Information Technology Adoption by Principals in Botswana Secondary Schools Angelina Totolo

Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2007 Information Technology Adoption by Principals in Botswana Secondary Schools Angelina Totolo

Follow this and additional works at the FSU Digital Library. For more information, please contact [email protected]

FLORIDA STATE UNIVERSITY

COLLEGE OF INFORMATION

INFORMATION TECHNOLOGY ADOPTION BY PRINCIPALS IN BOTSWANA SECONDARY SCHOOLS

ANGELINA TOTOLO

A dissertation submitted to the College of Information in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Degree Awarded: Summer Semester, 2007

The members of the Committee approve the dissertation of Angelina Totolo on June 15, 2007.

------Kathleen Burnett Professor Directing Dissertation

------Alysia Roehrig Outside Committee Member

------Eliza Dresang Committee Member

------Darrell Burke Committee Member

------Gary Burnett Committee Member

Approved:

Larry Dennis, Dean, College of Information

The Office of Graduate Studies has verified and approved the above named committee members.

This dissertation is dedicated to my family, to my husband, Dr. Otlogetswe Totolo, whose love, dedication and sense of humor encouraged me to go on. To my son Tshepo, who is always very supportive and loving in his quiet way. To my loving and very courageous daughter, Refilwe, who endured many years without the love of a mother at a tender age. Her strength and endurance made me all the stronger to continue my studies. Above all, I thank the Almighty God for his guidance throughout my studies.

iii

ACKNOWLEDGMENTS

I wish to express my sincere and deep appreciation to my major professor, Dr. Kathleen Burnett, who was my mentor and my advisor. I am deeply indebted to her unwavering support both professionally and personally, and to her prompt, perceptive and thorough review of my dissertation. Her smile each time I met her for our weekly meetings encouraged me to go on.

To my outside committee member, Dr. Alysia Roehrig, my committee members, Dr. Gary Burnett and Dr. Darrell Burke, I want to say thank you very much for enhancing my work with your expertise and knowledge.

I wish to express my appreciation to Dr. Eliza Dresang, my committee member, for her unfailing support both academically and personally. I am deeply indebted to her kindness, her concern for my progress and her professionalism.

To Mrs. Kemotho Mabe and Ms. Gaothobogwe Malepa, whose help with the surveys and the setting up of interview appointments was invaluable, I express my appreciation for their immeasurable contribution.

To Mannana Totolo, Molaodi Totolo, Tshegofatso Motsemme and her family, Constance Mabile and her family, Dineo Setshogo and her family, and Galeagelwe Baikepi, I wish to express my appreciation for their unfailing support throughout my studies.

To the participants, the principals of Botswana secondary schools, I say thank you very much for making this study a reality. To Mr. Abraham Senabye, for his invaluable help with the data collection; I express my sincere appreciation.

To the University of Botswana, I want to say thank you for giving me an opportunity to study.

TABLE OF CONTENTS

LIST OF TABLES...... vii

LIST OF FIGURES...... viii

ABSTRACT ...... ix

CHAPTER 1: PROBLEM CONTEXT ...... 1 Problem Statement...... 2 Significance of the Study...... 4 Methodology ...... 5 Instrument Pre-testing and Piloting...... 6 Research Objectives and Questions...... 11 Definition of Terms...... 12 Assumptions...... 17

CHAPTER 2: LITERATURE REVIEW ...... 19 Theories of IT diffusion and adoption...... 19 Barriers to IT Diffusion ...... 25 Botswana ...... 30 IT Leadership and Principals ...... 34 Problem in relation to the literature...... 36

CHAPTER 3: RESEARCH DESIGN AND METHODOLOGY ...... 38 Research Design...... 38 Methods ...... 43 Population of the Study...... 45 Data Collection and Dissemination...... 46 Techniques for maximizing the return rate of survey ...... 48 Techniques for collecting useful data from interviews ...... 48 Techniques to Ensure Reliable and Valid Data...... 49 Instrument Validity and Reliability...... 50

CHAPTER 4: SURVEY ANALYSIS AND RESULTS ...... 53 Description of the Research Population...... 54 Regression Tests ...... 57 Hypotheses Testing ...... 58 Distinct Hypothesis and Research Questions ...... 61 Analysis and Summary of the Survey...... 64 Chapter Summary...... 67

CHAPTER 5: INTERVIEW ANALYSIS AND RESULTS...... 69 Type of Interview Questions ...... 69 Computer Anxiety: Self Descriptions...... 72 Perceived Usefulness: Self Descriptions ...... 77 Perceived Ease of Use: Self Descriptions...... 79 Part 2: Future of computers in schools ...... 85 Uses of the computer: Self descriptions...... 86 Future plans for schools ...... 90 Analysis of Interview Responses...... 93

CHAPTER 6: CONCLUSIONS AND FUTURE RESEARCH ...... 100 Conclusion...... 105 Implications of the study...... 106 Limitations of the Study ...... 107 Future Research ...... 108

APPENDIX A: DESCRIPTIVE DATA...... 109

APPENDIX B: TEST OF NORMALITY...... 110

APPENDIX C: TAM STATEMENTS...... 111

APPENDIX D: REMOVED OUTLIERS...... 112

APPENDIX E: TOLERANCE AND VIF RESULTS...... 113

APPENDIX F :SCATTERPLOT...... 114

APPENDIX G: REGRESSION RESULTS...... 116

APPENDIX H: QUESTIONNAIRE ...... 117

APPENDIX I: CONSENT LETTER ...... 119

APPENDIX J: INTERVIEW CONSENT LETTER ...... 120

APPENDIX K: HUMAN SUBJECTS APPROVAL MEMORANDUM ...... 121

APPENDIX L: TIMELINE...... 122

APPENDIX M: INTERVIEW LEAD QUESTIONS ...... 123

REFERENCES ...... 124

BIOGRAPHICAL SKETCH...... 131

LIST OF TABLES

TABLE 1: VARIABLE TYPE AND SOURCE 1 ...... 42

TABLE 2: DEMOGRAPHIC DATA 1...... 55

TABLE 3: PU AND PEOU ITEM STATISTICS 1...... 60

TABLE 4: PART 1 INTERVIEW OBJECTIVES 1...... 71

TABLE 5: NERVOUSNESS ANXIETY 1...... 74

TABLE 6: FEELING AT EASE ANXIETY 1 ...... 75

TABLE 7: FEELING THREATENED ANXIETY 1 ...... 75

TABLE 8: SINKING FEELING ANXIETY 1 ...... 76

TABLE 9: UNDERSTANDABLE PEOU 1...... 83

TABLE 10: LEARNING EASY PEOU 1 ...... 83

TABLE 11: PART 2 INTERVIEW OBJECTIVES 1...... 86

TABLE 12: COMPUTER ANXIETY STATEMENTS 1 ...... 109

TABLE 13: PERCEIVED USEFULNESS (PU) 1...... 109

TABLE 14: PERCEIVED EASE OF USE STATEMENT 1...... 109

TABLE 15: DEPENDENT VARIABLE 1 ...... 109

TABLE 16: SHARPIRO-WILK TEST OF NORMALITY 1...... 110

TABLE 17: PU AND PEOU ITEM STATISTICS 1...... 111

TABLE 18: CASEWISE DIAGONISTICS 1 ...... 112

TABLE 19: MULTICOLLINEARITY RESULTS 1...... 113

TABLE 20: REGRESSION MODEL SUMMARY 1...... 116

vii

LIST OF FIGURES

FIGURE 1: TECHNOLOGY ACCEPTANCE MODEL...... 24

FIGURE 2: EXTENDED TAM MODEL ...... 25

FIGURE 3:EXTENDED TAM MODEL...... 42

FIGURE 4: BAR GRAPH FOR INTENTION TO ADOPT ...... 63

FIGURE 5: COMPUTER COMFORT SELF DESCRIPTIONS...... 77

FIGURE 6: PERCEIVED USEFULNESS...... 79

FIGURE 7: PERCEIVED EASE OF USE...... 84

FIGURE 8: COMPUTER USES BY ALL GROUPS ...... 90

FIGURE 9: FUTURE PLANS FOR THE SCHOOL ...... 94

FIGURE 10: SCATTER PLOT ...... 115

FIGURE 11: NORMAL P-P PLOT...... 115

viii

ABSTRACT

This research investigated the likelihood of computer technology adoption in Botswana, among school principals in secondary schools, who are assumed to be transformational leaders. The Technology Acceptance Model (TAM) survey and an interview were used to determine the perceptions of the school principals about accepting and using computer technology. The survey was used to predict and explain the principals’ acceptance of computers in relation to whether they find them useful, easy to use, and if they intended to adopt and use them. The in-depth follow up semi-structured interviews explored further the responses of the survey questionnaire. The results of the TAM survey based on the regression model with an R square of .273, show that there was substantial support that the participants who found computers easy to use and useful in their job intended to adopt and use them. The interviews addressed why principals rejected or adopted computers and how they intended to enhance computer use in the school. The results show that time constraints, phobia, lack of skills or training and the lack of practice with computers were identified as barriers to adoption in this study. The results of the study confirmed that the research population was not homogenous; there were early adopters, who showed characteristics of transformational leadership as well as late adopters and non adopters who were still learning how to use computers. Therefore training on the use of computers should include strategies to alleviate barriers to computer adoption. This study has implications for the Vision 2016 because Botswana has already made a significant investment in information technology and information technology infrastructure for its secondary schools (The Revised National Policy on Education, 1994).

CHAPTER 1: PROBLEM CONTEXT

The study examined information technology (IT) adoption readiness, more specifically computer technology, among principals in Botswana secondary schools. It is motivated by the promulgation of the Long Term Vision 2016 (hereafter, Vision 2016), which aims to make Botswana an information society (Long Term Vision 2016, 1997). The study employed the Technology Acceptance Model (TAM) as proposed by Davis (1986) to determine the perceptions of the school principals about accepting and using computer technology, using three constructs; perceived ease of use (PEOU), perceived usefulness (PU) and behavioral intention to adopt (BI). TAM has since been extended to include the computer anxiety construct (Compeau & Higgins, 1995). This study used the extended TAM, which includes computer anxiety as a variable of study. The aim of this research was to investigate the likelihood of computer technology adoption in Botswana, specifically among school principals in secondary schools. In this study, information technology is understood to be the umbrella term that encompasses all computer technology, as explained in the definition of terms section.

The principals are regarded as transformational leaders, which is why they were chosen for this research. According to Todd (1999), “Transformational leadership moves beyond managerial and instructional leadership to providing schools with strategies necessary to cope with change” (p. 4). Writing on the characteristics of transformational leaders, Todd (1999, p.5) says that they center on transforming the environment, developing a shared vision and strong philosophy, emphasizing school based management as well as keeping abreast with new trends and developments. Doyle & Smith (2001,Transformations, ¶ 1) share this view of transformational leadership when they describe transformational leaders as visionary people and change agents. Doyle & Smith (2001,Transformations, ¶ 1) argue that leaders emerge during a crisis or when there is a need to make an important decision for the survival of the organization. In this study, principals were assumed to be in a position to transform the school system through computer technology; therefore

knowing whether they intend to adopt computers is crucial for technology implementation. Transformational leadership is therefore a very important concept in the explanation of the adoption and use of computers by school principals, because according to this concept, leaders should be seen to be in the forefront of computer technology adoption and use. This entails having strong philosophies and policies in place to guide the process of computer adoption and effective use.

TThe problem statement, in the next section, expands on the nature of the problems associated with computer technology adoption and information technology acceptance in general. As explained above, computer technology is used as a subset of the umbrella term, information technology, in this study.

Problem Statement

The fact that Sub Saharan Africa is both technologically and economically less developed than countries in Europe and America, and that this has led to the slow adoption of information technology, is well-established in the literature (Odedra et al., 1993; Onyango, 2000; Peterson, 1991; Udo & Edoho, 2000). Although Sub Saharan Africa has not responded very well to information technology adoption and other modernization attempts in the past (Onyango, 2000; Peterson, 1998), governments in Africa continue to invest in information technology. Among the countries that are currently attempting to promote information technology adoption, is Botswana. Botswana has developed a Vision 2016, which aims at turning the country into an information society or an “educated, informed nation” by the year 2016 (Long Term Vision 2016, p. 71). The achievement of this goal of an informed nation implies the adoption, diffusion and use of information technology in the nation. As the vision document states, the adoption of information technology is paramount to the achievement of the Vision 2016 goals. One of the objectives of the Vision 2016 document is “All Batswana {Batswana refers to the people of Botswana} to use and apply the potential of computer equipment in their daily life” (p. 35). Therefore research on the perceptions of IT adoption among school principals, who are regarded as transformational leaders in this study, is timely.

IT adoption is a subject of worldwide discussion as evidenced by the extensive literature examining what factors influence its adoption failure or success. Flanagan & Jacobsen (2003) say that “Despite government spending on boxes and wires, technology integration in North American classrooms has ranged from uneven at best to nonexistent in some cases” (p. 125). Research from the African perspective (Akpan, 2000; Alemna, 1999; Heeks, 2002; Jain & Mutula, 2001; Jimba, 2000; Odedra et al, 1993; Onyango, 2000; Thapisa & Birabwa, 1998; United Nations Economic Commission for Africa, 2001) and the international perspective (Baskerville & Pries-Heje, 2001; Comin & Hobjin, 2004; Karahanna & Straub, 1998; Kukafka, et al, 2003; Legris, et al, 2003; Rogers 1995; Russell, 2004) has dealt with the causes of failure to adopt and use information technology successfully.

In the case of Botswana, information technology is a relatively new concept, and therefore adoption must precede use. As the above authors have noted, IT adoption and use has not been easily accomplished. Therefore, it is important that Botswana’s readiness to fulfill its objective of achieving an educated and informed nation as outlined in Vision 2016 be examined and assessed. Adoption of computers in the secondary schools is one indicator of readiness that can be measured early in the course of implementation of the Vision 2016.

This research asked the question: Given that information technology, more specifically computer infrastructure, is in place in Botswana’s secondary schools, do the principals of these schools intend to support adoption and use by themselves, the teachers, and the students? Since education is the most obvious means through which adoption and use can be positively influenced, early adoption and use by secondary school principals is likely to be a strong predictor of success or failure.

This study will add to the existing research on information technology adoption in general, and to that on Botswana, Sub Saharan Africa, and developing countries in general, as discussed in the next section.

Significance of the Study

Governments across Africa are grappling with problems related to computer technology implementation. Many people assume that once the hardware and the software have been purchased, computer technology adoption and its effective use will follow, but this has not proven to be the case in Africa, as demonstrated in the extensive literature examining the factors that influence IT adoption failure or success. Internationally, researchers have demonstrated that computer anxiety, lack of perceived usefulness, and lack of perceived ease of use of IT have led to low adoption and usage of information technology (Davis, 1989; Karahanna and Straub, 1998; Legris et al, 2003; Straub, et al, 1997; Szajna, 1996; Venkatesh and Morris, 2000). These factors have serious implications for policy implementation failure and seem to confirm the suggestion that while governments may be well-intentioned and sincere in their attempts to make technology available, this is not enough to guarantee its use (Odedra et al, 1993).

Scholars have written about the benefits of adopting new information technology in Africa, extolling the economic, social, and political benefits to be gained from its diffusion (Powell, 1992; Udo & Edoho, 2000). They have also pointed out that though IT adoption can bring economic benefits, the cost of implementing IT might not match the benefits of using it. Writing on the issue, Heeks (2002) says, In a very direct sense, failure is also a problem because of the opportunity costs of resource investment in failure, as opposed to success. Such opportunity costs are likely to be particularly high in DCs [Developing Countries] because of the more limited availability of resources such as capital and skilled labor.” (p. 103)

Botswana has already made a significant investment in information technology and information technology infrastructure for its secondary schools (The Revised National Policy on Education, 1994). It is important to find out prior to more extensive investments, whether the technology is being adopted by the principals, teachers and students for whom its use is intended. Botswana needs to know whether principals are successfully implementing adoption of the computer technology in place in their schools,

in order to inform strategic planning and implement cost-effective policy decision- making.

This will be the first study of this kind carried out in Botswana. Other studies have dealt with the information seeking behavior of various groups in Botswana (Grand, 1997; Moahi, 2000; Mooko, 2002) and establishing the Information and Communication Technologies (ICT) training needs of teachers in Botswana (Bose, 2004). Other literature treats IT policy infrastructure, in terms of delay in information technology adoption, thereby suggesting possible remedies in the form of policy formulation, human resources and physical infrastructure (Jain & Mutula, 2001). Notwithstanding the useful contribution by the various scholars writing on IT, there have not been any empirical studies conducted to find out what perceptions users have towards adopting IT and what implications their perceptions would have on IT policy implementation failure or success in Botswana. This study endeavored to explore the IT adoption scenario in Botswana through examination of its use by secondary school leaders.

In order to achieve the objectives of this study, questions were formulated according to the extended Technology Acceptance Model as shall be demonstrated in the next section.

Methodology

The research population is the principals of the 233 secondary schools in Botswana; this number includes all the government secondary schools in the country. Out of the 233 schools, 206 are community junior secondary schools (the equivalent of U.S. middle schools) and 27 are senior secondary schools (the equivalent of U.S. high schools). As far as the qualifications of the principals are concerned, there is no difference between the senior and junior schools; they hold the same qualifications and they all have equal chances of working in any of the two types of schools, therefore they move between the two types of schools. Any principal can be posted to any secondary school in the country, be it a junior or senior school. The use of the simple random sample for the combined school types is therefore justified in this study but further investigation on the data

collected will determine whether there is a difference between senior and junior school principals.

This population was selected because it is assumed that principals are leaders in education and therefore they play an important role in the adoption and use of computer technology. This assumption is based on the literature which has established that school principals are expected to be change agents and to be effective leaders in the information era (Telem, 1996; Todd, 1999; Yee, 1998, Yuen et al. 2003). Telem (1996), writing on the importance of leadership in the digital era says, MIS implementation research findings indicate that top management's, i.e. the principal's, support is associated with MIS implementation success. S/he should be involved in SMIS's initiation, implementation and day-to-day operation. Under his/her leadership, the school's computerization policy should be formulated and his/her and other school staff's information requirements should be clearly specified. (p. 89)

The Botswana government has recently equipped all secondary schools with computer laboratories (Revised National Policy on Education, 1994). Therefore this study, which aimed at finding out if computers were adopted in schools, is necessary for strategic planning and technology implementation in the school system.

Instrument Pre-testing and Piloting

In October 2004, a pre test survey was administered to 10 secondary school principals in the Gaborone schools. This was done in order to assess content, construct validity, and reliability of measures. The pre-test results revealed a positive response towards the adoption of computers as well as transformational leadership qualities amongst the school heads, which differed from the literature consulted. After the pre-test was administered and analyzed, the research instrument was altered by changing the “yes and no” questions to a scale with five options; strongly disagree, disagree, not sure, agree and strongly agree for the TAM questions and very low priority, low priority,

medium priority, high priority, and very high priority for leadership questions. Also added to the instrument were two open-ended questions in order to capture the school heads’ perceptions of the role of the school library in the integration of information technology in the school. This instrument was used to collect data from 24 principals and a research paper entitled, “Information Technology Adoption in Botswana Secondary Schools and its Implications on Leadership and School Libraries in the Digital Era” was written and presented at IASL Conference in Hong Kong (Totolo, 2005). This pilot study yielded results similar to those in the pre-test, which strengthens the certainty that this instrument is both valid and reliable. In addition, TAM has been used many times before, and has been found to be robust in explaining technology acceptance.

The pilot instrument was adopted for the present study with a few changes. Consistent with TAM constructs, the questions remained largely the same for the TAM, except for rewording and addition of one TAM question to capture behavioral intention aspects. The other change was the removal of transformational leadership questions, school library demographics and the open-ended questions. These were replaced by interviews to be conducted with a purposive sample of ten (10) of the survey participants, which would capture more in-depth information about the principals’ intentions to adopt and use computers. This study is therefore a mixed method study as explained below.

In October 2006, a pilot testing of the interview instrument was done by telephone. Six pilot interviews were carried out before the actual interviewing took place. Several cultural problems were experienced with the original interviewing method that was adapted from Dresang (1990) for this research because micro moments and backward chaining did not suit the conversational style of the principals in Botswana. Therefore the method of interview was changed to a semi structured telephone interview using the TAM constructs as the main structure of the interview. An example of the statements used in the interview is as follows, “I will start off by refreshing your memory about the type of interview we are going to have as per the letter of consent I sent with the questionnaires. I am thankful that you also responded to that questionnaire. Today I am

7 largely confirming the results of that questionnaire. I got different types of answers. Some people felt comfortable with computers and others didn't. Some thought the computer was useful for them and others didn't think so. The same goes for the ease of use, some people said it was easy to use and others said it wasn't. So I am just confirming the results from your perspective. Can I ask how you feel about computers? What is your comfort zone concerning computers?”

Mixed Methods Design

A mixed methods design, which combines quantitative data collection and analysis using the Technology Acceptance Model (TAM) survey, and qualitative data collection and analysis of interviews, was done. The survey was used both to explain causality between the dependent and independent variables, and to test hypotheses (Bryman, 1984). The interviews were used for an in-depth analysis of the participants’ perceptions towards adopting and using computer technology. The TAM survey as a tool for research is not enough to present a holistic picture about phenomena because it basically addresses whether people want to adopt a technology and not why or how they intend to adopt it. Therefore, telephone interviews were conducted to address the why and how questions. The mixed method, according to Morse (2003), is “the incorporation of various qualitative and quantitative strategies within a single project that may have either a qualitative or a quantitative theoretical drive” (p. 190). In compliance with Morse’s explanation, this study collected data using surveys and interviews. The use of two methods allowed the researcher to collect both qualitative and quantitative data. The Explanatory Design, which is used to collect information sequentially, was used in this study (Creswell, 2002). That is, quantitative information was collected using a survey, and then qualitative questions were developed from the gaps found in the survey. The theoretical thrust of the study was quantitative, therefore the survey questionnaire was used first to collect demographic and opinion related data, using TAM constructs, and the interviews were used next to collect in-depth responses to “enlighten or provide clues that are followed up within the core method” (Morse, 2003, p. 190), concerning the

principal’s perceptions towards adopting computers in the school. The interview questions consisted of modifications and expansions of the survey questions.

Survey

The social survey is the preferred instrument for this kind of research. This method is used to explain causality between variables, the relationship between the dependent and independent variables, as well as the testing of hypotheses and established theories (Bryman, 1984). In this case the theory tested is the Technology Acceptance Model. TAM, as a theoretical framework, has been previously used in many studies (Davis, 1989; Karahanna and Straub, 1998; Straub et al., 1997; Szajna, 1996; Venkatesh and Morris, 2000).

The survey is used to predict and explain user acceptance of computers in relation to whether users find them useful, easy to use, and if they intend to accept or adopt them. Parker (1974), writing on the purposes of the survey in information technology research says, Changes in information technology can be examined either as a dependent variable of social science interest (what are the causal factors in our culture leading to changes in information technology?) or as an independent variable (what social effects follow from changes in information technology).” (p. 590)

As mentioned above, the most used quantitative method instrument is the survey. Surveys have both strengths and weaknesses. The strengths are, 1. Surveys are good in external validity. This validity refers to how the approximate and presumed causal relationships can be generalized to and across different types of groups of people, settings, and times (Cook and Campbell, 1979). 2. Surveys are useful for describing the characteristics of large samples (Babbie, 2001). 3. Survey questions can be standardized and re-used for different contexts (Babbie, 2001).

