Evaluating a financial education curriculum as an intervention to improve financial behaviors and financial well-being of survivors of domestic violence: results from a longitudinal randomized controlled study.
Financial education programs that increase consumer knowledge about financial management skills have garnered greater attention, especially in times of economic crisis and distress for individuals, families, organizations, and communities (Gudmunson and Danes 2011). Most of the existing financial education programs were launched in the late 1990s and early 2000s with a variety of institutions (e.g., community organizations, cooperative extension services, businesses, and community colleges) targeting specific populations (e.g., women, welfare recipients, low-income families, adolescents, college students, and persons of color) (Vitt et al. 2000). While most of these programs conducted consumer satisfaction evaluations, very few had the resources or inclination to conduct rigorous evaluations (Braunstein and Welch 2002; Collins 2012; Collins and O'Rourke 2010; GAO 2012; Lusardi and Mitchell 2007). Even studies designed to evaluate impact had challenges with measurements that appropriately captured the specific construct of financial knowledge and behaviors. Moreover, most lacked a conceptual framework to guide the variables measured in the studies, compromising the relevance of studies in terms of identifying specific mechanisms and best practices.
Women are one such specific population at risk for experiencing situations that leave them in worse financial shape (Fonseca et al. 2012; Tamborini, lams, and Reznik 2011). Such situations may include lower earnings in the workplace, divorce, or outliving their spouse (Fonseca et al. 2012). Domestic violence also creates situations in which women may feel economically dependent on the abuser and trapped in the relationship (Borden et al. 2007; Sanders and Schnabel 2006; Zorza 1991). Research and practice in the field have identified abusive strategies to include physical and sexual assault, psychological badgering, emotional blackmail, isolation tactics, and threats to harm the children in an effort to coercively control an intimate partner (Stark 2007). Recent attention has been given by researchers to exploring economic abuse strategies abusers may also utilize to control their partner. Such strategies may include economic exploitation and economically controlling behaviors as well as employment sabotage (Adams et al. 2008; Postmus, Plummer, McMahon, Murshid, and Kim 2012). For example, the abuser may discourage or prevent her from working, harass and disrupt her at work, purposively ruin her credit score, demand to know how money was spent, spend money that was designated for bills, or make important financial decisions without seeking input from his partner (Adams et al. 2008; Postmus et al. 2012 Raphael 1999; Tolman and Rosen 2001).
With survivors identified as a population at risk for financial challenges, several nonprofit domestic violence organizations developed financial education programs to improve individual financial management skills and ultimately, to improve survivors' financial well-being. These financial educational programs included key topics on basic financial management skills, such as saving, budgeting, getting or repairing credit, cash flow management, purchasing a home, predatory lending practices, financing major purchases, investing, and wise spending habits--all topics considered key dimensions of a successful financial management education program (Vitt et al. 2000). In addition, these programs included information on economic abuse and how to create economic safety plans that included how to disentangle joint financial relationships with an abusive partner or how to identify community-based resources to assist with financial safety challenges (Postmus 2010). Such attention given to this at-risk population seeking services at domestic violence organizations provides a key moment for improving their financial well-being as these survivors explore how to live independently from their abusers.
Similar to other financial education programs, limited rigorous evaluations of these programs for domestic violence survivors exist. One program, the Redevelopment Opportunities for Women's Economic Action Program (REAP), was evaluated in 2001 using randomly assigned domestic violence organizations (Sanders, Weaver, and Schnabel 2007). A total of 67 survivors completed a pre-test and a post-test two weeks following the implementation of the curriculum. Results indicated that survivors who participated in REAP showed greater and significant differences in their knowledge when compared to survivors who did not participate in REAP (Sanders, Weaver, and Schnabel 2007). Though using randomly assigned groups, this study had a small sample size and conducted only one post-test following the implementation of the curriculum.
The Allstate Foundation, in partnership with the National Network to End Domestic Violence (NNEDV), also created a financial education program for domestic violence survivors called "Moving Ahead Through Financial Management." This curriculum was developed based on comprehensive research of similar financial literacy programs for domestic violence survivors and other at-risk populations. In addition, an exploratory study of an earlier version of this program was piloted in several states; the curriculum was further modified prior to its implementation (Postmus and Plummer 2010).
