FINSER_31-2_Complete_Issue


Financial literacy during a pandemic:
Does modality matter?

Greg Filbecka,*, Xin Zhaob

aPenn State Behrend, Black School of Business, 281 Burke, Erie, PA 16563, USA
bPenn State Behrend, Black School of Business, 256 Burke, Erie, PA 16563, USA

Abstract

We compare the performance of financial literacy programs launched by the CFA Society
Pittsburgh in residential settings (2017–2019) with virtual/hybrid programs during the COVID-19
pandemic (2020-2021). Pretest baseline knowledge assessment shows that female students scored
lower on subjective and objective financial knowledge questions and self-esteem. However, the
global pandemic did not impact the effectiveness of programs based on modality. Students experi-
enced a statistically significant improvement in all four assessment areas of financial literacy. The
largest gains in subjective and financial knowledge center on retirement planning. Objective knowl-
edge and self-esteem improvements occur most in credit and inflation. Female students experience
more significant gains in subjective knowledge, objective knowledge, and self-esteem. © 2023
Academy of Financial Services. All rights reserved.

JEL classifications: G53

Keywords: Financial literacy; Survey; Outreach; Education; Teaching modality

1. Introduction

The state of financial literacy and numeracy worldwide is in distress. Academic studies
affirm a significant lack of financial literacy exists across nearly all demographics. While fi-
nancial literacy statistics are important, the implications of financial literacy and numeracy
are far-reaching due to their impact on financial decisions. Financial literacy measures the
degree to which one understands critical financial concepts and possesses the ability and

*Corresponding author. Tel.: +1-814-898-6549.
E-mail address: mgf11@psu.edu (G. Filbeck)

1057-0810/23/$ – see front matter © 2023 Academy of Financial Services. All rights reserved.

Financial Services Review 31 (2023) 169–196



confidence to manage personal finances through appropriate, short-term decision-making
and sound, long-range financial planning, while mindful of life events, and changing eco-
nomic conditions.

The COVID-19 pandemic dominated headlines throughout 2020 and into 2021. High

school teachers had to rapidly transform in-class content delivery into modalities that

included synchronous and asynchronous online instruction. At-risk students were most likely

to have limited access to the internet and technology to participate fully. Many high schools

with financial literacy components in their curriculum had to postpone such instruction to

ensure coverage of the core subject matter. This study extends prior research on the effec-

tiveness of financial literacy education by comparing the in-person effectiveness of a finan-

cial literacy campaign launched by the CFA Society Pittsburgh during 2017–2019 with

online/hybrid instruction during 2020-2021. Online/hybrid programs are defined as those

that contain at least some virtual learning component to those programs that are delivered

completely in an in-person environment.
We conduct pretests to assess baseline knowledge for program participants before the pro-

gram administration. Overall, pretest results from financial literacy programs before and dur-

ing the pandemic show no overall differences regarding baseline financial literacy subjective

knowledge. However, we observe gender differences across pre-surveys as female students

scored lower than male students on subjective and objective financial knowledge. Females

also exhibited lower self-esteem. Following the program administration, we conduct post-

tests to assess gains in financial literacy due to the program. Both male and female students

experienced a statistically significant improvement in subjective and objective knowledge,

behavior, and self-esteem. The most significant gains in subjective and financial knowledge

center on retirement planning. Objective knowledge and self-esteem improvements occur

most in credit and inflation. Female students experience greater gains in subjective and

objective knowledge, and self-esteem than their male peers.
Students in both modalities experienced a statistically significant improvement in subjec-

tive and objective knowledge, behavior, and self-esteem. There is no statistical difference in

improvements between modalities except in financial behavior. The following section out-

lines existing literature and our hypotheses, followed by sample and methodology, results,

and conclusions.

2. Literature review

2.1. Financial literacy

Financial literacy is critical because the knowledge and skills enable the proper use, accu-
mulation, increase, and management of incomes while directly affecting countries’ econo-

mies. In context, financial literacy refers to the knowledge of financial concepts and

applications that are vitally important in everyday life (Semercioglu & Akcay, 2016).

Financial literacy programs primarily aim to understand individuals’ financial knowledge

170 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



better. Semercioglu and Akcay conclude that low-level financial literacy across numerous

countries is associated with a lack of financial training for individuals in their traditional

education experience.
The United States offers an example of a financial literacy system needing improvement.

Only 49% of Americans with a college education can answer basic questions regarding fi-
nancial literacy (Faulkner, 2017). Faulkner believes that the observed lack of financial liter-
acy is positively correlated because household spending in the United States has consistently
ranked among the highest globally. Essentially, lower financial literacy contributes to the
higher spending habits of individuals. Additionally, the average savings rate in the United
States has hovered around 5%, while the recommended level according to the U.S. Bureau
of Economic Analysis is often double this value (Faulkner, 2017).

Henager and Cude (2016) note increased academic research focusing on financial literacy
and renewed interest in financial education and related policy since the mid-2000s driven by
an increase in school-based financial education programs. In addition, an increase in state
mandates for financial education in high schools and the creation of entities (e.g., the
Financial Literacy and Education Commission and the Consumer Financial Protection
Bureau) addressing financial literacy illustrate an increase in attention to improving financial
literacy across the country. Henager and Cude consider how financial knowledge correlates
with short-term and long-term behavior segmented by age group. They discovered a positive
correlation between financial literacy and short-term and long-term financial behavior. An
example of short-term financial behavior is paying your monthly bills on time, while long-
term behavior could be saving for retirement. De Beckker, De Witte, and Van Campenhout
(2021) note that financial education does not result in better customer choices, despite dem-
onstrating improvements in financial literacy.

One purpose of implementing financial literacy programs across high schools is to increase

students’ knowledge of basic financial concepts and make smarter financial decisions as an

adult. Brown, Grigsby, Van Der Klaauw, Wen, and Zafar (2016) consider the effects of expo-

sure to financial training on early adults’ debt outcomes based on their high school’s inclusion

of financial literacy training. Brown et al. analyze large-scale changes in financial training expo-

sure using a sample of young Americans and their debt behaviors over the decade immediately

following high school training to see if financial trading has had any positive impact on their

behavior. They find that financial and quantitative education during high school has moderate

implications for young adults’ financial decisions aged 19 to 29.
Digital financial literacy (DFL), the education of financial literacy through digital plat-

forms, is likely to become an increasingly important aspect of education, especially in a
post-COVID-19 world. This transition means individuals can become more responsible for
their financial planning, including retirement and other goals (Morgan, Huang, & Trinh,
2019) via a digital learning environment. Since the early 2000s, many technological advan-
ces have emerged across the globe that have impacted several countries’ structures, yet vari-
ous educational systems have not adequately adapted to a digital learning environment.
Financial technology (fintech), using software, applications, and digital platforms to deliver
financial services to consumers and businesses through digital devices such as smartphones,
has become recognized as a promising tool to promote financial inclusion. Inclusion

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 171



involves necessary access to financial products and services provided to excluded house-
holds and small firms.

FinTech is revolutionizing the financial services industry at a very rapid pace. Views dif-
fer regarding the likely impact that FinTech is expected to have on personal financial plan-
ning, well-being, and societal welfare (Panos & Wilson, 2020). Many scholars have one
question whether the implementation of FinTech in the high school curriculum will have a
material effect on students’ improvement of financial knowledge. They argue that financial
literacy research should make financial education more effective through improved content
design and delivery. The disadvantage of enabling a more user-friendly and easily accessible
program online is that students might develop misconceptions of specific topics more often
than if they were engaged in an in-person program. For instance, Panos & Wilson indicate
that students had a persistent misunderstanding of the risk/reward tradeoff with derivatives,
more so than any other topic due to the program’s online nature. If enough student partici-
pants form this misconception, then a possibility exists that other misconceptions related to
basic financial knowledge could emerge. However, van Alten, Phielix, Janssen, and Kester
(2019) find that in flipped classroom settings in which students are required to watch online
videos, students achieved significantly higher assessed learning outcomes than students in
traditional classrooms, assuming face-to-face class time was not reduced compared to non-
flipped classrooms. Their study did not have any direct evidence of differences. Noetel,
Griffith, Delaney, Sanders, Parker, del Pozo Cruz, and Lonsdale (2020) find that substituting
video for existing teaching methods leads to smaller improvements in student learning. In
contrast, videos as a supplementary teaching tool led to stronger learning benefits.