TAM concepts use a survey method; however, the survey method is bound to have gaps that can be supplemented with other methods because no single method can answer all the questions about a phenomenon. According to Babbie (2001, p. 268), the weaknesses of surveys are as follows: 1. Surveys are inflexible because once an instrument is validated; it has to be administered as it is. 2. Surveys can be too artificial in that respondents can give wrong information which does not reflect the real situation. 3. Surveys do not measure social action; they only report what the respondents say.

The above survey weaknesses justify supplementing surveys with qualitative methods like interviews.

Interviews

The purpose of qualitative inquiry is to study society from a close-up view in order to explain why events under observation happen in that setting and how patterns or theories emerge from the observation of the particular society. This study included in-depth follow up semi structured interviews with a purposive sample of ten (10) principals to further explore their responses to the survey questionnaire. Although the results from the interviews cannot be generalized to the whole population under the study, this method has the advantage of increasing interaction between the researcher and the interviewee, which can enhance the quality of data collected. The interviews were chosen for this study because these allow the participant to express themselves within some structure but at the same time having the latitude to expand on their explanations (Babbie, 2001). According to Bill (1977) semi structured interviews can outperform a survey that is based on the same questions in terms of predictive efficiency (p.19). Also participants can share their experiences beyond what the researcher had anticipated, thereby making the data richer (Creswell, 2002).

Research Objectives and Questions

Research Objectives

1. To explain the perceptions of the principals towards adopting computer technology through the following: a. To find out if school principals display feelings of computer anxiety. b. To describe the principals’ perceptions of the usefulness of computers in their work. c. To find out if school principals find computers easy to use. d. To predict if school principals intend to use computer technology. 2. To find the gaps created by the survey analysis 3. To find out how the principals of these schools intend to support adoption and use by all users in the school.

Research Questions

Consistent with the extended TAM constructs, the distinct hypothesis to be tested is: Secondary school principals who are adopters of computers will: (1) adopt computer technology for use in their work; (2) exhibit positive feelings about the usefulness of computers in their schools; (3) exhibit positive feelings about the ease of using computers in their schools; and (4) lack anxiety about computer adoption and use.

The first research question and its sub question were based on TAM and are addressed in the survey, while the last question was dealt with through the interviews. The research questions were as follows; 1. What are the perceptions of the principals towards adopting computer technology in their schools? To address this question, four sub questions were posed: a. Do school principals report feelings of anxiety about computer adoption and use?

b. What are the principals’ perceptions of the usefulness of computers in their schools? c. Do school principals report that computers are easy to use? d. Do school principals intend to adopt and use computer technology in their schools?

The questions above dealt with the survey. The following questions (2 and 3) were addressed in the interviews. Question 2 was referred to as Part 1 and Question 3 became Part 2 in the interview analysis in Chapter 5 of this dissertation. 2. What are the gaps found in the analysis of the TAM survey? Three questions were posed for this question. a. Are school principals comfortable with using the computer? b. Do some principals think that computers do not increase their job performance? c. Do some principals think that it is not easy to learn how to use the computer?

3. What are the actions school principals intend to take to support computer adoption in the school? A question designed to solicit the principals’ perceptions towards computer adoption and use in the school is, “What are some actions you intend to take to support adoption and use of computer technology by yourself, the teachers, and students in your school?” To address this question, two questions were posed:

a. What are the principals’ self reported uses of the computer? b. What are the principals’ self reported plans for the future? Research Question 2 and 3 were answered using the qualitative interview method. Interview questions were formulated after analyzing the survey responses.

Definition of Terms

Definitions for the terms most often employed in the study (innovation diffusion, IT, ICT, and Information Technology diffusion, transfer, adoption, acceptance, and IT adopter) are provided next.

Innovation Diffusion

Innovation diffusion is an umbrella term used to talk about all technology diffusion, not just IT diffusion. Rogers (1995) and Barnett (1953) both call the spread of a new idea or invention innovation diffusion. According to Rogers (1995, p. 5), Diffusion “is the process by which an innovation is communicated through certain channels over time among members of a social system.” Barnett defines innovation as “any thought, behavior, or thing that is new because it is qualitatively different from existing forms” (p. 7). Barnett goes on to emphasize cultural change and how a new innovation alters the cultural setting. Rogers refers to diffusion in a general sense to mean, “a planned and spontaneous spread of ideas,” nonetheless, both Barnett and Rogers agree on the meaning of diffusion to refer to the spread of ideas. In this study, innovation diffusion will be used to refer to the spread of a new idea or innovation from one area to another, using the communication channels in the receiving society.

Transfer

The term diffusion seems to be used infrequently in the literature that was consulted dealing with information technology diffusion in Africa. Instead, the word transfer seems to be more prevalently used in the literature on the diffusion of information technology in the least developed countries. Scholars writing on Africa refer to IT diffusion as transfer (Udo & Edoho, 2000). Baskerville & Pries-Heje (2001) appear to use diffusion and transfer interchangeably. According to them, the diffusion of a bundle of technology (hardware) and knowledge (software) can be called technology transfer. Generally, all scholars who use the word transfer express the same concept as diffusion (Kedia & Bhagat, 1988; Klauss, 2000; Udo & Edoho, 2000). It can be argued that Barnett (1953) agrees with the scholars above because he says after an idea is transferred across borders, then it gets diffused. He uses both the words “transfer” and “diffusion.” There therefore seems to be overall agreement that both terms can be used to mean the spreading of an innovation from one culture to another. In this prospectus, diffusion and transfer shall be used to mean the spread of innovations from one culture to another.

Adoption and Acceptance

Adoption and acceptance seem to be used interchangeably in the literature consulted. Barnett (1953) contends that if an idea grows in popularity in the same place it originated, that process is called adoption or acceptance and if it is transferred to other places across ethnic boundaries, then that is “spreading, borrowing or diffusion” (p. 291). Barnett goes on to say that if two ethnic groups are involved in the transfer, then that is called “assimilation” or “acculturation”. He, like Rogers (1995), argues that, “a study of cultural change takes us beyond the appearance of a new idea into considerations of its acceptance and rejection” (p. 291) and that innovation diffusion takes place “on a mental plane,” thereby underlining the importance of perception, cognition, recall and affect during the innovation diffusion process (p. 181).

According to Dasgupta, et al, (2002, p. 87), information technology adoption is “defined as the decision to accept, or invest in a technology.” This definition includes the word accept, which probably explains why most scholars consulted use the words acceptance and adoption interchangeably. According to Shavo & Igbaria (2003), “The terms adoption, use, consumption, and acceptance are used interchangeably. Organizational adoption looks at adoption by aggregates: viz. departments, agencies, Strategic Business Units (SBUs), or companies” (Definitions, ¶ 1). The other important thing to note here is that one of the most used technology adoption models is called the technology “acceptance” model, which scholars use to examine technology “adoption”, thereby underlining the fact that adoption and acceptance mean the same thing (Gefen & Straub, 1997; Venkatesh, 2000; Venkatesh & Morris, 2000). In this prospectus, acceptance and adoption shall be used to mean the decision to accept and use an innovation, in this case information technology.

From the above explanation, there seems to be a difference between diffusion or transfer and adoption or acceptance. Transfer or diffusion appears to be the initial stage that comes before the adoption or acceptance stage. First of all, technology has to be brought into a nation or transferred (diffused), and then individuals or users choose to adopt or

accept the transferred technology. Therefore, it appears that the transfer of information technology happens at a national level, between governments involved in the transfer, and then the users and the organizations or institutions come in next to adopt and use the technology. In this prospectus, transfer will be used to talk about diffusion at an international and national level, while adoption will be used to talk about technology acceptance at an organizational and individual level, in accordance with the previous literature.

Information and Communication Technologies

Generally speaking, Information and Communication Technologies (ICTs) is an umbrella term, which refers to the equipment, infrastructure, and the communication used to make information in all its different formats, accessible to users. According to Rodriguez & Wilson (2000), ICTs refer to “the set of activities which facilitate by electronic means the processing, transmission and display of information” (p. 5). Akpan (2000) approaches the definition from four perspectives, that of services, applications, and hardware, and communication. The above definitions underline four components of the definition of ICTs; namely the machinery or equipment, the infrastructure or software applications, the communication and services made available by technology.

Information Technology

Information Technology (IT) can be defined as a phrase that covers all forms of technology that are used to create, store and distribute information in all its formats, mostly using computers. “According to NDCC 54.59.01 Information Technology means the use of hardware, software, services, and supporting infrastructure to manage and deliver information using voice, data, and video” (Definition of Information Technology, ¶ 1).

According to Friedman (1994),

The information technology field may then be defined as the social space structured around the production, use, definition, and control of information technology. Information technology is the base technology for information systems (IS). It includes the designs and characteristic materials used in computer hardware, software, and peripheral equipment as well as the bundle of techniques for developing, implementing, and maintaining computer-based information systems (Technology Fields and the Information Technology field ¶.4).

Friedman (1994), in the quote above, defines information technology as a field, and this gives the definition a specific context. However the basic theme underlying his definition is the use of computers for the manipulation of information.

In summary, ICTs and IT pose a challenge in their definition because of the various meanings attached to them by different scholars. Therefore the best approach is to use a functional definition. The use of the terms, ICTs and IT, do not seem to be very different. The definition of ICTs emphasizes the communication component and includes all forms of communication equipment, while the definition of IT emphasizes the use of computers as the medium of communication more than any other form of equipment. In this discussion there will be no difference in meaning between the two terms, both ICT and IT shall be taken to mean all hardware and software infrastructure that is used for the manipulation, creation, storage and communication of information in all its formats, using the computer and other supporting equipment.

Computer Technology

Most sources consulted do not define computer technology, although it is used in a substantial number of scholarly papers. All the reference sources consulted define the words computer and technology separately. Encyclopedia Britannica Online defines a computer as “Programmable machine that can store, retrieve, and process data” (¶ 1) and technology as “the application of scientific knowledge to the practical aims of human life or, as it is sometimes phrased, to the change and manipulation of the human environment” (¶ 1). The definitions of ICT and IT above both include the computer and

technology to refer to the machine and the communication or manipulation of information respectively. In this study, computer technology shall be used as a subset of both IT and ICT to refer specifically to the machine (computer) and all the uses of the computer to manipulate and use information (technology).

Adopter

In this study, an adopter is an individual who has accepted computer technology use. According to the Merriam-Webster Dictionary Online (2006), the word adopter means “to take an opinion, policy, or practice as one's own. ADOPT implies accepting something created by another or foreign to one's nature” (Synonyms, ¶ 1). This definition underlines the importance of accepting an innovation that does not originate in one’s place.

Computer Technology acceptance or adoption

According to the TAM, system usage behavior is determined by the intention to use a particular system, which in turn, is determined by the perceived usefulness and perceived ease of use of the system (Luarn and Lin, 2005). A person has accepted computer technology if they are willing to use it. Assumptions

School principals are chosen for this study because they are viewed as transformational leaders in the literature consulted (Doyle & Smith, 2001; Todd, 1999; Yee, 1998; Yuen et al., 2003). The use of Technology Acceptance Model constructs in this research is specifically meant to target leaders who are expected to have a positive influence on the implementation of government policy. The assumption here is that principals are in a position of influence both as opinion leaders and as administrative leaders in the education environment. Leadership by the principals in the digital era is crucial to the spread and use of computer technology in schools, therefore their technology acceptance situation is crucial for policy implementation planning. In a paper in which Yuen et al. (2003) explored the leadership issues in ICT implementation in schools, they came to the

17 conclusion that implementation involves stakeholders and issues of leadership. Writing on the same issue, Yee (1998), says that the role of the principal has changed from that of an instructional leader to a transformational leader and that a transformational leader, “focuses on building a shared vision, improving communication and making decisions collaboratively” (p. 57). Also, one of the objectives of the Vision 2016 is “All Batswana to use and apply the potential of computer equipment in their daily life” (p. 35). Computer acceptance by school principals is likely to facilitate the acceptance of computers by all the teachers and students, who look up to the leadership of principals, and this in turn will help with the achievement of this Vision 2016 goal.

The literature dealing with the context of the above problem statement, significance of this study, and the objectives of this research is discussed next.

CHAPTER 2: LITERATURE REVIEW

The literature reviewed in this section describes the theories of IT diffusion and adoption, the barriers to IT diffusion, the IT diffusion situation in Botswana, Botswana’s socio- political environment and education system, IT leadership by principals and the problem in relation to the literature.

Theories of IT diffusion and adoption

Diffusion and adoption theories have centered on the society and its environment, as well as the individual as a unit of analysis in order to assess the success of the diffusion. The literature on IT transfer in Africa has mostly centered on the impediments or barriers to IT diffusion at the national level. Thus, the literature on the barriers to IT transfer or diffusion in Africa has emphasized political, social, and economic barriers.

IT adoption literature covered in this chapter includes studies that explore IT diffusion internationally using innovation diffusion theories, and adoption or acceptance of IT using behavioral intention theories (specifically, the Technology Acceptance Model (TAM)), and IT transfer in Africa ( including Sub Saharan Africa, and emphasizing Botswana).

Diffusion Theories

Rogers (1995) introduced the Innovation Diffusion Theory (IDT) and he mentions the influence Tarde (1895) had on the origins of his theory, since Tarde came up with the Laws of Imitation and the Theory of Invention. Rogers (1995), following Tarde’s theory (1903), states that “…the more similar an innovation is to the ideas that have been accepted; the more likely the innovation will be accepted” (p. 40). Therefore, diffusion theories tend to address the circumstances that influence the adoption or rejection of an innovation.

In addition, Rogers mentions the German-Austrian and British schools of diffusionism in anthropology as contributors to his theory (Rogers, 1976, p. 290). Rogers uses innovation diffusion to conduct communication research (Rogers, 1976, p.292). However, this theory encompasses economic, social, communication, and behavioral issues, and has thus been adapted to research in all of these areas, which makes it difficult to pin it down to a single purpose or field. Innovation Diffusion Theory (IDT) is therefore best described as a “theoretical paradigm” (Rogers, 1976, p. 291). It has also been called a meta-theory (Surry & Farquhar, 1997).

Rogers’ Innovation Diffusion Theory (1995) has four concepts: innovation, communication channels, time and social system. Each of these concepts is represented by variables that can be used to predict whether a society will diffuse technology or new ideas that are introduced in that particular social system. According to Surry & Farquhar (1997), diffusion research investigates how the four major factors; innovation, innovation communication, time and the social system interact to facilitate or impede the diffusion of an innovation in a particular group. Surry & Farquhar go on to say that research carried out by Ryan & Gross in 1943 provided the genesis of the Innovation Diffusion Theory. Within the IDT meta-theory are four theories: Innovation-Decision Process, Individual Innovativeness, Rate of Adoption, and Perceived Attributes (Surry & Farquhar, 1997, p. 3). These theories have been broken down into five categories, also known as characteristics of an innovation as perceived by the individual. These are perceived attributes of innovation (relative advantage, compatibility, complexity, trialability and observability), type of innovation decision, communication channels, nature of the social system and the change agents (Rogers, 1995, p. 207).

IDT and Society as a Unit of Analysis

Scholars who have undertaken research using the four theories of diffusion as proposed by Rogers, have mostly turned to behavioral sciences to operationalize the innovation of diffusion concepts. This is because the behavioral sciences paradigm posits that an approach or a new idea can be accepted if it is designed taking into consideration the most important factors of user acceptance behavior in a given situation (Kukafka, et al,

2003). The unit of analysis of the IDT meta-theory is the society, and this approach enables the researcher to move the unit of analysis to the individual, in order to find out what characteristics of the society or its environment are likely to impede or facilitate the diffusion, acceptance or adoption of information technology by individual members of that society.

Innovation Diffusion Theory has guided research in Africa and other developing countries designed to explain and describe the lack of diffusion of innovations in societies. One example of its use is in explaining the diffusion of bottle-feeding in Africa (Rogers, 1995). However, as mentioned above, the main focus of IDT is anchored in the analysis of the process of “imitation” or diffusion in a society through examination of four concepts, namely: innovation, innovation communication, time, and the social system.

Although some IDT researchers have used behavioral sciences approaches to operationalize these concepts, they have not always included the individual’s “intentions” to accept or adopt an innovation. According to Kukafka, et al. (2003), a distinct group of researchers use behavioral intention theories to include and often focus on individual intentions. TAM is the behavioral intention theory most often adopted in this type of research.

The next section describes these behavioral intention theories, which examine the individual’s intention to accept a technology. These theories differ from the IDT above because while the IDT describes the society and its communication channels for innovation or technology adoption, the behavioral intention theories describe the individual and his or her intention to adopt a technology.

Behavioral Intention Theory

Research based on the individual’s intention to accept or reject information technology has been conducted extensively using intention-based theories and models. Many

scholars have carried out research to test behavioral intentions variables using various theories and models. Some of the models and theories, which appear many times in the literature, include: Social Cognitive Theory by Bandura (1986); and a Theory of Cognitive Dissonance by Festinger (1957). Both these theories are grounded in the behavioral sciences. Another most cited theory, which falls under behavior intention theories, is the Theory of Reasoned Action (TRA) first proposed by Fishbein & Ajzen in 1975 (Leonard, et al, 2004).

TRA posits that one’s intention to perform or not to perform an action (behavioral intention) is the immediate precursor to the actual behavior. The TRA model introduced two factors that affect behavioral intention: attitude toward the behavior and subjective norms (Leonard, et al, 2004). According to TRA (Leonard, et al, 2004), behaviors are under a person’s volitional control, therefore a person’s attitude towards a behavior (negative or positive) and social pressure (what is expected of them) determines the person’s willingness to perform a behavior (behavioral intention).

This theory was later modified by Ajzen (1991) to become the Theory of Planned Behavior (TpB). Ajzen (1991) came to the conclusion that behavior is not always under control nor totally voluntary as posited by the Theory of Reasoned Action. Ajzen argued that sometimes behavior is influenced by perceived behavioral control. Perceived behavioral control has been defined as “the perception of how easy or difficult it would be to perform the behavior” (Leonard, et al., 2004, p. 144). Therefore, Ajzen extended the Theory of Reasoned Action into the Theory of Planned Behavior (TpB). According to Ajzen (2005), the TpB posits that human action is guided by three kinds of considerations; behavioral beliefs, (attitude towards behavior) normative beliefs (perceived social pressure) and control beliefs (perceived behavioral control) and a combination of these considerations culminates in behavioral intention to perform a behavior. This school of thought draws its concepts from attitude and behavior research that focuses on the individual’s perceived or planned behavior, which can be used to predict use or adoption of an innovation or technology (Colvin and Goh 2005).

Technology Acceptance Model

Davis (1986) adapted the TRA model by “replacing TRA’s attitudinal determinants, derived separately for each behavior, with a set of two variables, perceived ease of use (PEOU) and perceived usefulness (PERUSE)” to come up with the Technology Acceptance Model (McFarland & Hamilton, 2006). The Technology Acceptance Model (TAM) posits that the individual’s intention to use is the “single best predictor of actual system usage” (Davis & Venkatesh, 1996, p. 20). TAM has been used extensively to predict and explain the individual’s adoption of information technologies.

TAM has been validated (Colvin & Goh, 2005, Davis & Venkatesh, 2000; Legris, et al, 2003) and extended to add context to the three main constructs of PEOU, PERUSE (shortened to PU) and intention to use or behavioral intention (BI) (Dasgupta Granger, & McGarry, 2002; Dishaw & Strong, 1999; Karahanna & Straub, 1998; McFarland & Hamilton, 2006). One of the suggested extensions of TAM introduced the computer anxiety construct, both as a mediator on perceived ease of use (Hackbarth, et al, 2003) and as an independent variable that has an influence on behavioral intention (Compeau & Higgins, 1995). Therefore, in addition to the TAM variables of perceived ease of use, perceived usefulness, and intention to adopt as first proposed by Davis (1986), scholars have introduced the computer anxiety variable to the model (Compeau & Higgins,1995; Hackbarth, et al, 2003; McFarland & Hamilton, 2006).

The purpose of TAM is to predict and explain user acceptance of a new technology. Acceptance is hypothesized to be determined by the attitudes or beliefs the user has towards a given technology. These attitudes or beliefs are measured by looking at the perceptions of the user in relation to perceived usefulness and perceived ease of use of the technology. According to TAM, the “external variables, such as the task, user characteristics, political influences, organizational factors, and the development process, are expected to influence technology acceptance indirectly by affecting beliefs, attitudes or intentions”( Szajna, 1996, p. 85). So, TAM looks at three constructs: the intention to adopt, perceived usefulness and perceived ease of use. Davis et al. (1989) posits that

PEOU is a causal antecedent to PU. This assertion has been tested and proved by various researchers who established that PEOU has a significant effect on PU (Davis et al., 1989; Gefen & Straub, 1997; Karahanna & Straub, 1999). Figure 1 below summarizes the TAM model.

Perceived usefulness

External Perceived ease Behavioral Actual usage variables of use intention

Figure 1: Technology Acceptance Model (Davis & Venkatesh, 1996 p. 20)

According to Venkatesh (2000), perceived ease of use has been defined as a person’s belief that using a particular technology does not require too much effort. He goes on to say perceived ease of use is an important factor influencing user acceptance and usage behavior of information technologies. Perceived usefulness is the extent to which a user believes that a certain technology will enhance the performance in his/her job and has thus been closely associated with motivation (Venkatesh, 2000).

Studies that have extended TAM have suggested that computer anxiety has an effect on behavioral intention to adopt computers. According to Compeau &Higgins (1995, p. 189), “computer anxiety is related to computer self-efficacy, which was found to exert a significant influence on individuals’ expectations of the outcomes of using computers, in terms of their emotional reactions to computers (affect and anxiety).” Computer anxiety has been defined as “the apprehension or fear that results when an individual is faced with the possibility of using an IS [information system]” (Hackbarth, et al, 2003, p. 224). Figure 2 below shows an extended TAM model to include computer anxiety.

Perceived usefulness

External Perceived ease Behavioral Actual usage

variables of use intention

Computer Anxiety

Figure 2: Extended TAM Model (Adapted anxiety construct from Compeau &Higgins, 1995)

The Technology Acceptance Model (TAM) has been replicated, tested and applied in most parts of the world. Researchers using this model have reported its robustness and suitability for explaining users’ intentions to adopt information technology (Dasgupta, Granger & McGarry, 2002; Gefen, & Straub, 1997; Venkatesh, 2000). TAM has been tested and proven successful in predicting 40% of system use (Colvin and Goh, 2004; Legris, Ingham, & Collerette, 2003). Evidence of TAM use in Africa is limited but nevertheless present (Meso, et al, 2005; Anandarajan et al, 2002). Therefore, it is very useful for predicting whether or not users intend to adopt technology.

Barriers to IT Diffusion

IT adoption barriers from the individual perspective

One of the problems identified in the literature on IT adoption is that some individuals might not want to accept or adopt and use technology. Fredland (2000), writing on the issue, says that attention has not been given to possible reactions of the intended recipients of the technology, nor their desires with regards to accepting new innovations.

Some scholars have established that the use of computers remains low, after the installation of software and hardware (Venkatesh & Morris, 2000). Karahanna & Straub (1998, p. 238), expressing the same concern say, Although there is little doubt that technological developments will occur at a fast rate, it is not immediately obvious that individual users of the new technology will be able to adopt and use new technological artifacts at the same pace.

