Financial Education as a Moderator Between Social Learning and Savings Intention/behavior among College Students

MISSING IMAGE

Material Information

Title:
Financial Education as a Moderator Between Social Learning and Savings Intention/behavior among College Students
Physical Description:
1 online resource (84 p.)
Language:
english
Creator:
Parker,William J
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Family, Youth and Community Sciences
Committee Chair:
Gutter, Michael S.
Committee Members:
Wysocki, Allen F
Radunovich, Heidi Liss

Subjects

Subjects / Keywords:
college -- education -- finance -- financial -- moderator -- personal -- savings -- students
Family, Youth and Community Sciences -- Dissertations, Academic -- UF
Genre:
Family, Youth and Community Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
This study utilizes the theory of planned behavior as the theoretical approach to explore whether financial education moderates the relationship between financial attitudes, subjective norms, and perceived behavioral controls, and savings intention/behavior among college students. Three research questions were proposed: 1) When controlling for other factors, will the antecedent constructs of TPB be significantly related to the likelihood that a student is saving/intending to save?; 2) Will the relationships between attitudes, subjective norms and perceived behavioral control (as blocks of variables) and savings/intention to save differ by whether the students have taken a personal finance course?; 3) Will the model that allows for financial education to be a moderator between attitudes, SN?s, PBC?s and savings behavior/intention be the most appropriate model? It was hypothesized that when controlling for other factors, 1) attitudinal factors, 2) subjective norm factors, and 3) perceived behavioral control factors (as individual blocks of variables) would be significantly related to the likelihood that a student is saving/intending to save; The relationship between 4) attitudes, 5) subjective norms, and 6) perceived behavioral controls (as a block of coefficients) and savings behavior/intention differ by whether the students have taken a personal finance course; and 7) The model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral controls and savings behavior/intention would be the most appropriate model. Binary logistic regression analysis, using a forced hierarchical approach, was used to test all hypotheses. Specifically, the Omnibus Test of Model Coefficients was evaluated following each step of the regression to determine the significance level for each block of variables (hypotheses 1-6), as well as the overall significance of the statistical model (hypothesis 7). Results indicate that the antecedent constructs of the TPB were found to be significantly related to savings behavior/intention among college students. Additionally, financial education did not moderate the relationships between financial attitudes, subjective norms, or perceived behavioral controls (as blocks of coefficients) and savings behavior/intention. Finally, the reduced model (without the interaction effects) was found to be the most appropriate model. Important implications are presented for practitioners and researchers.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by William J Parker.
Thesis:
Thesis (M.S.)--University of Florida, 2011.
Local:
Adviser: Gutter, Michael S.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043498:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 FINANCIAL EDUCATION AS A MODERATOR BETWEEN SOCIAL LEARNING AND SAVINGS INTENTION/BEHAVIOR AMONG COLLEGE STUDENTS By WILLIAM J. PARKER A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFI LLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 William J. Parker

PAGE 3

3 To my family, especially my mother, Julie Rawlings, and my father, Charles Parker II

PAGE 4

4 ACKNOWLEDGMENTS I am thankful for everyone that has had a hand in helping me complete my Heidi Radunovich, Dr. Al Wysocki and to Dr. Zeynep Copur for their guidance and support. I would especially like t o thank Dr. Michael Gutter, whose expertise, encouragement, and leadership have proved to be invaluable in helping me complete Finally, am grateful to all of my family, friends, and colleagues for their unconditional love and support

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 2 LITERATURE REVIEW ................................ ................................ .......................... 16 Theoretical Perspectives of Savings Behavior ................................ ........................ 16 The Theory of Planned Behavior (TPB) and Financial Behaviors ........................... 17 Behavior Formation and the TPB Model ................................ ................................ 18 Intention ................................ ................................ ................................ ............ 18 Attitude ................................ ................................ ................................ ............. 19 Subjective Norms ................................ ................................ ............................. 23 Perceived Behavioral Control ................................ ................................ ........... 24 Demographic Variables ................................ ................................ .................... 26 Attitudes, Subjective Norms, and Perceived Behavioral Control as Products of Social Learning ................................ ................................ ................................ .... 27 Financial Education and Savings Behavior ................................ ............................. 29 Financial Education as a Moderator between Attitudes, Subjective Norms, and Perceived Behavioral Control on Savings Behavior ................................ ............ 31 Research Questions and Hypo theses ................................ ................................ ..... 32 3 METHODS ................................ ................................ ................................ .............. 37 Sampling and Data Collection ................................ ................................ ................. 37 Dependent Va riables ................................ ................................ ........................ 38 Independent Variables ................................ ................................ ..................... 38 Demographic Variables ................................ ................................ .................... 42 Intera ction Variable ................................ ................................ .......................... 43 Analysis ................................ ................................ ................................ .................. 44 4 ANALYSIS ................................ ................................ ................................ .............. 47 Sample Description ................................ ................................ ................................ 47 Demographic Variables: ................................ ................................ ................... 47 Dependent Variable ................................ ................................ .......................... 47 Independent Variable s ................................ ................................ ..................... 48

PAGE 6

6 Procedural Summary ................................ ................................ .............................. 49 Binary Logistic Regression Analysis ................................ ................................ ....... 50 B lock 0: Demographics ................................ ................................ .................... 50 Block 1: Attitudes ................................ ................................ .............................. 51 Block 2: Subjective Norms ................................ ................................ ................ 51 Block 3: Perceived Behavioral Controls ................................ ........................... 52 Block 4: Financial Education ................................ ................................ ............ 53 Block 5: Interactions ................................ ................................ ......................... 53 Accept/Reject Hypotheses ................................ ................................ ...................... 53 Question 1 ................................ ................................ ................................ ........ 53 Attitudes and savings behavior/intention ................................ .................... 54 Subjective norms and savings behavior/intention ................................ ...... 54 Perceived behavioral controls and savings behavior/Intention ................... 54 Question 2 ................................ ................................ ................................ ........ 54 Attitudes moderated by financial education ................................ ................ 55 Subjective norms moderated by financial education ................................ .. 55 Perceived behavioral control moderated by financial education ................ 55 Question 3 ................................ ................................ ................................ ........ 56 5 CONCLUSIONS AND IMPLICATIONS ................................ ................................ ... 67 Conclusions ................................ ................................ ................................ ............ 67 Attitudes and Savings Behavior/Intention ................................ ......................... 67 Subjective Norms and Savings Behavior/Intention ................................ ........... 68 Perceived Behavioral Controls and Savings Behavior/Intention ....................... 68 Attitudes Moderated by Financial Education ................................ .................... 70 Subjective Norms Moderated by Financial Education ................................ ...... 70 Perceived Behavioral Control Moderated by Financial Education .................... 71 Financial Education as a Moderator ................................ ................................ 72 Implicat ions ................................ ................................ ................................ ............. 73 Limitations ................................ ................................ ................................ ............... 75 APPENDIX: EXPLANATION OF RACE VARIABLE ................................ ...................... 77 LI ST OF REFERENCES ................................ ................................ ............................... 80 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 84

PAGE 7

7 LIST OF TABLES Table page 3 1 Variables inc luded in hierarchical regression blocks ................................ .......... 46 4 1 Sample descriptive statistics ................................ ................................ ............... 57 4 2 Sample profile by whether students have taken a financial education course .... 58 4 3 Block 0: Omnibus tests of model coefficients ................................ ..................... 60 4 4 Block 0: Variables in the equation ................................ ................................ ...... 60 4 5 Block 1 ................................ ................................ ................................ ................ 60 4 6 Block 1 ................................ ................................ ................................ ................ 60 4 7 Block 2 ................................ ................................ ................................ ................ 61 4 8 Block 2 ................................ ................................ ................................ ................ 61 4 9 Block 3 ................................ ................................ ................................ ................ 62 4 10 Block 3 ................................ ................................ ................................ ................ 62 4 11 Block 4 ................................ ................................ ................................ ................ 63 4 12 Block 4 ................................ ................................ ................................ ................ 63 4 13 Block 5 ................................ ................................ ................................ ................ 64 4 1 4 Block 5 ................................ ................................ ................................ ................ 64 A 1 Other race cross tab ................................ ................................ ........................... 77 A 2 Asian ................................ ................................ ................................ .................. 78 A 3 African A merican ................................ ................................ ................................ 78 A 4 Hispanic ................................ ................................ ................................ .............. 79

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Behavior (TPB) ................................ ........................ 34 2 2 Antecedents of TPB as products of social learning ................................ ............ 35 2 3 TPB model including moderator effect ................................ ................................ 36

PAGE 9

9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science FINANCIAL EDUCATION AS A MODERATOR BETWEEN SOCIAL LEARNIN G AND SAVINGS INTENTION/BEHAVIOR AMONG COLLEGE STUDENTS By William J. Parker August 2011 Chair: Michael Gutter Major: Family, Youth, and Community Sciences This study utilizes the theory of planned behavior as the t heoretical approach to explore whether financial education moderates the relationship between financial attitudes, subjective norms, and perceived behavioral controls, and savings intention/behavior among college students. Three research questions were proposed: 1) When controlling for other f actors, will the antecedent constructs of TPB be significantly related to the likelihood that a studen t is saving/intending to save?; 2) Will the relationships between attitudes, subjective norms and perceived behavioral control (as blocks of variables) an d savings/intention to save differ by whether the students have taken a personal finance course ?; 3) Will the model that allows for financial education to be a moderator between attitudes, and savings behavior/intention be the most appropriate model? It was hypothesized that when controlling for other factors 1) attitudinal factors, 2) subjective norm factors, and 3) perceived behavioral control factors (as individual blocks of variables) would be significantly related to the likelihood that a student is saving/intending to save; The relationship between 4) attitudes, 5) subjective norms, and 6) perceived behavioral controls (as a block of coefficients) and savings behavior/intention differ by whether the students have taken a personal finance

PAGE 10

10 c ourse; and 7) The model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral controls and savings behavior/intention would be the most appropriate model. Binary logistic regression analysis us ing a forced hierarchical approach was used to test all hypotheses. Specifically, the Omnibus Test of Model Coefficients was evaluated following each step of the regression to determine the significance level for each block of variables (hypotheses 1 6), as well as the overall significance of the statistical model (hypothesis 7). Results indicate that the antecedent constructs of the TPB were found to be significantly related to savings behavior/intention among college students Additionally, financial edu cation did not moderate the relationships between financial attitudes, subjective norms, or perceived behavioral controls (as blocks of coefficients) and savings behavior/intention. Finally, the reduced model (without the interaction effects) was found to be the most appropriate model. Important implications are presented for practitioners and researchers.

