco, & 140

Whitmore 1994; Creswell, 2010). Kane (2013) proffered that validity shared two characteristics:

(a) a measure is evaluated over some population or subpopulation, and (b) the measure is widely accepted as legitimate in relation to the universal content of acceptability regarding a proposed topic (p. 61).

Haghighat (2005) demonstrated examples of obtaining validity evidence based on reliability using an instrument to measure stigmatization. He used the following steps to examine evidence of validity in the Standardized Stigmatization Questionnaire, version 1 (SSQ1; p. 144):

(1) simplified and rephrased items; (2) excluded items that were redundant, culturally limiting, or representing factual knowledge rather than attitude; (3) field tested items using rigorous methods to examine validity evidence based on reliability; and (4) conducted factor analyses that resulted in three factors of stigmatization (e.g., social, psychological, and evolutionary; see Haghighat,

2001). 141

CHAPTER 4. METHODOLOGY

This chapter describes the methodology used to develop an instrument to examine educator perceptions of African American males in public schools. The instrument includes an adapted social desirability scale to control for bias. The measure will also examine educator perceptions of African American males in two ways: (1) answering for the average person, and

(2) self-reporting personal perceptions. The research design, participants, instrumentation, research process procedures, research questions, data collection and analysis, study variables,

and limitations of the study are described in detail throughout this chapter.

Definition of Perceptions of African American Males

Foster (1995) defined perceptions of Black males as stereotypical beliefs, feelings, expectations, and fantasies held by the average person. Standing alone, Foster’s definition provided a starting point. Simson (2013) added implicit bias that results in stigmatization (e.g., rejection and isolation) as a factor that influences educator decision making. Hostile treatment creates stress for Black students causing them to act out (e.g., aggression, violence, defiance, disrespect, and other socially unacceptable behaviors). Stigmatization is defined by the hostile treatment that precipitates unfavorable outcomes for Black students. When African American students act out, it supports how some educators already perceived them based on race

(Nickerson, 1988; Simson, 2013). The present study uses findings from Hing, Chung-Yan,

Hamilton, and Zenna (2008) to add context to Foster’s (1995) definition of perceptions and

Simson’s (2013) stigmatization due to implicit biases. Hing et al. (2008) used a two-dimensional model that included political ideologies and implicit/explicit biases to predict discriminatory 142

behaviors within four prejudice profiles. Chapter 2, Table 2.3, listed the two dimensions of Hing et al. (2008) as the highs and lows of implicit and explicit biases.

I conclude that perceptions related to African American males include personal beliefs

(Pohan & Aguilar, 2001; Simson, 2013), stereotypical beliefs (Foster, 1995; Quinn, 2017); societal beliefs (Quinn, 2017; Wilson, Hugenberg, & Rule, 2017), feelings (Foster, 1995;

Haghighat, 2001; 2005; 2007; Washington, 1981), expectations (Fitzgerald, 2009; Pohan &

Aguilar, 2001), and fantasies (Foster, 1995).

Validity Evidence for Content: Developing the Public Educator Perceptions of African

American Males’ PreK-12 Survey

Findings from a qualitative study conducted by Foster (1995) reported data from n =

3130 participants regarding perceptions of Black men. He reported findings using tables and percentages from demographic data of study participants (e.g., educator/noneducator, Black female/White female/ non-Black female person of color). Overall perceptions of African

American males are negative (Fitzgerald, 2009; Foster, 1995; Wilson, Hugenberg, & Rule,

2017). The present study will determine if allowing people to answer on behalf of the average person is reliable in a quantitative study. Fosters study reported on six areas regarding negative perceptions of Black males. The present study used six categories, five of which were reported in the study conducted by Foster (1995).

In Chapter 2, the literature review indicated that findings reported by Foster (1995) were consistent with a universe of possibilities regarding perceptions of young Black males (Atwater,

Lau, Bass, Avolio, Camobreco, & Whitmore, 1994). Foster’s (1995) qualitative study of educator noneducator perceptions of Black men served as the inspiration for the present study. 143

The five perception categories that were derived from Foster’s study are crime, education, attitude, personality, and sexual prowess. A sixth category, emotional and behavioral disorders was created to form a special education (SpEd) category. The SpEd category was created by combining two categories (e.g., education and personality).

Exploratory Sequential Research Design

Findings from a qualitative study (see Foster 1995) aided in developing an instrument to examine educator perceptions of African American male students. According to Creswell (2012), an exploratory sequential research design involves mixed methodological approaches (see

Meijer, Verloop, & Beijaard, 2001). Qualitative methods in an exploratory sequential research design could include interviews (e.g., semi-structured interviews, concept-mapping; Creswell,

2012). Foster (1995) used a one-question survey to gather perceptions of African American males held by educators and non-educators. Quantitative methods were deployed in the present study using the five categories mentioned previously from Foster’s findings to create an instrument to examine educator perceptions. Descriptive and inferential statistics were used to explore the relationships between 2 dependent variables (personal perceptions held by educators, perceptions for the average person), and 1 independent variable (scores on the adapted brief social desirability scales; ABSDS).

The proposed instrument used a four-point scale to create a questionnaire to examine educators’ perceptions. Participants answered how the average person perceives African

American male students, as well as personal perceptions. Additionally, the measure included an adapted brief social desirability scale (ABSDS; see Haghighat, 2001) to determine if answers for the average person are more reliable than self-reporting personal perceptions. 144

Testing Research Hypotheses

It is possible that educators may not self-report personal perceptions of African American male students when directly asked in a survey. I believe that public educators, in order to appear to be good-natured, will report personal perceptions of Black male students more favorably than for the average person. Haghighat (2005) found this to be the case regarding the stigmatization of people with schizophrenia. Foster (1995) used one open-ended survey question asking participants to answer for the average person regarding perceptions of Black males.

Consequently, the present study will report the psychometric properties of the instrument used to examine educator perceptions of African American male students. Moreover, if perception scales

(e.g., perceptions of the average person, self-reported perceptions, and use of the adapted brief social desirability scale) work well to adequately measure overall perceptions of African

American male students, the following research questions can be examined:

(1) Is there a relationship between (a) perceptions of the average person and (b) self-

report scores among public educators of African American male students?

(2) Is there a difference in the relationship between (a) perceptions scores for the average

person and (b) self-report perceptions scores for participants with different scores on (c)

the adapted brief social desirability scales)?

Instrumentation

The Public Educator Perceptions of African American Male Students Survey

(PEPAAMS) PK12 consists of five parts: (1) qualifying questions that will determine whether participants will advance to the next section, (2) an adapted brief social desirability scale, (3) the average persons’ perceptions of African American males, (4) self-report perceptions of African 145

American males, and (5) personal and professional demographic information. Participants received invitations via e-mail that included a URL link directing them to the survey in Qualtrics

(See Appendix B).

Validity Exploration

An expert panel was assembled to appraise content of the measure regarding negative perceptions of African American males. The expert panel was made up of 15 professionals (e.g.,

PreK -12 teachers, professors, criminal justice professionals, or other related professions that are familiar with African American male perspectives (see Castillo et al., 2016; Creswell, 2012;

Galvan, 2002; Pohan & Aguilar, 2001). Expert panel members were required to hold at least a bachelor’s or higher in a related field. The criterion for changes and deletions was or near 80% majority agreement amongst expert panel members (Pohan & Aguilar, 2001; See Appendices D

& G).

As a graduate teaching assistant at a university in Northwest Ohio, I was able to ascertain professional e-mail addresses of over 4100 public school educators. This was a convenience sample (Galvan, 2002). The list of e-mail addresses was divided to conduct an initial sample of the measure, and for a broader for administration of the Public Educators’ Perceptions of African American Males PreK -12 Survey post-doctorate (see Creswell, 2010; Pohan &

Aguilar, 2001).

The target sample comprised public school teachers in Northwest Ohio. The ABSDS was created to control for social desirability bias within the PEPAAMS PK12 (see Haghighat, 2007;

Pohan & Aguilar, 2001). The PEPAAMS PK12 measure was piloted to examine evidence of validity for internal reliability using a small sample of practicing public educators in Northwest 146

Ohio. All subsequent pilots and distributions of the PEPAAMS PK12 will sample public educators in the largest school districts in the United States (e.g., Los Angeles, Chicago, Atlanta;

Creswell, 2010; Castillo, et al, 2016).

Assumptions

The hope is that the present study will contribute to a body of research in education that focuses attention on the plight of African American male students attending public schools in the

United States. I hold several assumptions from reviewing relevant literature. First, some participants may not feel comfortable with disclosing personal perceptions of African American males. They may choose not to participate (Haghighat, 2005; 2007; Crowne & Marlowe, 1960;

Strahan & Gerbasi, 1972; Van de Mortel, 2008; Wark & Galliher, 2007). Additionally, the research suggests that biases related to social desirability are likely to occur, which I assume will be the case when asking participants to report personal perceptions (Castillo et al., 2016; Pohan

& Aguilar, 2001). I do not expect the same issues from participants reporting perceptions on behalf of the average person.

Qualifying Questions

The survey began with two questions meant to qualify participants to continue with the study. In order to identify the target sample relevant to the present study, individuals not eligible to continue did not advance to the study based on answers given in Part I of the PEPAAMS

PK12 (see Appendix B). Participants were required to hold a valid teaching license for their respective state. Additionally, educators were required to be employed in full or half-time status working directly with students in some related capacity. Identifying the sample population allowed me to make inferences from collected data about participants in the study. Table 4.1 lists 147

three questions that participants were required to answer in order to proceed to Part II of the

Public Educators’ Perceptions of African American Males PreK -12. Participants meeting the criteria to advance to part 2 of the PEPAAMS PK12 survey automatically advanced to the adapted brief social desirability scale (ABSDS).

Table 4.1

Public Educators’ Perceptions of African American Males Survey - Qualifying Questions

Yes No Are you a licensed educator in a PreK – 12 Continue Survey concluded Public School in the United States? Are you currently teaching in a PreK – 12 Continue Survey concluded Public School full or part time?

Examining Validity Evidence for the Adapted Brief Social Desirability Scale

Research studies and surveys that contain sensitive topics are vulnerable to measurement error (Berliner & Calfee, 1996; Foster, 1995; Haghighat, 2007; Pohan & Aguilar, 2001; Van de

Mortel, 2008; Wark & Galliher, 2007). Controlling for measurement errors in self-report surveys is essential to increase confidence in the validity and reliability of a measure (Castillo, March,

Stockslager, & Hines, 2016). I used Haghighat’s (2007) brief social desirability scale (BSDS) to develop an adapted scale by using phrases that are related to professionals working in education.

Haghighat’s BSDS had five items.

In Table 4.2 the left column lists the original five items from Haghighat's (2007) BSDS.

The first 4-items were significantly representative of social desirability. The right column of

Table 4.2 shows the adapted version the BSDS for educators. Items 1 and 4 were changed to 148

reflect school based scenarios and to address response set bias (Castillo et al., 2016; Cronbach,

1946), and for attention check (Kung, Kwok, & Brown, 2017).

Table 4.2

Adapted Brief Social Desirability Scale – Part II of the Public Educator Perceptions of African

American Males PK12 Survey

Haghighat’s (2007) Brief Social Desirability Scale ABSDS for PEPAAM PreK -12 (BSDS) Survey 1.Would you smile at people every time you 1.Do you smile at every student you meet them? encounter during the school day? 2.Do you always practice what you 2.Do you always practice what you preach? preach? 3.If you say will do something, do you always keep your promise no matter how 3.Do you always keep your promise? inconvenient it might be? 4.Would you laugh or made fun of a 4.Would you ever lie to people student you did not like behind their back? 5.**Would you ever laugh at a dirty joke 5.Do you always admit when you make a people may make mistake?

Examining Evidence of Validity for Reliability Using the Adapted Brief Social Desirability Scale (ABSDS)

The present study examined validity evidence for reliability by conducting cognitive interviews using a convenience sample with people that were willing to participate. Interviews were conducted in a place where many professionals gather to eat and shop in Central Ohio (see

Haghighat, 2007). Participants were asked to answer questions as a person who wanted to appear to be good in the eyes of others. Cognitive interviews were conducted using qualitative methods 149

to address items that may be problematic in the processes of examining evidence of validity for the PEPAAMS PK12 (see Appendix I).

Part II - Answering for the Average Person and Self Reporting Personal Beliefs

Regarding African American Males

Table 4.3 lists the items adapted from findings reported by Foster (1995). Part III of the

PEPAAMS PK12 has two dependent variables: (1) perceptions of African American males reported for the average person and (2) self-report of personal perceptions of African American male students held by individual educators. There are five sub-variables for each dependent variable: (1) criminality, (2) educational attainment, (3) attitude, (4) personality traits, and (5) sexuality (See Appendix B).

Table 4.3

Perceptions of Black Males Categories and Items: Foster (1995)

special education/ emotional & criminality education attitude personality sexuality behavioral disorders 1. dumb 2. ignorant 1. larger penis 3. low intelligence 2. strong sex 1. dumb 1. belligerent 4. uneducated drive 1. criminal 2. ignorant 2. mean 5. unmotivated 1. lazy shiftless 3. good lovers 2. dishonest 3. low 3. vain 6. apathetic 2. irresponsible 4. rapes 3. steals intelligence 4. abusive 7. belligerent 3. drinks women 4. sells drugs 4. uneducated 5. aggressive 8. mean 4. on welfare 5. abuses 5. uses drugs 5. unmotivated 6. threatening 9. vain women 6. apathetic 7. violent 10. abusive 11. aggressive 12. threatening 13. violent 150

Public Educators’ Reporting Personal Perceptions of African American Males

Public educators’ professional beliefs might differ from their personal beliefs (Hing,

Chung-Yan, Hamilton, & Zenna, 2008, Pohan & Aguilar, 2001; Washington, 1981).

Assumptions that public educators may not express true personal beliefs when asked directly about perceptions of African American males’ functions as a hypothesis. Public educators will be asked to express their personal perceptions of African American males.

Parts III and IV - Measuring Educators’ Perceptions of African American Male Students

PK12 Scales and Scoring

The Adapted Brief Social Desirability Scale will be used as a qualifying variable to inform how dependent variables are described and interpreted in Parts III and IV of the Public

Educator Perceptions of African American Males Survey (PEPAAMS) PK12. The Adapted Brief

Social Desirability Scale (ABSDS) can range from 5 to 10. High scores reflect social desirability bias (Haghighat, 2007). A preliminary cut-off score for the ABSDS will be 7. Haghighat (2007) recommended that the brevity of the scale allows the researcher to use discretion when determining acceptable cut off scores. He recommended allowing participants to answer 1, but no more than 2 items in a way that reflects social desirability. If a participant answered an item in the ABSDS that reflects social desirability, a score of 2 was assigned for each item. A score of

1 on each item indicated that a participant is not likely to reflect social desirability bias and, therefore, will answer candidly regarding personal perceptions of African American male students. There are five items, participants could earn a minimum score of 5 and a maximum score of 10. 151

Briefly, the adapted scale reflected the original intent of Crowne & Marlowe (1960) to provide respondents with a list of "culturally acceptable" behaviors in the context of education, which are "relatively unlikely" to occur in real life situations (pp. 353-354). Validation of the adapted scale presented in this chapter will be based on two questions: (1) does this question represent a relevant experience of a typical educator in public schools; and (2) is it likely for a person to meet the criteria with 100% surety. Data collected in the validation examination are included in Chapter 5 of the present study. Cognitive interviews were used as the primary process to examine validity evidence for the ABSDS (Nicholls, Chapman, Shaw, Perkins,

Sullivan, Crutchfield, & Reed, 2011; Olson, 2006; Plano-Clark & Creswell, 2015).

Interpretation and Use Argument for Scales on the PEPAAMS PK12

A 4-point scale was used over a 5-point scale. Non-response bias is a risk for the

PEPAAMS PK12 because of subject sensitivity. Shortening 5-point scales eliminated neutral categories that allows for participants to avoid expressing their true feelings (Haghighat, 2005).

Furthermore, providing neutrality as an option creates a space for avoiding self-revelation

(Haghighat, 2005, p. 146). Using a 4-point scale over a 5-point scale allowed me to make better inferences regarding data.

Part III of the PEPAAMS PK12 consists of 44 items for participants to answer on behalf of the average person (see Foster, 1995). Respondents can choose from the following answer options to negatively worded items: (1) strongly agree, (2) agree, (3) disagree, and (4) strongly disagree. Parts III and IV of the PEPAAMS PK12 public educators responded to perceptions for the average person, as well as personal perceptions. Summarily, scales for Part III and IV were 152

the same which allowed public educators to self-report personal perceptions of African American males on a 4-point scale.

Statistical analyses for both parts III and IV of the PEPAAMS PK12 will primarily focus on mean scores and standard deviation of scores. After proper examination of validity and reliability, individual differences may be computed by adding the sum of scores for both parts of the measure. Data for participants whose scores fall within the range of high social desirability bias (e.g., >8) will be considered with (a) perceptions of the average person scores, or (b) personal value statements (political ideology views). If the ABSDS does not function as intended, personal value statements (ideological views) will replace the ABSDS as the independent variable. Personal value statements will only be used if the ABSDS does not perform as intended. For example, low social desirability scores (£7) might indicate that personal perception scores best reflect the true perceptions of public educators. However, social desirability scores 8 or above might suggest that scores for the average person are a reliable indicator of educator perceptions of African American males.

Part V - Perceptions of African American Males Survey: Demographic Data

The length of the PEPAAMS PK12 Survey was a concern. Engaging participants about personal perceptions of African American males might engender emotions of anger, resentment, and hostility (Fitzgerald, 2009; Foster, 1995; Washington, 1981). Surveys like the PEPAAMS

PK12 are susceptible to measurement errors, particularly non-response bias. Positioning the demographic section at the end of the survey is a strategy intended to control for non-response bias. Placing demographic questions at the end of a questionnaire is a strategy to keep participants engaged and lighten emotional loadings from sensitive topics (Burns, Duffett, Kho, 153

Meade, Adhikari, Sinuff, & Cook, 2008). Some researchers choose to use demographic questions to ease respondents into a questionnaire (Burns, et al., 2008, p. 247). However, they suggest that investigators may opt to place demographic questions at the end of the survey when questions posed in a measure are sensitive. The demographic section in the survey provides critical information to make conclusions about the sample population. For example, Hing et al. (2008) found that four prejudice-related ideologies were a predictor of an individual's potential to stigmatize when acting on explicit and implicit biases. The demographic section includes one category regarding personal values, which allowed me to make inferences regarding negative perceptions that might correlate to prejudice. Another category in the PEPAAMS PK12 Survey includes items relating to sexual orientation and family status (Pohan & Aguilar, 2001). Sexual orientation and family status provide another layer to filter how perceptions differ with alternate independent variables. Finally, the PEPAAM PK12 Survey asked participants to indicate if an immediate family member or spouse was a law enforcement officer, which allowed exploration of Wilson, Hugenberg, and Rules’ (2017) theory regarding threat perceptions of African

American males. Specifically, explorations relating the violence, threats, and potential for harm.

Procedures

Data collection commenced upon approval granted by the Institutional Review Board

(IRB; see Creswell, 2010; Haghighat, 2007). Expedited approval was requested because of the minimal impact on human subjects. After IRB approval, the assembling of expert panels commenced to begin content validation. Once content validation was satisfactorily examined, administration of the PEPAAMS PK12 commenced by requesting participation from Northwest 154

Ohio teachers via email with an attached hyperlink that redirected them to the survey in

Qualtrics.

Once an invitation was sent via e-mail, prospective participants received a reminder every week for three weeks. Completed surveys were downloaded from Qualtrics into IBMs

SPSS Software 22.0 wherein data was coded, screened, and simplified. No personal information was used to identify participants.

Informed Consent

Approval was granted by the Institutional Review Board and data collection commenced.

I did not anticipate any complications regarding the study of Educators' Perceptions of African

American Male students as I believed it posed minimal to no risk for participants (Creswell,

2012; see Appendix E). However, once the study commenced, I was contacted by the Office of

Research Compliance making me aware that they received complaints about the study. At the center of each complaint were two items on the survey, which made reference to perceptions, specifically, sexually explicit aspects of stereotypes. The research study was halted until a meeting was scheduled with the compliance office. After the meeting was held with the office of research compliance, modifications were made to the instrument. Informed consent was updated to include (1) a bolded statement alerting participants of sexually explicit references, (2) a disclaimer that advised participants to take the survey on a personal device – not on a work computer. Finally, before the items in question appeared in the survey, an option to skip the item and continue with the survey was added so that participants who might be offended by the content could bypass it and continue with the study (see Appendix A). 155

Participant anonymity was secured because respondents were not required to provide any personally identifiable information. Completed surveys provided the date, time of day, and the time it took to complete the survey. Data were generalized to the whole group using general demographic categories. Results did not report on individual cases. In sections requiring extended responses, themes, patterns, and conclusions were summarized as a part of the sample.

Informed consent provided the following to study participants in addition to the modifications described previously: (1) summary of the study, (2) participants’ right to withdraw at any time, (3) voluntary participation, (4) guarantee of privacy, and (5) right to know the outcome of the study (see Creswell, 2012, p. 159). Informed consent for this study appeared in the initial invitation message sent via email, as well as the homepage when the Qualtrics URL was activated. Once activated, participants were directed to the Public Educator Perceptions of

African American Males Survey PK12 in Qualtrics. Informed consent also included a brief description of the study, which included its purpose, benefits, risks, time commitment, and person to contact for further information. Prospective participants provided consent by activating the next button for the Public Educator Perceptions of African American Males PK12 Survey.

Participants were required to read brief and concise instructions to each section of the survey. At the conclusion, a thank-you message appeared.

Data Collection for Each Phase of the Instrument Development Process

Timelines for collecting data when developing an instrument are not precise and may vary by the results at each phase (see, e.g., Castillo et al., 2016; Haghighat, 2005; 2007; Pohan &

Aguilar, 2001). Ideally, data collection for all stages of instrument development process will take place during a 60 to 180-day period. Educators are more likely to respond to surveys during 156

October and November (Roberts, 2010). However, due to the unpredictable nature of approval time for institutional review board; participant responses; and any other unforeseen occurrences, the survey was administered late April through mid-May.