Based on the above deliberations, it can be concluded that indeed there is a problem concerning individuals who might not want to adopt and use computers after they are installed in their work place and this issue warrants further research.

Another problem that has been identified about individuals and computer adoption entails discomfort in the use of computers. It has been established that computer anxiety has an effect on the behavioral intention to adopt computer technology (Hackbarth, Grover & Yi, 2003). Orr (1993), expressing the same idea says, “the feeling of anxiety toward computers and computer use is common, affecting 30 to 40% of the population” (¶. 1). Computer anxiety as a factor in computer adoption is well documented in the literature (Compeau & Higgins, 1995; McFarland & Hamilton, 2006; Venkatesh, 2000).

Yet another issue identified by scholars writing on IT adoption in Sub-Saharan Africa is ‘dysfunctional behavior’, a situation where computers are not used effectively, adoption rates remain slow, yet governments support implementation through the purchase of relevant computer hardware and software (Odedra et al, 1993). Hu, et al. (2003), writing on IT adoption failure in schools, say, “The role of information technology in modern education has increased … but resistance to technology remains considerably high” (p. 228). From the above, it can be established that indeed there is a problem regarding individuals who resist adopting computer technology.

In conclusion, the diffusion and adoption of technology cannot be regarded as complete without the input of the individual for whom the system was put in place. Therefore it is important to find out what individuals think about adopting and using technology.

The next section introduces the barriers to information diffusion or transfer of technology in general, not just computers, as described in the literature on Africa. The barriers can be viewed from the national or governmental angle because they do not deal with the individual per se, they deal with weaknesses in the social, political and economic structure of the government, that lead to barriers in the diffusion or transfer of technology.

IT adoption failure in Africa

Heeks (2002), writing on the failure of information systems in Africa, says failure can be divided into two categories, total failure and partial failure. Heeks goes on to say that total failure is when an initiative never gets implemented or it is implemented and then abandoned, and partial failure is when an initiative does not achieve the major goals or it results in undesirable outcomes. Africa is lagging behind in its efforts of achieving an information society. The literature in this field has established that sub Saharan Africa is both technologically and economically least developed and that this has led to the slow transfer and adoption of information technology (Onyango, 2000; Udo & Edoho, 2000).

The barriers to IT diffusion and adoption can be divided into two types, the first being barriers at a national level and the second being barriers at an individual and organizational level. Literature on Africa and Sub Saharan Africa has mainly covered the national level of IT diffusion, which is mostly referred to as IT transfer, and has targeted social, economic, and political barriers. This literature is discussed next.

Socio-cultural Barriers

The social barriers mainly deal with access issues (sometimes referred to as the digital divide), the appropriateness of IT to African culture, Africa’s lack of communication channels, and lack of IT education and training. Other issues mentioned are the content of the information technology and the use of the English language, which most Africans are not familiar with, thereby creating a context that accelerates inequalities and complexity to the challenge of social inclusion (Chowdhury, 1998). Rice (2003), writing on the issue

says, “the lack of the application and diffusion of ICTs in LDCs (least developed countries) is exacerbated by the fact that more than 80% of all web sites are in English, a language understood by only about 10% of the world’s population” (p. 72). Also the lack of training, education, and all kinds of literacy needed for the use of IT seems to be a problem for the IT transfer process. The types of literacy involve basic reading and writing, basic computer literacy, and information literacy, which encompasses the ability to access and use information from all sources effectively. It is mentioned in the literature that three quarters of Africans in Sub Saharan Africa are illiterate, and without phone or electricity (Obijiofor, 1998; Chowdhury, 1998; Jensen, 2002). Also cited in the literature is the underutilization of existing computer resources, which is linked to illiteracy and poverty (Jensen, 2002).

Emanating from the above and associated with the social barriers to IT adoption is the issue of culture. Some scholars have expressed the fear that the culture of Africa is conservative and will therefore hinder IT adoption. These ideas have been dismissed on the grounds that Asian culture is also different from that of the West and yet IT has flourished there (Udo & Edoho, 2000). Other scholars have carried out research in some African villages and concluded that a conservative culture is not an impediment to IT adoption (Hudson, 2000). Therefore the conservative culture of Africa as an impediment to IT adoption will be ruled out in this discussion. Economic barriers

Some of the barriers echoed in all the literature consulted have to do with Africa’s underdevelopment, civil wars, corruption in the government and poverty. Statistics abound on the low density of telephone lines, the lack of clean water and electricity, to mention but a few (Akpan, 2000, Jensen, 2002; Oladele, 2001; Onyango, 2000). According to Odedra et al. (1993), Africa seems to be the “lost continent” of the information technologies (IT). Odedra et al go on to say that Africa is “the second largest continent and the least computerized” (p. 26).

The high costs of the IT equipment and the equally high costs of maintenance for a people who live in poverty is a real challenge. Consequently, Africa does not have the infrastructure or the skilled manpower to accelerate IT adoption. This situation has often been described as the digital divide. The digital divide is measured by “access to technologies, access to relevant and usable content, skills in using ICTs (knowledge and experience in use)” (Kebede, 2004, p. 274).

Political barriers

The political barriers deal with policy and political power, policy implementation failure, and the inadequacies of national information policies in information technology strategic planning and implementation. Chowdhury (1998), writing on information policy in Africa, has remarked upon the lack of awareness and enthusiasm of the policy makers, the absence of political and regulatory frameworks, and the lack of local networks for management. Other scholars approach the policy barrier from the economic point of view by suggesting economic empowerment through investment and trade, which should be made possible by the formulation of the right information policies (Udo & Edoho (2000). Still other scholars emphasize the liberalization of telecommunications policy as a way of stimulating growth in IT (Hudson, 2000; Jensen, 2002).

Most of Africa is characterized by a dearth of national information policies and this has contributed immensely to the failure of information technology adoption (Berman & Tettey, 2001; Korac-Kakabadse, et al, 2000; United Nations Economic Commission for Africa, 2001; Wilson & Wong, 2003). According to United Nations Economic Commission for Africa (2001), the process for setting up ICT policies and strategies is a work in process but it is rather slow because it requires concerted effort from all parties, mainly the national government for effective leadership and direction. United Nations Economic Commission for Africa (2001) goes on to say that the few national information policies that have been formulated have been marked by their lack of comprehensiveness in terms of content and coverage. Added to the lack of policy is the well-documented

crisis in Africa of poverty, which is intensified by social, political and economic problems.

Onyango (2000), summing up the magnitude of the political problems says that policy has tended to fail in Africa because there has been no thread of continuity or review process in all policy matters, instead, governments and other organs of state “lurched at fever-pitch speed from one idea to another in the implementation of these imported and unadjusted development blueprints” (p. 198). In other words, policy development in Africa has always posed problems because when blueprints and ideas from other nations are adopted without making them suitable for the culture and the situation, policy fails.

The literature above describes the social, economic, and political barriers to IT adoption in Africa. Botswana, which is in Sub Saharan Africa, has a different social, political, and economic environment to the one discussed above. Botswana’s economic, social, and political problems do not reach the magnitude of similar problems in other Sub Saharan African nations, and this fact is one of the reasons why the approach to the present research is different from the one suggested in the literature on Africa, as will be discussed later in this study.

Botswana

Economic, Political, and Social Atmosphere

Botswana’s political, social, and economic situation is different from most African countries and has thus made it one of the most democratic nations in Africa. Botswana achieved political and economic stability at the time of independence in 1966 and has successfully managed to sustain this stability for almost 35 years (Parsons, 1999). This situation makes Botswana different from all the other countries in Sub Saharan Africa that are dealing with poverty, civil strife, economic, and political difficulties.

Economically, Botswana has also been regarded as successful compared to some poor countries of Sub Saharan Africa. The economic situation changed after independence due to the discovery of minerals and the diversification of the economy. The main sources of income in Botswana are: mining, agriculture, tourism, and manufacturing. Diamonds, which were discovered in 1967, exceed all other minerals by export value. In terms of production, Botswana is one of the largest diamond producers in the world (Silitshena &McLeod, 1998; May, 1998). According to the Government of Botswana Website the combined earnings from diamonds has accounted for 77 percent of the total export earnings, or 45 percent of the GDP. The annual output of Botswana diamonds is 15 million carats. Botswana diamonds are in two categories, gemstones (for jewelry) and industrial diamonds (Silitshena & McLeod, 1998; May 1998). According to the Government of Botswana website, “Botswana has also been awarded the highest sovereign ratings in Africa by both Standard & Poor’s (S&P) and Moody’s Investors Service (Moody’s). S&P assigned Botswana a rating of “A” for long-term debt and Prime-1 (P1) for short-term debt” (Botswana Economy, Facts and Figures, ¶ 9). Parsons (1999) attests to this when he says, “Botswana has maintained one of the world’s highest economic growth rates since its independence, transforming it from one of the world’s poorest countries to a middle-income country today” (¶ 9). Therefore, according to United Nations in Botswana (2005), Botswana is a middle income country with a “GDP per capita of $4,104 (IMF2), a GDP of $5.831 billion (IMF1) and GDP Growth of 5.2% (2004, IMF2)” (¶. 9).

Politically, Botswana has often been praised for its good governance by several scholars (Beaulier, 2001; Molomo & Somolekae, 1999). Botswana is a multi-party democracy, with more than 13 political parties. Elections are held once every five years. Parsons (1999) has indicated that, “Botswana is Africa’s longest continuous multi-party democracy. Widely lauded as one of the most stable democracies on the African continent and in the third world, Botswana has been seen for many years as an exceptional case in an otherwise conflict-ridden region” (¶. 1).

Socially, Botswana is regarded as a peace-loving country and in Africa it is regarded as the least corrupt country (Parsons, 1999; Lekorwe, et al; 2001). At independence on 30 September 1966, Botswana adopted a philosophy called Kagisano, a Setswana word {Setswana is the national language of Botswana} that basically means social harmony, as its policy or pillar. Kagisano is a Setswana word that originates from the word Kagiso, which means peace (Department of Curriculum Development and Evaluation, (1988). However, one can argue that even before its codification, Batswana (people of Botswana) have always been peace loving. Some examples of proverbs in the Setswana language attest to this assertion. For example a common proverb in Setswana is “Ntwa kgolo ke ya molomo” which means: there is no need to solve a dispute through bloodshed; the best solution is through diplomacy or negotiation. Another example: “Matlo go swa mabapi” means that neighbors should always offer a helping hand to each other because if a fire destroys one household, if the neighbors do not cooperate, their household will also catch fire. These are just two examples among an array of proverbs that explain Batswana’s philosophy of community spirit. Most probably this philosophy explains why Botswana has not suffered civil wars and internal conflict that have characterized most of Africa.

Botswana’s information technology scene

Botswana has not formulated a national information policy yet, but it does have a Vision 2016 document. Botswana has embarked on an information technology diffusion or transfer strategy, which aims at making information a part and parcel of everyday activities that would improve the economic, social and political situation of the country, hence the Vision 2016. Also, the existence of two other documentary sources, the National Development Plan 9 (2003) and the Revised National Policy on Education (1994) underscore the importance of information technology diffusion in the nation as well as in schools. In 1997, Botswana’s government set up a Presidential Task Group, to work on a Vision for Botswana. The importance of lifelong learning is mentioned in the document. Also mentioned in the vision is the desire to produce citizens who are informed and are able to use information for solving daily problems. The step towards the development of an “educated, informed nation” (p. 71) is documented in the Botswana

Vision 2016 document. Among the goals included in the document, Botswana intends to “formulate a national information vision, policy and information technology strategies, as well as co-ordinate the currently fragmented information infrastructure in the country” (p. 35). Some of the Botswana Long-term Vision 2016 objectives are,

1. Acquire the best available technology in order to make information accessible to all; 2. Develop communication capability to enable participation in information age; 3. All Batswana to have access to information; 4. All Batswana to use and apply the potential of computer equipment in their daily life; 5. Botswana to recognize the importance of information and the development of efficient information systems and networks (p. 71).

Prior to the Vision 2016 document, the Botswana Government introduced reforms documented in the 1994 Revised National Education Policy. One of the objectives mentioned in the Revised National Education Policy is to equip all secondary schools with a computer laboratory. This objective has been achieved and schools in the country currently have a computer laboratory.

Education in Botswana

In the education sector, Botswana has made strides towards equitable education for all. The Ministry of Education has begun implementing the aims of Vision 2016. A brief outline of Botswana’s education system and structure indicates that there are (excluding pre-school education) seven years of primary education (equivalent to elementary school), three years of junior secondary education (equivalent to middle school), and two years of senior secondary education (equivalent to high school). A significant achievement has been made in the provision of educational opportunity at the junior secondary education level. About 95% of primary school leavers now go on to Form 1 (junior school). All education from primary (elementary) up to tertiary level (university

and vocational schools) is free for all Batswana (Revised National Education Policy, 1994). There is universal access to schooling from primary up to junior school (middle school) and government is working on providing universal access to all levels of school, by providing more senior secondary education (Revised National Education Policy, 1994). Altogether there are 233 secondary schools in Botswana (both junior and senior schools). 27 of these schools are senior secondary and 206 of these are junior secondary schools. Botswana’s government, through the Ministry of Education, has introduced three computer-based subjects, namely Computer Awareness and Design and Technology (core subjects) in both junior and senior schools and Computer Science as an examinable elective subject in senior schools (Revised National Education Policy, 1994).

Although Botswana has made efforts in education to bring about an information society, there are still some obstacles before that can be achieved. The next section discusses some of those obstacles.

IT Leadership and Principals

The adoption and use of computers in schools is a big change for the principals, teachers and the students, because not only does everybody need to adjust to computers and accept their use, but the entire learning culture is impacted. This calls for changes in the way principals lead during the implementation of computers in schools. Several authors have suggested that principals should be transformational leaders in the information era (Flanagan & Jacobsen, 2003; Telem, 1996; Todd, 1999; Yee, 1998; Yuen et al, 2003).

Flanagan & Jacobsen (2003), writing in support of the above statement, argue that Many principals have not been prepared for their new role as technology leaders, and have therefore struggled to develop both the human and technical resources necessary to achieve ICT outcomes in their schools. Very few principals have themselves used computers in meaningful ways with children, and therefore lack the requisite pedagogical vision and experience to guide teachers. (p. 127)

In addition to the above assertion, Yuen et al. (2003) in a study carried out exploring the leadership issues amongst early adopters of ICT implementation in Hong Kong schools found that indeed implementation involves stakeholders and issues of leadership.

Todd (1999) proposes that school principals should be transformational leaders in the information era. He writes that “for school principals, there are two essential keys: transformational leadership at the school level and transformational learning at the classroom level” (p. 4). Todd, writing on strategic leadership as encompassed in the transformational leader says, Strategic leadership focuses on the longer term, where the scale and scope of action are school wide rather than program-focused. It demands a clear perception of role and a sense of vision. It is indicated in a long-term and futures-oriented technology plan for the school, a program of staff development that provides access to technology, and developing the staff's knowledge and skills about the pedagogy of integrating information technology into the classroom. (p. 5)

Flanagan & Jacobsen (2003) however, warn that, While there are positive examples of technology being used to support student learning and to foster positive changes in schools, predictions that computers would revolutionize public education have not materialized. Merely installing computers and networks in schools is insufficient for educational reform. (p. 125)

Flanagan & Jacobsen go on to say that there are four themes that can be identified as barriers to technology integration. These are “pedagogical issues, concerns about equity, inadequate professional development, and lack of informed leadership” (p. 125).

Based on the above discussion, it can be concluded that leadership in IT implementation is a complicated issue, and principals as leaders are grappling with IT adoption issues concerning themselves as well as the teachers and the students. This makes the principal’s role in leading others even more crucial in the digital era. Yee (1998) emphasizing the above point says that the role of the principal has changed from that of

35 an instructional leader to a transformational leader and that a transformational leader “focuses on building a shared vision, improving communication and making decisions collaboratively” (p. 57).

Problem in relation to the literature

First, the issue of information technology diffusion and adoption in Botswana has not been studied extensively and the literature available is confined to descriptions of the situation in relation to the lack of information policy, infrastructure, and the relevant skills for the use of IT (Bose, 2004; Jain & Mutula, 2001), as well as government documents on what is planned to take place, in the form of mission statements (Long term Vision 2016; Revised National Education Policy, 1994; National Development Plan 9, 2003 ). This study will explore individual adoption of computers by the school principals in Botswana, as opposed to describing the situation, in order to predict whether users intend to adopt the computers.

Secondly, Botswana’s political, economic, and social situation differs from that of most Sub Saharan Africa as demonstrated above (Parsons, 1999) and this calls for a different approach to research based on the individual as a unit of analysis because the formulation of Vision 2016 places Botswana in a different level from those African states which do not have IT diffusion plans in place yet and are economically, politically, and socially challenged. The approach from the individual perspective of computer adoption should help explain the intentions of the principals towards accepting and using computers.

Lastly, with the installation of computer laboratories in all secondary schools (Revised National Education Policy, 1994), the introduction of Vision 2016 and computer based subjects in schools, Botswana has made the first step in the transfer or diffusion of technology at a national level, therefore the environment is conducive for research on the individual’s intention to adopt and use computer technology. In order to complete the technology diffusion or transfer process, the citizens must accept or adopt technology as well as use it. The present study examines the technology acceptance or adoption of the

school principals, who are assumed to be transformational leaders in their communities. One of the most used models for studying individual intentions to adopt technology is the Technology Acceptance Model. Davis( 1986) introduced TAM specifically to explain computer usage behavior. This justifies why TAM is suitable for studies in computer acceptance.

CHAPTER 3: RESEARCH DESIGN AND METHODOLOGY

Research Design

The thrust of this study was mainly quantitative in nature. It employed the survey as the primary data collection tool but since the survey method is limited in its representation of the whole picture, this study was complemented by interviews. The interview is a qualitative data collection tool. This research used the semi structured interview because it does not restrict participants and therefore allows a free flow of conversation within the structure (Creswell, 2002). The mixed method research design combining quantitative and qualitative data collection, helped explain what the computer adoption intentions of the principals were and how they intended to implement computer technology.

Survey

The quantitative method employed the Technology Acceptance Model (TAM) survey as proposed by Davis (1986) to determine the perceptions of the school principals about accepting and using the computer technology. The questionnaire designed for this study, consistent with TAM, was used to explain and predict the behaviors of school principals towards computer technology adoption (Davis, 1989; Legris, Ingham & Collerette, 2003; McFarland & Hamilton, 2006). Questions were modified by changing the name of the technology system tested from an existing pool of TAM research questions to solicit the extent of the principal’s computer technology acceptance in a survey. (In this case the computer replaced any other system tested previously using TAM).

In accordance with the literature in the field, the present study extended the original Technology Acceptance Model (TAM) by adding computer anxiety as an independent variable. Therefore, the study adopted an extension to TAM based on the research on computer anxiety and technology acceptance by several scholars (Compeau & Higgins, 1995; McFarland & Hamilton, 2006; Venkatesh, 2000). Scholars have added computer anxiety as an independent variable or predictor of intention to adopt a technology

(Compeau & Higgins, 1995), and as a mediator on perceived ease of use (Hackbarth, et al, 2003) in the TAM model. The study used computer anxiety as an independent variable or predictor. Igbaria & Nachman (1990) posit that computer anxiety is one of the strongest predictors of a negative attitude toward computers. The study presented and applied the Technology Acceptance Model (TAM) to determine system-specific perceived ease of use, perceived usefulness, anxiety towards computers and the intention to adopt and use computers by school principals.

The survey and the interview were conducted in both English and the national language (Setswana). Both English and Setswana are official languages in Botswana and all headmasters use both in their schools (Republic of Botswana National Development Plan 9, 2003).

Interview

The interview participants were chosen using the purposive sample technique. The purpose of having a purposive sample was to make sure that all ages, types of schools, both genders and all locations were represented. Also, different populations within one sample should also be represented (for example, adopters and non-adopters). A purposive sample for the interview was appropriate for this research. According to Babbie (2001), “sometimes it is appropriate to select a sample on the basis of knowledge of a population and the purpose of the study” (p.179). Since many participants fell under the above mentioned categories, the 10 participants were randomly selected within the purposively selected group. A random selection ensures that all participants have an equal chance of participation (Creswell, 2002).

The criteria for choosing participants included age, location (city or village), type of school (junior or senior secondary school), gender, and the level of comfort or discomfort around the computer as reported in the survey. One participant was aged between 30 and 40, one was 51 and above, and eight participants were between 40 and 50 years of age. The survey had 12.6% for the 51 and above age range, 13.5% in the 30-40 age range and

71.2% between 41 and 50 age range. So the majority of the participants fell in the 41 to 50 age range. Two participants were from senior secondary schools while the rest of the eight participants came from community junior secondary schools. The survey results had 85.6% of the participants from junior school principals and 14.4 % from senior secondary schools. The majority of the participants were from junior schools. In terms of adopting the computer, two participants did not intend to adopt, one was neutral (neither adopting nor rejecting computers), and the rest of the seven participants wanted to adopt computers. According to the survey results, 3.6% were neutral respondents, 90.1% were adopters and 6.3% were non-adopters. The majority of the respondents wanted to adopt computers. In the survey the gender result was 71.2% male and 28.8% female participants. In the interview, there were six male and four female participants. Seven participants resided in big villages, two in the city and one in a small village.

The data collection tool was a semi-structured telephone interview. This tool gave the researcher a structure within which to channel questions (Babbie, 2001). It also gave the participants an opportunity to expand on their responses, within a specified structure, and in that respect, useful data were collected. According to Bill (1977), semi structured interview data have high predictive effectiveness. The qualitative data from semi structured interviews were transcribed from the tape recordings and entered into the NVivo software program for coding, memoing and concept mapping to explore the patterns that emerged from the data (Babbie, 2001). Using NVivo, coded materials were reflected on, revised and re-coded until patterns emerged in the data.

The wording and content of the interview questions depended on the results of the survey. As explained above, all principals had the same required qualifications even though they were posted in either junior or senior schools, therefore they moved between schools, so they can be considered as a single, homogenous population rather than two heterogeneous populations. Participants were given fictitious names to protect the identity in the discussion of the interviews.

Variables of the Study

The variables for the proposed study are based on the extended Technology Acceptance Model (TAM), and are used to test the technology acceptance of the subjects. If the school principals do not report computer anxiety, and perceive computers as useful and easy to use, then they are more likely to accept computer technology. The variables and their operational definitions are as follows.

Computer anxiety Anxiety is associated with intense dread, apprehension, or nagging worry about the use of computers (Orr, 1993). If school principals dread computers they are not likely to want to adopt them.

Perceived ease of use of the computer (PEOU) Perceived ease of use has been defined as a person’s belief that using a particular technology, in this case computers, does not require too much effort (Venkatesh, 2000). If the school principals do not believe that computers are easy to use, they are not likely to adopt them.

Perceived usefulness of the computer (PU) Perceived usefulness is the extent to which a user believes that computer technology will enhance his/her job performance (Venkatesh, 2000). If the school principals do not believe that computers are useful in their jobs, they are not likely to adopt them.

Computer technology acceptance or adoption According to the TAM, system usage behavior is determined by the intention to use a particular system, which in turn, is determined by the perceived usefulness and perceived ease of use of the system (Luarn and Lin, 2005). A person has accepted computer technology if they are willing to use it. Table 1 below shows the variables and the sources of the TAM statements.