Wolla (2017) reports that FDIC research indicates that 80% of the states in the United
States have currently adopted some form of personal finance education standard, up from
42% in 1998. While the numbers show an improvement, these states do not mandate finan-
cial education. With only 29.7% of schools offering financial education, the student’s need
for financial education is not adequately satisfied. Wolla, a member of the Federal Reserve
Bank of St. Louis, notes that one of its resources, Soar to Savings, provides an online learn-
ing module that teaches essential personal finance and economics concepts. With a sample
of 3,061 sets of pretest and post-test scores of students across 100 schools, Wolla concludes
that a statistically significant improvement exists in scores from both the overall student
results and the school results. Overall, the results confirm that the paired samples at both lev-
els indicate that the Soar to Savings online module effectively increases financial knowledge
among high school students.

Although traditional printed materials and in-person classroom-style workshops are most
prevalent, technological advances have created online financial education opportunities in
recent years (Kim, Russell, & Schroeder, 2017). Government studies, however, show that
Wolla’s (2017) findings are not broadly applied to all schools since most of the schools do
not include online modules. While programs such as the Soar to Savings online module are
gaining momentum in high schools, authors often omit relevant research and theoretical
frameworks to develop such programs. Kim et al. show an imbalance of the benefits pro-
vided by online financial literacy programs.

While Wolla (2017) and Panos and Wilson (2020) show evidence in favor of students’
increase in financial knowledge after financial literacy programs, other studies question the

172 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



methodological rigor and whether the “improvement” of student knowledge diminishes over
time. Kim et al. (2017) argue that Wolla’s findings that programs do not adequately measure
the students’ knowledge retention after the post-test was completed. A key implication
described by Kim et al. is that many financial education programs, such as Soar to Savings,
lack an explicit theory to frame information delivery. This lack of framework explains why
students lose retention of learned financial concepts over time.

Technology is integrated into everyday life, and offering online financial education may
offer alternative and innovative ways to reach broader audiences. However, this process
requires more than just publishing previously printed materials online. Financial educators
and practitioners should design effective and interactive online programs and tools that build
knowledge, facilitate improved financial decision-making, and foster positive behavior
change.

Kim et al. (2017) provide several recommendations to incorporate financial concepts into
the school curriculum. First, teachers can use online tools and resources to extend in-person
educational encounters by allowing for more flexibility and frequency between student and
teacher. These digital resources can provide a more accessible way for students to get
involved in the program. Educators can also create websites that offer tools, lectures, webi-
nars, videos, downloadable documents, activities, worksheets, and other resources for “on-
demand” learning, which would be especially useful for schools that offer financial educa-
tion programs as an extracurricular activity. Second, standalone online educational programs
should be tailored to specific financial behaviors and audiences. Targeted programs (e.g.,
programs derived from surveys of student populations or social media campaigns for specific
user groups) may be more effective than general financial education. Third, multiple modal-
ities such as blogs, online games, chatrooms, and smartphone apps could be employed by
teachers to accommodate a range of various learning styles, which should increase students’
overall engagement with the program. Fourth, educators should structure programs with
reminders, alerts, and prompts to help students monitor their progress and keep them
engaged throughout the program’s entire duration.

Cameron, Calderwood, Cox, Lim, and Yamaoka (2014) suggest that younger generations
are poorly prepared for making potentially life-changing financial decisions. The previous
research provides multiple opportunities to enhance financial education programs to increase
individuals’ overall financial literacy knowledge. However, the most notable two findings
are related to the modality (online/hybrid vs. in-person) and the program’s length. First,
assuming the research discussed remains true, individuals enrolled in more extended finan-
cial education programs tend to retain more concepts as time passes. Continual education
improves students’ cognitive understanding and increases their financial attitude and appli-
cation of financial concepts with everyday life. Secondly, newer research conducted by
Panos and Wilson (2020), Wolla (2017), and Kim et al. (2017) shows that online modules
are effective in increasing students’ short-term financial literacy. Although each question
whether the modules have a lasting impact on their financial cognitive ability similar to
what in-person education programs show. Therefore, creating more extended financial edu-
cation programs with the flexibility of in-person or online modules can provide the most effi-
cient form of increasing students’ overall financial attitude, behavior, and literacy.

In this study, we investigate two issues: First, we compare the effectiveness of the CFA
Society Pittsburgh’s financial literacy program under a model of in-person instruction

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 173



(2017–2019) and virtual/hybrid instruction (2020-2021). Second, we explore whether gender
differences exist in in-person versus virtual/hybrid instruction effectiveness. We create pre-
and post-surveys to assess four key factors of financial success: self-esteem, perceived (sub-
jective) knowledge, behavior, and objective numeracy. The surveys reflect the work of
Filbeck, Zhao, and Pettner (2020), which were motivated by previous works by Lusardi and
Mitchell (2008, 2011) and Lusardi, Mitchell, and Curto (2010).

The in-person instruction data are collected during 2017-2019, while the virtual/hybrid
instruction data are during 2020-2021. This paper extends work done by Filbeck et al.
(2020) by including a teacher survey to assess the modality of the courses taught in 2020-
2021. Our hypotheses are as follows:

H1: The virtual/hybrid implementation of financial literacy education will significantly differ in
students’ performance compared to in-person learning.

H2A: Male students have better financial literacy in all four key areas before the survey in both
modalities.

H2B: Female students will show a greater improvement in knowledge, at the statistically significant
level, across both modalities than male students.

Based on the previous research, the first hypothesis statement should hold for the sample
students in this survey. First, if the hypothesis holds true, in-person financial education pro-
grams will prove more efficient than a virtual/hybrid instruction mode. The hypothesis aligns
with Amagir et al.’s (2018) research, which finds that person-to-person financial education is
positively associated with higher financial knowledge scores from their survey. Additionally,
Panos and Wilson (2020) find in their research that virtual/hybrid modes of financial education
courses failed to teach students in certain financial topics, stating a lack of direct student-to-
teacher contact as the vital reason. These financial topics test students in their objective and
subjective financial knowledge.

Our hypothesis H2A builds on prior research that has found a financial literacy gender
gap. In particular, Cupák, Fessler, Schneebaum, and Silgoner (2018) find that women score
lower than men on financial literacy, with a more pronounced gap in developed countries.
Additionally, Preston and Wright (2019) examine the financial literacy gap in Australia.
While the “human capital variables” (age and education) were not significant in explaining
the gap, “labor market variables” (including sector and occupation) were significant in
explaining the gap. One possible explanation for gender differences is how females respond
to financial literacy questions compared to males. Lusardi and Mitchell (2014) indicate that
females are disproportionately more likely than males to respond to a question with “I don’t
know.” Bucher-Koenen, Lusardi, Alessie, and van Rooij (2017) find that females tend to
give themselves lower scores in self-assessed knowledge. They conclude that “I don’t
know” reflects not simply the lack of knowledge but rather the lack of confidence in their
possessed knowledge.

Riener and Wagner (2017) find that girls are more likely to skip questions (“I Don’t
Know”) than boys if they deem them difficult. They find differences cease when extrinsic
rewards are provided. They argue that their findings are consistent with a stereotype threat
explanation. Support for a stereotype threat explanation hinges on the disappearance of the
gender gap when the task’s difficulty is made less salient. The quantitative nature of

174 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



financial literacy may also offer a possible explanation for females skipping questions.
Saygin and Atwater (2021) find that female test-takers skip significantly more questions
than males in quantitative areas, while no differences exist in nonquantitative areas. Overall,
in-person classes would increase students’ financial knowledge and behavior, ultimately
improving financial success and responsibility.

Our hypothesis H2B builds on the arguments that females and males would show a statis-
tically significant difference in learning with virtual/hybrid courses than in-person, with
females showing a more significant improvement in financial knowledge than males.
Pangestu and Karnadi (2020) and Gerrans and Heaney (2019) support this hypothesis by
concluding in their research that, since women are building from a lower financial literacy
base, a more significant improvement is observed when compared to men’s improvement.