Phase I

Data collection ran concurrently utilizing different samples to examine (1) validity evidence for the content of measures related to perceptions and (2) reliability of the ABSDS.

Participants were invited to participate in cognitive interviews for the Adapted Brief Social

Desirability Scale (ABSDS). The target group for this step in the examination of the ABSDS were willing to spend a minimum of ten minutes (Benson & Clark, 1983; Creswell, 2010).

Phase II

The Public Educators’ Perceptions of African American Males Survey PK12 has two dependent variables that include six sub-variables that make up overall perceptions. An expert panel comprised of various professionals in law and education who have either participated in presentations regarding educator perceptions of African American males, or who have personal and professional knowledge of the subject matter, appraised items in each category for representativeness, logical nomenclature of categories, provided feedback on items to determine if there are any redundancies. The time dedicated for validation of content ranged from 7 to 14 days.

After receiving feedback from the expert panel and panel member interviews, items were formalized and presented to the principal expert reviewer for feedback relating to the fidelity of revisions recommended by the expert panel. It was necessary to conduct personal interviews with 157

some members of the expert panel to get clarification regarding certain areas of concern (e.g., wording, scales, and content; see Appendices D & G).

Phase III

Phase III of data collection included the first distribution of the PEPAAMS PK12 with a small sample of educators in Northwest Ohio, which included all five parts of the measure. The

PEPAAM Survey PK12 consisted of the following parts: (a) Part I – two qualifying questions that determined progression to the remainder of the measure; (b) Part II – an adapted brief social desirability scale with 5 items; (c) Part III – public educators answered 22 items on behalf of the average person regarding perceptions of African American males; (d) Part IV – public educators answered 22 items regarding personal perceptions; and (e) Part V – participants answered demographic related items.

Phase IV: Ongoing Examination of Validity

In Phase IV of the instrument development process, a second distribution will be planned for administration at the conclusion of the present study with a broader sample that requires funding to support a large distribution with public educators in the largest school districts across the United States. Results of this study will be summarized for publication. Additionally, grant funding will be secured to expand the study to understand educator perceptions of African

American male students.

Data Analyses

The first step in the process of data analysis was to screen the data set for outliers and missing data. A preliminary screening of the data indicated that there were no problems with variables that were not equally distributed. Participants that did not provide answers for critical 158 variables were deleted from analysis. Exploratory factor analysis using maximum likelihood was conducted for dependent variables. Missing scores were replaced with the mean of scores from the sample because the sample size was sufficient for advanced statistical analyses. Chapter

5 outlines the interpretation and use argument (IUA) plan for employing appropriate statistical methods.

Descriptive Statistics

After data screening, descriptive statistics were calculated to examine measures of central tendency and variability for the data set. The mean was used to evaluate the categorical independent variable for the adapted brief social desirability scale, answering perceptions for the average person, and public educators’ self-reporting their personal perceptions of African

American males. Additionally, demographic data were used to describe participants in order to understand group differences in perceptions for both dependent variables. Measures of central tendency were used to examine the normality for personal values that replaced the independent variable because the ABSDS did not function as intended (Pohan & Aguilar, 2001).

Inferential Statistics

Inferential statistics were used to answer research questions (e.g., relationships between both dependent variables; Plano-Clark & Creswell, 2015). Additionally, inferential statistics helped to examine validity evidence for the adapted social desirability scale and to compare: [1] relationships between (a) reporting perceptions for the average person and (b) self-report perceptions of public educators, [2] differences in the relationship between (a) perceptions scores for the average person, (b) self-report perceptions, and (c) low/high scores on the adapted brief social-desirability scale. Table 4.4 lists specific test statistics used for this study, its symbol, use, interpretation, and example application. 159

Table 4.4

Test Statistics Used in Inferential Analyses

Statistic Name Symbol Use Significance if Example t-Test t Difference p£a Comparing between 2 average person groups with one and self-report dependent perceptions variable when controlling for social desirability bias

Multivariate F Differences Significance if Comparing Analysis of between 2 or p£a group (e.g., self- Variance more groups in report personal, (MANOVA) terms of average person, independent gender, race, variable age) Pearson r Group Significance if Compare Correlation differences in a p£a educators (e.g., categorical self-report scores independent and their variable prejudice profiles) Linear b Explore if the Significance if Exploring Regression independent p£a whether personal variable can value statements predict the value are an adequate of the dependent predictor of variable negative perceptions 160

The research question is intended to evaluate the relationship between public educators’ answering for the average person (see Foster, 1995) and self-report scores regarding personal perceptions (see Castillo et al., 2016; Pohan & Aguilar). Multivariate analysis of Variance

(MANOVA) was used to evaluate significant differences in mean scores for the two dependent variable(s), and between two or more groups (Mertler & Vannatta, 2013). MANOVA was used to explore mean differences related to the independent variable, social desirability bias scores; and the dependent variables, answering for the average person (AAP), and self-reporting personal perceptions (SRPP). The following list shows how group means were evaluated by independent and dependent variables: (1) low social desirability bias, (2) high social desirability bias, (3) perceptions for the average person/low social desirability bias, (4) self-report of personal perceptions/low social desirability bias, (5) perceptions for the average person/high social desirability bias, and (6) self-report perceptions/high SD bias. The present study used

MANOVA to evaluate mean differences that informed how decisions were made when interpreting average and personal perception scores.

MANOVA was used to replace the independent variable formerly represented by social desirability scores in three phases to examine significant effect. First, MANOVA was used to determine if there was significant effect between the 12 personal value statements (e.g., ideological statements), and the dependent variables (e.g., scores for the average person and personal perceptions). MANOVA was also used to replace the independent variable (personal value statements) with ideological values that were least likely to lead to marginalization and stigmatization as reflected in less negative perceptions of African American male students

(TLP). Next, the independent variable was replaced with personal value statements that were more likely to lead to discrimination or marginalization resulting in negative perceptions of 161

African American male students (e.g., not TLP) to determine if a significant simple linear regression equation is realized. Three of the 12 personal value statements comprise ideological values that are TLP. Nine of the 12 personal values statements made up ideological values that are not TLP. Chapter 5 will discuss how distinctions between TLP and not TLP were determined (see Table 5.12).

The paired samples t-Test was used to identify group differences between average and personal perceptions of African American males held by public educators (Mertler & Vannatta,

2013). In order to evaluate the research question, the independent variable and two dependent variables relating to public educators expressing personal perceptions of African American males were compared using the scores from the adapted brief social desirability scale (ABSDS). The independent variable (social desirability) and the dependent variable(s) AAP and self-report scores relating to perceptions revealed issues related to internal reliability.

Studies that focus on race and gender (e.g., Black males) are prone to specific challenges such as scrutiny of the findings that point to systemic racism as causal factors (Berliner & Calfee, 1996).

Secondly, developing an instrument to mitigate threats to validity and measurement error required an extensive review of literature that would inform a reasonable interpretation and use argument (Kane, 2013) in defense of decisions regarding use of scores and data interpretation.

The primary limitation of this study relates to topic sensitivity and the potential of measurement error. For example, a person holding biases towards African American males may choose not to continue in the study resulting in incomplete data. I anticipate non-response bias (see Castillo, et al., 2016; Nicholls, Chapman, Shaw, Perkins, Sullivan, Crutchfield, & Reed, 2011; Olson, 2006;

Sakshaug, Yan, & Tourangeau, 2010), followed by social desirability bias (Donaldson & Grant-

Vallone, 2002; Van de Mortel, 2008) as the most significant limitations that can result in measurement error. 162

Conclusion

Low participation rates amongst educators are more likely to occur during certain months of the year (Roberts, 2010). According to Roberts (2010), January and February are not ideal to seek participation in a study for educators due to professional pressures and the emotional slump after the holiday season. For example, educators are prone to have negative outlooks in January and February due to the “holiday depression effect.” Roberts (2010) further explained that data collection in March and April are not ideal as schools adjourn for Spring recesses. Additionally,

Ohio administers its State examinations during this time. May, June, July, and August, according to Roberts (2010), are the months wherein securing participants in a study are not ideal. This period is characterized by (1) end of the academic year, (2) summer recess, and (3) beginning of the school year. The PEPAAMS was not administered during an ideal time of the academic year, which I believe may have impacted participation rates, along with the sensitivity of the subject matter. September through November are the best times to administer surveys to educators (Roberts, 2010). The PEPAAMS PK12 was administered during the end of April through May 2019. 163

CHAPTER 5. RESULTS

Chapter 5 presents the results from reliability and validity tests and other psychometric property evaluations of the PEPAAMS PK12. This instrument was used to measure educator perceptions of African American male students in public schools when self-reporting responses for the average person and personal perceptions. This chapter begins with a description of the validity examination process for the adapted brief social desirability scale used to answer research question one. Additionally, Chapter 5 will report the results from the administration of the instrument at the center of the present study. Finally, the results from MANOVA testing are reported to answer research question two.

Cognitive Interviews for the Adapted Brief Social Desirability Scale (ABSDS)

The adapted brief social desirability scale (ABSDS) was a potentially significant factor used to interpret results from the perceptions of African American males survey PK12.

Specifically, the ABSDS affects how the interpretation and use argument (IUA; Kane, 2013) was developed. The ABSDS was integral to the results of Research questions 1 and 2.

Research question 1 tested the relationships between (1) educators reporting the perceptions of the average person, and (2) self-reporting personal perceptions of African

American males. Research question 2 examines the difference in the relationships between (a) perceptions scores for the average person and (b) self-report perceptions scores for participants, considering scores on the ABSDS. There were potential issues that had to be considered in order to derive a logical, clear IUA statement using results from the ABSDS. First, the present study answered how the scores from the ABSDS were used to make sense of the self-report scores for public educators. Next, results indicated that there were discrepancies in scores from the 164

ABSDS. Consequently, personal value statements were used as an alternative variable to determine which dependent variable should be used to draw conclusions and make inferences regarding perceptions of African American male students.

Examining Validity of the Adapted Brief Social Desirability Scale (ABSDS)

Cognitive interviews (CIs) were used to examine the viability of the Adapted Brief Social

Desirability Scale. The adapted scale was taken from Haghighat’s (2005) Brief Social

Desirability Scale (BSDS) for use with an instrument to examine stigmatization and marginalization of patients with schizophrenia. Haghighat’s five-item scale was used to exclude scores for participants that showed evidence of social desirability bias. Haghighat provided guidance on the cut off score for his brief social desirability scale ranging from a score of 1 to 5.

Scores on the ABSDS ranged from 5 to 10. High scores reflect social desirability bias (Crowne

& Marlowe, 1960; Haghighat, 2007; Strahan & Gerbasi, 1972). A preliminary cut-off score for the ABSDS was 7. Haghighat (2007) recommended that the brevity of the scale allows the researcher to use discretion when determining acceptable cut off scores. He recommended allowing participants to answer 1, but no more than 2 items in a way that reflects social desirability (See Appendix I).

I recruited participants for cognitive interviews. All participants were college graduates at various stages in their careers, including retired professionals. Participants were interviewed in three phases. First, I explained the purpose of the social desirability scale and how it relates to the study. Specifically, I explained that studies that involve difficult topics are subject to bias because people tend to provide answers that make them appear to be good to others (Crowne &

Marlowe, 1960; Foster, 1995; Haghighat, 2001; 2007; Strahan & Gerbasi, 1972). 165

Cognitive interviews allowed each participant to verbally deliberate the distinction between socially desirable answers without much thought. However, when discussing the answer that was not socially desirable, participants went through the process of establishing why an item would not be socially desirable. Each interview was highly structured. I began by giving a brief rationale why a social desirability scale was necessary in a research study regarding sensitive topics. Additionally, each participant was given the same example regarding my hypothetical dog, Lana. I communicated to each person that if asked if I ever left my dog outside in extreme weather conditions (e.g., heat or cold), I would say, “no,” because who wants to admit to leaving a dog out in those conditions. I want to appear to be good in the eyes of animal lovers.” Conversely, when speaking with each participant about the possibility that I could have left my dog out in extreme weather conditions, so the honest answer would be yes. I gave the same example to each participant; I could have gotten busy and forgot my dog was outside in the extreme heat or cold.

Results from cognitive interviews indicated that there may be reliability concerns with two items in the ABSDS. For example, when asked if they ever lie to students, respondents raised concerns regarding the context of lying to build rapport. Specifically, one respondent stated, “I think a lot of teachers do it [lie] to be silly or to build rapport with children.”

Additionally, in the validity examination phase of instrument development, respondents were concerned about whether a teacher would admit to giving misleading information regarding a student. Initial examination gave indication that the two items mentioned previously would not hold up well in examination of reliability. I concluded that the two items regarding lying to and 166

giving false information regarding students should be eliminated because the ABSDS is intended to mitigate dishonesty, rather than confuse participants with questions regarding honesty.

Analysis of Cognitive Interviews

After modeling the process of responding to a cognitive interview, each person was shown the computer screen to read each of the items. Each participant first gave the socially desirable answer, followed by providing an answer for each item that was not socially desirable.

For example, most participants discussed whether it is feasible to smile at every student encountered during the school day. In another example, participants concluded that, as much as they would like to, nobody practices everything that they preach. After ten cognitive interviews, all responses yielded similar cognitive reasoning to derive socially desirable answers versus answers that were not. The ABSDS used for the present study had seven questions. As a result of cognitive interviews and consultation with the methodologist of record, two questions were reworded, but ultimately removed. One question asked, “have you ever given false or misleading information regarding a student.” After one particular cognitive interview, a participant raised serious concerns that a teacher would never admit to lying regarding a student even if they had done so. The question was reworded to, “Have you ever given misleading information regarding a student.”

One question asked, “do you smile at every student during the school day.” An educator depending on their role can see hundreds, and in some cases thousands of students during the day. For example, a principal may see every student if they are highly visible throughout the school day. As written, the wording was ambiguous. To provide clarity, it was changed to reflect more specificity, “do you smile at every student you encounter during the school day.” 167

Addressing ambiguous wording of items in a survey can address potential measurement error

(Creswell, 2012).

In the original adapted social desirability scale for the present study, participants were asked, “have you ever laughed or made fun of a student you did not like.” Lack of clarity for this item could be interpreted that an educator would do this in front of students. We assume that most teachers have high moral and ethical standards. Most educators would not laugh or make fun of students in the presence of students. This item underwent a slight change to specify when an educator might laugh or make fun of a student they did not like. For example, the teacher’s lounge could be a place that significant dialogue about individual students might occur - outside the presence of students. The item was modified to reflect laughing or making fun of a student behind their back. It should be noted that one cognitive interview yielded contradictory information regarding the ethical and moral code of educators. One participant indicated that he, in fact, did experience an educator that laughed or made fun of students in the presence of students, and to other students, individually. This information was outside the scope of the study, and I assumed that laughing or making fun of a student in their presence was not a common occurrence.

Participants that subscribed to high moral and ethical values answered “no” for both social and non-socially desirable categories. Furthermore, the purpose of the scale is to detect social desirability bias (i.e., answers that do not represent the honest sentiments of participants).

Table 5.1 summarizes results from cognitive interviews that includes the socially and non-social desirable responses. Scoring for the ABSDS will be covered later in the chapter. 168

Table 5.1

ABSDS Results from Cognitive Interviews

Socially Not Item Desirable Desirable Answer Do you always practice what you preach? yes no

Do you always keep your promise? yes no

Have you ever laughed or made fun of a student you did not like behind their no yes back?

Do you always admit when you make a mistake? yes no

Do you smile at every student you encounter during the school day? yes no

As a final point regarding cognitive interviews, I used the terms “answer that is not socially desirable” and the “honest answer” interchangeably when interviewing participants in the validity examination phase for the ABSDS. This might pose a problem for critics because an answer could be not socially desirable and yet not be the honest answer. Additionally, an honest answer could be simultaneously both socially desirable and honest (see Appendix I).

Alternatively, any use of the scale for future research must be broached with caution if cognitive interviews are to be used to serve as the validation method for the ABSDS. Lack of continuity in the terms might have led to some confusion, even though participants did not articulate any misunderstanding in terminologies during interviews.

Expert Panel Review of PEPAAMS PK12

An expert panel, comprised of 15 people was assembled to take, examine, and provide feedback regarding content and construct validity of the PEPAAMS PK12. One person was 169

appointed and agreed to serve as the principal expert panel reviewer. The role of the principal expert reviewer (PER) was to engage in a presentation of all the feedback provided by expert panel reviewers. After I reviewed feedback from the panel, I made revisions to the scale. The

PER’s role was to examine the final instrument to determine whether all changes and feedback had been reasonably addressed.

Expert panel members were sent an invitation to participate as an expert reviewer via e- mail. Expert panel members consisted of professionals that I have known in professional roles as principal (e.g., educators that I supervised and principals who were colleagues), conference presenter (e.g., Educational Law Association, National Principal Leadership Institute), and other professionals whom I have encountered in some capacity within the last four years as a doctoral student. All expert panelists have at least a baccalaureate degree in an education or related social science field. The principal expert reviewer holds a doctoral degree in education and serves as the supervisor of special education in one of Ohio’s largest city school districts.

Expert Panel Feedback

I sent participation requests to 55 professionals to participate on the expert review panel.

Fifteen people responded back agreeing to participate as a panel member. The large panel pool was desired to extract a diversity of perspectives. The survey was constructed through Qualtrics.

I expected that some professionals would not respond in a timely manner. Within the first 48 hours, four members responded by completing the survey instrument. They also provided feedback as requested. Expert panel members were prompted to provide feedback in the following message within the survey: 170

“This is the section wherein I need the expertise of the expert panel. Please read each

question below and offer as little or as much feedback that will help me make this

instrument better. If you would like to speak to me directly about this survey, please feel

free to call me at 614-535-7773, or by e-mail at [email protected]. Thank you.”

Expert review panel members (ERPMs) were asked to answer nine questions to assist in refining the survey instrument. See Table 5.2. Additionally, they were asked to answer demographic questions to ensure that the panel represented (1) diversity of job positions within education and related fields, and (2) varying degree levels; e.g., profession, job title, age, gender, race/ethnicity, and highest degree held). 171

Table 5.2

Expert Review Panel Questions

1. Was the amount of time needed to complete this survey reasonable (e.g., is it too long?) If so, please explain.

2. Are directions clear? If not, please explain.

3. In your opinion, are questions scaled appropriately? (e.g., strongly agree, strongly disagree). If not, please explain.

4. Are questions clear and concise? If not, please explain.

5. Please list any questions that were not clear or concise (e.g., 1, 2, 4). Please provide recommendations to make it/them better.

6. Were there any confusing words in questions? If so, please explain.

7. Do you feel questions are worded in a culturally appropriate manner? (e.g., easily understood by members of all racial groups)? Please explain.

8. In your opinion, do the questions reasonably represent perceptions of African American males held by the average person in the United States? If not, please explain in the space provided.

9. In this space, please provide any additional comments, concerns, or questions that may not have been addressed.

First Round of Revisions

During the first round of expert review panel (ERP) participation, a few revisions were made to the instrument to clear up confusion. Most critical was the section of the instrument that asked participants to drag specific phrases to two boxes that did or did not represent their personal views. The item had criteria that would not allow participants to continue if a certain number of items were not placed in each box. Additionally, the section that asked for personal 172

First Round of Revisions

During the first round of expert review panel (ERP) participation, a few revisions were made to the instrument to clear up confusion. Most critical was the section of the instrument that asked participants to drag specific phrases to two boxes that did or did not represent their personal views. The item had criteria that would not allow participants to continue if a certain number of items were not placed in each box. Additionally, the section that asked for personal perceptions of African American males were changed from requiring a response (e.g., forced response) to requesting one (e.g., response requested). I was concerned that if a participant was feeling uneasy about responding to an item that aroused negative feelings, a forced response might increase the likelihood that the survey would be discontinued, rendering data unusable.

Forced responses were required for fewer areas of the survey: (1) qualifying questions, (2) perceptions of African American males for the average person, and (3) demographic questions.

Changes to Wording for Specific Items

The click and drag question in reporting personal perceptions was changed to select all that apply. For the item regarding African American males likely to “be on welfare” was changed to “receive government assistance,” adding specific examples like social security, supplemental nutrition assistance program (SNAP), and temporary assistance for needy families

(TANF). This change was consistent with the literature regarding value systems of social and political conservatives (Hing et al., 2008).

Surveys should contain questions that are clear and concise (Creswell, 2012). In addition to changing the term “welfare,” there was some concern regarding the terms “shiftless” and the

“average person.” The term shiftless was a non-standard or unfamiliar term. African American males perceived as shiftless was changed to “not ambitious.” The use of a principal expert for the 173

ERP was necessary to serve as a safeguard to ensure that all recommendations and comments were considered with fidelity. The term “average person” raised some concern. For example, one member of the ERP commented that when they think of the average person, they are led to think of a “White, Anglo-Saxon person” (WASP). Another participant wrote:

I find that trying to determine what the “average person” thinks is as difficult as trying to

disprove the average American. The [average person] in California and New York are a

group unto themselves; [white people] in the south are completely different from white

people in the North. People in Wyoming are nothing like people in Massachusetts. I’m

wondering if you want me to answer with regard to the average person in my circle of

friends/acquaintances? My area of the country?

After consulting with my dissertation advisor, who in turn consulted with the methodologist of record for the present study, I made the decision to leave the term “average person” intact, thus leaving it up to study participants to make the determination. Foster (1995) interpreted the “average person” as the person completing his survey. This study interprets the average person as the participant with high social desirability scores (e.g. greater than 7), high perceptions scores for the average person, and low personal perception scores.