Table 1: Variable type and source 1

Variable Type Source of statements

Perceived Ease of Use Independent Variable Davis (1989) Technology (PEOU) Acceptance Model (TAM) Perceived Usefulness (PU) Independent Variable Davis(1989) Technology Acceptance Model Computer Anxiety Independent Variable Venkatesh (2000); Compeau, & Higgins (1995); McFarland & Hamilton (2006) Behavioral Intention to Dependent Variable Davis(1989) Technology adopt (BI) Acceptance Model

The proposed research model, illustrated in Figure 3 below, extends TAM by adding computer anxiety as an independent variable. The model posits that the intention to adopt and use computers is jointly determined by perceived ease of use, perceived usefulness, and computer anxiety. The present study, consistent with previous studies, proposes the use of the extended TAM as shown in figure 3 below.

Perceived usefulness

External Perceived ease Behavioral Actual usage

variables of use intention

Computer anxiety

Figure 3:Extended TAM Model (adapted anxiety constructs from Compeau & Higgins, 1995)

Hypotheses

Consistent with the extended TAM constructs, the distinct and overarching hypothesis tested was: positive feelings towards the usefulness of computers, the ease of using computers and a lack of computer anxiety among secondary school principals are indicative of computer technology adoption and willingness to use computer technology. This hypothesis was broken down into four sub hypotheses. In accordance with TAM literature, the hypotheses to be tested were as follows: a. Hypothesis 1: There is a negative relationship between computer anxiety and the principal’s intention to adopt and use computers. b. Hypothesis 2: The principals’ perceived usefulness (PU) of the computer has a significant influence on their intention to adopt and use the computer in their work. c. Hypothesis 3: The principals’ perceived ease of use (PEOU) of the computer has a significant influence on their intention to adopt and use the computer. d. Hypothesis 4: There is a positive relationship between perceived ease of use (PEOU) and perceived usefulness (PU) of the computer and the intention to adopt and use computers. Methods

The study used a mixed method design which employs quantitative and qualitative methods to examine the principal’s intention to adopt and use computer technology.

Quantitative method

The quantitative method employed a survey based on the Technology Acceptance Model (TAM) as proposed by Davis (1989) to determine the perceptions of the school principals about accepting and using the computer technology. The survey questionnaire (see Appendix H) was used to collect demographic and opinion related data. Demographic variables, measured in the present study, include gender, qualifications, age, and type of school, for use in descriptive statistics, while the rest of the questionnaire dealt with TAM questions in order to predict whether principals intend to adopt computers. The survey

result was statistically analyzed prior to the formulation of the interview question, using SPSS to run multiple regression on the three independent variables (computer anxiety, perceived ease of use of the computer and perceived usefulness of the computer), and the dependent variable (computer adoption intention).

The survey, consistent with TAM surveys, was an instrument with multiple scales, which measured particular attitude dimensions. The principals placed a tick to show their choice. The survey included 23 statements such as: 1. Computers make me feel uneasy 2. I feel threatened when other people talk about computers 3. Interacting with the computer requires a lot of mental effort

Consistent with the literature on TAM, participants rated each statement on a seven-point likert scale with Strongly Agree and Strongly Disagree at the beginning and end of the scale. The statements were adapted from previous questionnaires from several studies (as listed in Table 1) to measure each of the constructs included in the model.

Qualitative method

The semi structured interview method is chosen in this research for its usefulness in contextual and in-depth inquiry, so as to explain why events occur as they do in a chosen study group, in this case the school principals. Interviews are used to answer “why” and “how” questions to complement the survey method which deals with “what” questions. The overall objective of the interviews was to establish how principals perceived computers in relation to adopting and using them in the whole school.

Content analysis method was used on the interviews. In this study, coding involved identifying four characteristics of the text; the frequency of events, the direction of the events, the strength of the statements and the size of text message for each event (Neuman, 2003). The interviews were analyzed using the NVivo, a content analysis software, to find the patterns in the data (Babbie, 2001). The NVivo software, allows the

researcher to search, link, model and perform code based theorizing. The interviews were analyzed in order to find out if any patterns and concepts were discernible in the data, and how those patterns impacted the adoption of computers by the school principals.

Population of the Study

The research population is the principals of the 233 secondary schools in Botswana. This number includes 206 community junior secondary schools (the equivalent of U.S. middle schools) and 27 senior secondary schools (the equivalent of U.S. high schools).

Research Sample

Probability sampling using simple random sampling strategy was used in this study because the population studied is dispersed around the country, therefore to make the study feasible and to generalize the results to the entire population, the simple random sample was proposed. Also the simple random sample has been found to be one of the best methods for inferential statistics (Babbie, 2001). The random sample targeted all principals in Botswana. The sample was drawn from a list of school names recorded in SPSS in order to use SPSS functions to randomize the sample. A sample of 145 participants was calculated using the following formula: Z2 * (p) * (1-p)/c2 where: Z = Z value (e.g. 1.96 for 95% confidence level), p = percentage picking a choice, expressed as decimal (.5 used for sample size needed), c = confidence interval, expressed as decimal (e.g., .04 = ±4) (Moore, 2004). According to the sample size calculator, the sample size should be 145 at a confidence interval of .05, a population of 233 and a confidence level of 95%.

The response rate of the questionnaire was very good at 114 participants, which is 78.6% of the expected sample of 145. Out of the 114 questionnaires received, 111 were usable; the other three were incomplete and were therefore discarded.

For interviews, a purposive sample of ten (10) subjects was selected from the three different groups identified in the survey results; namely the adopters, non-adopters and the neutral. The sample included representatives of the participants by gender, type of school, age, as well as the location of the school.

Data Collection and Dissemination Survey

The whole process of sending, and receiving surveys lasted for about six weeks. The survey was collected using three methods, surface mail, fax, and actual visits to principals during meetings and altogether 114 surveys were collected.

The questionnaire was mailed to all participants with a cover letter that briefly described the study and a stamped, self-addressed return envelope for the convenience of the participants (Dillman, 1978). Survey follow-ups were made by sending another questionnaire to those who had not responded. Other follow-up techniques were using the fax and phone in order to reach all the subjects. A thank you note was sent to all participants. This note served as both a thank you to those who responded and a reminder to those who still needed to fill in the questionnaire (Dillman, 1978). The mailed survey, according to Creswell (2002), is economical, as well as one of the convenient ways of reaching a geographically dispersed population. The fax and the phone are also economical, if the population surveyed is not too large, as is the case with the present study.

Another technique used to collect the questionnaires was attending a meeting held in a village called Palapye, where 80 principals representing both junior and senior secondary schools, from the central and northern part of Botswana were meeting. Principals have meetings and workshops on a regular basis throughout the academic year. Questionnaires were filled at that meeting of the principals from the north and central part of the country. The faxed and mailed questionnaires were a result of telephone follow-ups and these

mainly came from the north-western part of the country. The rest of the questionnaires were filled through individual visits to schools in the south-eastern part of the country.

Interviews

Semi structured in-depth interviews, which form the next level of the research, were done by telephone, using the Technology Acceptance Model constructs (Davis,1989) and computer anxiety variable (McFarland & Hamilton, 2006), to determine the factors that influence the acceptance or rejection of computer technology among school principals. The ten participants were chosen through purposive sampling in order to have a representative sample (Babbie, 2001). The 10 school principals included six male and four female participants from Botswana senior and junior secondary schools.

Each interview took 20-30 minutes. A telephone adapter that plugs into the phone and the tape recorder was used to capture the data (Creswell, 2002). Content analysis was performed on the interviews and data were transcribed and entered into the computer for analysis using NVivo software.

Babbie (2001, p. 263) mentions three advantages of the phone interview: 1. The researcher’s manner of dress does not affect the responses from participants 2. Participants can communicate a lot about themselves over the phone 3. Participants do not have to worry about personal safety because they do not need to meet strangers. The potential disadvantages, according to Creswell (2002), are that the researcher does not have direct contact with participants and this limits communication. Also telephoning might turn out to be costly. However, the advantages outweigh the disadvantages of telephone interviews.

Techniques for maximizing the return rate of survey

According to Dillman (1978), the three things to be mindful of when administering a survey are minimizing the costs for responding for the participants, maximizing the rewards for participation, and establishing trust that the rewards promised to the participants are delivered. Dillman suggests that minimizing the costs can be done through saving the participants’ time by making the survey simple. This research, in accordance with Dillman’s suggestions, uses a one and a half page long survey which involves mainly ticking 23 survey statements and circling the choices for demographic questions. To maximize the rewards and establish trust, consistent with Dillman’s suggestions, a personalized cover letter was included with each mailing. The letter contained statements that reassure the principals about the usefulness of filling in the questionnaire as well as the confidentiality of the data collected. Dillman (1978) also suggests the use of statements that support the participants’ values as well as liberal communication to thank the participants. These suggestions have been incorporated in the cover letter accompanying the questionnaire.

The return rate was good at 78.6% because in addition to sending the survey by mail, making follow-ups by fax and sending thank you cards to encourage those who had not responded yet, the principals were contacted in person at one meeting held in Palapye, and individual schools were also visited.

Techniques for collecting useful data from interviews

A method of interviewing using micro-moments and backward chaining as suggested by Dresang (1990) was suggested for the interview. This method is an adaptation of Brenda Dervin’s sense making approach to interviewing (Dresang). Since the aim of the present research was to probe the principals’ thoughts in order to find out how they felt about computer adoption, the adaptation of the two major steps, micro-moments and backward chaining, were suggested. These steps involve recalling or remembering major events in a participant’s situation, recording these moments as entries in the computer, and then choosing one of the events (micro-moments) at a time to discuss until all micro-moments

48 are exhausted (Dresang, 1990). However, due to cultural differences, after six pilot interviews, it became apparent that micro-moments and backward chaining was not effective among school principals in Botswana, who have a different culture from the Western one, where the method was first used. Therefore micro-moments and backward chaining techniques were not used for the 10 participants chosen for this study. Although 16 interviews were done, only 10 were used in this study. Since micro-moments and backward chaining did not work, the researcher used a semi structured interview, based on the TAM constructs to elicit responses that could be compared and to maximize the chances of collecting useful data.

Techniques to Ensure Reliable and Valid Data

According to Suskie (1992), even a seemingly straightforward survey can generate erroneous responses. Therefore, survey questions need to be piloted before the final administration; this will increase the likelihood that it is usable when it returns from the clients. Research has been conducted using the present instrument in a previous study by the present author (Totolo, 2005). The instrument used was adapted from an existing pool of TAM constructs, which had been used many times by several researchers and had proven to be valid (Venkatesh, 2000; Karahanna, Straub, & Chervany; 1999; Karahanna & Straub, 1998; Venkatesh & Davies, 2000; Dishaw, & Strong, 1999).

Lastly, the main weakness of the survey instrument is that it can be too artificial, therefore, it does not give a close-up view of society (Babbie,2001).To overcome the weaknesses of the present survey, interviews were used to complement the questionnaire and get in-depth information about a small subset of the subjects. In qualitative research, data are coded and checked against the categories derived from the data. In this case, the interviews were coded and checked against categories derived from the survey responses. This exercise helped with the validation of the data and the instrument.

Statistical regression, experimental mortality, testing and selection-maturation interaction have been controlled by using the survey method as opposed to the experimental design which uses testing (pre and post testing), and is likely to have maturation through interaction of subjects and the loss of subjects through mortality. Other possible threats are maturation because the subjects might have more exposure to computers over time. This also affects history but can be overcome by making the data collection period shorter, which is why as stated above, data were collected within a six week period.

Instrument Validity and Reliability Validity

The researcher selected a validated survey instrument, based on TAM, to increase the reliability of the instrument. Also the Cronbach’s Alpha test was performed to test the average internal consistency of each variable of the scale used for TAM (Moore, 2004). There was no need to test for the validity of the TAM instrument because the numerous studies, as shown in the literature review, employed the TAM instrument and established its validity. This encompasses construct, convergent and divergent validity as established on the TAM (Venkatesh and Davis, 2000), therefore TAM is a valid instrument. Validity refers to the approximate and presumed causal relationships that can be generalized to and across different types of groups of people, settings, and times (Cook and Campbell, 1979).

Reliability

Reliability is the main strength of quantitative designs because of statistical power, which allows for several tests on the data. The reliability of the TAM scales needed to be established for this study because the scales were modified by adding valid computer anxiety scales. Before performing regression tests, the reliability test was done on the TAM variables using the Cronbach alpha reliability test, consistent with prior psychometric properties evaluated on TAM (Venkatesh and Davis, 2000). Reliability is

50 the extent to which “a particular technique, applied repeatedly to the same object, would yield the same result each time” (Babbie, 2001, p.140).

The Cronbach alpha measures of internal consistency, before the removal of outliers, showed very high reliability at .909 for the six statements of perceived usefulness (PU) but very low reliability of .215 for computer anxiety, and .354 for the perceived ease of use (PEOU) variables. Overall the extended TAM model was placed at .642 for the extended TAM. The results of the computer anxiety and perceived ease of use depart from the previous TAM Cronbach’s alpha reliability tests results, which have consistently put the PEOU results above the acceptable 0.80 (Davis, Bagozzi, & Warshaw, 1989; Hendrickson et al, 1993) and anxiety at 0.92 (Hackbarth, Grover, & Yi, 2003). Therefore, two procedures were done to the data before the regression model could be tested. These procedures were the removal of outliers, and further tests of reliability of TAM and computer anxiety scales after the removal of outliers.

After removing 25 influential outliers, consistent with the TAM research (Venkatesh, 2000; Venkatesh & Davies, 2000), eight statements (four from each variable) were picked from the PU and PEOU variables, based on the means, lower standard deviation, acceptable correlation between the dependent and independent variables to perform reliability tests. The use of four statements for each variable is a common practice among scholars testing TAM (Davis, 1989; Hackbarth, Grover, & Yi, 2003). The statements chosen for the regression model appear on Table 17 (see Appendix C). The Cronbach's coefficient alpha test was applied to the variables used in the study to check the average internal consistency of each variable on the scale. The Cronbach alpha coefficients on the TAM constructs were quite high at .972 for the PU and .806 for the PEOU. These results compare very well to previous alpha results on TAM research (Davis, 1989, Davis & Venkatesh, 1996). However, the results of the Cronbach alpha for computer anxiety were very low and unacceptable at -.093.

According to Peterson (1994, p.381), Cronbach coefficient alpha coefficients have ranged from .5 to .6 for preliminary research to as high as .95 for applied research. Therefore, the

computer anxiety statements, which consistently exhibited an unacceptable and very low Cronbach alpha coefficient, were not used in the regression tests that tested the intention to adopt and use computers. In addition, earlier in the descriptive portion of this research, computer anxiety had more outliers (22) than PU and PEOU together (23), therefore it is not surprising that the computer anxiety scale was found to lack reliability. Based on the reliability tests before and after the removal of outliers, the computer anxiety variable did not seem to be reliable as an independent variable in the TAM model. Writing on the importance of reliability of scales, Keith (2006) says "reliability of a test, scale, survey, or any other measure places an upper limit on the correlation that the measurement can have with other measurements." Keith (2006) goes on to say that unreliable scales cause researchers to underestimate the effects of one variable on another (p.291). Therefore TAM was not extended using the computer anxiety variable in this study; instead the anxiety variable was explored in the qualitative part of this research. The quantitative portion of this study tested the original TAM model only, which had PU and PEOU as independent variables, and behavioral intention to adopt (BI) as the dependent variable, as first proposed by Davis (1986).

CHAPTER 4: SURVEY ANALYSIS AND RESULTS

The purpose of the quantitative portion of this study was to use the Technology Acceptance Model (TAM) to predict and describe the secondary school principals’ behavioral intention to adopt and use computers. This study examined computer technology adoption readiness. TAM was used to determine the perceptions of the school principals about accepting and using computer technology, using three constructs; perceived ease of use (PEOU), perceived usefulness (PU) and behavioral intention (BI) to adopt (Davis,1986). Several researchers have extended TAM to include the computer anxiety construct (Compeau & Higgins, 1995). This study used the extended TAM that includes computer anxiety as a variable of study. Altogether there were 114 randomly selected participants who responded to the survey questionnaire and 10 purposely selected interviewees.

Data were analyzed in two stages. The first stage was quantitative data analysis, using descriptive and inferential statistics. Descriptive statistics were used to describe the means, skewness, and the type of distribution using the Sharpiro-Wilk test of normality. Inferential statistics was used to do multicollenearity tests, and multiple regression tests for hypotheses testing. Data from the questionnaires were coded and analyzed using SPSS statistical software to develop descriptive and inferential statistics. Demographic information was used to provide percentages, and means to describe characteristics of the population, while inferential statistics tested the TAM hypotheses, as well as provided answers for the research questions. The testing of TAM variables employed multiple regression using SPSS.

The next stage was qualitative data analysis using content analysis techniques to characterize and classify the participants. The data were analyzed using four techniques, namely; the frequency of events, the direction of the events, the strength of the statements

and the size of text message for each event. Data were entered in the NVivo software for theorizing, modeling and concept building.

Description of the Research Population

Demographic Information

The purpose of the demographic data in this study is to give the reader a description of the population studied. Demographic information was collected for the gender, education, age and type of school variables. Gender yielded 71.2% male and 28.8% female participants. This result closely mirrors the true picture of the gender percentage of the principals of Botswana schools. According to the Ministry of Education school list, the population of principals is 21% female and 79% male. However the list has not been updated in the website to reflect the situation today (Ministry of Education Website, 2007).

About twenty one percent (21%) of the principals have diplomas (equivalent of an associate degree), 61% have junior degrees (the equivalent of a bachelor’s degree), and 18% have a masters degree. The high percentage in the junior degree category is expected because most teachers obtain a junior degree in order to teach in secondary schools. It is the standard qualification required of all teachers (Report of the National Commission on Education, 1993, p.337).

The age of the participants varied from 2.7% in the 20-30 range, 13.5% in the 31-40 range, 71.2% between 41 and 50, and 12.6% for the 51 and above range. The majority of the participants were between 41 and 50 years of age (71.2%).

Participants were drawn from the junior and senior secondary schools: 85.6% of the participants were junior school principals and 14.4 % were senior secondary school principals. This result is a close reflection of the number of schools of each type. There

are 206 junior (85.6%) and 27 senior secondary schools (14.4%). Table 2 below summarizes the demographic data described above.

Table 2: Demographic data 1

Demographics Sample Population%

Gender Male 71.2% Female 28.8% Education Diploma 21% Degree 61% Masters 18% Age 20-30 2.7% 31-40 13.5% 41-50 71.2% 51+ 12.6% Type of Junior school 85.6% school Senior school 14.4 %

Descriptive Data Analysis

Three independent variables (computer anxiety, perceived ease of use and perceived usefulness) and one dependent variable (behavioral intention to adopt) were tested to find out the perceptions of the principals regarding adopting computers in schools. Descriptive data on the three independent variables under observation: computer anxiety, perceived ease of use, and perceived usefulness, and the dependent variable: intention to adopt revealed both positive and negative skewness, and a large standard deviation. A large standard deviation means that there are relatively more participants scoring toward one extreme or the other (Moore, 2004), and this accounts for the high skewness of the data, as shown in Appendix A (Tables 12-15). The mean, standard deviation, and the skewness all reveal that the data were not normally distributed. This implied that more tests were needed to further examine the data for outliers and for the correction of the data for

further analysis using regression tests. Since data were skewed, this had an effect on the mean, which is sensitive to skewed data and outliers (Moore, 2004). Therefore, further analysis using the median, which is not sensitive to outliers, was necessary to further describe the findings.

To further examine the data for normality, following the descriptive data analysis, the Sharpiro-Wilk test of normality for checking the assumptions about a normal distribution was performed. The hypotheses tested included the null hypothesis: p>0.05 and alternative hypothesis: p<0.05. Shapiro-Wilk tests revealed a significance of p< 0.05, therefore there is not enough evidence to say that the data were normally distributed and the null hypothesis was accepted. This test confirmed that the data were not normally distributed and its results are displayed in Table 16 (see Appendix B).

To conclude the descriptive data analysis, the above tests have confirmed that the data were not normally distributed and that there were large standard deviations and extreme skewness in some of the items observed. The data had many extreme outliers and therefore this warranted further tests and treatment of the data before inferential statistics, using the regression model, could be tested. Since the regression model is sensitive to data that is not normally distributed (Moore, 2004), further tests and treatment entailed doing psychometric tests to test for the reliability of the scales, and for the removal of outliers.

The reliability of the TAM scales was established and the Cronbach alpha coefficients on the TAM constructs were quite high at .972 for the PU and .806 for the PEOU. TAM constructs were proved to be reliable measures with an alpha above .80 in this study. The next stage of analysis was performed on the three variables, perceived usefulness, perceived ease of use and intention to adopt and use computers, using the previously chosen statements (see Appendix C) to test the hypotheses and answer the research questions. Multiple regression was used because there was one dependent variable, intention to adopt and use, and two independent variables or predictors; perceived ease of use, and perceived usefulness. All items were measured on a 7-point Likert scale, where

1 was strongly disagree, 2; moderately disagree, 3; somewhat disagree, 4; neutral (neither disagree nor agree), 5; somewhat agree, 6; moderately agree, and 7; strongly agree. As explained above, the computer anxiety variable did not prove to be reliable as an independent variable for the TAM extension, therefore the testing of TAM was based on the original TAM variables. The computer anxiety variable, which had a large number of outliers, was explored further through interviews. Once the computer anxiety variable was removed, all 111 cases were included in the regression test in order to increase the performance of the regression model with more cases.

Regression Tests

The first step in inferential statistics was to perform casewise regression. This procedure removes influential outliers from the data set. This was done because earlier in this discussion, the Sharpiro-Wilk test had revealed a large number of extreme outliers in all the four variables under observation but once the computer anxiety variable was removed; all the 111 cases were re-examined using casewise regression. This type of regression selects the influential outliers for the model which can then be excluded from the data to correct it for regression tests, which assume normality of data. Seven outliers were shown to be influential in the TAM and were thus removed from the data. Regression tests were performed on the remaining 104 cases. Table 18 (see Appendix D) shows the outliers that were removed from the data based on the casewise regression test. After removing the outliers, several procedures were performed on the data to check the normality of data before applying regression tests. Consistent with multiple regression procedures, before the actual regression model was tested, the data were tested for multicollinearity. This is a situation where independent variables correlate with each other at a high level. Multicollinearity can result in misleading regression output (Keith, 2006). To check for multicollinearity, Tolerance, and Variance-inflation factor (VIF) were measured. Tolerance is a measure of the degree of independence of one IV (independent variable) from another IV. Tolerance can vary from 0 to 1 with 0 denoting complete independence and 1 showing total dependence. The larger the values of Tolerance the better (Keith. 2006). VIF is the “index of the amount that the variance of

each regression coefficient is increased.” Smaller VIF values are desired and the common rule of the thumb for a large value of VIF is 10 (Keith, 2006, p 201). For a regression model to be effective, the assumption of “no perfect collinearity” has to be met. The results on Table 19 (see Appendix E) show that the assumption of no perfect collinearity has not been violated. All the values for Tolerance are sufficiently distant from one and all the VIF values are below 10. The results show that the independent variables do not correlate very highly, therefore regression tests were carried out next.