PAGE 11

11 CHAPTER 1 INTRODUCTION Most college students are at a fascinating and influential stage in life. This stage is known as emerging adulthood (Arnett, 2000) Emerging adulthood is a unique transitional period, from adolescence to young adulthood, and is characterized by a own actions, and becoming increasingly financiall y independent (Arnett, 2004). According to the National Center for Education Statistics (2010), approximately degree, and 100,000 students will graduate with a Professional degree in 2011. Presumably, a large percentage of these graduates will be seeking employment, and thus, financial independence. While the economic conditions and employmen t outlook for college graduates are slowly improving, the current situation is still not good (NCES, 2010). With the unemployment rate hovering around nine percent (Bureau of Labor Statistics, 2011), these graduates will not only be competing for employmen t against fellow classmates, but also many former graduates that are still unemployed or under employed, experienced workers that have been laid off, and former retirees that may be in need of income to support their lifestyle. As of May, 2011, over two mi llion of the unemployed population in the United States is 2011) The uncertainty stemming from not having a job after graduation gives many student debt holders a justifiable reason for concern. For the 2007 2008 school year,

PAGE 12

12 the average amount borrowed by graduate and professional degree seekers was significantly higher (NCES, 2009). On top of having an exorbitant amount of student loan debt, the average college student has 4.6 credit cards and carries a balance of According to Mod Cycle Hypothesis (LCH), consumers want to smooth the marginal utility (happiness) of consumption over a lifetime. One assumption that the theory in fers is that people are expected to borrow against expected future earnings while they are young in order to smooth the marginal utility in a period where they earn less of an income It is common to begin borrowing as college students (student loan debt) with the expectation of beginning a career, earning an income, and acc umulating wealth upon graduation Unfortunately, many students will not find a job right away that will provide them with the level of income they anticipated while incurring the debt which will create challenges in reducing/repaying their debt. For this reason, a large number of college graduates are forced to move back in with their parents. These individuals are often referred to as boomerang children (Okimoto & Stegall, 1987). One thing college students can do to help alleviate some of the stress associated with an uncertain job market is to begin saving money while still in school, just in case their plan to begin their career right away does not pan out. By having a sufficient am ount of money in an emergency fund available a student can prevent having missed credit card payments, loan payments, and cover any other expenses that might arise.

PAGE 13

13 This money can serve as a short term financial buffer, and have a powerful impact on a stud being. Financial well being has been described in literature as the level of financial adequacy and security of an individual or family (Xiao, Sorhalndo, & Garman, 2006). Saving money in college has been found to increase financial sat isfaction and positive academic performance (Xiao, Tang, & Shim, 2009), which, So why are not all college students intentionally setting money aside for an emergency fund and saving for retirement? Previous research on this subject shows that students generally lack basic financial knowledge (Bakken, 1967; Chen & Volpe, 1998; Danes & Hira, 1987; Gutter, Garrison, & Copur, 2010; Jump$tart, 1997, 2002; Kim, 2000; Volpe, Chen, & Pavlicko, 1996). Despite this lack of knowledge, many college students will still come across opportunities to actively learn the skills needed to be financially independent ( Shim Barber, Card, Xiao, & Serido 2009). Some of these occasions will come in t he form of new financial challenges that require prudent decision making and a high level of responsibility (Lyons, Scherpf, & Roberts, 2006). For example, many college students face challenges related to making their financial aid/student loan disbursemen t cover all of the expected and unexpected expenses that are associated with attending college and living away from home over the course of a semester. If the money falls short of the amount needed, students may be at risk of over drafting their checking a ccount, missing debt payments, being sent to collections, racking up penalties and fees, and having to drop out of school. Researchers have found that a higher rate of students drop out as a result of debt and financial pressures than for academic failure (United College Marketing Services, 2006). Students in this

PAGE 14

14 undesirable financial position will either have to find a way to cut back their spending, supplement their income with a job, take out additional loans, ask their parents for financial support, an d/or apply for a credit card(s). If they fail to manage their money efficiently and effectively, saving money will be a difficult task. If college students are indeed unable to save money while in school, it is still important for them to plan on saving a fter college. Although many students do not save money while in school because of a lack of financial resources, others are not saving because of a classic self control issue known Myopia refers to short sightedness specifically in the field of consumer economics, Kivetz and Simonson (2002, p. 200) describe it as regulation necessary to intentionally put money aside for the future. Wh ere do college students learn savings behavior? There are several ways in which college students learn their behavior. For example, they may learn through personal experience (as mentioned previously); they may learn through self study; they may learn abou t savings by taking a personal finance class; they may learn through social learning opportunities (i.e. discussing financial management concepts with parents and/or peers; observing their parents and/or peers engaging in personal financial management acti vities) (Gutter, Copur, & Garrison, 2010). Without taking a formal class on financial education, individuals are at risk of learning false financial knowledge through social learning (Gutter, Copur, & Garrison, 2010). This could come as a result of witness ing friends or family member make unwise financial decisions or just being misinformed by the information they have seen and heard on financial

PAGE 15

15 matters. In terms of saving money while in school some students may be under the impression that if no one else (friends and peers usually) is doing it, it must not be important So the question that arises from this potentially mixed flow of information via social learning and formal financial education is this: h ow does it saving behavior? Thi s research aims to examine whether financial education is a moderator between the factors influencing savings behavior and actual savings behavior among college students. In some instances, saving money may not be feasible for some students. In fact, many need to take out additional loans or use a credit card to meet their financial obligations. For this reason, this study also look s at whether financial education is a moderator between the f actors influencing savings behavior and the intention to save among college students.

PAGE 16

16 CHAPTER TWO LITERATURE REVIEW Theoretical Perspectives of Savings Behavior Different perspectives have been used to frame studies of savings behavior. One perspective researchers have applied to savings behavior is normative theory. Cycle Hypothesis (LCH), which e xamines factors like age/life stage, perceptions of future (income and lifespan), and risk tolerance to predict savings behavior, is an example of a normative theory (Thaler & Benartzi, 2004). The optimization problem that the LCH attempts to solve is the problem of consumption smoothing over a lifetime. Another perspective is Benartzi, 2004, p. 3). These theories differ from normative theories in that they describe how things are, rather than how they ought to be. The descriptive perspective is more of a micro approach, which integrates intrapersonal/psychological factors into the resear ch methodology. An example of a descriptive theory is the behavioral life cycle hypothesis which is similar to the LCH but also incorporate s factors like self control and mental accounting (Shefrin & Thaler, 1988). Other examples of descriptive theories ar e the social learning theory (Bandura, 1977; Churchill & Moschis, 1979; Moschis & Moor, Copur, 2010) and Theory of Planned Behavior (Xiao, 2008; Xiao & Wu, 2008). A th ird perspective that has been used to frame studies on savings behavior is through the lens

PAGE 17

17 improve their decision r & Benartzi, 2004, p.3). If the purpose of the study were to understand how college students are supposed to be saving, a normative theory would be appropriate. If the purpose of the study were to help change savings behavior, a prescriptive theoretical a pproach would be appropriate. However, the purpose of this study is to describe the relationship of financial education on actual savings behavior and intention of college students, which can assist in identify ing factors and relationships that are moderat ed by financial education; for this reason, a descriptive theoretical approach is the best fit for answering the research question. Specifically, the constructs of the theory of planned behavior are operationalized in this study to examine whether formal f inancial education has a moderating effect on the relationship between the factors influencing savings and actual savings behavior among college students. The theory of planned behavior is a suitable theory for studying savings behavior for college studen ts because it allows for the evaluation of the attitudes that may affect their saving s behavior their feelings about the social norm pressure, and the factors that increase or decrease the perceived level of difficulty in depositing money into a savings/i nvestment account regularly. The Theory of Planned Behavior (TPB) and Financial Behaviors The theory of planned behavior (TPB) is an extension of the theory of reasoned action (TRA) (Fishbein & Ajzen, 1975), which is use d to predict and understand human b ehavior (Ajzen, 1991). While these behavioral theories have mostly been utilized in the field of health related behaviors (Schifter & Ajzen, 1985; Sparks, 1994; Povey, Conner, Sparks, James, & Shepherd, 2000), as previously mentioned, it has also been used to frame studies on various forms of financial behavior such as investment decision

PAGE 18

18 making, completing a debt management plan, risky and convenient credit card usage, and saving (East, 1993 ; Shim, Xiao, Barber, & Lyons, 2007; Rutherford & DeVaney, 2009; Xiao & Wu, 2006). Behavior F ormation and the TPB Model In this chapter, the constructs of the TPB (as shown in Figure 2 1 below) are thoroughly discussed. In addition, an explanation is given of how this theory is utilized to in the context of this study to predict and understand savings behavior of college students. Finally, the research questions and hypotheses are presented in this chapter. Intention actual behavior is the intention to perf orm the behavior (Ajzen, 1991). Therefore, to understand and predict how a particular behavioral act manifests, it is important to understand how an individual develops the intention to perform the behavior. In terms of this study, it is important to under stand how college students develop the intention to save money. The TPB identifies three independent factors that interact to formulate the behavioral intention: (1) attitude toward the behavior, (2) subjective norms, and (3) perceived behavioral control ( Beck & Azjen, 1991). Th e intention construct represents degree to which a person has formulated conscious plans to perform or not perform p. 14 ). T he l ikelihood of the behavior gets stronger (Ajzen, 1991, p. 181). The intention construct is the immediate

PAGE 19

19 Attitu de feels that the particular behavior is favorable or unfavorable (Beck & Ajzen, 1991). But exp ectancy by associating the behavior of interest with an anticipated outcome, or perhaps the cost of carrying out the behavior. By associating behaviors with expected consequences, people automatically develop an attitude toward a behavior. In doing so, favorable attitudes are formed for behaviors that have positive anticipated outcomes and unfavorable attitudes are formed for behaviors that have negative anticipated outcomes (Ajzen 1991). According to TPB, the extent that the person feels the attitude is favorable or unfavorable should influence the likelihood that the person perform s the behavior. For example, in a study by Godin, Valois, Lepage, and Desharmais (1992), the researc hers found to have favorable views of smoking were more likely to smoke cigarettes, whereas respondents who had unfavorable views of smoking were less likely to smoke cigar ettes. However, in this study, specific financial dispositions are used examined to in a study by Gutter Copur, and Garrison (2010), a direct link was found between financial dispositions and fin ancial to take financial risks, compulsive buying, materialism, and future orientation influenced behaviors such as budgeting, saving, and risky credit behaviors. This st udy use s the se

PAGE 20

20 same attitudes to examine their relationship with the savin gs behavior of college students. Additionally, this study examine s financial education has an impact on th ese relationship s (Grable, 2000, p. 625). propensity to take risks and savings behavior, using cross secti onal data. Participants size of 178 unmarried people between the ages of 22 64 years of age, the findings show that propensity to take financial risks does indeed influen ce savings. Specifically, Dahlback (1991) found a strong influence on the way people were invested in assets with differing degrees of risk (i.e. savings account vs. stocks). The risk tolerance (1991) study was comprised of 16 d ifferent items, which participants were supposed to read and respond to whether the item correctly describes them. However, Grable and Lytton (1999), provide d a more useful measure of risk tolerance by identifying eight dimensions of willingness to take fi ) guaranteed versus probable gambles, 2) general risk choice, 3) choice between sure loss and sure gain, 4) risk as related to experience and knowledge, 5) risk as a level of comfort, 6) speculative risk, 7) prospect theory, and 8) investme combining these dimensions to measure willingness to take financial risk, Grable and d a more accurate measurement. Their measurement has been utilized extensively throughout consumer economic litera ture It has been used on topics such as differences in financial behavior (Gilliam, Goetz, &

PAGE 21

21 Hampton, 2008) and investment risk & Lytto n, 2009). This disposition is relevant to savings behavior because ty pically the less a person is willing to take financial risk, the less they will be willing to invest in assets that may yield higher returns. primary response to negative event This disposition may be more prevalent in consumers today due to the easily accessible lines of consumer credit (Roberts & Jones, 2001). In a study by Roberts and Jones (2001), credit card usage was found to exacerbate the problem of compulsive buying. When the average college student carrying 4.6 credit cards, this financial disposition being. In the study by Roberts and Jones, 2001), t he researchers used a convenience sample of 406 college students from a small private college in Texas. The researchers used the Compulsive B uying S widely used in the literature consumer behavior. I f not controlled, c ompulsive buying can lead to overspending which can be a form of dissaving or negative saving Dissaving, can be any form of spending of an existing savings account or debt utilization Compulsive buying is relevant to savings behavior because if a person is spending money as a response to negative events or feelings, they may lack the self control that is necessary to intentionally put money into a savings or investment account. In the same vein, the person may be more tempted to utili ze funds that have already been set aside to make compulsive purchases.