Second Round of Revisions

At the conclusion of the window for expert review panel member participation, revisions were made to the instrument addressing reasonable concerns after compiling all comments and suggestions in a report to the principal expert reviewer (PER). Along with the concerns, comments, and suggestions, I reported to the PER how I addressed each area. I scheduled a meeting with the PER at her office. Table 5.3 summarizes the report from comments from expert review panel members. 174

Table 5.3

Summary of Comments from Expert Review Panel Members

Measurement Extremely Moderately Area of Emphasis Yes No Reasonable Reasonable amount of time needed to complete the survey 12 3 clear directions 10 5 appropriate scales 12 3 clear and concise questions 12 3 confusing words in questions 0 15 culturally appropriate wording 15 0 questions reasonably represent perceptions of African American males held by the 12 3 average person in America

I scheduled a meeting with the PER, forwarding her a copy of the report prior. Also included in the report were the following questions to guide in the process of evaluating whether changes reflected the concerns of the expert review panel: (1) do changes made to the survey instrument represent the comments, concerns, and feedback of the expert review panel? (2) If not, are you as the principal reviewer satisfied with the rationale for not making certain changes

(e.g., not adding a neutral category to the rating scale and not defining the average person for study participants)? and (3) are there any concerns you have as the principal reviewer that I should consider before the survey instrument is disbursed for administration? 175

The PER asked another member of the expert panel to sit in on the meeting. The discussion was led by reviewing comments and suggestions from ERP members. The principal expert reviewer engaged in a brief dialogue about her concern with not having a neutral category, at which point an explanation was given about participants’ opting to avoid expressing honest beliefs by including it in the scale. The explanation provided was acceptable to her and she was in agreement that the survey, in her opinion, was ready for dissemination.

Final Revisions to the PEPAAMS PK12

I had a concern that the principal expert reviewer’s lens in Pre K12 education might be subject to blind spots (Peshkin, 1988); so as a safeguard to protect the integrity of the instrument,

I contacted another member of the expert review panel, whose work in higher education would serve as a safety net to ensure that no potential pitfalls were left unexamined (Milner, 2007). The expert panel member’s work as a full time adjunct professor with students pursuing PK12 education degrees at a public university in Northwest, Ohio, made her a prime candidate to review the instrument prior to distribution. Additionally, she had a particular familiarity with the present study as she read and provided feedback of initial versions of Chapters 1 and 2.

I forwarded the expert panel review member a finished copy of the instrument without any discussion regarding changes that were made up to that iteration. The survey was sent via e- mail asking her to provide any relevant feedback that she deemed appropriate. Based on her feedback, valid points were raised regarding the instrument. First, the order of items in the adapted brief social desirability scale might interact with scores. Secondly, four words (e.g., when compared to non-Black males) for items in perceptions of African American males for the average person and self-reporting personal perceptions could create cognitive diffusion, which 176

could activate confirmatory bias (Nickerson, 1988). Asking participants to compare African

American males to non-Black males could exhaust participants, particularly those who take items literally. For example, it might be feasible that some participants might try to process the question from each individual non-Black male group (e.g., Hispanics and Asians).

I deemed that the recommended changes would strengthen the instrument. For each item, participants were asked to either express how the average American views African American males, or express their personal views regarding African American males, excluding the words,

“when compared to non-Black males.” Finally, a few minor grammar changes were made (e.g., changing singular to plural) prior to distribution.

First Distribution of the PEPAAMS PK12

In the initial distribution of the PEPAAMS PK12, a group of e-mails were sent via

Qualtrics e-mail on a Friday morning to 198 participants. A small initial distribution was intended to catch any unforeseen problems that were missed in the validation phase. One problem was detected and fixed. The item that requested participants’ specific role in education proved to be too complex for a matrix style item. Qualtrics would not allow participants to limit their responses to a precise detail (see Figure 5.1). 177

Figure 5.1. An example of the complex item before it was simplified to increase response rates.

The item was simplified to a multiple choice question that listed general job parameters

(see Figure 5.2). The matrix item was intended to extrapolate the most information about individuals’ roles in education for future research. 178

Figure 5.2. Simplified item information regarding participants’ role in education.

Prior to the change, the item in Figure 5.2 was too complex for the format that attempted to extract more information in a single item. Forced responses would not allow participants to continue through the survey if they did not provide a response to each item. This can be problematic when individuals are uncomfortable with the subject matter. An example of participants not being comfortable with the subject regarding African American males was evident in the first distribution of n =198 participants. One respondent e-mailed, “No thank you,” regarding the invitation to participate. Another participant sent an e-mail describing that they discontinued the survey because they were required to respond to an item asking them to select all that apply to how the average person views African American males (e.g., low intelligence, not educated, not motivated, aggressive, threatening, and violent). She did not agree with any of 179

the choices. All listed choices were negative. I made the decision to remove the forced response feature to increase participation. Initially, two sections of the survey required a response (e.g., adapted brief social desirability scale and reporting perceptions of the average person).

Second Distribution of the PEPAAMS PK12

Foster (1995) reported that some participants demonstrated strong emotions (e.g., anger, resentment, and evasiveness). He cited that due to anger some participants refused to participate in his study of educators and non-educators perceptions of African American males. The present study was not without controversy. The second distribution of the survey was sent to 4000 teachers in Northwest Ohio two days after the first distribution that was meant to catch any unforeseen issues. However, due to several complaints made to the Office of Research

Compliance and my committee advisor, the study was closed so that modifications could be made. Before the close of the study, 172 people participated, but only 92 people completed the survey in its entirety. The study had to be closed because participants were not warned regarding sexually explicit content that occurred in two items. Additionally, the largest urban district in

Northwest Ohio objected because prior approval was not secured before the invitation to participate was distributed to teachers.

Before the survey was reopened, the following changes were made. First, the informed consent was updated to warn participants of sexually explicit content and material, and to advise them to complete the survey from a personal device as recommended by IRB. Additionally, two items were added to the survey prefacing the items with explicit sexual content. Respondents were given the opportunity to decide if they wanted to engage the explicit items or skip them. A 180

skip logic function was activated to skip the items with explicit content, allowing participants to continue with the remainder of the study.

Third and Final Distribution of the PEPAAMS PK12

Prior to the redistribution of the study in Qualtrics, I modified the instrument to exclude

2240 educators from the region’s largest urban school district. The remaining 1956 teachers were assigned random numbers and the list was randomly sorted. After randomly sorting the entire list; the list was split in two (e.g., n=1000 and n=956). The list containing the e-mails of 956

Northwest Ohio teachers was randomly sorted again, with 500 e-mails taken from the list. The list containing 1000 e-mails was randomly resorted and 100 e-mails were taken from that list.

The total number of surveys sent in the last distribution was 600. Participants were invited to participate in the study via email in Qualtrics. The final distribution only yielded 15 participants, so I made the decision to use data collected from the first two distributions for analyses. No data was analyzed in this distribution due to low participation.

Overall Response Rate and Useable Data

Response rates associated with this study were a concern. Readers should keep in mind that out of 172 participants that attempted the survey from the first and second distributions, only

92 provided usable data. However, some statistical analyses in IBM’s SPSS version 22 used n =

89 for data analyses (e.g., independent sample t-Test and MANOVA testing) after correcting for degrees of freedom.

Nearly half of all participants elected not to answer all questions related to personal perceptions. Scores were deleted for participants that did not answer any of the questions central to the present study (e.g., perceptions for the average person and personal perceptions). After 181

deleting scores for missing data, the total sample size was n = 89 when degrees of freedom were factored in. Summarily, only forty-seven percent of participants chose not to continue with the study. It was expected that once participants learned that the survey measured educator perceptions of African American males that many would opt out.

Demographics of the Sample Population

Results of the study indicated that the majority of participants who elected to participate were White. Previously, it was indicated that in the first and second distribution of the survey,

172 people began the survey, approximately 53% completed the entire survey, providing useable data for analyses. Four people did not select a racial and ethnicity category. Eighty six and sixth- tenths percent of participants were female. Seventy one and sixth-tenths percent of all participants have earned a master’s degree or higher.

All participants that agreed to complete the entire survey had at least 3 years of teaching experience. Forty-four and six-tenths percent (n = 41) had at least 20-30 years of experience.

Twenty-six and one-tenth percent (n =24) had 10 – 20 years of experience, while 18.5% (n = 17) had between 3 and 9 years of experience in education.

Role in Education and Type of School District Represented

Study results indicated that the majority of participants worked in an urban school district, 64.1% (n = 59). This is significant because relevant research indicated that African

American males are most likely to attend an urban school district (Foster, 1995; Ladson-Billings

& Tate, 1995). Approximately 30.1% (n = 28) worked in a suburban school district. Ninety-two and four-tenths percent of participants ( n = 85) worked in general education roles 4th to 12th grades. Fourteen participants indicated that they worked in special education roles. The data do 182

not support enough people in any one category in special education to extrapolate more detail.

For example, only one participant indicated that they worked in special education for

Emotional/Behavioral Disabilities (EBDs).

Domestic and Political Affiliations

The data indicate that Northwest Ohio educators might be representative of the average educator working in urban school districts (e.g., marital status, sexuality, and political affiliation). Seventy-seven and two-tenths percent of study participants (n = 71) indicated that they were in a traditional marriage, compared to 2.2% (n = 2) that indicated that they were in a domestic partnership. Eighty-seven percent (n = 80) indicated that they identified as heterosexual, compared to 5.4% (n = 5) who identified as LGBTQ+. Half of all participants identified holding their political affiliation with Democrats, 50% (n = 46).

Other Attributes of Study Participants

A plurality of study participants, 45.7% (n = 42), lived in suburban communities.

Seventeen percent lived in rural communities. The aforementioned information is important because 64.1% (n = 59) of study participants work in urban school districts, but 64.7% (n = 57) of them live outside of the communities in which they work.

Exploratory Factor Analyses

Exploratory factor analyses (EFA) were conducted to understand the underlying structure of the instrument used to measure Public Educator Perceptions of African American male

Students PK12 for the average person and personal perceptions. EFA was appropriate in this case because heretofore these items were presented as themes derived from a qualitative study and were developed for use in the present quantitative study (See Foster, 1995). Results of EFA 183

offered insight into the psychometric qualities of the PEPAAMS PK12 and allowed me to interpret subscale scores to examine conspicuous aspects of perceptions of African American male students.

Factor analysis is useful to understand the structure of a set of variables (e.g., perceptions of African American male students related to crime), and to reduce a data set to a manageable size while retaining important nuanced information (e.g., items that do not perform well on factor loadings; Field, 2009, p. 628).

I conducted Exploratory Factor Analysis using maximum likelihood estimation in SPSS

Statistics, version 22 (IBM Corp., 2016). It was expected that factors would correlate. Factor analysis using maximum likelihood estimation was used to explore factors and examine scales to reveal internal consistency. Maximum likelihood estimation chooses the parameters that maximize the probability of a factor emerging in loadings (Field, 2009). In addition to this, analysis of scale reliability determined whether the PEPAAMS PK12 can be interpreted consistently under various conditions (Field, 2009). The Cronbach a = .970, indicated strong internal consistency for the instrument for perceptions of the average person. Overall perceptions of African American males are negative. I felt that calculating Cronbach alpha scores was not necessary at this time given the limited sample size. An oblique rotated solution using the Promax method with Kaiser normalization (Field, 2009), revealed three factors with eigenvalues greater than one for perceptions of the average person. Inspection of the scree plot

(see Figure 5.3) revealed an inflection point at three factors, confirming the three-factor solution. 184

Figure 5.3. Scree plot of exploratory factor analysis for perceptions of the average person.

Each of the three factors identified in Figure 5.3 explained 63.7% of the total variance in participant scores before rotation. After rotation, the three factors combined explained 76% of total variance. Table 5.4 displays the means, standard deviations, pattern coefficients (which represent the unique relationship between the item and the underlying factor after controlling for the other factors), and communalities for each item in the PEPAAMS PK12.

Analysis of the internal consistency for each subscale using Cronbach’s alpha indicates reliabilities either above .90 are suitable for the purposes of this instrument (Lance, Butts, &

Michels, 2006). The overall scale that included all items in the instrument demonstrates adequate 185

internal consistency of α = .97. Internal consistency analyses for the overall scale, and for each subscale, indicated that dropping two of the items associated with sexual prowess and education would allow the instrument to perform better after factor extraction. One item relating to sexual prowess for personal and average perceptions were deleted because African American males being good lovers did not perform well on the pattern matrix as it did not correlate to any factor loadings. The assumption that being a good lover is a positive trait and that did not fit the pattern for how overall perceptions of African American males are negative. Finally, but less impactful, an item (i.e., African American males perceived as dishonest) was included in both criminal and education subcategories for the average person. Scores for dishonesty were not considered in the education category. Dishonesty scores were, however, considered in the category relating to criminality. Making the aforementioned modifications contributed to the overall scale demonstrating adequate internal consistency.

Factor Loadings for Perceptions for the Average Person

Table 5.4 lists three factors that resulted from the analysis of maximum likelihood estimation, which shows that participants viewed the average person to have a unidimensional view of African American males that is overtly negative. In fact, the most negative perceptions that participants attributed to the average person in factor one is clustered around attitudes and personalities that causes them to be viewed as threatening, aggressive, belligerent, and mean.

Additionally, factor 1 (negative attitudes) revealed that the average person perceives Black males as having prominent deviant tendencies (e.g., violent, rapacious, and abusive). 186

Table 5.4

Means, Standard Deviations, Pattern Coefficients, and Communalities for the Average Person on the PEPAAMS PK12 Items

Pattern Coefficient Item M SD h2 Factor 1 – Negative Attitudes Threatening 2.74 .651 .367 -.118 -.118 .849 Aggressive 2.78 .651 .404 -.115 -.135 .867 Belligerent 2.67 .656 .178 -.029 -.083 .716 Mean 2.49 .660 .164 -.073 -.036 .688 Abusive 2.43 .640 .121 -.082 .001 .619 Vain 2.30 .628 .074 -.034 -.014 .462 Not Intelligent 2.59 .689 .091 .014 -.055 .615 Rape Women 2.18 .703 .065 -.002 -.026 .525 Violent 2.65 .662 .075 -.047 .036 .695 Dishonest 2.60 .596 .069 -.065 .092 .707 Factor 2 - Personal Characteristics Not Educated 2.76 .678 -.169 .392 -.073 .829 Use Government Assistance 2.91 .655 -.071 .201 -.053 .695 Lazy 2.72 .710 -.135 .323 -.044 .811 Not Ambitious 2.77 .638 -.055 .260 -.070 .811 Irresponsible 2.80 .628 .014 .239 -.116 .815 Ignorant 2.86 .628 -.028 .117 -.015 .684 Not Motivated 2.64 .643 .041 .070 -.039 .686 Factor 3 - Criminality Sell Drugs 2.61 .596 -.246 -.384 .964 .942 Use Drugs 2.65 .644 -.155 -.008 .276 .823 Dishonest 2.60 .596 .069 -.065 .092 .707

Note: Bold indicates highest pattern coefficient and h2 = communalities.

Factor 2 (personal characteristics) indicated that the average person views African

American males as lacking intellectual acuity that is evidenced by perceptions of them being uneducated because they are lazy, unambitious, irresponsible, and look for handouts from government programs (e.g., social security, disability, food stamps, cash assistance, etc.). When analyzing results considering factors 1 and 2, participants indicated that the average person 187

Factor 2 (personal characteristics) indicated that the average person views African

American males as lacking intellectual acuity that is evidenced by perceptions of them being uneducated because they are lazy, unambitious, irresponsible, and look for handouts from government programs (e.g., social security, disability, food stamps, cash assistance, etc.). When analyzing results considering factors 1 and 2, participants indicated that the average person believes that African American males have innate (e.g., genetic/biological) factors that contribute to them not being able to attain education as evidenced by not completing high school.

Factor 3 (Criminality) was correlated to African American males being perceived as criminals. For example, the pattern matrix in Table 5.4 shows that the average person believes that African American males will resort to selling drugs and other criminal behavior that result from perceived genetic and biological tendencies mentioned from factor 1 loadings. Table 5.5 shows factor with items combined and grouped that lists their means and standard deviations.

Table 5.5

Means and Standard Deviations for each Factor Number of M SD Items in Factor Factor 1 – Negative Attitudes 2.56 .51 10 Factor 2 – Personal Characteristics 2.77 .57 7 Factor 3 – Criminality 2.61 .56 3

Although overall perceptions of African American males are negative, perceptions are most negative regarding personal characteristics. Public educators believed that the average person believed that African American males were likely to receive assistance from government programs as indicated by a mean score of 2.91. 188

Items on the Public Educator Perceptions of African American Males Survey PK12 were negatively worded. Higher scores represented more negative views of African American males held by the average person. For example, a participant that strongly agreed that the average person believes that African American males benefit from government assistance will be assigned a score of 4. If they agree, their score would be 3. Positive perceptions of African

American males should be closer to 1. Participants believed that the average person perceives

African American males to be somewhat negative in every category.

Factor Loadings for Personal Perceptions Held by Public Educators

The Kaiser-Meyer-Oikin Measure of Sampling Adequacy provided a score of .920, which indicated that the sample size was sufficient enough to conduct adequate factor analyses.

Initial Eigenvalues indicated that 75% of variance was attributed to factor 1, and 7% for factor 2.

Variance for extracted sums were also similar for personal perceptions held by public educators

(e.g., factor 1 accounted for 75% of variance).

An oblique rotated solution using the Promax method with Kaiser normalization (Field,

2009), revealed two factors with eigenvalues greater than one for personal perceptions.

Inspection of the scree plot (see Figure 5.4) revealed an inflection point at two factors, confirming the two- factor solution. 189

Figure 5.4. Scree plot of exploratory factor analysis for personal perceptions.

To evaluate the dependent variable associated with personal perceptions held by public educators, I also conducted EFA using maximum likelihood estimation. It was expected that factors related to personal perceptions of educators would correlate. I selected an oblique rotated solution using the Promax method with Kaiser normalization (Field, 2009).

Analysis of the internal consistency for each subscale using Cronbach’s α indicates reliabilities above .90 are suitable for the purposes of this instrument (Lance, Butts, & Michels,

2006). The overall scale that included all items for personal perceptions held by public educators 190

indicated adequate internal consistency of α = .95. Internal consistency analyses for the overall scale, and for each subscale, indicated that dropping five items associated with education, personality, and sexual prowess would allow the instrument to perform better under various conditions.

Two items relating to education were deleted because African American male students perceived as ignorant and not educated did not perform well on the pattern matrix as they had small loadings, which decreased loadings in otherwise strong factors. Additionally, African

American male students likely to rape or abuse women did not perform well and were subsequently removed. Finally, but less impactful, one item, African American males perceived as usurpers of government assistance programs, was excluded because it did not perform well in initial factor loadings for personal perceptions held by public educators.

Scores were deleted for participants that did not answer any of the questions that were central to the study to examine the dependent variable related to personal perceptions. After deleting scores for missing data, the total sample size was 89 for public educators’ personal perceptions of African American male students. Each identified factor contained the items included in each aspect of perceptions and jointly explained 82% of the total variance in participant scores before and after rotation. Table 5.6 displays the means, standard deviations, pattern coefficients (which represent the unique relationship between the item and the underlying factor after controlling for the other factors), and communalities for each item in the PEPAAMS

PK12. 191

Table 5.6

Means, Standard Deviations, Pattern Coefficients, and Communalities of Personal Perceptions for the PEPAAMS PK12

Pattern Coefficient Item M SD h2 Factor 1- Personal Characteristics Mean 1.60 .578 .101 -.273 .863 Abusive 1.56 .602 .138 -.291 .896 Threatening 1.67 .687 .112 -.199 .872 Not Intelligent 1.52 .660 .041 -.047 .687 Vain 1.61 .596 .059 -.097 .771 Violent 1.64 .606 .102 -.103 .859 Belligerent 1.69 .650 .060 -.098 .774 Not Ambitious 1.63 .646 .076 -.072 .815 Aggressive 1.80 .726 .070 -.038 .801 Lazy 1.69 .717 .044 -.023 .702 Irresponsible 1.73 .653 .043 .029 .698 Steal 1.67 .653 .070 -202 .835 Factor 2 - Criminality Criminal Behavior 1.82 .732 .107 .717 .888 Sells Drugs 1.78 .735 .060 .337 .801 Note: Bold indicates highest pattern coefficient and h2 = communality. Cronbach’s a = 0.95

Pattern Matrix for Factor Loadings for Personal Perceptions of Public Educators

Factor loadings on the pattern matrix for personal perceptions of public educators lists 2 factors that resulted from an analysis of maximum likelihood estimation. The matrix shows that the strongest residuals cluster around negative personalities wherein African American male students are perceived as threatening and violent.

Factor 2 loadings were less conspicuous, with relatively weak residuals. In factor 2, only two items related to criminality (e.g., criminal behavior and selling drugs) appeared in the factor.

The residual patterns indicated that all items that comprise factor 1 were strong. Interestingly, the two items comprising factor 2 also loaded in factor one with similar strength. I chose to retain the 192

two aforementioned items in factor two separately to use as a reference point for future distributions of the PEPAAMS PK12.

Research Question 1: Relationship Between (a) Perceptions for the Average Person, and (b) Self-Report Scores

Research question 1 sought to discover if there was a relationship between (a) perceptions for the average person, and (b) self-report scores among public educators of African

American male students. A paired samples t-Test indicated that there was a statistically significant difference between the two dependent variables. As indicated by t(87)= 13.25 , p =

<.05.

Descriptive Statistics Results: Means and Frequencies

Items on the Public Educator Perceptions of African American Males PK12 were negatively worded. Higher scores represented more negative views of African American male students held by public educators. Scores for the average person showed that participants believed that the average person perceived African American males more negatively than do themselves, the public educators. Mean scores indicated that public educators claim to not agree with negative perceptions of African American male students. For example, when asked if public educators viewed African American males students as not being motivated, a mean score of 2.33 indicated that they disagree that Black male students lack motivation. Interestingly, public educators reported that the average person agreed with that perception as represented by a mean score of 2.64. 193

Table 5.7 shows that the mean score for perceptions of the average person were higher than personal perceptions of public educators. On the surface, it appears that public educator’s personal perceptions were less negative than the average person.