3. Sample and methodology

The CFA Society Pittsburgh has been active in financial literacy outreach for over a dec-
ade, with a financial literacy committee charged with curricular development and training.
Financial literacy leaders conduct Act 48 training sessions in several Pennsylvania locations
and deliver hour-long presentations on core financial literacy concepts as requested. The
CFA Society Pittsburgh provides participating high schools with instructional materials for a
semester-long equivalent course based on The Missing Semester (Kabala & Natali 2012) and
The Missing Second Semester (2020). Schools voluntarily participate and deliver their pro-
grams based on their schedule. Participating teachers are supplied with PowerPoint resour-
ces to accompany the book, along with a web-based portal (available through the CFA
Society Pittsburgh website) of best practices and exercises submitted from previous partici-
pating schools. Program participation is open to any high school. Each school determines
the nature of program delivery, which ranges from within the context of a subject class
(such as math) or as a standalone unit. The teachers decide at what point in high school to
offer the programs. In most cases, programs are delivered as for-credit courses.

After gathering the list of enrolled teachers, we assigned each a unique class code. Links
for pre- and post-surveys were provided to the participating teachers. In the introductory
email, instructors were given directions to assign each student a unique ID number, allowing
pre- and post-surveys to be matched for analysis. Participating schools agreed to administer
the pre-survey before any instructional delivery. Post-surveys were completed within a week
after completing the last instructional unit on financial literacy.

The participants of surveys that study the effectiveness of financial literacy education
consisted of 2,297 students (1,622 in 2017-2019 and 675 in 2020-2021) across 108 schools
(79 in 2017-2019 and 29 in 2020-2021). This sample is labeled the “whole sample.” A
description of each survey follows:

• Pre-survey: Distributed to students at the beginning of the financial education program by teach-
ers that measure students’ initial financial knowledge in four main categories: subjective knowl-
edge, objective knowledge, behavior, and self-esteem.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 175



• Post-survey: Distributed to students upon completion of the financial education program by
teachers that measure the same four categories from the pre-survey to analyze. Different input
variables are used for quantitative questions measuring objective knowledge.

We also conducted a teacher survey during 2020-2021 to obtain information about the indi-
vidual financial education programs, including program length, modality, and resources used
to teach the material. The surveys distributed were extensions of the work conducted by
Filbeck, Zhao, and Pettner (2020). The three surveys can be found in Appendixes A, B, and C.

Table 1 reports the descriptive statistics for the pre-survey and post-survey samples. The
pre-survey sample consists of 2,297 students completing the pre-survey, while the post-sur-
vey sample includes only 1,138 students who submitted both a pre- and post-survey.[1] Of
the pre-survey sample, 2,075 (90.3%) students are in their junior or senior year; in the post-
survey sample, 1,021 (89.7%) students are in their junior or senior year. Female students
account for approximately 47.0% and 48.6%, respectively, in the pre-survey and post-survey
samples. Additionally, students who participated during 2020-2021 make up 29.4% in the
pre-survey sample and 26.8% in the post-survey sample. Moreover, of the pre-survey (post-
survey) sample, 94.9% (98.0%) of students are in their junior or senior years participated
during 2020-2021, while the corresponding number is 88.0% (86.7%). This finding suggests
a higher percentage of students who participated during 2020-2021 with online/hybrid learn-
ing are in their junior or senior years.

3.1. Survey methodology

The surveys are modified based on Filbeck, Zhao, and Pettner (2020) based on the frame-
work established by Tang and Baker (2016), using path models to analyze the relationship

Table 1 Sample description

Grade

9th 10th 11th 12th Total

Panel A. Whole sample
Year 2017–2019
Female 36 39 147 536 758
Male 54 59 194 557 864

Year 2020–2021
Female 11 4 200 105 320
Male 3 16 213 123 355

Total 104 118 754 1321 2297

Panel B. Test sample
Year 2017–2019
Female 19 26 85 264 394
Male 30 36 112 261 439

Year 2020–2021
Female 1 1 106 52 160
Male 0 4 101 40 145

Total 50 67 404 617 1138

Note. Table 1 shows the number of students across different grade levels and favorite subjects for the whole
sample (Panel A) and the test sample (Panel B).

176 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



between characteristics and four financial literacy measures. The model shows independent
variables produce both direct and indirect effects on a dependent variable. Behavior
(Subjective) is the total score for the questions in financial behavior (subjective financial
knowledge) questions, while Objective (IDK) is the total scores for students who choose cor-
rect answers (“I Don’t know” answers) for objective questions. We use financial behavior
scores (Behavior), subjective financial knowledge scores (Subjective), correct answers in the
objective questions (Objective), and “I don’t know” answers in the objective questions
(IDK) as dependent variables and test the effect of the financial literacy program. Detailed
variable definitions can be found in Appendix D.

We introduce several control variables associated with our study. In all three path models,
we include gender (Female), grade level (Sophomore, Junior, and Senior), favorite subject
(English, Math, and Science), and GPA. Following Amagir et al. (2018), we also include
learning method preferences (Learning by doing [LBD], Listening, Discussing, and Visual)
as independent variables that may impact financial literacy program success.

Students’ pre- and post-survey questions have been divided into two major categories: fi-
nancial behavior and financial knowledge. Financial knowledge questions are further split
into objective and subjective financial knowledge. The surveys consist of 21 questions: three
financial behavior, six objective financial knowledge, and 12 subjective financial knowledge.
Subjective financial knowledge and financial behavior questions are rated on a 5-point scale
ranging from 1 ¼ strongly disagree to 5 ¼ strongly agree. The three financial behavior ques-
tions are “I like to save money more than I like to spend it,” “I have a checking and/or a sav-
ings account,” and “I have conversations with my parents regarding personal finance.”
Subjective financial knowledge questions involve a perceived understanding of financial
concepts. They include questions such as “I understand how to establish a financial plan,” or
“I understand the process by which my parents/guardians make financial decisions.”

Five broad categories of financial literacy make up the survey questions: interest (numer-
acy), compound interest, inflation, and credit. However, objective financial knowledge ques-
tions are conducted with “right” or “wrong” answers. Each objective financial knowledge
question contains at least one wrong answer and the option to choose “I Don’t Know.” The
questions are analyzed using two methods: willingness to answer and correctness. The first
method, willingness to respond, assigns a score of 1 for an answer of “I Don’t Know” and 0
for any other answer. Secondly, questions with the correctness method assign a score of 1
for each correct answer and 0 for any other answer.

The teacher survey comprises five informational questions regarding how the instructors
teach financial literacy courses in their respective high schools. Of the five informative ques-
tions, two focus on the mode of instruction, while the other three focus on the program’s du-
ration. Questions regarding the mode include: “How are you teaching the material?” and
“What methods did you use to teach the material?” Concerning durational questions, these
include: “How many total contact hours will you spend teaching financial literacy?” “How
often will students receive financial literacy instruction?” and “What best describes the total
length of your financial literacy instruction program.” For 2020-2021, according to the
teacher’s survey results, teachers conducted their programs in an online/hybrid mode
(instruction occurring in a combination of virtual and in-person formats).

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 177



4. Results

4.1. Pre-survey analysis

For the t-test of the pre-survey responses, two characteristics are analyzed: student gender
and the delivery mode of content. Student gender is divided into subgroups (female or
male), while the delivery mode is based on year: in-person (2017-2019) or virtual/hybrid
(2020-2021). Table 2 lists the initial results of the pre-surveys submitted by students catego-
rized by the four areas of financial knowledge tested: subjective knowledge, objective
knowledge, financial behavior, and self-esteem. We only report the total scores for each area
of financial knowledge in Table 2 as Cronbach’s a for each area of financial literacy ranges
between 0.71-0.88, which is at an acceptable level of reliability and suggests the category
totals are consistent measures of each question responses in that category. The detailed
responses for each question are reported in Appendix E: Table 2A.