The final version of this financial education curriculum contains five modules, including: (1) understanding financial abuse, (2) learning financial fundamentals, (3) mastering credit basics, (4) building financial foundations, and (5) creating budget strategies. There are three objectives of this program. The first overall objective is for participants to learn the basic vocabulary and knowledge with regard to money, credit, and financial management. The second objective is for participants to understand basic financial processes including loan applications, filing for bankruptcy, running a credit score, and filling out the accompanying financial paperwork. The last objective of the curriculum is specifically geared toward the needs of domestic violence survivors and includes information survivors would need to leave an abusive relationship. The topics incorporated material on disentangling joint financial relationships and repairing credit damaged by an abuser, locating safety and financial resources, and forming economic safety plans and strategies to protect them against their abuser. (1) The five modules in the curriculum provide the information to the staff to cover in these sessions; however, the material is not proscribed as to how long each module takes or how to present the materials. Such openness allows agencies the flexibility in how to deliver the materials based on their clients and schedule. During the "Training of Trainers" provided by NNEDV, staffs are encouraged to provide group and individual sessions when covering the material.
The study presented in this article evaluates this curriculum through the use of a longitudinal, randomized study that is guided by the reasoned action approach (RAA), a manifestation of the Theory of Planned Behavior (Fishbein and Ajzen 2010). Created through social psychology, RAA is a theory that guides prediction and explanation of individual behavior. It has been tested in over 1,000 empirical studies on topics ranging from alcohol behavior and job performance to condom use and interpersonal violence (Fishbein and Ajzen 2010). RAA suggests that there are various factors that influence one's change in behavior. These factors include one's attitudes or knowledge as well as one's intentions to change or perform a specific behavior. Most believe that if you provide information to someone, e.g., around financial management skills, he/she would take that information and change poor financial behaviors. What is missing, according to RAA, is an understanding of the individual's intentions to perform the behavior. Other factors that may influence behavior include (1) norms or beliefs about performing a specific behavior from peers or family of origin, (2) having the ability or the skills necessary to perform the behavior, and (3) one's confidence or self-efficacy in performing the behavior (Fishbein and Ajzen 2010).
An evaluation of a financial education program that applies the RAA frame would need to examine if the financial education program changes participants' knowledge, behavior, and intention to perform the behavior. In addition, evaluations should include financial well-being variables such as a decrease in financial stress. Hence, the hypotheses for this study include the following: Domestic violence survivors who attend classes on the "Moving Ahead Through Financial Management" curriculum, when compared to survivors who have not attended classes, will show greater improvements over time in (1) their financial knowledge; (2) their intentions to perform financial behaviors; and (3) their financial behaviors. Survivors who received the curriculum will also show a decrease over time in their financial stress.
The financial education curriculum was evaluated using a longitudinal randomized control study, with data collected over four time periods spanning 14 months. Face-to-face interviews were conducted prior to the implementation of the financial education curriculum, and served as baseline data. The data were recorded using SNAP software, an online survey tool. After completing a baseline interview, participants were randomly assigned within each agency to a treatment group who received the curriculum or to a control group who did not. Post-test interviews were conducted after approximately two months, eight months, and 14 months. Study participants were recruited from 14 different advocacy organizations located in seven different states and Puerto Rico. Advocates in these agencies distributed flyers to female survivors in order to provide them with information about the study. Participant eligibility criteria included: (1) must be female over the age of 18, (2) must have experienced some form of domestic violence (i.e., physical, psychological, sexual, or economic) at least once in the previous year, (3) have not attended a financial education class within the past two years, (4) are committed to attend the curriculum group if selected, and (5) are committed to participate in the research project whether or not they are selected for the curriculum group.
Participating domestic violence agencies included both shelters and nonresidential programs. NNEDV trained the staff from the agencies on the curriculum, and then the researchers instructed them to provide between four to eight group classes and at least one individual session in a two-month period. These instructions were developed in collaboration with several staff who had several years of experience teaching this curriculum to survivors. To achieve fidelity in the implementation of the curriculum and to allow for flexibility for the programs, the researchers instructed the staff to complete a checklist to ensure that all topics were covered. Participants were given a twenty dollar incentive for completing the Time 1 interview (hereafter T1), twenty-five dollars at Time 2 (hereater T2), thirty dollars at Time 3 (T3), and forty dollars at Time 4 (T4).
At T1, 457 women participated in the interviews, 300 women at T2, 279 at T3, and 245 at T4. The number of participants in each study group (control and treatment) remained fairly even in all time periods. One hundred and sixteen women (25 percent), who only completed the first interview, were dropped from the project. Women were dropped for various reasons, including: the agency did not provide the curriculum to the treatment group; the agency provided the curriculum to the control group; the participant did not complete the curriculum; the researcher was not able to get in contact with the participant; and the participant did not want to continue in the project. Considering the remaining 75 percent of the sample (n = 341), 15 percent of the women completed two interviews, 28 percent completed three interviews, and close to half (n = 195, 57%) completed all four interviews. This article focuses on the group of 195 women who completed all four interviews (treatment group, n = 94; control group, n = 101). By limiting this analysis to women in this longitudinal sample, we are able to show a very detailed picture of the impact of the curriculum over time.