PAGE 22

22 In the context of economic psychology and consumer research, Belk (1985, p. 265) define d and are also se material products that are highly valued by a consumer. A materialistic lifestyle has become a relevant part of the culture in many developed countries. If materialists are in fact co nsuming more, they are likely to be saving less. In a study by Watson (2003), the purpose of the research was to find out how people with differing levels of materialism influence their propensity to spend and/or save. Using a sample of 322 households from two geographic areas of Pennsylvania (one urbanized, and one non urbanized) Watson (2003) found that people that scored low on the Richins and Dawson (1992) materialism scale were more likely to invest in stocks, mutual funds, and bonds. Similarly, peopl e with low levels of materialism were more likely to be savers, rather than spenders. This may be a testament to the suggestion that materialistic people would rather purchase status goods than to save for future consumption. These findings were generaliza ble to households across the United States. Future orientation, the final disposition that is examine d extent to which individuals consider the potential distant outcomes of their current behaviors and the extent to which they (Strathman, Gleicher, Boninger, & Edwards, 1994, p. 743).

PAGE 23

23 decisions (Howlett, Kees, & Kemp, 2008). In a study by Howlett, Kees, and Kemp (2008), researchers found that consumers who were rated with a high degree of consideration of future consequences (CFC) on the CFC scale (developed by Strathman, et al. (1994) were more likely to participate in a 401(k) plan. This finding show ed that people who were future oriented are more likely to save in a manner that is consistent with their long term financial goals, which is highly consistent with the LCH model as well as other research on CFC (Joireman Sprott, & Spangenberg, 2005; Modigliani & Brumberg, 1954; Strathman, Gleicher, Bo nninger, & Edwards, 1994). The sample used was 89 graduating seniors from a public university in the south central United States. Subjective Norms Another construct of the t heory is subjective norms. Subjective (or perceived) perceived behavioral expectations of people the individual considers to be most important to them (i.e. parents, siblings, spouse, friends, supervisor, teachers, doctors, etc.). For example, in terms of this study, the perceived expectation to save money while in college by the referent groups comply with the perceived behavioral expectations. For instance, if the referent groups life expect him or her to be saving, the degree to which the student is motivated to comply with this expectation influence s the likelihood that the student is saving or intend ing to save money while in college. For this study, the referent group s that are include d

PAGE 24

24 expectations, collaborate to form a subjective norm (Ajzen, 1991). Perceived Behaviora l Control As mentioned early on in the chapter, the construct that is included in the TPB that distinguishe d it from the original TRA model, is perceived behavioral control. In assessing the limitations of the TRA, Ajzen (1985) found that it lacked the abi lity to account for non motivational factors that influence behavior such as the availability of opportunity and resources. By adding the construct of perceived behavioral control and forming the new model, Ajzen (1991) showed that both motivation and abil ity interact ed be utilized when predicting and understanding behaviors in which people lack complete volitional control. Perceived behavioral control is characterized b y the extent to which a person perceives his or her ability to perform a behavior (Azjen, 1991). Not only does this factor reflect the ease to which a behavior can be performed, it is also assumed to include past experience and expected obstacles (Azjen, 1 their own ability to perform a task is affected by the control beliefs the perceived occurrence of factors that either assist or hinder the performance of a behavior (Azjen, 1991). These beliefs are largely created by consi dering the amount of resources that are accessible and the number of barriers to performing the behavior that a person anticipates. These two factors together generate the control beliefs. The term or desired to perform a specific behavior. This can include money, time, skills, cooperation of others, etc. (Azjen, 1991). It makes sense then that the more resources an individual believes he/she possesses that help facilitate behavioral achievement and the fewer number of

PAGE 25

25 obstacles he/she anticipates, the more perceived control he/she has over the behavior (Azjen, 1991). Perry and Morris (2005), in a study on the role of self perception and knowledge on consumer behavior of individuals between the ages o f 20 and 40 with incomes below $75,000, their perceived level of control over the behavior, their financial knowledge, and their financial resources. Their study used data from the 1999 Freddie Mac Consumer Credit Survey, with a sample size of 10,997, which cannot be generalized to individuals earning moderate and high incomes due to the income stipulation. T his study look s at the student s confidence in their own ability to save money. The perceiv ed behavioral control construct is of utmost importance for this particular study for this reason: if students do not have enough money to cover their financial obligations while allowing for regular deposits into a savings/investment account, they may str ongly desire to save money but yet they are unable to do so because of a lack of resources. In the same vein, if they have large amounts of debt, they may perceive this as an obstacle that inhibits them from saving in the future. Other examples of factors save are her perceived self efficacy, amount of monthly income, amount of debt overall perceived financial knowledge and perceived knowledge of saving and investing less of a need to save money because they are still legally considered dependent. Students with low levels of self efficacy may not feel like they have the ability to manage their finances well, which could negatively influence savings behavior. Students with

PAGE 26

26 lower levels of income may perceive it to be more difficult to save. Students may perceive having high levels of debt to be an obstacle that in hibit s saving for the future. The literature on actual financial knowledge shows that a lack of knowledge is related to debt (Norvilitis, Merwin, Osberg, Roehling, Young, & Kamas, 2006) and financial well being (Lyons, 200 8); However, objective and subjective financial knowledge may have unique effects on decision making and financial behavior (Xiao, 2011), it is necessary to distinguish between the two. Because this construct measures perceived financial knowledge. In this study, perceived knowledge of saving and investing is also used as a perceived behavioral control variable. A student that does n o t feel they are competent enough to save/invest for the future may not feel confident enough to do so. Demographic Variables The developers of TPB assert that demographic factors do not have a direct impact on intention and actual behavior but they affect them indirectly through their imp act on attitudes, subjective norms, and perceived behavioral control (Xiao, 2008, p. 32). In effect, the variances stemming from theses psychological variables reflect the variances of the demographic variables (Xiao, 2008, p. 32). Therefore, demographics are accounted for but not explicitly represented in the TPB model (Ajzen & Fishbein, 1980). However, i n this study, demographics are represented as a separate block of control variables for the purpose of examining their individual relationship with saving s behavior. Previous research on savings behavior shows that certain demographic variables significantly influence the likelihood that a person is saving money regularly. For example, race, gender, marital status, enrollment status, and number of financial

PAGE 27

27 dependents have all been found to be significantly related to financial behavior (Gutter, Copur, & Garrison, 2010; Rutherford & DeVaney, 2009; Williams, 1991; Xiao & Noring, 1993). If significant relationships exist between these demographic variables an d savings behavior, this could provide major implications for financial educators and the indirect perceived behavioral controls, which are products of the social lear ning process. The next section presents the antecedents of the TPB as products of social learning. Attitudes, Subjective Norms, and Perceived Behavioral Control as Products of Social Learning To this point in the chapter, the rationale for using the TPB ha s been discussed with respect to its ability to guide this study in answering the research question Specifically, the constructs of this theory (attitudes, subjective norms, and perceived behavioral control) have been operationalized in the context of thi s study of savings intention and behavior of college students. The next section discuss es how the three major constructs are formed: through the process of social learning. Accor ocial l earning t heory, people learn their att itudes and behaviors by observing the behaviors and attitudes of important referent developing behaviors and attitudes is known as the process of socialization, which beg (McNeal, 1987; Moschis, 1985, 1987). When discussing the n otion of modeling, Bandura (1963 1977) state d that there are four integral sub processes or conditions that must b e met for modeling to be

PAGE 28

28 effective. The first condition is attention; it would be highly unlikely for an individual to imitate another individual without paying attention. The second condition is retention; he attitude or pattern of behavior for a long period of time. The third condition is reproduction, which refers to the ability of the imitator to reproduce the attitude of pattern of behavior. Lastly, a person must be motivated to acquire, retain, and exec ute a learned attitude or behavior. This motivati on, heory of planned b ehavior, can be a direct result of the the beliefs about the expected outcome. The s ocial l earning t heory has also been exten ded to the field of consumer finance. Ward (1974, p. 2) define d young people acquire the skills, knowledge, and attitudes relevant to their functioning in ature has been expanded to include things such as acquiring and developing values, attitudes, norms, skills, behaviors, motives, and knowledge which are related to consumption and financial management (Rettig & Mortenson, 1986). As mentioned in Gutter an d Garrison (2008, p. 74 75), empirical results by some of the leading researchers in the field of behavioral peers, school, family, and the media with family being the pr (Fox, Bartholomae, & Gutter, 2000; Lee & Hogarth, 1999; Shim, Xiao, Barber, & Lyons, 2008; Xiao et al., 2007). It is important to note, however, that financial socialization is far more inclusive than just learning how to be an e acquiring and developing values, attitudes norms, knowledge, and behaviors that

PAGE 29

29 contribute to the financial viability and well et al. 2009). Aside from the social learning aspe behavioral control; this construct also includes features of endowments. For the purpose of this study, the term endowments refers not only to some demographic factors but also factors such as whether or not stude nts have a part time job, have enough money to save, are still dependent upon their parents for financial support, etc. Financial Education and Savings Behavior attitudes, and/or beh Bartholomae, & Lee, 2005). Empirical results in the literature on financial education and savings shows a direct relationship between financial education and savings. For example, when eval uating the effectiveness of the High School Financial Planning Program (HSFPP), Danes, Huddleston Casas, and Boyce (1999), found that upon completion of the curriculum, about 40% of the students began writing financial goals for saving money. Three months later, the researchers conducted a follow up and found that 60% of the students had changed their savings patterns, with 80% reporting that they are saving for needs/wants, and 20% reporting that they save every time they receive money (Danes, Huddleston C asas, & Boyce, 1999). This study used a sam ple of 418 high school student and the data was collected by contacting high school teachers who were interested in using the HSFPP curriculum. In another study, Lyons, (2008) used a sample of 29,759 college stude nts from 10 campuses located in the Midwest to study the effects of financial education on risky financial behavior. The results indicated that college students were significantly less likely to engage in risky

PAGE 30

30 financial behavior, such as accruing credit c ard debt and missing credit card payments, if they were currently enrolled in, or have already taken a formal course in personal finance (Lyons, 2008); Engaging in these risky financial behaviors can make depositing money into a savings/investment account difficult. Thus, financial education may reduce risky financial behaviors and improve positive financial behaviors, but other personality factors also influence these things. Gutter and Renner (2006) conducted a study at a large Midwestern university to ev attitudes, and/or behaviors. The original intention of the study was to collect data right before the class began, immediately after the class had ended, and again, nin e months after the class had ended. The researchers wanted to test for longitudinal effects of the financial education. Although they received survey data from 270 students, collected in fall of 2006, researchers were only able to collect all three surveys from 77 of the students. Although the results of the study showed that six percent of the students who completed the follow up changed from having no intention in opening a retirement account to actually opening one, the results also showed that students were found to be less likely to save nine months after the study than they had been before taking the who were required to take the course, as well as students who elected to take the course, although between group comparisons were not made. Because of the small sample size, conclusions and generalizations are limited. Bernheim, Garrett, & Maki, (2001), in a study on state financial mandates and their affect on financial beha vior, compared curriculum mandates on financial choices of