The average responses are compared to the individual subgroups of gender and the
courses’ modality. Compared to the total average response score for subjective financial
knowledge questions, female students scored lower than males (statistically significant at the
1% level). Students who learned virtually/hybrid in 2020-2021 show a slightly higher score
than in-person learning, but the difference is not statistically significant. For financial behav-
ior, students who partook in a virtual/hybrid format are better financially behaved (statisti-
cally significant at the 1% level), while no statistically significant differences exist based on

Table 2 Differences based on student characteristics: Pre-survey results

Question type Average response Gender Modality

Male Female In-person Online/hybrid

Panel A. Financial subjective knowl-
edge questions

Total score 33.928 34.571 33.196*** 33.893 34.010

Panel B. Financial behavior questions
Total score 14.895 14.835 14.964 14.746 15.254***

Panel C. Objective questions (correct
answers)

Total score 2.982 3.253 2.707*** 2.934 3.100**

Panel D. Objective questions (“I don’t
know” answers)

Total score 1.667 1.411 1.977*** 1.717 1.545**

Note. Table 2 shows the differences of pre-survey student responses on financial knowledge, financial behav-
ioral, and objective questions across different gender and GPAs for the whole sample. We only report total
scores for each category as Cronbach’s alpha for each category ranges between 0.71-0.88, which is at an accept-
able level for internal consistency. Detailed responses for each question can be found in Appendix: Table 2A.
Gender is identified as a dummy variable equal to 1 if the student is female and 0 otherwise. Modality is denoted
as a dummy variable equal to one if the sample comes from an online/hybrid modality, the 2020-2021 sample,
otherwise, the dummy variable is zero, indicating an in-person structure based on the sample from 2017 to 2019.
We also conduct a t-test to test the differences in gender (i.e., male vs. female) and modality (in-person vs.
online/hybrid). We use asterisks to show the significance of the t-test result.
***, **, * indicate statistical significance at 0.01, 0.05, and 0.10 level, respectively.

178 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



gender. Females scored lower in correctness and self-esteem regarding objective financial

knowledge, while virtually/hybrid trained students scored higher in correctness and self-

esteem (statistically significant at the 1% level). This finding is consistent with our H2A in

that male students have better baseline financial literacy than female students before the sur-

vey in all areas except for financial behavior. Our results are consistent with Riener and

Wagner (2017) and Saygin and Atwater (2021) and may support the stereotype threat expla-

nation. This explanation would argue the gender gap vanishes if the task’s difficulty is made

less salient (Riener & Wagner, 2017).
Table 3 reports Pearson correlations between financial behavior and self-esteem (IDK),

subjective, and objective financial knowledge for both the whole sample and the two sub-
samples. Financial behavior, objective, and subjective knowledge are significantly correlated
with each other. IDK is significantly negatively correlated with the other three measures of
financial literacy, which indicates a positive correlation between self-esteem level and the
other three measures of financial literacy. Naturally, because of the setup of the self-esteem
measure, we find a negative correlation ("0.76) between IDK answers and objective knowl-
edge. Comparing in-person and virtually/hybrid learning modality, we do not observe signif-
icant differences in correlations between these four variables during 2017-2019 and 2020-
2021.

4.2. Post-survey analysis

The results of the pre- and post-surveys are compared using the post-survey sample of

833 students in 2017-2019 (in-person) and 305 students in 2020-2021 (virtual/hybrid).

Improvement is defined in several ways. Gains from subjective financial knowledge, objec-

tive financial knowledge, and financial behavior are defined as the post-survey scores minus

the pre-survey response scores. To gauge financial self-esteem, we define confidence gains

as a decrease in “I Don’t Know” responses in the post-survey minus the pre-survey. In other

words, students exhibit better self-esteem when they have fewer “I Don’t Know” answers in

the post-survey compared to the pre-survey.
Table 4 illustrates the t-test results by question and overall score for each of the four items

measured: subjective financial knowledge, financial behavior, objective financial knowledge,

and financial self-esteem. We report results of total scores of each financial literacy category

in Table 4 and detailed results for each question in Appendix F: Table 4A. Questions from

subjective knowledge and financial behavior are assessed based on 1-5, 1 ¼ I highly dis-

agree and 5 ¼ I highly agree. The results in Table 4 and Appendix F: Table 4A show

improvements with nearly every question in all categories, per question and total. For exam-

ple, the financial behavior question “I like to save money more than I like to spend it” had

an average pre-survey score of 3.54 and a post-survey score of 3.81. The increase shows

that, on average, students’ preference for the balance between saving and spending

improved, improving their overall financial behavior. Each of the improvements is statisti-

cally significant at the 1% level, except for one question in the financial behavior category “I

have a checking and/or savings account.” Given the COVID-19 pandemic that occurred in

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 179



2020-2021, the result for this question is reasonable given the reduction of in-person banking

services.
The most significant improvement of student responses to questions related to subjective

financial knowledge comes from understanding Roth IRA (a gain of 1.653) and retirement
(an increase of 1.376). All subjective financial knowledge questions are statistically

Table 3 Correlation coefficients between financial behavior and self-esteem, objective and subjective financial
knowledge

Question type Subjective Behavior Objective IDK answers

Panel A. Whole sample
Subjective
Corr 1.000
p-value

Behavior
Corr 0.357*** 1.000
p-value <.0001

Objective
Corr 0.286*** 0.267*** 1.000
p-value <.0001 <.0001

IDK answers
Corr "0.364*** "0.267*** "0.759*** 1.000
p-value <.0001 <.0001 <.0001

Panel B. Year 2017–2019
Subjective
Corr 1.000
p-value

Behavior
Corr 0.349*** 1.000
p-value <.0001

Objective
Corr 0.291*** 0.255*** 1.000
p-value <.0001 <.0001

IDK answers
Corr "0.366*** "0.255*** "0.753*** 1.000
p-value <.0001 <.0001 <.0001

Panel C. Year 2020–2021
Subjective
Corr 1.000
p-value

Behavior
Corr 0.384*** 1.000
p-value <.0001

Objective
Corr 0.276*** 0.293*** 1.000
p-value <.0001 <.0001

IDK answers
Corr "0.359*** "0.292*** "0.774*** 1.000
p-value <.0001 <.0001 <.0001

Note. Table 3 shows the correlation coefficients between financial behavior, self-esteem (IDK answers), and
objective and subjective financial knowledge.
***, **, * indicate statistical significance at 0.01, 0.05, and 0.10 level, respectively.

180 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



significant at the 1% level. Further analyzing the financial behavior questions’ responses, the
biggest gain derives from the importance of contributing to a retirement plan (an increase of
0.402). All questions are statistically significant at least the 5% level.

The results link confidence to answer a question (self-esteem) and correctness (objective
financial knowledge). Meanwhile, the most significant improvements seen from objective fi-
nancial knowledge regarding questions involving credit, the concept of an agreement to pur-
chase a product or service with the express promise to pay for it later (a gain of 0.251), and
knowledge of inflation rate (a gain of 0.200). The largest increase in correct responses
between pre- and post-surveys comes from self-esteem questions (how often students
answered “I Don’t Know” on the objective financial questions). These two questions also
show the greatest improvement in students’ self-esteem. That is, fewer students answered “I
Don’t Know” with those two questions than any other between the pre- and post-surveys.

Next, we test H1 and H2B with subgroup results. Table 5 reports the t-test results for dif-
ferent subgroups. Panel A reports the subgroups by gender. The statistically significant
improvement across all subgroups at the 1% level for both male and female groups indicates
a significant improvement after completing the financial literacy program for both males and
females. t-test results on the differences of improvements between males and females sug-
gest that female students experience more significant improvements in financial behavior,
objective knowledge, and self-esteem than their male peers. This finding is consistent with
Gerrans and Heaney (2019) and H2B.

Panel B of Table 5 reports the subgroup results by modality (in person, 2017-2019 vs.
online/hybrid, 2020-2021). The statistically significant differences between pre- and post-
survey in all areas of financial literacy and both modalities show that students have signifi-
cant improvements after the literacy program in both modalities. These findings reject H1
and are consistent with Wolla’s (2017) finding that online learning effectively increases fi-
nancial knowledge among high school students. Further, students’ financial behavior gains
are statistically significantly higher in 2020-2021 than the gains experienced during 2017-
2019. However, this result should be interpreted with caution as the sample has a potential
attrition issue because of COVID-19 and will be tested with more robustness checks.

Next, we run regression analysis to examine how student characteristics and other control
variables affect their knowledge and behavior gains. Table 6 reports the regression results.

