What's in Your Financial Package? Student Credit Card Use And

What’s in Your Financial Package? Student Credit Card Use and Economic Insecurity in College

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Benjamin D. Andrews

Graduate Program in Sociology

The Ohio State University

2017

Dissertation Committee:

Rachel Dwyer, Advisor

Claudia Buchmann

Vincent Roscigno

Copyrighted by

Benjamin D. Andrews

2017

Abstract

Since the turn of the 21st century, going to college has become increasingly financially difficult in the United States. Tuition prices continued to rise, state funding for higher education declined, and the mean family income declined or stagnated for all but the top

20 percent of families (Goldrick-Rab 2016). In a period where college has risen to be the preeminent way Americans can make a better life for themselves, it is becoming more difficult for Americans to pay for college. Financial aid does not cover as much of the price of college as it once did (Goldrick-Rab 2016), and college students are relying on financing methods like student loans more than ever before.

Student loans, however, are not the only credit-based financial strategy college students use to pay for college (Manning 2000, 2005). With the explosion of consumer credit access from the 1980s to the 2000s, college students are using credit cards, many times to bridge gaps in their budgets as they try to pay for college. College students use credit cards to pay for both indirect expenses (e.g., food and housing) and direct educational expenses (e.g., tuition and books) as they pursue a college degree (Nellie

Mae 2005; Sallie Mae 2009). While media, government, and scholarly attention has highlighted widespread college student credit card use (J. Manning 1991; R. Manning

2000; Norvilitis et al. 2006; Rubin 1998; United States General Accounting Office 2014), we know surprisingly little about differences in college student credit card ownership and use from a sociological perspective. ii

This dissertation uses data from the Education Longitudinal Study (2002-12) and the Study on Collegiate Financial Wellness (2017) to understand how students respond to increased financial responsibility in college by taking on credit card debt, and how using credit cards influence their educational trajectories. The first two papers of this dissertation examine questions related to who uses credit cards, in what ways, and in what situations. I examine college student credit card use through differences in who owns a credit card, who carries a balance on their credit card, and who pays for tuition with a credit card. The third paper focuses on linking college student credit card ownership and use to educational outcomes, specifically that of bachelor’s degree attainment.

Through the three empirical studies that make up this dissertation, I find that students from low socioeconomic backgrounds are more likely to practice risky credit card behavior like owning 2 or more credit cards and carrying a revolving credit balance from month to month, that students in financially precarious situations are more likely to pay for college tuition with a credit card, and that carrying a revolving credit balance decreases college students’ likelihood of attaining a bachelor’s degree.

Findings shed light on how social shifts in higher education funding, and how students respond to them, influence the educational attainment process. Namely, students without proper financial support during college resort to risky means of paying for college that ultimately influence their educational outcomes.

iii

Dedicated to my grandmother, Janice Andrews, who set a new educational norm for our family by going to college and earning a bachelor’s degree and who quietly admonished a younger me to practice piano 15 minutes a day, laying the foundation for the practice of daily writing that made this dissertation possible.

Acknowledgments

One night at the end of my second year in Ohio State’s Sociology PhD program I had a dream. I was standing in a restaurant, looking past the tables of people, out beyond the restaurant’s balcony, and into the distance. The restaurant sat on stilts in the middle of a large field of waist-high grass. There were trees that edged the field, trees from which I saw something exiting and moving toward the restaurant. The grass was parting as the thing approached. It was giant, but it moved fast, and the people in the restaurant seemed to understand two things: 1) that it was coming for them and 2) that it was bad. They left in one swift movement and—as happens in dreams—I stayed.

As the bad thing got closer I saw that it was a giant snake, at least a football field in length. A monster, really, especially for someone like me whose nightmares almost exclusively place me in a situation where there are snakes and no good way to get away from them. The snake slithered through the tall grass and disappeared underneath the stilted restaurant. As any reasonable person would do, I went out to the balcony to see if I could climb up the side of the building and distance myself from the thing. I grabbed onto the ledge of a door frame and lifted my body up, trying to shimmy up the wall and onto the roof, but apparently in the time it took me to walk onto the balcony, the snake had wrapped itself around the restaurant and sat coiled on the roof, its massive head perched above the door frame I was using as support. We stared at each other, its dead eyes giving

v me the impression that it was waiting for me to make a move. The dream ended and I woke up terrified.

In my fifth year in the PhD program, about two weeks before a full draft of my dissertation was due, I had another dream. I was in a kitchen this time. The uniform chrome color and industrial-sized appliances around me indicated that the kitchen was a commercial kitchen, probably in a restaurant. I was accompanied by five or six chefs, all wearing huge, white hats and being altogether jovial about something. They cheered and laughed as they collectively lifted a mammoth object onto the kitchen’s central chrome table. The object turned out to be a huge snake, one that looked just like the snake in my dream three years earlier. The way the snake flopped onto the table made me think it was dead, and when I saw the other chefs cutting into the snake to prepare it for dinner, I realized I was right. For the record, I’ve been having snake dreams since I was a child and this is the first ever dream that I remember where the snake wasn’t alive. Also for the record, I have never knowingly eaten a snake. Or prepared one for a customer in a restaurant.

As the chefs laughed and celebrated the soon-to-be feast, I was tasked with removing the snake’s head which was hanging limply off the table in front of me. Even though the snake was dead, I felt a shock of fear and hesitated to do my assigned task. I mustered up what little gumption I had, grabbed the snake’s head—much smaller now than what I remembered in my original dream—and used a knife to sever it from the body, only succeeding in removing part of the head due to my clumsy approach. The other chefs smiled and applauded anyway, and that was the end of that.

Those two dreams summarize my graduate school experience and dissertation writing process better than I could ever describe on my own. Much of graduate school has been a terrifying stalemate between me and the monster that is my graduate work.

And what successes I’ve experienced always seem to not be the result of my work ethic or ability, but rather the result of mentors and colleagues sharing their expertise and guidance with me in times when I needed help. Just like the chefs in the kitchen who killed the beast and let me take some of the credit even though my clumsy attempt at chipping in didn’t even come close to their mastery over the task, I consider this dissertation a culmination of many “chefs” who have helped and guided me through graduate school, who let me feel the excitement of being part of their group, and without whom I wouldn’t have the slightest idea how to finish a project like this.

To Rachel Dwyer, Claudia Buchmann, and Vinnie Roscigno: thank you for your generosity of time and willingness to share your expertise with me through good and bad seasons in my graduate career. To Anne McDaniel, D’Arcy Oaks, Krystyne Savarese, and my friends at the Center for the Study of Student Life: thank you for giving me the chance to contribute to an excellent work environment and for teaching me so many valuable skills that have laid the foundation for my future work in higher education. To

Cohort 2012: thank you for continually providing a supportive environment to ask questions, to collaborate, and to laugh.

To Erica Phillips and Rob VandenBerg: thank you for counseling me through difficult times in graduate school, for bunkering down with me in the trenches of statistics homework (Rob), and for giving me tremendous advice on how to write in a

vii professional manner and advocate for my education (Erica). To Laura DeMarco: thank you for starting our Dissertation Working Group. Had you not gathered us all together, I think I would still be staring at a blank page and wondering if I would ever finish graduate school. To Korrie Johnson: thank you for introducing me to the book How to

Write A Lot. That book changed my life. To Chris Munn: thanks for including me in your racquetball games, playing me left-handed just so it was a fair game, and being such a good friend, especially when I was so new to graduate school.

To David and Tiffani Hoover, Kale Booher, Caleb Craft, Justin and Nicole

Dickman, Logan Baker, Mary Wadell, and Chris Cannell: you all at one point or another provided me with the zest of life I needed to keep moving forward. You reminded me that life existed—lots of it—outside of graduate school. You filled my home with laughter and love. To Tim Andrews: thank you for sharing your life with me these last few years. I hope you never forget to call Matt Kuchers. To Dave and Emily Andrews: you’ve been an unbelievable support to me through life and I can’t even begin to detail all the ways you’ve given me opportunity when I didn’t deserve it and help when I needed it most. To Jan and Bid Andrews: thank you for setting the best example I’ve seen for what it looks like to age with grace and dignity, to not lose passion and humor in the later stages of life. Grammy, you are a continual inspiration to me in all areas of life, and

I hope I can one day influence my family even half as much as you have influenced our family.

To Dani: this dissertation, this degree, it is only possible because of your tireless support of me through the last five years. Words cannot explain the depth of my gratitude

viii for your presence in my life. I am proud to call you my wife and I cannot wait to see the adventures that unfold before us as we travel west. Your love, gentleness, and unfailing forgiveness inspires me to be a man worthy of the title “husband.”

Ray Bradbury once said that an important component of reading is finding an author who can take you through the dark. To all of the authors who have taken me through the dark times of graduate school: Stephen King, Jack London, Ray Bradbury,

Bram Stoker, Mary Shelley, Ernest Hemingway, Aldous Huxley, William Goldman, and many more. I consider you all fellow travelers through graduate school. You have introduced me to adventures I could have never imagined experiencing when I first started the PhD program. You have inspired me to write and create and live life in pursuit of adventure, and comedy, and even a touch of suspense.

To all of the chefs who have cooked up this dissertation through their support and knowledge and love, thank you. Bon appétit.

Vita

June 2008 ...... Rutherford B. Hayes High School

2012...... B.A. Sociology, Ohio Wesleyan University

2014...... M.A. Sociology, The Ohio State University

2013 to 2015 ...... Graduate Teaching Associate, Department

of Sociology, The Ohio State University

2015 to present ...... Research Analyst, Center for the Study of

Student Life, The Ohio State University

Publications

Baker, Amanda, Benjamin D. Andrews, and Anne McDaniel. Forthcoming. “The Impact of Student Loans on College Access, Completion, and Returns.” Sociology Compass.

Fields of Study

Major Field: Sociology

Table of Contents

Abstract ...... ii

Dedication ...... iv

Acknowledgements ...... v

Vita ...... x

Table of Contents ...... xi

List of Tables ...... xiii

List of Figures ...... xvi

Chapter 1: Introduction and Historical Overview ...... 1

Chapter 2: Patterns of College Student Credit Card Ownership and Use...... 19

Chapter 3: Paying for Tuition with a Credit Card...... 50

Chapter 4: How Student Credit Card Use Influences College Degree Attainment ...... 79

Chapter 5: Discussion and Conclusion ...... 110

References ...... 119

Appendix A: Materials for Chapter 2 ...... 129

Chapter 2 Variables...... 130

Chapter 2 Tables ...... 132

Chapter 2 Figures ...... 141

Appendix B: Materials for Chapter 3...... 144 xi

Chapter 3 Variables...... 145

Chapter 3 Tables ...... 149

Chapter 3 Figures ...... 155

Appendix C: Materials for Chapter 4...... 161

Chapter 4 Variables...... 162

Chapter 4 Tables ...... 164

Chapter 4 Figures ...... 172

Appendix D: Correlation Matrices...... 174

Correlation Matrices by Regression Table ...... 175

Appendix E: Study on Collegiate Financial Wellness ...... 186

Study on Collegiate Financial Wellness ...... 187

xii

List of Tables

Table A.1. Descriptive Statistics ...... 132

Table A.2. Demographic Differences in College Student Credit Card Ownership ...... 134

Table A.3. Demographic Differences in Owning Multiple Credit Cards ...... 135

Table A.4. Demographic Differences in Carrying a Revolving Credit Balance ...... 136

Table A.5. Logistic Regression Model for Whether Students Own a Credit Card in their

Name ...... 138

Table A.6. Logistic Regression Model for Whether Students Own Multiple Credit Cards in their Name...... 139

Table A.7. Logistic Regression Model for Whether Students Carry a Revolving Balance on their Credit Card ...... 140

Table B.1. Descriptive Statistics for ELS Variables ...... 149

Table B.2. Descriptive Statistics for SCFW Variables ...... 150

Table B.3. Logistic Regression Model for Whether Students Use a Credit Card to Pay for

Tuition Using ELS Data...... 151

Table B.4. Average Financial Literacy Scores by Whether Student Uses Credit Card to

Pay for Tuition ...... 152

Table B.5. Logistic Regression Models for Whether Students Use a Credit Card to Pay for Tuition Using SCFW Data ...... 153 xiii

Table B.6. Students’ Primary Reasons for Using Credit Card to Pay for Tuition ...... 154

Table C.1. Descriptive Statistics ...... 164

Table C.2. Logistic Regression Model for Bachelor’s Degree Attainment by 2012,

Testing Credit Card Ownership Variable ...... 166

Table C.3. Logistic Regression Model for Bachelor’s Degree Attainment by 2012,

Testing Number of Credit Cards Variable ...... 167

Table C.4. Logistic Regression Model for Bachelor’s Degree Attainment by 2012,

Testing Pays for Tuition with Credit Card Variable ...... 168

Table C.5. Logistic Regression Model for Bachelor’s Degree Attainment by 2012,

Testing Carries Revolving Credit Balance Variable ...... 169

Table C.6. Logistic Regression Models for Bachelor’s Degree Attainment by 2012 .... 170

Table C.7. Epanechnikov Kernel Propensity Score Matching Model for Bachelor’s

Degree Attainment by 2012 ...... 171

Table D.1. Correlation Matrix for Table A.5 ...... 175

Table D.2: Correlation Matrix for Table A.6 ...... 176

Table D.3. Correlation Matrix for Table A.7 ...... 177

Table D.4. Correlation Matrix for Table B.3 ...... 178

Table D.5. Correlation Matrix for Table B.5 ...... 179

Table D.6. Correlation Matrix for Table C.2 ...... 180

Table D.7. Correlation Matrix for Table C.3 ...... 181

Table D.8. Correlation Matrix for Table C.4 ...... 182

Table D.9. Correlation Matrix for Table C.5 ...... 183

xiv

Table D.10. Correlation Matrix for Table C.6 – Model 1 (All Respondents) ...... 184

Table D.11. Correlation Matrix for Table C.6 – Model 2 (Credit Card Holders) ...... 185

Table E.1: Sample Size Chart from cfw.osu.edu ...... 192

List of Figures

Figure A.1. Ownership of Multiple Credit Cards, by Socioeconomic Status ...... 141

Figure A.2. Whether Student Carries Revolving Credit Card Balance, by Socioeconomic

Status ...... 141

Figure A.3. Ownership of Multiple Credit Cards, by Institution ...... 142

Figure A.4. Whether Student Carries Revolving Credit Card Balance, by Institution ... 142

Figure A.5. Ownership of Multiple Credit Cards, by Hours Worked Weekly During the

2005-2006 School Year ...... 143

Figure A.6. Whether Student Carries Revolving Credit Card Balance, by Hours Worked

Weekly During the 2005-2006 School Year ...... 143

Figure B.1. Whether Student Pays for Tuition with a Credit Card, by Socioeconomic

Status ...... 155

Figure B.2. Whether Student Pays for Tuition with a Credit Card, by Hours Worked .. 155

Figure B.3. Whether Student Pays for Tuition with a Credit Card, by Institution ...... 156

Figure B.4. Whether Student Pays for Tuition with a Credit Card, by First Generation

College Student Status ...... 156

Figure B.5. Whether Student Pays for Tuition with a Credit Card, by Whether Student

Has Access to Parent Loans to Help Pay for College ...... 157

xvi

Figure B.6. Whether Student Pays for Tuition with a Credit Card, by Total Pell Grant

Amount ...... 157

Figure B.7. Whether Student Pays for Tuition with a Credit Card, by Race ...... 158

Figure B.8. Whether Student Pays for Tuition with a Credit Card, by Family Status .... 158

Figure B.9. Whether Student Pays for Tuition with a Credit Card, by Access to

Emergency Cash ...... 159

Figure B.10: Whether Student Pays for Tuition with a Credit Card, by Financial

Difficulty in School...... 159

Figure B.11: Whether Student Pays for Tuition with a Credit Card, by Whether Student

Worries About Having Enough Money to Pay for School ...... 160

Figure C.1. Credit Card Ownership and Behavior, by Socioeconomic Status ...... 172

Figure C.2. Credit Card Ownership and Behavior, by Hours Worked ...... 172

Figure C.3. Credit Card Ownership and Behavior, by Institution...... 173

Figure C.4. Bachelor’s Degree Attainment by 2012, by Credit Card Ownership and

Behavior ...... 173

xvii

CHAPTER 1: Introduction and Historical Overview

Introduction

In the context of the United States’ meritocratic system, education is often a necessary requirement for upward social mobility. Research within the status attainment tradition in sociology focuses on the critical role educational attainment plays in providing people with opportunities to be upwardly mobile. Beginning with the work of

Blau and Duncan (1967), status attainment research examines the link between an individual’s family background and their eventual occupational status, influenced by their educational attainment. Sewell et al. (1969) built on Blau and Duncan’s model by arguing that social-psychological factors matter in the process of social mobility as well. In both of these status attainment models, educational attainment is a key lynchpin for social mobility. What was so profound about Blau and Duncan’s (1967) research was the finding that though ascriptive characteristics (e.g., father’s occupational status, family socioeconomic status) matter for social mobility, they do not matter nearly as much as achieved characteristics (e.g., an individual’s educational attainment). Status attainment research has established that there is more social fluidity than originally thought (Beller

2009), and that educational attainment serves as a key component for social mobility

(Buchmann and Obinna 2014).

However, when access to education varies between social groups, the foundation of the meritocratic system begins to crumble. Upward social mobility opportunities in the

United States are limited in part due to disparate access to educational opportunities

(Alon and Tienda 2007). Many scholars note deep differences in school quality and resources, dependent on a child’s neighborhood context (Condron and Roscigno 2003;

Duncan and Murnane 2011, 2014). While debate persists in this area of research

(Downey, von Hippel, and Broh 2004; Downey, von Hippel, and Hughes 2008), scholars suggest that the primary and secondary educational system in America serves to promote already existing inequality (Bourdieu and Passeron 1990; Bowles and Gintis 2002).

However, some research suggests that postsecondary schooling may be an equalizer for disadvantaged groups in the United States (Brand and Xie 2010). That is, despite persistent inequality in primary and secondary schooling, if students from disadvantaged backgrounds gain access to and complete postsecondary schooling, they may have more equal footing for upward social mobility opportunities than they would have if they did not attend college. Indeed, the difference between the lifetime earnings of a high school graduate and a college graduate continue to widen as the returns to a college degree increase (McCall and Percheski 2010).

Given this scenario, making college accessible and a college degree attainable is increasingly important in order to equalize opportunity for upward social mobility in the

United States. However, increasing costs associated with postsecondary education makes going to college prohibitive for some groups. Many scholars have noted this inequality in access and have called for widespread funding so that anyone, no matter their social

2 background, can attend college (Goldrick-Rab and Kendall 2016). The first major example of this kind of funding is the government-funded GI Bill that gave funding for postsecondary education to World War II veterans. Since that time, opportunities to finance a college degree have proliferated, namely with the explosion of educational loans (Altbach, Gumport, and Berdahl 2011).

As college access increased over the course of the 20th century, a “college for all” ethos took hold in America. Nearly all students approaching high school graduation expect to attend college (Rosenbaum 2004). Because not all of these prospective college students come from backgrounds that make attending college financially plausible, and because of the preponderance of financialized methods for funding a college degree, taking on student debt has become a widespread part of the American college experience.

Research shows mixed evidence as to whether student debt helps aspiring college students over the long term (Dwyer, McCloud, and Hodson 2012; Houle and Berger

2015; Nau, Dwyer, and Hodson 2015), and much of the public discourse around this topic has decried the rising amount of student debt, calling it a “crisis” (Mitchell and

Jackson-Randall 2012). Regardless, credit-based financing has become a widespread part of financing postsecondary education, much like credit-based financing programs in housing and transportation.

While student loans are the most well-known credit-based option for financing a college degree, loans are not the only credit-based method students use to meet educational expenses. As access to higher education increased after World War II, so did access to consumer credit, specifically in the form of credit cards (Carruthers and

Ariovich 2010; Hyman 2012). These two developments have overlapped in important ways in the realm of higher education financing. As noted in other research, credit access has expanded as a substitute for expanding government support in the face of lower wages and using credit has become a necessary part of engaging in economic activity in the United States (Dwyer and Nau n.d.; Krippner Forthcoming). This same shift has occurred in the institution of higher education.

As tuition and other educational costs ballooned in conjunction with declining federal student aid, credit cards became a widespread method for financing costs associated with education and college living (Manning 2000). According to a recent study, 84 percent of undergraduate students have a credit card, and 92 percent of undergraduate students who have a credit card have used their credit to pay for “direct education expenses” like textbook and school supplies (Sallie Mae 2009). A smaller, but not insignificant percentage of undergraduate students use a credit card to pay for at least some part of their college tuition. These patterns show how “credit cards are used to bridge financial gaps in student budgets” (Manning 2000, p. 170).

This dissertation investigates the impact of increasing access to credit in conjunction with increasing access to higher education on college student financial behaviors and educational outcomes. Specifically, in what situations—and why—do college students use credit cards, especially to pay for direct educational expenses like tuition? And how do college students’ credit card spending behaviors connect to their subsequent educational outcomes like bachelor’s degree attainment? Better understanding the role credit cards play in the educational attainment process will help researchers,

4 higher education practitioners, and policymakers know how the changing landscape of higher education connects to students’ strategies for paying for college, and to what end.

But first, how did we get here? What historical developments led to widespread college student credit card use? The following section details societal changes over the past century that provided expansive access to both consumer credit and postsecondary education, changes that laid the historical foundation for the questions posed in this dissertation.

Historical Context for the Current Study

Prior to 1916, legal interest rates for personal loans were too low to incentivize banks to lend to individual consumers, simply because there was no profit potential.

When people encountered difficult times, they resorted to borrowing money from usurious loan sharks who lent money at illegally high interest rates. Seeing the flaws inherent in this system, the Russell Sage Foundation spearheaded original research in the area and advocated for a legal alternative for people suffering hard times and in need of a small loan (Carruthers and Ariovich 2010; Hyman 2012). This led to the Uniform Small

Loan Law in 1916 that raised legal interest rates on personal loans of $300 or less to 3.5 percent a month. Borrowers would pay interest on the unpaid balance, not the original amount. This law allowed for the spread of personal lending through the majority of the country by 1930 and because the interest was legally enforced, lenders could rely on the profits since borrowers were legally bound to paying back their debt, drastically reducing risks associated with lending to consumers (Hyman 2012).

The rise of the automobile in the 1920s introduced a novel development in the

American financial infrastructure. For the first time, customers could pay for a car through an installment plan and automobile companies could sell the remaining balance on a customer’s loan to a finance company. While installment lending had existed prior to the development of cars, purchasing automobiles through installment lending made this financial practice widespread. This development was the precursor to the shift toward increasing personal debt throughout the twentieth century (Hyman 2012).

In 1928, National City Bank began the first personal loan department in a bank.

This was the initial connection between commercial banking and consumer lending.

National City Bank did not initially think of this department as a profit-producing endeavor, but more of a “public outreach” program (Hyman 2012). And personal lending did not catch on until years later with the development of New Deal programming.

In order to stimulate the economy during the Great Depression, the Federal

Housing Administration developed the Title I loan program which loaned money to homeowners for home repairs and improvements. Homeowners saw these loans as affordable opportunities to improve their living standards, and the government guaranteed the principle of the loans to banks lending money to homeowners, greatly reducing the risk for lenders and increasing the amount of personal loans banks offered

(Hyman 2012).

In the face of declining and absent business loans, and because of the guaranteed profits insured by the government in personal loans, commercial banks opened up FHA

Title I loan departments and the connection between commercial banking and consumer

6 lending became ubiquitous. Banks across the United States opened up personal lending departments beyond the Title I departments as they saw the potential profit value in consumer lending (Hyman 2012).

In Debtor Nation (2012), Louis Hyman writes about this shift:

Loan balances rocketed from $33 million in 1933 to $129 million by 1936. Before Title I, less than 1 percent of commercial banks had a personal loan department. By the end of 1934, 71 percent of the ‘total banking resources’ did. Through ‘the example and the introduction of FHA modernization financing,’ one banking journal noted, banks ‘established such departments through which almost every type of personal or installment financing [could then] be arranged.’ (pp. 86-87)

This shift brought personal lending within the purview of commercial banks and pushed out other lending organizations (e.g., industrial banks, small loan companies) that had once been mainstays in personal lending. As banks expanded their borrowing options for customers, installment lending filtered into the purchasing process of other consumer goods outside of the automobile market (e.g., furniture). The expansion of consumer credit, however, was a concern during World War II because it had the potential to contribute to inflation while large proportions of resources were allocated to the war effort. President Franklin D. Roosevelt instituted Regulation W, which allowed the

Federal Reserve to regulate consumer borrowing. Regulation W placed heavy restrictions on installment lending, but businesses worked around these restrictions by using revolving credit accounts that were much more difficult to regulate. These revolving credit accounts were “open book” charge accounts that had no contract for repayment, which hid them from the regulatory eye of the government. This development was the precursor to the modern day credit card (Hyman 2012).

Seeing the proliferation of charge accounts (open credit) to avoid federal regulation, the federal government regulated consumer credit more severely, requiring stores to freeze charge accounts and turn them into installment plans after a certain period of time if the customer was not current on their payment, which brought these (now installment) accounts within the purview of Regulation W (Hyman 2012).

In order to comply with new federal regulations, businesses had to find an efficient way to track customers’ account balances in order to freeze them when necessary. Businesses used new technologies like the Addressograph and the Charga-

Plate to streamline and track customer payment. The Addressograph mechanized the process of labeling customers’ charge accounts and the Charga-Plate allowed businesses to efficiently identify a customer’s account through the account number printed on the card. By 1949, the Charga-Plate went hand-in-hand with revolving credit systems at businesses. Furthermore, revolving credit was growing in popularity because it brought with it newfound profit opportunities for businesses—an increase in sales and revenue from the interest on unpaid balances—and it allowed customers from less well-off backgrounds to have access to goods that had been previously prohibitively expensive.

Revolving credit democratized the American shopping experience (Hyman 2012).

At the same time that revolving credit was established as a mainstay in consumer lending, enrollment in higher education was exploding. After World War II, postsecondary education institutions saw low enrollment numbers balloon into massive enrollments over a brief period of time. In A History of American Higher Education

(2011), John Thelin writes about this transition:

In 1939-40, total student enrollment at all colleges and universities was just under 1.5 million. During World War II, regular student enrollments dipped substantially as a result of the military draft. The lack of students (and professors) led the president and faculty at Harvard to consider implementing a moratorium on enrollment and instruction at Harvard Law School and in other advanced graduate programs. All this changed after 1945. By 1949-50, total student enrollments had ballooned to almost 2.7 million—an increase of about 80 percent in one decade. This was no aberration, for the figure increased to about 3.6 million in 1960 and then doubled again over the next decade, reaching over 7.9 million in 1970. (p. 261)

During the Korean War, President Harry Truman, following the example set in

World War II, reenacted Regulation W to prevent inflation. However, this time only installment credit was regulated, not revolving credit. Revolving credit had been growing until this point and this restriction of installment lending and lack of restriction on revolving credit led to revolving credit becoming a ubiquitous practice around the country. After the Korean War, the focus was no longer on restricting consumer credit, but expanding access to credit to everyone as a way to give Americans the appropriate financial resources for upward social mobility (Hyman 2012).

At this time in American history, credit was seen in an undeniably positive light, as Hyman (2012) describes in this passage:

By the 1960s, credit access was deemed to be unequivocally beneficial. Credit use, far from marking one as immoral or unthrifty as it might have in the 1910s, denoted high social status and personal responsibility. In the 1960s, those without credit agitated for more ‘fair’ or ‘equal’ access. By the end of the decade, as access to credit became a social marker of independence and prosperity, various credit activists for women and people of color demanded access to credit. Those left out—middle-class women and working-class African Americans—wanted in. (p. 174)

Into the early 1970s, women experienced particular difficulty obtaining credit due to discriminatory practices in lending (Krippner Forthcoming). In marriage, women’s

9 credit histories were subsumed into their husbands’ and any credit offer to a wife required a co-signature by the husband. Even after marriage, lenders would often refuse to offer women credit within a year of divorce because they saw it as reflecting life instability— which restricted a pivotal resource from women during a time of need (Krippner

Forthcoming). Seeing these discriminatory practices, the National Organization of

Women (NOW) fought to end credit discrimination, leading to new legislation around credit access.

Passage of the Equal Credit Opportunity Act (ECOA) in 1974 made discrimination in credit lending on the basis of gender or marital status illegal, and subsequent amendments to the law in 1976 expanded access to all people regardless of their race, age, religious affiliation, or country of origin (Hyman 2012). The new law did not end discrimination in credit lending, however, but ushered in the practice of credit scoring to evaluate potential borrowers from a number of characteristics, including gender and race, that hid such discrimination within the statistical analysis of several traits that comprised credit scoring (Krippner Forthcoming).

Access to higher education was following a similar path. Calls for desegregation of colleges and universities in the 1960s were met with token acquiescence, many postsecondary institutions still excluding these groups from campus life (Thelin 2011).

However, expansion of access to credit simultaneously with expansion of access to higher education would intertwine in important ways in coming years.

At the same time that credit access was given to more people than ever before,

Americans began to experience economic turmoil and uncertainty due to

10 deindustrialization and rising unemployment. As the amount of debt an American carried on their credit card increased because of their decreasing ability to pay back what they borrowed, carrying long term credit card debt became widespread. Furthermore, banks earned the most profit on accounts that carried a balance from month to month; these customers were typically from lower socioeconomic backgrounds and disadvantaged racial groups. So, expanded access to marginalized groups meant higher profits for lenders and as interest rate caps were lifted in 1978, banks offered credit accounts to increasingly risky borrowers (Hyman 2012).

In Credit Card Nation (2000), Robert Manning writes about the unique developments during the 1970s that shifted credit card access and use in America:

As real wages began stagnating in the mid-1970s and inflation soared to double digits at the end of the decade, universal credit cards became the angel of mercy to struggling middle-class households. With the onset of federal deregulation of banking in 1980, the rising cost of real interest rates led banks to expand into increasingly profitable consumer financial services…Over the next two decades, the economic dislocations produced by U.S. industrial restructuring contributed to the successful marketing of consumer credit to traditionally neglected groups: low-income workers, blue-collar households, and college students. (pp. 112-113)

As more young people were going to college, thanks in part to programs like the

Pell Grant (Thelin 2011), college students became an increasingly strategic population to which lenders could market credit cards. Credit access expanded to college students in this time frame because of their potential future earning power. Prior to the late 1980s, credit cards were symbolic of advantaged living and sophisticated purchasing power.