PAGE 31

31 adults between ages 30 and 49 The goal of the study was to distinguish whether financial choices varied by whether the student was exposed to a state financial education mandate and the number of years since exposed to the mandate Data used for the analysis was based on a cross sectional survey of households from 1995. The age ranged between 30 and 49, which meant they presumably completed high school between 1964 and 1983. 2000 tele phone surveys were completed. Surveys collected information on standard economic and demographic information, including self reported rates of saving. The survey also asked respondents to name the state they graduated high school in, as well as other infor mation regarding their exposure to financial education. The final sample showed significant discrepancies in single individuals, females, non whites, and people who did not complete high school. Median household income was also much higher than the benchma rk (approximately 35% higher). Results of t h e study found the first systematic evidence of long term financial behavioral effects of the financial education curriculum mandates (Bernheim, Garrett, & Maki, 2000). Results also found that the mandates increas ed exposure to financial education, which ultimately elevates rates of savings and wealth accumulation during their adult lives (Bernheim, Garrett, & Maki, 2000). As with most studies of this nature, Bernheim, Gar rett, and Maki (2000), discuss the self rep ort survey method as a potential concern. This study adds to the body of evidence that education may be a powerful tool for increasing personal savings. Financial Education as a Moderator between Attitudes, Subjective Norms, and Perceived Behavioral Contro l on Savings Behavior T he most important question regarding this study is whether financial education acts as a moderator between psychological factors and financial behaviors. This

PAGE 32

32 question was posed in Gutter and Garrison (2008) regarding the relationshi p between norms and personal financial behaviors (p. 85). Gutter, Copur, and Garrison (2010) found that financial education is significantly related to both actual and perceived financial knowledge. In the same study, the researchers found financial educat ion to be significantly related to financial behaviors such as budgeting and saving (Gutter, Copur, & Garrison, 2010, p. 53). This study suggests that formal financial education via classes in school or in the community and informal education via social learning have a significant effect on subjective norms, dispositions, and behavior. However, given the mixed information people may receive from the classroom versus their other agents of socialization, it is u nclear how these forms of education work toge ther to impact behavior. Thus, an important question to ask is whether financial education acts as a moderator between the products of socialization (attitude s and motivational factors) and savings behavior/intent among college students (as shown in figure 2 3) This study looks to evaluate all three of the main TPB constructs and how financial education might influence their relationships with behavioral intention, as well as actual behavior. Research Questions and Hypotheses Q1. When controlling for other factors, will the antecedent constructs of TPB be significantly related to the likelihood that a student is saving/intending to save? H1. When controlling for other factors, attitudinal factors are significantly related to the likelihood that a student is saving/intending to save. H2. When controlling for other factors, subjective norm factors are significantly related to the likelihood that a student is saving/intending to save.

PAGE 33

33 H3. When controlling for other factors, perceived behavioral control factors are significantly related to the likelihood that a student is saving/intending to save. Q2. Will the relationships between attitudes, subjective norms and perceived behavioral control (as blocks of variables) and savings/intention to save differ by whether the students have taken a personal finance course (as shown in figure 2 3) ? H4. The relationship between attitudes (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course. H5. The rel ationship between subjective norms (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course H6. The relationship perceived behavioral controls (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course Q3. Will the model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral controls and savings behavior/int ention be the most appropriate model? H7. The model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral controls and savings behavior/intention will be the most appropriate model.

PAGE 34

34 Figure 2 T heory of P lanned B ehavior (TPB)

PAGE 35

35 Figure 2 2. Antecedents of TPB as products of social learning

PAGE 36

36 Figure 2 3. TPB model including moderator effect

PAGE 37

37 CHAPTER 3 METHODS Sampling and Data Collection The data used in this study was collected durin g the spring and fall semesters of 2008 as part of a larger research project, which examined the impact of financial education on financial behaviors on college students. The sample population for this study is college students above the age of 18 that rec eived their diploma from a high school located within the United States. A total of 15 universities from across the United States were chosen at random from the six policy categories identified by Gutter, Copur, and Garrison, (2010) based on the 2004 Natio nal Counc il on Economic Education report; the researchers used a stratified sampling technique. Randomized lists of student email addresses were obtained from each one of the universities selected, and occasionally, complete lists of email addresses for st udent populations were given to the researchers for use in their study. A cross sectional research design was utilized through the use of an online times over the span of one month. These three emails contained information regarding a chance for an incentive (every one thousandth survey that was completed would receive a $100 gift card), as well as an informed consent document. The emails also contained a link to the su rvey. Before the participants began the survey, they had to once again confirm their consent on the informed consent document. In all, 172,412 students received the three emails, with 16,876 students responding to the survey, which yields a response rate o f approximately 9.79 %. Respondents who did not attend high school in the United States, recipients of a GED, students who were home

PAGE 38

38 schooled, and students who failed to indicate which state they graduated high school were removed from the sample, resulting in a sample size of 15,797 students. From this sample, cases with missing values for the dependent and independent variables were removed, leavi ng a final sample size of 10,006 Dependent Variables Savings There is only one dependent variable for this s tudy: saving/intention to save. It is comprised of actual savings behavior and savings intention. In other words, it is made up of students who are saving and students who are not saving but intend to save at some point in the near future. Two questions we re combined to create this variable. The first question was regular basis into some sort of account (includes employer plans, mutual funds, yes or no question designed to measure actual savings behavior The second question was portion of the Students who answered yes to either question were included in this category. In other words, students who identified themselve s as saving regularly and students who were planning on saving (next month, in six months, or upon graduation) Independe nt Variables Following the TPB model, the independent variables for this study represent

PAGE 39

39 behavioral control on saving. Other independent factors include whether students h ave taken financial education, as well as the interaction of the attitudes, subjective norms, and perceived behavioral control variables by whether the students have taken financial education. Attitudes Four financial dispositions were measured in this st udy. These attitudinal variables make up Block 1 of the variables in Table 4 1. the statements on this page comes closest to the amount of financial risk that you are willi fin These were both statements, whereas the next three dispositions were measured using scales. The second disposition, compulsive buying, was measured using the C ompulsive seven statements that represent specific feelings and behaviors related to compulsive buying. The data collectors used six of the seven statements, leaving off the i tem that because of a lack of student responses to this item. (Gutter, Copur, & Garrison 2010, p. 27). Responses to the items ranged from 1= very often, to 5=never. Mor e severe compulsive buying feelings and behaviors will result in lower scores on the scale.

PAGE 40

40 reliability was reasonable. The third disposition, materialism, was measured usin (1992) Materialism Scale. The scale analyzes the concepts of centrality, happiness, and success, the three factors related to materialism. The researchers used 15 of the original 18 items on the scale, all ranging from 1=strongly agre e, to 5=strongly disagree. ve scores on the scale could range from 15 to 75, with lower scores representing lower levels of materialism and higher scores representing higher levels of materialism. The inter item reliability was found to be relatively high among college students (alp ha = .86). Future orientation was the final financial disposition measured. Researchers utilized the Consideration of Future Consequences (CFC) scale, d eveloped by Strathman et al. (1994). This 12 distant versus immediate consequences of possible behaviors. Answer choices ranged calculated to be .78. Subjective Norms. Three subjective norm variables were measured in this study. The first subjective norm variable, perceived norms of parents, was measured by asking This variab le was The second subjective norm variable, perceived

PAGE 41

41 norms of peers, was measured by asking the quest are variable, the perceived norm of a typical student, was measured by asking the question Do you think the typical student at your school is saving/investing in These variables were intentionally coded in this way for comparison purposes. If students did not believe their friends, parents, and or the typical student was savi Per ceived Behavioral Control. Six perceive behavioral control variables were measured in this study. These six variables make up block three of the analysis. The first perceived behavioral control variable financial self efficacy was assessed using six item factor, which was developed by Tang (1992). Responses were on a 7 point, Likert type scale (1=strongly disagree, 7=strongly agree). Higher perceptions of self efficacy will result in higher scores on the scale. Results from the six items were combined to create an overall self by This variable was coded into yes=1 and no or unsure=0. The third perceived behavioral control variable, debt,

PAGE 42

42 was measured by asking the students several questions related to the amount of debt they are currently carrying (i. e. credit card debt, student loan debt, car loan, mortgage, etc). The totals from these questions were combined to form overall debt l evels. The answer options for this question were in these ranges: $0, $1 $999; $1,000 $4999; $5,000 +; Not sure. Another perceived behavioral control variable, income, was measured by asking the students how much they receive in monthly income. Answer choices for this question were: $0; $1 $499; $500 $999; $1,000 +. Perceived knowledge of savings/investing was also measured as a perceived behavioral control saving/investing co better, the same, and worse. This variable was coded to compare students who rated their savings/investing knowledge to be better or worse than the typical student. The final perceived beha vioral control variable measured in this study is perceived overall financial knowledge. This was measure by asking students to rate their own level of financial knowledge rate your financial choices were better, the same, and worse. This variable was coded to compare students who rated their overall financial knowledge to be better or worse than the typical student Demographic Variables The demographic variables used in this study are race, sex, marital status, financial dependents, and enrollment status. Race was measured by asking the students to choose the race/ethnicity that best describes them. Answer choices te (non

PAGE 43

43 White (non Hispanic) Sex was measured by asking students to identify whether they The variable was coded to represent males=1 and female 0. Marital status was measured by asking whether students to choose the option that best This variable was coded to represent single=1 and other=0. The financial dependents variable was measured by asking the students how many financial dependents they are responsible for. This variable was coded to represent whether a student was responsi ble for at least one person other than Enrollment status was measured by asking the students if they were enrolled full time or part time. This variable was coded: full time=1, part time=0. Inter action Variable Financial Education seminar on personal finance issues in [the] community, religious institute, or 4H in education course ed Each financial education variable is controlled for on its own, and if significant, is multiplied by all of the other variables to form the interaction variables.

PAGE 44

44 Analysis To find out if each block of predictor varia bles described by the constructs of TPB (attitudes, subjective norms, perceived behavioral control ) are significantly related to the likelihood that a student is saving/intending to save ceteris paribus (Hypothesis 1, 2, 3 ), a binary logistic regression i s utilized using a forced hierarchical regression technique The significance of the blocks of variables is examined. The order and content of the blocks of coefficients is based on the constructs of the TRA and TPB model, with the addition of the financia l education variable demographics block of variables, and the interaction block, which are added for the purpose of this study. The only part of the block entry order that can be interchangeable is the attitude block and the subjective norm block. Analyse s will be ran for each order to ensure the order of these blocks do not change the outcome of the statistical equation. To find out if the relationship between attitudes, subjective norms, and perceived behavioral controls (as blocks of coefficients) and s avings or intent to save differs by whether the students have taken a financial education course ceteris paribus (Hypothesis 4, 5, 6), the interaction block of variables are tested for significance. In order to determine whether the model that allows for financial education to be a moderator is the most appropriate model (Hypothesis 7 ), the full model is compared to the reduced model. In the reduced model, the relationship between the predictor variables and the dependent variable (savings/intention to sav e) is examined controlling for each additional block of variables The full model which includes all of the reduced model blocks of variables, also includes the interaction effect of the independent variables on whether they have taken a course on person al finance This is done by including interaction terms between the independent variables and the

PAGE 45

45 independent variable, indicating that financial education has a moderating effect on the predictor variables.