Table 4 t-test results between pre- and post-survey

Question type Pre Post Diff T-stat

Panel A. Financial subjective knowledge questions
Total score 33.839 43.963 10.125 36.38***

Panel B. Financial behavior questions
Total score 14.944 16.365 1.421 15.11***

Panel C. Objective questions (correct answers)
Total score 2.970 3.979 1.009 18.55***

Panel D. Objective questions (“I don’t know” answers)
Total score 1.690 0.482 "1.209 "22.43***

Note. Table 4 shows the t-test results of student responses to financial behavior and knowledge questions
before and after the financial literacy educational efforts for the test sample.
***, **, * indicate statistical significance at 0.01, 0.05, and 0.10 level, respectively.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 181



T
ab
le

5
t-
te
st
R
es
u
lt
s
b
et
w
ee
n
p
re
-
an
d
p
o
st
-s
u
rv
ey

fo
r
d
if
fe
re
n
t
su
b
sa
m
p
le
s

S
u
b
sa
m
p
le
ty
p
e

F
in
an
ci
al
b
eh
av
io
r

S
u
b
je
ct
iv
e
q
u
es
ti
o
n
s

O
b
je
ct
iv
e
q
u
es
ti
o
n
s

S
el
f
es
te
em

P
re

P
o
st

P
o
st
-
P
re

P
re

P
o
st

P
o
st
-
P
re

P
re

P
o
st

P
os
t
-
P
re

P
re

P
o
st

P
o
st
-
P
re

P
an
el
A
.
S
u
b
sa
m
p
le
s
b
y
g
en
d
er

M
al
e
(1
)

3
4
.4
6
4

4
3
.6
0
2

9
.1
3
8
*
*
*

1
4
.9
01

1
6
.2
70

1
.3
6
9
*
*
*

3
.2
1
5

4
.0
7
2

0
.8
5
7
*
*
*

1
.4
1
0

0
.4
2
5

"
0
.9
8
5
*
*
*

F
em

al
e
(2
)

3
3
.1
7
7

4
4
.3
4
5

1
1
.1
6
8
*
**

1
4
.9
89

1
6
.4
66

2
.7
1
1
*
*
*

3
.8
8
1

3
.8
8
1

1
.1
7
0
*
*
*

1
.9
8
7

0
.5
4
2

"
1
.4
4
6
*
*
*

(1
)
"

(2
)

"
2
.0
3
0
*
*
*

"
0
.1
0
8

"
0
.3
1
3
*
*
*

0
.4
6
1
*
*
*

P
an
el
B
.
S
u
b
sa
m
p
le
s
b
y
y
ea
r

Y
ea
r
2
0
1
7
-2
01
9
(3
)

3
3
.7
8
0

4
3
.6
7
7

9
.8
9
7
*
*
*

1
4
.7
95

1
6
.3
34

1
.5
3
9
*
*
*

2
.9
3
6

3
.9
4
4

1
.0
0
7
*
*
*

1
.7
7
0

0
.5
5
5

"
1
.2
1
5
*
*
*

Y
ea
r
2
0
2
0
-
2
0
2
1
(4
)

3
3
.9
9
7

4
4
.7
3
9

1
0
.7
4
3
*
**

1
5
.3
49

1
6
.4
50

1
.1
0
1
*
*
*

3
.0
6
2

4
.0
7
5

1
.0
1
3
*
*
*

1
.4
7
6

0
.2
8
3

"
1
.1
9
2
*
*
*

(3
)
"

(4
)

"
0
.5
5
3
*
*
*

0
.4
3
8

"
0
.0
0
6

"
0
.0
2
3

N
o
te
.
T
ab
le
5
sh
o
w
s
th
e
t-
te
st
re
su
lt
s
o
f
st
u
d
en
t
re
sp
o
n
se
s
to

se
lf
-e
st
ee
m
,
fi
n
an
ci
al
b
eh
av
io
r,
su
b
je
ct
iv
e,
an
d
o
b
je
ct
iv
e
q
u
es
ti
o
n
s
b
ef
o
re

an
d
af
te
r
th
e
fi
n
an
-

ci
al
li
te
ra
cy

ed
u
ca
ti
o
n
al
ef
fo
rt
s
fo
r
d
if
fe
re
n
t
su
b
sa
m
p
le
s.

*
*
*
,
*
*
,
*
in
d
ic
at
e
st
at
is
ti
ca
l
si
g
n
ifi
ca
n
ce

at
0
.0
1
,
0
.0
5
,
an
d
0
.1
0
le
v
el
,
re
sp
ec
ti
v
el
y
.

182 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



Table 6 Regression results

Model (1) Model (2) Model (3) Model (4)

Dep. Var.:
Diff_BEHAV

Dep. Var.:
Diff_KNOW

Dep. Var:
Diff_OBJ

Dep. Var.:
Diff_IDK

Coefficient t-stat Coefficient t-stat Coefficient t-stat Coefficient t-stat

Panel A. OLS regression results on student and school district characteristics

Intercept 6.740 0.46 "20.697 "0.47 16.818 1.91* "8.520 "1.01
Year 2020 "0.036 "0.13 2.945 3.61*** 0.311 1.89* "0.351 "2.23**
Female 0.022 0.10 1.537 2.28** 0.274 2.02** "0.460 "3.53***
Upperclass "0.286 "0.76 1.606 1.45 "0.071 "0.32 0.120 0.56
GPA "0.162 "1.01 0.190 0.40 "0.149 "1.55 0.280 3.03***
Log (Population) "1.225 "4.42*** "6.492 "7.89*** "0.822 "4.96*** 0.741 4.66***
Log (Household_
Income)

0.611 0.42 7.633 1.78* "0.695 "0.80 0.000 0.00

Poverty "0.510 "0.05 31.186 1.09 "3.677 "0.64 "0.908 "0.16
Pct_College 2.788 1.89* 11.935 2.73*** 2.369 2.69*** "1.949 "2.31**

Model (1) Model (2) Model (3) Model (4)

Dep. Var.:
Diff_BEHAV

Dep. Var.:
Diff_KNOW

Dep. Var:
Diff_OBJ

Dep. Var.:
Diff_IDK

Coefficient Z-stat Coefficient Z-stat Coefficient Z-stat Coefficient Z-stat

Panel B. GLM regression results after controlling for fixed effect of classes, with standard error clustered at the class level

Intercept 3.087 4.17*** 7.549 3.71*** 1.091 2.57** "1.662 "3.18***
Year 2020 "0.782 "2.02** "1.062 "0.60 "0.055 "0.18 0.189 0.66
Female 0.169 1.07 2.313 3.66*** 0.397 3.09*** "0.526 "4.71***
Upperclass "0.359 "1.14 1.377 0.94 0.044 0.17 "0.020 "0.08
GPA "0.238 "1.69* 0.402 0.92 "0.073 "0.76 0.158 1.85*
English 0.014 0.05 "1.250 "2.20** "0.231 "2.33** 0.071 0.54
Math "0.082 "0.26 "0.138 "0.15 "0.247 "1.50 0.170 1.44
Science 0.119 0.33 "0.582 "0.78 "0.248 "1.97** 0.184 1.67*
LBD "0.207 "1.11 0.628 1.64 0.218 2.00** "0.116 "0.86
Listening "0.271 "1.17 "1.638 "1.83* "0.236 "1.50 0.300 1.94*
Discussing "0.125 "0.35 "1.424 "1.54 "0.101 "0.55 0.143 0.98
Visual "0.196 "1.13 0.675 1.04 0.209 1.39 "0.142 "1.10

Note. Table 6 shows the regression results of the test sample. Panel A reports the OLS regression results on stu-
dent and school district characteristics. Panel B reports the regression results on student characteristics after con-
trolling for fixed effect of classes, with standard errors clustered at the class level. Diff_BEHAV (Diff_SUBJ,
Diff_OBJ, Diff_IDK) is the difference between the student’s pre- and post-survey scores for the financial behavior
(knowledge, objective) questions. Year2020 is a dummy variable equal to 1 if it belongs to the 2020-2021 sample.
Female is a dummy variable equal to 1 if the student is a female student and 0 otherwise. Upperclass is a dummy
variable equal to 1 if the student is a junior or senior and 0 otherwise. English (Math, Science) is a dummy variable
equal to 1 if the student’s favorite subject is English (math, science) and 0 otherwise. LBD (Listening, Discussing,
and Visual) is a dummy variable equal to 1 if the student chooses learning by doing (listening, discussing with
peers, and features visual support) as a favorite instruction method, and 0 otherwise. GPA is a student’s grade point
average. Log(Population) is the log of the population in the school district. Log(Household_Income) is the log of
the median household income in the school district. Poverty is the poverty rate in the school district. Pct_College
is the percentage of students with parents/guardians who have attained a bachelor’s degree or higher.
***, **, * indicate statistical significance at 0.01, 0.05, and 0.10 level, respectively.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 183



Panel A reports the regression results on student characteristics and school district character-
istics. The dependent variable is Diff_BEHAV (Diff_SUBJ, Diff_OBJ, Diff_IDK), which is
the difference between the students’ pre- and post-study scores (post-minus pre) for the fi-
nancial behavior (subjective, objective) questions. All the other variables are listed in
Appendix D. The results show that students who participated in virtually/hybrid format and
female students gained most in all areas except financial behavior.