The next analysis was the testing of the TAM to find out whether principals in Botswana secondary schools intend to adopt or reject computers. The regression equation tested is Y = a+bX1 + bX2 + e. The result of the regression model shows an R square of .273. The normal P-P Plot of standardized residuals and the scatter plot of the three variables show a closer to normal distribution as shown on Figures 10 (a) and (b) (see Appendix F), therefore we can accept the results of the regression model because the assumption of a normal distribution was satisfied. The results have largely confirmed the findings of prior research that TAM is robust in explaining the user acceptance of the system. The results indicate that overall TAM is statistically significant at R of .522 and adjusted R² of .273, with the two independent variables (PU and PEOU) accounting for 27% of the variance in the intention to adopt and use computers (BI). The adjusted R² of .273 means that the model explains about 27% of the variance in usage intentions. The results of the variance explained by the model are shown on Table 20 (see Appendix G). The next step in the analysis was to test the hypotheses.

Hypotheses Testing

Hypothesis 1 hypothesized that there was a negative relationship between computer anxiety and the principal’s intention to adopt and use computers. This was not tested due to the unreliable scales with a Cronbach alpha score of .215 and -.093 and extreme outliers in the computer anxiety construct. This phenomenon was explored in the interview section of this dissertation.

Hypothesis 2 tested whether the principals’ perceived usefulness (PU) of the computer had a significant influence on their intention to adopt and use the computer in their work, while Hypothesis 3 tested whether the principals’ perceived ease of use (PEOU) of the computer had a significant influence on their intention to adopt and use the computer. Hypothesis 4 tested whether there was a positive relationship between perceived ease of use (PEOU) and perceived usefulness (PU) of the computer and the intention to adopt and use computers. Both PU and PEOU explain 27% of the variance in the intention to adopt and use computers, supporting Hypothesis 4, Hypothesis 2 and Hypothesis 3. Both PU and PEOU have an effect on BI, which pertains to the intention to adopt and use of computers.

Further evidence supporting Hypothesis 2 and 3 is based on the betas. The relative absolute magnitudes for a given item reflect their relative importance in predicting respondents’ intentions to adopt and use computers. The beta is the meaningful interpretive choice because it compares the relative importance of several variables in a regression model since each standard deviation increase in X should result in an X amount standard deviation in Y. The results on the beta of the individual item of the independent variables reveal a predictive power of the PU and PEOU variables as shown in Table 3 below. Each of the two independent variable items had a statistically significant effect on the intention to adopt and use, since the standardized regression coefficient beta for PU statement were .094, .084, .277 and –.488, indicating that each additional impact of PU, intention to adopt increases by the above points and the same explanation goes for PEOU with standardized regression coefficient beta for PEOU at .350, .416, -.200 and -.141, supporting H2 and H3 respectively. Therefore PU and PEOU are significant determinants of intended use in this study.

Betas are only compared within a model, not between. They are highly influenced by misspecification of the model and adding or subtracting variables in the equation will affect the size of the betas (Moore, 2004). The negative betas (slope) for the statement; "Increases productivity (PU) and two PEOU statements; “Understandable,” and “Learning easy," may be a result of the removal of outliers to make the data closer to a

59 normal distribution or they may indicate possible adoption issues among the research population (see table 3 below).The later was investigated in the analysis of the interviews. Table 3 shows the PU and PEOU Item statistics for the betas.

Table 3: PU and PEOU Item Statistics 1

PU/PEOU items Standardized

Coefficients Beta Perceived Usefulness

Accomplish tasks PU .277

Supports job PU .094

Increases performance PU -.488

Increases productivity PU .084

Perceived ease of use

Understandable PEOU -.200

easier to do job PEOU .416

Learning easy PEOU -.141

Easy to remember PEOU .350

Distinct Hypothesis and Research Questions

To conclude the hypotheses testing, consistent with the TAM constructs, the distinct hypothesis tested was “secondary school principals who are adopters of computers will: (1) adopt computer technology for use in their work; (2) exhibit positive feelings about the usefulness of computers in their schools; and (3) exhibit positive feelings about the ease of using computers in their schools.” Based on the results of the regression model with an R square of .273 (shortened to 27%), it can be concluded that there is substantial support that the participants who found the computers easy to use and useful in their job, intended to adopt and use them.

Research Questions

Question 1: What are the perceptions of the principals towards adopting computer technology in their schools? The following four sub-questions addressed this question, these are:

Question 1 (a): Do school principals report feelings of anxiety about computer adoption and use?

The question on feelings of anxiety about computer adoption and use also had mixed responses and many outliers (22) as indicated previously. However, the computer anxiety variable could not be tested statistically to determine if it can be used to predict intention to adopt and use computers because the low reliability (.215 and -.093) of the scales, based on Cronbach alpha test, indicated that the scale was not acceptable for statistical analysis. Therefore, there was no statistical evidence in this research to support the assertion that computer anxiety has a negative impact on the intention to adopt and use computers among school principals in Botswana. Computer anxiety was explored further through interviews, which are discussed in the next section.

Question 1(b): What are the principals’ perceptions of the usefulness of computers in their schools?

On the principals’ perceptions about the usefulness of computers, most participants reported that they found computers useful in their job. This finding was supported by the standardized regression coefficient betas for PU statements; Supports job PU (.094), Increases productivity (.084), Accomplish tasks (.277), indicating that principals agree that computes are useful. For the “Increases performance” (–.488) statement, the slope was negative, therefore it is not certain that principals believe that computers increase job performance. However the positive betas indicate the impact those PU statements have in predicting system usage or intention to adopt computers. Therefore it can be concluded that principals reported that computers are useful in their schools. There were 13 outliers, which indicate that not all participants find computers useful. This issue is explored further in the discussion of interviews.

Question 1(c): Do school principals report that computers are easy to use?

The question on the ease of use of the computer also revealed a positive response with the majority of the principals reporting that they found computers easy to use. This finding was supported by the standardized regression coefficient betas for PEOU statements; “Easy to remember” (.350), and “Easier to do job” (.416) indicating that each additional impact of PU, intention to adopt increases by the same amount. However the negative betas for “Understandable” (-.200) and “Learning easy” (-.141), are very important because they signal the researcher to possible problems with computer adoption. These were explored in the interviews. In this variable there were fewer outliers (5); the majority reported that they perceived computers to be easy to use. The outliers point to the fact that not all principals found computers easy to use. Therefore the negative betas above are an indication of a possible problem with the adoption process among school principals in Botswana.

Question 1(d): Do school principals intend to adopt and use computer technology in their schools?

The question on the perceptions of the principals regarding adopting computer technology in their schools produced mixed responses. Based on the results of TAM, the majority of principals reported that they intend to adopt computers. However, there were some outliers, which indicate that not all principals want to adopt and use computers. It can be concluded that the research group is divided into adopters (90.1%), neutral responders (3.6%) and non-adopters (6.3%), with adopters in the majority as proved by the Technology Acceptance Model results, which showed a variance of 27%. The interview is used to explore this issue further. Figure 4 (bar graph) below shows the distribution of the responses to “intention to adopt and use computers.” As was established, adopters are in the majority.

Intention to adopt computers

40 76.6 Percent

11.7

4.5 3.6 0 1.8 1.8 strongly moderately neutral slightly agree moderately strongly agree disagree disagree agree Intention to adopt computers

Figure 4: Bar graph for intention to adopt

Analysis and Summary of the Survey

Intention to adopt

The Technology Acceptance Model (TAM) survey above addressed whether school principals wanted to adopt computer technology. TAM was robust in predicting the principals’ intention to adopt and use computers. TAM has been replicated, tested and applied in most parts of the world. Researchers using this model reported its robustness and suitability for explaining users’ intentions to adopt information technology (Dasgupta, Granger & McGarry, 2002; Gefen, & Straub, 1997; Venkatesh, 2000). TAM has been tested and proven successful in predicting 40% of system use (Colvin and Goh, 2004; Legris, Ingham, & Collerette, 2003). The evidence of TAM’s robustness in explaining intention to adopt a system was established in this study with a 27% of the variance explained by the model. Therefore it can be concluded that those who find computers easy to use and useful in their job are more likely to adopt them. Another thing found in this research is that there are three populations within the research group. These are; the neutral principals, they were neither adopting nor rejecting computers (3.6%), adopters (90.1%) and non-adopters (6.3%) of computers. This finding warranted further investigation to better understand why some people adopt computers while some reject them. In this research, there was some indication that not all principals will adopt and use computers; therefore further investigation using interviews was necessary.

Perceived usefulness

The Perceived usefulness (PU) variable had outliers and this indicated possible adoption problems for some principals. Perceived usefulness is the extent to which a user believes that computer technology will enhance his/her job performance (Venkatesh, 2000). If the school principals do not believe that computers are useful in their jobs, they are not likely to adopt them. Although in three of the four PU statements tested in the TAM (Supports job, Increases productivity, and Accomplish tasks), principals confirmed the usefulness of the computer to them, in one statement (Increases performance (–.488) the beta was

64 negative. According to Pedhazur (1982, p.59), “when a beta is being tested for significance, the question being addressed is whether it differs from zero while controlling for the effects of the other independent variables.” Therefore the negative beta for “increases performance” differed from a zero and showed an inverse relationship. The effect of this statement causes the prediction on the BI (Behavioral Intention to adopt) to decrease by .488. This was a very interesting finding and it differs from previous literature on the same variable (Davis, 1989; Venkatesh, 2000) because there was not enough evidence that principals believed that computers made them perform better in their job. This was another interesting finding which is worth examining further in the interviews, to confirm it. The question raised here is, “why do some principals seem to think that computers do not increase their performance?”

Perceived ease of use

Perceived ease of use is an important factor that influences user acceptance and usage behavior of information technologies. Although the regression model was robust in explaining the principal’s intention to adopt and use computers at an R square of 27%, still there were outliers, therefore it can be concluded that not all principals report that that computers are completely easy to use and useful in their job. This assertion is supported by the negative betas (slope) for the PEOU statements; “Understandable” and “Learning easy.” However there were positive betas for two PEOU statements; “Easy to remember” and “Easier to do job”, indicating that PEOU does have an impact on the users’ intention to use the computer. The existence of outliers and the negative betas warranted further investigation through interviews to explore the PEOU phenomena and to further investigate the participants’ perceptions. The negative betas for “Understandable” and “Learning easy.” statements might mean that these did not have the expected effect of increasing the prediction on the BI (Behavioral Intention to adopt) in the TAM model, and this differs from previous literature on the same variable (Davis, 1989; Venkatesh, 2000). On the other hand, as previously mentioned betas are affected by misspecification of the model and adding or subtracting variables in the equation

(Moore, 2004). However, it is possible that participants have problems with learning how to use the computer, therefore the interview dealt with that issue. The question raised here is on learning or interacting with the computer. The specific question is, “Do principals seem to think that it is not easy to learn how to use the computer?”

Computer Anxiety

As explained above, the data in this research were not a normal distribution because of varying standard deviations and outliers. The computer anxiety variable had the most outliers (22). Outliers signal the researcher to the possibility of different populations within one sample. All the outliers are very important sources of information because they show that not all principals report that they wanted to adopt computers. One of the problems identified in the literature on IT adoption was that some individuals might not want to accept or adopt and use computer technology (Davis, 1989). Another problem that has been identified in the literature is discomfort and anxiety in the use of computers. Although the literature review established that computer anxiety has an effect on the behavioral intention to adopt computer technology (Hackbarth, Grover & Yi, 2003; Orr, 1993), this research did not have enough evidence to predict intention to adopt computers from the computer anxiety variable. The computer anxiety variable had a lot of outliers (22), and the Cronbach alpha coefficient values were too low at .215 and -.093.

There are two possible reasons why the reliability coefficient alpha was this low for previously proven reliability of the computer anxiety variable. Firstly, as Hendrickson et al (1993) pointed out in their research, subjects might feel obligated to give certain answers because it is the proper thing to do, and this can affect the reliability of the instrument. In this research, as previously pointed out, the data were not normally distributed; it was extremely skewed because the majority of the subjects reported positively to accepting and using computers, despite the outliers in the PEOU, PU and Computer anxiety variables, as noted before, in the contrary. Secondly, the very low coefficient alpha of the computer anxiety variable might point to the possibility that the variable does not fit perfectly as an independent variable alongside PU and PEOU into

the established TAM 1, which has only two independent variables, PU and PEOU and one dependent variable, behavioral intention to adopt and use a system. Therefore this construct was explored further in the interviews because the outliers revealed that there were mixed feelings about computer anxiety among school principals in Botswana. As Compeau &Higgins (1995, p. 189), have put it, “computer anxiety is related to computer self-efficacy, which was found to exert a significant influence on individuals’ expectations of the outcomes of using computers, in terms of their emotional reactions to computers (affect and anxiety).” All these findings warranted investigating further in the interviews. The question raised here is “Are school principals comfortable with using the computer?”

Chapter Summary

To conclude the hypotheses testing, consistent with the TAM constructs, the distinct hypothesis tested was, “secondary school principals who are adopters of computers will: (1) adopt computer technology for use in their work; (2) exhibit positive feelings about the usefulness of computers in their schools; and (3) exhibit positive feelings about the ease of using computers in their schools.” Based on the results of the regression model with an R square of .273, it can be concluded that there was substantial support that the participants who found the computers easy to use and useful in their job, intended to adopt and use them.

In summary, Chapter 4 presented the quantitative analysis of data collected through the administration of a survey questionnaire. Descriptive statistics and inferential statistics were provided to illustrate the characteristics of the respondents and the theoretical model testing. In the testing of the theoretical model, the psychometric properties of the measures were assessed in terms of reliability. The results of hypotheses testing and the research questions were also presented.

The next chapter deals with the gaps found in the survey analysis above. As explained in the methods section above, this study used an Explanatory Design to collect information

67 sequentially (Creswell, 2002). That is, quantitative information was collected using a survey, and then qualitative questions were developed from the gaps found in the survey. The next section, the interview discussion, fills in the gaps found in the survey discussion. Based on the analysis above, the questions raised from the survey discussion were as follows:

1. Why do some principals seem to believe that computers do not increase their job performance? 2. Do some principals think that it is not easy to learn how to use the computer? 3. Are school principals comfortable with using the computer?

To further investigate the above findings, objectives and questions were formulated in the interview section to address them. These are discussed in Chapter 5, which follows next.

CHAPTER 5: INTERVIEW ANALYSIS AND RESULTS

Based on the survey questionnaire results, ten participants, six male and four female, from senior and junior schools, were purposively selected to take part in telephone interviews. According to Creswell (2002), It is typical in qualitative research, to only study a few individuals or a few cases. This is because the overall ability of a researcher to provide an in-depth picture diminishes with the addition of each new individual or case. (p.197) Therefore, in accordance with Creswell’s statement above, ten participants seemed to be a reasonable number for the qualitative portion of the present study.

Purposive sampling was employed for this portion of the study. It was the best option in this research because the results of the survey, as previously shown, had demonstrated that there were three distinct populations in this research: the adopters, neutral respondents, and non-adopters. Therefore members of all three groups identified in the survey results would have equal chances of representation in the interviews across gender, type of school, location and age. According to Babbie (2001), purposive sampling is necessary if interview data collection should be based on the knowledge of the participant. In this study there were adopters, neutral respondents and non adopters of computer technology, therefore purposive sampling was warranted.

Type of Interview Questions

The TAM survey results confirmed the hypothesis that if users find a computer easy to use and useful they are more likely to adopt it. The survey results showed that perceived ease of use (PEOU) and perceived usefulness (PU) explain 27 % of the variance in the dependent variable: intention to adopt computers (BI). However, knowing whether a principal wants to adopt computers or not, was only the first step in this research. The next step was to fill in the gaps and to answer the questions raised in the survey analysis. Fredland (2000), discussing the issue, says possible reactions of the intended recipients of

the technology and their desires to accept new innovations need to be examined. The two objectives that were originally formulated for the qualitative part of the study were 1. To find the gaps created by the survey analysis 2. To find out how the principals of these schools intend to support adoption and use by all users in the school.

In the light of the findings in this research, the research objectives and questions above needed to be complemented with three (3) questions raised in the survey analysis to fill in the gaps found in the research to answer research question 2 and two questions (2) to answer research question 3. Therefore the interview consisted of two parts: Part 1 addressed questions raised from the TAM survey analysis (question 2) while Part 2 addressed the original question (question 3) raised at the beginning of the study, pertaining to the actions principals intended to take to spread the use of computers in the school.

Part 1: Technology Acceptance Model Constructs

Three (3) questions were raised in the TAM survey analysis. TAM results established that there are adopters and non-adopters within the research population. Also, the outliers for the computer anxiety, PEOU and PU variables signal the researcher that non-adopters of computer technology existed within the research population. The research questions for this section were designed to solicit self reports from principals on their feelings concerning the three independent variables.

To address each of the issues raised above, objectives and interview lead questions were formulated. The first objective addressed whether principals reported that they have computer anxiety. It asked the question, “Are school principals comfortable with using the computer?” The next objective dealt with whether the computer increases job performance and the question was, “Do some principals think that computers do not increase their job performance?” The last objective addressed the issue of learning how to

70 use the computer. The question asked was, “Do some principals think that it is not easy to learn how to use the computer?”

The objectives, questions, and interview lead questions asked appear in Table 4 below.

Table 4: Part 1 Interview objectives and 1

Objective Question Interview lead Question

To find out if principals are Are school principals “What is your comfort zone comfortable with computers comfortable with using around computers? Do you the computer? feel comfortable around computers?”

To investigate how useful Do some principals “Do you find the computer the computer is to school think that computers do useful to you as a school principals not increase their job principal?” performance?

To find out if learning how Do some principals “Do you find the computer to use the computer is a think that it is not easy easy to use or learn how to problem for principals to learn how to use the use?” computer?

Interview questions

The previous section introduced and discussed the interview questions as posed by the researcher as well as the objectives behind the questions. In this section the responses of

71 the participants to the three questions posed were discussed. To protect the identity of participants, participants were given false names. Question 1 is discussed next.

Computer Anxiety: Self Descriptions

Question 1: Are school principals comfortable with using the computer?

The Computer Anxiety variable was removed from the TAM because of its unreliable scales. It had Cronbach alpha coefficient values of .215 and -.093, which were not suitable for statistical analysis. According to Keith (2006), unreliable scales cause researchers to underestimate the effects of one variable on another. It therefore was necessary to explore the computer anxiety variable in the interviews. The main questions the researcher asked the participants were, “What is your comfort zone around computers?” and “Do you feel comfortable around computers?”

In response to the questions above, two participants, Setho and Tebogo, reported that they were very comfortable with the use of computers. They described themselves as “very comfortable.” Three participants, Otsile, Baeti and Botho, described themselves as “comfortable” around computers, while three participants, Selelo, Tshekiso and Batsile expressed some discomfort around computers. They said they were “not comfortable” around computers. Tshekiso, expressing the extent of the discomfort said, “I am a little bit comfortable but not very much.”

During the interview participants described the fear of computers using several adjectives. They used the words: afraid, fear, phobia, intimidation and lack of comfort. It can be concluded that that use of these words represent a real problem among the participants. Here is what Baeti said in relation to this fear, “Also people are afraid of new things and they consider the computer to be too complex and complicated beyond their understanding.” Taolo, confirming this fear as discussed above said,

“Eh.. personally, you see, I see it as a good thing, but you see, I have a phobia for computers and also I am old, I am just about to retire. You see the phobia is not about computers only; it’s about all the other new gadgets as well. So personally I have that phobia.”

Tebogo, introducing another synonym of phobia or fear said, “I have got no intimidation.” This phenomenon was confirmed again by Otsile when he said, “People still have this fear for computers.” The other one, referring to himself, said, “I am not comfortable because I do not have enough knowledge on its use”

The word phobia was introduced by the participants to answer a question that asked about their “comfort.” One of the participants, Ditshupo was non committal, and did not seem to label himself as comfortable or uncomfortable. Ditshupo said, “I do not have phobia as such” In response to the question asking about computer comfort. This kind of response was observed in the survey responses where some principals opted to be neutral, neither admitting that they experienced computer discomfort or not.

In summary, it can be concluded that according to the reports of the principals, some principals are comfortable around computers while others are not. One of the findings from the interviews is that there are distinct groups within the research population. There are two people who fall in one group of “Very Comfortable” users, three in a group of “Uncomfortable” users, three in a group of the “Comfortable” users, one in the “Neutral” position, and one in the stage of “Phobia.” This result ties in very well with the results of the survey where the computer anxiety variable had 22 outliers. This indicated that within the research population, there was a group of participants who reported that they experience some computer anxiety. Therefore, the interview findings were instrumental in establishing the different levels of comfort and discomfort around the computer. The reasons for the different stages of comfort and discomfort unfold throughout the discussion of the interviews. Four statements from TAM (Computer anxiety variable)

were used to analyze the findings statistically, consistent with the four statements used for both PU and PEOU. These statements were: 1. Working with a computer makes me nervous, so I avoid using it if I can 2. I feel threatened when other people talk about computers 3. I feel at ease in a computer training workshop 4. I get a sinking feeling when I think of trying to use a computer

Four tables showing the TAM statistics supporting the computer anxiety reported above in four different statements are shown below.

Table 5: Nervousness Anxiety 1

Valid Cumulative Frequency Percent Percent Percent Valid strongly disagree 87 78.4 78.4 78.4 moderately 6 5.4 5.4 83.8 disagree slightly disagree 7 6.3 6.3 90.1 neutral 7 6.3 6.3 96.4 slightly agree 2 1.8 1.8 98.2 moderately agree 1 .9 .9 99.1 strongly agree 1 .9 .9 100.0 Total 111 100.0 100.0

Table 5 above shows the TAM survey statistics that confirm and support the results of the interview that some participants had computer anxiety. The results show that the cumulative number of participants who reported feelings of nervousness around the computer is four (4) at 3.6%. Equally interesting is the seven (7) neutral respondents at 6.3 %. These results complement and qualify the interview findings discussed earlier.

The figures from the TAM statistics in Table 6 below complement the findings on the interviews that some participants do not feel at ease around computers and in workshops. Altogether 81 (73%) participants did not think that they are at ease around computers.

This is a very high percentage and probably explains the low Cronbach alpha findings discussed earlier. It is also possible that the statement was misunderstood by the participants.

Table 6: Feeling at ease ANXIETY 1

Valid Cumulative Frequency Percent Percent Percent Valid strongly disagree 8 7.2 7.2 7.2 moderately 6 5.4 5.4 12.6 disagree slightly disagree 10 9.0 9.0 21.6 neutral 6 5.4 5.4 27.0 slightly agree 12 10.8 10.8 37.8 moderately agree 21 18.9 18.9 56.8 strongly agree 48 43.2 43.2 100.0 Total 111 100.0 100.0

The statistics in Table 7 below also complement the findings in the interview that found substantial support from the self reports of the participants who “feel threatened when other people talk about computers.” A total of 10 (9%) participants agreed to feeling threatened and six (6) at 5.4% participants were neutral.

Table 7: Feeling threatened ANXIETY 1

Valid Cumulative Frequency Percent Percent Percent Valid strongly 84 75.7 75.7 75.7 disagree slightly 11 9.9 9.9 85.6 disagree neutral 6 5.4 5.4 91.0 slightly agree 4 3.6 3.6 94.6 moderately 5 4.5 4.5 99.1 agree strongly agree 1 .9 .9 100.0 Total 111 100.0 100.0

In Table 8 below, a total of eight (7.2%) participants reported that they got a sinking feeling when they thought of trying to use a computer. Five (4.5%) participants were neutral on this statement. These results support the findings of phobia and discomfort discussed in the interview analysis above.