PAGE 46

46 Table 3 1 Variables included in h ierarchical r egression b locks Block 0 Block 1 Block 2 Block 3 Block 4 Block 5 Demographics Attitudes Subjective Norms Perceived Behavioral Control Financial Education Moderator Variable Race Financial Risk Tolerance Parents Save Norm Financial Self efficacy Finan cial Education in High School Attitudes x Financial Education Gender Materialism Peers Save Norm Listed as dependent tax return Financial Education in the Community x Financial Education Marital Status Compulsive Buying Typical Student S ave Norm Perceived Financial Knowledge x Financial Education Have financial dependent(s) Future Orientation Perceived Savings Knowledge Enrollment Status Debt Income

PAGE 47

47 CHAPTER 4 ANALYSIS Sample D escription Demographic Variables: Upo n running the descriptive statistics for this sample (as shown in Table 4 1 and Table 4 2) of college students used in this study, the results show ed that this nationally representative sample is reasonable when compared to national demographic averages of college students (NASPA, 2008). The majority of students were enrolled full time (93.2% national average 90.58% ). Most students were white (82.79%, national average 69.8% white), most were female (67.4% national average 62.7%), and the majority were unm arried (83.13%, national average, 58.1% single). 7.97% of the sample had one or more financial dependents. Another characteristic of this sample is that 41.2% were taught a personal finance, either through a course in high school or in the community: 35% t ook a course in high school and 7.2% took a course in the community. All of these demographic variables have been identified in the literature as being related to financial behaviors, and specifically savings behavior For that reason, these variables are controlled for in this study. Other variables that can be considered demographic in nature are included in the analysis in the perceived behavioral control block of variables due to their effects regularly. De pendent Variable Saving and intending to save The results of the c ross 2 distribution show ed that there were overall significant differences in whether students were saving or intending to save by whether students have taken a personal finance

PAGE 48

48 2 = 8.634, p<.01) This finding was expected due to th e previous literature on financial education and its effects on savings behavior Independent Variables Attitude v ariables 2 test show ed that there are significant differences in certain financial dispositions by whether students have taken a course on personal finance. For example, there was a significant difference in students that were 2 2 =42.367, 2 =24.217, p<.05) All of these attitudinal variables have been identified in the literature as being related to financial behaviors, and specifically savings behavior. Thus, they are controlled for in the regression analysis. Subjective n orm v ariables The results of the 2 test show that there are significant differences in the subjective norm variables by whether students have taken a course on personal finance. The first subjective norm variable, perceived parents saving norm, significantly differs by financial educati 2 =15.367, p<.001). Another subjective norm variable, peers saving norm, significantly differs by financial education 2 =10.293, p<.0 1). Lastly, typical student saving s norm was also found to significantly differ s =16 .729, p<.001). Perceived b ehavioral c ontrol v ariables 2 test show ed significant differences in perceived behavioral control variables by whether students have taken a course on personal finance. Of the perceived behavioral control vari ables, significant differences were found among students that perceived their overall financial 2 =40.669, p<.001), and worse than a 2 =84.058, p<.001), by financial education. Significant differences were

PAGE 49

49 students who made $500 stu p<.001). Lastly, overall financial self efficacy was found to significantly differ by financial education. Procedural Summary The remaining portion of this chapter present s the results of the statistical tests that were used to examine the proposed hypotheses Hypotheses 1, 2, and 3, were tested by using a binary logistic regression, through a forced hierarchical approach to find out if the blocks of predictor variables significa ntly influenc e the likelihood that a student is saving or intending to save, ceteris paribus. H ypotheses 4, 5, and 6 were tested by using the same binary logistic regression through a forced hierarchical approach to find out whether the relationship betwe en savings or intent to save differed by whether students have taken a personal finance course in the community Finally, hypothesis 7 was tested by comparing the full and reduced models using the Omnibus Tests of Model Coefficients, which showed the signi ficance of the overall model after each step of the regression. The Omnibus Tests of Mod el Coefficients presented the 2 statistic, degrees of freedom, and the p value for the block of variables and the complete model for each step of the regression. Overall this model was a good fit for this study. The significance level for block entries remains a p<.001 significance le vel until the Block 4 entry, the financial education variables. The significance of the overall model remains a p<.001

PAGE 50

50 significance level until Block 5 ( the moderator variables ) is entered into the regression. The following paragraphs discuss the results o f these statistical tests. Binary Logistic Regression Analysis As mentioned previously, a binary logistic regression through a forced hierarchical approach was used to examine all three of research questions either by proving or disproving the hypothes es. In essence, six different binary logistic regressions were completed in order to predict the probability that a college student is saving, given each set of predictor variables, but this was done in a single hierarchical process. Blocks of vari ables w ere entered as groups that were entered into the regression in successive stages. As each block was subsequently added to the model, the previous block(s) remain ed in the model as control factors. The blocks of variables that were included in the regressio n include: Block 1 = Demographic variables; Block 2 = Attitude variables; Block 3 = Subjective norm variables; Block 4 = Perceived behavioral control variables; Block 5 = Financial education variables; and Block 6 = Interaction variables. These variables a re listed in (Table 3 1). The order that the blocks were entered into the model was determined by the constructs of the TRA and TPB models, with the other blocks being added in a way that made the most logical sense for this study. The only part of the blo ck entry order that can be interchangeable is the attitude block and the subjective norm block. Upon running the analysis, the order in which these blocks were entered did not change any of the results. Therefore, the attitude block was entered into the st atistical equation before the subjective norm block. Block 0: D emographics Demographic factors were added first in order to control for these variables for each of the regressions. As a whole, Block 0 was found to be a significant predictor of

PAGE 51

51 savings beha vior/intention ( 2 =37.05, p<.001) thus, leaving the entire model to be 2 = 37.05, p<.001) Both the model and block were positively related to savings behavior/intention. Of the variables in this block, only race (p<.001) and the constant (p <.001 ) were found to be significant and positively related to saving/intention to save more likely to be saving or intending to save. Block 1: A ttitudes The next block added to the r egression was one of the constructs from the original TRA model a ttitudes while controlling for demographics. When added, Block 1 was 2 =129.122, p<.001), keeping the entire model significant as well 2 =166.172, p<.001). Both the model and block were positively related to savings behavior/intention. Of the variables contained in this block, co mpulsive buying (p<.001), risk tolerance (willing to take no risk) (p<.001), and the constant (p<.001) were all found to be significant predictors of saving/intending to save. The risk tolerance variable (no risk) was found to have a negative relationship with savings/intention to save. This means that the more likely a student is unwilling to take any financial risk, the more likely the student wi ll be saving or intend to save The compulsive buying variable, which is reverse scored on the CBS scale but po sitively related to savings behavior/intention shows that students who score lower on the scale are more likely to be saving or intending to save. Block 2: S ubjective N orms The s ubjective norms block which is also a construct in the original TRA, was add 2 = 9 2.791 p<.001), 2 =258.963, p<.001). Both the model and

PAGE 52

52 block were positively related to savings behavior/intention. Of the variables contained in this block, perceived parents saving (p<.001), perceived friends saving (p<.001), and the constant (p<.001) were found to be significant predicto rs of saving/intending to save. This means that students who perceived their parents and/or peers to be sav ing were more likely to be saving or intending to save. Block 3 : Perceived B ehavioral C ontrols Next the TPB construct of perceived behavioral controls was added into the equation. 2 =88.601, p<.001), 2 =347.576, p<.001). Both the model and block were positively related to savings behavior/intention. The 13 predictors that were added into the model with this block doubled the degrees of freedom from 13 to 26. Of the variables contained in this block, overall financial knowledge (better than typical student) (p<.001), perceived saving knowledge (bet ter than typical student) (p<.01 ), and the constant (p<.01) were found to be significant predictors of savings behavior/intention. Having a better overall financial knowledge than the typical student was positively related to savings behavior/intention, indicating that students who perceived their were found to be more likely to be saving/intending to save. The perceived savings knowledge variable has a negative Beta, indicating an inverse relationship with savings behavior/inte ntion. This means that students who perceived their savings knowledge to saving.

PAGE 53

53 Block 4 : Financial E ducation The next group of variables added to the model was the f inancial education block. This was the last step of the reduced model. When added to the model, the block was 2 =353.07, p<.001) and positively related to savings behavior/intention Of the two variables in this block, personal finance taught in the community was found to be a signif icant predictor (p<.0 5 ) and positively related to savings behavior/intention The constant remained significant as well (p<.01) Block 5 : Interactions The las t block that was added to the regression was the interaction block. This block represents all of the variables that were added to the reduced model to create the full regression model. This block added all 26 variables, multiplied by the moderator variable personal finance in the community. This was done because personal finance in the community was found to be significant and personal finance in school was not, which also caused Block 4 to not be significant. When this block was added to the model, it was not found to be significant; In addition the model was no longer significant 2 =383.489, p<.0 84 ). None of the variables added to this group were found to be significant predictors of saving/intention to save. Accept/Reject Hypotheses Q uestion 1 When controlling for other factors will the antecedent constructs of TP B be significantly related to the likelihood that a student is saving/intending to save?

PAGE 54

54 Attitudes and s avings b ehavior/ i ntention Hypothesis 1 stated that when controlling for other factors, the block of attitudinal factors is significantly related to the likelihood that a student is saving/intending to save. When controlling for other factors, the block of attitudinal factors was found to be significantly related to the likelihood that a student is saving/intending to save 2 =129.122, p<.001). Thus, accept H1. Subjective n orms and s avings b ehavior/ i ntention Hypothesis 2 stated that when controlling for other factors, the block of subjective norm factors is significantly related to the likelihood that a student is saving/in tending to save. When controlling for other factors, the block of subjective norm factors was found to be significantly related to the likelihood that a student is saving/intending to save 2 =92.791, p<.001). Thus, accept H2. Perceived b ehavioral c ontrols and s avings b ehavior/Intention Hypothesis 3 stated that when controlling for other factors, the block of perceived behavioral control factors is significantly related to the likelihood that a student is saving/intending to save. When controlling for other fact ors, the block of perceived behavioral control factors was found to be significantly related to the likelihood that a student is 2 =88.613, p<.001). Thus, accept H3. Q uestion 2 Will the relationships between attitudes, subjective norms and perceived behavioral control (as blocks of variables) and savings/intention to save differ by whether the students have taken a personal finance course, ceteris paribus?

PAGE 55

55 Attitudes m oderated by f inancial e ducation Hypothesis 4 stated the relation ship between attitudes (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course. Reject H4 ( see explanation below) Subjective n orms m oderated by f inancial e ducation Hypothesis 5 stated that the relationship between subjective norms (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course. Reject H5 (see explanation below) Perceived b ehavioral c ontrol m oderated by f in ancial e ducation Hypothesis 6 stated that t he relationship perceived behavioral controls (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course. Reject H6 (see explanation below) Expl anation for H4, 5, 6: When added to the model, Block 5 (the financial education block) was not found to be a significant predictor. In this block the personal finance in the community variable was found to be significant (p<.001), but the personal finance in school variable was not. Because of this, the personal finance in the community variable was used as the moderating variable with all of the other predictor variables that were used in the reduced model forming interaction variables. None of the inter action variables were found to be significant, so H4, H5, and H6, are all rejected.