Similarly, students in school districts with a smaller population and a higher percentage
of parents/guardians with bachelor’s degrees experience a statistically significant gain in fi-
nancial behavior. This finding is encouraging, as Stolper and Walter (2017) point out that
the opportunity to relate financial literacy to various demographics in the context of their
spending behavior is key to program success. In financial knowledge, we find students com-
ing from higher-income households perform better. These results echo the findings of Kaiser
and Menkhoff (2017), who find that financial literacy programs are less effective for clients
with lower incomes and from low- and lower-middle-income economies. In such environ-
ments, altering financial behaviors in debt handling may be less effective.

The results in Panel A may be misleading on the conclusion of modality because schools
with less ability to conduct financial literacy programs during the COVID pandemic may not
be participating in the 2020-2021 period. Students who participated in the 2020-2021 pro-
gram may come from a school with better infrastructure to adapt to the challenges of online
learning. Therefore, we need to control for school or class differences in our regressions.
Next, we include other control variables such as favorite subjects and learning style and use
fixed effect regressions controlling for class differences.

Specifically, the class fixed effects allow the class dummy variable to differ and control
for the variations across classes. We also cluster standard errors at the class level. Table 1
shows a higher percentage of junior or senior students who participated in the 2020-2021
program. Therefore, we add Upperclass (i.e., a dummy variable equal to 1 if the student is a
junior or senior) as an independent variable to control for grade level in the regression.
Panel B of Table 6 reports the regression results. We use the same dependent variables as in
Panel A. The results show no statistically significant improvements except for financial
behavior for students who participated in virtually/hybrid learning programs. This finding is
consistent with Table 5 and rejects H1. A trio of findings is notable for female students.
First, female students whose favorite subject is not English tend to gain more subjective fi-
nancial knowledge. In addition, female students whose favorite subject is not English or
Science and who prefer more hands-on learning opportunities (learning by doing) gain more
objective financial knowledge. Finally, female students with higher GPA experience the
most significant self-esteem gain.

Results in Tables 4-6 use test samples (including pre- and post-survey results from stu-
dents who participated in both pre- and post-survey) and test the improvements (i.e., differ-
ences from pre- to post-survey scores) in students’ financial literacy. However, this setup
may be subject to potential self-selection and attrition bias in our sample. Specifically, from
Table 1, we can see that about 52% of female students and 51% of male students completed
both pre- and post-survey during 2017-2019, while the corresponding numbers for the 2020-
2021 sample are 50% and 41%, respectively. In other words, in the 2017-2019 sample, there

184 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



is roughly 51% to 52% attrition for both male and female students. While for the 2020-2021
sample, there is 50% attrition for female students, but about 59% attrition for male students.

Given that male students have higher baseline financial literacy and less scope for finan-
cial literacy improvement, this might bias the results of comparisons between the 2017-2019
sample and the 2020-2021 sample as presented in previous tables.[2] Also, the sample is sub-
ject to self-selection bias as schools with less ability to conduct financial literacy programs
during the COVID pandemic may not be participating in the 2020-2021 period. Students
who participated in the 2020-2021 program may come from a school with better infrastruc-
ture to adapt to the challenges of online learning.

To address this potential self-selection and attrition bias in our sample, we pool our whole
pre-survey data (from all students who participated in pre-survey) and post-survey data. We
use the standard diff-in-diff model and run regression models using outcome variables (i.e.,
Behavior, Subjective, Objective, and IDK) as dependent variables. We report the results in
Table 7. In our regressions, we add Post, Post # Female, and Post # Year2020 as additional
control variables. Post is a dummy variable that equals 1 if the data are post-survey results and
0 otherwise. Post# Female is an interaction term and captures the change in the outcome vari-
able (i.e., Behavior, Subjective, Objective, and IDK) from the pre-survey period to the post-
survey period for female students, relative to the change in outcome variable for male students.
Similarly, Post # Year2020 is an interaction term that captures the change in outcome varia-
bles from the pre-survey period to the post-survey period for the 2020-2021 program, relative
to the change in outcome variable for the 2017-2019 program. Other control variables are
defined the same as in Table 6 and Appendix D. We also include student fixed effects to con-
trol student characteristics differences, with standard errors clustered at the student level.

The statistically significant coefficient on Post in Table 7 suggests that students show stat-
istically significant improvements in all four financial literacy areas in virtually/hybrid learn-
ing and in-person programs after controlling for potential attrition bias. Moreover, the
statistically insignificant regression coefficient on Post # Year2020 in Table 7 using
Subjective, Objective, and IDK as dependent variable (i.e., Models 2-4) suggest that there
are no statistically significant differences of improvements in students subjective knowledge,
objective knowledge, and self-esteem in virtual/hybrid format compared with the in-person
format. This finding is consistent with our previous results in Table 6 and rejects our H1.
The statistically significant coefficients Female in Models 2-4 also confirm our previous
results that male students have a higher level of financial literacy in all areas except for fi-
nancial behavior than female students, which supports our H2A. More importantly, female
students experienced the most gains in all financial literacy programs except for financial
behavior, as evidenced by the statistically significant interaction term Post # Female. This
supports our H2B.

5. Conclusions

This study’s primary purpose is to investigate the effectiveness of high school financial
literacy education programs in a virtual/hybrid environment instead of an in-person setting.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 185



T
ab
le

7
R
eg
re
ss
io
n
re
su
lt
s:
D
if
fe
re
n
ce
-i
n
-d
if
fe
re
n
ce

m
o
d
el

M
o
d
el
(1
)

M
o
d
el
(2
)

M
o
d
el
(3
)

M
o
d
el
(4
)