Table 5.7

Descriptive Statistics for Variables

Minimum Maximum Standard Dependent Variables N SS Mean Score Score Deviation Average Person 89 1.00 4.00 229.90 2.61 .55 Perceptions Personal Perceptions 88 1.00 3.24 151.59 1.70 .52

Independent Variable

Social Desirability Bias 89 5.00 9.00 628.00 7.06 1.02 Note: Scores were rounded to the nearest hundredth

The PEPAAMS PK12 included two sections that required participants to give overall perceptions of African American males for the average person and report personal perceptions.

Participants were asked to select all that apply to how the they and the average person perceive

African American males overall. Table 5.8 provides a summary of data that lists overall perceptions of African American males. 194

Table 5.8

Overall Perceptions of African American Male Students by Percent and Frequency

Category Average Person Personal Perceptions % ƒ % ƒ low intelligence 41.3 38 2.2 2 aggressive 53.3 49 10.9 10 threatening 58.7 54 * * violent 40.2 37 * * belligerent * * 2.2 2 mean * * 2.2 2 abusive * * 2.2 2

*not selected by study participants for a given category

The table above indicates that participants believed that overall perceptions of the average person were negative with regard to characteristics that could cause African American males to be viewed as a threat. Data also indicated that personally, educators overall perceive

African American males as having lower intelligence 41.3%, n = 38, and lack motivation to achieve. Additionally, 10.9%, n = 10 personally reported that they view African American males as being aggressive.

Inferential Statistics Results

To determine if the statistical test was appropriate to calculate differences, skew and kurtosis levels were evaluated to determine if they fell within the acceptable values for a t-Test

(e.g., skew < |2.0| and kurtosis < |9.0|; Posten, 1984). The values for skew and kurtosis were .451 195

and -.506, respectively. The assumption was considered and satisfied, t(87) = 13.25, p < .001.

Thus, it is concluded that public educators were less likely to express personal views of African

American male students negatively compared to self-reporting for the average person. Cohen’s d was estimated at 1.68 which is a large effect based on guidelines set by Cohen (1992).

Summarily, I can be relatively certain that the significance between the two dependent variables can be used to investigate the second research question.

When considering if there is a statistical difference between the two dependent variables, a paired samples t-Test indicated that public school educators may have more negative perceptions of African American male students but are not likely to disclose them. Specifically, there was a significant difference in the scores for average person (M = 2.61, SD = .52), and personal perceptions (M = 1.71, SD = .55); t(87)= 13.25 , p = <.05. The reason for statistical differences could not be confirmed by the ABSDS, which was included to mitigate measurement error and detect social desirability bias.

Scores on the ABSDS did not indicate participants’ likelihood to express social desirability bias when asked about personal perceptions regarding African American male students. A score of 5 (the lowest possible score) indicated that a participant is not likely to reflect social desirability bias. Any score higher than 7, with ten being the highest possible score), satisfied the criterion for using the scores of the average person to express the personal perceptions of public school educators in Northwest Ohio. Furthermore, the lack of participation of public educators when given the opportunity to participate in the study is consistent with the literature regarding non-participation bias when faced with a topic that causes emotional or 196

cognitive discomfort (Nicholls, Chapman, Shaw, Perkins, Sullivan, Crutchfield, & Reed, 2011;

Olson, 2006).

Research Question 2: Differences Between (a) Perceptions Scores for the Average Person

and (b) Self-Report Perceptions Scores for Scores on the ABSDS

Research question 2 sought to understand whether differences in the relationships between (a) perceptions scores for the average person and (b) self-report perceptions scores for participants were due to different scores on the adapted brief social desirability scale (ABSDS).

To test research question 2, a one-way MANOVA indicated that different social desirability scores did not have any statistical significance on the dependent variables, which rejected the belief that social desirability scores would help me determine whether it was more reliable to use responses for the average person above educators’ personal responses. One way of analyzing results is to assume that public educators’ personal perceptions are less negative than the reported perceptions of the average person. At the outset of this study, I wanted to differentiate between three groups of educators using the ABSDSs impact on perception scores: (1) educators who are not honest about their negative perceptions (i.e., high scores on the ABSDS/high scores for negative personal perceptions), (2) educators who have negative perceptions, but are honest about them (i.e., low scores on the ABSDS/ high scores for personal negative perceptions), and

(3) educators whose perceptions are not negative, as reflected by low social desirability scores

(i.e., low scores on the ABSDS/low personal perception scores).

Testing Assumptions for MANOVA

Multivariate Analysis of Variance (MANOVA) creates a linear combination between the two dependent variables to determine if there are any differences across the levels of the 197

independent variable on that linear combination of the two dependent variables. Simply put, I wanted to discover if different scores on the ABSDS had any relationship to perceptions of the average person and personal perceptions. Before this determination could be made, I needed to make sure that assumptions of normality were satisfactory for the two dependent variables (i.e., personal perceptions and personal perceptions held by public educators).

In order to test assumptions for outliers, I performed a linear regression to examine

Mahalanobis distance, which has a standard critical value of 13.82. Anything above the critical value would indicate that there are outliers, which would reject assumptions for normality, rendering MANOVA testing inadequate to evaluate differences in scores on the ABSDS in relation to perception scores. Additionally, the Shapiro-Wilk test indicated a significance score of .148 for perceptions of the average person, indicating normality. However, for personal perceptions held by public educators, the Shapiro-Wilk test indicated significance less than .05, which rejects assumptions of normality for the dependent variable pertaining to personal perceptions held by public educators. A significance score less than .05 suggested that personal perceptions is not a normally distributed variable.

Statistical significance in the Shapiro-Wilk test necessitated further testing for the assumption of multicollinearity using bivariate correlation analysis. The Pearson Correlation allowed me to determine that both dependent variables correlated adequately enough to proceed with MANOVA testing as indicated by an r value of .290. Any score with an r value > .2 is acceptable for testing the assumption of multicollinearity, r (86) = .29, p = .006 (Castillo, et al.,

2016; Mertler & Vannatta, 2013). 198

Moderate correlation for both dependent variables, suggested the appropriateness of a

MANOVA. Alternately, the Box’s M value of 5.867 was associated with a p-value of .947, which was interpreted as non-significant based on Huberty and Petoskeys’s (2000) guideline (p <

.005). Thus, I concluded that matrices between groups were assumed to be equal for the purposes of MANOVA testing.

Results from MANOVA Testing

A one-way multivariate analysis of variance (MANOVA) was conducted to test the research question that different social desirability scores can be used to determine which set of perceptions scores are most reliable to use. Results indicated that there was no significant

MANOVA effect as evidenced by Pillais’ Trace = .081, F(8, 166) + .871, p > .001. The multivariate effect size was estimated at .040, which implies that .004% of the variance in the canonically derived dependent variable was accounted for by social desirability bias.

Results from research question 2 determined that no significant difference existed between the two dependent variables (personal perceptions and perceptions for the average person) when considered jointly with the independent variable (social desirability scores),

F(8,78) + .360, p = .229, h2 = .122.

For perceptions of the average person, whose overall mean score was significantly higher than personal perceptions, there was no statistical significance when scores were different on the

ABSDS, thus concluding that any expansions to the present study would proceed without the adapted brief social desirability scale as it had no bearing to inform decision making whether it was more reliable to use one set of perceptions scores over the other. 199

Descriptive Statistics for Results on the ABSDS

As a reminder, Foster (1995) asked participants to reflect perceptions of Black males based on the average person. The research questions seek to answer whether it is feasible to use self-report scores for the average person or personal perceptions as reported by participants of

African American males. For the Adapted Brief Social Desirability Scale (ABSDS), scores greater than 7 were considered to demonstrate social desirability bias. Table 5.9 shows the means and standard deviations for each item contained in the ABSDS. The closer scores are to 1 indicates lack of social desirability bias.

Table 5.9

Mean and Standard Deviations for Items on the ABSDS

M SD Do you always practice what 1.42 .496 you preach? Do you always admit when 1.36 .483 you make a mistake Do you always keep your 1.34 .475 promises? Do you smile at every student you come in contact with 1.46 .501 during the school day? Have you ever laughed or made fun of a student you did 1.48 .503 not like behind their back?

The adapted brief social desirability scale (ABSDS) informed interpretation and use of scores for perceptions for the average person and personal perceptions of African American males. Mean scores for average person and personal perceptions were compared to determine if there are statistical differences. Differences in perceptions for the average person and personal 200

perceptions were compared to scores on the ABSDS and were used to determine which data is more reliable for use to draw conclusions and make inferences regarding the sample population.

Scores greater than 7 on the ABSDS would have resulted in use of scores for the average person.

Because there were no statistical differences in (a) perceptions of the average person and (2) self- report of personal perceptions of African American males, when considered jointly with (c) high scores (scores greater than 7) on the ABSDS, scores that described personally held views were considered as an alternative to determine which set of scores will be used (scores for the average person or personal perceptions) to draw conclusions regarding the data. Later in this Chapter, readers will be informed how personally held views were used to interpret educator perceptions of African American males.

The ABSDS was computed into one variable by summing the scores. Low scores (5 or less) indicated a participant’s unlikely disposition to display social desirability bias. High scores

(8 or greater) indicated a participants’ disposition to answer in a way to appear to be good to others, this is known as social desirability bias. Table 5.10 provides descriptive statistics for both dependent variables (personal perceptions held by public educators and the average person) and the independent variable (the Adapted Brief Social Desirability Scale). 201

Table 5.10

Mean Differences, Standard Error, and Significance

ABSDS Mean Differences by Score Standard p 5 6 7 8 9 Error Value DVs (Perceptions) Average Person .264 -.264 -.402 -.117 -.349 .235 >.05

Personal Perceptions .154 -.154 -.148 -.079 .079 .255 >.05

Statistical significance is represented by a p Value of <.005

A MANOVA was used to examine relationships between negative perceptions of African

American male students, specifically, how scores can be interpreted when considering social desirability bias based on the differences in personal perceptions held by public educators and their reporting of perceptions of the average person. There was no MANOVA effect between both dependent variables and the independent variable. More specifically, the ABSDS failed to predict whether participants would express their personal views of African American males candidly, and whether the ABSDS was a reliable predictor of personal perception scores held by public educators.

I concluded that the adapted brief social desirability scale, which served as the independent variable for the present study does not adequately describe the differences in relationships and differences between the 2 dependent variables and the independent variable.

Overall analysis of descriptive data indicates that study participants might have been more comfortable reporting overall perceptions of African American males for the average person over 202

their personal perceptions. One item asked participants about their comfort level completing the survey, as well as their willingness to be contacted by the researcher to complete an interview.

The majority reported being very comfortable or comfortable in completing the survey. Only

21.4% reported that they were very uncomfortable or uncomfortable completing the survey.

Seventy-four percent reported they were comfortable or very comfortable completing the survey.

When asked about having a willingness to be contacted by the researcher for a personal interview regarding the study, 43.3% said they would be willing to participate in a follow-up interview with the researcher.

Alternate Analysis Using Personal Value Statements: Ideological Values

Personal values reflect ideological attitudes not measured in the ABSDS and perception scales of African American males. The majority (9 items) on this scale describe negative attitudes that are likely to result in discrimination or marginalization of outgroup members (see

Hing, et al., 2008; Sears & Henry, 2003; Sniderman, Brody, & Kuklinski, 1984; Sniderman &

Tetlock, 1986; Wood, 1994; Zaller, 1992). Any combination of responses for personal value statements (ideological values) where responses favor social/political conservatism, and aversive racism (see Hing, et al., 2008) will result in the use of average perceptions scores to reflect the views of public educators. Later in this Chapter, I will discuss how personal value statements are interpreted using MANOVA, bivariate correlation, and simple linear regression equations to interpret data.

Ideological Values

Ideological values were important to the present study because research indicated that educators

are more likely to express views related to egalitarianism (i.e., equality for all races); 203

however, when faced with the opportunities to promote them in professional practice, these values are not always apparent (Hing, et al., 2008, Quinn, 2017). Table 5.11 separates personal values into the appropriate ideological category.

Table 5.11

Ideological Items from the PEPAAMS PK12 (Hing, et al., 2008)

Egalitarianism Aversive Racism Social/Political Conservatism racism not a bad a portrayed programs to help minorities politically liberal on issues in media not needed

politically liberal on issues should not sit/kneel during the support equality for all racial pledge groups some races receive more Bible infallible word of God races should have equal access gov’t support oppose American values threat to Democracy conservative on political issues tax dollars should not be used for welfare obedience and respect should be taught in schools

Table 5.12 describes political values for the item in the PEPAAM PK12, which asked participants to select all that apply to personally held views. This information is critical to the argument that educators who have negative perceptions of African American males students should have a higher number of items that subscribe to social/political conservatism and aversive racism. The following delineate the twelve items that comprise the scale of social and political values: (1) five items subscribe to social/political conservatism and aversive racism, (2) two items describe political and social conservatism, (3) two items describe social conservatism, and

(4) three items describe egalitarianism (Dovidio et al., 2002; Hing et al., 2008). 204

Table 5.12

Definitions for Ideological Views for the PEPAAMS PK12

Code Description Political Values (Hing, et al., 2008) programs not needed to help minorities 1 political/social conservativism (Not TLP) leveling playing field racism is not a big issue as portrayed in political/social conservativism & aversive 2 media racism (Not TLP) Americans should not sit/kneel during political/social conservativism & aversive 3 anthem and pledge racism (Not TLP) Bible is the infallible, inspired word of 4 social conservatism (Not TLP) God opposition to traditional American values a political/social conservativism & aversive 5 threat to our democracy racism (Not TLP) children should learn respect and 6 social conservatism (Not TLP) obedience in school 7 liberal on many political issues egalitarianism (TLP)

8 conservative on political issues political/social conservatism (Not TLP)

9 equality for all racial and ethnic groups egalitarianism (TLP) all races should have equal opportunities in 10 egalitarianism (TLP) the U.S. some races receive more government political/social conservativism & aversive 11 support than others racism (Not TLP) tax dollars should not be used for political/social conservativism & aversive 12 cash/food assistance for families on racism (Not TLP) welfare

Note: Truly Low Prejudice (TLP) 205

The data indicated that 34.8% of participants believed that Americans should not sit or kneel during the National Anthem or pledge of allegiance, a value indicative of both social and political conservatism. Twenty-one and seven tenths percent believe the that Bible is the infallible, inspired word of God. This is also a characteristic of social and political conservatism.

Approximately 57% of participants believed that children should learn obedience and respect in school. This is indicative of social conservatism. The data also show that 38% believed that some races receive more government support than others, a politically conservative value.

Data from the PEPAAMS PK12 show that 88% of participants, n=81 believe that all races should have equal opportunities in the United States. Additionally, 85.9%, n = 79, support equality for all racial and ethnic groups. It can be concluded that principles and beliefs that inform behavior may play out differently in public schools for African American males. Table

5.13 shows how participants selected value statements that corresponded to ideological beliefs. 206

Table 5.13

Personal Value Selections, Ideological Categories, Percentages, and Frequencies

Personal Value Selections Ideology Percent Frequency

Racism is not a big of an issue as AR, PC, SC, MR 7.6% n = 7 portrayed in media

Programs to help minorities are not SC, PC 5.4% n = 5 needed to level the playing field Americans should not sit/kneel SC, PC 34.8% n = 32 during the anthem or pledge of allegiance The Bible is the infallible, inspired SC 21.7% n = 20 word of God Opposition to traditional American SC, PC 5.4% n = 5 values is a threat to democracy Obedience and respect are important SC, PC 56.5% n = 52 values to be taught in school Conservative on many political SC, PC, 20.7% n = 19 issues Liberal on many political issues AR, TLP 51.1% n = 47 I support equality for all racial and TLP 85.9% n = 79 ethnic groups All races should have equal TLP 88% n = 81 opportunity in the United States Some races receive more AR, SC, PC, 38% n = 35 government support than do others Tax dollars should not be used for PC, SC 7.6% n = 7 cash and food assistance for families on welfare Key for Ideological Personal Value Statements: PC – politically conservative, SC – Socially Conservative, AR – Aversive Racist, TLP – Truly Low Prejudice. 207

Alternate Analyses to Examine the Relationship and Differences Between Both Dependent

Variables and the Independent Variable

An underlying question left unanswered as a result of non-significance of the ABSDS, was which dependent variable best represents participants’ true perceptions of African American male students. When constructing the Public Educator Perceptions of African American Males

Survey (PEPAAMS) PK12, it was anticipated that low response rates for personal perceptions may impact data interpretation. The PEPAAMS PK12 included a section containing 12 individual statements. Study participants were invited to select all items that applied to their personal values. These items were taken from Hing. (2008), which categorized ideological values into prejudice profiles that predicted the likelihood of discrimination and bias.

There were 12 personal value statements that participants could have selected. They could have selected as few as 1 and as many as 12. Nine of the 12 statements indicated ideological values that could lead to discrimination and bias. Only 3 of the 12 statements expressed individual views that would not likely lead to discrimination. Hing et al. (2008) characterized individuals who espouse these beliefs as truly low prejudice, thus not likely to discriminate or show bias. As such, people that are TLP are least likely to have negative perceptions of African

American male students.

When replacing personal value statements as the independent variable, a significant

MANOVA effect was discovered. Both dependent variables were deemed to be significant when considered jointly with the new independent variable (personal value statements), F(8,78) + .784, p = .005, h2 = .240. To be sure that this assumption was correct, post hoc testing revealed 208

significant MANOVA effect as evidenced by Pillais’ Trace = .360, F(16, 156) + 2.141, p < .01 for both dependent variables. The multivariate effect size was estimated at .180, which implies that 18% of the variance in the canonically derived dependent variables was accounted for by ideological views.

I wanted to see how scores held up with regard to participants who were least likely to discriminate. Once again the independent variable, which consisted of the three items that comprised TLP, replaced personal value statements (ideological views) taken from the

PEPAAMS PK12. Results indicated significant MANOVA effect as evidenced by Pillais’ Trace

= .294, F(4, 162) + 6.990, p < .01 for both dependent variables (average and personal perceptions). The multivariate effect size was estimated at .147, which implies that 14% of the variance in the canonically derived dependent variables was accounted for by TLP views.

Truly low prejudice people have egalitarian values and are least likely to discriminate. I wanted to examine if TLP characteristics are a likely predictor of less negative perceptions of

African American males. A simple linear regression was calculated to predict personal perceptions of African American males based on TLP personal value statements, b = -.363, t(84)

= 11.08, p <. 001, with an R2 value of 1.32. A significant regression equation was found F (1,

84) = 12.584, p <.000. Truly low prejudice educators are predicted to have more favorable perceptions of African American males compared to the average person as represented by personal perception scores from the PEPAAMS PK12.

Overall results from Tests of Between-Subjects showed significance for personal values of participants when considered jointly on the perceptions of the average person, F(2,81) + 4.66, p = .01, h2 = .103. Additionally, there was also a statistically significant MANOVA effect 209

between TLP scores for personal perceptions held by public educators, F(2,81) + 7.47, p = .001,

η2 = .156. Furthermore, educators that are TLP are likely to have positive perceptions of African

American males as indicated by the differences in personal perception scores. In like manner, educators with personal value statements that express ideological views that are likely to discriminate (not TLP) will have negative perceptions of African American males as expressed by means scores for the average person. I conclude that it is appropriate to use personal perception scores for educators who are TLP. Alternatively, I conclude that perceptions for the average person are more appropriate to use for educators who are not TLP.

If people that are not TLP are more likely to discriminate or show bias, I wanted to understand how the 9 items that expressed ideologies that were not TLP would function when compared to perceptions of the average person. Heretofore, results indicated that people that are

TLP will have more favorable perceptions of African American males, as indicated by personal perception scores whose mean was significantly lower than perceptions of the average person. I attempted to determine if non TLP personal value statements were a predictor of perceptions of the average person. Results from a simple linear regression equation showed significance; however, a simple linear equation failed to provide additional statistical information to explain the significance. In order to understand why, I conducted a bivariate analysis using the Pearson correlation to determine how non TLP personal values correlated with perceptions of the average person, which I believed to be the personal perceptions of educators who are not TLP. Results from the Pearson correlation r (68) = -.037, p = .767, did not explain the significance in the linear regression equation due to non TLP not being equally distributed as an independent variable. 210

Personal value statements showed significant MANOVA effect when considered jointly on both dependent variables. Failure of bivariate correlation analysis, in my opinion, did not nullify the significance of MANOVA effect. I wanted to examine the items that surfaced in the research regarding symbolic racism that could explain significant interaction with negative perceptions of African American males. To get a baseline understanding if non TLP values can predict negative perceptions of African American males, I selected one item from the list of 9 that expressed non TLP values. I wanted to see if participants that believed that racism in the United

States is not as big of an issue as portrayed in the media predicted negative perceptions of African

American males. A simple linear regression was calculated to predict personal perceptions of

African American males based on one TLP personal value statement regarding the severity of racism in the United States. Again, significant effect was discovered; however, non TLP values as a variable, was abnormally distributed which explained why bivariate analysis did not show correlation. Based on lack of normality, appropriate statistics could not be calculated.