PAGE 56

56 Q uestion 3 Will the model that allows for financial education to be a moderator between attitudes subjective norms, and perceived behavioral controls and savings behavior/ intention be the most appropriate model? Appropriate m odel Hypothesis 7 stated that the model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral controls and savings behavior/intention will be the most appropriate model. The model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral controls and savings behavior/intention was not found to be the most appropriate model. The reduced model (and all of the individual blocks within the reduced model) was found to be significant. After the final block of variables, Block 6 interaction variables, was added, the mode l lost its significance. Thus, H7 is rejected

PAGE 57

57 Table 4 1. Sample d es criptive s tatistics Variable Mean St. Dev Dependent Saving or Intent 96.86% .17444 Independent Fin. Risk Tolerance No Risk 16.51% .37130 Above avg. and substantial risk 25.57% .43625 Materialism 43.23 5.83857 Compu lsive Buying 24.47 3.97800 Future Orientation 16.51 .37130 Parents save norm 71.59% .45101 Peers save norm 23.73% .42542 Typical student save norm 6.47% .24592 Perceived Saving Knowledge Better than typical student 17.16% .37709 W orse than typical student 52.20% .49954 Overall Financial Knowledge Better than typical student 12.30% .32843 Worse than typical student 59.51% .49090 Financial Self efficacy 30.11 7.84701 Debt $1 $999 6.08% .23897 $1,000 $4999 5.10% .22001 $5,000 + 10.57% .30747 Not sure 1.87% .13534 Income 34.98% 10823 $1 $499 16.22% 10823 $500 $999 11.71% 10823 $1,000 + 34.98% 10823 turn 66.38% .47244 Race 82.79% .37751 Sex 32.60% .46876 Marital Status 83.13% .37452 Financial dependents 7.97% .27089 Enrolled Full Time 93.20% .25176 Financial Education 41.20 % .478

PAGE 58

58 Table 4 2. Sample profile by whether students have taken a fi nancial education course Mean/Prop Significance Test Variable Have taken a class Have not taken a class Dependent Saving or Intent Yes 97.5% 96.50% No 2.5% 3.50% Independent Fin. Risk Tolerance No Risk 15.1% 17.30% Other 84.9% 82.70% Above avg. and substantial risk 25.6 % 25.60% Other 74.4% 74.40% Materialism Minimum (15), Maximum (75) 43.38 43.14 t=2.017* Compulsive Buying Minimum (6), Maximum (30) 24.62 24.42 Future Orientation Minimum (13), Maximum (25) 22.64 22.57 Parents save norm Yes 73.9% 70.30% Other 26.1% 29.70% Peers save norm Yes 25.5% 22.80% Other 74.5% 77.20% Typic al student save norm Yes 7.5% 5.80% Other 92.2% 94.20% Perceived Saving Knowledge Better than typical student 59.0% 48.50% Other 41.0% 51.50% Worse th an typical student 12.8% 19.60% Other 87.2% 80.40% Financial Self Efficacy 85.191*** Minimum (6), Maximum (42) 30.11 29.04 Fin. Knowledge Better than typical student 63.6% 57.30% Other 36.4% 42.70% Worse than typical student 8.4% 14.40% Other 91.6% 85.60%

PAGE 59

59 Table 4 2. Continued Mean/Prop Significance Test Debt $1 $999 5.9% 6.20% Othe r 94.1% 93.80% $1,000 $4999 4.9% 5.20% Other 95.1% 94.80% $5,000 + 9.6% 11.10% Other 90.4% 88.90% Not sure 2.0% 1.80% Other 98.0% 98.20% Income $1 $499 35.5% 34.70% Other 64.5% 65.30% $500 $999 14.0% 17.40% Other 86.0% 82.60% $1,000 + 9.9% 12.7%. Other 90.1% 87.30% Listed as a dependent on return 70.6% 64.10% Other (no, unsure) 29.4% 35.90% Race White 85.3% 81.40% Other 14.7% 18.60% Sex Male 36.3% 30.60% Female 63.7% 69.40% Marital S tatus Single 86.8% 81.10% Other 13.2% 18.90% Financial dependents None 93.1% 91.40% One or more 6.9% 8.60% Enrollment Status Full time 94.2% 92.60% Part t ime 5.8% 7.40%

PAGE 60

60 Table 4 3. Block 0 : Omnibus t ests of m odel c oefficients Chi square Df Sig. Step 1 Step 37.050 5 .000 Block 37.050 5 .000 Model 37.050 5 .000 Table 4 4. Block 0: Variables in the equation B Sig. Exp(B) Step 1a Financial depen dent .384 .067 .681 race .720 .000 2.055 sex .210 .098 1.234 Marital status .019 .911 .981 Enrollment status .063 .774 1.065 Constant 2.800 .000 16.445 Table 4 5 Block 1: Chi square Df Sig. Step 1 Step 129.122 5 .000 Block 129.122 5 000 Model 166.172 10 .000 Table 4 6 Block 1: B Sig. Exp(B) Step 1a Financial dependent .327 .122 .721 race .556 .000 1.743 sex .030 .820 .970 Marital status .031 .858 .970 Enrollment status .003 .990 .997 materialism .004 .712 .996 Compulsive buying .069 .000 1.071 Future orientation .067 .039 .936 Willingness to take financial Risk (No risk) 1.269 .000 .281 (Above Average/Substantial) .052 .743 .949

PAGE 61

61 Table 4 7 Block 2 Chi square Df Sig. Step 1 Step 92.791 3 .000 Block 92.791 3 .000 Model 258.963 13 .000 Table 4 8 Block 2 B Sig. Exp(B) Step 1a Financial dependent .315 .141 .730 race .351 .008 1.421 sex .045 .736 .956 Marital status .114 .513 .893 Enrollment status .058 .795 1.060 materialism .002 .803 .998 Compulsive buying .055 .000 1.056 Future orientation .063 .052 .939 Willingness to take financial Risk (No risk) 1.103 .000 .332 (Above Average/Substantial) .153 .336 .858 Parents Save Norm 1.009 .000 2.742 Friend Save Norm .654 .001 1.924 Typical Student Save Norm .169 .554 1.185

PAGE 62

62 Table 4 9 Block 3 Chi square Df Sig. Step 1 Step 88.613 13 .000 Block 88.613 13 .000 Model 347.576 26 .000 Table 4 10. Block 3 B Sig. Exp(B) Step 1a Financial dependent .362 .099 .696 race .326 .016 1.385 sex .091 .506 .913 Marital status .023 .900 1.023 Enrollment status .132 .574 1.141 materialism .004 .673 .996 Compulsive buying .013 .436 1.014 Future orientation .032 .320 .968 Willingness to take financial Ri sk (No risk) 1.007 .000 .365 (Above Average/Substantial) .234 .146 .792 Parents Save Norm .920 .000 2.510 Friend Save Norm .601 .002 1.823 Typical Student Save Norm .219 .448 1.245 Listed as a dependent on .139 .332 .870 Overall Financial Knowledge (better than average student) .078 .620 .925 (Worse than average student) .510 .001 1.665 Monthly Income ($1 499) .189 .171 1.208 ($500 999) .192 .298 1.212 ($1,000 +) .036 .876 1.036 Debt ($1 999) .249 .301 1.283 ($1,000 4,999) .161 .550 1.175 ($5,000+) .201 .371 1.223 (Not Sure) .425 .166 .654 Financial Self efficacy .016 .104 1.016 Perceived Saving Knowledge (Worse than average student) .289 .068 1.335 (Better than average student) .465 .002 .628

PAGE 63

63 Table 4 11 Block 4 Chi square Df Sig. Step 1 Step 5.494 2 .064 Block 5.494 2 .064 Model 353.070 28 .000 Table 4 12. Block 4 B Sig. Exp(B) Step 1a Financial dependent .383 .081 .682 race .334 .014 1.397 sex .114 .406 .892 Marital status .026 .886 1.027 Enrollment status .138 .555 1.148 materialism .004 .683 .996 Compulsive buying .014 .419 1.014 Future orientation .033 .317 .968 Willingness to take financial Risk (No risk) 1.005 .000 .366 (Above Average/Substantial) .240 .136 .787 Parents Save Norm .920 .000 2.509 Friend Save Norm .584 .002 1.793 Typical Student Save Norm .211 .466 1.235 Listed as a dependent on .134 .349 .875 Overall Financial Knowledge (better than average student) .066 .6 75 .936 (Worse than average student) .191 .167 1.211 Monthly Income ($1 499) .199 .283 1.220 ($500 999) .039 .865 1.040 ($1,000 +) .234 .333 1.263 Debt ($1 999) .147 .585 1.158 ($1,000 4,999) .191 .395 1.211 ($5,000+) .448 .144 .639 (Not S ure) .505 .001 1.658 Financial Self efficacy .016 .110 1.016 Perceived Saving Knowledge (Worse than average student) .266 .093 1.305 (Better than average student) .460 .002 .631 High school personal finance .130 .329 1.139 Community personal fin ance .528 .048 1.695 Constant 2.613 .010 13.640

PAGE 64

64 Table 4 13 Block 5 Chi square Df Sig. Step 1 Step 30.419 26 .251 Block 30.419 26 .251 Model 383.489 54 .084 Table 4 14. Block 5 B Sig. Exp(B) Step 1a Financial dependent .316 .173 .729 race .372 .008 1.450 sex .111 .430 .895 Marital status .005 .978 1.005 Enrollment status .184 .441 1.202 materialism .004 .734 .996 Compulsive buying .015 .395 1.015 Future orientation .024 .477 .977 Willingness to take financial Risk (No risk) .993 .000 .370 (Above Average/Substantial) .310 .058 .733 Parents Save Norm .937 .000 2.552 Friend Save Norm .524 .007 1.689 Typical Student Save Norm .160 .583 1.173 Listed as a dependent on .136 .356 .873 Overall Fina ncial Knowledge (better than average student) .056 .726 .945 (Worse than average student) .213 .134 1.237 Monthly Income ($1 499) .214 .262 1.238 ($500 999) .005 .982 1.005 ($1,000 +) .147 .546 1.158 Debt ($1 999) .196 .493 1.216 ($1,000 4,999 ) .195 .404 1.216 ($5,000+) .551 .073 .576 (Not Sure) .523 .001 1.686 Financial Self efficacy .014 .156 1.014 Perceived Saving Knowledge (Worse than average student) .308 .058 1.360 (Better than average student) .402 .009 .669 High school fin ancial education .113 .396 1.120

PAGE 65

65 Table 4 14. Continued B Sig. Exp(B) Community financial education 10.463 .095 1.674 Community financial education raceDUM .529 .474 .589 Community financial education INsexDUM .012 .986 1.012 Community finan cial education INstatusDUM 1.313 .155 3.718 Community financial education Financial dependents .437 .603 .646 Community financial education Enrollment status .862 .488 .422 Community financial education risk tolerance (no risk) .316 .636 729 Community financial education risk tolerance (above average/substantial) 1.503 .192 4.493 Community financial education materialism .028 .554 .972 Community financial education compulsive buying .041 .663 .960 Community financial educat ion Future orientation .282 .162 .754 Community financial education parents saving .253 .696 .777 Community financial education friends saving 1.411 .237 4.098 Community financial education typical student saving 16.829 .997 5.263 Communi ty financial education Financial self efficacy .038 .401 1.039 Community financial education financial knowledge (better) .490 .516 .612 Community financial education Overall financial knowledge (worse) .262 .763 .770 Community financial educ ation Saving Knowledge (better) 1.306 .256 .271 Community financial education Saving knowledge (worse) 2.244 .057 .106 Community financial education listed as dependent on .111 .888 .895 Community financial education ($ 1 999) 17.997 .997 6.313