Marketing to college students was a strategic move as these students connected credit card use with being upwardly mobile after college and living the good life. Furthermore,

11 colleges typically received a cut of the revenue in return for providing lenders with student addresses and information (Hyman 2012).

The predominant distributers of credit cards up until the 1980s were retail and department stores. Customers registered for charge cards at specific stores and learned how to engage in consumer behavior through revolving credit by using store-specific credit cards. Into the 1980s, credit cards sourced by third party lenders like banks were not widely used, but mainstream credit card use shifted from retail stores to banks as technological development gave third party lenders access to more efficient tools and therefore access to an advantage in the lending process. With highly variable interest rates and balances on option accounts, a centralized credit provider became possible because they had the machinery—namely computers—that could efficiently organize customers’ accounts and print bills. The emergence of the bank card—previously unsuccessful—had now come at the right time (Hyman 2012).

Credit card lending increased in its profitability through the 1980s and 1990s, and banks began to engage in the process of securitizing the balances on these credit accounts through third party investors. As securitization increased, credit lending became less risky, and lenders found increasingly sophisticated ways to make profits even from the riskiest borrowers. Credit card use exploded in the 1980s and 1990s and a stagnant labor market kept borrowers in debt. By the late 1990s, Americans no longer had the choice of whether to participate in the credit system. Even without a credit card, every American was given a Fair Isaac Corporation (FICO) credit score that influenced their ability to pay for education, housing, and other basic necessities (Hyman 2012).

As the credit card market became saturated in the 1980s and 1990s, credit card companies aggressively marketed to college students. By 1998, around two-thirds of undergraduate students at four-year colleges and universities had a credit card (Manning

2000). The number of college students who owned credit cards would continue to grow over the next ten years (Sallie Mae 2009).

By 2000, 158 million credit card holders owned 1.5 billion consumer credit cards, an enormous number, both per person and overall (Manning 2000). Credit card advertisements targeted at college students increased during the 1990s and into the early

2000s. Credit cards gave college students access to financial independence that contributed to the experience of going to college and becoming independent from parents and authority figures. Advertisements presented credit card debt as a normal part of college life (Manning 2000).

Higher education enrollment continued to expand in the late 1990s and early

2000s, with total postsecondary enrollment reaching 16 million students (Thelin 2011).

The cost of a college degree rose substantially over this time period in tandem with stagnating wages and lower state-level government support for higher education, decreasing families' ability to pay for their children’s college education and decreasing students’ ability to pay for their own education through part time and summer work. In this predicament, college students resorted to borrowing money to compensate for the diminishing support available to them (Manning 2000). Manning writes about this shift and its impact on college student credit card use:

Since the mid-1970s, rapidly escalating college costs and declining financial aid and real wages have forced students increasingly to rely on credit cards to help 13

pay for their college educations. This has led to a new trend in which credit card debts are being revolved—paid off with federal student loans or even with private debt consolidation loans. For growing numbers of students, credit cards are becoming a savior for financing their educations—especially in public schools. For others, consumer credit initially offers freedom but may become a financial shackle by the end of their college career. The most unfortunate may find that their only option for regaining personal control in the just-do-it culture of credit dependency is to withdraw from school and work full-time in order to pay off their debts. Indeed, official dropout rates (attributed to low grades) include growing numbers of students who are unable to cope with the stress of both their debts and the part-time jobs they must take to service those debts. For others, the reality of their credit card indebtedness may not be realized until after graduation when prospective employers question their past financial recklessness or when they must accept a sharp decline in their standard of living. (pp. 191-192)

Yet despite the problems pointed out in Manning’s account, credit card companies continued to market to college students, even establishing a physical presence on campuses and offering free prizes to students who signed up for a credit card (Karger

2005). Educators highlight credit card debt as a growing factor related to college students’ reasoning for dropping out of school (Hancock, Jorgensen, and Swanson 2013;

Karger 2005; Manning 2000). Furthermore, some research shows evidence that college students use additional credit cards to manage their current credit card debt, creating a bleak picture for college students trying to get out of debt and afford college (Norvilitis and MacLean 2010). In 2009, 84 percent of undergraduate students owned a credit card—compared to two-thirds of undergraduates a decade earlier—with average balances of $3,173 (mean) and $1,645 (median) (Sallie Mae 2009). Seeing the problems of the connection between credit card companies and college students without the financial resources of many post-college Americans, researchers and government officials took action to lobby for government restriction on college student credit card access.

On May 22, 2009, President Barack Obama signed into law the Credit Card

Accountability, Responsibility and Disclosure (CARD) Act which, among other things, restricted credit card access for consumers who are less than 21 years of age if they cannot either demonstrate that they have an independent income or do not have a co- signer who is 21 or older (U.S. Congress 2009). The Credit CARD Act also limited credit card companies’ ability to market to students on campus, but whether these provisions actually influenced credit card companies’ access to students under 21 is unclear

(Matthews 2013). For example, credit card companies allow college students to list student loans as part of their independent income, thereby qualifying them for credit cards under the new law.

College students continue to have access to and to use credit cards while pursuing their postsecondary degrees. While this practice would have been foreign to most college students just 25 years ago, college student credit card ownership and use is widespread.

We know little about the situations in which college students choose to use credit cards and how this relates to their educational attainment process. This dissertation sheds light on this issue by empirically investigating these questions. The sections of this dissertation are outlined below.

Dissertation Outline

Prior research shows that credit access and use matters for life outcomes (Hodson,

Dwyer, and Neilson 2014; Krippner Forthcoming; Nau et al. 2015). This dissertation

15 considers a specific period of the life course—college—as an interesting place of empirical inquiry into the connection between credit access and use and life outcomes.

Chapter 1 has contextualized the main inquiry within the current sociological research on credit access and higher education. This chapter incorporated an historical investigation of the development of access to higher education and access to consumer credit through the financialization of the American economy in the post-World War II

United States.

Chapter 2 of the study is the first empirical paper. This paper describes some of the factors that might influence patterns of credit card use among college students.

Questions in this paper include: What are the demographic differences between college students who use credit cards and those who do not? What are the differences between college students who carry a monthly balance on their credit cards and those who own a credit card but do not carry a balance? Are there different financial situations between college students who use credit cards and those who do not? Are there different financial situations between college students who carry a monthly balance on their credit cards and those who do not? This paper adds to current research on college student credit card use that does not always include information on students’ financial situations and demographic factors that might influence said use.

Chapter 3 of the study is the second empirical paper. One specific development in this research area is the advent of college students using credit cards to pay for some of their tuition. This pattern is particularly concerning because it might suggest that students in financially strapped situations will resort to risky measures to finance their college

16 degree, to the potential detriment of finishing college. This paper investigates this financial strategy as a way to understand whether students are using credit cards because of financial restrictions such that credit cards serve as a way to bridge a necessary financial gap, or because of differences in levels of financial literacy. Meaning, do financial situations or financial knowledge better explain student use of credit cards to pay for tuition? Other questions in this paper include: Who participates in this financial strategy, and in what circumstances? What are the primary reasons college students use this financial strategy to fund their degrees? How do students’ levels of financial knowledge vary? And how do these variations connect to students using a credit card to pay for tuition?

Chapter 4 of the study is the third empirical paper. This paper extends the analyses of the first two empirical papers by investigating how patterns of college student credit card use impact educational outcomes. This paper is aimed at understanding how students’ responses to increased financial responsibility in college (e.g., taking on loans, credit card debt) influence their educational trajectories. This paper considers questions including: How does carrying credit card debt influence educational outcomes like degree attainment? Are specific spending strategies related to credit card use (e.g., carrying a revolving credit balance, having several credit cards, paying for tuition with a credit card) related to educational outcomes?

Chapter 5 concludes the study with a discussion of the findings and a synopsis of the significance of the study’s findings for current policy and future inquiry into the topic. This discussion will reflect on the salience of the study’s findings for scholars of

17 higher education, practitioners, and policy makers. This dissertation aims to investigate this topic of research in a way that is accessible to both sociology and higher education audiences.

CHAPTER 2: Patterns of College Student Credit Card Ownership and Use

Introduction

In the face of a thin social safety net, Americans turn to finance strategies to navigate tricky economic waters. For example, recent research shows that growing economic inequality in tandem with a financialized economy has led to Americans using credit to manage gaps in their budgets (Dwyer and Nau n.d.; Fligstein and Goldstein

2015). This type of financial behavior is increasingly common; young adults today are carrying more unsecured debt than previous generations of young adults (Houle 2014).

Households across the socioeconomic spectrum have increased use of financial products (e.g., mortgages, home equity loans, mutual funds, credit cards), but not at the same rate. Higher socioeconomic groups use financial products and services at higher rates in order to maintain and expand current lifestyles, while lower socioeconomic groups use these products at lower rates, and typically in times of economic trouble

(Fligstein and Goldstein 2015). Regardless of the motive or situational context,

Americans are using financial strategies to fund critical purchases throughout the life course. One example of this pattern is the widespread student use of educational loans to finance higher education.

Beginning with the National Defense Act of 1958 in which the federal government first offered student loans for postsecondary education, students now have the ability to finance a college degree through federal Stafford loans, Perkins loans, institutional emergency loans, Parent Loan for Undergraduate Students (PLUS) loans, and private loans, among other options like home equity or 401k loans. The financialization of the American economy has influenced institutions of higher education, both in the sources of their revenue and in the methods by which students pay to attend

(Eaton et al. 2016). Debt is now a part of the college experience for many students, and the cumulative student loan debt continues to rise. Total student loan debt in the United

States is now over $1 trillion, greater total debt than credit card debt or automobile loans

(Kantrowitz 2016).

The preponderance of available student loan options shows evidence of the impact of financialization on institutions of higher education. But student loans are not the only finance-related options available to students looking to fund their college degree and meet educational expenses. In Credit Card Nation, Robert Manning shows that

“credit cards are used to bridge financial gaps in student budgets,” (2000, p. 170). In the face of rising tuition and declining government funding of higher education, students resort to strategies similar to those used by many Americans in a financialized economy in order to finance their college degrees. Not only do college students use credit cards to fund indirect educational expenses, but a sizeable share of college students even use credit cards to directly fund their college tuition.

We know that college students use these financial strategies to navigate higher education, but we know surprisingly little about the differences between students who use credit cards and those who do not. Furthermore, we know little about how particular credit card practices might be overrepresented among certain demographic groups. For instance, the historical development of credit cards across racial groups documented banks soliciting credit cards to non-white Americans because of their higher likelihood of carrying a balance on their credit card, and therefore higher profits for the bank (Hyman

2012). Because credit cards charge higher interest rates than many credit-based financial products, if some demographic groups are more likely to carry a revolving balance and pay more interest from using a credit card, these patterns are likely to have a connection to economic inequality in America through inhibiting wealth building among some

Americans and not others.

Because some research has documented a compensatory effect of college for these social disparities (Brand and Xie 2010; Jackson and Reynolds 2013), it is important for us to better understand all of the factors that may maintain social inequality through prohibiting disadvantaged social groups from graduating from college. Similar to the historical differences in who was given access to credit and who benefited most from having access to credit, this paper investigates demographic differences in college student credit card ownership and use in order to better understand how this issue intersects with social inequality during this time in the life course.

College student credit card use is a particularly salient topic of study considering recent rising tuition costs in conjunction with stagnating government funding for higher

21 education and decreasing opportunity in the labor market for families to support their children in college (Goldrick-Rab 2016). These societal shifts leave college students with limited options to pay for college, and banks and credit card companies have played an important role in bridging these financial gaps for many students (Manning and Kirshak

2005). So, how do students use credit cards? Who is more likely to own and use a credit card? And in what situations? This paper seeks to provide some clarity on this issue through an empirical analysis of current and recent college students.

College Students and Credit Cards

Research has observed a drastic increase in college student ownership and use of credit cards from the late 1980s through the 1990s (Bird, Hagstrom, and Wild 1999;

Robb and Sharpe 2009), and a continued rise in ownership and use into the 2000s (Sallie

Mae 2009). Studies document between 60 and 85 percent of college students in the past two decades owning a credit card (Manning 2000; Manning and Kirshak 2005; Nellie

Mae 2005; Sallie Mae 2009; United States General Accounting Office 2001).

The shift from marketing credit cards to economically established groups to economically insecure groups like college students occurred largely through the deregulation of retail banking in the late 1970s and early 1980s (Manning and Kirshak

2005). Banks and other financial organizations that offered consumer lending noticed they could make substantial profit gains from customers who carried revolving credit balances on credit cards because interest rate caps were lifted from more conservative rates in prior years (Hyman 2012). This newly accessible profit source encouraged these

22 companies to offer credit to increasingly risky customers, notably college students who often did not have an established income or a clear way to pay down revolving credit card debt (Manning 2000). This practice of offering college students credit cards continues today.

A prominent reaction to this development is one of disdain; many people think that credit card companies are targeting the college student population because they are inexperienced financially, somewhat gullible, and will easily sign up for a credit card

(Burnsed 2010; Manning 2000). On the other hand, some suggest that these companies think that this group of students are a better investment because of their higher than average earnings potential, and therefore their greater chance of paying off any credit card debt they may accrue over time, particularly after they graduate and get a good- paying job (Hayhoe et al. 2005). Research has shown that despite near ubiquitous credit card use among college students, most students use credit cards responsibly and do not engage in risky financial behaviors simply because they have access to credit (Lyons

2004).

Many researchers and higher education practitioners consider credit card debt problematic because they see credit card debt as an additional layer of debt to the growing student loan debt—debt that a college student takes with them after college and debt that influences a number of college and post-college decisions (The Education

Resources Institute and The Institute for Higher Education Policy 1998). Research shows that debt is related to college graduation rates, marriage and family decisions, future wealth attainment, health problems, and future access to credit (Dwyer, Hodson, and

McCloud 2013; Dwyer, McCloud, and Hodson 2012; Hodson, Dwyer, and Neilson 2014;

Houle and Berger 2015; Manning 2000; Nau, Dwyer, and Hodson 2015). Student debt receives widespread critique for burdening students with more than they can financially handle, and credit cards receive similar judgments considering their substantially higher interest rates comparative to student loans, and the lack of ability to defer payments until the student is no longer enrolled in school.

Current research finds that individuals with fewer financial resources are more likely to utilize riskier credit sources (Manning 2000), which then puts them at a further disadvantage in terms of becoming financially stable and experiencing upward social mobility because those credit sources come with high interest rates. And with banks and credit card companies commonly soliciting college students to sign up for a credit card, many researchers, higher education practitioners, and policy-minded individuals have expressed concern that these companies were financially benefiting from a vulnerable population (Bianco and Bosco 2002; Hawkins 2012; Hayhoe et al. 2005; United States

General Accounting Office 2014).

However, methodological limitations in the research up to this point limits our ability to fully understand the scope of college student credit card ownership and use on campus. This paper aims to readdress some of the questions that have been asked in the extant research with nationally representative data in order to evaluate our understanding of this issue in a more robust manner.

Methodological Limitations in the Literature

Despite widespread attention to credit cards on college campuses (Manning 2000;

Rubin 1998; United States General Accounting Office 2001, 2014), research on college student credit card use has several methodological limitations that prevent current knowledge in this area from having a more complete understanding of this topic. Perhaps the most salient methodological hindrance to trustworthy findings in this area of research is the preponderance of single institution studies (Gnizak, Meier, and Stark 2004; Jamba-

Joyner, Howard-Hamilton, and Mamarchev 2000; Jones 2006; Lyons 2004; Saez 2008;

Sotiropoulos and d’Astous 2012; Wang and Xiao 2009). Because so many of these studies have been conducted using convenience samples at one university, their findings can only be extrapolated to make claims about that university, provided these samples are representative of the broader student population at that university. In order to learn about college student credit card use more broadly, research must transition to using national datasets that can tell us more about these patterns across the several thousand existing institutions of higher education.

Typical within these single institution studies are small sample sizes (Gnizak et al.

2004; Jamba-Joyner et al. 2000; Jones 2006; Saez 2008; Sotiropoulos and d’Astous 2012;

Wang and Xiao 2009) that are often collected from one particular sub-group of students

(e.g., an introductory statistics course or first-year orientation) (Jamba-Joyner et al. 2000;

Jones 2006; Wang and Xiao 2009). These sampling practices further limit the ability of these research findings to speak to credit card ownership and use among a national group of college students. And even among studies that sample across a national group of universities, there still exists questionable sampling practices. For example, one particular

25 yearly study on the way college students use financial services surveys only 12 students at every university they include in their sample (Student Monitor 2016).

Furthermore, because credit cards produce so much profit for banks and credit card companies, some research reports questionable findings due to the support of those financially interested in seeing credit cards continue to be used widely among college students. For example, Barron and Staten’s (2004) study originated from questionable funding sources from credit card companies that likely influenced their results, as noted in Manning and Kirshak (2005). Research on a financial vehicle that produces profit gains for banks and credit card companies to the extent credit cards do will likely continue to contain biased studies influenced by funding sources interested in growing credit card use across the American population.

While there are substantial limitations to the research findings in this content area, there are also examples of studies that use better data and methods to back their empirical analyses (Adams and Moore 2007; Gutter and Copur 2011; Hodson et al. 2014; Lusardi,

Mitchell, and Curto 2010). Furthermore, extant research, despite the limitations, has provided a foundation for research in the area of college student credit card use. The following section provides a snapshot of the research on college student ownership and use in order to set up the current empirical inquiry.

Patterns in College Student Credit Card Ownership and Use

Some research in this area has investigated college student credit card ownership and use as an independent variable associated with certain outcomes reflected in the

26 sample studied. For example, research shows that college student credit card debt is related to higher levels of financial stress and lower levels of financial well-being (Grable and Joo 2006; Gutter and Copur 2011). Another study finds college student credit card debt, particularly high levels of debt (greater than $1,000), is associated with risky health behavior like binge drinking and substance use (Nelson et al. 2008). High levels of credit card debt are also associated with lower mental and physical health (Berg et al. 2010).

Furthermore, credit card debt as a factor leading to college student attrition from postsecondary education has been well publicized (Hayhoe et al. 2005; Manning 2000;

Rubin 1998). The lion share of research in this area, however, deals with college student credit card ownership and use as a dependent variable with a number of variables predicting when a student will be in credit card debt, and to what extent.

College students with student loans are more likely to own a credit card than students without loans, and are also more likely to carry a revolving balance on a credit card compared to students without loans (The Education Resources Institute and The

Institute for Higher Education Policy 1998). Students from traditionally disadvantaged backgrounds are more likely to carry high levels of credit card debt and have revolving balances on their credit cards. For example, college students from low socioeconomic backgrounds, students of color, and women are more likely to carry a revolving balance on their credit cards (Grable and Joo 2006; Jamba-Joyner et al. 2000). This pattern holds true among students with high levels of credit card debt. Lyons (2004) classifies students with more than $1,000 in credit card debt as financially at risk and finds that financially at risk students are more likely to be black, Hispanic, female, and receiving need-based

27 financial aid. These findings are borne out in other studies as well (Grable and Joo 2006;

Lyons 2008).

Additional research considers how college students’ attitudes, financial knowledge, and personality characteristics contribute to patterns of credit card ownership and use. Norvilitis and collaborators (2006) find that college students’ ability to delay gratification and their attitudes toward taking on debt are associated with credit card debt levels. In the same study, Norvilitis et al. argue that financial knowledge emerges as the preeminent predictor for credit card debt among the variables considered in their models.

Another study shows that college students with parents who argue about finances and students who are comfortable making minimum payments on a credit card each month have higher levels credit card debt (Hancock, Jorgensen, and Swanson 2013). Virtually all of the studies that consider how college students’ attitudes, financial knowledge, and personality characteristics relate to credit card debt levels suggest that more robust financial education is needed on college campuses.

This paper builds on this prior literature by examining some of these questions with a nationally representative data set and interrogating predictors of college student credit card ownership and use from a sociological perspective. That is, this paper investigates several variables potentially related to college student credit card ownership and use in order to see which demographic variables and indicators of a college students’ social situation consistently relate to certain patterns of college student credit card use.

The following section details the unique approach of this paper within the context of this area of research.

The Current Study

As noted in one recent study (Sotiropoulos and d’Astous 2012), much of the research in the area of college student credit card use explains differences in patterns of use as due to differences in certain psychosocial factors that influence college students’ decisions to own and use a credit card (e.g., lack of self-control, favorable attitudes toward debt). Scant research in this area aims to understand what social factors might influence college students’ credit card ownership and use. This paper investigates differences in credit card ownership and use by demographic groups like socioeconomic status by focusing on students’ financial situations as a particular way to investigate the social factors that might be related to college student credit card ownership and use.

Specifically, this paper considers college students’ financial aid levels, work hours per week, and family support as potential factors through which socioeconomic disadvantage may manifest and connect to credit card ownership and use. Meaning, students from lower socioeconomic backgrounds may need to use credit cards as a way to bridge gaps in their budget, even if they are working and have financial aid.

This paper adds necessary information to this conversation by expanding the variables that have been considered to relate to college student credit card ownership and use by incorporating student financial aid variables and employment status. Furthermore, it provides a national look at college student credit card patterns that have largely been examined by single institution studies. This paper also contributes a unique focus on credit card ownership and revolving credit balances which might be a particularly

29 important issue that is often overlooked in favor of the amount of credit card debt a student holds. Research shows how revolving credit is a particularly problematic version of credit card use because it means, especially for populations without an established credit record, that the repayment amount is much higher than the initial purchase and can be financially burdensome for those from financially strapped situations (Hyman 2012).

While this paper contributes a unique conceptual approach to college student credit card ownership and use by investigating these social factors, it also performs several basic demographic analyses with a nationally representative dataset to understand better trends in college student credit card ownership and use. Among college students, who is most likely to own a credit card? Who is most likely to carry a balance from month to month? How do these patterns vary between demographic groups?

I predict that students from traditionally disadvantaged and underrepresented backgrounds will be more likely to carry a revolving credit balance because of the historical precedent of this established in prior research (Hyman 2012). I also predict that college students’ employment status will be a primary predictor of college student revolving credit card debt as working more hours may be a way college students navigate trying to pay off their credit card debt.

The following sections outline the empirical strategies I take to examine these questions in an attempt to offer a more robust understanding of college student credit card ownership and use with a nationally representative dataset that contains data from college students who were recently enrolled in college.

Methods

Data

In order to examine this paper’s empirical questions, I use data from the

Education Longitudinal Study of 2002 (ELS: 2002) from the National Center for

Education Statistics (NCES). The ELS dataset is a nationally representative dataset that surveys students who were sophomores in high school in 2002 and follows them as they go to college, transition into the workforce, and make other life transitions. The ELS spans ten years from 2002 to 2012. The ELS data are an ideal source of information for answering this paper’s questions about college student credit card ownership and use because of their national scope and how recently they have been collected compared to much of the research in this area that occurred in the late 1990s leading up to 2000 (The

Education Resources Institute and The Institute for Higher Education Policy 1998;

United States General Accounting Office 2001).

The study sample (n = 7,190) includes respondents who were enrolled in a postsecondary institution in 2006. Respondents in this sample have information on credit card ownership and use, as well as financial aid information, employment status, and information on several demographic indicators. Most of the analytical models in this paper contain sample sizes of over 7,000 respondents; models contain slight variations in sample sizes due to small differences in missing data between variables represented in some models and not represented in other models. This variation is minor, however, as the restricted sample contains very little missing data in the variables considered in this paper.

Variables

I use variables from the ELS data that report respondents’ credit card ownership and patterns of use in 2006; these variables serve as dependent variables throughout the paper. These variables include the number of credit cards a student has in their name and whether they carry a revolving balance on their credit card from month to month. I restructured the number of credit cards variable into two variables. One is a dichotomous variable that divides students into credit card owners and students who do not own a credit card. The second is a categorical variable that divides students into three categories: students who do not own a credit card, students who own 1 credit card, and students who own 2 or more credit cards due. I group students who own 2 credit cards with students who own up to 7 credit cards in one category due to small cell sizes in credit card ownership above 2 credit cards.

While the variable asking respondents about the revolving credit balance was originally asked only to students who indicated that they owned a credit card in their name, I merged this variable with the credit card ownership variable to include all students in the sample, even if they indicated that they do not own a credit card. This way, the analyses offer further nuance within this study by comparing differences between college students’ ownership of credit cards and patterns of credit card use between three groups: 1) those who do not own a credit card, 2) those who own a credit card but pay off any balance they have every month, and 3) those who own a credit card and carry a balance from month to month. I see these three groups as qualitatively different groups and think that this variable configuration tells us more than a continuous

32 credit card debt variable that groups students who pay off their debt each month in the same category (no credit card debt) as students who do not own a credit card.

Independent variables in the analyses include the respondent’s socioeconomic status, the amount of hours they worked per week during the 2005-2006 academic year, the type of postsecondary institution they first attended, whether they delayed entry into college after high school graduation by seven months or more, whether they are a first generation college student, race, gender, family status (e.g., whether they are married or have been previously married and/or whether they have a biological child), and several financial aid variables that include total amount of student loans, total amount of Pell

Grants, and whether parent loans are paying for part of their college education.

I use the ELS provided framework as a starting place for the socioeconomic status variable included in the analytical models. Father’s level of education, mother’s level of education, family income, father’s occupation, and mother’s occupation are equally weighted and standardized to form one socioeconomic variable with four ordered quartiles. After examining cell counts in each socioeconomic quartile, I operationalized the final socioeconomic status variable as a three category variable that collapses the bottom two quartiles into one “Lower” category and translates the top two quartiles into

“Middle” and “Upper.” This operationalization is largely for making the analytical models as parsimonious as possible.

I collapse two separate variables that describe respondents’ marital status and whether they are parents into one family status variable, coded as a dichotomous variable where 1 means that the respondent is currently married, has formerly been married,

33 and/or has a biological child by 2006. Complete variable descriptions and descriptive statistics for all variables can be found in Appendix A.

Statistical Analyses

First, I conduct demographic analyses through a series of binary comparisons using cross-tabulations and Pearson chi-squared tests. I use these methods to report basic demographic patterns of college student credit card ownership and use and describe the salient differences between student groups and their likelihood of owning a credit card, the number of credit cards owned by different groups, and their likelihood to participate in certain credit card behaviors like carrying a balance from month to month.

Second, I evaluate these demographic and other relevant independent variables within logistic regression models in order to test which of these variables retains a significant relationship to these credit card variables after controlling for the other pertinent variables. In this way, I aim to explore which of these factors emerges as a salient predictor of college student credit card ownership and use. All multivariate regression models in this paper have been tested for multicollinearity concerns using mkcorr and collin commands in Stata. These tests reported no issues related to multicollinearity. Correlation matrices for each regression model in this paper are presented in Appendix D.

Limitations

In order to expand the sample size wide enough to include as many respondents as possible, I use minimal restrictions for ELS respondents in the final analytical models. As long as the respondent is enrolled in a postsecondary institution in 2006, they are

34 included in the models. This restriction, of course, is not something to simply ignore as it excludes all students who never attended a postsecondary institution after high school, and it does not include students who delayed entry into college more than two years after high school. Because I open up the sample qualifications to include more respondents than other papers in this dissertation, this paper has unique limitations to the empirical analyses and findings.

For example, students may have varied paths into and out of college between high school—most of these students graduated in 2004—and the 2006 enrollment census date that this sample shares. While some students may have enrolled immediately into college and are part way through their second year in college, others might have just enrolled in college a couple of months prior to 2006. These students will have had entirely different experiences that might influence how they approach credit card ownership and use.

This paper cannot make claims about the varied pathways students take through higher education and how that relates to college students’ credit card ownership and use, but it can give a broad look at a variety of students who were enrolled in college in 2006 when the ELS credit card questions were asked. Students may be varied in their stage of postsecondary schooling in 2006, but the sample shares that they are enrolled in a postsecondary institution in 2006, the same year that the ELS asked about their credit card ownership and use, which means that the empirical sample looks explicitly at currently enrolled college student credit card ownership and use patterns.

These shortcomings notwithstanding, this paper provides a broad look at college students enrolled within the past decade and how they used credit cards. The variety that

35 the few sample restrictions introduces into the sample allows for an appropriate reflection of the continually varied pathways students take through higher education (Goldrick-Rab

2006). Furthermore, placing fewer restrictions on the empirical sample allows for a higher number of students from lower socioeconomic or disadvantaged backgrounds to be represented in the sample since these students are also more likely to delay entry into college after high school graduation (Bozick and DeLuca 2005; Roksa and Velez 2012).

While this paper’s models are rather simple and straightforward, they give the reader a broad understanding and detailed picture of credit card ownership and use on college campuses in America.

Results

The following results show several interesting patterns in the ELS data that suggest college students from disadvantaged or lower-resourced backgrounds are more likely to own a credit card and use it in a financially risky manner (e.g., owning multiple credit cards or carrying a revolving balance). The demographic analyses offer three different perspectives on credit card ownership and use and show that even when differences in credit card ownership seem to favor advantaged student groups (e.g., starting at a 4-year institution relates to higher levels of college student credit card ownership), investigating whether students own multiple credit cards and whether they carry a revolving credit balance on their credit card reveals that disadvantaged student groups are more likely to practice these riskier financial behaviors.

Logistic regression analyses extend the demographic analyses and aim to clarify which of the initial demographic models maintain their statistical salience once all of the demographic variables are combined into a single model. Findings from these models vary and different variables arise as important indicators of college student credit card ownership and use. What is most striking in these models, however, is the staying power that a student’s work status (i.e., the number of hours they worked per week in the 2005-

06 academic year) has across all three logistic regression models. Without fail, work status significantly relates to college student credit card behaviors, with students who work more than 20 hours per week exhibiting the riskiest credit card behavior compared to students who work 1-20 hours a week and students who do not work

These findings support the hypotheses outlined at the beginning of this paper that predict students from underrepresented and disadvantaged groups will be more likely to exhibit risky credit behaviors like carrying a credit balance from month to month as well as the prediction that a student’s employment status will be a primary predictor of college student credit card ownership and use patterns in the final models.

Demographic Analyses1

Table A.2 shows statistically significant demographic patterns in college student credit card ownership. The percentages presented in Table A.2 were tested within a cross- tabulation comparison using Pearson chi-squared tests to explore whether the variation in

1 Figures A.1 through A.6 in Appendix A show select findings from Tables A.2 through A.4 in a chart format to provide a different visual representation, particularly of risky credit card behaviors. 37 credit card ownership was statistically significant between subgroups in the sample. See

Table A.2 in Appendix A.

Among the analytical sample considered, 47.23 percent of students own a credit in their own name. Credit card ownership varied significantly by socioeconomic status, work hours, institution type where the respondent first started college, first generation student status, total student loan amount, race, and gender.