MANOVA testing indicated that there was no significant MANOVA effect for participant perception scores for the average person), F(8,78) + .360, p = .229, h2 = .122 whose mean was significantly higher than personal perceptions when testing non TLP personal values. Simply put, personal values that are not TLP did not indicate MANOVA effect on perceptions for the average person. Therefore, I conclude that educators that are not TLP have more negative perceptions of

African American male students. Conversely, there was a significant difference between ideological scores for educators who were not TLP related to personal perceptions held by public educators, F(8,78) + .784, p = .005, h2 = .240. I concluded that perceptions for educators who are not TLP are not best represented by personal perception scores for this study. 211

Table 5.14 lists mean scores and standard error based on results reported by public educators, which expresses their perceptions for the average person, as well as personal views of African American male students. The list show means scores (based on a 4-point scale) from greatest to least. Additionally, results indicate that negative self-report perceptions are higher for the average person compared to personal perceptions held by public educators. The statement listed in the table represent personal views that are indicative of educators’ ideological beliefs. Furthermore, results indicated that public educators are more likely than not to reflect their true perceptions of African American male students through perceptions of the average person as reflected by significant differences in mean scores (e.g., t(87)= 13.25, p

= <.05). Ideological views reflect a likelihood that public educators could be prone to discriminate or show bias toward African American male students as evidenced by negative perceptions of African American males. 212

Table 5.14

Descriptive Statistics, Means and Standard Deviations/Error for Ideological Values

Items that are most Likely to Mean Standard Error Discriminate or Show Bias Average Personal Average Personal Person Perception Person Perception Programs to help minorities not needed 4.00 3.06 * * to level the playing field Racism is not as big of an issue as 2.61 2.00 .184 .197 portrayed in media Americans should not sit or kneel during the Pledge of Allegiance or 2.69 1.36 .134 .078 National Anthem The Bible is the infallible, inspired 2.59 1.94 .099 .088 word of God People who oppose traditional American values are a threat to 2.57 1.63 .111 .180 democracy Obedience and respect are important 2.50 1.75 .168 .212 values that should be taught in schools I am politically conservative on many 2.72 1.70 .211 .161 political issues Some races receive more government 2.54 1.92 .249 .231 support than others Tax dollars should not be used for cash and food assistance for families on 2.07 1.53 .167 .529 welfare Items that are most Likely to Show Standard Deviation Truly Low Prejudice I am liberal on many political issues 2.78 2.35 .693 .550 I support equality for all racial and 2.42 1.73 .508 .537 ethnic groups All races should have equal 2.76 1.56 .469 .482 opportunities in the United States *indicates that standard error was not calculated 213

Summary

This chapter reported the results of exploratory factor analysis using the maximum likelihood method, which indicated that perceptions for the average person loaded with three factors (negative attitudes, personal characteristics, and criminality). Using the same statistical method to analyze personal perceptions, EFA showed that negative perceptions of African

American male students loaded with two factors (personal characteristics and criminality).

Overall factor analysis revealed adequate internal consistency for average and personal perceptions. The items that did not load well were removed to strengthen loadings for items that were consistently strong before and after extraction.

When evaluating research questions 1 and 2, results revealed that there was a statistical difference between reported average and personal perceptions of African American male students. Notwithstanding, attempting to use the ABSDS as the independent variable did not explain the significance between both dependent variables using multivariate analyses of variance (MANOVA). Further investigation revealed that the statistical difference between both dependent variables could be explained my examining personal values statements, which comprise educators’ ideological and political values. 214

CHAPTER 6. DISCUSSION, CONCLUSIONS, AND RECOMMENDATIONS

This chapter will discuss the results of research questions 1 and 2. Additionally, I will use what I learned from data analyses to discuss implications for leadership, policy, and practice within public education in the United States. Also, we will revisit Milner’s (2007) framework introduced in Chapter 1, using systems theory to provide missing links that research findings cannot answer. Systems theory can offer remedies to explain social construction of race processes within systems that shift between tight and loose coupling interposing dominate hegemony over racial minority groups, which results in controlled narratives for how issues related to race, and gender are interpreted. Finally, discussion of future research will serve as a guide for distributions and modifications for an expanded study with a larger and more diverse sample size.

Results indicated that the ABSDS did not function as intended. The discussion in this chapter will emphasize what could be done differently for it to function properly, or whether it should be eradicated altogether in future distributions of the PEPAAMS PK12. There were statistical differences between personal perceptions and perceptions for the average person reported by public educators in the sample population. This chapter will unpack possible reasons for the differences and how future distributions will proceed based on results from this study.

Differences in Personal Perceptions and the Average Person

Mean scores for perceptions of the average person and personal perceptions were statistically different. Results from an independent samples t-Test calculated the differences as t(87) = 13.25, p < .001. The overall mean score for perceptions for the average person was M =

2.61. Overall mean scores for personal perceptions was M = 1.70. I am confident that differences 215

in means scores are attributed to educators feeling less comfortable with expressing personal perceptions when asked directly. Conversely, I am equally as confident that asking educators to report perceptions for the average person is directly linked to their personal perceptions, when accompanied by a component (ideological views) that can differentiate between educators that are least likely to have negative perceptions of African American male students from those that are not.

Foster (1995) asked both educators and non-educators to list perceptions of African

American males for the average person. His findings concluded that perceptions of the average person were the personal perceptions of educators and non-educators alike. I questioned whether the results would be different if he simply asked for their personal perceptions. Based on differences in mean scores, I agree that Foster drew appropriate conclusions regarding educators’ perceptions of African American males in the absence of a scale to detect bias.

Scores for personal perceptions and perceptions of the average person could have ranged from 4 to 1. Four represented that participants strongly agreed with negative perceptions. Three corresponded with agreement with regarding negative perceptions. A score of 2 indicated that educators disagreed with negative perceptions, followed by a score of 1 that indicated strong disagreement with negative perceptions in favor of African American males. Differences in mean scores for personal and average person perceptions, coupled with the failure of the ABSDS to explain differences in scores, resulted in further examination into which set of perceptions adequately represents educators’ personal perceptions of African American male students. 216

Negative Perceptions of African American Male Students: Aggressive and Threatening

Results from this study indicated that public educators reported negative perceptions of

African American males were negative related to aggressiveness, M = 2.78 for the average person. This is consistent with the literature regarding how African American males are constantly viewed as a threat because they are aggressive (Foster, 1995; Wilson, Hugenberg, &

Rule, 2017). Aggressiveness is often equated with how severe Black males are disciplined in school. The consequences of African American males perceived as aggressive could result in harsher punishment for discipline infractions and referrals to special education for emotional and behavioral disorders (Fitzgerald, 2009; Foster, 1995; Oelrich, 2012; Simson, 2008).

Perceived aggressiveness is not a lone negative trait attributed to Black males, this study also revealed that Black males are viewed as a threat, M = 2.74 for the average person. This is especially concerning because educators might be prone to distort perfectly normal behaviors as aggressive or threatening (Hing et al, 2008; Simson, 2013). African American males are most likely to attend inner city schools (Beckford, 2016). Relevant research indicated that the presence of law enforcement in inner city schools are more likely to result in African American males being inducted into the criminal justice system (Beckford, 2016; Noguera, 2012). Wilson,

Hugenberg, & Rule (2017) concluded that negative perceptions and judgment almost always ends with law enforcement justifying the use of force against Black males. However, when

African American males are viewed as a threat to educators, who according to this study might be a White female, fear is the motivating factor that accompanies negative perceptions (Wilson,

Hugenberg, & Rule, 2017). This assertion is supported through related literature that White females are less satisfied with their jobs when they are teaching in schools comprised of minority 217

outgroups (i.e., African American males; Fairchild, Tobias, Corcoran, Djukic, Kovner, &

Noguera, 2012).

In addition to Black males perceived as aggressive and threatening, this study found that educators perceived African American males to be belligerent and violent. These findings suggest that misinterpretations and cultural misunderstandings might increase the likelihood that

Black males might be subject to unnecessary isolation and marginalization in school (Fitzgerald,

2009; Kunjufu, 2005; Simson, 2013). One way this might occur is in referrals for special education and screening for psychotropic drugs (Fitzgerald, 2009). Extant literature revealed that school teams charged with making referrals for African American males for special education eligibility, distorted behavioral observations. For example, researchers found that when Black males questioned judgment of adults, they were often viewed as belligerent, defiant, and non- compliant. Furthermore, these observations were not applied uniformly with other demographic groups (Fitzgerald, 2009; Simson, 2013). Table 6.1 allows readers to see how factors loaded for both educators’ personal perceptions and perceptions of the average person that were presented separately in Chapter 5. Additionally, the table shows mean scores that reflect the degree of negative perceptions. Finally, the information contained in the table will help readers follow the discussion and impact of perceptions on future research. 218

Table 6.1

Corresponding Factors for Perceptions for the Average Person and Personal Perceptions

Pattern Pattern Coefficient Coefficient Item M SD h2 M SD h2 Factor 1 – Negative Attitudes Factor 1 – Negative Attitudes Average Person Personal Perceptions Threatening 2.74 .651 .367 -.118 -.118 .849 1.67 .687 .112 -.199 .872 Aggressive 2.78 .651 .404 -.115 -.135 .867 1.80 .726 .070 1.80 .726 Belligerent 2.67 .656 .178 -.029 -.083 .716 1.69 .650 .060 -.098 .774 Mean 2.49 .660 .164 -.073 -.036 .688 1.60 .578 .101 -.273 .863 Abusive 2.43 .640 .121 -.082 .001 .619 1.52 .660 .041 -.047 .687 Vain 2.30 .628 .074 -.034 -.014 .462 1.61 .596 .059 -.097 .771 Not Intelligent 2.59 .689 .091 .014 -.055 .615 1.52 .660 .041 -.047 .687 Rape Women 2.18 .703 .065 -.002 -.026 .525 Violent 2.65 .662 .075 -.047 .036 .695 1.64 .606 .102 -.103 .859 Dishonest 2.60 .596 .069 -.065 .092 .707 Factor 2 - Personal Characteristics Not Educated 2.76 .678 -.169 .392 -.073 .829 Use 2.91 .655 Government -.071 .201 -.053 .695 Assistance Lazy 2.72 .710 -.135 .323 -.044 .811 1.69 .717 .044 -.023 .702 Not Ambitious 2.77 .638 -.055 .260 -.070 .811 Irresponsible 2.80 .628 .014 .239 -.116 .815 Ignorant 2.86 .628 -.028 .117 -.015 .684 Not Motivated 2.64 .643 .041 .070 -.039 .686 Factor 3 - Criminality Factor 2 Criminality Sell Drugs 2.61 .596 -.246 -.384 .964 .942 1.78 .735 .060 .337 .801 Use Drugs 2.65 .644 -.155 -.008 .276 .823 Dishonest 2.60 .596 .069 -.065 .092 .707 219

Negative Perceptions of African American Males: Personal Characteristics

Results from this study suggests that there is a link to African American males perceived to have negative attitudes (e.g., threatening, aggressive, belligerent, violent, mean, and dishonest) and personal characteristics. Factor 2 (personal characteristics) indicated that the average person perceives African American males as having a predisposition to use government assistance, M =

2.91. Additionally, educators reported that the average person perceives African American males to be irresponsible, ignorant, and not educated. Perceptions of lack of education is the result of high school dropout rates, which according to the literature is a result of suspension, expulsions, and the school to prison pipeline phenomenon (Fairchild, et al., 2012; Fitzgerald, 2009; Foster,

1995; Horsford, 2011; Kunjufu, 2005; Noguera, 2012; Oelrich, 2012; Simson, 2013; Woodson &

Harris 2019). These findings are important because the case can be made that negative attitudes and personal characteristic perceptions can be combined as one variable to predict the likelihood that African American males could face higher referrals to special education for learning impairments and emotional and behavioral disorders (Fitzgerald, 2009; Foster, 1995; Simson,

2013; Woodson & Harris, 2019). Negative perceptions of African American males are steeped in bias that emanates from an individuals’ experiences from birth to college that shape their understanding of the world (Dovidio et al., 2002; Flavell, 1963; Haghighat, 2001; Nickerson,

1988).

Negative perceptions of African American males will ultimately affect the ways in which educators behave toward them (Simson, 2013). Additionally, perceptions can even affect the way in which Black males receive advocacy in the face of racial injustice (Quinn, 2017).

Furthermore, negative perceptions that motivate educator behaviors can result in emotional harm for Black males as result of marginalization and stigmatization in school (Quinn, 2017; Simson, 220

2013). For me this bears witness with experiences that I have observed in public schools. On one particular occasion, I witnessed a teacher yelling at an African American male student in third grade for not standing for the pledge of allegiance. The teacher was furious because the

African American male was in her opinion, disrespecting our country and the flag. Later in the day, the student was in the office on two separate incidences with discipline referrals from the same teacher.

Fundamental outcomes for students with emotional and behavior disorders are a diminished capacity for learning, resulting in poor academic performance, school truancy, increased rates of suspensions and expulsions, alcohol and drug abuse, as well as increased risk of not completing high school (Abubakur-Abdullateef, 2017; Foster, 1995; Shakoor & Chalmers,

1991; Horsford, 2011; Simson 2008). Summarily, it can be concluded that overall perceptions of

Black males are negative, and that negative perceptions adversely affect African American males in all facets of school. Perceptions of Criminality

Results from this study indicated that African American males are perceived to have criminal behavior. For example, Table 6.1 shows that African American males are perceived to engage in criminal behaviors such as using drugs, M = 2.65, selling drugs, M = 2.61, and are dishonest, M = 2.60. The literature suggests that adverse effects of negative perceptions of

African American males can adversely affect mental health (Beckford, 2016; Watkins, et al.,

2010; Quinn, 2017; Simson, 2013). Black males are less likely to seek out intervention for issues related to mental health. Furthermore schools are less likely to provide treatment for mental health issues (Watkins, et al., 2010). Increased drug use (e.g., marijuana) is an example how self- medication is the result of the mental stressors of marginalization and stigmatization faced by 221

African American males in schools (Watkins, Walker & Griffith, 2010). Perceptions of African

American males as criminals included selling drugs and dishonesty. Both comprised factor 3 for criminality along with using drugs. Selling drugs had the highest mean related to personal perceptions of public educators (M = 1.78, SD = .735). Criminality associated with Black males can be attributed to Blackness itself. Research suggests that Black males are often pictorializedas criminals (Foster, 1995; Linter, 2004; Wilson, et al., 2017; . Media photos often depict Black males as bigger, stronger, and capable of harm (Wilson et al., 2017). Media images can converge with confirmatory biases to create fixed ideals for how Black males are perceived in school and society (Foster, 1995; Jackson & Crawley, 2003, Nickerson, 1988; Wilson et al., 2017).

Perceptions of Black males are often shaped by images in the media, dominant White suppositions, and outcomes in schools (Foster, 1995). According to a Pew Research study, 20% of Americans get news from newspapers, 38% online, 25% from radio, and 57% from television.

Racial beliefs held by individuals are often developed before adulthood. Subsequently, an individual may develop political ideologies based on beliefs that have been developed before adulthood (Wood, 1994). Anti-Black affect can be developed on an individual basis and can motivate political beliefs (Davis & Engel, 2001; Dovidio, et al., 2002; Sniderman & Piazza,

1994). Political values may be anti-Black, but not intended to be racist. As an example, consider an article published on February 29, 2016, in The Washington Post. The headline read, "Donald

Trump’s father was arrested after a Klan riot in Queens." The title of the article, while salacious, is more relevant to the political values that comprised the group of White men, some of whom 222

were self-professed Klan’s men. According to The Washington Post, the men distributed fliers rallying native-born White men to speak out to protect: "one flag, the American flag; one school, public school; and one language, the English language.” The article cited by The Washington Post revealed a profound element of perceptions; the convergence of social and political values with the education system.

Implications from Exploratory Factor Analysis on Personal Perceptions

Factor analysis for personal perceptions yielded two factors (negative attitudes and criminality). Due to the limited sample size for this study and the statistical differences in perception means, I thought it was more meaningful to focus this discussion regarding results on factor analysis loadings for perceptions of the average person. For the sake of this study, perceptions of the average person are indicative of personal perceptions. Results from

MANOVA testing indicated that differences in personal perceptions scores were related to personal value statements that make up a persons’ ideological values (see also Hing et al., 2008).

A little later in this chapter, I will discuss the relationship personal perception scores have with personal value statements.

Overall personal perception scores revealed that educators do not agree with negative perceptions related to negative attitudes and criminality. More specifically, educators strongly disagreed that African American males are not intelligent and abusive as indicated by a mean score of 1.52 and a standard deviation of .660 respectively. Differences in mean scores for the average and personal perceptions can be interpreted two ways. First, lower means in personal 223

perceptions could indicate that educators are not likely to admit to having negative perceptions of African American males; therefore, they will report more favorable perceptions. Alternatively, educators that are least likely to discriminate as evidenced by personal value statements might accurately report higher perceptions of African American males. Failure of the ABSDS to help me differentiate differences in mean scores requires a discussion regarding how this instrument can proceed in future research.

The Adapted Brief Social Desirability Scale

The adapted brief social desirability scale was intended to help me make a decision whether it was more appropriate to use scores for the average person or personal perceptions. I discovered that the ABSDS did not function as intended. I adapted the ABSDS from Haghighat’s

(2001) brief social desirability scale (BSDS). First, the ABSDS did not perform well because I believe the process to adapt it to educators was overcomplicated and unnecessary (adapting items to fit educators). Haghighat (2001) used an abbreviated version of Strahan and Gerbasi’s (1972) condensed social desirability scale (See Appendix H). I thought that by changing two items to fit a school related scenario would help me make the ABSDS more relevant to educators. The two items changed to fit a school-based context (e.g., speaking to every student you come in contact with, and laughing at students behind their backs) did not load well when examined as loaded factors using EFA of maximum likelihood. Table 6.2 shows each item on the ABSDS examined as factors before and after extraction for EFA using maximum likelihood estimation. 224

Table 6.2

Social Desirability Item Communalities Before and After Extraction

Initial After Extraction 1. Do you always practice what you preach? .205 .999

2. Do you always admit when you make a mistake? .216 .835

3. Do you always keep your promises? .268 .286

4. Do you smile at every student you come in contact with during the .074 .077 school day? 5. Have you ever laughed or made fun of a student you .032 did not like behind their .037 back

Items 4 and 5 on the ABSDS did not load well in comparison to items 1 and 2, which had the highest residuals. Additionally, item 3 was not as strong as items 1 and 2, which adds credence to the argument of Gerbasi and Strahan (1972) who purported that the Marlowe-Crown social desirability scale was not necessarily reliable in its totality (i.e., 33-item scale; see

Appendix F). Consequently, I agree that a revised social desirability scale could have performed well; however, I should have taken items directly from Strahan and Gerbasi’s 10 -item scale to test for reliability with this study.

I did not run additional statistical tests to see if a two-item scale would have allowed the

ABSDS to function as intended. Future testing could incorporate a revised ABSDS with items 1 225

and 2 of the current scale to test Haghighat’s (2001) assertion that a social desirability scale can be reliable with only two items. Further testing would need to take place with a larger, more diverse sample size to determine scale reliabilities of the ABSDS. However, due to items 4 and 5 loading poorly, any iterations of the ABSDS would exclude them. Moreover, of more interest to the analysis of results is the MANOVA effect discovered with personal value statements

(ideological values). I find it more beneficial to proceed with future testing using a scale that does not have the ABSDS due to its failure to function as intended for this study.

Although disappointing, it was not a surprise that the ABSDS did not perform well. The discussion around the complexity of perceptions, particularly related to race and gender cannot be moderated by a scale to detect social desirability bias. Results allowed me to conclude that it is highly feasible that study participants did not demonstrate social desirability bias. A social desirability scale to detect marginalization of patients with mental illness is only marginally useful when examining perceptions of African American males that comprises social, political, psychological, and biological factors. Looking deeper at the complexity of perceptions related to race and gender, results of this study indicate that social and political factors operating in complex systems may explain how Black males are perceived in public schools (Milner, 2007;

Sears & Henry, 2003; Wood, 1994).

The Relationship Between Perceptions and Political Ideologies

Personal perceptions and perceptions of the average person, which served as the dependent variables for this study were deemed to be significant when considered jointly with personal value statements, F(8,78) + .784, p = .005, h2 = .240. Based on these results, I assert that personal value statements will allow me to use the perceptions for the average person as the 226

personal perceptions of public educators. Additionally, I concluded that personal perceptions of educators are more negative than reported; however, they are less likely to express them. Finally, as a result of the MANOVA effect found when examining personal value statements in relation to personal perceptions, I concluded that educators that are TLP are least likely to have negative perceptions of African American males; therefore, personal perceptions can be attributed to personal values that are more egalitarian.

I used personal value statements to make certain assumptions regarding public educators who personally hold more negative perceptions than they reported in this study but expressed their personal perceptions through the average person. For example, 38% of participants indicated that they believed that some races receive more government support than others.

Educators reported that the average person agrees that African American males are likely to use government assistance (e.g., social security, food stamps, TANF) as evidenced by a mean score of 2.91. This supports findings that indicated that personal value statements are an appropriate indicator of personal perceptions. Simply put, participants in this study can be classified into two categories: truly low prejudice (TLP) and those that are not TLP. The PEPAAMS PK12 had an item that asked participants to select all that apply to how the average person perceives African

American males overall. Results indicated that overall the average person believes that African

American males are (1) threatening, (2) aggressive, (3) lack intelligence, and (4) violent. I conclude that the findings support further investigation that educators who have negative perceptions of African American males might be more likely than not to make special education

(SpEd) referrals for specific learning disabilities (SLD), and emotional and behavioral disorders

(EBDs). 227

I do not believe that every educator has negative perceptions of African American males.

Using items that comprise truly low prejudice (TLP) profiles as the independent variable, results indicated that educators with TLP profiles are less likely to have negative perceptions, which could explain the statistical differences in the means of personal perceptions and average perceptions of African American males. Significant MANOVA effect suggests that educators with TLP values are least likely to have negative perceptions of African American male students, which allows me to use findings related to personal perceptions of African American males with educators who are truly low prejudice. I chose to apply the same criteria for TLP educators who were asked to express their personal perceptions by selecting all that apply to how African

Americans are perceived overall. I conclude that educators who are TLP would not be likely to make special education referral for any category.

Personal Value Statements and Political Affiliation

Findings from the present study were not able to link political affiliations to personal or average perceptions. Two areas of this study asked participants to indicate their political views or affiliations. Participants were asked to indicate if they were liberal or conservative on a variety of political issues. Fifty one and one-tenth percent of participants indicated that they were liberal on many political issues compared to 20.7% who expressed being conservative on a variety of issues.

In the demographic section of the PEPAAMS PK12, participants were requested to indicate their political affiliation. Participants who identified as democrats were 51.2%.

Republicans comprised 17.1%, and Independents made up 20.7%. Libertarian, Green, and Other categories made up 7.3% of responses. People that are TLP are most likely to be liberal on many 228

issues; however, educated liberals are more likely to have complex belief systems that favor policies that disadvantage minority outgroups (Sears & Henry, 2003; Sniderman, 1988; Wood,

1994).