PAGE 66

66 Table 4 14 Continued B Sig. Exp(B) Community financial education ($1,000 4,999) .704 .488 .495 Community financial education Debt ($5,000+) .509 .612 1.664 Community financial education Debt (not sure) 18.849 .998 4 .351 Community financial education Income ($1 499) 1.161 .124 .313 Community financial education Income ($500 999) .660 .459 .517 Community financial education Income ($1,000+) .013 .991 1.013 Constant 2.306 .027 10.038

PAGE 67

67 CHAPTER 5 CONCLUS IONS AND IMPLICATION S Conclusions This study examined financial education as a potential moderator between the products of social learning and savings intention/behavior among college students. The study was based on data collected in 2008 via an online su rvey. The sample population consisted of college students from across the United States that graduated from high school s with in the United States. Logistic regression analysis was used to identify whether financial education moderated the relationships bet ween blocks of variables representing financial attitudes, subject norms, and perceived behavioral control and savings behavior. The results of this research present several important conclusions. Attitudes and Savings Behavior/Intention Hypothesis 1 stat ed that when controlling for other factors, the block of attitudinal factors is significantly related to the likelihood that a student is saving/intending to save. The results of the binary logistic regression showed that the block of attitudinal variables ( Block 1) is significantly related to the likelihood that a student is saving/intending to 2 =129.122, p<.001). For this reason, hypothesis 1 was accepted. Al though compulsive buying and willingness to take financial risk (no risk variable; negative r elationship on savings behavior/intention) were the only variables found to be significant as individual factors, the y were significant enough to cause the group of attitudes as a whole to be significant ly related to sa vings behavior/intention. It can be c oncluded, then, that overall, attitudinal factors are significantly related to the likelihood that a student is saving or intending to save Additionally, students who rated low in compulsive buying and students who were not willing to take financial risk were found to

PAGE 68

68 be more likely to be saving or intending to save. These findings were consistent with Gutter, Copur, and Garrison, (2010), who found compulsive buying and n o willingness to take financial risk to be significant predictors of savings behavior. The findings were also consistent with the theory of planned behavior mode l (Ajzen, 1991) as shown in chapter two. Subjective Norms and Savings Behavior/Intention Hypothesis 2 stated that when controlling for other factors, the block of subjective norm f actors is significantly related to the likelihood that a student is saving/intending to save. The results of the binary logistic regression showed that the block of subjective norm variables (Block 2) is significantly related to the likelihood that a stud ent is 2 =258.963, p<.001). For this reason, hypothesis 2 was accepted. The two variables that were shown to be significant predictors of savings/intention to save, individually, were parents saving norm and peers saving norm. It can be concluded, then, that overall, students who perceive their parents and peers to be saving/investing regularly are more likely to be saving/intend to save. The result s on perceived norm influe nces and social norm pressure. This conclusion theory of planned behavior model (1991). Perceived Behavioral Controls and Savings Behavior/Intention Hypothesis 3 stated that when controlling for other factors, the block of perc eived behavioral control factors is significantly related to the likelihood that a student is saving/intending to save. The results of the binary logistic regression showed that the block of perceived behavioral control variables (Block 3) is significantly related to the 2 =88.601, p<.001). For this

PAGE 69

69 reason, hypothesis 3 was accepted. Individual perceived behavioral control factors that were found to be significant predictors of savings behavior/intentio n were overall financial knowledge (better than typical student), and perceived savings knowledge (better than typical student; negatively related to savings behavior/intention). Although these were the only two coefficients found to be significant factors they were significant enough to make the entire block of perceived behavioral control factors, significantly related to the likelihood that a student is saving or intending to save It can be concluded, then, that overall, that students who believe their overall financial knowledge to intending to save. Alternatively, students who view their saving/investing knowledge to esting knowledge are less likely to be saving or intending to save. The finding regarding overall perceived financial knowledge is consistent with the literature on financial knowledge (Perry & Morris, 2005 ; Xiao, 2011), which states that financial knowle behavior, is significantly related to savings behavior. Further, this study found that financial knowl edge were more likely to be saving or intending to save, which is consistent with the findings in Gutter, Copur, and Garrison, (2010). Students who perceive their knowledge of savings/investing may be less likely to be saving in college because of their c onfidence in receiving a high rate of return in future years may make saving during college seem unnecessary These findings are consistent with LCH model (Modilgliani & Brumberg, 1954), which suggests that the

PAGE 70

70 amount of savings is largely influenced by a expected rate of return. The results of these three hypotheses answer the first research question, that when controlling for other factors, the antecedent constructs of TPB are significantly related to the likelihood that a student is saving or in tending to save. Attitudes Moderated by Financial Education Hypothesis 4 stated the relationship between attitudes (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course. This hypoth esis was rejected. The results of the binary logistic regression showed that financial education (as a block of variables) is not a significant predictor of savings behavior/intention. Of the two variables in this block, personal finance taught in school w as not found to be a significant predictor; however, personal finance taught in the community was found to be a significant predictor. For this reason, personal finance taught in the community was used as the moderating variable. When multiplied by each in dependent attitudinal variable, none of the resulting interaction variables were found to be significant predictors of savings behavior/intention Therefore, while financial education in the community is not significant moderat or on the relationship betwee n attitudes, as a block of coefficients ( or as individual variables ), and savings behavior/ intention it does increase the likelihood of savings/intention to save among college students. Subjective Norms Moderated by Financial Education Hypothesis 5 stated that the relationship between subjective norms (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course. This hypothesis was rejected. The results of the binary logistic regression sho wed that financial education in the community was found to be a

PAGE 71

71 significant predictor of savings behavior/intention. However, when multiplied by each independent subjective norm variable, none of the resulting interaction variables were found to be signifi cant. Therefore, while financial education in the community is not significant moderator on the relationship between subjective norms as a block of coefficients (or as individual variables), and savings behavior/intention, it does increase the likelihood of savings/intention to save among college students. Perceived Behavioral Control Moderated by Financial Education Hypothesis 6 stated that t he relationship perceived behavioral controls (as a block of coefficients) and savings or intent to save differs by whether the students have taken a personal finance course. This hypothesis was rejected. The results of the binary logistic regression showed that financial education in the community was found to be a significant predictor of savings behavior/intention. However, when multiplied by each independent perceived behavioral control variable, none of the resulting interaction variables were found to be significant. Therefore, while financial education in the community is not significant moderator on the relation ship between perceived behavioral control as a block of coefficients (or as individual variables), and savings behavior/intention, it does increase the likelihood of savings/intention to save among college students. The results of hypotheses 4, 5, and 6 answer the second research question, that when controlling for other factors, the relationships between attitudes, subjective norms, and perceived behavioral controls (as blocks of variables) and savings/intention to save does not differ by whether the st udents take a personal finance course. This was the main research question and the primary purpose of this study. However, o ne point of concern arises with the results of this research question and the rejection of these

PAGE 72

72 hypotheses : the financial education in the community variable is full of heterogeneity. The item used on the survey that was used to collect data ever taken a course, program, or seminar on personal finance issues in [the] community, religious institute, or 4H in o Students who answered yes to this question could have attended a broad range of personal finance events ranging from one 30 minute presentation to a course or program that spans over several months. The she e r amount of heterogeneity that likely exists in this variable could possibly be the reason why there were no significant relationships between the interaction variables and savings behavior/intention. This is discussed further in the limitations section. Financial Ed ucation as a Moderator Hypothesis 7 stated that the model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral controls and savings behavior/intention will be the most appropriate model. The re sults of the binary logistic regression showed that each block of variables in the reduced model were significant; thus, indicating the model was a good fit for the study. When the last block of variables (Block 5, the interaction variables) was added to t he reduced model to create the full/interaction model, the model was no longer significant. It can be concluded, then, that the model that allows for financial education to be a moderator between attitudes, subjective norms, and perceived behavioral contro ls and savings behavior/intention is not the most appropriate model the reduced model is a better fit. For this reason, hypothesis 7 was rejected

PAGE 73

73 Implications Several important implications come about as a result of this study. For parents, it is import financial behavior. They should be mindful of their financial decision making, not just for their own well being, but for the future financial well being of their children. Als o, parents should also take note that personal financial education through the community was found to be a significant predictor of savings behavior/intention. They may wish to send their children to a money camp or other community event that focuses on pe rsonal finance. For practitioners, while this study was not successful in proving that a financial education moderates the relationship between products of social learning and savings behavior/intention, it did reveal many other significant relationships. For example, financial education taught in the community was found to be a significant predictor of savings behavior/intention. Thus practitioners that work in communities, such as county extension agents, 4H agents/volunteers, and other community professi onals/volunteers, should feel encouraged, and know their work is paying off. If these community practitioners are looking for programs or workshops that have been found to encourage positive financial behaviors, they may look to starting a personal finance program/event. Also, practitioners should use the findings to re evaluate their teaching style, the material they are using, and the overall effectiveness of their program. For policy makers, an important implication from this study is for them to potenti ally increase funding for community based financial education outreach. They may decide to hire more community professionals and provide them with larger budgets in order to implement more personal finance programs.

PAGE 74

74 For researchers, future research on fin ancial education should be more specific when operationalizing a financial education variable. To further research this topic or to improve upon this study, researchers should look at some of the aspects of these community based financial programs. For exa mple, th ey should narrow their search by inquiring about the topics covered in the program, as well as the length and/or intensity of the program As mentioned previously, programs implemented in the community can consist of a 30 minute presentation on sav ing for financial goals to a program with higher intensity that covers concepts like savings vehicles and compounding interest over the course of a few months. Simply asking if the students have attended a course/program/seminar on personal finance just al lows for too much heterogeneity in the variable. Additionally, the content, intensity, standards, and /or the educational methods used to teach personal finance to students in high school may need to be reconsidered to improve effectiveness, as suggested by Mandell and Klein (2009). Researchers should determine which methods get the best results (i.e. online stock market interaction game instead of or in addition to a traditional lecture on stock market.) Making a class relevant, interactive, and fun, may in motivation to learn the material, which may be a factor that is offered more in the community setting more so than in the classroom. Another implication for researchers is the need to investigate the issue of financial education as a m oderator between the products of social learning and savings behavior longitudinal data. While financial education in school was not found to be a significant factor, it may become significant over time as the students move on in life, becoming more respon sible for their own financial behavior. In other words, as suggested by

PAGE 75

75 Mandell and Klein (2009), the information learned in a high school personal finance position, financially, to utilize what they have learned. A further concept to investigate in the context of the study of savings behavior is motivation. Are students who attend financial education events in the community going to these events because they desire th e information being presented (as opposed to students who take a class in school)? This idea has been supported in literature in a study by Mandell and Klein (2007), who found motivation to be a factor in increasing financial literacy in adults. Limitation s As mentioned previously, the way in which the financial education in the community variable is defined could be a reason for concern. By asking whether or not a student has attended a financial educational event outside of school without inquiring furthe r about specific details such as which topics were covered, the intensity of the program, and the duration of the program, this allows for a fair amount of heterogeneity. bro ad range of financial programs and events. O ne of the major limitations of the study stems from the use of a self report survey as the data collection instrument. With self reported data, it can be difficult at times to determine whether the data is repre senting whether financial education affects actual behavior or just the way people answer questions regarding financial behavior. In this study, it is a possibility that students who have taken a course in financial education may feel like they should be s aving, so they may feel the need to portray that they know more about saving or are acting on the knowledge, regardless of their actual

PAGE 76

76 knowledge/behavior. Self report can also lead to students over estimating their abilities, when comparing themselves to others. For example, when reporting their knowledge of savings/investing, students may not want to admit that they are average or below average compared to their peers or the typical college student. It can be a difficult task for some to willingly classif y their self, or even to perceive their self, as being less than average at something Another limitation of this study stems from the sample demographics. The sample is comprised of 82.79% white students which is approximately 13% more than the national a verage of 69.8% (NASPA, 2008). Another demographic discrepancy that may have affected the implications is the high proportion of students who are unmarried. There were 83.13% in this sample, compared to the national average, which is 58.1 %. This means the results of the analyses may be biased towards white students as well as single students. Furthermore, while a large dataset was utilized for the analysis, upon breaking down the race variable into African American, Hispanic, Asian, and other, and using Wh ite as the reference variable, the results showed that there were very few respondents from each category that had taken a class in the community (see Appendix A ) If these numbers were larger, comparisons could be made using four different racial/ethnic c ategories; however, because the purpose of this study was to first and foremost address whether financial education was a moderator between products of social learning and savings behavior/intention, the variable was coded dichotomously (White=1, other=0) to include more students in the sample that have taken financial education in the community.