Students from lower socioeconomic status backgrounds are more likely to own a credit card in their name than students from middle and upper class backgrounds.

Students who worked more than 20 hours per week during the 2005-06 academic year were more likely to own a credit card than students working 20 hours per week or fewer and students who did not work. Students who first start college at a 4-year postsecondary institution are more likely to own a credit card than students who first start college at a less than 4-year institution. Female students and students with greater levels of student loans are more likely to own a credit card, as are Asian students and students who are categorized in the “Other” race category.

I ran four other cross-tabulation models for students who delayed entry into college by seven months or more, total Pell Grant amount, whether the respondent used parent loans to help pay for school in 2006, and whether the respondent was married, previously married, and/or had a biological child. None of these models showed statistically significant differences in credit card ownership along these lines.

Table A.3 reports statistically significant differences in college student credit card ownership with a particular focus on students who own multiple credit cards, defined

38 here as students who own two or more credit cards. The percentages presented in Table

A.3 were tested using cross-tabulation models and Pearson chi-squared tests to explore whether subgroups differed in their likelihood of owning multiple credit cards. See Table

A.3 in Appendix A.

Overall, 28.20 percent of students in this sample owned one credit card in their name and 19.04 percent of students owned two or more credit cards in their name.

Ownership of multiple credit cards varied significantly by socioeconomic status, work hours, institution type where the respondent first started college, first generation student status, total student loan amount, total Pell Grant amount, race, and gender.

Students from lower socioeconomic backgrounds were more likely to own two or more credit cards than students from middle and upper class backgrounds. Students who worked more than 20 hours per week were more likely to own two or more credit cards than those who worked less or did not work. Interestingly, though students who started at a 4-year institution are more likely to own a credit card—as shown in both Table A.2 and

Table A.3—students who start college at a less than 4-year institution are more likely to own two or more credit cards. First generation college students, non-White students, female students and students with loans are more likely to own two or more credit cards.

Students with a moderate amount of Pell Grants ($1-$10,000) are more likely than students without Pell Grants and students with high levels of Pell Grants to own two or more credit cards.

I ran three other cross-tabulation models for students who delayed entry into college by seven months or more, whether the respondent used parent loans to help pay

39 for school in 2006, and whether the respondent was married, previously married, and/or had a biological child. None of these models showed statistically significant differences in multiple credit card ownership along these lines.

Table A.4 shows the final set of demographic analyses that investigates college student credit card use by testing differences in students’ likelihood of carrying a credit balance from month to month, also called a revolving balance. Like Tables A.2 and A.3,

Table A.4 reports statistically significant results from Pearson chi-squared tests in cross- tabulation models. See Table A.4 in Appendix A.

Over half of the sample does not own a credit card. About a third of students own a credit card and pay off any balance on their credit cards every month. Nearly 15 percent of students carry a revolving balance on their credit cards. These patterns vary between subgroups in the sample. Specifically, whether a student carries a revolving credit balance on their credit card varied significantly by socioeconomic status, work hours, institution type where the respondent first started college, whether a student delayed entry into college after high school, first generation student status, total student loan amount, total Pell Grant amount, race, gender, and family status.

Students from lower socioeconomic backgrounds are more likely to carry a revolving credit a balance compared to students from middle and upper class backgrounds. Students who work more than 20 hours are more likely to carry a balance on their credit card from month to month than students who work less than 20 hours a week or who do not work. Despite being less likely to own a credit card than students who start college at a 4-year institution, students who start college at a less than 4-year

40 postsecondary institution are more likely than students who start college at a 4-year institution to carry a revolving credit balance. Students who delay entry into college after high school graduation, first generation college students, students with high levels of student loans, non-White, non-Asian students, and female students are all more likely to carry a balance on their credit cards from month to month. Students with moderate levels of Pell Grants and students who have ever been married or have a biological child are more likely to carry a revolving credit balance as well.

I ran one other cross-tabulation model for students who used parent loans to help pay for school in 2006. This model did not show a statistically significant difference in carrying a revolving balance on a credit card between students who used parent loans and those who did not.

Building on these initial descriptive analyses, the following sections report results from logistic regression models that interrogate which of these variables considered in the demographic analyses serve as important predictors for college student credit card ownership and use.

Logistic Regression Models

Demographic analyses are helpful for understanding basic patterns in the data related to college student credit card ownership and use, but they lack the ability to control for other variables that may influence the relationship between student demographic indicators and credit card variables. Tables A.5, A.6, and A.7 extend the analyses in the prior section by using the outcome variables in Tables A.2 through A.4 as outcome variables for three logistic regression models, all with the same independent

41 variables. All three models in these tables are binary logistic regression models and Table

A.6 and A.7 consider a restricted sample of students who own a credit card. Table A.6 and A.7 are restricted in order to isolate the important difference between owning a credit card and using it in a non-risky manner versus owning a credit card and practicing risky credit card use (e.g., owning two or more credit cards, carrying a revolving credit card balance). This is not to say that these models test why students use credit cards in this manner, but rather what variables predict this kind of credit card use as an outcome.

Tables A.5 through A.7 present three full analytical models that test predictors for three credit card outcome variables. As a precursor to testing these full models, I tested simplified models that restricted independent variables to demographic control variables such as socioeconomic status, race, gender, and family status. I ran these simplified models to investigate demographic patterns that may have been hidden in the full analytical models. For example, if socioeconomic status was a significant predictor of credit card ownership in a simpler model but a nonsignificant predictor in a model that contained employment status, this might suggest that socioeconomic disadvantage manifested in lower socioeconomic students’ need to work during college to financially support themselves. The simpler models are reported next to the full models in each table.

See Table A.5, A.6, and A.7 in Appendix A.

Table A.5 considers the outcome of credit card ownership as a binary dependent variable, either a student owns a credit card or not. The demographic analyses showed significant variation in college student credit card ownership by a number of variables, and Table A.5 shows which of these variables maintains statistically significant variation

42 in credit card ownership after controlling for a host of variables in the model. Students who work more than 20 hours a week are more likely to own a credit card than students who work 1-20 hours a week. Students who do not work are less likely to own a credit than students who work 1-20 hours a week. Starting at a 4-year institution and having high levels of student loan debt are both associated with a higher likelihood of owning a credit card. Asian and Hispanic students are more likely than White students to own a credit card, while Black students are less likely to own a credit card than White students.

Female students are more likely to own a credit card than male students.

Table A.6 uses the same set of independent variables in the model but changes the outcome variable to whether a student owns multiple credit cards. This model focuses its analysis on a restricted sample that includes only students who own credit cards. This methodological decision is appropriate here because Tables A.5 through A.7 are largely concerned with showing robust statistically significant patterns related to which independent variables predict risky credit card behavior. So, the difference between students without a credit card and those with a credit card, having been addressed in

Table A.5, is not as interesting to Tables A.6 and A.7. Table A.6 uses a dichotomous dependent variable where 0 = a student owning one credit card and 1 = a student owning two or more credit cards.

Table A.6 shows a statistically significant difference in the socioeconomic status variable where Model 1 did not. Students from the upper class category are less likely than students from the middle class category to own two or more credit cards. Similar to

Table A.5, Table A.6 shows that students who work more than 20 hours a week are more

43 likely to own two or more credit cards than students who work 1-20 hours a week. In contrast, students who do not work are less likely to own two or more credit cards than students who work 1-20 hours a week. Hispanic students are more likely to own two or more credit cards, compared to White students. Female students are more likely to own two or more credit cards than male students.

Table A.7 shifts the analysis to a binary outcome variable that measures whether a student carries a revolving balance on their credit card from month to month. This variable is by nature different in its implications; carrying a revolving balance on a credit card may be a closer connection to risky credit card behavior in ways that owning a credit card, even multiple credit cards, may not be. If a college student were to own several credit cards but pay their credit balances off every month, they may experience a financial privilege from access to such credit that would not be the experience of students who become financially strapped because of growing credit card debt.

Similar to Table A.6, Table A.7 shows that students from the highest socioeconomic category are less likely to carry a revolving balance than students from the Middle class category. Differences in students’ work status continue to be significant predictors of college student credit card behavior across all three models. Students who work more than 20 hours per week are more likely to carrying a credit balance from month to month, compared to students who work 1-20 hours a week. Students who do not work are less likely to carry a revolving balance than students who work 1-20 hours a week. Students with no student loans and students with no Pell Grants are less likely to

44 carry a credit card balance from month to month. Students with high levels of student loans and Black students are more likely to carry a revolving credit balance.

Discussion

The analyses conducted in this paper reveal several important findings. Namely, college students from disadvantaged backgrounds (e.g., lower socioeconomic backgrounds, first generation college students, and non-White students) are more likely to exhibit riskier credit card behavior like owning two or more credit cards and carrying a revolving credit balance from month to month. This finding confirms prior research that focused on demographic differences in college student credit card ownership and use

(Lyons 2004, 2008).

Perhaps the most interesting finding of this paper is that college students’ employment status is the most consistent indicator of credit ownership and use patterns among college students. Students who work more than 20 hours per week are more likely to own a credit card, more likely to own two or more credit cards, and more likely to carry a revolving credit balance from month to month.

While this paper does not test how exactly work is related to college student credit card ownership and use, it brings to light an interesting topic in the field of higher education and the broad topic of students finding ways to pay for college. It is unclear in these models, due to the limitations of the data in the ELS, whether students are owning and using credit cards because they feel more confident in their ability to pay back their debts since they work more, or if they work more because they find themselves in

45 financially precarious situations where they owe a substantial amount of money on their credit cards. Recent research talks at length about the many ways college students try to pay for college, including resorting to credit cards to bridge budget gaps (Manning 2000;

Manning and Kirshak 2005) and working during college to pay for the tuition and other college-related expenses (Goldrick-Rab 2016).

Without having qualitative information on college student decisions to use credit cards, for what reasons, and why they work while in college, it is impossible to answer some of the questions posed as a result of the findings in this paper. It is important to note, however, that even though many of the statistically significant independent variables in the demographic analyses lose their statistical significance when added to the cumulative logistic regression models, we should take heed that underrepresented college student groups and students from disadvantaged backgrounds are most likely to practice risky credit card behaviors. While prior research with similar findings has concluded that this calls for a concerted effort to develop financial counseling resources for students on campus (Jamba-Joyner et al. 2000), research has not yet seriously considered that these students might likely be in these kinds of situations because they have lower financial resources to begin with and credit cards are an easy, accessible option to get needed money, and fast.

There are a myriad of concerns in the research about students from disadvantaged backgrounds navigating the hierarchy of postsecondary education and advocating for higher education administrators and practitioners to look for ways to support these students in their attempt to earn a college degree (Mayhew, Rockenbach, and Bowman

2016). Part of supporting these students is offering them the financial support they need to pay for college so that they do not feel the need to work more hours in a job that may distract from their studies and ultimately lead to attrition from college. This paper has likely posed more questions than it has answered, but it finds interesting patterns that show credit card ownership and use as a salient issue for college students, one that higher education practitioners should consider more closely when thinking about how to support these students.

Conclusion

Further research is needed in many areas of this topic of study. Without good qualitative data on the interplay between college students’ credit card ownership and use and the ways they might navigate paying for college, we are limited in our ability to speak to student financial decisions and struggles while in college. How do college students go about the process of deciding to use their credit card in the face of financial trouble, or even to use their credit card as a way to access goods or a type of lifestyle that would have been otherwise unattainable? How do these decisions intersect with their understanding of the role credit cards play in their ability to pay for college in the context of a myriad of financial instruments such as student loans? And how do students from disadvantaged backgrounds navigate the tricky financial waters of using credit cards to better their life chances, with the distinct possibility that it could backfire on them? These are a handful of questions that should be pursued by additional research.

Recent legislation has tried to protect college students by limiting the ability of credit card companies to solicit to college students (Hawkins 2012), but it is unclear whether this legislation is making a difference for college students who continue to sign up for credit cards (Matthews 2013). Despite this kind of legislation, credit cards are increasingly integrated into American life, and this includes the college experience. No matter the future of credit cards on campus, it is vital that we research and learn more on this topic in order to support students as best as we can through financial decisions and patterns of credit use that may be riskier than a student initially thinks.

For scholars and higher education practitioners interested in the role of work in the life of a college student, this paper gives a glimpse into the importance that employment status may play in influencing a college students’ decision to own and use a credit card. Furthermore, as an increasing number of nontraditional students attend higher education, working while in college will continue to be a widely accepted part of the college experience. This paper shows that working more than 20 hours per week is associated with riskier forms of credit card ownership and use and prior work shows that working too much in college can hinder a student’s progress to the degree (Mayhew et al.

2016).

Future research and higher education practice should take steps to address the interplay between college students’ financial actions like working or signing up for a credit card. And in the midst of this examination, scholars, practitioners, and educators should pay heed to the differences that may be apparent in these situations between students from more privileged backgrounds and those from less privileged backgrounds.

Investigating these questions will ultimately help us support college student learning and success.

CHAPTER 3: Paying for Tuition with a Credit Card

Introduction

Over the course of the 20th century, a bachelor’s degree has become an essential credential to have in order to land a good job (Hout 2012; Rosenbaum 2004). However, over the past three decades, college tuition has increased while government funding for higher education has declined (Mitchell and Leachman 2015). In the place of government funding for higher education is increased access to financial vehicles like student loans to pay for college (Goldrick-Rab 2016). At the crossroads of these changes exist prospective college students vying to gain access to a better life by financing their degrees via credit

(e.g., student loans, credit cards).

Recent work draws attention to college student credit card use to pay for educational expenses (Nellie Mae 2005). In addition to paying for educational expenses like textbooks and school supplies, students also use credit cards to fund recreational, leisure, and consumer pursuits while in college (Manning 2000). Substantial research has investigated the topic of college student spending habits and credit card use (Dale and

Bevill 2007; Lyons 2004, 2007; Palmer, Pinto, and Parente 2001; Wang and Xiao 2009;

Xiao et al. 2011), much of it advocating for limiting the influence of credit card companies on college campuses (Burnsed 2010).

While the majority of college students use credit cards for educational expenses like textbooks, recent data reports that college students also use credit cards to directly fund their schooling by charging for at least some part of their tuition (Sallie Mae 2009).

Because credit cards carry a higher interest rate than student loans, and because they do not have a period of deferred payment while a student is enrolled in school, credit cards are a particularly risky method of payment that students resort to in order to attend college.

Sara Goldrick-Rab and Nancy Kendall (2006) note that when financial aid policies fall short of their intended outcomes, students are limited in their ability to pay for college, and therefore “focus their energies on working and raising funds rather than studying and attending classes,” which hurts their chances of attaining a college degree

(p. 2). Some research shows this pattern in relation to college student credit card use. As students’ credit card debt accumulates, they look to work and non-school funding sources in order to finance their education, which ultimately takes their time and attention away from the classroom and works as an obstacle to college completion (Dale and Bevill

2007; Jones 2006).

If low-income college students resort to riskier means of financing a college degree because they lack safer means for funding their education, they may experience additional obstacles to completing a college degree and establishing financial resources from which they can pull in times of financial pressure post-college. That is, students from lower socioeconomic backgrounds come into college without the kind of family financial support their higher socioeconomic peers have access to, and may use risky

51 financial tools to bridge budget gaps that might serve to further the economic disparities between these groups. For example, if paying for college tuition with a credit card creates a situation where a college student carries a high-interest balance that takes several months to pay off, or that leads to default, the student will have an even harder time establishing a good credit history and a more difficult road to economic security, with or without a college degree. And even if this college student finds a way to pay off their credit card debt through taking on more work hours, this practice may get in the way of completing the college degree, which financially hamstrings the student with respect to future earning potential.

Little is known, however, about the portion of college students who use credit cards to pay for school. In order to better understand how students navigate the postsecondary education landscape using various financial strategies, investigating college student credit card use for tuition is an important next step. Who is most likely to use credit cards to pay for college tuition? Why do students choose to use credit cards to pay for tuition; simply because they have access to credit cards, or because that is their only option?

This paper investigates college student credit card use to pay for tuition as a specific financial practice during a time of stagnating government funding of higher education and lower wages available in the labor market. Students may use credit cards to pay for tuition because of the rewards offered by banks and credit card companies for using credit cards to purchase goods, but they may also pay for tuition with a credit card if they do not have access to safer means of paying for college. To date, no empirical

52 studies have examined this financial practice, so this paper aims to provide initial exploration into this topic.

College Students, Credit Cards, and Tuition

As discussed in other parts of this dissertation, substantial research has examined how college student credit card use for consumption is associated with student attitudes toward credit cards, parental norms related to finances, employment patterns, and a number of other variables (Hancock, Jorgensen, and Swanson 2013; Hayhoe et al. 2005;

Lyons 2004, 2007, 2008; Robb and Sharpe 2009; Wang and Xiao 2009; Xiao et al. 2011,

2011). Limited research, however, has considered the role credit cards play in directly financing a student’s education through tuition and fees. Concerns have primarily focused on consumer debt as problematic in the lives of already-indebted college students. That is, educational debt is enough of a concern by itself, adding consumer debt into the mix only complicates matters. Scant research has considered students who use credit cards to pay for college tuition. This omission in the research is significant because paying for tuition with a credit card may be indicative of particularly precarious student situations.

Several recent news articles have highlighted a study by Creditcards.com that surveyed 300 colleges and universities in an attempt to understand these institutions’ approaches to letting students pay for tuition with credit cards (Gallegos 2016; Mercado

2016; Mulhere 2016). The study showed that many colleges allow students to pay for tuition with a credit card, typically with a 2.5 to 3.5 percent fee charged to the purchase in order to cover fees required by the credit card issuer. These articles concluded that paying

53 for tuition with a credit card is a bad deal because of the required fees and encouraged their readers to stay away from this financial practice.

Interestingly, not all colleges charge fees at the same rate, and many do not charge any fees when students use credit cards to pay for tuition (Gallegos 2016). According to the Creditcards.com survey, private colleges are least likely to allow their students to pay for tuition with a credit card, and public colleges are most likely to allow student to pay for tuition with a credit card for a fee. Community colleges, however, are the most likely to allow students to pay for tuition with a credit card, and often without a required fee.

One article suggests that community colleges do not charge a fee so that they can attract more students to enroll by offering more flexible opportunities to pay (Gallegos 2016).

In a survey of 800 college students between the ages of 18 and 24, Sallie Mae

(2016) reports that five percent of students use a credit card to pay for college, with an average charge amount of $1,615. In an earlier Sallie Mae (2009) report, nearly all (92%) college students used credit cards to pay for direct educational expenses like textbooks and school supplies. Furthermore, this report showed that 30 percent of their sample paid for tuition with a credit card. Students who paid for direct educational expenses with a credit card charged an average amount of $2,200.

The smattering of media coverage of this topic and the few reports produced by

Sallie Mae give us a general introduction to the issue of college students using credit cards to pay for tuition, but nearly all of the media coverage has focused on the one study by Creditcards.com and Sallie Mae reports pull from small samples, often limited to students who are borrowing private loans from Sallie Mae. In short, we know almost

54 nothing about this topic and should not jump to conclusions with the kind of data and information currently in circulation. Yes, paying an additional fee to an already high tuition bill is not a financially savvy choice, especially considering that it would likely negate any rewards a credit card company would offer for using their services to pay for college. Stopping at this conclusion, however, does not provide us with the kind of knowledge we need to understand how this financial practice fits into the structure of social inequality as manifested in higher education.

Given increasing costs associated with higher education, lower socioeconomic status students may resort to paying for some part of their college education with credit cards because they do not have as many family resources to pull from and credit cards, while riskier, are potentially a more easily accessible form of credit. Additionally, if community colleges are more likely to accept credit card payment for tuition without any additional fees, the students they serve may be more likely to pay for tuition with a credit card. However, if this riskier form of credit is associated with negative outcomes both during and after college, these students may occupy a particularly vulnerable position as they attempt to complete college and become upwardly mobile.

Credit cards, on the other hand, may provide college students access to financial resources they would not otherwise have and could aid them in their quest to pay for college. If using a credit card to pay for tuition allows a student with few other financial resources to make it through college, it might be an unrecognized positive resource for helping lower-resourced students through higher education. Either way, we know virtually nothing about this practice and this paper provides an initial exploration into

55 demographic and financial differences in this practice in order to understand how college students differentially use credit cards to pay for tuition. This paper focuses on socioeconomic differences in this practice, whether a student’s level of financial knowledge or access to financial resources predict using a credit card to pay for tuition better, and whether this practice varies between institution types with a particular focus on community colleges.

The Current Study

This project investigates which student populations are most likely to use credit cards to pay for tuition with a focus on what situations predict when students resort to this financial practice as a way to pay for college. Meaning, this paper identifies among which groups this practice is most prevalent, while also examining the situational factors that might influence whether students exhibit this kind of financial behavior. Further, this project explores students’ understanding of credit card use as a viable option for paying for college.

To be more specific, this paper investigates college student credit card use for tuition as a way to understand whether students are using credit cards because of financial restrictions such that credit cards serve as a way to bridge a necessary financial gap, or because of differences in levels of financial literacy. In other words, do financial situations or financial knowledge better explain student use of credit cards to pay for tuition? Other questions in this paper include: Who participates in this financial strategy, and in what circumstances? What are the primary reasons college students use this

56 financial strategy to fund their degrees? How do students’ levels of financial knowledge vary? And how do these variations connect to students using a credit card to pay for tuition?

This paper adds necessary information to this area of research because little research has been done on this topic and what research has been done is limited by the methodological approaches in those projects. I use two datasets to answer the range of questions posed in this paper, one is a nationally representative dataset and another is a national study on college students’ financial wellness that asks several specific questions about college students using credit cards to pay for tuition.

I predict that students’ socioeconomic status will be associated with their likelihood to use credit cards to pay for at least some of their tuition. More specifically, students from lower socioeconomic status backgrounds will be more likely to use credit cards as a way to pay for college tuition. Additionally, because of the institutional differences in which colleges allow their students to pay for tuition with a credit card without any fees, I predict that this practice will be more common among students attending community colleges. Finally, I predict students’ financial situations will predict better students’ likelihood to pay for college tuition with a credit card than their financial knowledge, though financial knowledge will play a role in influencing whether students exhibit this financial behavior.

Methods

Data

In order to examine this paper’s empirical questions, I use data from two datasets: the Education Longitudinal Study of 2002 (ELS: 2002) from the National Center for

Education Statistics (NCES) and the Study on Collegiate Financial Wellness (SCFW) from the Center for the Study of Student Life (CSSL) at The Ohio State University.

The ELS dataset is a nationally representative dataset that surveys students who were sophomores in high school in 2002 and follows them as they go to college, transition into the workforce, and make other life transitions. The ELS spans ten years from 2002 to 2012. NCES sampled over 15,000 students in 750 high schools in 2002 to achieve a nationally representative sample of high school sophomores. In 2004, NCES followed up with the base year respondents and surveyed several additional respondents in order to maintain a nationally representative sample and to compensate for attrition.

The ELS are an excellent source for learning more about the topic of college student credit card use to pay for tuition because of their national scope and their relatively recent collection. Analyses using the ELS will provide a general overview of this financial practice before diving deeper into some of the empirical questions of this paper with the

SCFW.

The study sample from the ELS (n = 3,321) includes respondents who were enrolled in a postsecondary institution in 2006 and who own a credit card. Respondents in this sample have information on whether they used a credit card in their name to pay for tuition, as well as information on their employment status, financial aid, what kind of postsecondary institution they first attended after high school, whether they delayed entry into college, and several demographic indicators.

The SCFW is a multi-institutional study administered in February 2017. The

SCFW collects data from institutions of higher education across the United States via an online survey. The survey has three primary foci:

1. “How are financial attitudes (including stress), financial behavior, and financial

knowledge related to academic success, decisions to borrow, and career selection?

2. How is student loan debt related to the issues of student financial stress,

enrollment success, decisions to borrow, career selection, investment in

education?

3. What factors (e.g. self-efficacy, financial knowledge, ability to repay, financial

behaviors, family socioeconomic status, seeking financial advice) moderate the

relationships outlined in questions 2 and 3?” (SCFW 2016).

The SCFW was administered on 85 campuses across 64 higher education institutions, from community colleges to 4-year universities. The institutions in the study are primarily public institutions. One particular benefit of using the SCFW to answer this paper’s analytical questions is that the survey instrument contains several questions catered to understanding more about college student credit card use to pay for tuition. In my role as an employee at CSSL, I was given the opportunity to develop a module of questions related to student credit card use, including credit card use for tuition. In this way, the SCFW is an ideal complement to the ELS analyses because it provides nuanced information that the ELS does not contain about its respondents. Furthermore, the SCFW includes excellent questions about students’ financial situations, how they pay for their education, and their financial knowledge.

The study sample from the SCFW (n = 5,360) includes respondents who were enrolled in one of the postsecondary institutions surveyed in February 2017 and who had information on whether they used a credit card to pay for tuition as well as information on their financial situation in college, their level of financial literacy, financial aid indicators, and several other independent variables that comprise the final empirical model.

Variables

The ELS data ask about college students’ use of credit cards to pay for tuition.

The ELS question that asks students whether they use a credit card to pay for tuition was asked only to students who reported they owned a credit card in their own name. Similarly, the SCFW question that asks students whether they use a credit card to pay for tuition was asked to students who reported they use credit cards as a way to pay for general college expenses. This paper’s analyses primarily investigate the difference in who pays for tuition with a credit card among students who own a credit card. While this group of students is a restricted group from all college students, whether or not they own a credit card, it is an appropriate sample to use within the context of this paper’s analyses

60 because this paper focuses on this particular financial strategy and not the broader topic of credit card ownership among college students.

Independent variables in the ELS analyses include the respondent’s socioeconomic status, whether they first attended a 4-year postsecondary institution, the amount of hours they worked per week during the 2005-2006 academic year, whether they delayed entry into college after high school graduation by seven months or more, race, gender, whether they are a first generation college student, whether they are married or have been previously married, whether they have a biological child, total amount of student loans, total amount of Pell Grants, and whether parent loans are paying for part of their college education.

“Middle” and “Upper.” This operationalization is largely for making the analytical models as parsimonious as possible.

I collapse two separate variables that describe respondents’ marital status and whether they are parents into one family status variable, coded as a dichotomous variable

61 where 1 means that the respondent is currently married, has formerly been married, and/or has a biological child by 2006.

Similar to the ELS analyses, the main variable of inquiry in the SCFW data is whether students use a credit card to pay for tuition. The SCFW asks this question to all students who indicated that they use credit cards to pay for at least some of their educational expenses. Because not everyone indicated that they use credit cards to pay for educational expenses, fewer respondents were asked the tuition-specific question, limiting the initial sample who answered this question. However, I combined the tuition- specific variable with another question asking students whether they owned a credit card and marked students who owned a credit card but did not indicate they used a credit card to pay for educational expenses as students who did not use a credit card to pay for tuition. In this way, this variable in the SCFW analyses reflects the same format of the credit card for tuition variable in the ELS analyses, containing only students who report that they own a credit card.

Independent variables in the SCFW analyses include students’ access to emergency cash if needed, students’ experiences of financial difficulties in college, whether students worry about having enough money to pay for college, and several questions that test students’ financial knowledge. Control variables in the SCFW include whether the student has been offered or received a Pell Grant, total student loan amount borrowed, work hours per week, whether the student is considered a dependent student of their parents for federal student aid, whether student is financially responsible for a child, spouse, or other family member, respondent’s current income, race, gender, and whether

62 respondent attends a 4-year institution. Complete variable descriptions and descriptive statistics for all variables can be found in Appendix B.

Statistical Analyses

Similar to the structure of Chapter 2’s analyses, this paper’s analyses will begin with several bivariate cross-tabulations and Pearson chi-squared tests using the ELS data to give a descriptive picture of who uses a credit card to pay for tuition. After these initial descriptive analyses, I conduct a logistic regression model using the variable asking whether students use a credit card to pay for tuition as the outcome variable of interest.

That is, the logistic regression model tests which of the descriptive analyses remain statistically significant predictors of whether students use credit cards to pay for tuition, controlling for the variety of other variables considered in the initial descriptive analyses.

Next, I use SCFW data to bring additional nuance to the empirical analyses of this paper. After describing broad patterns in college student credit card use for tuition, I use the SCFW questions that directly ask students who pay for tuition with a credit card why they engage in this financial strategy to get at information about this topic that the ELS cannot provide. Furthermore, logistic regression analyses with the SCFW data will test whether students’ financial knowledge or financial situations better predict the outcome of whether a student pays for tuition with a credit card.

All multivariate regression models in this paper have been tested for multicollinearity concerns using mkcorr and collin commands in Stata. These tests reported no issues related to multicollinearity. Correlation matrices for each regression model in this paper are presented in Appendix D.

Limitations

Both datasets used in this study contain unique limitations which hopefully balance out by using both datasets in one empirical study. These limitations are noted here.

The ELS data span ten years (2002-2012) but focus their credit card questions in

2006 to capture college student credit card use. Because these credit card questions are specific to this year, the results reflect college student credit card use during a specific timeframe that may have changed in the last ten years. For example, the Credit CARD

Act of 2009 (Hawkins 2012; Matthews 2013) introduced legislation aimed at restricting the influence of credit card companies on college campuses and limiting college student access to credit cards until they are 21. The ELS data on college student credit card use for tuition, collected three years before the Credit CARD Act of 2009, may reflect patterns of college student credit card use that have changed since the introduction of this new legislation.

The SCFW data provide a much more recent picture of college student credit card use, having been collected in February 2017, but do not reflect patterns within a nationally representative dataset. The SCFW collected data from 63 institutions across the United States, but these institutions all self-selected into participating in the study, limiting the claims the results from these data can make on the national scope of college students paying for tuition with a credit card.

Taken together, these datasets offer an ideal comparison for the study. The ELS data give us a broad idea of college student credit card use to pay for tuition while the

SCFW data give us a nuanced, and recent description of why students engage in this financial strategy. Furthermore, despite the limitations inherent in these two datasets, this study is the first large scale empirical study to investigate this topic and provides a much more trustworthy empirical investigation than the news coverage and small surveys related to this topic to date (Gallegos 2016; Mercado 2016; Sallie Mae 2009, 2016). This study makes a needed contribution to understanding this topic with reliable data sources in order to extend current work on college student credit card use that has to date ignored the specific practice of college students using credit cards to pay for tuition.