Participants from the longitudinal sample (n = 195) represent 13 of the 14 agencies from seven states (New York, New Jersey, Rhode Island, Connecticut, Texas, Wisconsin, and Iowa) and Puerto Rico. They ranged in age from 21 to 62 years with an average age of 38. Sixty percent of the participants identified themselves as Latina or Hispanic; over 20 percent identified as black, non-Hispanic. Over 14 percent identified themselves as White, non-Hispanic. Less than 5 percent identified themselves as "other" (i.e., Asian, Multi-ethnic). Of the 195 participants, over 45 percent answered "no" when asked if they were born in the United States; over 87 percent of those immigrants (n = 77) had lived in the United States for five years or more. Over half of the participants (51.3 percent) were not working; most participants (86 percent) were not in school. Over 82 percent of the participants were financially responsible for children, with most of the participants responsible for three children or less. In addition, most of the participants (85.1 percent) made less than $25,000 in the past year for their entire household. Finally, the sample had an average of almost 12 years of education with almost 45 percent reporting post-high school education. Please see Table 1 for a description of the sample. The data are divided by group membership (i.e., treatment or control).
To understand the difference between the participants who completed all four interviews (n = 195) and those that did not (n = 262), a number of statistical tests were run on the variables discussed earlier. The results indicated that individuals who completed all four interviews (n = 195) were older by about two years. There was also a statistically significant difference among ethnic groups. A higher percentage of women that identified as "Latinas" and a lower proportion of those identifying as "other" completed all four interviews compared to those that completed between one and three interviews. No other significant differences between the groups were found in any other variables including all of the financial variables.
The survey instrument consisted of numerous validated or revised scales as well as demographic questions. The scales used in this article include financial knowledge, financial intentions, financial behaviors, and financial stress. These scales were chosen to represent the variables outlined by the RAA.
The Financial Knowledge scale was designed to measure an individual's perceived financial knowledge and is based on a similar scale created and tested with a different population of domestic violence survivors learning the "Moving Ahead Through Financial Management" curriculum (Postmus, Hetling, and Hoge 2013). Four items were added to the original 13 items to reflect the full content of the financial education curriculum. Based on an exploratory factor analysis, two of the 17 items were dropped, leaving a 15-item scale (a = .896). The response categories for the questions used a 1-5 scale that ranged from 1 (Strongly Disagree) to 5 (Strongly Agree).
The Financial Intentions scale (and subsequent Financial Behaviors scale) was created by the researchers for this study as one variable comprising the RAA as outlined in the literature review. Items for this scale were derived from the literature and from the behaviors addressed in the curriculum (Postmus et al. 2013). Financial intentions were measured by asking the respondents 14 questions regarding their financial intentions or motivation to perform a particular financial behavior in the upcoming month. For example, participants were asked "How likely are you in the next month to pay your bills on time?" and "How likely are you in the next month to follow a weekly or monthly budget?" Response options ranged from 1 (Never) to 5 (Always). After performing an exploratory factor analysis, four items were removed, leaving 10 items (a = .801).
The Financial Behavior scale was also created by the researchers based on financial behaviors identified in the literature and those that were addressed in the curriculum (Postmus et al. 2013). Financial behaviors were measured by asking the respondents the same 14 questions asked in the Financial Intentions scale; however, this scale asked participants about their financial behaviors in the past month. For example, participants were asked "How often in the last month did you pay your bills on time?" and "How often in the last month did you follow a weekly or monthly budget?" Response options ranged from 1 (Never) to 5 (Always). After performing an exploratory factor analysis, four items were removed, leaving the final 10 items (a = .804).
The Financial Strain Survey (Aldana and Liljenquist 1998) is an 18-item scale that measures different areas of financial strain. Participants were asked to indicate how often the items applied to them over the past 12 months. Participants indicated such frequency using a 5-point scale with answers ranging from 1 (Never) to 5 (Always). In this sample of female domestic violence survivors, the survey demonstrated high internal reliability (a = .798). Item numbers 1, 2, 3, and 15 were recoded as they were negatively worded items.
Repeated Measures Analysis of Variance (RM-ANOVA) was conducted in the statistical software package SPSS 21. The purpose of the analyses was to determine whether there were significant differences between those participants who received the curriculum and those who did not on scale variables of interest. These analyses also determined whether those differences were a result of the passage of time or if they could be attributed to participation in the curriculum. In order to test the assumption of sphericity, Mauchly's test was inspected for each variable of interest. For cases in which the assumption of sphericity was violated, the Greenhouse-Geisser correction was used to reduce the likelihood of a Type I Error. Measures of effect sizes (Cohen's d and partial eta-squared) are interpreted using Cohen's (1988) guidelines. According to Cohen's guidelines, effect sizes can be considered small (d = 0.2 and eta-squared = 0.01), medium (d = 0.5 and eta-squared = 0.06), or large (d = 0.8 and eta-squared = 0.14). Cohen cautions, however, that these guidelines should not be interpreted as rigid rules as behavioral science is a very diverse field (Cohen, 1988).