PAGE 77

77 APPENDIX EXPLANATION OF RACE VARIABLE Findings This appendix will present the cross tabs analysis of the race variables. In the analysis, the race variable was c oded as White=1, other=0. When race was broken down further into African American, Hispanic, Asian, and Other, while using White as the reference variable, there were not enough from each group that had taken a personal finance course in the community to p rovide meaningful results that are relevant to this study. Have you ever taken a course, program, or seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. Other Race Cross tabulation Ta ble A 1 Other race cross tab Count Other Race Total .00 1.00 Have you ever taken a course, program, or seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. 0 9466 311 9777 1 969 39 1008 Total 10435 350 10785

PAGE 78

78 Have you ever taken a course, program, or seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. Asian Cross tabulation Table A 2. Asian Count Asian Tota l .00 1.00 Have you ever taken a course, program, or seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. 0 9283 494 9777 1 956 52 1008 Total 10239 546 10785 Have you ever taken a course, program, or seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. African American Cross tabulation Table A 3. African American Count African American Total .00 1.00 Have y ou ever taken a course, program, or seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. 0 9457 320 9777 1 918 90 1008 Total 10375 410 10785

PAGE 79

79 Have you ever taken a course, program, o r seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. Hispanic Cross tabulation Table A 4 Hispanic Count Hispanic Total .00 1.00 Have you ever taken a course, program, or seminar on personal finance issues in your community, religious institution, or 4H -in other words not through school. 0 9303 474 9777 1 958 50 1008 Total 10261 524 10785

PAGE 80

80 LIST OF REFERENCES Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckmann (Eds.), Action contro l: From cognition to behavior p. 1 39 Heidelberg: Springer. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human D ecision Processes 50, 179 211. Ajzen, I., & Madden, T. J. (1986). Prediction of goal directed behavior: Attitudes, intentions, and perceived behavioral control. Journal of Experimental Social Psychology 22, 453 474. Allen, M. W., Edwards, R., Hayhoe, C. R., & Leach, L. (2007). Imagined interactions, family money management patterns and coalitions, and attitudes toward money and credit. Journal of Family and Economic Issues 28, 3 22. Arnett, J. J. (2004). Emerging adulthood: The winding road from late tee ns through the twenties. Oxford: Oxford University Press. Bandura A. of imitative responses. Journal of Personality and Social Psychology, 1 (6), 589 5 95. Bandura, A. 1977. Self ef ficacy: Toward a unifying theory of behavioral change. Psychological Review 84, 191 215. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs: Prentice Hall xiii, p.617 Bernheim, B. D., Garrett, D. M., & Maki, D. M. (2001). Education and saving: The long term effects of high school financial curriculum mandates. Journal of Public Economics, 80, 435 465. Bureau of Labor Statistics (2011). Unemployment rate for college grads Retrieved February 11, 2011 from http://www.bls.g ov/news.release/empsit.t04.htm Churchill, G. A., & Moschis, G. P. (1979). Television and interpersonal influences on adolescent consumer learning. Journal of Consumer Research 6 (1): 23 35. Danes, S. M., & Hira, T. K. ( 1987). Money management knowledge of college students. Journal of Student Financial Aid, 17 (1): 4 16. Danes, S. M., Huddleston Casas, C. A., & Boyce, L. (1999). Financial planning curriculum for teens: Impact evaluation. Financial Counseling and Planning, 10 (1): 2 5 37

PAGE 81

81 Journal of Consumer Research, 19 (3): 459 469 Fox, J. J., Bartholomae, S., & Gutter, M. S. (2000). What do we know about socialization? Consumer Interest Ann ual, 46 p. 1 2. Garrison, S. (2010). Gender differences in financial socialization and will ingness to take financial risks Working Paper, Graduate t hesis p. 1 70. Gutter, M. S., Copur, Z., Garrison, S. (2010). Financial capabilities of college students from states with varying financial educa tion policies. National Endowment of Financial Education Full Report. 2019; Thirty eighth edition. Grable, J. E. (2000). Financial risk tolerance and additional factors that affect risk taking in everyday money matters. Journal of Business and Psychology, 14 (4), 625 630. Grable, J. & Lytton, R. H. (1999). Financial risk tolerance revisited: The development of a risk assessment instrument. Financial Services Review, 8 163 181. Gutter, M.S. & Garrison, S. (2008). Perceived norms, financial education, and col lege student credit card behavior. Journal of Consumer Education 24, 73 88. Hilgert, M. A., Hogarth, J. M., & Beverly, S. G. (2003). Household financial management: The connection between knowledge and behavior. Federal Reserve Bulletin, 89 309 322. Joir eman, Jeff, David Sprott, and Eric Spangenberg. 2005. Fiscal Responsibility and the Considerationof Future Consequences. Personality and Indivi dual Differ ences, 39 (32): 1159 1168. Jump$tart Coalition. (1997). High school seniors lack financial smarts show s survey. American Savings Education Council News Release. Jump$tart Coalition. (2002). Financial literacy declining among 12th graders, coalition urges states to include personal finance in curriculum standards. Retrieved November 10, 2006, from http://www.jumpstartcoalition.org Kim, J. 2000. The effects of workplace financial education on personal finances and work outcomes. Unpublished doctoral dissertation Virginia Polytechnic Institute and State: Black sburg. Lee, J., & Hogarth, J. (1999). The price of money: consumers' understanding of APRs and contract interest rates', Journal of Public Policy and Marketing, 18 (1): 66 76. Lyons, A. C., Scherpf, E., & Roberts, H. (2006). Financial education and communic ation between parents and children. Journal of Consumer Education, 23 64 76.

PAGE 82

82 Mandell, L., & Klein, L. S. (2007). Motivation and financial literacy. Financial Services Review 16 p.106 116. McNeal, J. V. (1987). Children as consumers: insights and implica tions Lexington, MA: Lexington Books. Modigliani, F. & Brumberg, R. (1954): 'Utility analysis and the consumption function: An interpretation of cross section data'. In: Kurihara, K.K (ed.): Post Keynesian Economics Moschis, G. P. (1985). The role of fam ily communication in consumer socialization of children and adolescents. Journal of Consumer Research, 11 (4): 898 913. Moschis, G. P. (1987). Consumer socialization: A life cycle perspective Lexington, MA: Lexington Books. Moschis, G. P., & Moore, R. (198 4). Anticipatory consumer socialization. Academy of Marketing Science 12 (4): 109 123. NCES, National Center for Educational Statistics (2010). Projection of Educational Statistics to 2020. Retrieved April 25, 2011 from http://nces.ed.gov/datab/quickstats Norvilitis, J. M., Merwin, M. M., Osberg, T. M., Roehling, P. V., Young, P. & Kamas, M. M. (2006). Personality factors, money attitudes, financial knowledge, and creditcard debt in college students. Jou rnal of Applied Social Psychology, 36 (6), 1395 1413. Okimoto, J. D., & Stegall, P. J. (1987). Boomerang kids: How to live with adult children who return home. Boston, MA: Little, Brown. Povey, R., Conner, M. C., Sparks, P., James, R., & Shepherd, R. (20 00). The theory of planned behavior and healthy eating: Examining additive and moderating effects of social influence variables. Psychology and Health, 14 991 1006. Richins, M. L., & Dawson, S. (1992). A consumer values orientation for materialism and its measurement: Scale development and validation. Journal of Consumer Research, 19 (3): 303 316. Rutherford, L. G., & DaVaney, S. A., (2009). Utilizing the theory of planned behavior to utilize convenience use of credit cards. Journal of Financial Counseling and Planning, 20 (2), 48 63. study of usage rates and trends 2009. Wilkes Barre, PA: Author. Schifter, D. E., & Ajzen, I. (1985). Intention, perceived control, and weight loss: An application of the theory of planned behavior. Journal of Personality and Social Psychology, 49, 843 851.

PAGE 83

83 Shim, S., Barber, B. I., Card, N. A., Xiao, J. J., Serido, J. (2009). Financial socialization of first year college students: The roles of p arents, work, and education. Journal of Youth and Adolescence, 39 (12), 1457 1470. Sparks, P. (1994). Food choice and health: Applying, assessing and extending the theory of planned behavior. In D. R. Rutter & L. Quine (Eds.), Social Psychology and Health: European Perspectives (pp. 25 45). Hants, England: Avebury. Strathman, A., Gleicher, F., Boninger, D. S., & Edwards, C. S. (1994). The consideration of future consequences: Weighing immediate and distant outcomes of behavior. Journal of Personality of Soci al Psychology 66 (4): 742 752 Tang, L. (1992). The development of a short money ethic scale: Attitudes toward money and pay satisfaction revisited. Personality and individual Differences, 19 (6), 809 816. United College Marketing Services. (2006). College Credit Card Statistics. Retrieved 20 May, 2010 from http://www.ucms.com/college credit card statistics.htm nd measurement. Journal of Consumer Policy, 11 419 433. Volpe, R. P., Chen, H., & Pavlicko, J. J. (1996). Personal investment literacy among college students: A survey. Financial Practice and Education, 6 (2): 86 94. Xiao, J. (2008). Applying Behavior Theo ries to Financial Behaviors in J. Xiao (Ed.), Handbook of Consumer Finance Research p 69 81 Kingston RI : Springer. Xiao, J. J., Sorhalndo, B., & Garman, T. E. (2006). Financial behaviours of consumers in credit counseling. International Journal of Consumer Studies, 30 (2), 108 121. Xiao, J. & Wu, J. (2008). Completing Debt Management Plans in Credit Counseling: An Application of the Theory of Planned Behavior. Journal of Financial Counseling and Planning. 19 (2), 29 45. Warshaw, P. R., & Davis, F. D. (1985). Disentangling behavioral intention and behavioral expectation. Journal of Experimental Social Psychology, 21 (3), 213 228. Ward, S. (1974). Consumer socialization. Journal of Consumer Research, 1 1 14.

PAGE 84

84 BIOGRAPHICAL SKETCH William Parker graduated from C itrus High School (Inverness, FL) in 200 5 He began coursework at the University of Florida in 2005, graduating with a Bachelor of Science in f amily, y outh, and c ommunity s ciences in the summer of 2009. He began his graduate studies at the University of Fl orida in fall of 2009, pursuing a Master of Science degree in f amily, y outh, and c ommunity s ciences, with a concentration in family financial management.