Results

The following results show several interesting patterns in both the ELS and

SCFW datasets that suggest college students from lower socioeconomic backgrounds and students in financially precarious situations are more likely to pay for tuition with a credit card. The ELS analyses offer a broad examination of national trends in college student credit card use to pay for tuition while the SCFW analyses provide additional nuance to the ELS findings and better data on students’ financial situations to show when students are most likely to engage in this financial practice.

The final logistic regression models in each set of analyses show that students from disadvantage backgrounds, students in financially precarious situations, and students who work a high amount while in college are more likely to use a credit card to pay for tuition. The SCFW analyses give detail about students’ financial situations and show that students who report being likely to come up with $400 cash in an emergency,

65 students who have not had financial difficulties at their current institution, and students who do not worry about having enough money to pay for school are all less likely to pay for tuition with a credit card than their peers in more precarious financial situations.

Furthermore, in the final SCFW logistic regression model, students’ financial knowledge is not a significant predictor of whether students use a credit card to pay for tuition.

These findings support hypotheses earlier in this paper that suggested that paying for tuition with a credit card is more commonly done in financially precarious situations and may be more particular to students who come from lower-resourced backgrounds.

And despite the widespread calls for additional financial literacy training on college campuses, these findings suggest that students’ level of financial knowledge is not as likely of a predictor of risky credit card practices as are the students’ financial situations.

Analyses Using ELS Data

Figures B.1 through B.8 show statistically significant demographic patterns in college student credit card use to pay for tuition. All of these figures were tested in a bivariate comparison within a cross-tabulation using Pearson chi-squared tests for significance. For the overall sample, about 16 percent of students who own a credit card use a credit card to pay for tuition.

Figure B.1 shows differences in whether college students use credit cards to pay for tuition by socioeconomic status. See Figure B.1 in Appendix B.

Students from lower socioeconomic backgrounds pay for college tuition with a credit card at higher rates than students from middle and upper class backgrounds. Over a fifth of the students from the lower class background in the analytic sample reported

66 using credit cards to pay for tuition. This finding supports this paper’s hypotheses that this practice would be more represented among students from lower socioeconomic backgrounds.

Figure B.2 shows differences in whether college students use credit cards to pay for tuition by employment status. See Figure B.2 in Appendix B.

Students who worked more hours per week during the 2005-2006 academic year are more likely to report using a credit card to pay for tuition. Specifically, students who report working more than 20 hours per week pay for tuition with a credit card at higher rates than students who do not work and students who work 20 hours or less per week.

The percentage of students who pay for tuition with a credit card is almost the same between students who do not work and students who work 20 hours or less per week.

This pattern suggests that working more than 20 hours per week is uniquely associated with higher rates of credit card use to pay for tuition in this comparison.

Figure B.3 shows differences in whether college students use credit cards to pay for tuition by institution type first attended. See Figure B.3 in Appendix B.

As expected, students who start at a less than 4-year postsecondary institution are more likely to pay for tuition with a credit card. Nearly a quarter of students who start at a less than 4-year institution use a credit card to pay for tuition. This finding supports earlier hypotheses that this financial strategy would be more highly represented in less than 4-year institutions like community colleges.

Figure B.4 shows differences in whether college students use credit cards to pay for tuition by first generation college student status. See Figure B.4 in Appendix B.

First generation college students are more likely that non-first generation college students to use credit cards to pay for tuition. While this paper has not explicitly focused discussion on first generation students’ approaches to paying for college, this finding suggests that this financial practice may be more highly represented in disadvantaged student groups, broadly defined.

Figure B.5 shows differences in whether college students use credit cards to pay for tuition by whether students pay for college using parent loans. See Figure B.5 in

Appendix B.

Students who use parent loans to pay for their college tuition in 2006 are less likely to use credit cards to pay for their tuition. Students without access to parent loans to help pay for their college are more likely to pay for tuition with a credit card. While the causal arrow between these two variables cannot be clearly defined in a bivariate analysis, this finding supports this paper’s hypotheses that suggest students will be more likely to pay for tuition using a credit card if they have access to fewer financial resources.

Figure B.6 shows differences in whether college students use credit cards to pay for tuition by total Pell Grant amount. See Figure B.6 in Appendix B.

Students who have a moderate amount of Pell Grants by 2012 are the most likely to have paid for tuition using a credit card, compared to students with no Pell Grants and students with a high amount of Pell Grants. Students with no Pell Grants are the least likely to pay for tuition with a credit card, and students with a high amount of Pell Grants are less likely to use a credit card to pay for tuition compared to students with a moderate

68 amount of Pell Grants, but not by much. Because Pell Grants are granted to students with the highest amount of financial need, this finding also supports the idea that students are more likely to pay for tuition with a credit card when they have access to fewer financial resources. Furthermore, the decline in this financial practice from the moderate Pell Grant category to high Pell Grant category may suggest that students with access to more financial support through Pell Grants do not resort to paying for college with a credit card as much. This implication, however, is preliminary until examined in more rigorous statistical analyses.

Figure B.7 shows differences in whether college students use credit cards to pay for tuition by race. See Figure B.7 in Appendix B.

In general, non-White students are more likely to use a credit card to pay for tuition than White students. As Figure B.7 shows, there are differences in the percentage of students paying for tuition with a credit card in each category, with Asian and Hispanic students most likely to pay for tuition with a credit card.

Figure B.8 shows differences in whether college students use credit cards to pay for tuition by family status. See Figure B.8 in Appendix B.

Students who have been previously or are currently married and/or have a biological child are more likely to pay for college tuition with a credit card. This finding is statistically significant at the p < .05 level, but it is important to note that very few students are represented in this category (n = 114).

Figures B.1 through B.8 have given a general idea of how paying for college tuition with a credit card varies between demographic groups. I ran three other cross-

69 tabulation models for students who delayed entry into college by seven months or more, gender, and total student loan amount. None of these models showed statistically significant differences in whether students used credit cards to pay for tuition along these lines.

In order to test these general demographic patterns in a more rigorous statistical model, I use a logistic regression analysis that includes all of these variables as independent variables predicting the dichotomous outcome variable of whether students use a credit card to pay for tuition. Table B.3 reports the results from this logistic regression model. See Table B.3 in Appendix B.

When all of the variables considered in the cross-tabulation analyses are added into the same logistic regression model, only two variables retain their statistically significant association with the outcome variable of whether students use a credit card to pay for tuition: socioeconomic status and employment status. Students from the highest socioeconomic background are the least likely to use a credit card to pay for tuition.

Students who work more than 20 hours per week are the most likely to use a credit card to pay for tuition. These findings support this paper’s hypotheses that students from lower socioeconomic backgrounds would be more likely to use credit cards to pay for tuition and that students from situations with fewer financial resources—possibly the reason why some students need to work more than 20 hours per week while in college—would be more likely to engage in this financial strategy.

Definitive conclusions from these findings, cannot be made because the ELS lacks qualitative data describing why students use credit cards to pay for tuition. I turn

70 now to statistical analyses using the SCFW data to gain a better understanding of the reasons why students engage in this financial strategy and whether financial situations or financial knowledge better explains college student credit card use to pay for tuition.

Analyses Using SCFW Data

Figures B.9 through B.11 show statistically significant bivariate relationships in the SCFW data between whether students pay for tuition with a credit card and a few variables that indicate their financial situation. All of these figures were tested in cross- tabulations using Pearson chi-squared tests for significance. For the overall sample, about

14 percent of students who own a credit card use a credit card to pay for tuition, similar to the findings in the ELS dataset.

Figure B.9 shows students' likelihood of paying for tuition with a credit card by their ability to access $400 cash in an emergency. See Figure B.9 in Appendix B.

Students who report they would be unlikely to come up with $400 cash in the case of an emergency are more than twice as likely to report that they pay for tuition with a credit card than students who report they would be likely to come up with $400 emergency cash.

Figure B.10 compares students’ self-reported financial difficulty at their current postsecondary institution with students’ likelihood to pay for tuition with a credit card.

See Figure B.10 in Appendix B.

Students who report that they have experienced financial difficulty at their current school are significantly more likely to say that they use a credit card to pay for tuition

71 than students who report that they have not experienced financial difficulty at their current school.

Figure B.11 compares whether students worry about having enough money to pay for school with their likelihood to report paying for tuition with a credit card. See Figure

B.11 in Appendix B.

Students who report that they worry about having enough money to pay for school are much more likely to report that they pay for tuition with a credit card than students who report that they do not worry about having enough money to pay for school.

Figures B.9 through B.11 all show that students in more financially precarious situations are more likely to pay for tuition with a credit card.

Table B.4 shows the average financial literacy scores between students who pay for tuition using a credit card and those who do not. The financial literacy score is a cumulative score between 0 and 6. As shown in Table B.4, there is no statistically significant difference between these two groups of students in terms of average financial literacy scores. See Table B.4 in Appendix B.

These results show initial answers to the question of whether a student’s financial situation or financial knowledge matter more for whether they use a credit card to pay for tuition, but further statistical examination is needed. Table B.5 reports three logistic regression models interrogating this question further. Each model in Table B.5 builds on the prior model to show how various control variables influence the outcome variable of whether students use credit cards to pay for tuition. See Table B.5 in Appendix B.

Model 1 in Table B.5 offers a basic analysis of students’ likelihood of paying for tuition with a credit card that only includes the main independent variables in the inquiry: whether the respondent is likely to come up with $400 cash in an emergency, whether the respondent has had financial difficulties while in school, whether the respondent worries about having enough money to pay for school, and the respondents’ cumulative score on the survey’s financial literacy measures. Model 1 shows that all of these variables are initially significant predictors of college student credit card use to pay for tuition.

Students who are unlikely to come up with $400 emergency cash, students who have had financial difficulties while in school, students who worry about having enough money to pay for school, and students with higher financial literacy scores are more likely to pay for tuition with a credit card.

Model 2 in Table B.5 introduces three additional variables into the model, all related to a student’s financial aid and resources. These variables include students’ current income levels, total student loan amount borrowed, and whether they have been offered or accepted a Pell Grant. Model 2 shows a decrease in the coefficient on all three financial situation variables from Model 1, but they all retain statistical significance. The financial literacy measure coefficient increases slightly but loses some of the statistical power that it held in Model 1. Students with incomes of $5,000 and more are more likely to pay for tuition with a credit card than students with an income of $1-$4,999. Students with a total student loan amount of $10,000 or more are less likely than students with total student loan amounts of $1-$9,999 to pay for tuition with a credit card.

Model 3 in Table B.5 adds in the final control variables to complete the full logistic regression model with the SCFW data. The added variables include whether the respondent is considered a dependent student, whether the respondent is financially responsible for others, race, work hours per week, and institution type. Similar to the analyses with the ELS data, students who work more than 20 hours per week are more likely than students who work 1-20 hours a week to pay for tuition with a credit card. As hypothesized, students who attend 4-year institutions are less likely to use a credit card to pay for tuition. Students who are financially responsible for others, students who are not considered dependents of their parents, students with high incomes, students who have not been offered or have not accepted Pell Grants, and non-White students are all more likely to pay for tuition with a credit card.

Interestingly, the full model retains significant relationships between the three financial situations variables at the top of the model but loses the statistically significant relationship between students’ financial knowledge and credit card use to pay for tuition.

This shift supports the hypothesis that students’ financial situations are more significant predictors of students’ likelihood to use a credit card to pay for tuition compared to students’ financial knowledge. However, I originally hypothesized that financial literacy differences between students would still retain a statistically significant relationship with students’ likelihood to pay for tuition with a credit card, even if it was not as influential of a predictor as students’ financial situations. That hypothesis is not supported by these analyses.

Reasons for Using a Credit Card to Pay for Tuition

In the SCFW survey, students who reported that they have ever used a credit card to pay for tuition were asked a follow up question, “What is the primary reason you used credit cards to pay for your college tuition?” Table B.6 offers a frequency table detailing student responses to this question. See Table B.6 in Appendix B.

Nearly 40 percent of students who use credit cards to pay for tuition report they do so because their financial package did not cover the entire cost of tuition. Students expanded on this sentiment in their write-in responses to the “Other (please specify)” option, common responses including that students had to pay for tuition with a credit card to stay enrolled while they waited for a financial aid payment to come in, that there were fees associated with tuition that financial aid did not cover, and that they had no options other than using their credit card pay for college. These results confirm the patterns laid out in the logistic regression models in Table B.5. Generally speaking, students seem more likely to pay for tuition with a credit card when they have few other financial options to pay for college.

Discussion

The analyses conducted in this paper reveal several important findings. Notably, college students from low socioeconomic backgrounds and financially fragile situations are more likely than their higher socioeconomic and financially stable peers to pay for tuition with a credit card. Even when taking into account students’ financial knowledge, financial aid variables, and other demographic variables, students’ financial situations remain strong predictors of whether students use a credit card to pay for tuition. The ELS

75 analyses show this pattern in a broad manner while the SCFW analyses give us a closer look at why students engage in this risky financial practice.

One pattern that carries over from the ELS analyses to the SCFW analyses is that students who work high hours (20 or more hours per week) while in college are more likely to engage in risky credit card practices like paying for tuition with a credit card.

This finding corroborates prior research that finds this pattern as well (Lyons 2007). The finding that college students do not need as much financial education as they do financial help to get them out of financially precarious situations runs counter to that of much of the research in this area (Hancock et al. 2013; Norvilitis et al. 2006; Robb and Sharpe

2009). That is, this paper argues for a more distinctly sociological approach to this topic, one that would assess the structural factors of funding available to students who are traversing the higher education landscape instead of social-psychological factors as the main interest of study. Recent research pushes scholarship toward this kind of approach

(Goldrick-Rab 2016) and with the ever changing structure of higher education in the

United States, we cannot lose sight of this approach.

About 15 percent of college students in the two datasets examined here pay for tuition with a credit card. This 15 percent of college students is specific to college students who own a credit card in both samples, but it is still a sizeable portion of college students participating in this practice. Furthermore, more than 15 percent of students who hold a credit card and find themselves in financially fragile situations or particular environments are engaging in this financial strategy. For example, while this paper does not focus at length on the institutional differences in this practice outside of including the

76 institution type variable in the final analytical models, nearly 30 percent of the students in the SCFW who attend a less than 4-year postsecondary institution pay for tuition with a credit card.

The students who are most likely to have trouble getting access to higher education and be able to persist through college to the degree are more likely to pay for tuition with a credit card. Because credit cards are high interest financial tools that are non-deferrable, the practice of paying direct educational expenses like tuition with such a risky financial tool suggests that these students encounter continual challenges during their time in college. If higher education is to play the role of reducing inequality in society, scholars and practitioners should pay attention to student financial behaviors as a way to pinpoint areas of growth in order to help historically underprivileged social groups access higher education and successfully navigate through it.

Conclusion

This paper is just a beginning with regard to research in the area of college student credit card use for direct educational expenses. Substantial research has been conducted on the general topic of college student credit card use, many times with methodological limitations and little attention to students using credit cards to pay for college (Hayhoe et al. 2005; Manning 2000; Manning and Kirshak 2005; Robb and

Sharpe 2009). Investigating college student credit card use to pay for tuition with two unique and well-suited datasets to answer the empirical questions of this paper is a good

77 start to expanding this field of research to help us better understand how college student credit card use fits in with the broader picture of inequality in society.

Additional research should be conducted in this area. Because of the differences in which institutions allow students to pay for tuition with a credit card without additional fees, additional research should investigate the way that community college students navigate paying for their education. Community colleges aim to provide a way for students who normally would not have access to higher education to go to college and interrogating some of these questions further would add much needed knowledge to our understanding of how to support community college students best as they navigate this unique part of the higher education institution.

Ultimately, the main conclusion of this paper is that students without financial resources to pay for college resort to financially risky behaviors to make it through college. This finding gives credence to the position that students should be recipients of increased funding for higher education and, as a system, we should think about ways to reduce the cost of attending college while increasing the financial support available to college students vying to make a better life for themselves. Particularly for students from underprivileged backgrounds, getting them financial resources at a pivotal time in college may prevent them from financial risk that could increase the obstacles they already face to finding a place in the labor market that could help them earn a decent living post- college.

CHAPTER 4: College Costs and Credit Cards: How Student Credit Card Use

Influences College Degree Attainment

Introduction

College students are on the search for convenient and accessible ways to pay for college. As the responsibility of paying for college has increasingly shifted to families, families are using credit-based strategies to finance college in order to bridge gaps in their budget between the cost of college and their financial resources. A primary example of this trend is the widespread use of student loans to finance college degrees.

Since WWII, the responsibility for paying for college has transitioned from the government to families. While government support for college funding increased through the passage of the Servicemen’s Readjustment Act of 1944 (the GI Bill) and the Higher

Education Act in 1965, it has lessened over time. For example, while Pell Grants were initially created to cover the cost of college for low-income students, virtually all current

Pell Grant recipients take out loans to cover some portion of their college costs

(Goldrick-Rab 2016). At the same time, college tuition has continued to rise, leaving an increasingly large burden of college payment on the shoulders of families. As a result, scholars increasingly worry that this shift in who pays for college has introduced significant financial precarity into the college-going process (Broton and Goldrick-Rab

2016; Goldrick-Rab and Broton 2015). Some students even sacrifice food in order to pay the costs associated with attending college (Broton and Goldrick-Rab 2015).

Loans, however, are not the only method students use to bridge monetary gaps in this area. College students are also using credit cards to pay for educational expenses

(Nellie Mae 2005). Because credit cards carry higher average interest rates than educational loans and do not allow users to defer the credit card debt they accumulate, that college students are using such high risk methods to fund their educational expenses may be an indicator of college students’ financially precarious situations. Furthermore, carrying non-deferrable, high-interest debt may have negative consequences that could exacerbate these students’ already insecure financial situations. For instance, if the immediate pressure of credit card debt dissuades students from persisting in college, they may drop out and be left with college debt but no degree. In contrast, however, access to credit may allow students from low income backgrounds to pay for important educational needs that they may have not had the resources to pay for otherwise. Access to credit might help students who might not have originally attended college due to their financial limitations to attend. Either way, the development of college student credit card access might have important implications for these students’ educational outcomes.

Considering the potential role credit cards may play in the life of college students—primarily as a factor influencing educational outcomes—investigating patterns in college student credit card use and its associated outcomes is an important endeavor. In this paper, I investigate how college students’ patterns of credit card use are related to their educational outcomes, particularly their progress to the bachelor’s degree.

Broadly, conversations about the ways in which college students finance their degrees inform our understanding of social mobility and inequality because the bachelor’s degree is an increasingly important tool for bettering life chances. Substantial research in the status attainment literature in sociology documents the important role educational attainment plays in a person’s likelihood to be upwardly mobile in society

(Beller 2009; Blau and Duncan 1967; Buchmann and Obinna 2014; Sewell et al. 1969).

Furthermore, the economic returns to a bachelor’s degree continue to outpace the economic returns to a high school diploma, making a bachelor’s degree necessary to be competitive in the labor market (Hout 2012; McCall and Percheski 2010; Rosenbaum

2004). Knowing about the ways in which students attain education helps us understand how they become socially mobile later in life and how social inequality functions in society.

If the way students finance their degree influences their likelihood to attain a degree, we might have a better understanding of the ways in which inequality is perpetuated in society through the educational attainment process. Furthermore, if we better understand how students in financially precarious situations use credit to finance necessary purchases, we might better be able to serve these students and help them acquire the appropriate resources to finance their purchases, including college.

Research in the fields of financial counseling and planning, consumer studies, public policy, sociology, and public health has explored college student credit card spending behaviors, associated health and educational outcomes, and the influence of family backgrounds on credit card use. This paper aims to extend this existing body of

81 research by considering how college student credit card use influences educational outcomes.

Gaining a deeper understanding of college student credit card use will ultimately shed light on the ways in which college students deal with economic insecurity while pursuing higher education and will inform our understanding on how methods of financing a college degree influence degree attainment. Because attaining a college degree is increasingly seen as an essential credential to land a good job (Rosenbaum

2004), disparities in degree attainment connect to disparities in income and wealth levels between social groups. While this paper focuses specifically on college student credit card use, this topic is embedded in and influenced by the broader structure of social inequality and insecurity in the United States. This paper is of interest to higher education scholars interested in how students navigate college and to sociologists interested in the ways in which outside social forces influence the educational attainment process.

College Students and Credit Card Use in the 2000s

Research has investigated the impact of increased credit access among college students by considering their credit card use (Hancock et al. 2013; Hayhoe et al. 2005;

Manning 2000). College students are taking on debt to pay for their degree and using credit to finance their purchases. This societal shift is a concern for many. Some sources have called the rampant increase in student loan use and continual rise in cumulative student loan debt a “crisis,” (Mitchell and Jackson-Randall 2012). Conversations about

82 college student credit card use share similar concerns that credit card debt only adds to the increasing debt load on students in the young adult stage of the life course.

In addition to using credit cards to pay for educational expenses like textbooks and school supplies, students also use credit cards to fund recreational, leisure, and consumer pursuits while in college (Manning 2000). Considering recent research on students experiencing financially precarious situations in college (Broton and Goldrick-

Rab 2016; Goldrick-Rab and Broton 2015), students may use credit cards to fund their purchase of food and other essential needs. Substantial research has investigated college student spending habits and credit card use (Dale and Bevill 2007; Lyons 2004, 2007;

Palmer, Pinto, and Parente 2001; Wang and Xiao 2009; Xiao et al. 2011), much of it advocating for limiting the influence of credit card companies on college campuses

(Burnsed 2010).

Research points to several reasons that college students acquire credit card debt.

Some scholars argue that students acquire credit card debt due to deficiencies in financial knowledge or literacy while others argue that students’ personality traits or attitudes toward money are related to their likelihood of using credit cards and of using them in an irresponsible manner. One study that aimed to decipher between these two lines of thinking found that financial knowledge and attitudes toward money were both significantly related to student levels of credit card debt (Norvilitis et al. 2006). Several studies consider other factors like demographic differences as potentially influential in the number of credit cards and amount of credit card debt college students acquire, but many of them suffer from omitted variable bias by not including college students’

83 financial situations in their analytical models (Hancock, Jorgensen, and Swanson 2013;

Norvilitis et al. 2006). Furthermore, the lion’s share of the current research takes for granted the assumption that credit card debt is necessarily negative, when credit cards where often used as a catalyst for economic security and mobility in the 1970s and 1980s for people without access to financial resources during hard times (Hyman 2012).

Research in this area is in need of an empirical investigation of the ways in which college student credit card use fits into the broader picture of the educational attainment process. It is insufficient to only know precursors to college student credit card use.

Without considering how this behavior connects to and influences a student’s educational attainment, we have an incomplete picture of the role credit cards play in the status attainment process. If students use credit cards as a way to make it through college and to the bachelor’s degree, acquiring credit card debt in college may not be as severe of a concern for scholars of educational inequality. However, if college students’ credit card spending patterns influence bachelor’s degree attainment—particularly if it influences it negatively—college student credit card use may be a mechanism through which social inequality in educational attainment is perpetuated. That is, if college students use credit cards to pay for educational expenses in the face of increasing financial responsibility for their education, and this financial strategy negatively influences their degree attainment chances, this may be evidence that increasing personal financial responsibility in higher education has disproportionately negative effects on those who have lower financial resources.

Attaining a College Degree: What Matters?

Extant research points to several factors that are related to a college student’s likelihood of attaining a degree. Research in this area often shows contradictory findings, but several patterns emerge that show certain factors play significant roles in whether a student attains a college degree.

Between-institution differences have been shown to matter for rates of college student degree attainment. For example, students who start college at a two-year institution are less likely than students who start college at a four-year institution to attain a bachelor’s degree. However, students who transfer from two-year to four-year schools are just as likely to attain a bachelor’s degree as those who begin their education at a four-year school. (Mayhew, Rockenbach, and Bowman 2016). Similarly, institutional quality plays a role in the bachelor’s degree attainment process. Students who attend more prestigious schools are more likely to attain a bachelor’s degree than students who attend less prestigious schools, net of other factors (Alon and Tienda 2005). Other than these major institutional differences, between-institution factors do not seem to influence degree attainment rates as much as within-college factors.

Several within-institution or student-level factors play a role in the college degree attainment process. For example, students who work more hours during college are more likely to leave school during their first year and less likely to attain a bachelor’s degree than students who work fewer hours during college (Andrews n.d.; Bozick 2007; Roksa

2011). Furthermore, students who are enrolled part-time are less likely to attain a college degree than students who are enrolled full-time (Roksa and Keith 2008).

Some research investigates the relationship between students’ participation in co- curricular activities known as “high-impact” practices and their subsequent levels of engagement (Andrews n.d.; Kuh 2008, 2009). This area of research finds that students who participate in activities like research with a faculty member, study abroad, and mentoring—among other activities—are more engaged in their college experience, and the assumption, then, is that these students will be more likely to persist to complete a college degree. However, whether student involvement in these activities actually relates to degree attainment has not been tested empirically.

Other student-level factors play a role in the degree attainment process. Mayhew,

Rockenbach, Bowman (2016, p. 416) note, “Of all within-college effects, academic achievement in college has the strongest and most consistent impact on retention, persistence, and graduation within various student populations.” That is, high-achieving students are more likely to attain a college degree compared to lower-achieving students.

Furthermore, students’ demographic characteristics are related to their likelihood of attaining a college degree. For example, students from lower-socioeconomic backgrounds, non-white students, and males are less likely to complete a bachelor’s degree than students from higher-socioeconomic backgrounds who are white and who are female (Bowen, Chingos, and McPherson 2009; Buchmann and DiPrete 2006).

Research has paid particular attention to the matter of financial factors playing a role in the college degree attainment process. Because students pay to attend college, a large amount of scholarly attention has been paid to the relationship between students’

86 degree attainment and factors like tuition costs, student loans, and financial aid packages.

I consider these findings in a distinct section because of its pertinence to this paper.

Financial Factors

While significant research has studied the connection between financial aid and college students’ entry into postsecondary education (Alon 2005; Singell and Stater

2006), as well as the connection between financial aid and student persistence in college

(Alon 2007; Goldrick-Rab et al. 2012; Haynes 2008), fewer studies have considered the relationship between financial aid and student’s likelihood of completing college. Among those that have, there are mixed findings (Mayhew et al. 2016).

The link between financial aid and college completion varies depending on the type of aid considered. For example, several studies find that grants and scholarships are a particularly effective type of financial aid for increasing students’ likelihood of completing a college degree, especially for students who typically have lower graduation rates (e.g., students from lower-socioeconomic backgrounds, students of color) (Alon

2007; Mayhew et al. 2016).

Research on the connection between student loans and college completion is mixed. Some studies show a positive effect of student loans on students’ likelihood of graduating (Jackson and Reynolds 2013; Johnson 2013), while others show a negative effect (Kim 2007; Paulsen and St. John 2002). Other studies show that college completion outcomes vary by the amount a student borrows in loans (Dwyer, Hodson, and McCloud 2013; Dwyer, McCloud, and Hodson 2012; Hu and Kramer II 2015; Zhan

2014). Several of these studies show that students who take on loan debt up to $10,000

87 experience a positive effect toward graduation, but that loan amounts above $10,000 have diminishing returns for students’ likelihood of graduating.

Goldrick-Rab, Harris, and Trostel (2009) suggest that findings in this area are conflicted because of the complicated connection between students’ background characteristics that make them differentially eligible for financial aid. Since many widely used data sets do not have good measures of both students’ background characteristics and their financial aid packages in college, studies find varying effects of financial aid on educational outcomes. Furthermore, some studies suggest that what is most important is not the amount of financial aid offered to a student, but rather the amount of tuition they are responsible for after their financial aid has been accounted for (Mayhew et al. 2016;

Paulsen and St. John 2002). That is, no matter the amount of financial aid or types of financial aid a student has access to, disparities in the gap between a student’s financial aid package and tuition costs are what influence students’ different educational outcomes.

The Current Study

This paper fills a knowledge gap in research related to college student credit card use and college student degree attainment. Research related to college student credit card use considers this behavior largely without considering the possible connection it may have with these students’ educational outcomes. Furthermore, research related to college degree attainment largely ignores the role college student credit card use may have in this process. Scant attention has been paid to the connection between college student credit card use and subsequent degree attainment.

In a dated Chicago Tribune article, a college administrator remarks, “We lose more students to credit card debt than academic failure,” (Rubin 1998). Dale and Bevill

(2007, p. 121) write that college students’ credit card debt “can lead to [students] dropping out of school,” but do not empirically test this claim. To date, one study has specifically addressed the relationship between college students’ credit card behavior and their subsequent degree attainment outcomes (Zhan 2014). Zhan finds no statistically significant relationship between students’ level of credit card debt and college graduation when taking into account students’ educational loans. However, Zhan does not take into account differences in youths’ credit card behaviors, specifically that of carrying credit card debt from month to month.

This paper adds necessary information to this conversation by 1) linking academic conversations related to college student credit card use to those related to college degree attainment and 2) extending Zhan’s (2014) investigation of this topic with additional empirical questions and an updated data source. This paper is aimed at understanding how students’ responses to increased financial responsibility in college (e.g., taking on loans, credit card debt) relate to their educational outcomes. This paper considers questions including: Are specific credit card spending strategies related to educational outcomes (e.g., having several credit cards, paying for tuition with a credit card)? How does carrying credit card debt relate to educational outcomes like bachelor’s degree attainment?

Students who carry debt across months might experience persistent financial strain as opposed to those who use credit cards and pay off their balance every month.

This strain might have an impact on students’ educational outcomes because they might make enrollment and persistence decisions based on their financial situations. While I cannot test for this mechanism in the connection between college student credit card use and degree attainment, it provides one theoretical justification for why carrying a revolving credit balance might relate to college students’ chances for attaining a bachelor’s degree.

I predict a significant relationship between college students’ credit card use and their educational outcomes. More specifically, I hypothesize that carrying credit card debt from month to month will be associated with lower rates of bachelor’s degree attainment by 2012. While I examine several types of credit card use (e.g., having multiple credit cards, paying for tuition with a credit card), I predict carrying a credit card balance from month to month will be most likely to have significant association with students’ educational outcomes.

The following sections document the empirical strategies and their subsequent results I use to interrogate these questions and shed light on the relationship between college student credit card use and their subsequent educational outcomes.

Methods

Data

I use data from the Education Longitudinal Study of 2002 (ELS: 2002) to interrogate this paper’s empirical questions. The ELS dataset is a nationally representative dataset from the National Center for Education Statistics (NCES). The

ELS follows students who were sophomores in high school in 2002 for ten years (2012) as they transition into the workforce or go to college, and as they graduate, drop out, get married, work, and make other life transitions. Due to their longitudinal nature, the ELS data are an ideal source for answering this paper’s questions. The ELS data are especially useful for this dissertation’s questions related to college students’ educational outcomes.