An examination of the data in all four time periods indicated that there was very little missing data with less than 5 percent missing for all variables; for many variables, the proportion of missing data was zero. As there were few missing values (less than 5 percent for all variables), those values were considered missing at random; pair-wise deletion was used for handling the missing values.
The results from the RM-ANOVAs are included in Tables 2 and 3.
Table 2 provides the means and standard deviations for the outcomes of interest over time as well as Cohen's d for the difference at T4. Table 3 summarizes the results of the RM-ANOVA analysis. The following discussion integrates findings from Tables 2 and 3 to illustrate significance and change over time in the variables of interest.
For financial knowledge, participants in the treatment group started with a mean score of 2.67, lower than the control group who started with a mean score of 2.81. However, this difference was not statistically significant. Mauchly's test indicated that the assumption of sphericity had been violated, W = .67, [chi square] (5) = 77.278, p < .001 ; therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (e = .77). Both groups showed improvement over time, and the effect of time on financial knowledge was found to be significant, F (2.313, 446.313) = 163.518, pc.001, partial [[eta].sup.2] = .459. There was also a significant interaction between time and group F (2.313, 446.313) = 48.809, p < .001, partial [[eta].sup.2] = .202, with those in the treatment group scoring statistically significantly better over time on the measure of financial knowledge than those in the control group. At T4, the treatment group had an average score of 4.01 in comparison to 3.27 for the control group. The effect size (d = 1.05) for this difference in means echoes the partial eta-squared results and indicates a large effect of the curriculum on financial knowledge.
Participants started at a similar place for financial intentions, with mean scores of approximately 3.08 for the treatment group and 2.97 for the control group in T1. The difference in scores at T1 was not statistically significant. Mauchly's test indicated that the assumption of sphericity had been violated, W = .88, [chi square] (5) = 23.555, p < .001, therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (e = .92). The results showed that there was a significant effect of time on financial intentions for both groups F (2.754, 526.102) = 26.098, p < .001, partial [[eta].sup.2] = .120. The interaction between time and group was also found to be significant, F (2.754, 526.102) = 7.919, p<.001, partial [[eta].sup.2] =.040, with those in the treatment group scoring statistically significantly better over time on the measure of financial intentions than those in the control group, with the treatment group scoring 3.59 in comparison to 3.24 for the control group in T4. The effect size (d = .46) for this difference in means as well as the partial eta-squared results miss the conventional cut-off for moderate impact by a couple hundredths of a point, indicating a close to moderate effect of the curriculum on financial intentions.
An examination of financial behaviors showed that participants in both groups started at a similar place with mean scores of 2.84 for the treatment group and 2.76 for the control group, with no statistically significant difference between those scores. Mauchly's test indicated that the assumption of sphericity had been violated, W = .80, [chi square] = 43.818, p < .001, therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity ([epsilon] = .86). A significant effect of time on financial behaviors was found, F (2.586, 496.486) = 36.952, p < .001, partial [[eta].sup.2] = .161. There was also a significant interaction effect of time and group on financial behaviors, F (2.586, 496.486) = 8.094, p < .001. partial [[eta].sup.2] = 040, with those in the treatment group scoring statistically significantly better over time on the measure of financial behaviors than those in the control group. At the end of the follow-up period, in T4, the treatment group had an average score of 3.53 in comparison to 3.12 for the control group. The effect size (d = .52) for this difference in means coupled with the partial eta-squared results shows a marginally moderate effect of the curriculum on financial behaviors. Although the scores of both groups improved overtime, the treatment group experienced a large increase between T1 and T2, while the control group's change was more constant and less notable, indicating a stronger effect of the curriculum in T2.
For financial strain, participants started at a similar place with mean scores of approximately 2.85 for each group. Mauchly's test indicated that the assumption of sphericity had been violated, W = .82, [chi square] = 37.094. p < .001 ; therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity ([epsilon] = .88). The scores of both groups decreased over time, and the results showed that the effect of time on financial strain was statistically significant, F (2.645, 510.531) = 112.250, p < .001, partial rc2 = .368. The results also showed that there was a significant interaction effect between time and group on financial strain F (2.645, 510.531) = 13.445, p<.001, partial [[eta].sup.2] = .065, with those in the treatment group scoring statistically significantly better over time on the measure of financial strain than those in the control group. At the end of the follow-up period, the treatment group scored 1.98 in comparison to 2.26 for the control group in T4. The effect size (d = .44) for this difference in means shows a close to moderate effect in T4 while the partial eta-squared results reveal a clearly moderate effect of the curriculum on financial strain over the full follow-up period.