Egalitarian traits generally describe people that are TLP. For this study there were three traits that described TLP, (1) liberal on political issues, (2) supporters of equality for all racial groups, and (3) believers that all races should have equal access. Results from this study indicate that people that are not TLP also can have egalitarian traits; however, inclusion of any of the traits that are identified by marginalization or stigmatization should not appear in any combination of the TLP profile. A truly low prejudice person is not relegated to liberal views; however, a person can still be TLP with only two of the three egalitarian traits; however, they should not have any ideological views that promote stigmatization (e.g., belief that programs to help minorities are not needed to level the playing field).

Negative Perceptions of African American Male Students and Ideological Views

In Chapter 2, the literature review introduced four prejudice profiles. So far in this

Chapter, I have only discussed TLP. Aversive racism, economic/social conservatism, and political conservatism are combined into one category, not TLP. Modern racism was dropped altogether because relevant literature indicates that old-fashioned racism has been transformed into more sophisticated forms of system operations (Hing, et al., 2008; Sears & Henry, 2003;

Sniderman, 1988; Wood, 1994). Negative perceptions of Black males in public schools are a complex system that cannot be oversimplified. Relevant research explored the role that symbolic attitudes and political values play in constructing beliefs and perceptions about outgroups

(Wood, 1994). 229

Negative perceptions of African American males involve varying degrees of implicit and explicit bias. Implicit biases can motivate attitudes (Hing et al., 2008; Simson, 2013). Implicit and explicit biases influence behaviors (Dovidio et al., 2002; Hing et al., 2008). Relevant research suggests that anti-Black affect describes how individual beliefs influence perceptions that oppose support for African American students (Davis & Engel, 2001; Dovidio, et al., 2002;

Sniderman & Piazza, 1994). Perceptions of low intelligence of African American males reflects the low academic expectations that educators have and thus not committed to teaching them at high levels (Ladson-Billings & Tate, 1995; Noguera, 2012; Simson, 2013).

Negative beliefs about certain groups form prior to adulthood (Pohan & Aguilar, 2004;

Sears & Henry, 2003; Wood, 1994). As individuals evolve and matriculate through higher education in the United States, they are exposed to a reframing of values that in many instances are not racist in the context of old-fashioned racism; however, these political or traditional values drive in-group aspirations to preserve their advantage over lesser outgroups (Hing, et al., 2008;

Quinn, 2017; Sears & Henry, 2003; Wood, 1994).

Quinn (2017) framed the conversation around racial attitudes and racial politics to describe perceptions held by educators that support marginalizing African American students and other minoritized outgroups. Sears and Henry (2003) and Wood (1994) both purported that symbolic racism shares beliefs regarding anti-black affect (opposing support for equality for them in organizations) and traditional political values that drive perceptions. Hing et al. (2008) provided further evidence that symbolic racism is often encapsulated in conservative political beliefs that rely on individual appraisals (perceptions) of individual qualities that are in turn used as a metric for validating ingroup superiority rather than devaluing outgroup members’ 230

inferiority. Some educators unconsciously preserve similar interests in that they may understand that overestimating perceptions of themselves in the context of ingroup privilege may converge with pre-adult indoctrination that African Americans (i.e., Black students) do not measure up to the qualities ascribed to the racially dominant group (Haghighat, 2005; Nickerson, 1988; Quinn,

2017).

In Chapter 5, personal value statements from the PEPAAMS PK12 were isolated into their respective ideological categories (e.g., political and social conservatism; see Table 5.11), however, they can act as interchangeable components under all categories wherein stigmatization or marginalization can be realized (e.g., aversive racism; Dovidio, Gaertner, Kawakami &

Hodson, 2002; Hing et al., 2008). Conversely, people who have characteristics related to aversive racism are most likely to operate upon complex systems of thinking that allows them to have liberal ideologies, yet still retain beliefs that could result in the marginalization of minoritized groups. For example, Kinder & Sears (1981) asserted that Americans have become racially egalitarian; however, they continued by inferring that some of those same racial egalitarians opposed affirmative action, which is symbolic of racism. Similarly, educated

Americans, (e.g., PreK12 and postsecondary educators) also share racially egalitarian principles, but do not support government backed programs that promote racial equality (Sniderman, Brody,

& Kuklinski, 1984).

Political conservatives are equally opposed to programs that require government support to disadvantaged groups even though they say they espouse beliefs that support equality for all

(Hing, et al, 2008, Sears & Henry, 2003; Quinn, 2017). Results indicated that personal values of educators are more than likely to predict their perceptions of African American male students. 231

Summarily, I conclude that political ideologies are a powerful tool that can be used to separate negative perceptions of African American males from educators who are TLP and those that are not. Future research that uses the PEPAAMS PK12 that replaces personal value statements instead of the ABSDS will allow me to investigate this phenomenon further.

Recommendations for Future Research

Results from this study clash with time-tested social science studies that suggest negative beliefs can be explained and supported from empirical research. Current research indicates that social phenomena (e.g., lack of success of African American males in public schools) can be explained away with supporting and opposing evidence (e.g., race, gender, and socio-economic status combined, not isolated contributes to lack of success in school; see Berliner & Calfee,

1996; Fitzgerald, 2009; Robbins, Mercer, and Meyers, 1967; Wilson, Hugenberg, & Rule, 2017).

I argue that negative perceptions and outcomes for African American males will not be solved solely by scholarly research that provides supporting evidence regarding the realization of pressing issues. Lord, Ross, and Lepper (1979) suggested that rationality, enlightenment, and social consensus are not likely to emerge from objective data about perceptions and attitudes toward African American males. This study was meant to be a starting point to fill a void where empirical research is lacking. I admit that this study is incomplete without a follow up study that activates lessons learned. For example, in a follow up study, the ABSDS will be replaced by personal value statements that express ideological values.

Expanding the study for future distributions without a social desirability scale will strengthen the measure because sufficient evidence suggests that asking participants’ to answer for the average person is a call for them to reflect upon themselves if they are not TLP. It is a 232

human phenomenon for groups to identify with people that most represent their personal, biological, or economic interests (Haghighat, 2005; Nickerson, 1988; Sniderman, Brody, &

Kuklinski, 1984; Sniderman & Piazza, 1993; Sniderman, Piazza, Tetlock & Kendrick, 1991;

Sniderman & Tetlock, 1986; Wilson, Hugenberg, & Rule, 2017). When members of minority outgroups do not reflect personal, biological, and economic interests of the racial majority, evidence points to complex systems that have been erected to stigmatize and marginalize them.

An educator completing a measure such as the PEPAAMS PK12 will answer for the average person as they would for themselves if they are not TLP. Conversely, any educator that is TLP will have more favorable perceptions of African American males; therefore, personal perception scores will appropriately represent their beliefs. This assertion was confirmed during the qualitative methods used to validate the measure used in the present study. Members of the expert panel understood that the average person may vary depending on where they live, were born and raised, as well as the types of families they were raised in (e.g., southeastern regions of the United States versus the west coast).

In order to fully understand the impact of findings from the present study, future studies will commence so as to do further testing with a larger sample size that include participants from the largest school districts in America. The present study included gender, race, and years of service in education, but did not include age. More analysis must be done to understand how educator demographics can interact with perception scores by adding an item in the demographic section that asks participants to disclose their age. 233

Exploring Age in Future Distributions of the PEPAAMS PK12

In 2012, the latest data available from the National Center for Education Statistics

(NCES), indicated that 76.3% of all public school teachers were female; and of these, 81.9% were White. NCES data also disaggregated age ranges of public school teachers. First, 15.3% of

America’s teachers were under the age of 30. Additionally, 28.9% of teachers’ ages ranged from

30-39. Similarly, 25.1% of teachers were 40-49, and 23.1% ages ranged from 50-59. Finally,

7.6% of America's teachers were 60 and over in 2012. Using the 2012 data from NCES, 77% of all teachers in America's public school's ages ranged from 30 - 59. The birth year of America's largest teaching majority ranges from 1953 to 1982.

To further understand age and teacher demographics in public schools in the United

States, in 2012, teachers ranging in ages 30 to 39, were born between years 1982 and 1973.

Teachers who were 40 to 49 in 2012, were born between years 1963 and 1972. Approximately

30.7% of America's teachers were born between 1952 and 1962. As of 2012, the largest majority of public school teachers, 55.8%, were born during the height of major historical tides in

America (e.g., Brown v. Board of Education and the Civil Rights Movement). Brown v. The

Board of Education initiated the era of desegregation in public schools.

A review of literature prior to the distribution of the PEPAAMS PK12 pointed out that teacher demographics can allow researchers to gain insight to negative experiences of African

American children as a result of desegregation. More specifically, treatment of Black males became an area where empirical research could answer how experiences of Black students explained suspension, expulsion and graduation data. Furthermore, in the age of mass 234

incarceration, the research suggests that African American male outcomes in public schools are reflected in the school to prison pipeline (Noguera, 2012; Simson, 2013).

Response Rates

I hoped for higher response rates for the PEPAAMS PK12. The total amount of participants for the first and second distributions of the survey was 172 people. However, only

92 people completed the survey for useable data. Fifty-three percent of participants who participated in this study show that educators in Northwest Ohio are predominately white and female. Additionally, Northwest Ohio is a union town as evidenced by its geographical location

(an hour away from Detroit), which is the home of major automotive industry. Additionally,

Northwest, Ohio’s largest school district has union protections for every employee class, including principals. The majority of the sample population reflected current data regarding teachers in public schools in America as predominately Caucasian and female.

A larger sampling is needed to test assumptions associated with this study. I did not have the finances needed to access larger databases of public educators in the United States. A broad and diverse sample of 5,000 educators would be ideal for a study of public educator perceptions of Black males compared to the 92 participants used to conclude findings for this study.

Projected costs associated with accessing public educators in larger cities and school districts such as Atlanta, Chicago, Detroit, Los Angeles, Miami, New York, and Philadelphia will be approximately $20,000. The data indicate that Northwest Ohio educators may not be representative of the average educator working in urban school districts (e.g., on measures such as marital status, sexual orientation, and political affiliation). Seventy-seven and two-tenths percent of study participants indicated that they were in a traditional marriage, compared to 2.2% 235

that indicated that they were in a domestic partnership. Eighty-seven percent indicated that they identified as heterosexual, compared to 5.4% who identified as LGBTQ+. The majority of participants identified holding their political affiliation with Democrats, 51.2%. I expect that a larger sampling might provide a more robust contrast from Northwest Ohio educators.

Diverse Population and Sample

A plurality of study participants, 47.7%, lived in suburban communities. In future research, I want to see if the phenomena of suburban and rural living situations of educators are similar to this study wherein 17% lived in rural communities. This information is important because 64.1% of participants work in urban school districts, but 64.7% of them live outside of the communities in which they work. If demographic trends are similar in a broader study, it might explain that lack of association and community connections may reflect in educator perceptions (Foster, 1995).

A sample that is larger and more diverse might help explain the impact of personal values relating to educator perceptions. When educators are disconnected from the communities in which they teach, it is more probable that they may hold more negative perceptions of African

American males. Social distance characterizes how close educators feel to students in other racial groups (Hing et al., 2008; Quinn, 2017).

Confirmatory Factor Analyses in Future Research

There are two areas that I will examine in future distributions using the PEPAAMS

PK12. First, I will compare results of this study using Exploratory Factor Analysis with maximum likelihood estimation. Factor analysis using maximum likelihood estimation explored factors and examined scales to reveal internal consistency. Results from maximum likelihood 236

estimation chose parameters that led to the three factors emerging (negative attitudes, personal perceptions, and criminality) for perceptions of the average person, and two factors (negative attitudes and criminality) for personal perceptions. I would like to see how factors load with a larger more diverse sample size.

In addition to exploratory factor analysis using maximum likelihood, future distributions will use Confirmatory factor analysis (CFA) to determine how factors related to negative perceptions of African American males overlap with factors from EFA (Mertler & Vannatta,

2013). Confirmatory factor analyses will help me determine if the PEPAAMS PK12 supports a correlated, three-factor structure for the average person and two-factors for public educators’ personal perceptions of African American male students (Castillo et al., 2016). Results from

CFA in future distributions will help me make decisions for factor retention to eliminate redundancies and the degree to which negative perceptions overlap. For example, results from

EFA in this study revealed an error I made by including dishonesty in categories related to criminality and education. I will make modifications in future distributions of the PEPAAMS

PK12 with the modified instrument to reexamine internal consistency using EFA and CFA.

Future Social Science Research and Policy

Similar to the imperative given by critical race scholars who issued an urgent call for cultural and ethnic minority social science researchers to reclaim the narrative of oppressed groups by telling their stories (e.g., storytelling associated with qualitative research; Delgado,

1995, Solorzano & Yosso, 2001), I will answer the call to engage in more empirical research to add quantitative value to the stories provided by qualitative research. Social science research can provide objective quantitative data for use in public discourse; however, the amount which exists 237

might not yield change without considering the relationships that influence the socio-political discourse.

Previous mention was made regarding established research in social discourse about

“burning issues” (e.g., the fluidity of examining perceptions of Black males as students to Black males and leaders in education; Lord, Ross, & Lepper, 1979). This is not feasible without considering correlated links present in the social policy agenda. Social policy, public attitudes, beliefs about policy, and the role of social science research are the elements to be considered to shape public discourse about oppressed African American males and the call for action to ameliorate negative outcomes for them in public spaces (Lord, Ross, & Lepper, 1979).

This study answers Milner’s (2007) imperative that urges researchers to shift from self- awareness to system convergence. Findings from this research study provide a deeper look into how history (legitimized dominant hegemony) and politics shape a racialized and cultural system of operations, which impacts African American males students. This study also serves as a starting point to examine public school systems related to outcomes for African American males by creating a link that connects personal values to political ideologies that are shaped by perceptions. The next step in system examination is to extend findings from this study to concrete action steps that can be taken to ameliorate marginalization and stigmatization associated with how Black males are perceived in school. An example of how this can occur is to legitimize research findings that racial and social dominance that has gone unexamined has exacerbated the cultural divide between pedagogy and social contracts that continues to place

African American male students at risk for isolation (Cross, 2003; Gillborn, 2005; Hing et al.,

2008; Ladson-Billings & Tate, 1995; Milner, 2007; Simson, 2013). It is not the intent to 238

complicate the present study findings by suggesting that racism motivates educator perceptions; however, it cannot be denied that racism drives the value systems of America’s largest and most powerful institutions (Harris, Crenshaw, Gotanda, Peller & Thomas, 2012). While not the focus of the present study, extant research suggests that racial attitudes amongst educators in the racial majority are more negative than those in minority outgroups (Foster, 1995; Quinn, 2017). This study expanded the definition of perceptions to include personal and societal beliefs that are stereotypical, which include implicit and explicit biases. The definition of perceptions also includes feelings, expectations, fantasies and individual values (Foster, 1995; Haghighat, 2001;

2005; 2007; Pohan & Aguilar, 2001; Quinn, 2017; Sears & Henry, 2003; Simson, 2013;

Washington, 1981; Wilson, Hugenberg, & Rule, 2017; Wood, 1994).

Milner (2007) provided a safe space for researchers to examine the permanence of race in organizations. As complex systems, institutions mirror the values of society by choice or by chance (Bess & Dee, 2007). Organizational complexity is realized by the individuals who comprise them. Educator demography is an important factor when considering organizational complexity because the majority of educators, approximately 71%, are female. With 81% of the women being White, it lends the question to what role should the exploration of racism play when considering educator perceptions.

The post-Obama era has opened a conversation about White frustration dressed as nationalism (Lyons, 2017). White nationalism and its influential leaders play a role in rearticulating American values. White Nationalists are a pronounced voice defining American values of the past, present, and future, diverging from feminism, multiculturalism, and sexual 239

identity liberalism. They are an example of how personal values regarding race and political beliefs can shape the political landscape (Berlet & Lyons, 2000; Lyons, 2017).

The permanence of issues pertaining to race should be considered by shifting emphasis for detecting old-fashioned “redneck” racism (see Wood, 1994) to a more ambiguous trend in discovering the prevalence of symbolic racism (Hing, et al., 2008; Sears & Henry, 2003; Wood,

1994). Old-fashioned racism is a strongly discouraged phenomena in public school districts; however, anti-Black affect that motivates politically conservative beliefs are alleged to be non- racist (Wood, 1994). Disentangling race and racism is difficult because of the causal relationship between racism and values (Wood, 1994, p. 682). Anti-Black affect is the feeling of dislike and hostility (Sears, 1988). Traditional conservative values are driven by ideals espoused to individualism, self-reliance, hard work, obedience to authority, discipline, economic prudence, punctuality, and delay of sexual and social gratification (Sears & Henry, 2003). Furthermore, symbolic racism is a combination of anti-Black affect and traditional conservative values that seem to be the mantra of White Nationalist ideology (Berlet& Lyons, 2000; Lyons, 2017; Sears

& Henry, 2003; Sniderman, 1988; Wood, 1994). Research is split whether symbolic racism can be classified as racism. Wood (1994) inferred that racist views are held, and maybe expressed because individual(s) do not know that they are racist. He continued by offering that a person may not realize that views and positions that were adopted by the racially dominant group were established to preserve a position of privilege. Therefore, symbolic racism is racism if it is determined by motive and intent, and not mere expression and position because a person might have racist views and actions without intending to be racist. 240

Finally, I have not found one specific study that makes the connection between perceptions of educators and implications for special education referrals for EBDs. The present study marks a beginning to this uncharted research topic. Subsequent research studies will begin to broach the conversation regarding if negative perceptions of African American male students predict the likelihood that they could be referred to special education for emotional and behavioral disorders (EBDs) as well as for learning disabilities (Fitzgerald, 2009; Woodson &

Harris, 2019). Now that an initial distribution has been completed, the PEPAAMS PK12 will undergo another set of revisions that will take into account the factor loadings that yielded an overall negative perception score, which includes a dependent variable that will be used to categorize EBDs (e.g., threatening, aggressive, abusive, not intelligent, violent, not educated, and not motivated).

Recommendations for Leadership and Practice

The PEPAAMS PK12 was built to examine the perceptions of educators at all levels of school systems. A particular item within the PEPAAMS PK12 was included for participants to select value statements (ideological beliefs; Hing et al., 2008) that applied to their personal views. Items were computed as dichotomous variables (likely and least likely to discriminate or show bias). Chapter 5 listed these statements and were coordinated to their respective category.

Educators can be biased and can be discriminatory without intending to be racist (Sears & Henry,

2003; Sniderman, 1988; Wood, 1994). Leadership within public school systems should understand how social contracts are motivated by perceptions, attitudes, beliefs, and stereotypes.

Social contracts form from organized systems of thinking that inform organizational behavior

(Bess & Dee, 2007; Bolman & Deal, 2017; Jackson & Crawley, 2003). Leadership in the 241

structural frame reflects in its mission, vision, values, and goals, which become evident when social contracts are operationalized (Bolman & Deal, 2017). Leaders within organizations shape policy by the values they set as priorities for achieving organizational goals (Bess & Dee, 2007).

The leaders’ values are tightly coupled with the mission and vision of the organization. The goal of understanding how educators perceive African American males in public schools is to shift the system from tight coupling in relation to leader values to a more fluid (loose) system that considers the values of the most vulnerable populations within public school systems (Bess &

Dee, 2007; Bolman & Deal, 2017; Wieck, 1976).

Leadership in Relation to Policy, and Practice

A major tenet of critical race thought is that racism is an embedded psychology in

American culture. In fact, racism is the engine that keeps American systems operational (Bell,

1998, Davis, 2012). In short, it is more efficient to address systemic bias from an organizational or systems corrections theory perspective and its relationship to policy and practice. Before doing so, it is important to understand social contracts and how leaders and members in organizations appropriate beliefs and knowledge in professional practice. Social contracts can be reflected as (1) ready-to-sign (e.g., firmly held negative perceptions of African American male students), (2) quasi-completed (e.g., loosely coupled perceptions that may vary on environment), or (3) co-created (e.g., perceptions of African American male students are appropriate and emphasize restorative justice; Jackson & Crawley, 2003; Simson, 2013). Social contracts are one key aspect to understanding how to address public school systems.

Research studies like Woodson and Harris (2019) suggested that bivariate variables (e.g., student race and gender) are an example of ready-to-sign contracts that determine how behaviors 242

are interpreted as a result of negative perceptions of African American male students. This study found that negative perceptions of African American males are likely to be held by educators that have personal views that promote marginalization and stigmatization. As an example, the

PEPAAMS PK12 found that 56.5% of educators believed that obedience and respect are important values that should be taught in public schools. The literature indicates that African

American male students are not passive, stationary learners (Kunjufu, 2005; Noguera, 2012;

Simson, 2013). African American males are assertive and demonstrate leader-like qualities that can cause then to be perceived as aggressive and threatening (Foster, 1995). If views are expressed that oppose the status quo, this poses a problem for Black males in public schools.

Ready-to-sign contracts leave no room for African American males to be perceived as leaders and creative thinkers.

When student behaviors are deemed to be difficult, teachers are significantly more likely to refer African American male students to special education (Foster, 1995; Oelrich, 2012;

Woodson & Harris, 2019). Missing from the research is the specific disability to which African

American males are most likely referred or eligible to qualify for special education. Evidence suggests that African American males are more likely to be eligible for special education for multiple disabilities (Woodson & Harris, 2019); other health impairments (OHI; Fitzgerald,

2009), and emotional or behavioral disorders (EBDs; See Foster, 1995, Oelrich, 2012). However, reshaping public schools as systems with co-created social contracts, leaders can shape policies that examine how perceptions derived from socially closed-contracts (ready-to-sign) disadvantage African American male students related to discipline, educational placements, and social justice. 243

Organizational Repairs in Public School Districts

Organizational learning can be defined by the outcomes related to the way individuals in an organization perpetuate the organization’s values and norms (Bess & Dee, 2007; Levitt &

March, 1988). When an organization (e.g., public school districts) are unsuccessful in realizing challenges related to creating successful environments for African American males, it can be attributed to cognitive dysfunctions of members within the organization. Cognitive dysfunctions are products of standardized and accepted practices that have been perpetuated over time. When these practices are challenged, it becomes necessary for organizations to initiate cognitive repairs that will encourage members to relearn or learn effective strategies to address the problem. The outcome of this study revealed that African American males are most likely to be viewed negatively and subsequently marginalized in public schools. The cognitive repair must identify specific mental processes used to normalize behaviors that purport cognitive hypotheses that contributes to the marginalization and stigmatization of African American males.