The study sample (n = 7,061) includes respondents who graduated from high school in 2004, enrolled in postsecondary schooling within six months of graduating high school, and were still enrolled in college in 2006. Respondents in this sample have information on credit card ownership and use, as well as degree attainment by 2012.

Primary analytical models include over 5,800 respondents, some who own a credit card and many who do not. Because some of the analyses in this paper only consider educational outcomes among students who own a credit card in their name, the final analytical models include smaller a sample size of 2,773 respondents.

Variables

The ELS data have variables that report students’ credit card ownership and patterns of use; these variables are central to the analyses of this paper. More specifically, these variables include how many credit cards a student has in their name, whether they carry a balance on their credit card from month to month, and whether the student has used their credit card to pay for tuition. These credit card variables are asked to all respondents, whether or not they own a credit card, so all of these variables offer respondents the option of marking that they do not own a credit card in their name. That is, these variables do not simply compare different credit card behaviors among students

91 who own a credit card, but they compare credit card behaviors among students who own credit cards, in reference to students who do not own a credit card. Credit card variables serve as independent variables in the analyses while variables related to students’ educational outcomes serve as dependent variables.

In order to measure students’ educational outcomes, I use ELS variables that measure the respondent’s level of education as of 2012. Respondents’ level of education by 2012 is operationalized as a dichotomous variable that divides respondents into those who have attained a bachelor’s degree and those who have not.

Important control variables from the ELS data include socioeconomic status, institution type, financial aid variables (e.g., student loan amounts, Pell Grant amounts), full-time or part-time enrollment status, family status that includes the respondent’s marital status and whether the respondent has children, and high school GPA/academic achievement variables. The ELS data provide the framework for socioeconomic status variables. Father’s level of education, mother’s level of education, family income, father’s occupation, and mother’s occupation are equally weighted and standardized to form one socioeconomic variable with four ordered quartiles. After examining cell counts in each socioeconomic quartile, I operationalized the final socioeconomic status variable as a three category variable that collapses the bottom two quartiles into one “Lower” category and translates the top two quartiles into “Middle” and “Upper.” This operationalization is largely for making the analytical models as parsimonious as possible, but it is also appropriate because after restricting the sample to immediate enrollers who were still enrolled in college in 2006, many students from the lowest

92 socioeconomic quartile were omitted from the analytical sample. Complete variable descriptions and descriptive statistics for all variables can be found in Appendix C.

Statistical Analyses

First, I offer descriptive analyses that report rates of credit card ownership and use across different groups in the sample. These descriptive analyses are meant to provide an overview of credit card behavior among the college student groups in this sample. I use basic statistical techniques like cross-tabulations and Pearson chi-squared tests to note significant patterns in the data within these analyses.

Second, I use logistic regression models to investigate which credit card behaviors are significantly related to bachelor’s degree attainment, controlling for other important variables. After examining relevant credit card use variables, I continue empirical analyses with a focus on statistically significant credit card variables from initial logistic regression models, particularly that of carrying credit card debt from month to month. All multivariate regression models in this paper have been tested for multicollinearity concerns using mkcorr and collin commands in Stata. These tests reported no issues related to multicollinearity. Correlation matrices for each regression model in this paper are presented in Appendix D.

As a way to get at the causal connection between carrying credit card debt and educational outcomes, I use propensity score matching to compare students who are similar except for whether they carry credit card debt. I run propensity score matching models for the outcome variable of whether or not students attained a bachelor’s degree by 2012. This model uses the credit card variable indicating whether students carry a

93 balance on their credit card from month to month as the treatment variable. That is, this model investigates whether carrying a credit card balance from month to month influences educational outcomes when other relevant variables are equal. This propensity score model considers a smaller sample that includes only students who own a credit card. While prior statistical analyses in this paper compare students who do not own a credit card with students who own a credit card but do not carry a revolving balance and furthermore compare those findings to students who own a credit card and carry a balance from month to month, these models compare the influence of carrying revolving credit card debt only among credit card holders.

Several studies document the reliability and use of propensity score matching methods in social science research (Melguizo, Kienzl, and Alfonso 2011; Morgan et al.

2010; Rosenbaum and Rubin 1983, 1985; Titus 2007). In sum, propensity score matching allows statistical analyses to make causal claims because the models match respondents with similar propensities to be in the treatment group but who differ in whether they are in the treatment or control group. In the context of this paper, carrying a balance on a credit card from month to month is not distributed randomly among the student population, but propensity score matching can mimic random distribution of this treatment variable and an experimental design because it matches students who are similarly likely to carry a balance on their credit card, based on a set of covariates, and makes counterfactual comparisons regarding educational outcomes between students who carry a balance and students who do not carry a balance on their credit card. I use the

Stata command psmatch2 to calculate these models and perform these analyses.

Limitations

Perhaps the primary limitation to this study is the way the sample is limited to a select group of college students: students who enrolled in college within six months of graduating from high school and who were still enrolled in college in 2006. Prior work documents the connection between delayed entry into college and degree attainment

(Bozick and DeLuca 2005; Goldrick-Rab and Han 2011; Roksa and Velez 2012), showing that students who delay entry into college are less likely to attain a bachelor’s degree than students who enroll immediately after graduating from high school. By omitting students who delay entry into college, this paper focuses on a group of students who fall into a more “traditional” college attendance pathway. Students’ college trajectories are increasingly varied and often interrupted (Goldrick-Rab 2006), and this paper may not speak to this development in college going patterns like it might if delayers were included in the analytical sample.

By restricting the analytical sample to students who enrolled in college within six months of graduating from high school and who were still enrolled in college in 2006, this paper’s findings can only speak to the relatively advantaged students who fit this criteria. Nearly 28 percent of students who started college in the fall term of 2014 failed to persist to the fall term of 2015 (National Student Clearinghouse Research Center

2016). A substantial portion of students do not make it to their second year of college, and requiring this paper’s analytical sample to still be enrolled in college in 2006 does not capture the educational outcomes of the many students who do not make it that far in college.

Despite these shortcomings, this paper provides important knowledge on the connection between college students’ credit card ownership and use and subsequent educational outcomes. In fact, restricting the analytical sample in this fashion allows the empirical models to speak with more authority on the connection between credit card use and educational outcomes. Because the ELS survey asks students about their credit card use in 2006, including students who were not enrolled in college in 2006 in the analytical sample does not appropriately reflect college student credit card use patterns.

Furthermore, limiting the sample to students who enrolled in college within six months of graduating from high school helps to standardize the time the analytical models gives students to attain a bachelor’s degree, which levels the playing field for possible educational outcomes by giving all students eight years to attain a bachelor’s degree.

These analytical decisions may limit this paper’s ability to speak to the broad range of college students attending college with varying enrollment trajectories, but it allows the empirical models to speak specifically to the connection between college student credit card use—while in college—and educational outcomes like bachelor’s degree attainment.

Results

I report results in the order of the empirical analyses as developed and described above. In sum, the descriptive analyses show that credit card ownership is more common among students from the lower socioeconomic status group, students who started at a 4- year institution, and students who work more than 20 hours per week. Carrying a balance on a credit card is more common among students from the lower socioeconomic status

96 group, students who started at a less than 4-year institution, and students who work more than 20 hours per week. When considering the descriptive connection between college student credit card use and bachelor’s degree attainment, students who carry a balance on their credit card from month to month are least likely to attain a bachelor’s degree by

2012.

This descriptive finding of the connection between a student carrying revolving credit and lower chances of attaining a bachelor’s degree will remain constant throughout the empirical models of this paper. Though I test for the connection between bachelor’s degree attainment and four different credit card ownership and use variables, carrying a credit balance from month to month is the only credit card-related variable that reports a significant relationship with bachelor’s degree attainment. This finding is true in both the logistic regression models and the propensity score matching models.

These findings support the hypotheses outlined at the beginning of this paper that carrying a revolving credit balance will have a significant relationship with students’ educational outcomes. The following sections develop this summary and present specific results in a more detailed manner.

Descriptive Analyses

Figures C.1 through C.3 show patterns in credit card ownership and behavior along important demographic differences. Namely, Figure C.1 examines differences between socioeconomic status groups; Figure C.2 examines differences between students’ work status—specifically that of how many hours they work during the school year—and Figure C.3 examines differences by the type of institution a student starts their

97 postsecondary education. Credit card ownership and behavior statistically significantly varies between all of the groups shown in Figures C.1 through C.3.

Figure C.1 shows notable socioeconomic differences, both in who owns a credit card and who carries a balance on their credit card from month to month. The majority of students from the upper socioeconomic category do not own a credit card in their name.

This group of students is also less likely to carry a balance from month to month on their credit card if they do own one. In contrast, students from the lower socioeconomic category are more likely to both own a credit card and carry a revolving balance if they own one. See Figure C.1 in Appendix C.

Figure C.2 shows the relationship between students’ level of work hours in the

2005-2006 academic year and their credit card ownership and use. Students who work more hours are more likely to own a credit card and carry a balance on their credit card from month to month. Over sixty percent of students who did not work during this academic year did not own a credit card in 2006. See Figure C.2 in Appendix C.

Figure C.3 documents differences in students’ credit card ownership and use between institution types, specifically whether or not they started their postsecondary education at a 4-year institution or a less than 4-year institution. Students who begin college at a less than 4-year institution are less likely to own a credit card than students who begin college at a 4-year institution. They are, however, more likely to carry a balance from month to month when they do own a credit card. See Figure C.3 in

Appendix C.

All of these figures have shown significant differences between demographic groups in credit card ownership and use. While these demographic differences are interesting, these patterns do not inherently tell us anything about the impact of these differences. Figure C.4 shows differences in students’ bachelor’s degree attainment by

2012, by their credit card ownership and use variables. Students who carry a balance on their credit card are less likely to attain a bachelor’s degree by 2012 compared to students who do not own a credit card and compared to students who own a credit card but do not carry a balance from month to month. See Figure C.4 in Appendix C.

Building on these initial descriptive analyses, the following sections report results from statistical models that interrogate the role of credit card ownership and use within the context of students’ educational outcomes.

Credit Card Use and Educational Attainment

Tables C.2 through C.5 report results from four logistic regression models considering the four credit card variables of interest to this paper. These models show which credit card variables are significantly related to students’ educational outcomes, specifically that of bachelor’s degree attainment, in order to isolate pertinent variables for subsequent analyses. As Tables C.2 through C.5 show, after controlling for a number of other variables, owning a credit card, using a credit card to pay for tuition, and carrying several credit cards are not significantly related to students’ bachelor’s degree attainment.2 See Tables C.2 through C.5 in Appendix C.

2 As a note, prior iterations of these—and all of the following—analytical models included measures of students’ race and gender. While there are important theoretical justifications for including these variables in these models because of the way credit card 99

Students who carry a balance on their credit card from month to month, however, are significantly less likely to attain a bachelor’s degree compared to students who own a credit card but do not carry a revolving balance. More specifically, controlling for a host of other factors, students who own a credit card and carry credit card debt from month to month are nearly three-fourths as likely to attain a bachelor’s degree as students who own a credit card but do not carry a revolving balance. There is no statistically significant difference, however, in bachelor’s degree attainment between students who own a credit and pay off their balance every month and those who do not own a credit card.

Control variables report expected findings across the four models considered in

Tables C.2 through C.5. Students from the upper socioeconomic category are more likely to attain a bachelor’s degree in 2012 than students from the middle socioeconomic category. These models find no significant difference in bachelor’s degree attainment between the lower and middle socioeconomic categories. Working more than 20 hours per week, having a biological child or being married or previously married, and being enrolled part time are related to lower chances of attaining a bachelor’s degree in 2012.

Starting at a 4-year institution and having higher high school academic performance indicators—GPA and standardized test scores—are related to higher chances of attaining a bachelor’s degree in 2012.

access developed historically (Hyman 2012), these groups consistently did not statistically vary in their bachelor’s degree attainment rates. I imagine that these models might have reported significant differences along gender and race lines if they considered data from the 1980s and possibly 1990s when these groups were receiving increased access to credit cards and growing in the amount of credit card owners represented from their respective social groups. In order to present parsimonious models, these variables are excluded from the final analytical models. 100

Interestingly, financial control variables report findings that students with no Pell

Grants or no student loans, as well as students with high amounts of Pell Grants or student loans, are both more likely to attain a bachelor’s degree than students with moderate amounts of Pell Grants or student loans. These findings may reflect the relatively high financial resources of students without Pell Grants or student loans and the effectiveness of higher Pell Grant or student loan amounts providing financial support for students in financial need during college. In contrast, having a moderate amount of Pell

Grants or student loans may indicate a financial need that is not entirely covered by these financial vehicles.

Considering these findings from the logistic regression models, the subsequent analyses exclusively utilize the variable indicating whether students carry a balance on their credit card from month to month to investigate how college student credit card spending behaviors are related to educational outcomes. That is, while research on college student credit card use comes in a variety of forms, this paper focuses specifically on the pattern of students carrying credit card debt from month to month in order to investigate whether the connection between this type of credit card use and students’ educational outcomes holds significance in more rigorous empirical models.

Table C.6 reports two logistic regression models that focus on the association between carrying a credit card balance from month to month and educational outcomes.

Model 1 is a copy of the full model in Table C.5, placed in Table C.6 in order to compare

Model 1 findings with Model 2. Model 1 includes all students while Model 2 restricts the analytical sample to students who own a credit card.

101

The primary purpose for Table C.6 is to show the connection between students’ credit card behavior and educational outcomes across different subsets of the sample. In prior iterations of Model 1, I explored statistical differences between students who carry a credit balance from month to month compared to students who do not own a credit card as a reference group. These analyses showed no significant difference between these two groups, but shifting the reference group to students who own a credit card but pay off their balances every month revealed a significant difference in bachelor’s degree attainment in 2012 between students who carry a balance and students who own a credit card but do not carry a balance. Using this finding as motivation, Model 2 investigates these patterns with a restricted sample of students who own a credit card to further interrogate the difference in bachelor’s degree attainment between students who own a credit card but have different patterns of use. See Table C.6 in Appendix C.

Model 2 reports similar findings as Model 1 and carrying a credit balance from month to month remains a statistically significant indicator of lower bachelor’s degree attainment by 2012. Model 2 reports slightly different findings across control variables.

The coefficients for students from the upper socioeconomic category, for students with no Pell Grants, and for students with student loans greater than $10,000 all increased but remained significant. The coefficients for students who work more than 20 hours per week, for students who started at a 4-year institution, for students with Pell Grants totaling more than $10,000, and for students with a high school GPA higher than 3.0 all decreased but remained significant. In Model 2, students who have a biological child, are married or have been previously married in 2006 are no longer represented as having

102 significantly different bachelor’s degree attainment rates compared to those without children and who have never been married. Similarly, students who have no student loans no longer report significant differences in bachelor’s degree attainment by 2012 compared to students with a moderate amount of student loans.

These models offer a comparison of credit card behavior among students who own a credit card in their name and provides the opportunity to investigate this difference in educational outcomes among credit card holders in a more sophisticated analytical model. The restricted Model 2 in Table C.6 sets up the next empirical investigation of this paper. In order to get a deeper understanding of the differences between these student groups, I consider comparisons in educational outcomes between students who own a credit card but do not carry a balance and students who carry a balance from month to month on their credit card using a more rigorous statistical tool: propensity score matching. Further comparing differences in credit card use among students who own a credit card might give us unique insights into the significance of carrying a revolving credit balance for students’ educational outcomes. The following section translates the models in Table C.7 from logistic regression analyses to propensity score matching models in order to investigate a causal connection between carrying a revolving credit card balance and educational outcomes.

Propensity Score Matching Models

In order to investigate the connection between students’ credit card behaviors and their subsequent educational outcomes, I use the quasi-experimental method of propensity score matching to extend the analyses and findings of this paper. Respondents

103 were matched according to the covariates in the logistic regression models and differed in whether they report that they carry a credit card balance from month to month (treatment group) or do not (control group). All students in these propensity score matching models are credit card owners.

Table C.7 reports findings from an Epanechnikov kernel propensity score matching model considering the outcome of bachelor’s degree attainment by 2012.

Kernel matching reduces bias in propensity score matching models by weighting control matches through an inverse association with the matches’ respective distances from their treatment counterparts. That is, kernel matching weights matches that are closer together more heavily so that it emphasizes the comparisons that best reflect similar cases on all covariates except for whether students carry a balance on their credit card from month to month. Because of this advantage, I report kernel matching estimates exclusively as opposed to other propensity score matching model types. See Table C.7 in Appendix C.

The treatment effect of carrying a balance on a credit card is significant and negative when considering bachelor’s degree attainment by 2012. As Table C.7 shows, the probability of attaining a bachelor’s degree by 2012 among students who carry a credit card balance from month to month is .552, compared to a .613 probability of these students attaining a bachelor’s degree had they not carried a credit card balance in college. The treatment effect of carrying a credit card balance from month to month, then, is a decrease in a student’s probability of attaining a bachelor’s degree by 6.1 percentage points.

104

Discussion

The analyses conducted in this paper reveal several important findings. Namely, that college students use credit cards is not problematic as much as how college students use credit cards. Carrying a credit balance from month to month appears to be a particularly problematic way to use credit cards. Also, even after controlling for and matching on financial aid variables like Pell Grant and student loan amounts, college student credit card use still has an impact on students’ educational outcomes. Significant attention has been paid to the influence of grants and student loans on the college-going and college-completing process (Goldrick-Rab 2016; Mayhew et al., 2016), but this paper suggests that additional financial factors like students’ credit card use should be considered in this area of research as well.

The results show consistent statistical differences in educational outcomes between students who own a credit card and carry a revolving credit balance and the two other student groups investigated in this paper: students who own a credit card and do not carry a credit balance from month to month and students who do not own a credit card.

Students who carry a revolving credit balance on their credit card are consistently less likely to attain a bachelor’s degree by 2012 compared to students who own a credit card but pay off their monthly credit balances.

This finding is potentially a cause for concern if, as shown in the descriptive analyses in Figures C.1 through C.4, disadvantaged groups are more likely to carry a credit balance from month to month. Credit cards have become a widespread method for bridging gaps in family budgets in the face of stagnating wages and labor market

105 opportunity, especially for those who have fewer financial resources (Hyman 2012;

Manning 2000). This paper shows that, at least among college students, carrying a revolving credit balance is detrimental to achieving important educational outcomes, even after controlling for financial variables like amount of Pell Grants and student loans a student has by 2012.

While I do not test questions related to why students carry a revolving credit card balance, history and research have shown that credit card companies encourage and reward credit card holders who carry a balance from month to month because they are the most profitable customers, especially if they are from lower financially-resourced backgrounds because of the higher interest rates attached to their accounts (Hyman

2012). As college students find themselves navigating the financial landscape of higher education as tuition increases are common and state funding for postsecondary education is stagnating, credit card companies have stepped in this gap in financial resources to offer easy access to needed money. This paper, however, shows that we should be concerned if students are using this credit access to fund their education, thinking that it will help them build better lives by helping them get through college, and yet reaping negative consequences from their credit card use.

Students who get into debt from college but do not attain a college degree are left between a rock and a hard place because they have debt without any ability to translate that investment into additional labor market value. Similarly with credit card debt, students who use credit cards to fund their college degree, whether directly or indirectly, might temporarily have greater access to education that they might not otherwise be able

106 to afford, but have higher chances of not being able to complete a college degree. In this case, these students are left with high-interest, non-deferrable debt that influences their ability to acquire financial resources in order to fund future important purchases, and straps their current financial options with the burden of ever growing debt.

The propensity score matching models show that the influence of carrying a credit card balance from month to month decreases a student’s likelihood of completing important educational outcomes by less than 10 percentage points, a relatively small amount, the finding is significant and robust. While this pattern in credit card use may not be one of the major factors influencing college students’ educational outcomes, it plays a role in the broader picture of the educational system and process in the United States.

Conclusion

This paper adds additional empirical knowledge to social science and education research areas interested in the overlap between college students’ financial situations and educational outcomes. As families become increasingly responsible for paying for college, research in this area is pivotal to helping scholars and practitioners understand how the changing landscape of higher education funding is influencing student outcomes.

Furthermore, the more we understand about student outcomes during this critical stage of the life course, the more we can understand about the ways in which higher education plays a role in the system of social stratification in the United States.

While this paper cannot speak to why students choose to carry a balance on their credit card from month to month—whether or not it is because students have limited

107 financial options and credit cards provide easily accessible money—it provides robust evidence that students’ financial situations impact their educational outcomes. With a particular focus on college students’ credit card use, this paper shows evidence for a link between carrying a balance on a credit card and a lower likelihood of completing a bachelor’s degree. If students from disadvantaged backgrounds resort to using credit cards and carrying credit balances that hinder their ability to complete a college degree, the broader system of higher education funding may be doing a disservice to the very students it aims to serve by offering them a chance to experience upward social mobility through a college education.

Numerous factors influence college students’ educational outcomes. Using robust statistical techniques, this paper shows that college student credit card use—particularly that of carrying a credit balance from month to month—is one of these factors. This paper’s analyses document college student credit card use in 2006, prior to the Credit

Card Accountability, Responsibility and Disclosure Act of 2009, which shifted credit card access for those under 21, making it more difficult to acquire a credit card. Further research should be conducted to see if these patterns have shifted since the legal change, and if under 21 college students are still obtaining and using credit cards at high rates because their parents are willing to co-sign for the credit cards, questions arise about how universities and researchers might educate co-signers and their students on the potential benefits and drawbacks of credit card use in college. Finally, while students who carry credit balances from month to month have bleaker educational outcomes, future research might investigate how these outcomes translate to financial behaviors and outcomes later

108 in the life course, and how they fit into the broader picture of social inequality in the

United States.

109

CHAPTER 5: Discussion and Conclusion

Attaining a bachelor’s degree increases degree holders’ labor market value and earning power and leads to a number of positive outcomes later in life (Hout 2012).

While the institution of education has often come under fire for contributing to and perpetuating inequality in society (Bourdieu and Passeron 1990; Bowles and Gintis

2002), postsecondary education has been lauded as a compensatory mechanism that reduces inequality in society. Students who are least likely to attend college are the most likely to benefit from going to college (Brand and Xie 2010). This unique contribution of higher education within the context of social inequality has led to a proliferation of research on how to help students be successful in college and attain a college degree

(Mayhew et al. 2016).

This dissertation examines questions related to college student success by investigating the specific topic of college student credit card ownership and use. The simultaneous expansion of access to consumer credit and higher education in 20th century

America has created a situation where college students are exposed to—and regularly use—credit-based financial tools like loans and credit cards as they navigate paying for college (Hyman 2012; Manning 2000; Thelin 2011). A significant amount of research has brought attention to college student loan borrowing (Addo, Houle, and Simon 2016;

110

Dwyer, Hodson, and McCloud 2013; Dwyer, McCloud, and Hodson 2012; Zhan 2014) and additional research has considered the topic of college student credit card behavior

(Hayhoe et al. 2005; Manning 2000; Manning and Kirshak 2005; Robb 2011).

These areas of research, however, often omit important considerations typically incorporated in the other research focus. For example, while research on college student loan borrowing often does a good job of incorporating variables that account for college students’ financial backgrounds and situations, this area of research rarely incorporates patterns of college student credit card ownership and use (but see Zhan 2014). And while research on college student credit card ownership and use focuses on this particular financial behavior, this area of research rarely incorporates variables that account for students’ financial resources and backgrounds.

This dissertation fills this gap in the research by examining patterns of college student credit card ownership and use within the context of varying financial situations.

In addition, this dissertation investigates college student success by connecting patterns of student credit card ownership and use to bachelor’s degree attainment. Understanding the factors that contribute to varying chances of attaining a bachelor’s degree is an important endeavor to understand the broader picture of status attainment in society, as established in prior research (Beller 2009; Blau and Duncan 1967; Buchmann and

Obinna; Sewell et al. 1969).

The main finding of this dissertation is that students from lower socioeconomic backgrounds and students in financially precarious situations are more likely to practice risky credit card behaviors such as owning 2 or more credit cards, carrying a revolving

111 credit balance from month to month, and paying for tuition with a credit card.

Furthermore, carrying a revolving credit card balance is negatively related to bachelor’s degree attainment, suggesting that college students who are already in a precarious financial situation and resort to risky credit card practices experience additional hurdles to attaining a bachelor’s degree, further reducing their chance to be upwardly socially mobile later in life.

The following sections detail specific findings from the three empirical studies in this dissertation, review the implications of these findings, and connect these findings to future research, practice, and policy.

Summary of Results

Chapter 2 of this dissertation extends prior research on college student credit card ownership and use by interrogating questions about demographic differences in credit card behaviors with a nationally representative dataset. This chapter focuses on the social factors that may contribute to differences in college student credit card ownership and use as opposed to the psychosocial factors most of the research in this area focuses on as primary predictors of college student credit card behavior.

The main findings of Chapter 2 show that students from lower socioeconomic backgrounds and students who work more than 20 hours per week are more likely to exhibit riskier credit card behavior such as owning 2 or more credit cards and carrying a revolving credit balance from month to month. Initial demographic analyses prior to the final analytic models show that several disadvantaged student groups (e.g., first

112 generation college students, non-White students) are more likely to engage in risky credit card practices. These relationships in the data drop out in the final, full analytic models while socioeconomic status and work hours maintain significant relationships to risky credit card behaviors, possibly suggesting that these disadvantaged student groups are resorting to working more as a way to access more financial resources in these situations.

Regardless of the exact mechanism behind these patterns, Chapter 2 establishes that students from disadvantaged backgrounds use credit cards in a riskier manner than students from advantaged backgrounds.

Chapter 3 of this dissertation focuses on the particular practice of college students using a credit card to pay for their tuition. This chapter extends prior research in this area in a broad sense because it is investigating the topic of college student credit card ownership and use. Furthermore, it investigates the practice of paying for tuition with a credit card as a new inquiry in this research area because minimal work has considered this financial practice as a specific outcome of interest. This chapter focuses on exploring whether college students’ financial situations or levels of financial knowledge better predict paying for tuition with a credit card. Chapter 3 uses two datasets that complement one another to bring a more robust understanding to this topic, one dataset that is nationally representative and one dataset that collects recent, detailed information on college students’ use of credit cards to pay for tuition.

The main findings of Chapter 3 show that students in financially precarious situations (e.g., students who cannot come up with $400 cash in an emergency, students who experience financial difficulties in college) are more likely to pay for tuition with a

113 credit card. What is more, these financial situation variables remain statistically significant predictors of college student credit card use to pay for tuition even after controlling for students’ level of financial knowledge. Much of the research in this area focuses on getting college students access to increased financial education but this chapter suggests that the focus should be first on resourcing students with needed financial support in order to place them in financially stable situations so they do not resort to financially risky practices that could then hinder their progress to the degree or their future wealth building potential.

Chapter 4 of this dissertation extends the empirical investigations of Chapters 2 and 3 by interrogating whether these credit card practices influence college students’ educational outcomes, namely that of bachelor’s degree attainment. While Chapters 2 and

3 document demographic differences in who is likely to participate in risky credit card practices, Chapter 4 explores whether these credit card practices have an effect on students’ likelihood to attain a bachelor’s degree in order to understand whether the demographic differences in college student credit card ownership and use matter for student success in college.

The main findings of Chapter 4 show that students who carry a revolving credit balance from month to month are less likely to attain a bachelor’s degree than students who do not carry a revolving balance. Interestingly, students who own several credit cards and who pay for tuition with a credit card do not experience a diminished likelihood of attaining a bachelor’s degree. So, the practice of carrying a revolving balance on a credit card emerges as a particularly notable indicator of risk in terms of

114 college students attaining positive educational outcomes. Certainly, these credit card practices may overlap. For example, students who pay for tuition with a credit card might pay that balance off over time instead of all at once. These findings indicate that focusing on credit card practices like carrying revolving credit is an important part of this research area, one that is often overlooked in order to focus on total college student credit card debt.

Taken together, these findings show that the ways in which college students use credit cards are connected to their financial situations and certain credit card practices lead to detrimental outcomes in terms of bachelor’s degree attainment. The following section details several implications of these findings.

Implications of Findings

The findings in this dissertation support the idea in prior research that credit serves different purposes for people along different socioeconomic positions in society

(Fligstein and Goldstein 2015). Notably, college students in financially precarious positions are more likely to engage in risky credit card practices that may lead to revolving debt that diminishes their chances of attaining a college degree, thereby limiting their future earning power in the labor market. Advantaged college students might use credit cards to establish a good credit history that will serve them well as they look to borrow for a car loan or a mortgage after college. In this way, structures of social inequality manifest in the area of college student financial decisions and influence the educational attainment process that is so crucial in the broader status attainment process.

115

Perhaps the most prominent implication of the findings in this dissertation is that increasing financial resources, particularly for students in financially precarious situations, is a must if we want to support college students toward bachelor’s degree attainment. While prior research has focused on increasing college students’ financial knowledge, this dissertation argues that the best use of financial resources to promote college student success is to directly support their education through increasing the funding available to them so they do not resort to using credit cards to bridge gaps in their budgets.

Furthermore, higher education researchers and practitioners should think about college student credit card use as an important topic of discussion with students and one that might indicate precarious financial situations. Conversations related to student loan borrowing tend to dominate discussions of college student finances and while this dissertation does not argue that the focus on student loan borrowing is inappropriate, it does argue that a focus on college student credit card ownership and use should also be included in conversations around college student finances.

In terms of policy, much of the conversation related to this issue has centered around limiting credit card companies’ influence on college campuses. While this dissertation supports this legislation in the sense that risky credit card behavior can be a hindrance to college students’ educational outcomes, it more directly argues that limiting these companies’ access to college students is only addressing a symptom of a broader problem: students do not have the needed financial resources to pay for college and its associated expenses. Policy conversations should shift from a focus on these companies

116 and turn to focus on increasing financial resources for college students, especially for those who are in precarious financial situations.

Contributions of Dissertation and Future Directions

This dissertation contributes several findings to this area of research and creates additional opportunities for research that builds on the findings herein. First, this dissertation has the methodological advantage of using a large, nationally representative data source in the ELS data to confirm some of the preliminary findings in single institution studies that show disadvantaged college students use credit cards in a riskier manner. Furthermore, it shows that research and practice should focus first on addressing students’ financial situations and resources before looking to financially educate students and build their financial knowledge.