As an experimental study, the results strongly supported the hypotheses and indicated that the differences observed post-curriculum are likely caused by participation in the curriculum and not by other factors. The two study groups were randomly assigned and baseline characteristics were similar on the key variables. The analyses demonstrated that while both groups experienced improvements over time, the treatment group had significantly better outcomes on all measures immediately after participation in the curriculum. Moreover and perhaps more importantly, the impact of the curriculum persisted over the full 12-month, post-curriculum, follow-up period. Although the control group also showed improvement over time, the smaller and slower change is likely related to other possible services received from the domestic violence organization.
An examination of our effect size findings indicates that the impact of the curriculum is not only statistically significant, but also practically important with a very large effect on financial knowledge, a moderate effect on financial behaviors and financial strain, and a close to moderate effect on financial intentions when comparing T1 to T4. Specifically, the treatment group had an average improvement of between a half point and to over a full point on self-reported financial knowledge, financial intentions, and financial behavior, measured on a five-point scale. In comparison, control group members improved by only half as much on each measure. Similarly, financial strain among the treatment group decreased by 0.88 points compared to a decrease of only 0.32 for the control group. In practical terms, these changes indicate that survivors who participated in the curriculum had a stronger grasp of financial management than those who did not, including understanding how to manage their money to actually changing behaviors that demonstrate better financial practices. Such a grasp may leave survivors more confident in their ability to, at least, be more financially independent and, at the most, to be independent from an abusive relationship. The immediate impact of the curriculum in T2 is particularly important in the context of abuse. While the control group also experienced gradual improvements on all measures over time, the abrupt change from T1 to T2 measures for the treatment group supports the unique impact that the curriculum may have on domestic violence survivors.
The ability to generalize these findings to other populations is limited by several issues. First, the agencies in the study chose to participate. It is possible that the success of the curriculum rests with the support from the agencies and the advocates charged with providing the curriculum. Second, all of the women in the study volunteered to participate. Although the random assignment of participants allows us to rule out motivation as a determinant of the observed outcomes, we cannot assume that the curriculum would have an equal impact on women who did not want to participate or who were mandated to take the curriculum. Just as notably, the women in the sample were receiving services from a domestic violence agency; hence, the results cannot be generalized to all women who experience domestic violence. In addition, the demographic characteristics of the sample are not reflective of domestic violence survivors across the country; instead, the sample has a higher number of Latinas because of the clientele of the programs from which survivors were recruited. Further research is needed to understand the potential differences which may exist with survivors from different cultures and different immigrant statuses. Finally, some of the measures chosen were not previously validated scales; additional validation is needed to ensure the scales appropriately capture the construct being measured.
Despite the mentioned limitations to sample generalizability in particular, the observed results are both statistically significant and practically large, particularly regarding the impact of the curriculum on financial knowledge. The findings also add to the growing body of literature on the subject. First, the design of the study responds to calls from other researchers (e.g., Collins and O'Rourke 2010; GAO 2012) for more rigorous evaluation of financial literacy programs. Second, our findings echo those of the REAP evaluation (Sanders, Weaver, and Schnabel 2007) even though we used a larger sample and conducted multiple post-tests. Such similarities further support the efficacy of implementing financial literacy programs for domestic violence survivors.
In regard to future research, the use of the RAA for framing the study shows great promise as a theoretical model for other evaluations of financial education programs. As applying the RAA to changing financial behaviors is a new concept, we chose to focus on the changes in variables over time instead of testing the relationship between variables. Further testing of the relationship between the variables and the paths through which they are related are still needed to fully test the RAA model. Regardless, RAA holds promise as a theory to understand how to improve or change individuals' financial behavior by focusing on improving their financial education as well as improving their intentions and decreasing any financial strain they may be experiencing.
PRACTICE AND POLICY IMPLICATIONS
The notable and lasting impact of participation in the curriculum for this study sample has critical implications for other agencies serving domestic violence survivors as well as other programs aimed at improving financial well-being among their clientele. Many survivors have limited access to resources and very little base knowledge and financial experience, and may be making strides to become financially independent for the first time. The Allstate Foundation's curriculum begins with an explanation of economic abuse and safety planning as a way to address concerns specific to domestic violence survivors who may be in crisis and focused on their immediate financial well-being and safety. At the same time, the curriculum is very comprehensive and includes topics ranging from budgeting to investing, hence moving beyond a focus on crisis management in the context of domestic violence to future planning and economic advancement. Other domestic violence survivors may benefit from this curriculum. As economic empowerment becomes a more common goal for agencies serving this population, practitioners should consider financial education, specifically this curriculum, as part of their core services to help survivors learn the fundamentals of financial management, experience less financial stress, and take necessary steps such as changing financial intentions and behaviors that might lead them to becoming financially independent.