D
ep
.
V
ar
.:
B
eh
av
io
r

D
ep
.
V
ar
.:
S
u
b
je
ct
iv
e

D
ep
.
V
ar
:
O
b
je
ct
iv
e

D
ep
.
V
ar
.:
ID

K

C
o
ef
fi
ci
en
t

Z
-S
ta
t

C
o
ef
fi
ci
en
t

Z
-S
ta
t

C
o
ef
fi
ci
en
t

Z
-S
ta
t

C
o
ef
fi
ci
en
t

Z
-S
ta
t

In
te
rc
ep
t

1
1
.2
0
9
1

2
7
.5
1
*
*
*

3
2
.7
2
5

3
2
.4
3
*
*
*

1
.1
5
4

5
.3
4
*
*
*

2
.6
9
4

1
3
.1
8
*
*
*

Y
ea
r2
0
2
0

0
.7
0
2

3
.3
4
*
*
*

0
.9
1
9

1
.6
3

0
.1
8
2

1
.4
2

"
0
.2
1
7

"
1
.7
7
*

P
o
st

1
.5
2
5

1
0
.0
3
*
*
*

8
.8
5
3

2
1
.7
0
*
*
*

0
.7
6
8

1
0
.2
6
*
*
*

"
0
.9
2
7

"
1
4
.2
5
*
*
*

P
o
st
#

Y
ea
r2
0
2
0

"
0
.4
7
2

"
2
.4
6
*
*

0
.7
1
4

1
.3
6

"
0
.0
2
4

"
0
.2
1

"
0
.0
9
4

"
1
.0
1

F
em

al
e

"
0
.0
4
7

"
0
.3
4

"
1
.4
6
2

"
4
.3
8
**
*

"
0
.6
5
4

"
8
.4
4
*
*
*

0
.6
3
6

7
.8
8
*
*
*

P
o
st
#

F
em

al
e

0
.0
6
4

0
.3
5

2
.1
4
6

4
.1
9
**
*

0
.3
9
1

3
.8
3
*
*
*

"
0
.4
5
0

"
4
.6
7
*
*
*

U
p
p
er
cl
as
s

0
.7
5
1

3
.8
3
*
*
*

"
0
.0
4
1

"
0
.0
9

0
.1
6
8

1
.6
1

"
0
.1
1
8

"
1
.1
3

G
P
A

0
.7
6
0

9
.3
7
*
*
*

0
.2
0
6

0
.9
9

0
.5
1
8

1
0
.6
8
*
*
*

"
0
.3
2
5

"
7
.0
4
*
*
*

E
n
g
li
sh

"
0
.3
7
4

"
2
.1
0
*
*

"
0
.1
7
7

"
0
.4
1

"
0
.3
0
9

"
3
.0
3
*
*
*

0
.1
8
3

1
.9
7
*
*

M
at
h

"
0
.2
2
7

"
1
.5
1

0
.0
2
4

0
.0
6

"
0
.0
6
2

"
0
.7
0

0
.0
3
9

0
.5
2

S
ci
en
ce

"
0
.3
6
1

"
2
.3
6
*
*

"
0
.1
4
3

"
0
.3
7

0
.0
8
8

1
.0
3

"
0
.0
5
9

"
0
.7
6

L
B
D

0
.4
2
8

3
.2
5
*
*
*

0
.8
7
4

2
.6
5
**
*

0
.2
6
8

3
.7
2
*
*
*

"
0
.0
7
4

"
1
.1
6

L
is
te
n
in
g

0
.0
6
1

0
.4
6

0
.6
7
7

1
.9
9
**

0
.0
0
6

0
.0
8

"
0
.1
0
1

"
1
.4
3

D
is
cu
ss
in
g

0
.0
7
7

0
.5
5

0
.5
9
5

1
.6
9
*

0
.0
1
5

0
.1
7

"
0
.0
1
2

"
0
.1
6

V
is
u
al

0
.2
6
1

1
.7
0
*

"
0
.1
0
5

"
0
.2
7

0
.0
8
7

0
.9
5

0
.0
3
6

0
.4
2

N
o
te
.
T
ab
le

7
sh
o
w
s
th
e
re
g
re
ss
io
n
re
su
lt
s
u
si
n
g
a
d
if
f-
in
-d
if
f
m
o
d
el
.
S
p
ec
ifi
ca
ll
y
,
w
e
u
se

o
u
tc
o
m
e
v
ar
ia
b
le
s
(i
.e
.,
B
eh
av
io
r,
S
u
b
je
ct
iv
e,

O
b
je
ct
iv
e,

an
d

ID
K
)
as

d
ep
en
d
en
t
v
ar
ia
b
le
s
w
it
h
p
o
o
le
d
p
re
-
an
d
p
o
st
-s
ur
v
ey

d
at
a.

W
e
re
p
o
rt
th
e
re
g
re
ss
io
n
re
su
lt
s
o
n
st
u
d
en
t
ch
ar
ac
te
ri
st
ic
s
af
te
r
co
n
tr
o
ll
in
g
fo
r
fi
x
ed

ef
fe
ct

o
f
st
u
d
en
ts
,
w
it
h
st
an
d
ar
d
er
ro
rs

cl
u
st
er
ed

at
th
e
st
u
d
en
t
le
v
el
.
B
eh
av
io
r
(S
u
b
je
ct
iv
e,

O
b
je
ct
iv
e,

ID
K
)
is
th
e
st
u
d
en
t’
s
su
rv
ey

sc
o
re
s
fo
r
th
e
fi
na
n
ci
al

b
eh
av
io
r
(k
n
o
w
le
d
g
e,
o
b
je
ct
iv
e)

q
u
es
ti
o
n
s.
Y
ea
r2
0
2
0
is
a
d
u
m
m
y
v
ar
ia
b
le

eq
u
al

to
1
if
it
b
el
o
n
g
s
to

th
e
2
0
2
0
-2
02
1
sa
m
p
le
.
P
o
st
is
a
d
u
m
m
y
v
ar
ia
b
le

eq
u
al

to
1
if
th
e
su
rv
ey

sc
o
re

is
fr
o
m

p
o
st
-s
ur
v
ey

an
d
0
o
th
er
w
is
e.

P
o
st
#

Y
ea
r2
0
20

is
an

in
te
ra
ct
io
n
te
rm

o
f
P
o
st
an
d
Y
ea
r2
0
2
0
.
F
em

al
e
is
a
d
u
m
m
y
v
ar
ia
b
le

eq
u
al
to

1
if
th
e
st
u
d
en
t
is
a
fe
m
al
e
st
u
d
en
t
an
d
0
o
th
er
w
is
e.
P
o
st
x
F
em

al
e
is
an

in
te
ra
ct
io
n
te
rm

o
f
P
o
st
an
d
F
em

al
e.
U
p
p
er
cl
as
s
is
a
d
u
m
m
y
v
ar
ia
b
le
eq
u
al

to
1
if
th
e
st
u
d
en
t
is
a
ju
n
io
r
o
r
se
n
io
r
an
d
0
o
th
er
w
is
e.

E
n
g
li
sh

(M
at
h
,
S
ci
en
ce
)
is
a
d
u
m
m
y
v
ar
ia
b
le

eq
u
al

to
1
if
th
e
st
u
d
en
t’
s
fa
v
o
ri
te

su
b
je
ct

is
E
n
g
li
sh

(m
at
h,

sc
ie
n
ce
)
an
d
0
o
th
er
w
is
e.
L
B
D
(L
is
te
n
in
g
,
D
is
cu
ss
in
g
,
an
d
V
is
u
al
)
is
a
d
u
m
m
y
v
ar
ia
b
le
eq
u
al
to

1
if
th
e
st
u
d
en
t
ch
o
o
se
s
le
ar
n
in
g
b
y
d
o
in
g
(l
is
te
n
in
g
,

d
is
cu
ss
in
g
w
it
h
p
ee
rs
,
an
d
fe
at
u
re
s
v
is
u
al

su
p
p
o
rt
)
as

a
fa
v
o
ri
te

in
st
ru
ct
io
n
m
et
h
o
d
,
an
d
0
o
th
er
w
is
e.

G
P
A

is
a
st
u
d
en
t’
s
g
ra
d
e
p
o
in
t
av
er
ag
e.

L
o
g

(P
o
p
u
la
ti
o
n
)
is

th
e
lo
g
o
f
th
e
p
o
p
u
la
ti
o
n
in

th
e
sc
h
o
o
l
d
is
tr
ic
t.
L
o
g
(H

o
u
se
h
ol
d
_
In
co
m
e)

is
th
e
lo
g
o
f
th
e
m
ed
ia
n
h
o
u
se
h
o
ld

in
co
m
e
in

th
e
sc
h
o
o
l
d
is
tr
ic
t.

P
o
v
er
ty

is
th
e
p
o
v
er
ty

ra
te

in
th
e
sc
h
o
o
l
d
is
tr
ic
t.
P
ct
_
C
o
ll
eg
e
is
th
e
p
er
ce
n
ta
ge

o
f
st
u
d
en
ts
w
it
h
p
ar
en
ts
/g
ua
rd
ia
n
s
w
h
o
h
av
e
at
ta
in
ed

a
b
ac
h
el
o
r’
s
d
eg
re
e
o
r

h
ig
h
er
.

*
*
*
,
*
*
,
*
in
d
ic
at
e
st
at
is
ti
ca
l
si
g
n
ifi
ca
n
ce

at
0
.0
1
,
0
.0
5
an
d
0
.1
0
le
v
el
,
re
sp
ec
ti
v
el
y
.