Table 8: Sinking feeling ANXIETY 1

Valid Cumulative Frequency Percent Percent Percent Valid strongly disagree 77 69.4 69.4 69.4 moderately 12 10.8 10.8 80.2 disagree slightly disagree 9 8.1 8.1 88.3 neutral 5 4.5 4.5 92.8 slightly agree 4 3.6 3.6 96.4 moderately agree 3 2.7 2.7 99.1 strongly agree 1 .9 .9 100.0 Total 111 100.0 100.0

The next section discusses the question on the usefulness of the computer to the principals. Figure 5 below shows the groups as described above and their self description about their comfort zone around computers.

Figure 5: Computer Comfort Self descriptions

Perceived Usefulness: Self Descriptions

The next question dealt with the usefulness of the computer to the school principals. Three statements, out of four PU statements (as discussed previously in the survey analysis), had positive betas, and only one statement on “Increases performance” had a negative beta. If the possibility of the misspecification of scales is ruled out as the cause of the negative beta (Moore, 2004), then it is possible that some principals might not

think that the computer has the potential to increase their job performance. This question was based on the survey results which showed a negative beta of -114, for the “Increases performance” statement, which was an inverse relationship (Moore, 2004). Therefore, this statement was investigated further in the interview.

Question 2: Do some principals think that computers do not increase their job performance?

The PU variable had the least outliers (5), indicating that most principals perceived the computer to be useful to them. In response to the interview question on the perceived usefulness of the computer, all participants described the computer as useful. The interview question that asked about the usefulness of the computer was responded to positively. The question was, “Do you find the computer useful to you as a school principal?” All ten participants reported that the computer was very useful to them. Most of the adjectives used express the extent of its usefulness to them. Some of these are, “essential and necessary”, “very useful” and “highly useful.” Two participants expressed very strongly how useful the computer is to them. One of them responded by saying, “Oh Yes! Ah! In these times, is there anybody who can say they do not see its usefulness? Using the computer is not a choice.” The other participant said, “Extremely, it is not a question of usefulness, it is part of my life as a school head.”

All the adjectives and statements summarize the extent of the usefulness of the computer. Therefore, there is no evidence from the responses that the principals report that computers do not increase work performance. It is possible that since betas are sensitive to adding and removing cases in a model, the negative beta on “Increases performance” was a result of the removal of outliers (Moore, 2004; Pedhazur (1982). Figure 6 below shows the responses.

Figure 6: Perceived usefulness

Perceived Ease of Use: Self Descriptions

The last question of Part 1 of the interviews focused on the PEOU variable. Two negative betas on “Understandable” and “Learning easy” showed an inverse relationship between the BI and PEOU variables (Pedhazur (1982). This result signaled the researcher to

possible problems with the ease of using the computer. This was investigated in the interviews and the question asked was, “Do you find the computer easy to use or learn how to use?” Varied responses were reported to this question, with some principals saying they find it easy to use and others reporting that they do not find it easy to use.

Question 3: Do some principals think that it is not easy to learn how to use the computer?

The responses to question 3 indicated that there were two major camps within the research group. The first camp of participants consisted of all of those who had previously reported that they are very comfortable around computers and two participants who said they were comfortable with computers. They said that they find the computer easy to use. The reasons given for the ease of use by the members in this camp were practice and the interactive nature of the computer. Baeti, who said she was comfortable about computers, also said: “It is easy to use because it interacts with you. If you make a mistake it tells you.” The main reasons given here are practice and the interactive nature of the computer. In the second camp were those who did not find the computer easy to use. This camp had one member, Otsile, who had earlier said that he was comfortable about computers, all three members who were uncomfortable, Selelo, Tshekiso and Batsile, the phobia participant, Taolo and the neutral member, Ditshupo, making it six members altogether.

Participants in the second camp confirmed that the learning and use of computers poses some challenges. One principal, commenting on the learning and use of computers said that, “It’s a real problem.” Otsile expressed that he is “still struggling to learn how to use the computer.” Selelo responded that: “at the moment it’s not easy for me, I don't find it easy to use.” This is supported in the literature on computer use in schools (Flanagan and Jacobsen, 2003). On examining the data, several factors or problems were discovered. The factors or causes of lack of perceived ease of use, discussed next, explain why it was regarded as a real problem to learn how to use computers. They complement the theme on computer anxiety discussed earlier.

One problem that surfaced during the interviews is that of lack of practice on the computer. The participants mention lack of practice by themselves and all of them seem to agree that practice and usage of computers is essential if one wants to master their use. One participant, commenting on lack of practice by themselves says, “Practice makes perfect.” Another participant summarized the practice factor in one metaphor: “I think the greatest point is that computers are not like riding a bicycle, if you don't practice, you forget, the skills sort of diminishes as you reduce the usage and practice.” This metaphor explains the fact that learning how to use the computer is a dynamic process, it is not static, therefore the skills one has need to be constantly renewed through practice and usage. Unlike the skills of learning the bicycle (which are routine), the rules of learning the use of the computer keep on changing. One other participant supports this statement when he says, “You can’t say you are computer literate without using it.” Yet another participant said, “They say practice, practice, practice can help you. Right now I wouldn't say I am comfortable because of lack of practice. Right now I have lost most skills I had because of lack of practice.” The participant mentioned the word practice five times and one would think it emphasizes the importance attached to practice.

The time factor appeared to be yet another barrier to the learning and use of the computer. The participants mentioned lack of time from two points of view. Some participants felt that they did not make the time to learn and use the computer while others felt that the employers ought to create the time for them to learn. Those who subscribe to the first view said, “I created time to learn.” The other participant blamed himself for the lack of time; “I do not give myself enough time to be given more information on the use of the computer. There are some opportunities if you give yourself time to learn you can learn.”

Those who thought the employers should create time wanted training to be formalized. One said, “I think eh training should be formalized you know, because if it is formalized, eh… everybody will find time to do it.” Another participant, confirming the above said, “The problem that we faced was that there was limited time during the course.” Other participants said, “If we could have formal courses, maybe two weeks of training could

help” and “Yes, let me say training..eh we keep on going to workshops here and there and when you are at those workshops computer training is included but its not enough because you just go there for one day.” Batsile mentioned that they did not “get enough training on them.” Ditshupo, whose response to the comfort with computers question was neutral, said that during training he “could not grasp” most of the things introduced. This quote from Ditshupo is revisited in the analysis section because it is possible that this participant is really one of the uncomfortable members although he did not explicitly say so in this section Ditshupo’s situation is discussed in the analysis. Taolo (phobia participant) had not tried to learn how to use the computer because of the dread of computers and “all the new gadgets.” He said, “you have to be bold or “segatlhamela masisi” to learn how to use it.” {Segatlhamela masisi is a setswana phrase for describing a brave and fearless person}. Therefore time and training seem to be intertwined in this theme.

Based on the above mixed responses to the question on the ease of using the computer, there were indications that indeed there were some principals who did not find the computer easy to learn how to use. Six members out of the ten participants reported that the computer was not easy for them. As for the causes of the lack of ease of use, some members mentioned that the training was not adequate, others mentioned lack of practice, while Taolo, had a phobia about computers and therefore could not learn how to use them.

In summary, the two camps discussed above agree that one of the reasons for lack of ease of using the computer is lack of practice. Those who did not find it easy to use mentioned lack of training and anxiety or phobia around computers, and time constraints, in addition to practice. Table 9 below shows that a total of 22 people (from adding those who disagree) did not think that “the interaction with the computer is clear and understandable.” The cumulative percentage was 19.8% which is quite substantial, thereby explaining the outliers and the negative beta for the statement as well as complementing the findings of the interviews.

Table 9: Understandable PEOU 1

Valid Cumulative Frequency Percent Percent Percent Valid strongly disagree 6 5.4 5.4 5.4 moderately 8 7.2 7.2 12.6 disagree slightly disagree 8 7.2 7.2 19.8 neutral 17 15.3 15.3 35.1 slightly agree 18 16.2 16.2 51.4 moderately agree 31 27.9 27.9 79.3 strongly agree 23 20.7 20.7 100.0 Total 111 100.0 100.0

Table 10 below shows the TAM statistics for the perceived ease of use statement for “Learning to use a computer is easy for me.” A total of 28 people and a cumulative percentage of 25.2% did not think that the computer is easy to learn how to use. This finding explains the outliers and the negative beta discussed earlier for this statement and complements the findings of the interviews.

Table 10: Learning easy PEOU 1

Valid Cumulative Frequency Percent Percent Percent Valid strongly disagree 6 5.4 5.4 5.4 moderately 9 8.1 8.1 13.5 disagree slightly disagree 13 11.7 11.7 25.2 neutral 11 9.9 9.9 35.1 slightly agree 16 14.4 14.4 49.5 moderately agree 27 24.3 24.3 73.9 strongly agree 29 26.1 26.1 100.0 Total 111 100.0 100.0

The above section concludes the analysis of the interview questions included in Part 1, which established whether participants reported that they were comfortable around computers, and whether they reported that computers were easy to use and useful to them as school principals. The interview participants largely confirmed the results of the

83 survey, which had shown earlier in the dissertation that even though the majority of principals wanted to adopt and use computers, there were some outliers, indicating that there was a likelihood of a different group of users within the research population.

Figure 7 below shows the self reported statements from the participants regarding perceived ease of use of the computer.

Figure 7: Perceived ease of use

Part 2: Future of computers in schools

The responses to the questions in Part 1 provided data with which to describe the principals’ feelings about adopting and using computers. Part 1 established that there is some computer anxiety among principals, and that not all principals find the computer easy to use or learn how to use. Also data on the level of discomfort around computers was also discussed. This section built on that information to classify and describe the principals’ actions and plans for the use of computers in schools in order to answer question 3 of the research questions suggested in this study.

Part 2 dealt with the objective formulated for the qualitative portion of the study. As already mentioned, the main research objective and question here were “to find out how the principals of these schools intend to support adoption and use by all users in the school” and “what are the actions school principals intend to take to support computer adoption in the school?” respectively.

This question is broad because not only does it address the actual actions the principals took to spread computer technology in schools, it also investigates the assumptions suggested earlier in this study about transformational leadership. In this study, principals are assumed to be in a position to transform the school system through computer technology. Todd (1999, p.4), writing on leadership says, “transformational leadership moves beyond managerial and instructional leadership to providing schools with strategies necessary to cope with change.” It is assumed that principals are leaders in their schools and therefore they play an important role in the adoption and use of computer technology.

To address the above, the main question was broken into two sub-questions and objectives. These are discussed in the next section. The first question asked about the uses of the computer while the second question asked the respondents to describe the plans they had for the school. Table 11 below presents the objectives, sub questions and interview lead questions.

Table 11: part 2 interview objectives an 1

Objective Question Interview lead Question

To find out how participants What are the principals’ self 1. “We have confirmed utilize the computer, reported uses of the that you find the computer? computer very useful. What do you use it

for?” 2. “Apart from storing and retrieving data what other uses do you put the computer to?”

To find out if principals What are the principals’ self “What are you future plans for intend to spread computer use reported plans for the future? spreading the use of the in the school computer in the school and for

educational purposes?”

Uses of the computer: Self descriptions

The first question in Part 2 asked about the uses of the computer. It was possible that an examination of the self reported uses of the computer would shed more light on the principals’ future plans for computers. Also an analysis of what uses a principal puts the computer to would provide data for classifying and describing the activities. Therefore, all the principals were asked to list their uses of the computer.

Question 1: What are the principals’ self reported uses of the computer?

In Part 1 above, three participants, Selelo, Ditshupo and Taolo did not use computers personally, instead their computing needs were accomplished by a secretary. This information brought about a change in the way this section was analyzed because these three participants would now form yet another group, that of non-adopters. From this point onwards, the three members are referred to as the non-adopters for the purpose of the discussion of this section. This is because they do not use the computer, so they cannot describe what they use it for. Secondly, since they did not use the computer they could not speak about implementing plans in the school. Therefore the analysis of this question was limited to the responses of the rest of the seven who did use computers, therefore the responses of the non-adopters were not considered.

The question the researcher asked the participants was, “We have confirmed that you find the computer very useful. What do you use it for?” On some occasions the researcher needed to do a follow-up to probe further. A follow-up question was, “Apart from storing and retrieving data, what other uses do you put the computer to?”

The responses to this question were very instrumental for putting activities into categories. Those participants who had earlier said they were uncomfortable and those who were comfortable around computers tended to agree on the uses of the computer and between them they had three activities they all reported. These were: writing confidential letters, storage of records, and retrieval of records. For example, in response to the question on the uses of the computer, Baeti said: “Mainly I use it to store and retrieve information when I need it.” Botho said, “I normally use it to store information, and let me say for record keeping.”

Some of the participants in these two groups had access to the Internet while others did not. It is possible that lack of access to the Internet was a limiting factor, as one participant reported in the interview. However when the same participant was later asked what the plans for the future were, the Internet was not mentioned. In addition, Botho, who reported that he used the computer for record keeping, had access to the Internet in the school. Therefore, while access to the Internet is a possible limiting factor, it is not discussed in that respect in this question. For the purpose of this category of users, it

suffices to say that all users identified typing confidential letters, storage and retrieval as the main uses of the computer. In this research, these participants were categorized as late adopters, because the assumption was that the principals should be transformational leaders. Principals, whose use of the computer is limited to word processing and storage of files, cannot be regarded as transformational leaders because of their limited visions of the uses of the computer. Transformational leaders are described as “change agents,” in relation to using technology (Doyle & Smith, 2001, Transformations, ¶ 1).

The two participants, Setho and Tebogo, who said they were very comfortable with computers, identified more varied uses for the computer than all the other participants. More than ten uses were identified. This is what Setho said:

Generally, almost it runs my office, it runs my office. (He laughs). It’s my databank, I dispose of correspondence with it, I invent and apply my things that I think of, I download materials I believe are needed in the school for the teachers and students and I give them. I carry our research with it, you know what, it runs my office, I do not hold a pen anymore, no, I don't carry things around, I have elected that the computer carry out those duties for me.

This statement summarizes and describes the extent of the usefulness of the computer to the particular principal. In this paragraph, the principal describes very advanced uses of the computer involving the use of the Internet for educational purposes, administrative duties, decision making (invent and apply), research, collaboration with the teachers and the students through shared knowledge (downloading), and communication. The other very comfortable participant, Tebogo, also mentioned equally advanced uses of the computer involving communication by e-mail, connecting the Internet in the school library for research purposes, building databases, as well as networking all computers in the school. Another very interesting thing to note is that one participant, Batsile, who described herself as uncomfortable with computers, reported that she uses the Internet for e-mail and searching the Internet, in addition storage and retrieval of information. This is what was reported: “I type my minutes …, write confidential letters to teachers, send e-

mail of minutes and search for information on the Internet.” All three participants have one thing in common: their uses of the computer go beyond word-processing and storage of information. They reported many uses for the computer while the other participants reported only three uses. Further, the uncomfortable and comfortable users were limited to using the computer for filing or storage of administrative information and word- processing of documents, while the very comfortable users mentioned communication by e-mail, research on the Internet, connecting the Internet in the school library for research purposes, building databases, downloading pertinent information for use by teachers and students from the Internet and using it for problem solving as well as using it for administrative purposes.

In conclusion, the very comfortable users went beyond using the computer as a word- processing and storage tool. Therefore in this research, they were labeled as early adopters. This is based on the definitions of transformational leaders discussed in the literature review in this study. Todd (1999) summarizes the characteristics of this category of leaders very well: Strategic leadership focuses on the longer term, where the scale and scope of action are school wide rather than program-focused. It demands a clear perception of role and a sense of vision. It is indicated in a long-term and futures-oriented technology plan for the school, a program of staff development that provides access to technology, and developing the staff's knowledge and skills about the pedagogy of integrating information technology into the classroom. (p. 5)

In summary, the findings above shed some light on what seems to be a clear demarcation between early adopters and late adopters. Also it has become clearer too that the three participants who did not use computers because the secretary did the work for them can be classified as the non-adopters because they did not intend to use the computer personally. Based on the above, it is quite possible that the link between the early adopter and the late and non-adopter is really the decisive factor between transformational leaders and those who are not. This is explained further in the analysis. Figure 8 below shows the different uses of the computer by the three groups.

Figure 8: Computer uses by all groups

Future plans for schools

Part 1 and 2 dealt with the main qualitative questions posed in the study, on the actions the principals intended to take to spread computer use. Question 2 of Part 2 introduced the self reported future plans of the principals in response to the interview question, “What are your future plans for spreading the use of the computer in the school and for educational purposes?” Part 2, question 1 above established the categories of users according to the reported uses of the computer. The categories established are early adopters, late adopters and non-adopters. In this section, these categories are used to clarify and expand on the discussion above.

Question 2:

What are the principals’ self reported plans for the future?

The responses to the question on the future plans revealed noteworthy differences amongst the very early adopters, the late adopters and the non-adopter.

The early adopters listed eighteen items they planned to have in their schools. Some of their future plans included: the use of the Internet both in the computer laboratories and in the library, networking all computers, disposing of administrative duties, as well training both teachers and students. Setho (early adopter) was already running a program on computer interaction for administrative staff and had made learning and using the computer a “performance” item for all staff. He said staff would be evaluated each year on the use of the computer. An example of the seriousness with which early adopters dealt with implementing change is when Setho, responding to a question on future plans, said:

Aaa! That’s very important. Now at the moment, for students we are limited because our prescription says computer awareness. But for teachers, eh now let me start with management. For the top five managers in the school, I have made it a performance issue and I have already signed for it, that using and learning or both are performance issues and for the rest of the teachers, I expect it to be

cascaded from top management down because at the moment we are lucky, for the past two weeks we are running a program on the interaction with computer. Our intention is that next year this time all of them should have the skills to do word processing and database processing and basic skills on computers because they are available now.

This statement describes the importance the early adopters attached to the future of computers in the school and seem to describe the qualities of transformational leaders as described earlier in the study (Telem, 1996; Todd, 1999; Yee, 1998, Yuen et al. 2003). The type of items the early adopters discussed are very advanced because they focus on using the Internet to improve learning, building shared vision and skills for teachers and students, decision making and the use of the library to enhance research and learning. By contrast, the late adopters focused on improving computer skills for students by increasing computer time and ensuring that students become confident in computer use.

The non-adopters did not have direct answers to the future plans, having admitted to not using the computer rendered it unnecessary to ask about the future of something they had not adopted. However, when they were asked whether they wanted to add anything to the interview before closing it, their reports reflected the importance of computers in schools. Here is what Taolo, who had earlier said he had a phobia for computers, said about the future, “I want to say that the computer is very essential in schools nowadays. We are in a new era, as I said before. You do not want to carry files all the time from one corridor to another. So the computer has many uses therefore we must have it in schools and we must all learn how to use it.” Selelo, who said she was uncomfortable with computers, referring to the future said, “I can just say nowadays the computer is very important, we are living in a world of computers, therefore it is very important that we have computers in our offices and that we are trained very well, not just one day training.”

In summary, the responses to the question on the future plans for the school were instrumental in expanding the theme that developed in this section. It became clearer in the analysis of these responses that there were distinct groups within the research

population. These groups may be characterized as early adopters, late adopters and non- adopters. The interview section confirmed the results of the Technology Acceptance Model which established that those who find the computer easy to use and useful are more likely to adopt it. As was discussed before, the findings from the TAM, which established that at a variance of 27%, PU and PEOU variables predict the principals’ computer acceptance, were not sufficient to establish why some principals would want to adopt or reject computers. The interviews filled this gap making it clearer that early adopters and late adopters adopt while non-adopters reject computers. The next section deals with the full analysis of the interviews, in the light of the findings discussed.

Analysis of Interview Responses

The telephone interviews were administered to 10 participants. The overarching research objectives for the interview section were, “To find the gaps created by the survey analysis” and “To find out how the principals of these schools intend to support adoption and use by all users in the school.” In this study principals were assumed to be in a position to transform the school system through computer technology; therefore knowing whether they intend to adopt computers is crucial for successful technology implementation.

In the light of the findings of this research, the research objectives above needed to be complemented with three (3) questions raised in the survey analysis to fill in the gaps found in during the survey analysis and two (2) questions to find out the future plans of the principals. Therefore the interviews consisted of two parts: Part 1 addressed questions raised from the TAM survey analysis, while Part 2 addressed the original question raised at the beginning of the study pertaining to the actions principals intend to take to spread the use of computers in the school. The analyses of the interviews were instrumental for a number of preliminary findings as discussed next.

The self reported future plans of the principals appear in Figure 9 below.

Figure 9: Future Plans for the school

Findings in Part 1

The TAM results had outliers, indicating the presence of different groups in the research population, the self descriptions of the participants were used to differentiate the groups. Four distinct groups were identified based on their self reported confidence levels around computers. The participants were grouped by like or similar characteristics to make the discussion more meaningful. The groups were based on the participants’ descriptions of themselves in response to the first interview question on computer comfort or discomfort. From the responses to the interview questions by the ten participants, the researcher labeled each according to their self-reported status. There were three participants who said they were comfortable around computers, these were labeled as “comfortable”, “2comfortable,” and “3comfortable.” Those participants who reported that they were very comfortable were labeled as “very comfortable” and “2 very comfortable.” One participant mentioned having a phobia and was labeled “phobia”, while the participant who tended to be neutral in his responses was labeled “neutral.” Three participants said that they were not comfortable around computers and were labeled as “not comfortable,” “not comfortable 3,” and “some computer discomfort.” Based on these self descriptions, the participants were grouped into “Very Comfortable”, “Comfortable”, “Uncomfortable”, “Neutral” and “Phobia” for purposes of analyzing the characteristics of each group. One of the findings in this research was that there seemed to be different levels of comfort around computers. These were “very comfortable” and “comfortable.” There were also different levels of discomfort, namely; “uncomfortable” and “phobia.” Also established in the interview is the neutral participant. This participant did not say how he felt about computers, in terms of comfort or discomfort. He chose to say that he did not have phobia. However, the analysis of the interview reports later revealed that he falls into the non-adopter group (because the secretary used the computer) and that his discomfort with computers put him under the uncomfortable users. To summarize, four categories of comfort and discomfort were established and these are very comfortable users, comfortable users, uncomfortable users and users who had a phobia for computers.

This intense fear for computers has been identified in the literature as computer anxiety. Compeau &Higgins (1995, p. 189), writing on computer anxiety mentioned that, “computer anxiety is related to computer self-efficacy, which was found to exert a significant influence on individuals’ expectations of the outcomes of using computers, in terms of their emotional reactions to computers (affect and anxiety).” This assertion is supported in this research. Based on the above, preliminary findings point to the fact that phobia and varying degrees of discomfort exist amongst some principals. Several findings support this assertion about the principals who reported that they had some kind of discomfort around computers. These are: 1. In the TAM analysis, the computer anxiety variable had many outliers. This indicated that the research population was not homogenous, and that there were distinct groups within the research population. 2. The self reports by some of the participants introduced several words that the researcher had not used: intimidation, fear, afraid and phobia. Some participants described themselves as uncomfortable around the computer and as having a phobia for computers. Compeau &Higgins (1995, p. 189), writing on computer anxiety mentioned that, “computer anxiety is related to computer self-efficacy, which was found to exert a significant influence on individuals’ expectations of the outcomes of using computers, in terms of their emotional reactions to computers (affect and anxiety).” This assertion is supported in this research.