Unlike other robust evaluations of financial literacy programs that focus on more financially stable adult populations (Collins and O'Rourke 2010; Hastings, Madrian, and Skimmyhorn 2012; Lusardi and Mitchell 2007), this study indicates that teaching financial education with domestic violence survivors is not only theoretically important but also empirically effective. Though the women in this study come from lower socioeconomic statuses and, as survivors leaving abusive relationships, would be less financially stable or have limited access to resources, the findings from this study suggest that successful financial education and intervention are possible and should even be considered required for this population. Our findings indicate that financial education at key moments, such as at the very beginning of establishing an independent life from abuse, can lead to improved financial knowledge, intentions, and behaviors, and decreased financial strain.
To conclude, our evaluation of the Allstate curriculum supports recent developments in the broader financial education movement to reach different populations (e.g., adolescents, adults, and older adults) and adds domestic violence survivors as another population group needing financial education to improve their financial behaviors. This particular curriculum takes a comprehensive approach to content from basic, short-term tools to complex, long-term strategies. While admittedly a far step from our current evaluation of a curriculum designed specifically for survivors of domestic violence, future policy discussions on developing financial capacity should consider these findings in possible relationship to other programs, such as those for high school students or first year college students. Such programs also are designed to provide a broad introduction to financial education and are aimed at developing financial capability early in life. The basic framework of the "Moving Ahead Through Financial Management" curriculum, particularly its comprehensive approach to financial education, may serve as a strong example of a program that aims to do something similar. By including education on economic abuse, such programs may even prepare these young adults to be watchful of potentially abusive partners.
Adams, Adrienne E., Cris M. Sullivan, Deborah Bybee, and Megan R. Greeson. 2008. Development of the Scale of Economic Abuse. Violence Against Women, 14 (5): 563-588. doi: 10.1177/1077801208315529.
Aldana, Steven G. and Wendy Liljenquist. 1998. Validity and Reliability of a Financial Strain Survey. Journal of Financial Counseling and Planning, 9 (2): 11-19.
Borden, Lynne M., Sun-A Lee, Joyce Serido, and Dawn Collins. 2007. Changing College Students' Financial Knowledge, Attitudes, and Behavior through Seminar Participation. Journal of Family and Economic Issues, 29 (1): 17.
Braunstein, Sandra and Carolyn Welch. 2002. Financial Literacy: An Overview of Practice, Research, and Policy. Federal Reserve Bulletin, 88 (11): 445-457.
Cohen, Jacob. 1988. Statistical Power Analysis for the Behavioral Sciences. 2nd edition. Hillsdale, NJ: Lawrence Erlbaum Associates.
Collins, J. Michael. 2012. The Impacts of Mandatory Financial Education: Evidence from a Randomized Field Study. Journal of Economic Behavior & Organization, 95: 146-158.
Collins, J. Michael and Collin O'Rourke. 2010. Financial Education and Counseling: Still Holding Promise. Journal of Consumer Affairs, 44: 483-498.
Fishbein, Martin and leek Ajzen. 2010. Predicting and Changing Behavior: The Reasoned Action Approach. New York: Psychology Press, Taylor & Francis Group.
Fonseca, Raquel, Kathleen J. Mullen, Gema Zamarro, and Julie Zissimopoulos. 2012. What Explains the Gender Gap in Financial Literacy? The Role of Household Decision Making. Journal of Consumer Affairs, 46 (1): 90-106.
GAO. 2012. Financial Literacy: Enhancing the Effectiveness of the Federal Government's Role. Washington, DC: U.S. Government Accountability Office.
Gudmunson, Clinton G. and Sharon M. Danes. 2011. Family Financial Socialization: Theory and Critical Review. Journal of Family and Economic Issues, 32: 644-667.
Hastings, Justine S., Brigitte C. Madrian, and William L. Skimmyhom. 2012. Financial Literacy, Financial Education and Economic Outcomes. The National Bureau of Economic Research [cited Working Paper No. 18412. Available from http://www.nber.org/papers/w18412.
Lusardi, Annamaria and Olivia S. Mitchell. 2007. Financial Literacy and Retirement Preparedness: Evidence and Implications for Financial Education. Business Economics, 42 (1): 35-44.
Postmus, Judy L. 2010. Economic Empowerment of Domestic Violence Survivors. Harrisburg, PA: VAWnet, a project of the National Resource Center on Domestic Violence, Pennsylvania Coalition Against Domestic Violence.