The findings in this dissertation encourage future researchers to integrate patterns of college student credit card ownership and use into studies of college student financial decisions and educational outcomes. Furthermore, due to the limitations of the datasets in this dissertation, I only begin to scratch the surface of how students navigate financial decision-making in college. Many students in the SCFW data report using credit cards to pay for tuition because their financial aid did not cover the full cost of tuition, but do these students also take additional action to pay for college? Actions like working more or reducing their credit load so that they attend college part time? While this dissertation cannot answer these questions directly, it suggests that there is a complex decision- making process that college students navigate as they aspire to pay for an increasingly

117 expensive education in the face of diminishing financial support. Future research should interrogate how students go about these decisions and how research, practice, and policy can best support students in this process.

In addition, the resilient finding of students’ work hours being connected to their credit card ownership and practices should be interrogated further, probably with qualitative data that can speak to how students shift from several different financial strategies in their attempt to pay for college. College students are commonly working while going to class and navigating co-curricular activities. How do college students make decisions about how much they should work, and in what circumstances? When is working more hours deemed too risky such that college students decide to use their credit cards to cover expenses? And when is paying with a credit card deemed too risky such that college students work more to cover expenses? There are a variety of unanswered questions in this area of research that this dissertation suggests should be pursued in future work.

As college students work toward attaining a degree in a social environment where they encounter several financial barriers hindering their educational persistence, higher education scholars, practitioners, and policy-makers should work toward supporting these students in this mission. This dissertation suggests that marshaling financial support to students in financially precarious positions is a particularly useful method for helping these students avoid risky financial practices and attain a college degree.

118

References

Adams, Troy and Monique Moore. 2007. “High-Risk Health and Credit Behavior Among 18- to 25-Year-Old College Students.” Journal of American College Health 56(2):101–8.

Addo, Fenaba R., Jason N. Houle, and Daniel Simon. 2016. “Young, Black, and (Still) in the Red: Parental Wealth, Race, and Student Loan Debt.” Race and Social Problems 8(1):64–76.

Alon, Sigal. 2005. “Model Mis-Specification In Assessing The Impact of Financial Aid on Academic Outcomes.” Research in Higher Education 46(1):109–25.

Alon, Sigal. 2007. “The Influence of Financial Aid in Leveling Group Differences in Graduating from Elite Institutions.” Economics of Education Review 26(3):296– 311.

Alon, Sigal and Marta Tienda. 2005. “Assessing the ‘Mismatch’ Hypothesis: Differences in College Graduation Rates by Institutional Selectivity.” Sociology of Education 78(4):294–315.

Altbach, Philip G., Patricia J. Gumport, and Robert O. Berdahl. 2011. American Higher Education in the Twenty-First Century: Social, Political, and Economic Challenges. JHU Press.

Andrews, Benjamin D. Unpublished manuscript. “Delayed Enrollment and Student Involvement: Linkages to College Degree Attainment.”

Barron, John M. and Michael E. Staten. 2004. “Usage of Credit Cards Received through College Student-Marketing Programs.” Journal of Student Financial Aid 34(3):7– 26.

Beller, Emily. 2009. “Bringing Intergenerational Social Mobility Research Into the 21st Century: Why Mothers Matter.” American Sociological Review 74:507-528.

Berg, C. et al. 2010. “Health-Related Characteristics and Incurring Credit Card Debt as Problem Behaviors Among College Students.” The Internet Journal of Mental Health 6(2):1–9.

119

Bianco, Candy A. and Susan M. Bosco. 2002. “Ethical Issues in Credit Card Solicitation of College Students – The Responsibilities of Credit Card Issuers, Higher Education, and Students.” Teaching Business Ethics 6(1):45–62.

Bird, Edward J., Paul A. Hagstrom, and Robert Wild. 1999. “Credit Card Debts of the Poor: High and Rising.” Journal of Policy Analysis and Management 18(1):125– 33.

Blau, P. M. & Duncan, O. D. 1967. The American Occupational Structure. Wiley, New York.

Bourdieu, Pierre and Jean-Claude Passeron. 1990. Reproduction in Education, Society and Culture. SAGE.

Bowen, William G., Matthew M. Chingos, and Michael S. McPherson. 2009. Crossing the Finish Line: Completing College at America’s Public Universities. Princeton University Press.

Bowles, Samuel and Herbert Gintis. 2002. “Schooling in Capitalist America Revisited.” Sociology of Education 75(1):1–18.

Bozick, Robert. 2007. “Making It through the First Year of College: The Role of Students’ Economic Resources, Employment, and Living Arrangements.” Sociology of Education 80(3):261–285.

Bozick, Robert and Stefanie DeLuca. 2005. “Better Late Than Never? Delayed Enrollment in the High School to College Transition.” Social Forces 84(1):531– 54.

Brand, J. E. and Y. Xie. 2010. “Who Benefits Most from College?: Evidence for Negative Selection in Heterogeneous Economic Returns to Higher Education.” American Sociological Review 75(2):273–302.

Broton, Katharine and Sara Goldrick-Rab. 2015. “To Cut Costs, College Students Are Buying Less Food and Even Going Hungry.” The Conversation.

Broton, Katharine and Sara Goldrick-Rab. 2016. “The Dark Side of College (Un)Affordability: Food and Housing Insecurity in Higher Education.” Change: The Magazine of Higher Learning 48(1):16–25.

Buchmann, Claudia and Thomas A. DiPrete. 2006. “The Growing Female Advantage in College Completion: The Role of Family Background and Academic Achievement.” American Sociological Review 71(4):515–541.

120

Buchmann, Claudia and Denise Obinna. 2014. “Status Attainment.” In Wiley Blackwell Encyclopedia of Sociology 2nd Edition, edited by George Ritzer. Williston, VT: Blackwell Publishing.

Burnsed, Brian. 2010. “New Rules Place Barriers Between Students, Credit Card Issuers.” US News & World Report.

Carruthers, Bruce G. and Laura Ariovich. 2010. Money and Credit: A Sociological Approach. Polity.

Condron, Dennis J. and Vincent J. Roscigno. 2003. “Disparities within: Unequal Spending and Achievement in an Urban School District.” Sociology of Education 76(1):18–36.

Dale, Larry R. and Sandra Bevill. 2007. “An Analysis of the Current Status of Student Debt: Implications for Helping Vulnerable Students Manage Debt.” Academy of Educational Leadership Journal 11(2):121-127.

Downey, Douglas B., Paul T. von Hippel, and Beckett A. Broh. 2004. “Are Schools the Great Equalizer? Cognitive Inequality during the Summer Months and the School Year.” American Sociological Review 69(5):613–35.

Downey, Douglas B., Paul T. von Hippel, and Melanie Hughes. 2008. “Are ‘Failing’ Schools Really Failing? Using Seasonal Comparison to Evaluate School Effectiveness.” Sociology of Education 81(3):242–70.

Duncan, Greg J. and Richard J. Murnane. 2011. Whither Opportunity?: Rising Inequality, Schools, and Children’s Life Chances. Russell Sage Foundation.

Duncan, Greg J. and Richard J. Murnane. 2014. “Growing Income Inequality Threatens American Education.” The Phi Delta Kappan 95(6):8–14.

Dwyer, Rachel E., Randy Hodson, and Laura McCloud. 2013. “Gender, Debt, and Dropping Out of College.” Gender & Society 27(1):30–55.

Dwyer, Rachel E., Laura McCloud, and Randy Hodson. 2012. “Debt and Graduation from American Universities.” Social Forces 90(4):1133–55.

Dwyer, Rachel E. and Michael Nau. Unpublished manuscript. “The Anti-Social Safety Net: Credit Cards, Insecurity, and Inequality in the U.S.”

Eaton, Charlie et al. 2016. “The Financialization of US Higher Education.” Socio- Economic Review 1-29.

Fligstein, Neil and Adam Goldstein. 2015. “The Emergence of a Finance Culture in American Households, 1989–2007.” Socio-Economic Review 13(3):575–601. 121

Gallegos, Demetria. 2016. “Pay College Tuition With a Credit Card? Forget It.” Wall Street Journal, December 12. Retrieved January 30, 2017 (http://www.wsj.com/articles/pay-college-tuition-with-a-credit-card-forget-it- 1481511841).

Gnizak, Charles J., Robert Meier, and Jerrold Stark. 2004. “College Students and Debt: Credit Cards and Student Loans in Western Kansas.” Kansas Policy Review 26(1):22–24.

Goldrick-Rab, Sara. 2006. “Following Their Every Move: An Investigation of Social- Class Differences in College Pathways.” Sociology of Education 79(1):67–79.

Goldrick-Rab, Sara. 2016. Paying the Price: College Costs, Financial Aid, and the Betrayal of the American Dream. Chicago ; London: University Of Chicago Press.

Goldrick-Rab, Sara and Katharine M. Broton. 2015. “Hungry, Homeless and in College.” The New York Times, December 4. Retrieved November 27, 2016 (http://www.nytimes.com/2015/12/04/opinion/hungry-homeless-and-in- college.html). Goldrick-Rab, Sara and Seong Won Han. 2011. “Accounting for Socioeconomic Differences in Delaying the Transition to College.” The Review of Higher Education 34(3):423–45.

Goldrick-Rab, Sara, Doug Harris, Robert Kelchen, and James Benson. 2012. Need-Based Financial Aid and College Persistence Experimental Evidence from Wisconsin. Rochester, NY: Social Science Research Network.

Goldrick-Rab, Sara, Douglas N. Harris, and Philip A. Trostel. 2009. “Why Financial Aid Matters (or Does Not) for College Success: Toward a New Interdisciplinary Perspective.” Pp. 1–45 in Higher Education: Handbook of Theory and Research, vol. 24.

Goldrick-Rab, Sara and Nancy Kendall. 2016. “The Real Price of College.” Retrieved June 30, 2016 (http://www.mtsac.edu/president/board- reports/The_Real_Price_of_College_The_Century_Foundation.pdf).

Grable, John E. and So-hyun Joo. 2006. “Student Racial Differences in Credit Card Debt and Financial Behaviors and Stress.” College Student Journal 40(2):400–408.

Gutter, Michael and Zeynep Copur. 2011. “Financial Behaviors and Financial Well- Being of College Students: Evidence from a National Survey.” Journal of Family and Economic Issues 32(4):699–714.

122

Hancock, Adam M., Bryce L. Jorgensen, and Melvin S. Swanson. 2013. “College Students and Credit Card Use: The Role of Parents, Work Experience, Financial Knowledge, and Credit Card Attitudes.” Journal of Family and Economic Issues 34(4):369–81.

Hawkins, Jim. 2012. “The CARD Act on Campus.” Wash. & Lee L. Rev. 69:1471.

Hayhoe, Celia Ray, Lauren J. Leach, Myria W. Allen, and Renee Edwards. 2005. “Credit Cards Held by College Students.” Journal of Financial Counseling and Planning 16(1). Haynes, R. 2008. “The Impact of Financial Aid on Postsecondary Persistence: A Review of the Literature.” Journal of Student Financial Aid 37(3).

Hodson, Randy, Rachel E. Dwyer, and Lisa A. Neilson. 2014. “Credit Card Blues: The Middle Class and the Hidden Costs of Easy Credit.” The Sociological Quarterly 55(2):315–40.

Houle, Jason N. 2014. “A Generation Indebted: Young Adult Debt across Three Cohorts.” Social Problems 61(3):448–65.

Houle, Jason N. and Lawrence Berger. 2015. “Is Student Loan Debt Discouraging Homeownership among Young Adults?” Social Service Review 89(4):589–621.

Hout, Michael. 2012. “Social and Economic Returns to College Education in the United States.” Annual Review of Sociology 38(1):379–400.

Hu, Xiaodan and Dennis A. Kramer II. 2015. “The Impact of Student Loans on Baccalaureate Degree Attainment: A Gender Perspective.” Paper presented at the 40th Annual Meeting of the Association for Education Finance and Policy, February 14, Washington, DC.

Hyman, Louis. 2012. Debtor Nation: The History of America in Red Ink. Reprint edition. Princeton University Press.

Jackson, Brandon A. and John R. Reynolds. 2013. “The Price of Opportunity: Race, Student Loan Debt, and College Achievement.” Sociological Inquiry 83(3):335– 68.

Jamba-Joyner, Lisa, Mary Howard-Hamilton, and Helen Mamarchev. 2000. “College Students and Credit Cards: Cause for Concern?” Journal of Student Financial Aid 30(3). Retrieved (http://publications.nasfaa.org/jsfa/vol30/iss3/2).

Johnson, Matthew T. 2013. “Borrowing Constraints, College Enrollment, and Delayed Entry.” Journal of Labor Economics 31(4):669–725.

123

Jones, Joyce E. 2006. “College Students’ Knowledge and Use of Credit.” Journal of Financial Counseling and Planning 16(2). Retrieved June 22, 2016 (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2248804).

Kantrowitz, Mark. 2016. “Why the Student Loan Crisis Is Even Worse Than People Think.” Time, January 11. Retrieved October 4, 2016 (http://time.com/money/4168510/why-student-loan-crisis-is-worse-than-people- think/).

Karger, Howard. 2005. Shortchanged: Life and Debt in the Fringe Economy. San Francisco, CA : Berkeley, CA: Berrett-Koehler Publishers.

Kim, Dongbin. 2007. “The Effect of Loans on Students’ Degree Attainment: Differences by Student and Institutional Characteristics.” Harvard Educational Review 77(1):64–100.

Krippner, Greta R. Forthcoming. “Democracy of Credit: Ownership and the Politics of Credit Access in Late-Twentieth Century America.” American Journal of Sociology.

Kuh, George D. 2008. High-Impact Educational Practices: What They Are, Who Has Access to Them, and Why They Matter. Washington, D.C.: Association of American Colleges & Universities.

Kuh, George D. 2009. “What Student Affairs Professionals Need to Know About Student Engagement.” Journal of College Student Development 50(6):683–706.

Lusardi, Annamaria, Olivia S. Mitchell, and Vilsa Curto. 2010. “Financial Literacy among the Young.” Journal of Consumer Affairs 44(2):358–80.

Lyons, Angela C. 2004. “A Profile of Financially At-Risk College Students.” Journal of Consumer Affairs 38(1):56–80.

Lyons, Angela C. 2007. “Credit Practices and Financial Education Needs of Midwest College Students.” Networks Financial Institute Working Paper (2007–WP):23.

Lyons, Angela C. 2008. “Risky Credit Card Behavior of College Students.” Pp. 185–207 in Handbook of consumer finance research. Springer. Retrieved June 22, 2016 (http://link.springer.com/chapter/10.1007/978-0-387-75734-6_11).

Manning, Jack. 1991. “Credit Cards Become Big Part of Campus Life.” The New York Times, February 9. Retrieved February 18, 2017 (http://www.nytimes.com/1991/02/09/education/credit-cards-become-big-part-of- campus-life.html).

Manning, Robert D. 2000. Credit Card Nation. Reprint edition. Basic Books. 124

Manning, Robert and Ray Kirshak. 2005. “Credit Cards on Campus: Academic Inquiry, Objective Empiricism, or Advocacy Research?” Journal of Financial Aid. Retrieved (http://scholarworks.rit.edu/article/560).

Matthews, Mary Beth. 2013. “The Credit CARD Act of 2009 – Four Years Later | Arkansas Law Notes.” Retrieved February 2, 2017 (http://media.law.uark.edu/arklawnotes/2013/08/23/the-credit-card-act-of-2009- four-years-later/).

Mayhew, Matthew J., Alyssa N. Rockenbach, and Nicholas A. Bowman. 2016. How College Affects Students : 21st Century Evidence That Higher Education Works. Newark, US: Wiley.

McCall, Leslie and Christine Percheski. 2010. “Income Inequality: New Trends and Research Directions.” Annual Review of Sociology 36:329–47.

Melguizo, Tatiana, Gregory S. Kienzl, and Mariana Alfonso. 2011. “Comparing the Educational Attainment of Community College Transfer Students and Four-Year College Rising Juniors Using Propensity Score Matching Methods.” The Journal of Higher Education 82(3):265–91.

Mercado, Darla. 2016. “Using Your Credit Card to Pay College Tuition Is a Bad Deal.” CNBC, August 24. Retrieved February 28, 2017 (http://www.cnbc.com/2016/08/24/dont-use-your-credit-card-to-pay-college- tuition.html).

Mitchell, Josh and Maya Jackson-Randall. 2012. “Student-Loan Debt Tops $1 Trillion.” Wall Street Journal, March 22. Mitchell, Michael and Michael Leachman. 2015. “Years of Cuts Threaten to Put College out of Reach for More Students.” Center on Budget and Policy Priorities 1–26.

Morgan, Paul L., Michelle L. Frisco, George Farkas, and Jacob Hibel. 2010. “A Propensity Score Matching Analysis of the Effects of Special Education Services.” The Journal of Special Education 43(4):236–54.

Mulhere, Kaitlin. 2016. “Should You Use a Credit Card to Pay College Tuition?” Money. Retrieved February 28, 2017 (http://time.com/money/4463405/credit-card-pay- college-tuition/).

National Student Clearinghouse Research Center. 2016. First-Year Persistence and Retention Rates, Fall 2015. Retrieved February 15, 2017 (https://nscresearchcenter.org/snapshotreport-persistenceretention22/).

125

Nau, Michael, Rachel E. Dwyer, and Randy Hodson. 2015. “Can’t Afford a Baby? Debt and Young Americans.” Research in Social Stratification and Mobility 42:114– 22.

Nellie Mae. 2005. “Undergraduate Students and Credit Cards in 2004.” Undergraduate Students and Credit Cards in 2005: An Analysis of Usage Rates and Trends. Retrieved June 22, 2016.

Nelson, Melissa C., Katherine Lust, Mary Story, and Ed Ehlinger. 2008. “Credit Card Debt, Stress and Key Health Risk Behaviors among College Students.” American Journal of Health Promotion 22(6):400–407.

Norvilitis, Jill M. et al. 2006. “Personality Factors, Money Attitudes, Financial Knowledge, and Credit-Card Debt in College Students.” Journal of Applied Social Psychology 36(6):1395–1413.

Norvilitis, Jill M. and Michael G. MacLean. 2010. “The Role of Parents in College Students’ Financial Behaviors and Attitudes.” Journal of Economic Psychology 31(1):55–63.

Palmer, Todd Starr, Mary Beth Pinto, and Diane H. Parente. 2001. “College Students’ Credit Card Debt and the Role of Parental Involvement: Implications for Public Policy.” Journal of Public Policy & Marketing 20(1):105–113. Paulsen, Michael B. and Edward P. St. John. 2002. “Social Class and College Costs: Examining the Financial Nexus Between College Choice and Persistence.” The Journal of Higher Education 73(2):189–236.

Robb, Cliff A. 2011. “Financial Knowledge and Credit Card Behavior of College Students.” Journal of Family and Economic Issues 32(4):690–98.

Robb, Cliff A. and Deanna L. Sharpe. 2009. “Effect of Personal Financial Knowledge on College Students’ Credit Card Behavior.” Journal of Financial Counseling and Planning 20(1). Retrieved June 22, 2016 (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2224225).

Roksa, Josipa. 2011. “Differentiation and Work: Inequality in Degree Attainment in U.S. Higher Education.” Higher Education 61(3):293–308.

Roksa, Josipa and Bruce Keith. 2008. “Credits, Time, and Attainment: Articulation Policies and Success after Transfer.” Educational Evaluation and Policy Analysis 30(3):236–54.

Roksa, Josipa and Melissa Velez. 2012. “A Late Start: Delayed Entry, Life Course Transitions and Bachelor’s Degree Completion.” Social Forces 90(3):769–94.

126

Rosenbaum, James E. 2004. Beyond College For All: Career Paths for the Forgotten Half. Russell Sage Foundation.

Rosenbaum, Paul R. and Donald B. Rubin. 1983. “The Central Role of the Propensity Score in Observational Studies for Causal Effects.” Biometrika 70(1):41–55.

Rosenbaum, Paul R. and Donald B. Rubin. 1985. “Constructing a Control Group Using Multivariate Matched Sampling Methods That Incorporate the Propensity Score.” The American Statistician 39(1):33–38.

Rubin, Bonnie Miller. 1998. “College Students Charge Right Into Valley Of Debt.” Chicago Tribune, August 16.

Saez, Megen. 2008. “Cashless College: Credit Card Debt Among College Students and the Leadership Role of Academic Institutions.” Dissertation, University of Phoenix.

Sallie Mae. 2009. “How Undergraduate Students Use Credit Cards.” Retrieved June 22, 2016 (http://static.mgnetwork.com/rtd/pdfs/20090830_iris.pdf).

Sallie Mae. 2016. How America Pays for College 2016. Retrieved February 28, 2017 (http://news.salliemae.com/files/doc_library/file/HowAmericaPaysforCollege201 6FNL.pdf).

Sewell, William H., Archibald O. Haller, and Alejandro Portes. 1969. “The Educational and Early Occupational Attainment Process.” American Sociological Review 34(1):82–92.

Singell, Larry and Mark Stater. 2006. “Going, Going, Gone: The Effects of Aid Policies on Graduation at Three Large Public Institutions.” Policy Sciences 39(4):379– 403.

Sotiropoulos, Veneta and Alain d’Astous. 2012. “Social Networks and Credit Card Overspending Among Young Adult Consumers.” Journal of Consumer Affairs 46(3):457–84.

Student Monitor. 2016. Financial Services. Retrieved February 21, 2017 (http://files.studentmonitor.com/s16/s16FSExec.pdf).

The Education Resources Institute and The Institute for Higher Education Policy. 1998. Credit Risk or Credit Worthy? College Students and Credit Cards, a National Survey. Retrieved June 22, 2016 (http://www.ihep.org/sites/default/files/uploads/docs/pubs/creditriskworthy.pdf).

127

Thelin, John R. 2011. A History of American Higher Education, 2nd Edition. 2nd edition. Baltimore: Johns Hopkins University Press.

Titus, Marvin A. 2007. “Detecting Selection Bias, Using Propensity Score Matching, and Estimating Treatment Effects: An Application to the Private Returns to a Master’s Degree.” Research in Higher Education 48(4):487–521.

United States General Accounting Office. 2001. “Consumer Finance: College Students and Credit Cards.” Retrieved February 18, 2017 (http://eric.ed.gov/?id=ED454785).

United States General Accounting Office. 2014. “Credit Cards: Marketing to College Students Appears to Have Declined.” Retrieved February 18, 2017 (http://www.gao.gov/assets/670/661121.pdf).

U.S. Congress. Senate and House of Representatives. Credit Card Accountability Responsibility and Disclosure Act of 2009. H.R. 627. 111th Congress, 1st Session, 2009.

Wang, Jeff and Jing J. Xiao. 2009. “Buying Behavior, Social Support and Credit Card Indebtedness of College Students.” International Journal of Consumer Studies 33(1):2–10.

Xiao, Jing Jian, Chuanyi Tang, Joyce Serido, and Soyeon Shim. 2011. “Antecedents and Consequences of Risky Credit Behavior among College Students: Application and Extension of the Theory of Planned Behavior.” Journal of Public Policy & Marketing 30(2):239–245.

Zhan, Min. 2014. “The Impact of Youth Debt on College Graduation.” Journal of Sociology and Social Welfare 41:133.

128

Appendix A: Materials for Chapter 2

129

Chapter 2 Variables

The ELS data have questions about college student credit card use that include: 1. “How many credit cards do you have in your own name that are billed to you? (If none, enter zero.)” . 0-6 . 7 or more credit cards 2. A recoded variable from the above measure that collapses numbered categories into a dichotomous variable that denotes credit card ownership. . Respondent does not own a credit card . Respondent owns a credit card 3. A recoded variable from the above measure collapsing numbers 2-7 into one category for students who own 2 or more credit cards. . Respondent does not own a credit card . Respondent owns 1 credit card . Respondent owns 2 or more credit cards 4. “Do you usually pay off your credit card balance each month, or carry the balance over from month to month?” This question was asked to any respondent in the second follow up (2006) who had at least one credit card. . Pay off balance . Carry balance 5. A recoded variable from the above measure expanding the variable to include a reference category that includes students who do not own a credit card. . No credit card . Credit card, does not carry balance . Carries balance on credit card Important control variables from the ELS data include: 1. Socioeconomic status . Lower . Middle . Upper 2. Race . White . Asian . Black . Hispanic 130

. Other 3. Gender . Male . Female 4. Whether student delayed entry into college . Enrolled in college within six months of graduating from high school . Delayed entry into college by 7 months or more after graduating from high school 5. First generation student status . Respondent has at least one parent who attended college . Respondent is the first person in their family to attend college 6. Hours worked weekly 2005-06 . Did not work . 1-20 hours . More than 20 hours 7. Institution type . Respondent first attended a less than 4-year postsecondary institution . Respondent first attended a 4-year postsecondary institution 8. Total student loans in 2012 . No student loans . $1-$10,000 . Greater than $10,000 9. Total Pell Grants in 2012 . No Pell Grants . $1-$10,000 . Greater than $10,000 10. Whether students’ postsecondary education is paid in part with parent loans . No . Yes 11. Respondent has biological child or is currently or has been previously married in 2006 . No . Yes

131

Chapter 2 Tables

Table A.1: Descriptive Statistics Variables Mean S.D. Min Max N Respondent has credit card in own name 0.47 0.50 0 1 7,480 How many credit cards respondent has in own name 0.82 1.20 0 7 7,480 Respondent owns multiple credit cards - - 0 2 7,480 Respondent does not own a credit card 0.53 - - - 3,947 Respondent owns 1 credit card 0.28 - - - 2,109 Respondent owns 2 or more credit cards 0.19 - - - 1,424 Respondent carries balance on credit card from month to - - 0 2 7,459 month No credit card 0.53 - - - 3,947 Credit card, does not carry balance from month to month 0.33 - - - 2,459 Carries balance from month to month 0.14 - - - 1,053 Socioeconomic status - - 0 2 7,190 Lower 0.32 - - - 2,304 Middle 0.26 - - - 1,863 Upper 0.42 - - - 3,023 Race - - 0 4 7,190 White 0.64 - - - 4,582 Asian 0.11 - - - 823 Black 0.10 - - - 734 Hispanic 0.10 - - - 737 Other 0.04 - - - 314 Female 0.56 0.50 0 1 7,207 Hours worked per week 2005-06 - - 0 2 7,477 Did not work 0.30 - - - 2,248 1-20 hours 0.39 - - - 2,829 More than 20 hours 0.32 - - - 2,400 First generation college student 0.16 0.37 0 1 7,201 Respondent has biological child or is married or previously 0.03 0.18 0 1 7,496 married in 2006 Delayed entry into college 0.10 0.30 0 1 7,532 Started at a 4-year postsecondary institutions 0.72 0.45 0 1 7,527 (Continued)

132

Table A.1: Continued Total Pell Grant amount - - 0 2 7,532 No Pell Grants 0.31 - - - 2,321 $1-$10,000 0.25 - - - 1,915 Greater than $10,000 0.44 - - - 3,296 Total loans - - 0 2 7,532 No loans 0.36 - - - 2,734 $1-$10,000 0.15 - - - 1,152 $10,000 0.48 - - - 3,646 Used parent loans to pay for some of college 0.19 0.39 0 1 7,496

133

Table A.2: Demographic Differences in College Student Credit Card Ownership Does Not Own a Owns a Credit Card VARIABLES Credit Card

Overall (n = 7,480) 52.77% 47.23%

Socioeconomic status*** Lower (n = 2,289) 50.11% 49.89% Middle (n = 1,848) 52.87% 47.13% Upper (n = 3,005) 55.24% 44.76%

Hours worked weekly 2005-06*** Did not work (n = 2,232) 61.69% 38.31% 1-20 hours (n = 2,819) 51.44% 48.56% More than 20 hours (n = 2,394) 45.99% 54.01%

Institution type** Respondent started at a less than 4-year institution (n = 55.69% 44.31% 2,065) Respondent started at a 4-year institution (n = 5,412) 51.66% 48.34%

First generation student status** Respondent is not a first generation college student (n = 53.67% 46.33% 6,013) Respondent is a first generation college student (n = 1,140) 49.21% 50.79%

Total student loans*** No student loans (n = 2,713) 56.58% 43.42% $1-$10,000 (n = 1,140) 54.82% 45.18% Greater than $10,000 (n = 3,627) 49.27% 50.73%

Race*** White (n = 4,557) 54.25% 45.75% Asian (n = 812) 46.06% 53.94% Black (n = 728) 56.46% 43.54% Hispanic (n = 734) 51.09% 48.91% Other (n = 311) 48.87% 51.13%

Gender*** Male (n = 3,138) 56.31% 43.69% Female (n = 4,020) 50.30% 49.70% *** p<0.001, ** p<0.01, * p<0.05

134

Table A.3: Demographic Differences in Owning Multiple Credit Cards Does Not Own Owns 2 or Owns 1 Credit a Credit Card More Credit Card VARIABLES Cards

Overall (n = 7,480) 52.77% 28.20% 19.04%

Socioeconomic status*** Lower (n = 2,289) 50.11% 25.12% 24.77% Middle (n = 1,848) 52.87% 26.57% 20.56% Upper (n = 3,005) 55.24% 31.18% 13.58%

Hours worked weekly 2005-06*** Did not work (n = 2,232) 61.69% 27.24% 11.07% 1-20 hours (n = 2,819) 51.44% 30.05% 18.52% More than 20 hours (n = 2,394) 45.99% 26.90% 27.11%

Institution type*** Respondent started at a less than 4-year institution 55.69% 23.29% 21.02% (n = 2,065) Respondent started at a 4-year institution (n = 51.66% 30.04% 18.29% 5,412)

First generation student status*** Respondent is not a first generation college student 53.67% 28.75% 17.58% (n = 6,013) Respondent is a first generation college student 49.21% 24.47% 26.32% (n = 1,140)

Total student loans*** No student loans (n = 2,713) 56.58% 28.38% 15.04% $1-$10,000 (n = 1,140) 54.82% 24.02% 21.14% Greater than $10,000 (n = 3,627) 49.27% 29.36% 21.37%

Total Pell Grants*** No Pell Grants (n = 2,307) 53.27% 29.30% 17.43% $1-$10,000 (n = 1,902) 50.79% 26.87% 22.34% Greater than $10,000 (n = 3,271) 53.56% 28.19% 18.25%

Race*** White (n = 4,557) 54.25% 28.29% 17.16% Asian (n = 812) 46.06% 33.13% 20.81% Black (n = 728) 56.46% 22.12% 21.43% Hispanic (n = 734) 51.09% 24.25% 24.66% Other (n = 311) 48.87% 29.58% 21.54%

Gender*** Male (n = 3,138) 56.31% 29.86% 13.83% Female (n = 4,020) 50.30% 26.69% 23.01% *** p<0.001, ** p<0.01, * p<0.05