The PEPAAMS PK12 indicated that 85.9% of public educators support equality for all racial and ethnic groups. Additionally, 88% of participants purported that all races should have equal opportunities in the United States. An example related to cognitive dysfunctions in public schools is that despite the majority of educators who espouse beliefs of equality and access, this is not reflected in student outcomes. Schools have yet to remedy how teacher quality and access to high quality educational opportunities in public schools are realized in schools and communities serving minoritized student populations (Ladson-Billings & Tate, 1995).

Cognitive repairs are divided into two classes: motivational and cognitive (Heath,

Larrick, & Klayman, 1988). For example, motivational repairs focus on the energy and 244

enthusiasm to which members engage in a task. Cognitive repairs are comprised of mental processes used to decide how to pursue a task. More importantly, what individuals do are a result of cognitive functions influenced by organization and social structures (Heath, Larrick, &

Klayman, 1988, p. 31). It can be concluded that the values of organizations are reified in its social structures, resulting in cognitive processes. Schools and school districts are urged to examine social structures that result in negative perceptions of African American males that drive decision-making (Fitzgerald, 2009; Simson, 2013).

Leadership and Social Contracts in Organizations

When there is no room for negotiation and judgments or decisions have been made regarding African American males, this reflects social contracts that are ready-to-sign (e.g., fixed and non-negotiable; Jackson & Crawley, 2003). The concept of ready-to-sign contracts (RtSC) are exemplified in high scores on negative perceptions. RtSC contracts are pre-negotiated and further negotiations are not permitted. For example, educators have specific perceptions of

African American males and deviating from these beliefs are futile and evidence gathering is included in all interactions to reify beliefs (Jackson & Crawley, 2003; Nickerson, 1988; Wilson, et al., 2017). From a systems theory perspective, ready-to-sign contracts are routine. Public school systems should not favor organizational cultures where ready-to-sign contracts represent a common trait used in decision-making. Organizational repairs are a priority repair that can address racial bias, stigmatization, marginalization, and cultural misunderstandings that result from ready-to-sign contracts (Bess and Dee, 2007; Jackson & Crawley, 2003; Levitt & March

1988; Oelrich, 2012; Simson, 2013). 245

Beliefs or perceptions that are partly open and partly fixed (i.e., a person having negative perceptions of Black males, but they are questioning why they have them) are indicative of quasi-completed contracts. Individuals in this stage retain the right to maintain the popular world-views, but they may also straddle the social fence. Additionally, a desire to maintain some sense of control reflects the vulnerability of partially open contracts. I proffer that developing leaders in public school districts are the key elements that can completely open, or permanently close, quasi-completed social contracts (Jackson & Crawley, 2003; Nickerson, 1988; Weick,

1976). There may be some educators at every level who are open to considering alternative perspectives as evidenced by diminished negative perceptions of African American male students; however, influential leaders within organizations can undo progress by leadership behaviors that are motivated by bias. Whether perceptions of African American male students are positive or negative, organizational leaders have influence over people and contribute to policy that shapes organizational culture.

The benefit of understanding perceptions related to African American male students is to move leaders and educators from quasi-completed contracts to co-created contracts. Co-created contracts are characterized by mutual satisfaction (Jackson & Crawley, 2003). Mutual satisfaction and understanding are the optimal place to be on the relationship spectrum. In co- created contracts, individuals do not hide from vulnerabilities and differences. On the contrary, they are open to share in the dynamic interfaces of differentiation, thriving on the mutual satisfaction that does not include obligation. Summarily, inherent differences drive robust understanding and mutual respect and affection that African American males need to thrive in school (Karpinski, 2006; Watkins, Walker & Griffith, 2010). 246

Perceptions of African American Males and School Leadership

Researching perceptions of African American male students in public schools is symbolic of researching African American male educators within public education (Milner, 2007). I believe that the same issues of stigmatization and marginalization of African male students also impact Black male educators, particularly those in leadership positions. Foster (1995) highlighted that participants in his study were keenly aware how negative perceptions of Black males impacted them as members of the group at the center of research. Diversity in school leadership that includes members that are most susceptible to marginalization and stigmatization are important to promote co-created contracts.

Leadership within schools and districts is key to building capacity to intervene in education where the average teacher is Caucasian and female. Educators must not be criticized for the environmental factors that shape the way in which they view the world; however, school districts must make greater efforts to ensure that leadership within schools and districts reflect its student population. Educational data trends suggest that public school staffs serving large minoritized student populations are not reflective of the student demographics they serve. For example, looking at race and ethnicity composites for school principals from 1987 to 2012, the percentage of African American principals has been steady, ranging from 9 to 10% (National

Center for Educational Statistics, 2012).

It will be virtually impossible for schools to increase the number of African American male teachers for first-line instruction by significant percentages. Notwithstanding, school districts can make deliberate efforts to increase the number of African American male principals in schools with larger numbers of at-risk male students. Not only should school districts be 247

proactive in recruiting and training principals, they should also be equally deliberative in providing culturally relevant leadership development and support for those principals.

I am making a recommendation to increase the number of African American male principals in public schools with a word of caution for senior leaders in school districts. Recruitment and placement of Black males as school principals without understanding the possible pitfalls that will limit success could allow for a faulty narrative to emerge about inferior leadership capacity.

Quinn (2017) points to the powerful influence that teachers’ unions play in shaping school policy. I argue that unions do more than shape policy, they can actually influence the type of leaders that will be acceptable to powerful union affiliates.

Teachers’ unions are paid to advocate for the interests of their members. To understand group interests of public educators, I would be less than honest if I did not express my fears that group interests expose minoritized children to inferior education in public schools. Teachers from the racial majority that believe that Black students cannot learn, might not exert the effort to educate them to high levels (Fairchild, et al., 2012; Ladson-Billings & Tate, 1995; Linter,

2004; Quinn, 2017). Furthermore, they will most likely resist an African American male principal whose job it is to act as an instructional leader (e.g., appraising teacher performance).

From personal experience, I can attest that African American male principals will most likely face challenges from teachers in the racial majority, and subsequently unions who work as lobbyists to school district leaders. Additionally, I proffer that union leaders can persuade district leaders regarding the effectiveness of African American male principals. The PEPAAMS PK12 can function as an abstract tool to measure how African American male principals might be perceived. Future research regarding African American male principals could provide a narrative 248

for how supervisors, school staffs, and Black males themselves view support for their leadership.

I hypothesize the following regarding perceptions of African American male principals: (1) supervisory and staff evaluations will be less favorable compared to non-Black counterparts, (2) decreased levels of support for challenging situations will be a common theme (e.g., more challenging leadership assignments and lack of material support to ensure success), and (3) as principals, Black males will have similar comments for how to navigate the political landscape in public school districts.

Education is a major source in promoting support for racial equality (Sniderman &

Tetlock, 1986). Racial equality in public schools is exposed to scrutiny because teachers in the racial majority give a nod to rhetorical support for racial equality but will do little to act upon it.

Extant research concludes that White educators often opposed government intervention to establish policies to level the playing field for African Americans and other racial minorities

(Milner, 2007; Quinn, 2017). Simply put, Sniderman and Tetlock (1986) and Hing et al. (2008) posit that some members of the racial majority oppose implementation of initiatives that support racial equality. I will extend it further; teachers in the racial majority may resist anyone that represents hope for racial equality (e.g., African American male educators; see Lewis, 2008).

Developing the PEPAAM Survey PK12 was the logical first step in expanding research regarding African American males in education. If I subscribe to findings by Wilson, Hugenberg, and Rule (2017) and Foster (1995), perceptions of African American males do not improve with time.

I conclude by emphatically purporting that the purpose of the instrument developed at the center of this study is not to administer to educators in order to label them as racist. On the 249

contrary, it is intended to be a tool used to develop a comprehensive program to provide leadership development for school district leaders and support for large urban school districts who are in academic distress who would benefit by having strong, African American male leaders to shift the narrative from a deficit structure to a empowered, egalitarian structure

(Milner, 2007; Peshkin, 1988).

Leaders in public school districts should heed the challenge for leaders to face the ocean to understand how to lead school districts and staff schools with effective teachers and principals. Kouzes and Posner (2002) described the ocean as the elements of unpredictability that require school leaders to understand the students and families that they serve. Diverse schools and families are symbolic of the ocean. School staff members are the fixtures on the shore that are the focus of decisions made by district leaders. District leaders are discouraged from leading from the vantage of the adults, but to face the areas of unpredictability and uncertainty and begin to lead with the most fragile, but also most vulnerable and unpredictable students in mind –

African American male students as well as other children in racial minorities. 250

REFERENCES

Alexander, M. (2011). The New Jim Crow. Ohio St. J. Crim. L., 9, 7.

American Educational Research Association, American Psychological Association, & National

Council on Measurement in Education. (2014). Standards for educational and

psychological testing. Washington, DC: America Educational Research Association.

Atkinson, G., & Nevill, A. M. (1998). Statistical methods for assessing measurement error

(reliability) in variables relevant to sports medicine. Sports medicine, 26(4), 217-238.

Atwater, L., Lau, A., Bass, B., Avolio, B., & Camobreco, J. (1994). The Content, Construct, and

Criterion-Related Validity of Leader Behavior Measures. VMI Research Labs.

Lexington, VA.

Baumgartner, L. M., & Johnson-Bailey, J. (2008). Fostering awareness of diversity and

multiculturalism in adult and higher education. New Directions for Adult and Continuing

Education, 2008(120), 45-53.

Bell, D. (2008). Faces at the bottom of the well: The permanence of racism. Basic Books.

Berlet, C., & Lyons, M. N. (2000). Right-wing populism in America: Too close for comfort.

Guilford Press.

Berliner, D. C., & Calfee, R. C. (Eds.). (1996). Handbook of educational psychology. Routledge.

Bertrand, M., & Mullainathan, S. (2001). Do people mean what they say? Implications for

subjective survey data. American Economic Review, 91(2), 67-72.

Bobo, L. (1994). The scar of race.

Bolman, L. G., & Deal, T. E. (2017). Reframing organizations: Artistry, choice, and leadership.

John Wiley & Sons. 251

Carr, P., & Klassen, T. (1997). Different Perceptions of Race in Education: Racial Minority and

White Teachers. Canadian Journal of Education / Revue Canadienne De

L'éducation, 22(1), 67-81. doi:1. Retrieved from http://www.jstor.org/s`/1585812 doi:1

Carré, J. M., McCormick, C. M., & Mondloch, C. J. (2009). Facial structure is a reliable cue of

aggressive behavior. Psychological Science, 20(10), 1194-1198.

Carrington, B., Tymms, P., & Merrell, C. (2008). Role models, school improvement and the

‘gender gap’—do men bring out the best in boys and women the best in girls?. British

Educational Research Journal, 34(3), 315-327.

Castillo-Diaz, M. & Padilla, J.-L. (2013). How cognitive interviewing can provide validity

evidence of the response processes to scale items. Social Indicators Research, 11(3). 963-

975. Doi:10.1007/s11205-012-0184-8.

Castillo, March, Stockslager & Hines (2015) Measuring Educators’ Perceptions of Their Skills

Relative to Response to Intervention: A Psychometric Study of a Survey Tool

Assessment for Effective Intervention, 41(2), 94-108.

Chapman, L. J. (1967). in observational reporting. Journal of Verbal

Learning and Verbal Behavior, 6(1), 151-155.

Clark, V.L.P., & Creswell, J.W. (2015). Understanding research: A consumer’s guide.

Merrill/Pearson Educational.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159.

Cohen, J. (1998). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: L.

Erlbaum Associates. 252

Colfax, J. D., & Stemberg, S. F. (1972). The perpetuation of racial stereotypes: Blacks in mass

circulation magazine advertisements. Quarterly, (36)8-18.

Connelly, L. M. (2016). Fisher's exact test. MedSurg Nursing, 25(1), 58.

Crenshaw, K. (1995). Critical race theory: The key writings that formed the movement. The New

Press.

Crenshaw, K. W. (2010). Twenty years of critical race theory: Looking back to move

forward. Conn. L. Rev., (43)1253.

Creswell, J.W. (2012) Educational Research: Planning, Conducting, and Evaluating Quantitative

and Qualitative Research. (4th ed.). Upper Saddle River, NJ: Pearson Education.

Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods

approaches. Sage publications.

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika,

16(3), 297-334.

Cross, B.E. (2003). Learning or unlearning racism: Transferring teacher education curriculum to

classroom practices. Theory into practice, 42(3), 203-209.

Crowson, H. M., DeBacker, T. K., & Davis, K. A. (2008). The DOG Scale: A valid measure of

dogmatism? Journal of Individual Differences, 29(1), 17-24. doi:10.1027/1614-

0001.29.1.17.

Desimone L. M. (2009). Improving impact studies of teachers’ professional development:

Toward better conceptualizations and measures. Educational Researcher, 38(3), 181–199.

doi:10.3102/0013189X08331140. 253

Delgado, R. (1995). Critical race theory: The cutting edge. Philadelphia, PA: Temple University

Press.

De Marco, N. (1995). Cut on the bias. Times Educational Supplement, April 7, 1995.

Desimone L. M. (2009). Improving impact studies of teachers’ professional development:

Toward better conceptualizations and measures. Educational Researcher, 38(3), 181–199.

doi:10.3102/0013189X08331140

Donaldson, S. I., & Grant-Vallone, E. J. (2002). Understanding self-report bias in organizational

behavior research. Journal of business and Psychology, 17(2), 245-260.

Dovidio, J. F., Gaertner, S.L., Kawakami, K., & Hodson, G. (2002). Why can’t we just get

along? Interpersonal biases and interracial distrust. Cultural Diversity and Ethnic

Minority Psychology, 8(2), 88-102. Retrieved from

http://rave.ohiolink.edu/ejournals/article/314032854

Dovidio, J. F. (2001). 2000 SPSSI Presidential Address: On the nature of contemporary

prejudice: The Third Wave. Journal of Social Issues, 57(4), 829-849. Retrieved from

http://rave.ohiolink.edu/ejournals/article/314032854

Du Bois, W. E. B. (Fourth Ed.). (1935). Black Reconstruction in America: Toward a history of

the part which black folk played in the attempt to reconstruct democracy in America,

1860-1880. Routledge.

Duckitt, J. (2006). Differential effects of right wing authoritarianism and social dominance

orientation on outgroup attitudes and their mediation by threat from and competitiveness

to outgroups. Personality and Social Psychology Bulletin, 32(5), 684-696. 254

Eberhardt, J. L., Davies, P. G., Purdie-Vaughns, V. J., & Johnson, S. L. (2006). Looking

deathworthy: Perceived stereotypicality of Black defendants predicts capital-sentencing

outcomes. Psychological Science, 17(5), 383-386.

Field, A. (2009). Discovering statistics using SPSS. Sage publications.

Fitzgerald, T.D. (2009). Controlling the African-American school-age male: Psychotropic

medications and the circumvention of public law 94-142 and Section 504. Urban

Education, 44(2), 225-247.

Flavell, J. H. (1963). The developmental psychology of Jean Piaget. Princeton, N.J.: Van

Nostrand.

Fleming, F.M. (2005). The subtlety and fragility of educational achievement: The African-

American male path to the Ph.D. Retrieved from ProQuest. (3169197)

Foner, E. (1990). A short history of reconstruction, 1863-1877. HarperCollins Publishers

Foster, H.L. (1995). Educators’ and noneducators’ perceptions of African-American males:

A survey. Journal of African American Men, 1(1), 37-70.

Franklin, J. H., & Moss, A. A., (1943; 1988). From slavery to freedom (6th ed.). New York

Gable, R. A., Park, K. L., & Scott, T. M. (2014). Functional behavioral assessment and students

at risk for or with emotional disabilities: Current issues and considerations. Education

and treatment of children, 37(1), 111-135.

Gillborn, D. (2005) Education policy as an act of white supremacy: Whiteness, critical race

theory and education reform. Journal of Education Policy, 20(4), 485-505.

Green, J.P. & M.A., Winters (2006). Leaving boys behind: Public high school graduation rates

(Report No. 48). New York, N.Y: Center for Innovation 255

Green, R. L. (1969). Racial Crisis in American Education.

Guion, R. (1977). Content validity: The source of my discontent. Applied Psychological

Measurement, 1(1), 1–10.

Haghighat, R. (2007). The development of the brief social desirability scale (BSDS). Europe’s

Journal of Psychology, 3(4).

Haghighat, R. (2005). The development of an instrument to measure stigmatization: factor

analysis and origin of stigmatization. The European journal of psychiatry, 19(3), 144-

154.

Hagiwara, N., Kashy, D. A., & Cesario, J. (2012). The independent effects of skin tone and facial

features on Whites' affective reactions to Blacks. Journal of Experimental Social

Psychology, 48(4), 892-898.

Harris, A. P., Crenshaw, K., Gotanda, N., Peller, G., & Thomas, K. (2012). Critical race

theory. International Encyclopedia of the Social & Behavioral Sciences.

Heath, C., Larrick, R.P., & Klayman, J. (1998). Cognitive repairs: How organizational practices

can compensate for individual shortcomings.” In Review of Organizational Behavior.

Hentschel, T., Shemla, M., Wegge, J., & Kearney, E. (2013). Perceived diversity and team

functioning: The role of diversity beliefs and affect. Small Group Research, 44(1), 33-61.

doi:10.1177/1046496412470725

Horsford, S.D. (2011). Vestiges of desegregation: Superintendent perspectives on educational

inequality and (dis)integration in the post-civil rights era. Urban Education Hills, 46(1),

34-54. 256

Huberty, D. J., & Petoskey, M. D. (2000). Multivariate analysis of variance and covariance. In

Handbook of applied multivariate statistics and mathematical modeling, pp. 183-208.

Academic Press. https://doi.org/10.1016/B978-012691360-6/500008-2.

Irvine, R. W., & Irvine, J. J. (1983). The impact of the desegregation process on the education of

Black students: Key variables. The Journal of Negro Education, 52(4), 410-422.

Jackson, R.L. & Crawley, R.L. (2003). White student confessions about a Black male professor:

A cultural contracts theory approach to intimate conversations about race and worldview.

The Journal of Men’s Studies, 12(1), 25-41.

Jacobs, H. A. (2009). Incidents in the life of a slave girl: Written by herself (Vol. 119). Harvard

University Press.

Jamieson, P., & Romer, D. (2008). The changing portrayal of adolescents in the media since

1950. Oxford University Press.

Kahle, L. R. & Berman, J. J. (1979). Attitudes cause behaviors: A cross-lagged panel analysis.

Journal of Personality and Social Psychology, 37(3), 315-321.

Kelly, T. J., Bullock, L. M., & Dykes, M. K. (1977). Behavioral disorders: Teachers' perceptions.

Exceptional Children, 43, 440-444.

Kern, L., Hilt, A. M., & Gresham, F. M. (2004). An evaluation of the functional behavioral

assessment process used with students with or at risk for emotional and behavioral

disorders. Education and Treatment of Children, 440-452. URL:

http://www.jstor.org/stable/42899816 Accessed: 29-10-2016 18:58 UTC

Khanna, N. (2010). “If You’re half black, you’re just black”: Reflected Appraisals and the

persistence of the one-drop rule. The Sociology Quarterly, 51(1), 96-121. 257

Kinder, D. R., & Sears, D. O. (1981). Prejudice and politics: Symbolic racism versus racial

threats to the good life. Journal of personality and social psychology, 40(3), 414.

Kunjufu, J. (1985). The conspiracy to destroy Black boys. Document Resume Ed 285 928 Ud

025, 682, 20.

Ladson-Billings, G., & Tate, W.F. (1995). Toward a critical race theory of education. Teachers

College Record, 97(1), 47-68.

Ladson-Billings, G., & Tate IV, W. F. (1995). Toward a critical race theory of education.

In Critical race theory in education (pp. 21-41). Routledge.

Lambert, A. J., & Chasteen, A. L. (1997). Perceptions of disadvantage versus conventionality:

Political values and attitudes toward the elderly versus Blacks. Personality and Social

Psychology Bulletin, 23(5), 469-481.

Lanier, J. & Wittmer, J. (1977). Teacher prejudice in referral of students to EMR programs. The

School Counselor, 24(3), 165-170.

Lavery, M. R., Jong, C., Krupa, E., & Bostic, J. (2019). Developing an assessment with validity

in mind. In J. Bostic, E. Krupa & J. Shih (Eds.), Assessment in Mathematics Education

Contexts: Theoretical Frameworks and New Directions. New York, NY: Routledge

Ledesma, M. C., & Calderón, D. (2015). Critical race theory in education: A review of past

literature and a look to the future. Qualitative Inquiry, 21(3), 206-222.

Lefevre, C. E., Lewis, G. J., Perrett, D. I., & Penke, L. (2013). Telling facial metrics: facial width

is associated with testosterone levels in men. Evolution and Human Behavior, 34(4), 273-

279. 258

Lintner, T. (2004). The savage and the slave: Critical race theory, racial stereotyping, and the

teaching of American history. Journal of Social Studies Research, 28(1), 27.

Loftus, G. R., & Masson, M. E. J. (1994). Using confidence intervals in within-subject designs.

Psychonomic Bulletin and Review, 1, 476-490.

Lynn, M. (2002). Critical race theory and the perspectives of Black men teachers in the Los

Angeles public schools. Equity & Excellence in Education, 35(2), 119-130.

Lyons, M. N. (2017). Ctrl-alt-delete: The Origins and Ideology of the Alternative Right.

Somerville, MA: Political Research Associates, January 20.

Ma, D. S., & Correll, J. (2011). Target prototypicality moderates racial bias in the decision to

shoot. Journal of Experimental Social Psychology, 47(2), 391-396.

Marotto, R. A. (1977). "Posin' to be chosen": An ethnographic study of ten lower-class African-

American male adolescents in an urban high school (Doctoral dissertation, State

University of New York at Buffalo, 1977). Dissertation Abstracts International, 39-03,

1234A.