Regarding the PU variable, there was no evidence from the responses that computers do not increase the principal’s work performance. It is possible that since betas are sensitive to adding and removing cases in a model, the negative beta on “Increases performance” was a result of removing outliers (Moore, 2004; Pedhazur (1982). Another possibility is social pressure, as Hendrickson et al. (1993) pointed out in their research, subjects might feel obligated to give certain answers because it is the proper thing to do. Therefore it is possible that the participants feel the pressure to conform to the objectives of the Long term Vision 2016. One of the objectives of Vision 2016 is “Botswana to recognize the importance of information and the development of efficient information systems and networks” (Long term Vision 2016, p. 71).

The PEOU variable revealed that not all users find the computer easy to use or learn how to use. This finding from the self reports of the participants strengthened the earlier findings in the TAM, where there were 17 outliers. As mentioned previously, the reasons for discomfort around computers unfolded in the PEOU section to shed more light on possible reasons for the computer anxiety issue discussed above. The first finding related to this variable is that there are two kinds of users, those who perceive computers to be easy to use and those who do not. On examining the self reports, it was found that six members out of the ten participants did not think that the computer is easy to use or learn how to use. Second, within the aforementioned groups, there were actual users of computers and those who did not use computers or who preferred to have their secretary perform computer-related tasks. Of the 10 participants interviewed, three did not use the computer personally. This was another interesting finding because it ties in very well with the research in the area, in which it has been established that, Although there is little doubt that technological developments will occur at a fast rate, it is not immediately obvious that individual users of the new technology will be able to adopt and use new technological artifacts at the same pace.” (Karahanna & Straub (1998, p. 238)

The reasons given for lack of ease in using the computer included lack of practice or usage, lack of training, time constraints, and anxiety or phobia around computers. Therefore the PEOU finding complemented the findings on computer anxiety reported earlier. It can, therefore, be assumed that there was some computer anxiety amongst principals and that one of the reasons for it was negative perceived ease of use of the computer. This finding is in line with previous research in the area, which established that computer anxiety has an effect on the behavioral intention to adopt computer technology (Hackbarth, Grover & Yi, 2003).

Findings in Part 2

Part 2 of the interview section dealt with the actions principals intended to take to spread the use of the computer in the school. Two questions were formulated to establish the

future plans principals had, the first on the uses of the computer and the second on the future plans principals intended to implement.

One finding on the uses of the computer is that there was a clear distinction between very comfortable users on one hand and the comfortable and uncomfortable groups on the other hand. The uncomfortable and comfortable users limited their use of computers to filing or storage of administrative information and word-processing of documents, while the very comfortable users mentioned communication by e-mail, research on the Internet, connecting the Internet in the school library for research purposes, building databases, downloading pertinent information for use by teachers and students from the Internet and using computers for problem solving as well as using it for administrative purposes. Based on this preliminary finding, it seems that it may be possible to distinguish those principals who are transformational leaders from those who are not.

For the purpose of this study, the very comfortable principals are referred to as early adopters and the uncomfortable and comfortable principals are referred to as late adopters. A third group of non-adopters emerged through the analysis of the interview data. Three participants had not adopted computers personally, but instead relied on the services of the secretary to perform computer tasks. Commenting on the use of the computer, one of these principals said that the secretary does the job for them (this participant was labeled as “neutral” in the study), the other principal said that there was no computer in the office and the computer laboratory was too far away, while the third one said that he had a phobia for all new gadgets, including the computer.

The findings in the next question, on the future plans, clarified and expanded on the previous findings. In the responses to this question, the distinction between early adopters, late adopters and non-adopters became clearer.

The finding here was that, of the three groups, the early adopters had between them eighteen items they were planning to have in their schools. Some of their future plans included: the use of the Internet both in the computer laboratories and the library,

networking all computers, disposing of administrative duties, as well training both teachers and students. The late adopters mentioned improving computer skills for students by increasing computer time and ensuring that students become confident in its use as their future plans. These two groups confirmed the findings discussed earlier that some principals can be regarded as transformational leaders.

Another important finding is that those principals who did not adopt computers because of phobia and varying degrees of discomfort around computers, and lack of perceived ease of use of the computer, could not have a say in the future of computers in their schools. This finding is very important for strategic planning and policy implementation.

Data from the interview analysis complemented the survey analysis and shed light on the issues raised. The next chapter concludes the findings from the survey and interview analysis.

CHAPTER 6: CONCLUSIONS AND FUTURE RESEARCH

Diffusion and adoption theories study the society and its environment, as well as the individual, to assess the success of the diffusion or transfer of innovations. The IT transfer literature on Africa mostly covered the barriers to IT diffusion at the national level (Akpan, 2000; Alemna, 1999; Heeks, 2002; Jain & Mutula, 2001; Jimba, 2000; Odedra et al, 1993; Onyango, 2000; Thapisa & Birabwa, 1998).Transfer or diffusion appears to be the initial stage that comes before the adoption or acceptance of a new innovation. Technology has to be brought into a nation or transferred (diffused), and then individuals or users choose to adopt or accept the transferred technology. Therefore, it appears that the transfer of information technology happens at a national level, between governments involved in the transfer, and then the adoption follows next, when users and the organizations or institutions come in next to adopt and use the technology. IT transfer of diffusion at a national level emphasizes political, social, and economic issues.

The government of Botswana took the first step in the transfer of computer technology, by equipping all secondary schools with computers (The Revised National Policy on Education, 1994). This is the national level of the technology transfer, the next step in the diffusion or transfer concerns the user of the computer technology. The present study concentrated on the individual as a unit of analysis, using behavioral intention theories, specifically the Technology Acceptance Model (Davis, 1986), because the diffusion and adoption of technology cannot be regarded as complete without the input of the individual for whom the system was put in place. Barnett (1953), writing on of the diffusion of new innovations says, “A study of cultural change takes us beyond the appearance of a new idea into considerations of its acceptance and rejection” (p. 291). Barnett also says that an innovation diffusion takes place “on a mental plane” (p. 181) and this underlines the importance of the perceptions of the individual, who is accepting or adopting a new technology. One of the objectives of the Vision 2016 document is “All Batswana {Batswana refers to the people of Botswana} to use and apply the potential of computer equipment in their daily life” (p. 35).Therefore it was important to find out

100

what individual principals think about adopting and using computer technology in this study.

TAM Survey Results

The purpose of TAM is to predict and explain user acceptance of a new technology. Therefore three findings in the survey analysis can be attributed to the Technology Acceptance Model.

Firstly, the TAM survey results were instrumental in predicting computer usage in this study. The school principals who perceived computers to be easy to use and useful in their job wanted to adopt and use them. As Venkatesh and Davis (1996, p. 20) put it, TAM posits that the individual’s intention to use is the “single best predictor of actual system usage.” This finding is in line with previous TAM findings discussed in this study. It was clear that perceived ease of use and perceived usefulness contribute to computer acceptance. However, some principals reported that they did not want to adopt computers. What was not clear is why some principals chose to adopt and use computers and others did not want to. TAM findings did not answer that, and interviews filled this gap.

Secondly, the TAM survey was useful for showing the outliers in the research group. Identification of outliers led the researcher to investigating them further, because the outliers showed that some principals want neither to accept computers nor to use computers. Hu, et al. (2003), writing on IT adoption failure in schools, say: “The role of information technology in modern education has increased … but resistance to technology remains considerably high” (p.228). Therefore, it can be established that indeed there was a problem regarding individuals who resisted adopting computer technology in this study. The number of outliers in this research indicated that there was some resistance to computers. This problem was also identified in the literature review in this study. Davis et al. (1989), writing in confirmation of the problem, said, “Unfortunately, resistance to end-user systems by managers and professionals is a

101

widespread problem” (p. 982). The gap created by this finding pertained to describing the different groups within the research population. The interviews filled in this gap through the classification of the participants.

Lastly, the TAM survey revealed that there were three groups within the research population. There were adopters, non adopters and those who were neutral participants. The latter were neither adopting nor rejecting computers; they chose to be neutral in their survey response. This was one gap the survey left unfilled and the interviews probed this issue further, in order to investigate which category the neutral respondents fall into.

There were a number of gaps in the TAM because the knowledge that principals intended to adopt or reject computers needed to be complemented with the reasons why they chose to adopt or reject. Therefore interviews were conducted to fill the gaps. The interviews addressed why principals wanted to adopt or not adopt computers and how they intended to enhance computer use in the school. As Davis et al. (1989, p. 982) described, “To better predict, explain, and increase user acceptance, we need to better understand why people accept or reject computers.”

Interview Results

It was established in the TAM analysis that some principals wanted to adopt computers while others did not. The interview analysis established that one reason that some principals did not want to adopt was a phobia for computers. One participant said, “You see the phobia is not about computers only, it’s about all the other new gadgets as well. So personally I have that phobia.” This reflects a problem established in the literature on computer adoption. Orr (1993), attested to this by saying, “the feeling of anxiety toward computers and computer use is common, affecting 30 to 40% of the population” (¶. 1). This problem is well documented in the literature dealing with the rejection of technology. Odedra et al, (1993), call this problem ‘dysfunctional behavior’, a situation where computers are not used effectively, adoption rates remain slow, yet governments

102

support implementation through the purchase of relevant computer hardware and software. Flanagan and Jacobsen (2003) writing on the same issue argue that:

Unfortunately, technology planning has too often been limited to the goal of acquiring hardware and software. Schools have focused on purchasing equipment, setting up labs and wiring their buildings, without considering the substantial organizational and cultural changes that are necessary to support appropriate use of technology to enhance student learning. (p. 127)

Another finding from the interviews is that principals fall into categories or groups; they have different adoption rates and different uses as well as plans for the computer. Three such groups identified were the early adopters, characterized by advanced information processing for decision making, research on the Internet, and plans for training staff as well as Internet connection to the library. By contrast, the late adopters’ interactions with computers were limited to typing, storage and retrieval of administrative records. The non-adopters indicated that the secretary does the job for them, so it can be assumed that they understand that they are non-adopters.

The assumption behind choosing the principals as the research population was that principals are transformational leaders and therefore are expected to lead in the adoption of computers. The results above have confirmed that early adopters show the characteristics of transformational leadership while late adopters and non- adopters have not caught up with the changes yet. Telem (1996), describing transformational leaders, says, MIS implementation research findings indicate that top management's, i.e. the principal's, support is associated with MIS implementation success. S/he should be involved in SMIS's initiation, implementation and day-to-day operation. Under his/her leadership, the school's computerization policy should be formulated and his/her and other school staff's information requirements should be clearly specified. (p. 89)

103

Based on the above description, it can be concluded that the principals in this study who described themselves as “very comfortable” are also transformational leaders. What follows next is a list of the findings in the interview section.

List of Findings

Based on the self descriptions of the principals,

1. There are those who are very comfortable, comfortable, and uncomfortable as well as those who have a phobia for computers within the research population. 2. The principals introduced the words, phobia, fear, afraid and intimidation in response to a question asking about the comfort around computers. These words explain the comfort zone of the participants who had a computer anxiety. 3. Six out of ten participants did not think that it was easy to use or learn how to use the computer. The reasons given for this perception were phobia, lack of practice, lack of time, and lack of training or skills. 4. The comfortable and uncomfortable groups reported using the computer for typing confidential letters and record keeping, while the very comfortable users went beyond storage and retrieval of administrative records to mention using the Internet for research, connecting the Internet to the school library, networking computers, training all users, communication, database processing as well as using the computer for decision making. 5. The future plans of the comfortable and uncomfortable users were restricted to getting more computers for the school and increasing the students’ skills, while the very comfortable users mentioned downloading pertinent information for the students and teachers from the Internet, training teachers and students on computer use, improving access by networking computers, connecting the Internet in the school library as well as creating and processing information.

104

Conclusion

To conclude, the results of the survey and interview answered the question asked at the beginning of the study. The question posed was “Given that information technology, more specifically computer infrastructure, is in place in Botswana’s secondary schools, do the principals of these schools intend to support adoption and use by themselves, the teachers, and the students?”The TAM survey results confirmed that principals in Botswana secondary schools, who perceived that computers are useful and easy to use, are more likely to adopt them, at an R square of 27%. The survey had also revealed that for each of the independent variables there were outliers, which was an indication of different types of groups in a research population. The data collected through interviews complemented the data collected through the survey and helped to fill gaps in understanding that the survey analysis left ambiguous or unfilled. Three categories of users were established: early adopters, late adopters, and non-adopters. Responses to the interview questions allowed the researcher to develop and refine categories identifying different groups within the research population. Early adopters distinguished themselves from late adopters by their more varied usage of computers and their future plans for the schools. The early adopters seemed to fit into the category of transformational leaders, as discussed in this study.

In conclusion, the barriers to IT transfer in Africa mainly dealt with the national issues pertaining to social, political and economic barriers in the diffusion or transfer of information technology technology (Akpan, 2000, Jensen, 2002; Oladele, 2001; Onyango, 2000). It was established that Africa is both technologically and economically least developed and that this has led to the slow transfer and adoption of information technology (Onyango, 2000; Udo & Edoho, 2000). Some of the socio-economic barriers have to do with Africa’s underdevelopment, civil wars, corruption in the government and poverty (Jensen, 2002). The political barriers dealt with policy implementation failure, and the inadequacies of national information policies Chowdhury (1998).

105

Botswana’s information technology diffusion scene is different from that of most of Africa because the government has installed computers in all the secondary schools (Revised National Policy on Education, 1994). Therefore, the problems or the barriers to information technology adoption found in this research mainly deal with computer anxiety, and lack of perceived ease of use of the computer among the principals. Computer anxiety has been discovered to vary from one individual to another, based on the self descriptions using words like “phobia”, “intimidation”, “fear”, and “afraid.” For perceived ease of use, learning how to use the computer was found to cause problems for some of the principals. Four factors were reported by the participants. These pertain to lack of practice, time constraints, lack of skills, and lack of training. Based on the above discussion, the information technology adoption barriers among school principals in Botswana qualify and add to the barriers associated with diffusion problems in Africa. Scholars identified low adoption of computers in Africa (Peterson, 1998) and the present study identified the causes or factors contributing to the low adoption. Also the literature in the field identified computer anxiety as a cause for information technology rejection; this study confirmed and qualified these reasons by identifying the different degrees of anxiety as well as the different groups within the research population. Lastly, another reason given in the literature was “dysfunctional behavior” (Odedra et al,2000), where governments have installed computers and yet they are not used. This study confirmed that some principals are not adopters; they used the services of the secretary for all their computer needs. These findings have implications for policy implementation.

Implications of the study

This study has shown that it is not enough to install computers in schools without considering the implications for the users. Users may choose to accept or reject any technology. Therefore strategic planning and implementation strategies should be part of the computer implementation process. Fredland (2000), writing in support of this issue, says that the reactions and desires of the intended recipients of a technology should be considered when an innovation is introduced.

106

Phobia, lack of skills and the lack of practice with computers were identified as barriers to adoption in this study. Therefore training on the use of computers should include strategies to alleviate these barriers. Flanagan & Jacobsen (2003), writing in support of such strategies, say that there are four themes that can be identified as barriers to technology integration. These are “pedagogical issues, concerns about equity, inadequate professional development, and lack of informed leadership” (p. 125).

This study has implications for the Vision 2016. Botswana has already made a significant investment in information technology and information technology infrastructure for its secondary schools (The Revised National Policy on Education, 1994). In the information era, leadership by the principals is crucial to the spread and use of computer technology in schools. The principals’ technology adoption situation is important for policy implementation. In their discussion of leadership in ICT implementation in schools, Yuen et al. (2003) argued that ICT implementation involves stakeholders and issues of leadership. Therefore, the adoption of computers is necessary for development and the overall improvement of the educational system.

Limitations of the Study

The purpose of this study was to predict the likelihood of technology acceptance by school principals who are an important group of opinion leaders in education. However, since this research targeted only secondary school principals, further research involving teachers and students in secondary as well as primary schools would be required to fully characterize the current computer adoption scene in schools. Furthermore, while this study assumed that school principals are opinion leaders in their communities, it recognizes that they do not represent the full range of such leaders in any given community. Therefore, to fully characterize the current computer adoption scenario in communities in Botswana, further research involving a random sample of opinion leaders would be required. While a holistic picture of the current computer acceptance scene in these segments of Botswana culture might be desirable, however, there was no reason to assume that such a picture would be a better predictor of general future acceptance than

107

that established through surveying and interviewing the nation’s primary educational leaders.

Future Research

Several opportunities of further research are possible in the future. This research has established the computer adoption situation amongst principals in secondary schools. Future research might investigate the adoption of computers by teachers and students, so as to get the whole picture of adoption in schools.

Also, the present study focused on the individual and his or her intention to adopt a system. It might be interesting to shift the focus from the individual to the society by investigating factors like the innovation, innovation communication channels, time and the social system, and how these interact to facilitate or impede the diffusion of a new technology in a chosen group.

Furthermore, another possible topic of research is investigating the kind of computer training or skills imparted to the teachers and the principals in the school, in order to find the link between pedagogy, transformational leadership and professional development for both the teachers and the principals.

Lastly, only two principals mentioned the school library and connection to the Internet. Another study might investigate the connection between computer use, the mission of the school library and life-long learning. The school library plays an important role in education.

108

APPENDIX A: DESCRIPTIVE DATA

Table 12: Computer Anxiety Statements 1 Mean Skewness Standard deviation Nervousness 1.54 2.414 1.197 Discomfort 1.59 2.416 1.358 Feeling threatened 1.78 1.823 1.522 Feeling at ease 5.37 -1.012 1.968 Sinking feeling 1.74 2.008 1.373 Comfortable 5.46 -1.148 1.944 Feeling uneasy 2.11 1.450 1.734

Table 13: Perceived Usefulness (PU) 1 Mean Skewness Standard deviation Improves work quality 6.05 -1.927 1.716 Control over work 5.84 -1.472 1.558 Accomplish tasks 5.94 -1.689 1.574 Supports job 5.73 -1.311 1.555 Increases productivity 5.95 -1.700 1.536 Increases performance 5.85 -1.573 1.515

Table 14: Perceived ease of use statement 1 Mean Skewness Standard deviation Job easier 5.94 -1.573 1.479 Cumbersome 2.38 1.072 1.874 Learning easy 4.97 -.644 1.866 Frustrating 2.16 1.369 1.708 Rigid and inflexible 2.17 1.436 1.778 Remember 4.96 -.803 1.878 Mental effort 4.05 -.157 2.205 Understandable 4.96 -.732 1.758 Effort for skill 5.00 -1.089 1.958 Overall 6.38 -2.759 1.667 Table 15: Dependent variable 1 Mean Skewness Standard deviation Intention to adopt 6.38 -2.759 1.490

109

APPENDIX B: TEST OF NORMALITY

Table 16: Sharpiro-Wilk Test of Normality 1

Tests of Normality

Shapiro-Wilk Statistic df Sig. Nervousness ANX .524 111 .000 Discomfort ANX .507 111 .000 Feeling threatened ANX .577 111 .000 Feeling at ease ANX .791 111 .000 Sinking feeling ANX .610 111 .000 Comfortable ANX .770 111 .000 Feeling uneasy ANX .687 111 .000 Improving work qualityPU .623 111 .000 Control over work PU .757 111 .000 Accomplish tasks PU .709 111 .000 Supports job PU .782 111 .000 Increases productivity PU .715 111 .000 Increases performance PU .755 111 .000

easier to do job PEOU .740 111 .000 Cumbersome to use PEOU .744 111 .000

Learning easy PEOU .879 111 .000 Interaction frustrating PEOU .717 111 .000 Rigid and inflexible PEOU .701 111 .000 Easy to remember PEOU .862 111 .000 Requires mental effort PEOU .877 111 .000 Understandable PEOU .890 111 .000 Effort to be skillful PEOU .863 111 .000 Overall perceived ease of use .831 111 .000 Intention to adopt computers .474 111 .000 a Lilliefors Significance Correction

110

APPENDIX C: TAM STATEMENTS

Table 17: PU and PEOU Item Statistics 1

Pearson correlation Independent (significant at variable Statement Mean Std. Deviation 0.01 Perceived Accomplish Usefulness tasks PU 5.98 1.590 .363**

Supports job PU 5.82 1.466 .334**

Increases performance PU 5.78 1.593 .313**

Increases productivity PU 5.95 1.560 .354**

Perceived ease Understandable of use PEOU 5.06 1.718 .054

easier to do job PEOU 5.91 1.558 373**

Learning easy PEOU 5.03 1.854 .120

Easy to remember PEOU 5.05 1.850 297**

111

APPENDIX D: REMOVED OUTLIERS

Table 18: Casewise Diagonistics 1

Case Intention to adopt Number Std. Residual computers Predicted Value Residual 12 -3.628 1 6.03 -5.035

49 -4.152 1 6.76 -5.761

77 -4.152 1 6.76 -5.761

85 -2.646 2 5.67 -3.672

87 -2.582 1 4.58 -3.583

90 -2.646 2 5.67 -3.672

101 -3.105 1 5.31 -4.309

Dependent Variable: Intention to adopt computers

112

APPENDIX E: TOLERANCE AND VIF RESULTS

Table 19: Multicollinearity results 1

Model Sig. Collinearity Statistics

Tolerance VIF

(Constant)

Understandable PEOU .560 1.785

Easy to remember PEOU .623 1.605

Learning easy PEOU .663 1.509 easier to do job PEOU .156 6.396

Accomplish tasks PU .257 3.895

Supports job PU .332 3.008

Increases productivity PU .150 6.671

Increases performance PU .160 6.234

Dependent Variable: Intention to adopt computers

113

APPENDIX F :SCATTERPLOT

Scatterplot

Dependent Variable: Intention to adopt computers

-1 Value

-2

-3 Regression Standardized Predicted -4

-6 -4 -2 0 2 Regression Standardized Residual

114

Normal P-P Plot of Regression Standardized Residual

Dependent Variable: Intention to adopt computers

1.0

0.8

0.6

0.4 Expected Cum Prob Cum Expected

0.2

0.0 0.0 0.2 0.4 0.6 0.8 1.0 Observed Cum Prob

Figure 10: Scatter Plot

Figure 11: Normal P-P Plot

115

APPENDIX G: REGRESSION RESULTS

Table 20: Regression Model Summary 1

Adjusted R Std. Error of

Model R R Square Square the Estimate

1 .522(a) .273 .212 .737 a Predictors: (Constant), Increases performance PU, Understandable PEOU, Learning easy PEOU, Easy to remember PEOU, Supports job PU, Accomplish tasks PU, easier to do job PEOU, Increases productivity PU b Dependent Variable: Intention to adopt computers

116

APPENDIX H: QUESTIONNAIRE

Name of the School (OPTIONAL)------

Please circle your choice for the following:

Gender: 1. Male 2. Female

Type of school: 1. C.J.S.S. 2. S.S.S.

Please place a tick on the right box

Age 9 Education 9 20-30 Diploma 31-40 Degree 41-50 Masters 51 and above PhD.

Please rate the following statements using the key below.

Key:

1 = strongly disagree 2 = moderately disagree 3 = slightly disagree 4 = neutral (neither agree or disagree) 5 = slightly agree 6 = moderately agree 7 = strongly agree

Please place a tick in the right box.