135

Table A.4: Demographic Differences in Carrying a Revolving Credit Balance Credit Card, Does Not Own Credit Card, Does Not Carry a Credit Card Carries Balance VARIABLES Balance

Overall (n = 7,459) 52.92% 32.97% 14.12%

Socioeconomic status*** Lower (n = 2,281) 50.28% 30.69% 19.03% Middle (n = 1,844) 52.98% 31.56% 15.46% Upper (n = 2,997) 55.39% 35.30% 9.31%

Hours worked weekly 2005-06*** Did not work (n = 2,224) 61.92% 30.67% 7.42% 1-20 hours (n = 2,811) 51.58% 35.33% 13.09% More than 20 hours (n = 2,390) 46.07% 32.22% 21.72%

Institution type*** Respondent started at a less than 4-year institution 55.74% 27.10% 17.16% (n = 2,063) Respondent started at a 4-year institution (n = 51.84% 35.21% 12.94% 5,393)

Delay status*** Respondent enrolled in college within 6 months of 52.54% 33.73% 13.74% graduating high school (n = 6,698) Respondent delayed entry into college by 7 months 56.24% 26.28% 17.48% or more after high school graduation (n = 761)

First generation student status*** Respondent is not a first generation college student 53.82% 32.97% 13.21% (n = 5,996) Respondent is a first generation college student 49.34% 32.10% 18.56% (n = 1,137)

Total student loans*** No student loans (n = 2,701) 56.83% 34.06% 9.11% $1-$10,000 (n = 1,140) 54.82% 29.04% 16.14% Greater than $10,000 (n = 3,618) 49.39% 33.39% 17.22%

Total Pell Grants*** No Pell Grants (n = 2,301) 53.41% 34.90% 11.69% $1-$10,000 (n = 1,898) 50.90% 30.77% 18.34% Greater than $10,000 (n = 3,260) 53.74% 32.88% 13.37%

Race*** White (n = 4,549) 54.34% 33.48% 12.18% Asian (n = 809) 46.23% 41.29% 12.48% Black (n = 725) 56.69% 23.17% 20.14% Hispanic (n = 730) 51.37% 29.45% 19.18% Other (n = 309) 49.19% 32.36% 18.45% (Continued) 136

Table A.4 Continued Gender*** Male (n = 3,128) 56.49% 32.26% 11.25% Female (n = 4,010) 50.42% 33.32% 16.26%

Family status*** Respondent is not currently married, has not ever 52.78% 33.30% 13.92% been married and does not have a biological child (n = 7,192) Respondent is currently or has previously been 55.56% 23.81% 20.63% married and/or has a biological child (n = 252) *** p<0.001, ** p<0.01, * p<0.05

137

Table A.5: Logistic Regression Model for Whether Students Own a Credit Card in their Name (Odds Ratios Reported) (1) (2) VARIABLES Basic Full Model Model

Socioeconomic status (reference category: Middle) Lower 1.11 1.08 (0.09) (0.09) Upper 0.95 1.05 (0.07) (0.08) Race (reference group: White) Asian 1.29** 1.38*** (0.12) (0.13) Black 0.85 0.80* (0.08) (0.08) Hispanic 1.17 1.24* (0.12) (0.13) Other 1.21 1.19 (0.18) (0.18) Female 1.25*** 1.20** (0.08) (0.07) Respondent has ever been married and/or has a biological child 0.88 0.95 (0.14) (0.16) Respondent started at a 4-year institution 1.36*** (0.10) Delayed entry into college 0.97 (0.10) First generation college student 1.11 (0.11) Hours worked weekly during 2005-06 school year (reference category: 1-20 hours) Did not work 0.64*** (0.05) More than 20 hours 1.33*** (0.10) Total student loans (reference category: $1-$10,000) No student loans 0.89 (0.09) Greater than $10,000 1.24* (0.11) Respondent used parent loans to pay for college 0.99 (0.08) Total Pell Grant amount (reference category: $1-$10,000) No Pell Grants 0.90 (0.08) Greater than $10,000 0.99 (0.08) Constant 0.79*** 0.61*** (0.05) (0.08)

Observations 7,129 7,060

BIC 2,123,793 2,061,674 Difference of 62,119 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

138

Table A.6: Logistic Regression Model for Whether Students Own Multiple Credit Cards in their Name (Odds Ratios Reported) (1) (2) VARIABLES Basic Full Model Model

Socioeconomic status (reference category: Middle) Lower 1.07 0.94 (0.12) (0.12) Upper 0.60*** 0.69** (0.07) (0.08) Race (reference group: White) Asian 0.92 1.06 (0.12) (0.15) Black 1.23 1.20 (0.19) (0.19) Hispanic 1.56** 1.59** (0.23) (0.23) Other 1.15 1.18 (0.25) (0.25) Female 1.80*** 1.76*** (0.17) (0.17) Respondent has ever been married and/or has a biological child 1.23 1.26 (0.30) (0.31) Respondent started at a 4-year institution 1.10 (0.12) Delayed entry into college 1.01 (0.16) First generation college student 1.26 (0.17) Hours worked weekly during 2005-06 school year (reference category: 1-20 hours) Did not work 0.68** (0.09) More than 20 hours 1.60*** (0.17) Total student loans (reference category: $1-$10,000) No student loans 0.92 (0.14) Greater than $10,000 1.15 (0.15) Respondent used parent loans to pay for college 0.92 (0.11) Total Pell Grant amount (reference category: $1-$10,000) No Pell Grants 0.94 (0.12) Greater than $10,000 0.89 (0.10) Constant 0.54*** 0.44*** (0.06) (0.09)

Observations 3,354 3,323

BIC 969,447 940,518 Difference of 28,929 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

139

Table A.7: Logistic Regression Model for Whether Students Carry a Revolving Balance on their Credit Card (Odds Ratios Reported) (1) (2) VARIABLES Basic Full Model Model

Socioeconomic status (reference category: Middle) Lower 1.20 1.15 (0.14) (0.15) Upper 0.61*** 0.74* (0.08) (0.09) Race (reference group: White) Asian 0.73* 0.85 (0.11) (0.13) Black 1.70*** 1.41* (0.26) (0.23) Hispanic 1.35* 1.30 (0.20) (0.20) Other 1.57* 1.54 (0.34) (0.35) Female 1.21 1.15 (0.12) (0.12) Respondent has ever been married and/or has a biological child 1.73* 1.64 (0.43) (0.43) Respondent started at a 4-year institution 0.84 (0.10) Delayed entry into college 1.23 (0.20) First generation college student 0.86 (0.12) Hours worked weekly during 2005-06 school year (reference category: 1-20 hours) Did not work 0.71* (0.10) More than 20 hours 1.49*** (0.17) Total student loans (reference category: $1-$10,000) No student loans 0.54*** (0.09) Greater than $10,000 1.37* (0.19) Respondent used parent loans to pay for college 1.05 (0.13) Total Pell Grant amount (reference category: $1-$10,000) No Pell Grants 0.65** (0.09) Greater than $10,000 1.04 (0.13) Constant 0.41*** 0.46*** (0.05) (0.10)

Observations 3,334 3,304

BIC 895,268 852,326 Difference of 42,942 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

140

Chapter 2 Figures

Figure A.1: Ownership of Multiple Credit Cards, by Socioeconomic Status Ownership of Multiple Credit Cards, by Socioeconomic Status (n = 7,480) 100% 13.58 24.77 20.56 80% 31.18 26.57 60% 25.12

40% 50.11 52.87 55.24 20%

0% Lower Middle Upper

Does not own a credit card Owns 1 credit card Owns 2 or more credit cards

Figure A.2: Whether Student Carries Revolving Credit Card Balance, by Socioeconomic Status Whether Student Carries Revolving Credit Card Balance, by Socioeconomic Status (n = 7,459) 100% 9.31 19.03 15.46 80% 35.30 30.69 31.56 60%

40% 50.28 52.87 55.39 20%

0% Lower Middle Upper

No credit card Credit card, does not carry balance Carries balance on credit card

141

Figure A.3: Ownership of Multiple Credit Cards, by Institution Ownership of Multiple Credit Cards, by Institution (n = 7,480) 100% 21.02 18.29 80% 23.29 30.04 60%

40%

55.69 51.66 20%

0% Started at less than 4-year institution Started at 4-year institution

Does not own a credit card Owns 1 credit card Owns 2 or more credit cards

Figure A.4: Whether Student Carries Revolving Credit Card Balance, by Institution Whether Student Carries Revolving Credit Card Balance, by Institution (n = 7,459) 100% 17.16 12.94 80% 27.1 35.21 60%

40% 55.74 51.84 20%

0% Started at less than 4-year institution Started at 4-year institution

No credit card Credit card, does not carry balance Carries balance on credit card

142

Figure A.5: Ownership of Multiple Credit Cards, by Hours Worked Weekly During the 2005-2006 School Year Ownership of Multiple Credit Cards, by Hours Worked Weekly During 2005-2006 School Year (n = 7,480) 100% 11.07 18.52 27.11 80% 27.24 30.05 60% 26.90

40% 61.69 51.44 20% 45.99

0% Did not work 1-20 hours More than 20 hours

Does not own a credit card Owns 1 credit card Owns 2 or more credit cards

Figure A.6: Whether Student Carries Revolving Credit Card Balance, by Hours Worked Weekly During the 2005-2006 School Year Whether Student Carries Revolving Credit Card Balance, by Hours Worked Weekly During 2005-2006 School Year (n = 7,459) 100% 7.42 13.09 21.72 80% 30.67 35.33 60% 32.22

40% 61.92 51.58 20% 46.07

0% Did not work 1-20 hours More than 20 hours

No credit card Credit card, does not carry balance Carries balance on credit card

143

Appendix B: Materials for Chapter 3

144

Chapter 3 Variables

The ELS data have questions about college student credit card use that include: 1. “Have you used your credit card to pay any portion of your tuition?” This question was asked to any respondent with at least one credit card who ever attended post-secondary school. Important control variables from the ELS data include: 1. Socioeconomic status . Lower . Middle . Upper 2. Race . White . Asian . Black . Hispanic . Other 3. Gender . Male . Female 4. Whether student delayed entry into college . Enrolled in college within six months of graduating from high school . Delayed entry into college by 7 months or more after graduating from high school 5. First generation student status . Respondent has at least one parent who attended college . Respondent is the first person in their family to attend college 6. Hours worked weekly 2005-06 . Did not work . 1-20 hours . More than 20 hours 7. Institution type . Respondent first attended a less than 4-year postsecondary institution . Respondent first attended a 4-year postsecondary institution 8. Total student loans in 2012 . No student loans 145

. $1-$10,000 . Greater than $10,000 9. Total Pell Grants in 2012 . No Pell Grants . $1-$10,000 . Greater than $10,000 10. Whether students’ postsecondary education is paid in part with parent loans . No . Yes 11. Respondent has biological child or is currently or has been previously married in 2006 . No . Yes

The SCFW data also ask about college students’ use of credit cards to pay for tuition, with some additional follow up questions that investigate students’ reasons for and beliefs about using credit cards to pay for educational expenses, including tuition: 1. “Have you ever used a credit card in your name to pay for your college tuition?” (Y/N) 2. “What is the primary reason you used credit cards to pay for your college tuition? . My financial aid package didn’t cover all my tuition . I missed a deadline to apply for financial aid/student loans . I had to use my tuition money for an emergency . I didn’t want to take on any more student loans . Paying with a credit card is easier than other methods . I always pay some of my tuition with my credit card(s) . Other (please specify) ______” With regard to financial situation variables and financial literacy variables, the SCFW asks the following questions: 1. “How likely is it that you could come up with $400 in cash in the event of a financial emergency during the school year?” . Unlikely . Likely 2. “I have experienced financial difficulties while enrolled at my current institution.” . Disagree . Agree 3. “I worry about having enough money to pay for school.” . Disagree . Agree **Questions 4-9 were combined into a cumulative financial literacy score from 0- 6. Students received 1 point for each question they answered correctly. 146

4. “Imagine that the interest rate on your savings account is 1% per year and inflation is 2% per year. After 1 year, would you be able to buy more than today, exactly the same as today, or less than today with the money in this account? . More than today . Exactly the same as today . Less than today . Don't know” 5. “Suppose you have $100 in a savings account and the interest rate was 2% per year. After 5 years, how much would you have in the account if you left the money to grow? . More than $102 . Exactly $102 . Less than $102 . Don’t know” 6. “Suppose you borrowed $5,000 to help cover college expenses for the coming year. You can choose to repay this loan over 10 years, 20 years, or 30 years. Which of these repayment options will cost you the least amount of money over the length of the repayment period? . 10-year repayment option . 20-year repayment option . 30-year repayment option . Don’t know” 7. “All paycheck stubs show your gross pay (the total amount you earned before any taxes were taken out for the pay period) and your net pay (the amount of your check after all taxes). The taxes that are commonly taken out include federal, state and local income tax, Social Security tax, and Medicare tax. On average, what percentage of your income would you expect to receive as take-home pay? . 100% . 90-99% . 80-89% . 70-79% . Don't know” 8. “Over a long period of time, which of the following types of investments will give you the highest rate of return on average? . Savings account . Stocks . Bonds . Don’t know” 9. “True/False: Maxing out your credit card will negatively impact your credit score, even if you make the minimum monthly payments. . True . False 147

. Don’t know” Important control variables from the SCFW include: 1. Whether respondent has been offered or received a Pell Grant. . No . Yes 2. Total student loan amount borrowed . No student loans . $1-$9,999 . $10,000+ 3. Current annual income . No income . $1-$4,999 . $5,000+ 4. Work hours per week . Not employed . 1-20 . More than 20 5. Whether respondent is considered a dependent student of their parents for federal financial aid . No . Yes 6. Whether respondent is financial responsible for a child, spouse or other family member . No . Yes 7. Whether respondent attends a 4-year postsecondary institution . No . Yes 8. Gender . Female . Male 9. Race . White . Black . Hispanic . Asian . Other

148

Chapter 3 Tables

Table B.1: Descriptive Statistics for ELS Variables Variables Mean S.D. Min Max N Respondent pays for tuition with a credit card 0.16 0.37 0 1 3,531 Socioeconomic status - - 0 2 3,358 Lower 0.34 - - - 1,142 Middle 0.26 - - - 871 Upper 0.40 - - - 1,345 Race - - 0 4 3,358 White 0.62 - - - 2,085 Asian 0.13 - - - 438 Black 0.09 - - - 317 Hispanic 0.11 - - - 359 Other 0.05 - - - 159 Female 0.59 0.49 0 1 3,369 Hours worked per week 2005-06 - - 0 2 3,517 Did not work 0.24 - - - 855 1-20 hours 0.39 - - - 1,369 More than 20 hours 0.37 - - - 1,293 First generation college student 0.17 0.38 0 1 3,365 Respondent has biological child or is married or previously 0.03 0.18 0 1 3,529 married in 2006 Delayed entry into college 0.09 0.29 0 1 3,533 Started at a 4-year postsecondary institutions 0.74 0.44 0 1 3,531 Total Pell Grant amount - - 0 2 3,533 No Pell Grants 0.31 - - - 1,078 $1-$10,000 0.26 - - - 936 Greater than $10,000 0.43 - - - 1,519 Total loans - - 0 2 3,533 No loans 0.33 - - - 1,178 $1-$10,000 0.15 - - - 515 $10,000 0.52 - - - 1,840 Used parent loans to pay for some of college 0.19 0.39 0 1 3,520

149

Table B.2: Descriptive Statistics for SCFW Variables Variables Mean S.D. Min Max N Respondent pays for tuition with a credit card 0.14 0.35 0 1 14,541 Respondent has access to $400 cash in an emergency 0.58 0.49 0 1 26,861 Respondent has experienced financial difficulty in school 0.59 0.49 0 1 24,388 Respondent worries not enough money for school 0.64 0.48 0 1 25,287 Financial literacy measure 3.28 1.62 0 6 22,278 Respondent’s income - - 0 2 20,155 No income 0.19 - - - 3,851 $1-$4,999 0.38 - - - 7,569 $5,000+ 0.43 - - - 8,735 Total student loan amount - - 0 2 13,075 No student loans 0.08 - - - 1,058 $1-$9,999 0.29 - - - 3,819 $10,000+ 0.63 - - - 8,198 Whether offered or received Pell Grant 0.54 0.50 0 1 19,080 Dependent status 0.69 0.46 0 1 19,444 Financially responsible for child, spouse, or other family 0.14 0.34 0 1 25,031 member Race - - 0 4 26,987 White 0.61 - - - 16,529 Black 0.08 - - - 2,079 Hispanic 0.10 - - - 2,569 Asian 0.10 - - - 2,727 Other 0.11 - - - 3,083 Male 0.33 0.47 0 1 26,675 Work hours per week - - 0 2 25,217 Not employed 0.36 - - - 9,126 1-20 0.39 - - - 9,847 More than 20 0.25 - - - 6,244 Respondent attends 4-year institution 0.76 0.43 0 1 27,241

150

Table B.3: Logistic Regression Model for Whether Students Use a Credit Card to Pay for Tuition Using ELS Data (Odds Ratios Reported) (1) (2) VARIABLES Basic Full Model Model

Socioeconomic status (reference category: Middle) Lower 1.33* 1.23 (0.19) (0.20) Upper 0.64** 0.72* (0.10) (0.11) Race (reference group: White) Asian 1.31 1.38 (0.21) (0.23) Black 0.95 0.93 (0.19) (0.19) Hispanic 1.23 1.22 (0.23) (0.23) Other 0.85 0.89 (0.25) (0.26) Female 0.82 0.79 (0.10) (0.10) Respondent has ever been married and/or has a biological child 1.33 1.32 (0.41) (0.41) Respondent started at a 4-year institution 0.76 (0.11) Delayed entry into college 0.71 (0.15) First generation college student 1.00 (0.17) Hours worked weekly during 2005-06 school year (reference category: 1-20 hours) Did not work 0.84 (0.15) More than 20 hours 1.34* (0.19) Total student loans (reference category: $1-$10,000) No student loans 1.13 (0.22) Greater than $10,000 1.03 (0.18) Respondent used parent loans to pay for college 0.83 (0.13) Total Pell Grant amount (reference category: $1-$10,000) No Pell Grants 0.84 (0.14) Greater than $10,000 0.96 (0.14) Constant 0.22*** 0.26*** (0.03) (0.07)

Observations 3,352 3,321

BIC 649,990 636,559 Difference of 13,431 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

151

Table B.4: Average Financial Literacy Scores by Whether Student Uses Credit Card to Pay for Tuition Does not use credit card Uses credit card to pay

to pay for tuition for tuition

Average financial literacy score (scale: 0-6) 3.41 3.37

152

Table B.5: Logistic Regression Models for Whether Students Use a Credit Card to Pay for Tuition Using SCFW Data (Odds Ratios Reported) (1) (2) (3) Basic Financial Full VARIABLES Model Aid Model

Respondent can come up with $400 cash in an emergency 0.60*** 0.56*** 0.61*** (0.02) (0.04) (0.05) Respondent has had financial difficulties while in school 1.78*** 1.67*** 1.74*** (0.10) (0.20) (0.19) Respondent worries they will not have enough money to pay for school 2.66*** 2.41*** 2.45*** (0.20) (0.24) (0.28) Financial literacy measure 1.05** 1.07* 1.06 (0.02) (0.03) (0.03) Respondent’s income (reference category: $1-$4,999) No income 1.28 1.01 (0.20) (0.20) $5,000+ 2.56*** 1.49** (0.33) (0.20) Total student loan amount (reference category: $1-$9,999) No student loans 0.69 0.69 (0.16) (0.19) $10,000+ 0.83 0.89 (0.09) (0.09) Whether offered or received Pell Grant 1.03 0.80* (0.08) (0.07) Dependent 0.49*** (0.05) Responsible for child, spouse, or other family member 1.27* (0.14) Race (reference group: White) Black 1.61* (0.32) Hispanic 1.63** (0.29) Asian 1.52 (0.37) Other 1.44** (0.17) Male 0.94 (0.09) Work hours per week (reference category: 1-20) Not employed 1.06 (0.14) More than 20 1.60*** (0.18) Respondent attends 4-year institution 0.67 (0.16) Constant 0.06*** 0.03*** 0.06*** (0.01) (0.01) (0.03)

Observations 13,384 5,830 5,363 Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

153

Table B.6: Students’ Primary Reasons for Using Credit Card to Pay for Tuition What is the primary reason you used credit cards to pay for your college tuition? (n = 2,045)

My financial aid package didn’t cover all my tuition 39.1% I missed a deadline to apply for financial aid/student loans 4.3% I had to use my tuition money for an emergency 4.1% I didn’t want to take on any more student loans 10.4% Paying with a credit card is easier than other methods 15.3% I always pay some of my tuition with my credit card(s) 11.1% Other (please specify) 15.7%

Total 100%

154

Chapter 3 Figures

Figure B.1: Whether Student Pays for Tuition with a Credit Card, by Socioeconomic Status Whether Student Pays for Tuition with a Credit Card, by Socioeconomic Status (n = 3,356) 100% 11.38 20.86 17.59 80%

60% 88.62 40% 79.14 82.41

20%

0% Lower Middle Upper

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

Figure B.2: Whether Student Pays for Tuition with a Credit Card, by Hours Worked Whether Student Pays for Tuition with a Credit Card, by Hours Worked Weekly during 2005-2006 School Year (n = 3,515) 100% 12.87 13.52 21.75 80%

60%

87.13 86.48 40% 78.25

20%

0% Did not work 1-20 hours More than 20 hours

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

155

Figure B.3: Whether Student Pays for Tuition with a Credit Card, by Institution Whether Student Pays for Tuition with a Credit Card, by Institution (n = 3,529) 100% 13.85 23.50 80%

60%

86.15 40% 76.50

20%

0% Started at less than 4-year institution Started at 4-year institution

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

Figure B.4: Whether Student Pays for Tuition with a Credit Card, by First Generation College Student Status Whether Student Pays for Tuition with a Credit Card, by First Generation College Student Status (n = 3,363) 100% 15.12 22.15 80%

60%

40% 84.88 77.85

20%

0% Non-First Generation College Student First Generation College Student

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

156

Figure B.5: Whether Student Pays for Tuition with a Credit Card, by Whether Student Has Access to Parent Loans to Help Pay for College Whether Student Pays for Tuition with a Credit Card, by Whether Student Has Access to Parent Loans to Help Pay for College (n = 3,518) 100% 16.95 13.80 80%

60%

40% 83.05 86.20

20%

0% Does Not Use Parent Loans Uses Parent Loans

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

Figure B.6: Whether Student Pays for Tuition with a Credit Card, by Total Pell Grant Amount Whether Student Pays for Tuition with a Credit Card, by Total Pell Grant Amount (n = 3,531) 100% 12.80 19.55 16.88 80%

60%

40% 87.20 80.45 83.12

20%

0% No Pell Grant $1-$10,000 More than $10,000

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

157

Figure B.7: Whether Student Pays for Tuition with a Credit Card, by Race Whether Student Pays for Tuition with a Credit Card, by Race (n = 3,356) 100% 14.53 20.78 16.72 20.11 15.82 80%

60%

40% 85.47 79.22 83.28 79.89 84.18

20%

0% White Asian Black Hispanic Other

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

Figure B.8: Whether Student Pays for Tuition with a Credit Card, by Family Status Whether Student Pays for Tuition with a Credit Card, by Family Status (n = 3,527) 100% 16.11 23.68 80%

60%

40% 83.89 76.32

20%

0% Never Married, Does Not Have Child Previously/Currently Married and/or Has Child

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

158

Figure B.9: Whether Student Pays for Tuition with a Credit Card, by Access to Emergency Cash Whether Student Pays for Tuition with a Credit Card, by Access to Emergency Cash (n = 14,504) 100% 10.08 20.65 80%

60% 89.92 40% 79.35

20%

0% Unlikely that respondent could come up Likely that respondent could come up with with $400 emergency cash $400 emergency cash

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

Table B.10: Whether Student Pays for Tuition with a Credit Card, by Financial Difficulty in School Whether Student Pays for Tuition with a Credit Card, by Financial Difficulty in School (n = 14,411) 100% 7.42 18.68 80%

60% 92.58 40% 81.32

20%

0% Respondent has not had financial Respondent has had financial difficulties at difficulties at school school

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

159

Figure B.11: Whether Student Pays for Tuition with a Credit Card, by Whether Student Worries About Having Enough Money to Pay for School Whether Student Pays for Tuition with a Credit Card, by Whether Student Worries About Having Enough Money to Pay for School (n = 14,504)

100% 6.17 19.19 80%

60% 93.83 40% 80.81

20%

0% Respondent does not worry about having Respondent worries about having enough enough money to pay for school money to pay for school

Credit card, does not use it to pay for tuition Uses credit card to pay for tuition

160

Appendix C: Materials for Chapter 4

161

Chapter 4 Variables

The ELS data have questions about college student credit card use that include: 1. “How many credit cards do you have in your own name that are billed to you? (If none, enter zero.)” . 0-6 . 7 or more credit cards 2. “Do you usually pay off your credit card balance each month, or carry the balance over from month to month?” This question was asked to any respondent in the second follow up (2006) who had at least one credit card. . Pay off balance . Carry balance 3. A recoded variable from the above measure expanding the variable to include a reference category that includes students who do not own a credit card. . No credit card . Credit card, does not carry balance . Carries balance on credit card 4. “Have you used your credit card to pay any portion of your tuition?” This question was asked to any respondent with at least one credit card who ever attended post-secondary school. . Yes . No 5. A recoded variable from the above measure expanding the variable to include a reference category that includes students who do not own a credit card. . No credit card . Credit card, does not pay for tuition with credit card . Pays for tuition using credit card In order to measure students’ educational outcomes, I use these ELS variables: 1. Respondent’s highest level of education as of the third follow up (2012) . Some PS attendance, no PS credential . Undergraduate certificate . Associate’s degree . Bachelor’s degree . Post-Baccalaureate certificate . Master's degree/Post-Master's certificate 162

. Doctoral degree 2. A recoded variable from the above measure simplifying degree attainment to whether or not the respondent attained a bachelor’s degree: . No bachelor’s degree by 2012 . Bachelor’s degree by 2012 Important control variables from the ELS data include: 1. Socioeconomic status . Lower . Middle . Upper 2. Race . White . Asian . Black . Hispanic . Other 3. Gender . Male . Female 4. Hours worked weekly 2005-06 . Did not work . 1-20 hours . More than 20 hours 5. Institution type . Respondent first attended a less than 4-year postsecondary institution . Respondent first attended a 4-year postsecondary institution 6. Total student loans in 2012 . No student loans . $1-$10,000 . Greater than $10,000 7. Total Pell Grants in 2012 . No Pell Grants . $1-$10,000 . Greater than $10,000 8. Respondent has biological child or is currently or has been previously married in 2006 . No . Yes 9. High school GPA . 3.00 or lower . 3.01-4.00 High school composite math/reading score (20.91-81.04)

163

Chapter 4 Tables

Table C.1: Descriptive Statistics Variables Mean S.D. Min Max N Respondent has credit card in own name 0.47 0.50 0 1 6,752 How many credit cards respondent has in own name 0.82 1.19 0 7 6,752 Respondent uses credit card to pay for tuition - - 0 2 6,677 No credit card 0.53 - - - 3,549 Credit card, does not pay for tuition with credit card 0.39 - - - 2,619 Pays for tuition with credit card 0.08 - - - 509 Respondent carries balance on credit card from month to - - 0 2 6,732 month No credit card 0.53 - - - 3,549 Credit card, does not carry balance from month to month 0.34 - - - 2,264 Carries balance from month to month 0.14 - - - 919 Attained bachelor's degree by 2012 0.64 0.48 0 1 7,124 Socioeconomic status - - 0 2 6,800 Lower 0.31 - - - 2,086 Middle 0.26 - - - 1,787 Upper 0.43 - - - 2,927 Race - - 0 4 6,800 White 0.65 - - - 4,390 Asian 0.12 - - - 795 Black 0.10 - - - 653 Hispanic 0.10 - - - 666 Other 0.04 - - - 296 Female 0.56 0.50 0 1 6,816 Hours worked per week 2005-06 - - 0 2 6,748 Did not work 0.32 - - - 2,162 1-20 hours 0.39 - - - 2,631 More than 20 hours 0.29 - - - 1,955 Respondent has biological child or is married or previously 0.03 0.16 0 1 6,764 married in 2006 Started at a 4-year postsecondary institutions 0.76 0.43 0 1 7,118 Total Pell Grant amount - - 0 2 7,124 (Continued)

164

Table C.1: Continued No Pell Grants 0.32 - - - 2,260 $1-$10,000 0.25 - - - 1,768 Greater than $10,000 0.43 - - - 3,096 Total loans - - 0 2 7,124 No loans 0.36 - - - 2,558 $1-$10,000 0.15 - - - 1,045 $10,000 0.49 - - - 3,521 High school GPA higher than 3.0 0.63 0.48 0 1 6,668 High school composite math/reading score 55.38 8.55 20.91 81.04 7,061

165

Table C.2: Logistic Regression Model for Bachelor’s Degree Attainment by 2012, Testing Credit Card Ownership Variable (Odds Ratios Reported) (1) (2) (3) VARIABLES Credit Card Demographic Full Variable Variables Model

Respondent owns a credit card 1.03 1.10 0.98 (0.07) (0.08) (0.08) Socioeconomic status (reference category: Middle) Lower 0.75** 0.84 (0.07) (0.09) Upper 1.55*** 1.40*** (0.15) (0.14) Hours worked weekly 2005-06 (reference category: 1-20 hours) Did not work 0.92 1.06 (0.08) (0.11) More than 20 hours 0.46*** 0.58*** (0.04) (0.05) Respondent has biological child, is married or previously 0.35*** 0.44** married in 2006 (0.08) (0.11) High school GPA higher than 3.0 2.83*** 2.38*** (0.23) (0.20) High school composite math/reading score 1.05*** 1.03*** (0.01) (0.01) Respondent started at a 4-year institution 2.07*** (0.19) Respondent enrolled part time in 2006 0.30*** (0.05) Total student loans (reference category: $1-$10,000) No student loans 1.36* (0.17) Greater than $10,000 2.75*** Total student loans (reference category: $1-$10,000) (0.32) No student loans 1.66*** (0.18) Greater than $10,000 1.66*** (0.17) Constant 1.66*** 0.09*** 0.05*** (0.07) (0.02) (0.02)

Observations 6,710 6,028 5,881

BIC 1,405,419 1,265,093 Difference of 140,326 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