Postmus, Judy L. and Sara-Beth Plummer. 2010. Validating The Allstate Foundation's National Model on Helping Survivors of Violence Achieve Economic Self-Sufficiency: Final Report. New Brunswick, N.J.: Rutgers University, School of Social Work, Center on Violence Against Women & Children.
Postmus, Judy L., Sara-Beth Plummer, Sarah McMahon, Nadine Murshid, and Misung Kim. 2012. Understanding Economic Abuse in the Lives of Survivors. Journal of Interpersonal Violence, 27: 411-430.
Postmus, Judy L., Andrea Hetling, and Gretchen L Hoge. 2013. Evaluating the Impact of the "Moving Ahead Through Financial Management" Curriculum: A Randomized Control Study. New Brunswick, NJ: Center on Violence Against Women & Children, Rutgers University, School of Social Work.
Raphael, J. 1999. Keeping Women Poor: How Domestic Violence Prevents Women from Leaving Welfare and Entering the World of Work. In Battered Women, Children and Welfare Reform: The Ties That Bind, edited by Ruth A. Brandwein (31-43). Thousand Oaks, CA: Sage.
Sanders, Cynthia K. and Meg Schnabel. 2006. Organizing for Economic Empowerment of Battered Women: Women's Savings Accounts. The Journal of Community Practice, 14 (3): 47-68.
Sanders, Cynthia K., Terry L. Weaver, and Meg Schnabel. 2007. Economic Education for Battered Women: An Evaluation of Outcomes. Affilia, 22 (3): 240-254.
Stark, Evan. 2007. Coercive Control: How Men Entrap Women in Personal Life. New York: Oxford University Press.
Tamborini, Christopher R., Howard M. lams, and Gayle L. Reznik. 2011. Women's Earnings before and after Marital Dissolution: Evidence from Longitudinal Earnings Records Matched to Survey Data. Journal of Family and Economic Issues, 33: 69-82. Online First, doi: 10.1007/s 10834-011-9264-1.
Tolman, Richard M. and Daniel Rosen. 2001. Domestic Violence in the Lives of Women Receiving Welfare: Mental health, Substance Dependency and Economic Well-Being. Violence Against Women, 1 (2): 141-158.
Vitt, Lois A., Carol Anderson, Jamie Kent, Deanna M. Lyter, Jurg K. Siegenthaler, and Jeremy Ward. 2000. Personal Finance and the Rush to Competence: Financial Literacy Education in the U.S. Middleburg, VA: Institute for Socio-Financial Studies.
Zorza, Joan. 1991. Woman Battering: A Major Cause of Homelessness. Clearinghouse Review, 25: 421-429.
Judy L. Postmus (email@example.com) is an Associate Professor & Director, Center on Violence Against Women & Children, Rutgers University. Andrea Hetling (firstname.lastname@example.org) is an Associate Professor, Rutgers University. Gretchen L. Hoge (email@example.com) is a Doctoral Candidate, Center on Violence Against Women & Children, Rutgers University. This project was supported by The Allstate Foundation, Economics Against Abuse Program. Points of view in this document are those of the authors and do not necessarily represent the official position or policies of The Allstate Foundation.
(1.) More information about the financial literacy program, including detailed descriptions of each curriculum module can be found here: http://www.clicktoempower.org/financial-tools/curriculumdownload.aspx.
TABLE 1 Demographic Characteristics of the Sample Full sample Variables (n=195) Age [mean (SD)] 37.5(8.85) Ethnicity Hispanic/Latina 60 Black or African American 21 White, Non-Hispanic 14.4 Other 4.6 Immigrant to the United States 45.1 If immigrant, time lived in the United States Less than 1 year 2.2 1 year to less than 5 years 10.2 5 years to less than 10 years 18.1 10 years or more 69.3 Employment Unemployed 51.3 Part time 25.1 Full time 23.6 Currently in school 13.9 Years of education [mean (SD)] 11.72(3.76) 1-8 years 15.7 9-12 years 39.7 13-16 years 37.9 17-20 years 6.7 Annual income $0-$ 10,000 44.1 $10,001-$15,000 27.7 $15,001-$25,000 13.3 $25,001-$35,000 7.2 More than $35,000 7.2 Marital status Married 15.9 Civil union/partnership 3.1 Separated 30.8 Divorced 19.5 Widowed 0.5 Single (includes dating) 30.3 Financially responsible for children 82.1 One child 24.1 Two children 27.2 Three children 19.5 More than three children (4-15) 11.2 Percentage or Mean Treatment