Marwick, A. & Lewis, R. (2017, May 25). Media manipulation and disinformation online. New

York: Data & Society Research Institute.

Maxwell, J. (2013). Qualitative research design: An interactive approach (3rd ed.). Thousand

Oaks: Sage Publications.

Mertler, C.A. & Vannatta, R.A. (2013). Advanced and multivariate statistical

methods: Practical application and interpretation (5th ed.). Los Angeles: Pyrczak.

Mezirow, J. (1991). Transformative dimensions of adult learning. San Francisco: Jossey-Bass. 259

Milner IV, H. R. (2007). Race, culture, and researcher positionality: Working through

dangers seen, unseen, and unforeseen. Educational researcher, 36(7), 388-400.

Minnesota Department of Education. (1989). A study comparing state and local eligibility

criteria for LD, EMH, and E/BD programs in Minnesota. St. Paul, MN: Minnesota Dept,

of Education, January.

Muijs, D. (2006). Measuring teacher effectiveness: Some methodological reflections.

Educational Research and Evaluation, 12(1), 53–74. doi:10.1080/13803610500392236

Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review

of general psychology, 2(2), 175-220.

Noguera, P. A. (2012). Saving Black and Latino boys: What schools can do to make a

difference. Phi Delta Kappan, 93(5), 8-12.

Oelrich, N. M. (2012). A New Idea: Ending Racial Disparity in the Identification of Students

with Emotional Disturbance. SDL Rev., 57, 9.

Peshkin, A. (1988). In search of subjectivity—one's own. Educational researcher, 17(7), 17-21.

Pitre, N., Stewart, S., Adams, S., Bedard, T., & Landry, S. (2007). The use of puppets with

elementary school children in reducing stigmatizing attitudes towards mental

illness. Journal of Mental Health, 16(3), 415-429.

Posten, H. O. (1984). Robustness of the two-sample t-Test. In D. Rasch and M. L. Tiku (Eds.),

Robustness of statistical methods and nonparametric statistics (pp. 92-99). Dordrect,

Germany: Reidel.

Press, T.A. (1988,). Study finds African-Americans twice as liable to school penalties as Whites.

The New York Times 260

Quinn, D. M. (2017). Racial Attitudes of PreK–12 and Postsecondary Educators: Descriptive

evidence from nationally representative data. Educational Researcher, 46(7), 397-411.

Rescorla, L. A., Achenbach, T. M., Ginzburg, S., Ivanova, M., Dumenci, L., Almqvist, F., &

Erol, N. (2007). Consistency of teacher-reported problems for students in 21

countries. School Psychology Review, 36(1), 91-111.

Rudman, L.A. (2004) Social justice in our minds, homes, and society: The nature, causes, and

consequences of implicit bias. Social Justice Research, 17(2), 129-142.

Saucier, G. (2000). Isms and the structure of social attitudes. Journal of personality and Social

Psychology, 78(2), 366-385.

Sears, D.O., & Henry, P.J. (2003). The origins of symbolic racism.” Journal of personality and

social psychology, 85,(2), 259-275.

Serafini, G., Pompili, M., Haghighat, R., Pucci, D., Pastina, M., Lester, D., & Girardi, P. (2011).

Stigmatization of schizophrenia as perceived by nurses, medical doctors, medical

students, and patients. Journal of psychiatric and mental health nursing, 18(7), 576-585.

Sexton, R C. (1969). The feminized male: Classrooms, White collars & the decline of manliness.

New York: Random House.

Smith, R. W. (1990). Ethnic images (GSS Top Rep. No. 19). University of Chicago, National

Opinion Research Center

Sniderman, P. M., & Piazza, T. (1993). The scar of race. Cambridge, MA: Harvard University

Press.

Sniderman, P.M., Brody, R.A., & Kuklinski, J.H. (1984). Policy reasoning and political values:

The problem of racial equality. American journal political, 75-94. 261

Sniderman, P.M., & Tetlock, P.E. (1986). Reflections on American racism. Journal of Social

Issues, 42(2). 173-187.

Sniderman, P.M., Piazza, T., Tetlock, P.E., & Kendrick, A. (1991). The new racism. American

Journal of Political Science, 423-447. https://www.jstor.org/stable/2111369.

Solorzano, D. G., & Yosso, T. J. (2001). From racial stereotyping and deficit discourse toward a

critical race theory in teacher education. Multicultural Education, 9(1), 2–8.

Steeh, C. (1995). “Paul M. Sniderman and Thomas Piazza, The Scar of Race. PUBLIC

OPINION QUARTERLY, 59,

Terry, L. M. (2007). Keeping black boys out of special education.

Tracy, S. J. (2010). Qualitative quality: Eight “big-tent” criteria for excellent qualitative

research. Qualitative inquiry, 16(10), 837-851.

Upton, G. J. (1992). Fisher's exact test. Journal of the Royal Statistical Society. Series A

(Statistics in Society), 155(3), 395-402. Doi:10.2307/2982890

Van de Mortel, T. F. (2008). Faking it: social desirability response bias in self-

report research. Australian Journal of Advanced Nursing, the, 25(4), 40-48.

Wald, J., & Losen, D. J. (2003). Defining and redirecting a school-to-prison pipeline. New

Directions for youth development, 2003(99), 9-15.

Watkins, D. C., Walker, R. L., & Griffith, D. M. (2010). A meta-study of black male mental

health and wellbeing. Journal of Black Psychology, 36(3), 303-¬330.

doi:10.1177/0095798409353756

Whites retain negative view of minorities, a survey finds. The New York Times, January 1,

1991, p. B10. 262

White, M., & Charry, J. (Eds.). (1966). School disorders, intelligence, and social class. New

York: Teachers College Press

Wilson, J. P., Hugenberg, K., & Rule, N. O. (2017). Racial bias in judgments of physical size

and formidability: From size to threat. Journal of personality and social psychology,

113(1), 59.

Wood, F. H., Johnson, J. L., & Jenkins, J. R. (1986). The Lora Case: Non-biased referral,

assessment, and placement procedures. Exceptional children, 52(4), 323-331.

Wood, J. (1994). Is “symbolic racism” racism? A review informed by intergroup behavior.

Political psychology, 673-686.

Woodson, L. & Harris, M. (2018) Teacher and Student Demographic Variables Which Predict

Teacher Referrals of Males for Special Education Evaluation. Journal of At-Risk Issues,

21(1), 32-43.

Yosso, T. (2002). Critical race media literacy: Challenging deficit discourse about Chicanas/os.

Journal of Popular Film and Television, 30(1), 52-62.

Young, M. D., & Skrla, L. (Eds.). (2012). Reconsidering feminist research in educational

leadership. SUNY Press

Weick, K. E. (1976). Educational organizations as loosely coupled systems. Administrative

Science Quarterly, 21(1), 1-19.

Zaller, J. (1992). Reasoning and choice: Explorations in political psychology. The American

Political Science Association. 263

APPENDIX A. INFORMED CONSENT LETTER

Dear Fellow Educator,

Recently teachers in school districts across Northwest Ohio were invited to participate in a research study that will examine educator perceptions of an underrepresented demographic group of students in public schools. If you have already completed the online survey, thank you for your participation. If not, I am asking for your kind consideration to participate in this study.

The primary focus of the study is to examine perceptions of Pre K through 12th-grade public educators.

Your participation is totally voluntary; however, if you choose to participate, you will be adding value to social science research wherein there is little quantitative research studies. Specifically, this study will inform future research for recommendations in teacher education training and providing professional development for schools and districts.

Risks: Risk of participation is no greater than basic critical thought processes regarding inferences and perceptions of people, as experienced in daily routines. However, two items contain information that some participants may find to be offensive as they contain sexually explicit language and material. You will be given the opportunity to skip these items and proceed with the rest of the survey if you so desire. It is recommended that this survey be completed from a personal device.

Rights: Your voluntary participation is anonymous, and you are free to withdraw at any time. To secure your anonymity, close your internet browser once you have completed the online survey. Additionally, it would be helpful to clear your Internet browser and page history. Completing the survey indicates your consent to participate.

How You Can Participate: The survey will take approximately 10 minutes to complete as you are guided through a series of questions. Again, there is no way to identify you or your institution. All data will be kept in a password-protected file, only accessible to me.

Thank you in advance for your participation in this study. If you have any questions about this research study, please feel free to call me directly at (614) 535-7773 or email me at [email protected] You may also contact Dr. Patrick Pauken at (419) 372-2550 or at [email protected]. Questions or concerns about your rights as a research participant should be directed to the Bowling Green State University Institutional Review Board (IRB) at [email protected] or at (419) 372-7716. The study has received approval through the BGSU IRB Committee. 264

Sincerely,

D. Christopher Scott Doctoral Candidate Bowling Green State University 265

APPENDIX B. PEPAAM SURVEY PK12

Public Educator Perceptions of African American Male Students PK12

Section One Qualifying Questions (Yes/No)

1. Are you employed full or part-time in a public school district? 2. Do you hold a valid license or certificate in the area in which you are employed in a public school or district?

Section Two Adapted Brief Social Desirability Scale

1. Do you always practice what you preach? 2. Do you always admit when you make a mistake? 3. Do you always keep your promises? 4. Do you always smile at every student you come contact with during the school day? 5. Have you ever laughed or made fun of a student you did not like behind their back?

Section Three Public Educators’ Perceptions of African American Males for the Average Person

Criminality

The average person perceives that African American males are: (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree)

1. Likely to have criminal behavior 2. Dishonest 3. Steal 4. Likely to sell drugs 5. Use drugs 6. Education

The average person perceives that African American males are: (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree) 266

1. Not as intelligent 2. Ignorant 3. Not as educated 4. Not motivated 5. Apathetic 6. Personality 7. Personality

The average person perceives that African American males are: (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree)

1. Lazy 2. Not ambitious 3. Irresponsible 4. Likely to receive government assistance (e.g., social security, SNAP/food stamps, TANF/temporary assistance for needy families, etc.)

Attitude The average person perceives that African American males are: (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree)

1. Belligerent 2. Mean 3. Vain 4. Abusive 5. Aggressive 6. Threatening 7. Violent

Sexual Prowess The average person perceives that African American males are: (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree) 267

1. Likely to rape women 2. Likely to abuse women 3. Have strong sex drives

Overall Perceptions of African American Males

Select all that apply regarding how you believe the average American perceives African American males:

1. Low intelligence 2. Not educated 3. Not motivated 4. Aggressive 5. Threatening 6. Violent

This section asks for you to express your personal perceptions. All responses are anonymous, so please feel free to be candid.

Crime

Rate your personal perceptions about African American males. (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree)

1. They are likely to be dishonest 2. They are likely to have criminal behavior 3. They are likely to steal 4. They are likely to sell drugs 5. They are likely to use drugs

Education Rate your personal perceptions about African American males. (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree) 268

1. They are not as intelligent 2. They will not complete high school 3. They lack motivation 4. They are apathetic 5. They are ignorant

Personality

Rate your personal perceptions about African American males. (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree)

6. They are lazy 7. They are not ambitious 8. They are irresponsible 9. They are likely to receive government assistance (e.g., social security, SNAP/food stamps, TANF/temporary assistance for needy families, etc.)

Attitude

Rate your personal perceptions about African American males. (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree)

10. They are belligerent 11. They are mean 12. They are vain 13. They are abusive 14. They are aggressive 15. They are threatening 16. They are violent

Sexual Prowess

Rate your personal perceptions about African American males. (* SA – strongly agree, A – Agree, D – disagree, SD – strongly disagree) 1. They are likely to rape women 269

2. They are likely to abuse women

Overall Personal Perceptions Held by Public Educators

Select ALL items that apply to your overall personal perceptions of African American males. 1. Not as intelligent 2. They have lower education 3. They lack motivation 4. They are apathetic 5. They are belligerent 6. They are mean 7. They are abusive 8. They are aggressive 9. They are prone to be violent Demographic Section

How do you identify

Female (1) Male (2) Transgender (3) Other. Please specify in the box (4)

Please select race/ethnicity categories that you feel apply to you. White (1) Black or African American (2) American Indian or Alaska Native (3) Asia (4) Native Hawaiian or Pacific Islander (5) Other (please specify) (6)

Please select the highest degree you have earned.

Less than a Bachelors (1) Bachelors (2) Masters (3) Educational Specialist (4) Doctorate (e.g., PhD., EdD., PsyD) (5) Other (please specify) (6) Please select your years of experience in education.

Less than 3 years (1) 3 to 9 years (2) 10 - 20 years (3) 20 - 30 years (4) Over 40 years (5)

Select the type of school district in which you work.

urban (1) suburban (2) rural (3) public/community charter school (4)

Please select all that apply.

I live in a suburban community (1) my partner or significant other is a police officer (2) I live in a rural community (3) someone in my immediate family is a police officer (4) 270

I live in the city (5)

Please select all that apply to your domestic situation. Legally Married (1) Domestic Partnership (2) Divorced (3) Separated (4) Single (5)

Which statement best describes you? LGBTQ (1) Heterosexual (2)

Select all that describe your personal views.

Programs to help minorities are not needed to level the playing field (1) Racism is not as big of an issue as portrayed in the media (2) Americans should not sit/kneel during the pledge of allegiance and national anthem (3) The Bible is the infallible, inspired word of God (4) People who oppose traditional American values are a threat to our democracy (5) Obedience and respect are important values that all children should learn in school (6) I am liberal on many political issues (7) I am conservative on many political issues (8) I support equality for all racial and ethnic groups (9) All races should have equal opportunities in the United States (10) Some races receive more government support than others (11) Tax dollars should not be used for cash and food assistance for families on welfare (12)

True or False: I do not have any problem with my child, sibling, or close relative marrying outside of their race. False (1) True (2)

Which best describes your political affiliation? Democrat (1) Republican (2) Independent (3) Libertarian (4) Green Party (5) Other, please list (6) How comfortable were you taking this survey? Very comfortable (1) Comfortable (2) Uncomfortable (3) Very uncomfortable (4) Would you be open to participating in a personal interview with the researcher?

If yes, (please provide your personal e-mail address to be contacted by the researcher) (1) If no, thank you for your participation thus far.(2) 271

APPENDIX C. IRB APPROVAL LETTER 272 273

APPENDIX D. INITIAL EXPERT PANEL INVITATION

Expert Panel Invitation (Initial)

Dear Expert Professional,

My name is Chris Scott; I am a doctoral student in the Leadership Studies Program in the School of Education Foundations Leadership & Policy Studies at Bowling Green State University (Ohio). I am respectfully requesting your consideration to participate in my research study regarding educator perceptions. The present study will fulfill a partial requirement for a dissertation proposal and subsequent defense. You are being invited to participate because of your expert knowledge of education as a public-school educator or related professional. Your participation is entirely voluntary. Your responses are entirely confidential, and no personally identifiable information is necessary for this study. The estimated completion time for this study is approximately 15 minutes educator perceptions.

If you would like to add to a limited body of research regarding perceptions of public school educators' and agree to participate, click the following hyperlink

Thank you,

Together as one as public-school educators,

D. Christopher Scott 274

APPENDIX E. INVITATION AND INFORMED CONSENT

Initial Invitation and Informed Consent to Public Educator Perceptions PK12 Survey

Dear Fellow Educator,

My name is Chris Scott and I am conducting a research study as a partial fulfillment of a Doctor of Education in Leadership Studies from Bowling Green State University in the school of Educational Foundations Leadership & Policy. Dr. Patrick D. Pauken, school director, is my advisor.

I am studying educator perceptions of an underrepresented demographic group of students in public schools. The primary focus of the study is to examine educator perceptions of public educators. Your participation is totally voluntary. Your relationship with BGSU will not be impacted if you choose not to participate.

Risks: Risk of participation is no greater than basic thought processes experienced in daily routines.

Rights: Your voluntary participation is anonymous, and you are free to withdraw at any time. To secure your anonymity, close your internet browser once you have completed the online survey. Additionally, it would be helpful to clear your Internet browser and page history. Completing the survey indicates your consent to participate.

How You Can Participate: The link in this e-mail will direct you to a link that will take you to the electronic questionnaire in Qualtrics. The survey will take approximately 10 minutes to complete

All data will be kept in a password-protected file, only accessible to me. To take the survey, please click on the following link provided at the conclusion of this message.

Thank you in advance for your participation in this study. The present study has been approved through BGSU’s Institutional Review Board If you have any questions about this research study, please feel free to call me directly at (614) 535-7773 or email me at [email protected] You may also contact Dr. Patrick Pauken at (419) 372-2550 or at [email protected]. Questions or concerns about your rights as a research participant should be directed to the Bowling Green State University Institutional Review Board (IRB) at [email protected] or at (419) 372-7716.

If the link below does not activate, please copy and paste into your browser. https://bgsu.az1.qualtrics.com/jfe/form/SV_dhWK2RzrQm8VQTH

Sincerely, 275

D. Christopher Scott Doctoral Candidate Bowling Green State University 276

APPENDIX F. MARLOWE-CROWNE SOCIAL DESIRABILITY SCALE

Marlowe-Crowne Social Desirability Scale (1960)

Listed below are a number of statements concerning personal attitudes and traits. Read each item and decide whether the statement is true or false as it pertains to you personally.

1. Before voting I thoroughly investigate the qualifications of all the candidates. (T)

2. I never hesitate to go out of my way to help someone in trouble. (T)

3. It is sometimes hard for me to go on with my work if I am not encouraged. (F)

4. I have never intensely disliked anyone. (T)

5. On occasion I have had doubts about my ability to succeed in life. (F)

6. I sometimes feel resentful when I don't get my way. (F)

7. I am always careful about my manner of dress. (T)

8. My table manners at home are as good as when I eat out in a restaurant. (T)

9. If I could get into a movie without paying and be sure I was not seen I would probably do it.

(F)

10. On a few occasions, I have given up doing something because I thought too little of my

ability. (F)

11. I like to gossip at times. (F)

12. There have been times when I felt like rebelling against people in authority even though I

knew they were right. (F) ,

13. No matter who I'm talking to, I'm always a good listener. (T)

14. I can remember "playing sick" to get out of something. (F) 277

15. There have been occasions when I took advantage of someone. (F)

16. I'm always willing to admit it when I make a mistake. (T)

17. I always try to practice what I preach. (T)

18. I don't find it particularly difficult to get along with loud mouthed, obnoxious people. (T)

19. I sometimes try to get even rather than forgive and forget. (F)

20. When I don't know something I don't at all mind admitting it. (T)

21. I am always courteous, even to people who are disagreeable. (T)

22. At times I have really insisted on having things my own way. (F)

23. There have been occasions when I felt like smashing things. (F)

24. I would never think of letting someone else be punished for my wrong- doings. (T)

25. I never resent being asked to return a favor. (T)

26. I have never been irked when people expressed ideas very different from my own. (T)

27. I never make a long trip without checking the safety of my car. (T)

28. There have been times when I was quite jealous of the good fortune of others. (F)

29. I have almost never felt the urge to tell someone off. (T)

30. I am sometimes irritated by people who ask favors of me. (F)

31. I have never felt that I was punished without cause. (T)

32. I sometimes think when people have a misfortune they only got what they deserved. (F)

33. I have never deliberately said something that hurt someone's feelings. (T) 278

APPENDIX G. REVISED EXPERT PANEL INVITATION LETTER

Expert Panel Invitation Letter (Revised)

Dear Expert Panel Contributor:

First, allow me to clarify my statement on you completing the survey in the YouTube link. PLEASE COMPLETE THE SURVEY! You may be able to detect any problems that participants might encounter during the piloting phase of the study.

I am honored to extend this invitation for you to participate as a member of the expert panel.

Time is the biggest concern for most of you. Believe it or not, taking the survey and providing expert feedback ranges from 8 to 15 minutes. If you choose to be very detailed, please plan on spending 20 minutes.

My dissertation encompasses reporting the psychometric properties of the instrument that I created to examine educator perceptions of African American males Pre K-12. As expert contributors, I am asking that you watch the two-minute YouTube link, which is posted below. Afterward, please complete the expert panel survey. Please see the enlarged link in red below.

Your professional role makes you a prime candidate to offer expert feedback for this study.

I look forward to interacting with your feedback.

If you would like a draft of the proposal document, send me a message and I will forward you a copy.

With many thanks, D. Christopher Scott 279

APPENDIX H. BRIEF SOCIAL DESIRABILITY SCALE

Haghighat (2001) Brief Social Desirability Scale (BSDS)

1. Would you smile at people every time you meet them?

2. Do you always practice what you preach to people?

3. If you say to people that you will do something, do you always keep your promise no matter

how inconvenient it might be?

4. Would you ever lie to people?

5. Would you ever laugh at a dirty joke people may make? 280

APPENDIX I. COGNITIVE INTERVIEW SCRIPT

Cognitive Interview Script

Hi, my name is D. Chris Scott, you can call me Chris. I am a social science researcher. My area of emphasis is critical race studies and perceptions as well as images of Black males. I am studying educator perceptions of African American males.

When studying such controversial topics like this, self-report surveys are vulnerable to a phenomenon called social desirability bias. Social desirability bias is when I give answers so I can look good to you, which may not be my honest opinion or my views. I have adapted a brief social desirability scale for educators to help me determine how to interpret their scores on the instrument that I created to examine educator perceptions. I need your help to understand how educators will answer. I am asking you to complete the adapted brief social desirability scale. At the end you will have the opportunity to give me some feedback on the scale. For example, if asked if you ever left a pet outside in extreme conditions. You might answer, “no” because who wants to admit to possibly mistreating an animal. I know I wouldn’t. There is another column asking for your honest answer. It is very possible that I may have left my dog Lana out in the cold unintentionally. But if I admitted that, I might not look good in the eyes of animal lovers. Do you get the point?

In the next column you will give the honest answer that is not socially desirable. But it’s okay because your honest response is needed for me to run statistical tests to determine the answer that is most socially desirable.

Finally, the questions are written for educators, but don’t worry, you do not have to be an educator to complete the scale. In fact, it is helpful to me if you are not an educator. If you are an educator no worries.

Thank you for helping me validate the adapted brief social desirability scale.