166

Table C.3: Logistic Regression Model for Bachelor’s Degree Attainment by 2012, Testing Number of Credit Cards Variable (Odds Ratios Reported) (1) (2) (3) VARIABLES Credit Card Demographic Full Variable Variables Model

Number of credit cards in own name 0.94* 1.03 0.99 (0.02) (0.03) (0.03) Socioeconomic status (reference category: Middle) Lower 0.75** 0.84 (0.07) (0.09) Upper 1.55*** 1.40*** (0.15) (0.14) Hours worked weekly 2005-06 (reference category: 1-20 hours) Did not work 0.92 1.06 (0.08) (0.11) More than 20 hours 0.46*** 0.58*** (0.04) (0.05) Respondent has biological child, is married or previously 0.35*** 0.44** married in 2006 (0.08) (0.11) High school GPA higher than 3.0 2.84*** 2.38*** (0.23) (0.20) High school composite math/reading score 1.05*** 1.03*** (0.01) (0.01) Respondent started at a 4-year institution 2.06*** (0.19) Respondent enrolled part time in 2006 0.30*** (0.05) Total student loans (reference category: $1-$10,000) No student loans 1.36* (0.17) Greater than $10,000 2.75*** (0.32) Total student loans (reference category: $1-$10,000) No student loans 1.66*** (0.18) Greater than $10,000 1.66*** (0.17) Constant 1.78*** 0.09*** 0.05*** (0.07) (0.03) (0.02)

Observations 6,710 6,028 5,881

BIC 1,405,717 1,265,096 Difference of 140,621 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

167

Table C.4: Logistic Regression Model for Bachelor’s Degree Attainment by 2012, Testing Pays for Tuition with Credit Card Variable (Odds Ratios Reported) (1) (2) (3) VARIABLES Credit Card Demographic Full Variable Variables Model

Whether respondent uses credit card to pay for tuition (reference category: Respondent owns a credit card but does not pay for tuition with credit card) Respondent does not own a credit card 0.85* 0.83* 0.98 (0.06 (0.07) (0.08) Owns credit card, pays for tuition with credit card 0.57*** 0.73* 0.80 (0.07) (0.11) (0.12) Socioeconomic status (reference category: Middle) Lower 0.76** 0.85 (0.07) (0.09) Upper 1.51*** 1.40*** (0.14) (0.14) Hours worked weekly 2005-06 (reference category: 1-20 hours) Did not work 0.97 1.07 (0.09) (0.11) More than 20 hours 0.46*** 0.58*** (0.04) (0.06) Respondent has biological child, is married or previously 0.32*** 0.44** married in 2006 (0.08) (0.11) High school GPA higher than 3.0 2.80*** 2.39*** (0.23) (0.20) High school composite math/reading score 1.05*** 1.03*** (0.01) (0.01) Respondent started at a 4-year institution 2.06*** (0.19) Respondent enrolled part time in 2006 0.30*** (0.05) Total student loans (reference category: $1-$10,000) No student loans 1.36* (0.17) Greater than $10,000 2.75*** (0.32) Total student loans (reference category: $1-$10,000) No student loans 1.66*** (0.18) Greater than $10,000 1.66*** (0.17) Constant 1.95*** 0.11*** 0.05*** (0.10) (0.03) (0.02)

Observations 6,636 5,962 5,879

BIC 1,384,237 1,263,411 Difference of 120,826 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

168

Table C.5: Logistic Regression Model for Bachelor’s Degree Attainment by 2012, Testing Carries Revolving Credit Balance Variable (Odds Ratios Reported) (1) (2) (3) VARIABLES Credit Card Demographic Full Variable Variables Model

Revolving credit balance status (reference category: Respondent owns a credit card but does not carry a revolving balance) Respondent does not own a credit card 0.83* 0.86 0.92 (0.06) (0.07) (0.08) Owns credit card, carries balance from month to month 0.63*** 0.86 0.74* (0.06) (0.10) (0.09) Socioeconomic status (reference category: Middle) Lower 0.75** 0.84 (0.07) (0.09) Upper 1.53*** 1.39** (0.14) (0.14) Hours worked weekly 2005-06 (reference category: 1-20 hours) Did not work 0.92 1.06 (0.09) (0.11) More than 20 hours 0.46*** 0.58*** (0.04) (0.06) Respondent has biological child, is married or previously 0.36*** 0.46** married in 2006 (0.09) (0.12) High school GPA higher than 3.0 2.83*** 2.36*** (0.23) (0.20) High school composite math/reading score 1.05*** 1.03*** (0.01) (0.01) Respondent started at a 4-year institution 2.07*** (0.19) Respondent enrolled part time in 2006 0.30*** (0.05) Total student loans (reference category: $1-$10,000) No student loans 1.34* (0.17) Greater than $10,000 2.79*** (0.32) Total student loans (reference category: $1-$10,000) No student loans 1.65*** (0.18) Greater than $10,000 1.66*** (0.17) Constant 1.99*** 0.11*** 0.06*** (0.11) (0.03) (0.02)

Observations 6,690 6,013 5,867

BIC 1,400,972 1,261,521 Difference of 139,451 in BIC provides very strong support for full model. Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05

169

Table C.6: Logistic Regression Models for Bachelor’s Degree Attainment by 2012 (Odds Ratios Reported) (1) (2) Attained Attained Bachelor’s Bachelor’s Degree Degree – Credit Card VARIABLES – Full Sample Holders

Revolving credit balance status (reference category: Respondent owns a credit card but does not carry a revolving balance) Respondent does not own a credit card 0.92 --- (0.08) --- Owns credit card, carries balance from month to month 0.74* 0.73* (0.09) (0.09) Socioeconomic status (reference category: Middle) Lower 0.84 0.84 (0.09) (0.12) Upper 1.39** 1.49** (0.14) (0.22) Hours worked weekly 2005-06 (reference category: 1-20 hours) Did not work 1.06 1.00 (0.11) (0.15) More than 20 hours 0.58*** 0.54*** (0.06) (0.07) Respondent has biological child, is married or previously 0.46** 0.55 married in 2006 (0.12) (0.22) Respondent started at a 4-year institution 2.07*** 2.05*** (0.19) (0.28) Respondent enrolled part time in 2006 0.30*** 0.30*** (0.05) (0.07) Total Pell Grant amount (reference category: $1-$10,000) No Pell Grants 1.65*** 1.76*** (0.18) (0.27) Greater than $10,000 1.66*** 1.59** (0.17) (0.23) Total student loans (reference category: $1-$10,000) No student loans 1.34* 1.35 (0.17) (0.25) Greater than $10,000 2.79*** 2.81*** (0.32) (0.47) High school GPA higher than 3.0 2.36*** 2.01*** (0.20) (0.25) High school composite math/reading score 1.03*** 1.03** (0.01) (0.01) Constant 0.06*** 0.08*** (0.02) (0.04)

Observations 5,867 2,773 Standard errors in parentheses *** p<0.001, ** p<0.01, * p<0.05 170

Table C.7: Epanechnikov Kernel Propensity Score Matching Model for Bachelor’s Degree Attainment by 2012 (Average Treatment Effect for the Treated (ATT) is reported) OUTCOME VARIABLES Treated Controls Difference Standard Error

Attained bachelor’s degree by 2012 .552 .613 -.061*** .021

*** p<0.001, ** p<0.01, * p<0.05

171

Chapter 4 Figures

Figure C.1: Credit Card Ownership and Behavior, by Socioeconomic Status Credit Card Ownership and Behavior, by Socioeconomic Status (n = 6,438) 100% 9.08 18.98 14.66 80% 35.20 32.76 31.66 60%

40%

49.36 52.58 55.72 20%

0% Lower Middle Upper

No credit card Credit card, does not carry balance Carries balance on credit card

Figure C.2: Credit Card Ownership and Behavior, by Hours Worked Credit Card Ownership and Behavior, by Hours Worked Weekly during 2005-2006 School Year (n = 6,702) 100% 7.76 12.81 21.46 80% 31.7 35.79 60% 32.65

40% 60.54 51.40 20% 45.89

0% Did not work 1-20 hours More than 20 hours

No credit card Credit card, does not carry balance Carries balance on credit card

172

Figure C.3: Credit Card Ownership and Behavior, by Institution Credit Card Ownership and Behavior, by Institution (n = 6,730) 100% 16.65 12.74 80% 27.7 35.45 60%

40% 55.65 51.8 20%

0% Started at less than 4-year institution Started at 4-year institution

No credit card Credit card, does not carry balance Carries balance on credit card

Figure C.4: Bachelor’s Degree Attainment by 2012, by Credit Card Ownership and Behavior Bachelor's Degree Attainment by 2012, by Credit Card Ownership and Behavior (n = 6,732) 100%

80% 55.17 65.03 68.82 60%

40%

20% 44.83 34.97 31.18 0% No credit card Credit card, does not carry Carries balance on credit balance card

No bachelor's degree Bachelor's degree

173

Appendix D: Correlation Matrices

174

Correlation Matrices by Regression Table

Table D.1: Correlation Matrix for Table A.5 Owns Credit Work 4-Year Delayed First Generation Total Total Pell Used Parent Family Class Race Female Card Hours Institution PS Entry Student Loans Grants Loans Status Owns Credit Card 1.00 Class -0.04 1.00 Work Hours 0.12 -0.19 1.00 4-Year Institution 0.04 0.26 -0.24 1.00 Delayed PS Entry -0.02 -0.13 0.08 -0.30 1.00 First Generation Student 0.03 -0.52 0.10 -0.16 0.07 1.00

175 Total Loans 0.07 -0.03 0.01 0.19 -0.10 -0.01 1.00

Total Pell Grants -0.01 -0.11 -0.01 -0.12 0.07 0.08 -0.42 1.00 Used Parent Loans 0.01 0.05 -0.02 0.15 -0.06 -0.04 0.21 -0.24 1.00 Race 0.02 -0.20 0.03 -0.08 0.08 0.10 0.02 0.10 -0.02 1.00 Female 0.06 -0.07 0.06 -0.01 -0.01 0.04 0.05 0.00 0.00 0.01 1.00 Family Status -0.01 -0.10 0.06 -0.12 0.17 0.09 -0.05 0.06 -0.06 0.05 0.10 1.00

Table D.2: Correlation Matrix for Table A.6 Owns 2 or Work 4-Year Delayed First Generation Total Total Pell Used Parent Family Class Race Female More CCs Hours Institution PS Entry Student Loans Grants Loans Status Owns 2 or More CCs 1.00 Class -0.17 1.00 Work Hours 0.18 -0.23 1.00 4-Year Institution -0.09 0.22 -0.27 1.00 Delayed PS Entry 0.02 -0.11 0.12 -0.31 1.00 First Generation Student 0.11 -0.52 0.10 -0.14 0.06 1.00 Total Loans 0.06 -0.06 0.00 0.16 -0.09 0.02 1.00 Total Pell Grants 0.00 -0.11 -0.01 -0.09 0.05 0.08 -0.38 1.00 Used Parent Loans -0.01 0.04 -0.05 0.13 -0.06 -0.02 0.19 -0.22 1.00 Race 0.08 -0.21 0.04 -0.06 0.05 0.11 0.05 0.10 0.01 1.00 Female 0.15 -0.07 0.06 -0.05 -0.01 0.04 0.04 -0.01 -0.02 0.01 1.00 Family Status 0.04 -0.08 0.04 -0.11 0.13 0.07 -0.02 0.08 -0.06 0.03 0.10 1.00

176

Table D.3: Correlation Matrix for Table A.7 Carries Credit Work 4-Year Delayed First Generation Total Total Pell Used Parent Family Class Race Female Balance Hours Institution PS Entry Student Loans Grants Loans Status Carries Credit Balance 1.00 Class -0.17 1.00 Work Hours 0.19 -0.23 1.00 4-Year Institution -0.12 0.22 -0.27 1.00 Delayed PS Entry 0.07 -0.11 0.12 -0.31 1.00 First Generation Student 0.07 -0.52 0.11 -0.13 0.06 1.00 Total Loans 0.12 -0.06 0.00 0.16 -0.09 0.02 1.00 Total Pell Grants 0.01 -0.11 -0.01 -0.09 0.05 0.08 -0.38 1.00 Used Parent Loans 0.03 0.04 -0.04 0.13 -0.06 -0.02 0.19 -0.22 1.00 Race 0.12 -0.21 0.05 -0.06 0.05 0.11 0.05 0.10 0.01 1.00 Female 0.07 -0.07 0.07 -0.05 -0.01 0.04 0.04 -0.01 -0.02 0.01 1.00 Family Status 0.07 -0.08 0.04 -0.11 0.13 0.07 -0.01 0.08 -0.06 0.03 0.10 1.00

177

Table D.4: Correlation Matrix for Table B.3 Paid Tuition Work 4-Year Delayed First Generation Total Total Pell Used Parent Family Class Race Female with CC Hours Institution PS Entry Student Loans Grants Loans Status Paid Tuition with CC 1.00 Class -0.11 1.00 Work Hours 0.10 -0.23 1.00 4-Year Institution -0.11 0.22 -0.27 1.00 Delayed PS Entry 0.01 -0.11 0.12 -0.31 1.00 First Generation Student 0.07 -0.52 0.10 -0.14 0.06 1.00 Total Loans -0.02 -0.06 0.00 0.16 -0.09 0.02 1.00 Total Pell Grants 0.04 -0.11 -0.01 -0.09 0.05 0.08 -0.38 1.00 Used Parent Loans -0.04 0.04 -0.05 0.13 -0.06 -0.02 0.19 -0.22 1.00 Race 0.04 -0.21 0.04 -0.06 0.05 0.11 0.05 0.10 0.01 1.00 Female -0.02 -0.07 0.06 -0.05 -0.01 0.04 0.04 -0.01 -0.02 0.01 1.00 Family Status 0.02 -0.08 0.04 -0.11 0.13 0.07 -0.02 0.08 -0.06 0.03 0.10 1.00

178

Table D.5: Correlation Matrix for Table B.5 Pays Financial Pell 4-Year Emergency Not Enough Financial Incom Student Dependen Financially Work Tuition with Difficult Grant Race Male Institutio Cash Money Literacy e Loans t Status Responsible Hours CC y s n Pays Tuition 1.00 with CC Emergency -0.12 1.00 Cash Financial 0.12 -0.28 1.00 Difficulty Not Enough 0.14 -0.27 0.34 1.00 Money Financial 0.02 0.20 -0.07 -0.12 1.00 Literacy Income 0.11 0.09 0.04 -0.01 0.19 1.00 Student 0.02 -0.03 0.11 0.05 0.08 0.12 1.00 Loans Pell Grants 0.03 -0.12 0.15 0.05 -0.03 0.03 -0.01 1.00 Dependent -0.18 0.01 -0.03 -0.01 -0.12 -0.21 -0.09 -0.16 1.00 Status Financially 0.13 -0.03 0.04 0.00 0.10 0.18 0.03 0.17 -0.39 1.00 Responsible Race 0.08 -0.09 0.06 0.05 -0.10 -0.07 -0.07 0.17 -0.04 0.06 1.00

179 Male -0.02 0.12 -0.08 -0.11 0.24 0.05 0.01 0.00 -0.02 -0.03 0.00 1.00

Work Hours 0.13 0.01 0.10 0.07 0.07 0.58 0.07 0.03 -0.17 0.12 -0.02 -0.05 1.00 4-Year -0.14 0.07 0.01 -0.02 -0.01 -0.08 0.09 -0.07 0.23 -0.23 -0.18 -0.01 -0.08 1.00 Institution

Table D.6: Correlation Matrix for Table C.2 Bachelor's Owns Credit 4-Year Enrolled Total Total Pell Femal Family HS HS Cognitive Class Work Race Degree Card Institution Part Time Loans Grants e Status GPA Score Bachelor's 1.00 Owns CC -0.01 1.00 Class 0.24 -0.06 1.00 Work -0.24 0.12 -0.20 1.00 4-Year College 0.34 0.03 0.25 -0.24 1.00 Enrolled PT -0.26 0.01 -0.12 0.18 -0.24 1.00 Total Loans 0.17 0.06 -0.05 0.01 0.18 -0.11 1.00 Total Pell -0.09 0.00 -0.11 -0.02 -0.12 0.05 -0.43 1.00 Race -0.10 0.04 -0.19 0.03 -0.07 0.08 0.04 0.09 1.00 Female 0.02 0.06 -0.07 0.06 0.00 -0.01 0.04 0.00 0.01 1.00 Family Status -0.11 0.01 -0.09 0.07 -0.08 0.08 -0.05 0.07 0.04 0.07 1.00 HS GPA 0.32 0.02 0.14 -0.16 0.31 -0.19 0.06 -0.08 -0.14 0.13 -0.04 1.00 HS Cog Score 0.31 0.00 0.34 -0.20 0.37 -0.20 0.06 -0.12 -0.23 -0.07 -0.09 0.42 1.00

180

Table D.7: Correlation Matrix for Table C.3 Bachelor's Number of 4-Year Enrolled Total Total Pell Femal Family HS HS Cognitive Class Work Race Degree Credit Cards Institution Part Time Loans Grants e Status GPA Score Bachelor's 1.00 Number of -0.05 1.00 CCs Class 0.24 -0.13 1.00 Work -0.24 0.17 -0.20 1.00 4-Year College 0.34 -0.03 0.25 -0.24 1.00 Enrolled PT -0.26 0.03 -0.12 0.18 -0.24 1.00 Total Loans 0.17 0.08 -0.05 0.01 0.18 -0.11 1.00 Total Pell -0.09 0.01 -0.11 -0.02 -0.12 0.05 -0.43 1.00 Race -0.10 0.07 -0.19 0.03 -0.07 0.08 0.04 0.09 1.00 Female 0.02 0.12 -0.07 0.06 0.00 -0.01 0.04 0.00 0.01 1.00 Family Status -0.11 0.04 -0.09 0.07 -0.08 0.08 -0.05 0.07 0.04 0.07 1.00 HS GPA 0.32 -0.02 0.14 -0.16 0.31 -0.19 0.06 -0.08 -0.14 0.13 -0.04 1.00 HS Cog Score 0.31 -0.06 0.34 -0.20 0.37 -0.20 0.06 -0.12 -0.23 -0.07 -0.09 0.42 1.00

181

Table D.8: Correlation Matrix for Table C.4 Bachelor's Pay for Tuition 4-Year Enrolled Total Total Pell Femal Family HS HS Cognitive Class Work Race Degree with CC Institution Part Time Loans Grants e Status GPA Score Bachelor's 1.00 Pay for Tuition with -0.10 1.00 CC Class 0.23 -0.12 1.00 Work -0.25 0.10 -0.23 1.00 4-Year College 0.32 -0.12 0.22 -0.25 1.00 Enrolled PT -0.26 0.08 -0.12 0.19 -0.25 1.00 Total Loans 0.17 0.00 -0.07 0.01 0.14 -0.10 1.00 Total Pell -0.07 0.02 -0.11 -0.02 -0.09 0.02 -0.38 1.00 Race -0.08 0.04 -0.21 0.04 -0.05 0.09 0.07 0.07 1.00 Female -0.01 0.00 -0.08 0.06 -0.05 0.00 0.04 0.00 0.01 1.00 Family Status -0.10 0.03 -0.08 0.04 -0.08 0.06 -0.03 0.08 0.03 0.08 1.00 HS GPA 0.28 -0.05 0.13 -0.16 0.28 -0.19 0.04 -0.06 -0.14 0.12 -0.05 1.00 HS Cog Score 0.30 -0.10 0.34 -0.23 0.34 -0.21 0.02 -0.09 -0.23 -0.08 -0.07 0.41 1.00

182

Table D.9: Correlation Matrix for Table C.5 Bachelor's Carries 4-Year Enrolled Total Total Pell Femal Family HS HS Cognitive Class Work Race Degree Balance Institution Part Time Loans Grants e Status GPA Score Bachelor's 1.00 Carries -0.05 1.00 Balance Class 0.23 -0.10 1.00 Work -0.24 0.16 -0.20 1.00 4-Year College 0.35 0.00 0.25 -0.24 1.00 Enrolled PT -0.26 0.02 -0.12 0.18 -0.24 1.00 Total Loans 0.17 0.09 -0.05 0.01 0.18 -0.10 1.00 Total Pell -0.09 0.01 -0.11 -0.02 -0.12 0.05 -0.43 1.00 Race -0.10 0.08 -0.19 0.03 -0.07 0.08 0.04 0.09 1.00 Female 0.02 0.08 -0.07 0.06 0.00 -0.01 0.05 0.00 0.01 1.00 Family Status -0.11 0.02 -0.09 0.07 -0.09 0.08 -0.04 0.07 0.03 0.07 1.00 HS GPA 0.32 -0.02 0.13 -0.16 0.31 -0.19 0.06 -0.08 -0.14 0.13 -0.04 1.00 HS Cog Score 0.31 -0.04 0.34 -0.20 0.37 -0.20 0.06 -0.12 -0.23 -0.07 -0.09 0.42 1.00

183

Table D.10: Correlation Matrix for Table C.6 – Model 1 (All Respondents) Bachelor's Carries 4-Year Enrolled Total Total Pell Femal Family HS HS Cognitive Class Work Race Degree Balance Institution Part Time Loans Grants e Status GPA Score Bachelor's 1.00 Carries -0.05 1.00 Balance Class 0.23 -0.10 1.00 Work -0.24 0.16 -0.20 1.00 4-Year College 0.35 0.00 0.25 -0.24 1.00 Enrolled PT -0.26 0.02 -0.12 0.18 -0.24 1.00 Total Loans 0.17 0.09 -0.05 0.01 0.18 -0.10 1.00 Total Pell -0.09 0.01 -0.11 -0.02 -0.12 0.05 -0.43 1.00 Race -0.10 0.08 -0.19 0.03 -0.07 0.08 0.04 0.09 1.00 Female 0.02 0.08 -0.07 0.06 0.00 -0.01 0.05 0.00 0.01 1.00 Family Status -0.11 0.02 -0.09 0.07 -0.09 0.08 -0.04 0.07 0.03 0.07 1.00 HS GPA 0.32 -0.02 0.13 -0.16 0.31 -0.19 0.06 -0.08 -0.14 0.13 -0.04 1.00 HS Cog Score 0.31 -0.04 0.34 -0.20 0.37 -0.20 0.06 -0.12 -0.23 -0.07 -0.09 0.42 1.00

184

Table D.11: Correlation Matrix for Table C.6 – Model 2 (Credit Card Holders) Bachelor's Carries 4-Year Enrolled Total Total Pell Femal Family HS HS Cognitive Class Work Race Degree Balance Institution Part Time Loans Grants e Status GPA Score Bachelor's 1.00 Carries -0.13 1.00 Balance Class 0.23 -0.16 1.00 Work -0.25 0.18 -0.23 1.00 4-Year College 0.32 -0.11 0.22 -0.24 1.00 Enrolled PT -0.26 0.05 -0.12 0.19 -0.25 1.00 Total Loans 0.17 0.12 -0.07 0.01 0.15 -0.10 1.00 Total Pell -0.07 0.02 -0.11 -0.02 -0.09 0.03 -0.38 1.00 Race -0.08 0.13 -0.21 0.04 -0.05 0.09 0.07 0.07 1.00 Female -0.01 0.08 -0.08 0.06 -0.05 0.00 0.04 0.00 0.00 1.00 Family Status -0.09 0.05 -0.07 0.05 -0.08 0.06 -0.03 0.07 0.03 0.08 1.00 HS GPA 0.28 -0.14 0.13 -0.16 0.28 -0.19 0.04 -0.06 -0.14 0.12 -0.06 1.00 HS Cog Score 0.30 -0.14 0.34 -0.23 0.34 -0.21 0.02 -0.09 -0.23 -0.08 -0.06 0.41 1.00

185

Appendix E: Study on Collegiate Financial Wellness

186

Study on Collegiate Financial Wellness

History and Purpose

The Study on Collegiate Financial Wellness (SCFW) is a survey run by the

Center for the Study of Student Life (CSSL) at The Ohio State University. The SCFW was originally named the National Student Financial Wellness Study and has collected data across multiple institutions of higher education 2014 and 2017. According to the study’s website (cfw.osu.edu) the purpose of SCFW “is to gain a more thorough and accurate picture of the financial wellness of students throughout the United States.” The

SCFW focuses on the following research questions as primary drivers of the survey’s inquiry:

“1) How are financial attitudes (including stress), financial behavior, and financial

knowledge related to academic success, decisions to borrow, and career selection?

2) How is student loan debt related to the issues of student financial stress,

enrollment success, decisions to borrow, career selection, investment in

education?

3) What factors (e.g. self-efficacy, financial knowledge, ability to repay, financial

behaviors, family socioeconomic status, seeking financial advice) moderate the

relationships outlined in questions 2 and 3?”

187

There are several key concepts that provide the foundation for the SCFW inquiry.

For example, the SCFW includes questions related to students’ financial education, financial stress, financial management behaviors, student loans, consumer debt, and financial knowledge, among other questions. The SCFW helps participating institutions to understand better how their students are navigating the financial landscape in higher education in order to better meet the needs of their students.

The SCFW data collection in 2014 surveyed students across 51 colleges and universities. These institutions were all in the United States and included public, private, two-year and four-year colleges. A total of 163,714 students received the survey and

18,795 students completed the survey, for a response rate of 12 percent. The response rate varied by institution from a 4 percent to 26 percent.

Project development for the 2017 SCFW focused on expanding data collection across the United States by including a larger number of colleges and universities in the survey sample. Furthermore, the survey instrument included additional and updated questions that included new financial wellness scales, updated student debt measures, and expanded inquiry related to college student credit card use. The credit card use questions incorporated in the 2017 administration of the SCFW are the focus of Chapter 3 in this dissertation.

Professional Affiliation

Over the course of the past two years I have worked as a Graduate Student

Researcher and then a Research Analyst at the Center for the Study of Student Life. As the study’s administrators worked to improve and expand upon the 2014 instrument for

188

the 2017 data collection, they gave me the opportunity to add questions related to the

20127 survey instrument. Having worked on financial wellness related projects using the

Education Longitudinal Study (ELS) data, I recognized this offer as a unique opportunity to investigate further questions related to college student credit card use that I could not answer with the ELS data alone.

Building on the ELS

Chapter 3 of this dissertation investigates the particular practice of college students using credit cards to pay for tuition. College student credit card use is a topic of interest to many parties—as taken up in this dissertation project—the practice of paying for tuition with a credit card has largely been left untouched by empirical studies. The

ELS has a question that asks students whether they have paid for tuition with a credit card in their own name but does not inquire further on students’ reasons for use or their attitudes toward using credit cards as a financial strategy in order to pay for tuition.

Considering this limitation in the ELS data, I developed several questions that asked

SCFW participants about the practice of paying for tuition with a credit card to add to the

SCFW instrument for the 2017 data collection.

I added the following questions to the 2017 SCFW instrument in order to investigate college students who pay for tuition with a credit card as a part of this dissertation:

“Have you ever used a credit card in your name to pay for your college tuition?”

. Yes

. No

189

“What is the primary reason you used credit cards to pay for your college

tuition?”

. My financial aid package didn’t cover all my tuition

. I missed a deadline to apply for financial aid/student loans

. I had to use my tuition money for an emergency

. I didn’t want to take on any more student loans

. Paying with a credit card is easier than other methods

. I always pay some of my tuition with my credit card(s)

. Other (please specify) ______

“Do you think it is a good idea or a bad idea for college students to use credit

cards to pay for educational expenses?”

. Good idea

. Good in some ways, bad in others

. Bad idea

The first two questions are original questions unique to the SCFW. The third question is modeled after a question on the Survey on Consumer Finances run by the

Federal Reserve. Chapter 3 of this dissertation focuses on the first two questions for its empirical analyses but does not incorporate the third question that gets at attitudes toward credit card use for educational expenses.

These questions, in addition to the financial situation questions asked in the

SCFW provided multi-institutional data uniquely catered to investigating the topic of college students paying for tuition with a credit card. Without the SCFW data, this

190

dissertation is limited in its ability to understand this pattern of credit card use among college students. However, I utilize the SCFW and the ELS as complementary datasets that provide a national look at college students who use credit cards to pay for tuition

(ELS) and investigate college students’ reasons for this pattern of credit card use and the situations in which they are most likely to resort to credit cards to pay for tuition

(SCFW). Chapter 3 in this dissertation utilizes both datasets to provide the first empirical look at this credit practice.

SCFW 2017 Data Collection

The SCFW 2017 data is made up of a convenience sample of campuses that elected to participate in the study. CSSL focused its marketing efforts for the survey on higher education groups interested in college students’ financial wellness. CSSL employees recruited SCFW participants by marketing through financial wellness listservs, by networking at a financial wellness symposium held at The Ohio State

University in 2016, and by recruiting participants of the 2014 SCFW data collection.

Because of the nature of these groups, the 2017 SCFW participants are more likely to be colleges and universities that already have a financial wellness program on campus or that want to establish a financial wellness program. CSSL staff worked with a mix of faculty, staff, institutional researchers, and financial aid or financial wellness practitioners at the participating higher education institutions in order to facilitate the data collection process. The final group of survey participants included 64 schools—22 two- year colleges and 42 four-year universities. Because several of these schools administered

191

the survey on more than one of their campuses, the SCFW collected data across 85 campuses throughout the United States.

Depending on the size of each school’s undergraduate student population, the

SCFW was sent to varying sample sizes between different schools. The following table is taken from cfw.osu.edu and shows sample sizes by an institution’s undergraduate student population size.

Table E.1: Sample Size Chart from cfw.osu.edu

The SCFW was open from February 6, 2017 to February 27, 2017 for most schools. A total of 271,191 surveys were distributed to students by email during the data collection process and a total of 27,965 students responded. The overall response rate was

11.5 percent and response rates varied by institution from a low of 3.4 percent to a high of 31.3 percent. Four-year schools (11.6 percent response rate) and schools that offered incentives (11.8 percent response rate) had higher response rates than two-year schools

(6.4 percent response rate) and schools that did not offer incentives (5.5 percent response rate). 43 schools offered incentives of a combined total worth of $20,908. The average incentive amount among these 43 schools was $486.

At the end of the SCFW 2017 data collection, CSSL staff begun the data cleaning and management process. In my role as a Research Analyst at CSSL, I contributed to the 192

data cleaning and management process and integrated the SCFW data into my empirical analyses for Chapter 3 of this dissertation. The SCFW data provide a robust and unique perspective to answering the questions posed in this dissertation and complements the nationally representative nature of the ELS data within the empirical analyses.

193