Variables (n = 94) Age [mean (SD)] 37.4(9.18) Ethnicity Hispanic/Latina 62.8 Black or African American 18.1 White, Non-Hispanic 13.8 Other 5.3 Immigrant to the United States 50.0 If immigrant, time lived in the United States Less than 1 year 2.1 1 year to less than 5 years 8.5 5 years to less than 10 years 19.1 10 years or more 70.2 Employment Unemployed 45.2 Part time 29.0 Full time 25.8 Currently in school 18.1 Years of education [mean (SD)] 11.4(3.94) 1-8 years 18.1 9-12 years 39.3 13-16 years 34.0 17-20 years 8.6 Annual income $0-$ 10,000 42.6 $10,001-$15,000 24.5 $15,001-$25,000 14.9 $25,001-$35,000 9.6 More than $35,000 8.5 Marital status Married 12.8 Civil union/partnership 4.3 Separated 36.2 Divorced 19.1 Widowed 0.0 Single (includes dating) 27.7 Financially responsible for children 79.8 One child 26.7 Two children 32.0 Three children 24.0 More than three children (4-15) 17.3 Control Variables (n = 101) Age [mean (SD)] 37.7(8.57) Ethnicity Hispanic/Latina 57.4 Black or African American 23.8 White, Non-Hispanic 14.9 Other 4.0 Immigrant to the United States 40.6 If immigrant, time lived in the United States Less than 1 year 2.4 1 year to less than 5 years 12.2 5 years to less than 10 years 17.1 10 years or more 68.3 Employment Unemployed 56.4 Part time 21.8 Full time 21.8 Currently in school 10.0 Years of education [mean (SD)] 12.0(3.56) 1-8 years 13.2 9-12 years 39.7 13-16 years 41.8 17-20 years 5.1 Annual income $0-$ 10,000 46.0 $10,001-$15,000 31.0 $15,001-$25,000 12.0 $25,001-$35,000 5.0 More than $35,000 6.0 Marital status Married 18.8 Civil union/partnership 2.0 Separated 25.7 Divorced 19.8 Widowed 1.0 Single (includes dating) 32.7 Financially responsible for children 84.2 One child 31.8 Two children 34.1 Three children 23.5 More than three children (4-15) 10.6 TABLE 2 Univariate Statistics for Variables in Analysis Means (Standard Deviations) T1 T2 T3 Financial knowledge Treatment group 2.67 4.04 3.95 (n = 94) (.82) (.62) (.66) Control group 2.81 3.09 3.31 (n = 101) (.79) (.77) (.76) Financial intentions Treatment group 3.08 3.76 3.62 (n = 93) * (.81) (.68) (.87) Control Group 2.97 3.11 3.26 (n = 100) * (.78) (.85) (.71) Financial behaviors Treatment group 2.84 3.45 3.48 (n = 93) * (.92) (.76) (.84) Control group 2.76 2.82 3.12 (n = 101) (.87) (.88) (.79) Financial strain Treatment group 2.86 2.12 2.03 (n = 94) (.59) (.58) (.61) Control group 2.84 2.62 2.40 (n = 101) (.65) (.68) (.65) T4 Cohen's d at T4 Financial knowledge Treatment group 4.01 1.05 (n = 94) (.57) Control group 3.27 (n = 101) (.82) Financial intentions Treatment group 3.59 .46 (n = 93) * (.75) Control Group 3.24 (n = 100) * (.76) Financial behaviors Treatment group 3.53 .52 (n = 93) * (.78) Control group 3.12 (n = 101) (.80) Financial strain Treatment group 1.98 .44 (n = 94) (.65) Control group 2.26 (n = 101) (.61) * Please note that for this analysis, pair-wise deletion was used for missing data. TABLE 3 Repeated Measures Analysis of Variance Effect MS df Error df F P Financial knowledge Time 47.982 2.313 446.313 163.518 *** Time x group 14.322 2.313 446.313 48.809 *** Financial intentions Time 8.552 2.754 526.102 26.098 *** Time x group 2.595 2.754 526.102 7.919 *** Financial behaviors Time 13.236 2.586 496.486 36.952 *** Time x group 2.899 2.586 496.486 8.094 *** Financial strain Time 22.907 2.645 510.531 112.250 *** Time x group 2.744 2.645 510.531 13.445 *** Partial Eta-Squared Greenhouse- Effect Results Geisser Financial knowledge Time 0.459 *** Time x group 0.202 *** Financial intentions Time 0.120 *** Time x group 0.040 *** Financial behaviors Time 0.161 *** Time x group 0.040 *** Financial strain Time 0.368 *** Time x group 0.065 *** * p<. 05: ** p< .01; *** p < .001.
|Printer friendly Cite/link Email Feedback|
|Author:||Postmus, Judy L.; Hetling, Andrea; Hoge, Gretchen L.|
|Publication:||Journal of Consumer Affairs|
|Date:||Mar 22, 2015|
|Previous Article:||Does "banking the unbanked" help families to save? Evidence from the United Kingdom.|
|Next Article:||A balance sheet perspective on financial success: why starting early matters.|