117

1 2 3 4 5 6 7 Working with a computer makes me nervous, so I avoid using it if I can Computers make me uncomfortable. I feel threatened when other people talk about computers I feel at ease in a computer training workshop I get a sinking feeling when I think of trying to use a computer I feel comfortable working with computers Computers make me feel uneasy Using a computer improves the quality of what I do. Using a computer gives me greater control over my work. The computer enables me to accomplish tasks more quickly The computer supports critical aspects of my job Using a computer increases my productivity Using a computer increases my job performance Using a computer makes it easier to do my job I find a computer cumbersome to use Learning to use a computer is easy for me Interacting with a computer is frustrating The computer is rigid and inflexible to interact with It is easy for me to remember how to perform tasks using the computer Interacting with the computer requires a lot of mental effort My interaction with the computer is clear and understandable I find it takes a lot of effort to become skilful at using the computer Overall I find the computer easy to use. Given that I have access to a computer, I intend to use it

Thank you very much for completing the survey.

118

APPENDIX I: CONSENT LETTER

November 3, 2005

Dear School Principal,

I am a citizen of Botswana currently working on a PhD. degree in the College of Information at Florida State University. I am interested in exploring the information technology adoption phenomena in Botswana secondary schools. No studies have investigated how the perceptions of the users of computers impact the adoption or acceptance of computer technology in Botswana and how that affects the achievement of the Long Term Vision 2016, which aims at making Botswana an information society. Knowing the perceptions of computer users towards adopting computer technology can facilitate information policy implementation. Your responses to the attached survey are very important because the results can be used to inform Vision 2016 objectives at the school and national levels.

Your participation involves completing the attached survey, which has two parts, demographic and technology acceptance questions. A follow up interview shall be administered to a few selected participants. You can expect to spend approximately ten minutes or less completing the one page survey. Completing the survey might be very easy to do because you simply place a tick in the relevant box. Once you have completed the survey, place it in the attached envelope and post it using the self addressed envelope enclosed. To maintain your anonymity, your name will not appear on the survey and your responses will not be shared with anyone else. These procedures are meant to protect the confidentiality of your responses. There are no known risks involved with participation.

Your participation in this study is voluntary. If you choose not to participate or to withdraw from the study at any time, there will be no penalty. It will not affect your standing within your school and the Ministry of Education in Botswana. The results of the research study may be published, but your name will not be used.

If you have any questions concerning the research study, please call me at 267-355 2470 or e-mail me at [email protected]. The results of this survey will form part of the discussions in a dissertation to be presented in partial fulfillment of the PhD. Program in the College of Information at Florida State University. Your return of completed surveys will be considered your consent to participate. Thank you in advance!

Sincerely,

Angelina Totolo, (Ph.D. candidate)

119

APPENDIX J: INTERVIEW CONSENT LETTER

November 3, 2006

Dear School Principal,

Your participation involves responding to the attached interview questions on technology acceptance. This is a follow up to the survey questionnaire to which you responded. You can expect to spend approximately 30 minutes or less to complete the interview.

To maintain your anonymity, your name will not appear on the interview discussion and analysis, and instead false names will be assigned to protect your identity. Also, your responses will not be shared with anyone else. The responses are audio taped and they will be used by the researcher alone. The tapes will be destroyed as soon as the dissertation is successfully defended and submitted for binding. These procedures are meant to protect participants to the extent allowed by law regarding confidentiality. There are no known risks involved with participation.

If you have any questions concerning the research study, please call me at 267-355 2470 or e-mail me at [email protected]. The results of this survey will form part of the discussions in a dissertation to be presented in partial fulfillment of the PhD. Program in the College of Information at Florida State University. Your return of a signed letter will be considered your consent to participate. Thank you in advance!

Sincerely,

Angelina Totolo, (Ph.D. candidate)

120

APPENDIX K: HUMAN SUBJECTS APPROVAL MEMORANDUM

121

APPENDIX L: TIMELINE

First 2 weeks: send questionnaires by surface mail, and make follow ups by fax, and phone.

Next 2 weeks: make follow ups by sending another questionnaire in the surface mail and by fax, and phone.

Next 2 weeks: Start visiting principals’ workshops or meetings to ask them to fill in the questionnaire.

Next 12weeks: Data entry from questionnaires and analysis

Next 12 weeks: Conduct interviews, transcribe and analyze data

Last 12 weeks: Analysis and writing

122

APPENDIX M: INTERVIEW LEAD QUESTIONS

• Good morning to you. Thank you for giving me an opportunity to interview you.

• This is a follow up to the questionnaire I circulated from which I got varied answers from the school heads. There were three basic things I wanted to find out from the school heads. The first one is comfort zone around the computer, how comfortable they feel in fact. The other one was whether it is useful to them or not. The third one was whether they find it easy to use. This is a follow-up to find out how you feel about computers and what your comfort zone is, how comfortable do you feel around computers?

• Do you find that the computer is useful to you as a school head?

• What kind of uses do you have for it?

• Apart from storing and retrieving data what other uses do you put the computer to?

• As an individual school head, is it easy for you to use? Some school heads had mixed answers on this question. Some said it is easy to use and others said it isn't. What is your take on this one?

• About your computer knowledge, I see that you are very comfortable with the computer. How did you learn the computer skills and how did you reach this comfort zone?

• What recommendations do you have for a better understanding of computers for school heads? Some headmasters had indicated that they had not mastered them yet.

• What are you future plans for the spreading the use of the computer in the school and for educational purposes? Do you have some plans right now or you are still thinking about some plans to put in place?

• It has been very good talking to you. Is there anything else you would like to add before we conclude?

• Thank you very much for your time and the information you gave me today. Have a good day.

123

REFERENCES

Anandarajan, M., Igbaria, M., & Anakwe, U.P. (2002). IT acceptance in less-developed country: a motivational factor perspective. International Journal of Information management, 22(1), 47-65. Ajzen, I. (1991). The theory of planned behavior. Organizational Behaviour & Human Decision Processes, 50, 179-211. Ajzen, I. (2005). Theory of planned behavior. Retrieved June 11, 2005, from http://www.valuebasedmanagement.net/methods_ajzen_theory_planned_behaviour.ht ml Akpan, P. (2000). Africa in the age of a global network society. African Studies Quarterly, 4(2). Retrieved November 20, 2005 from http://web.africa.ufl.edu/asq/v4/v4i2a1.htm Alemna, A.A. (1999). The impact of new information technology in Africa. Information Development, 15(3), 167-170. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, N.J: Prentice-Hall. Babbie, E. (2001). The practice of social research. Australia: Wadsworth. Barnett, H.G. (1953). Innovation: The basics of cultural change. New York: McGraw-Hill. Baskerville, R., & Pries-Heje, J. (2001). A multiple-theory analysis of a diffusion of information technology case. Information Systems Journal, 11(3), 181-212. Beaulier, S.A. (2003). Explaining Botswana’s success: The role of post-colonial policy. Cato Journal, 23(2), 227-240. Berman, B.J. & Tettey, W.J. (2001). African states, bureaucratic culture and computer fixes. Public Administration and Development, 21, 1-13. Bill,J.M. (1977). Effects of varying structure and methods on the validity of self-report measures. Research Intelligence,3 (1), 19-21. Bose, K. (2004). Computer Training Programme for primary school teachers in teacher training institutions of the southern region of Botswana. Research in Post Compulsory Education, 9 (3), 401-416. Botswana Economy, facts and figures. (n.d.). Retrieved April 4, 2006, from http://www.gov.bw/government/index.html Brewer, J.K. & Workman, D.R. (2003). Introductory Statistics for Researchers (7th ed.). Boston, MA: Pearson Custom Publishing. Bryman, A. (1984). The debate about quantitative and qualitative research: A question of method or epistemology? The British Journal of Sociology, 35 (1), 75-92. Chowdhury, G.G. (1998). The Changing face of Africa’s information and communication scenario. International Information and Library Review, 30, 1-21.

124

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale: Lawrence Erlbaum Associates. Colvin, C. A., & Goh, A. (2005). Validation of the technology acceptance model for police. Journal of Criminal Justice, 33 (1), 89-95. Comin, D., & Hobjin, B. (2004). Cross-country technology adoption: Making the theories face the facts. Journal of Monetary Economics, 51 (1), 39-83. Compeau, D.R., & Higgins, C.A. (1995). Computer self-efficacy: Development of a measure and initial test. Management Information Systems Quarterly, 19(2), 189-211. Cook, T.D., & Campbell, D.T. (1979). Quasi-Experimentation: Design and analysis issues for field settings. Boston: Houghton Mifflin Company. Cresswell, J.W. (2002). Educational research: Planning, conducting, and evaluating quantitative and qualitative research. London: Merrill Prentice-Hall. Dasgupta, S., Granger, M., & McGarry, N. (2002). User acceptance of e-collaboration technology: An extension of the Technology Acceptance Model. Group Decision and Negotiation, 11, 87-100. Davis, F.D. (1986). A technology Acceptance Model for empirically testing new end-user information systems: Theory and results (Doctoral dissertation, Sloan School of Management, Massachusetts Institute of Technology, 1986). Davis, F.D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13 (3), 319-339. Davis, F.D.,Bagozzi, R.P., & Warshaw, P.R. (1989). User acceptance of computer technology: A comparison of two theoretical models. Management Science, 35(8),982-1003. Davis, F.D., & Venkatesh, V. (1996). A critical assessment of potential measurement biases in the technology acceptance model: three experiments. International Journal of Human Computer Studies, 45, 19-45. Department of Curriculum Development and Evaluation. (1988). Draft Module 9: Kagisano and Botswana’s Four National Principles. Gaborone: Author. Dillman, D.A. (1978). Mail and telephone surveys. New York: John Wiley & Sons. Dishaw, M.T., & Strong, D.M. (1999). Extending the technology acceptance model with the task-technology fit constructs. Information and Management, 36 (1), 9-21. Doyle, M. E., & Smith, M. K. (2001). Classical leadership. Retrieved November 19, 2005, from http://www.infed.org/leadership/traditional_leadership.htm Dresang, E.T. (1990).Interviewing using micro-moments and backward chaining. In Evaluation strategies and techniques for public library children’s services: Sourcebook. Jane Robbins, Holly Willet, M.J. Wiseman and Doug Zweizig. Madison, WI: University of Wisconsin –Madison. School of Library Studies. Encyclopedia Britannica Online (2006). Retrieved May 6, 2006, from http://www.britannica.com/search?query=computer+technology&submit=Find&sour ce=MWTEXT

125

Festinger, L. (1957). A Theory of Cognitive Dissonance. Stanford, CA: Stanford University Press. Flanagan, L., & Jacobsen, M. (2003). Technology leadership for the twenty-first century principal. Journal of Educational Administration, 41 (2). 124-142. Fredland, R.A. (2000). Technology transfer to the public sector in developing states. The Journal of Technology Transfer, 25 (3), 265-275. Friedman, A.L. (1994).The information technology field: Using fields and paradigms for analyzing technological change. Human Relations 47(4), 367-393. Gefen, D., & Straub, D.W. (1997). Gender Differences in the perception and use of e-mail: An extension to the technology acceptance model. MIS Quarterly, 21 (4), 389-400. Grand, Balulwami. (1997). An Investigation of the Information Seeking Behavior of Nurses in Botswana. Digital Dissertations. (UMI No. AAT 9736406) Hackbarth, G., Grover, V., & Yi, M.Y. (2003). Computer playfulness and anxiety: Positive and negative mediators of the system experience effect on perceived ease of use. Information and Management, 40 (3), 221-232. Heeks, R. (2002). Information Systems and Developing Countries: Failure, Success, and Local Improvisations. The Information Society, 18(2), 101-112. Hendrickson, A.R; Massey, P.D., & Cronan, T.P. (1993). On the test-retest reliability of Perceived usefulness and Perceived ease of use scales. MIS Quarterly, 17 (2), 227- 230. Hu, P.J., Clark, T.H.K., & Ma, W.W. (2003). Examining technology acceptance by school teachers: a longitudinal study. Information and Management, 41 (2), 227-241. Hudson, H.E. (2000). From African village to global village: Lessons in bridging the African digital divide. Retrieved April 11,2005, from http://www.tprc.org/abstracts00/africanvillage.pdf Igbaria, M., & Nachman, S.A. (1990). Correlates of user satisfaction with end user computing. Information & Management, 19(2), 73-82. Jain, P., & Mutula, S.M. (2001). Diffusing information technology in Botswana: A framework for Vision 2016. Information Development, 17 (4), 234-241. Jensen, M. (2002). Information and communication technologies (ICTs) in Africa-A status report. UN ICT Task Force, “Bridging the Digital Divide in the 21st Century”. Third Task Force Meeting, United Nations headquarters. 30 September-1 October 2002. Jimba, S.W. (2000). Information technology and the dialectics of poverty in Africa. New Library World, 101(6), 253-262. Kaplan, B., & Duchon, D. (1988). Combining qualitative and quantitative methods in information systems research: A case study. MIS Quarterly, 12 (4), 571-586. Karahanna, E., & Straub, D.W. (1998). The psychological origins of perceived usefulness and ease-of-use. Information and Management, 35, 237-250. Kebede, G. (2004). The information needs of end users of Sub Saharan Africa in the digital information environment. The International Information and Library Review, 36(3), 273-279.

126

Kedia, B.L., & Bhagat, R.S. (1988). Cultural constraints on transfer of technology across nations: Implications for research in international and comparative management. Academy of Management Review, 13 (4), 551-571. Klauss, R. (2000). Technology transfer in education – application to developing countries. The Journal of Technology Transfer, 25, 277-287. Korac-Kakabadse, N., Kouzmin, A., & Korac-Kakabadse, A. (2000). Information technology and development: Creating “IT harems”, fostering new colonialism or solving “wicked” policy problems? Public Administration and Development, 20(3), 171-184. Kukafka, R., Johnson, S.B., Lifante, A., & Allgrante, J.P. (2003). Grounding a new information technology implementation framework in behavioral science: A systematic analysis of the literature on IT use. Journal of Biomedical Informatics, 36 (3), 218-227. Legris, P., Ingham, J., & Collerette, P. (2003). Why do people use information technology? A critical review of the technology acceptance model. Information and Management, 40 (3), 191-204. Lekorwe, M., Molomo, M., Molefe, W., & Moseki, K. (2001). Public Attitudes towards democracy, governance, and economic development in Botswana. Afrobarometer working paper no.14. A comparative series of national public attitude survey on democracy, markets and civil society in Africa. Retrieved November 20, 2005, from http://www.afrobarometer.org/papers/Afropaper%20No14.pdf Leonard, L.N.K., Cronan, T.P., & Kreie, J. (2004). What influences IT ethical behavior intentions—planned behavior, reasoned action, perceived importance, or individual characteristics? Information and Management, 42 (1), 143-158. Long Term Vision 2016: Towards prosperity for all. (1997). Gaborone: Presidential Task Group. Luarn, P., & Lin, H. (2005). Toward an understanding of the behavioral intention to use mobile banking. Computers in Human behavior, 21(6), 873-891. May D. (1998). Geography of Botswana. Gaborone: Macmillan Botswana. McFarland, D.J., & Hamilton, D. (2006). Adding contextual specificity to the technology acceptance model. Computers in Human Behavior. 22(3), 427-447. Merriam-Webster Dictionary Online (2006). Retrieved May 1,2006, from http://www.m- w.com/dictionary/adopter Meso, P., Musa, P., & Mbarika, V. (2005). Towards a model of consumer use of mobile information and communication technology in LDCs: the case of sub-Saharan Africa. Information Systems Journal, 15(2), 119-146. Moahi, Kgomotso (2000). A study of the information behavior of health planners, administrators and mangers in Botswana, and implications for the design of a national health information system (NHIS). Digital Dissertations. (UMI No. AAT 9985000) Molomo, M.G., & Somolekae G. (2001). Sustainable electoral democracy in Botswana.Proceedings of An International IDEA-SADC Conference: Towards Sustainable Democratic Institutions in Southern Africa. International IDEA. 142–165.

127

Mooko, N.P. (2002). A study of the family information needs and information seeking behaviors of rural women in Botswana. Digital Dissertations. (UMI No. AAT 3078863) Moore, D. 2004. The Basic Practice of Statistics (3rd ed.). New York: W.H. Freeman and Company. Morse, J. (2003). Principles of mixed and multi-method research design. In A. Tashakkori. & C. Teddlie (Eds.), Handbook of mixed methods in social and behavioral research (pp.189-208). London: Sage Publications. Neuman,W.L. (2003). Social research methods: Qualitative and quantitative approaches. Boston: Pearson Education, Inc. Obijiofor, L. (1998). Africa’s dilemma in the transition to the new information and communication technologies. Elsevier Science, 30 (5), 453-462. Odedra, M., Lawrie, M., Bennett, M., & Goodman, S. (1993). Sub-Saharan Africa: A technological desert. Communications of the ACM, 36 (2), 25-29. Official Portal for North Dakota State Government (2006). Definition of information technology. Retrieved on April 25, from http://www.nd.gov/itd/planning/definition.html Oladele, B.A. (2001). The imperatives of challenges for Africa in the knowledge age: Status and the role of national information policy. Proceedings of the 67th IFLA Council and General Conference, August 16-25. Onyango, R.A.O. (2000). Global information and Africa: On the crest of a mirage? Library Management, 21(4), 197-204. Orr, L.V. (1993). Computer anxiety. Retrieved November 19, 2005, from http://www.usm.maine.edu/~com/lindap~1.htm Parker, E.B. (1974). Implications of new information technology. The Public Opinion Quarterly, 37 (4),.590-600. Parsons, N. (1999). Botswana History Pages: Web Encyclopedia-Botswana. Retrieved November 19, 2005, from http://www.eisa.org.za/WEP/bot2.htm Pedhazur, E.J. (1982).Multiple regression in behavioral research. London: Harcourt Brace College Publishers. Peterson, R.A. (1994). A meta-analysis of Cronbach’s Coefficient Alpha. The Journal of Consumer Research, 21 (2), 381-391. Peterson, S.B. (1991). From processing to analyzing: Intensifying the use of microcomputers in development bureaucracies. Public Administration and Development, 11 (5), 491- 510. Peterson, S.B. (1998). Saints, demons, wizards and systems: why information technology reforms fail or underperform in public bureaucracies in Africa. Public Administration and Development, 18, 37-60. Powell, P. (1992). Information technology evaluation: Is it different? Journal of the Operational Research Society, 43 (1), 29-42. Republic of Botswana Report of the National Commission on Education. (1993).

128

Gaborone: Government Printer. Republic of Botswana Government Paper No. 2. (1994). The Revised National Policy on Education. Gaborone: Government Printer. Republic of Botswana National Development Plan 9 (2003). National Development Plan 9 2003/04-2008/09. Gaborone: Government Printer. Rice, M. (2003). Information and communication technologies and the global digital divide: Technology transfer, development, and least developing countries. Comparative Technology Transfer and Society, 1(1), 72-87. Rodriguez, F., & Wilson, E., III. (2000). Are poor people losing the information revolution? (InfoDev Working Paper). Washington, DC: World Bank Group. Retrieved November 19, 2005, from http://www.cidcm.umd.edu/ICT/papers/are_poor_countries_losing.pdf Rogers, E.M. (1976). New product adoption and diffusion. The Journal of Consumer Research, 2( 4), 290-301. Rogers, E.M. (1995). Diffusion of innovations (4th ed.). New York: The Free Press. Rose, G.M. & Straub, D.W. (1998). Predicting general IT use: Applying TAM to the Arabic World. Journal of Global Information Management, 6 (3), 39-46. Russell, D.M. (2004). People and information technology in the supply chain; social and organizational influences on adoption. International Journal of Physical Distribution and Logistics Management, 34 (2), 102-122. Scales and Standard Measures (n.d.) Retrieved from http://www2.chass.ncsu.edu/garson/PA765/standard.htm Shavo, C., & Igbaria, M. (2003). Information technology adoption. Retrieved November 19, 2005, from http://www.dekker.com/sdek/abstract~db=enc?content=10.1081/E-ELIS- 120008816&refs=true#refs713531939 Silitshena, R.M.K., & McLeod, G. (1998). Botswana: A physical, social and economic geography (2nd ed.). Gaborone: Longman Botswana. Straub, D., Keil, M., & Brenner, W. (1997). Testing the technology acceptance model across cultures: a three country study. Information and Management, 33, 1-11. Surry, D.W., & Farquhar, J. D. (1997). Diffusion theory and instructional technology. Journal of Instructional Science and Technology 2(1). [On-line] Retrieved April 19, 2006, from http://www.usq.edu.au/electpub/e-jist/docs/old/vol2no1/article2.htm Szajna, B. (1996). Empirical evaluation of the revised Technology Acceptance Model. Management Science, 42 (1), 85-92. Tarde, G.,de (1903). The Laws of Immitation. New York: H. Holt and Company. Telem, M. (1996). MIS implementation in schools: A systems socio-technical framework. Computers and Education, 27 (2), 85-93. Thapisa, A.P.N., & Birabwa, E. (1998). Mapping Africa’s initiative at building an information and communications infrastructure. Electronic Networking Applications and Policy, 8 (1), 49-58.

129

Todd, R.J. (1999). Transformational leadership and transformational learning: Information literacy and the World Wide Web. NASSP Bulletin, 83(605), 4-12. Totolo, A. (2005). Information technology adoption in Botswana secondary schools and implications on leadership and school libraries in the digital age. In S. Lee, P. Warning, D. Singh, E. Howe, L. Farmer and S. Hughes (Eds.), IASL Reports 2005: Information leadership in a culture of change (p. 78). Erie, PA: International Association of School Librarianship. Udo G. J., & Edoho, F. M. (2000). Information technology transfer in African nations: An economic development mandate. The Journal of Technology Transfer, 25 (3), 329- 342. United Nations Economic Commission for Africa. (2001). African Information Society Initiative: An Action Framework to Build Africa’s Information and Communication Infrastructure. Retrieved November 19, 2005, from http://www.uneca.org/ United Nations in Botswana. (2005). Retrieved November 19, 2005, from http://www.unbotswana.org.bw/about_b.html Venkatesh, V. (2000). Determinants of perceived ease of use: Integrating control, intrinsic motivation, emotion into the technology acceptance model. Information Systems Research, 11 (4), 342-365. Venkatesh, V., & Davis, F.D. (2000). A theoretical extension of the TAM: Four longitudinal field studies. Management Science, 46(2), 186-204. Venkatesh, V., & Morris, M.G. (2000). Why don’t men ever stop to ask for directions? Gender, social influence, and their role in technology acceptance and usage behavior. MIS Quarterly, 24 (1), 115-39. Wilson, E.J., & Wong, K. (2003). African information revolution: a balance sheet. Telecommunications Policy, 27, 155-177. Yee, D.L. (1998). Chalk, chips and children. Educational leadership: journal of the Department of Supervision and Curriculum Development, N.E.A., 55(7), 57-60. Yuen, A.H.K., Law, N., & Wong, K.C. (2003). ICT implementation and school leadership; Case studies of ICT integration in teaching and learning. Journal of Educational Administration, 41(2), 158-170.

130

BIOGRAPHICAL SKETCH

The author holds a Masters in Library Science from Syracuse University, a Bachelor of Arts from the University of Botswana and a Certificate in School Librarianship from the University of Botswana. She is a lecturer at the University of Botswana in the Department of Library and Information Studies, on study leave to read for a PhD. in the College of Information at Florida State University. Her interests are in information technology adoption and how it can enhance the use of information in the solution of Africa’s problems. A papers related to the dissertation topic presented to a conference is; Totolo, A. (2005). Information technology adoption in Botswana secondary schools and implications on leadership and school libraries in the digital age. In S. Lee, P. Warning, D. Singh, E. Howe, L. Farmer and S. Hughes (Eds.), IASL Reports 2005: Information leadership in a culture of change (p. 78). Erie, PA: International Association of School Librarianship.

131