186 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



Financial literacy improvement is measured in four areas: subjective financial knowledge, fi-
nancial behavior, objective financial knowledge, and self-esteem.

The pre-survey results taken by students before beginning the financial education program
are initially analyzed using a t-test. Overall, the results show no statistical difference in pre-
survey financial subjective knowledge between the two subperiods based on modality.
Virtually/hybrid trained students scored higher in financial behavior, correctness, and self-
esteem. However, female students scored lower on subjective and objective financial ques-
tions and exhibited lower self-esteem regarding objective financial knowledge. The only
area in which females did not underperform males in the pre-survey was financial behavior.

To test the effectiveness of financial literacy programs, a t-test was implemented between
the pre- and post-survey results taken after completion of the course. The t-test analyzes the
four major topic areas listed above. Overall, male and female students experienced a statisti-
cally significant increase in all four topics. Interestingly, the one area in which no differences
occurred was whether students have a checking account. Given the reduction in in-person
banking during the pandemic, this finding is not surprising.

With the four areas studied, subjective financial knowledge gains center on retirement plan-
ning. Similarly, the greatest gains in financial behavioral questions are tied to the importance
of contributing to a retirement plan. Objective knowledge improvements and self-esteem
improvements occur most credit and inflation. While statistically, significant improvement
occurred for both genders, female students experience a greater improvement in financial
behavior, objective knowledge, and self-esteem than male peers.

Modality does not appear to impact the effectiveness of the financial literacy programs as
all areas of financial literacy and in both modalities show that students have significant
improvements. Thus, online learning is an effective tool for increasing financial literacy
among high school students.

Students in school districts with smaller populations and a higher percentage of parents
holding bachelor’s degrees experience a statistically significant gain in financial behavior.
Results should be interpreted with some caution as schools not participating in the financial
literacy program during the pandemic may have been in more socioeconomically challenged
areas than in more urban areas.

Based on the analysis extracted from the surveys, statistically, significant improvements
in subjective financial knowledge, financial behavior, objective financial knowledge, and fi-
nancial self-esteem lead us to conclude that the CFA Society Pittsburgh financial literacy
program successfully increases students’ chances of financial success. The analysis affirms
that one area the pandemic failed to impact was the effectiveness of financial literacy pro-
grams as no statistical differences exist between improvements based on modality.

These findings add to the existing literature in financial literacy, demonstrating that edu-
cational techniques including online and hybrid learning, which were often required during
the COVID-19 pandemic, do not diminish effectiveness when used for financial literacy pro-
grams. Based on these findings, we argue that school districts that currently do not require fi-
nancial literacy components within their curriculum consider making changes based on the
societal ramifications of such deficiencies and the significant improvements associated with
successful financial literacy programs, such as the focus of our study.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 187



Notes

1 To test whether there are selection biases between students who participated in
both pre- and post-survey and students who participated only in pre-survey, we
run T-test on the differences of students’ pre-survey results between students
who participated in the post-survey and students who participated in the pre-sur-
vey but not post-survey. We find no statistically significant differences in their fi-
nancial behavior, subjective and objective financial knowledge, and self-esteem
in the pre-survey results between these two groups. The results are available
upon request.

2 We thank an anonymous referee for valuable comments and suggestions on this issue.

Appendix A

Pre-survey questions

Class Code:
Student ID:
Gender:
GPA:
Grade:
Favorite subject in school:
____ English
____ Math
____ Social Studies
____ Science

Questions:

1. I like to save money more than I like to spend it.
2. I understand how to establish a financial plan.
3. I think financial literacy is important for my future.
4. I have a checking and/or a savings account.
5. I have conversations with my parents regarding personal finance.
6. I understand the process by which my parents/guardians make financial decisions.
7. I know how to determine the appropriate total costs associated with the colleges/universities I

am interested in attending.
8. I understand the process by which loan repayments take place including the impact of interest,

delinquency, and default.
9. I understand the process by which credit card charges and repayment schedules can impact the

level of financial debt levels.
10. When it comes to purchasing a car, I know how to determine how much of a car I can afford.
11. I understand how to evaluate the cost-benefit analysis of training for the job I would like to per-

form after completing school.
12. I know what a Roth IRA is and how it works from a taxation standpoint.
13. I know how to create a savings plan based on the ability to estimate monthly living expenses.
14. I know how to plan financially for retirement.

188 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



Learning preferences:

I am able to master material when instruction includes:

1. Learning by doing/manipulating objects
2. Listening
3. Discussing with peers
4. Features visual support (e.g., PowerPoint slides)

Objective questions:

1. Is it safer to put your money into one investment or put your money into multiple investments?
2. If you invest $100 in a Roth IRA and earn 10% per year for 3 years, how much would it be worth

at the end of three years.
3. If you use a credit card in January for a total of $300, which payment option will result in the

lowest amount of overall interest paid.
4. Suppose you decide to buy a BMW for $50,000. If you take out an auto loan for 5 years with 5%

interest, how much total will you pay per year?
5. In the future, the cost of things you buy doubles AND your income also doubles. How much will

you be able to buy in the future in comparison to today?
6. Suppose you have $30,000 in student loans. Which payment option would result in the lowest

amount of overall interest paid?

Appendix B

Post-survey questions

Class Code:
Student ID:

Questions:

1. I like to save money more than I like to spend it.
2. I understand how to establish a financial plan.
3. I think financial literacy is important for my future.
4. I have a checking and/or a savings account.
5. I have conversations with my parents regarding personal finance.
6. I understand the process by which my parents/guardians make financial decisions.
7. I know how to determine the appropriate total costs associated with the colleges/universities I

am interested in attending.
8. I understand the process by which loan repayments take place including the impact of interest,

delinquency, and default.
9. I understand the process by which credit card charges and repayment schedules can impact the

level of financial debt levels.
10. When it comes to purchasing a car, I know how to determine how much of a car I can afford.
11. I understand how to evaluate the cost-benefit analysis of training for the job I would like to per-

form after completing school.
12. I know what a Roth IRA is and how it works from a taxation standpoint.
13. I know how to create a savings plan based on the ability to estimate monthly living expenses.
14. I know how to plan financially for retirement.
15. I think it is important to contribute to a retirement plan (ex. Roth IRA, 401k, etc.)

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 189



Learning preferences:

I am able to master material when instruction includes:

1. Learning by doing/manipulating objects
2. Listening
3. Discussing with peers
4. Features visual support (e.g., PowerPoint slides)

Objective questions:

1. Which is less risky: Investing your money into one investment or multiple investments?
2. If you invest $100 in a Roth IRA and earn 5% per year for 3 years, how much would it be worth at

the end of three years.
3. If you use a credit card in January for a total of $500, which payment option will result in the low-

est amount of overall interest paid.
4. Suppose you decide to buy an Audi for $50,000. If you take out an auto loan for 5 years with 5%

interest, how much total will you pay per year?
5. In the future, the cost of things you buy doubles BUT your income remains the same. How much

will you be able to buy in the future in comparison to today?
6. Suppose you have $40,000 in student debt. Which payment option would result in the lowest

amount of overall interest paid?

Appendix C

Teacher survey questions

Class code:
Teacher ID:
Gender:

Questions:

1. How are you teaching the material?
2. How many total contact hours will you spend teaching financial literacy?
3. How often will students receive financial literacy instruction?
4. What best describes the total length of your financial literacy instruction program?
5. What methods did you use to teach the material? (Select all that apply).
6. Did you take a finance-related course in high school or college?
7. What subject do you teach?
8. What suggestions do you have about the program that we can improve next year?

190 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



Appendix D

Variable definitions

Dependent variables:
Subjective Total score for the subjective questions in the survey
Objective Total score for the objective questions in the survey with correct answers
Behavior Total score for the financial behavior questions in the survey
IDK Total score for the objective questions in the survey with “I don’t know”

answers
Diff_Subj The difference between the students’ pre- and post-study scores (post minus

pre) for the subjective questions.
Diff_Obj The difference between the students’ pre- and post-study scores (post minus

pre) for the objective questions with correct answers.
Diff_Behav The difference between the students’ pre- and post-study scores (post minus

pre) for the financial behavior questions.
Diff_IDK The difference between the students’ pre- and post-study scores (post minus

pre) for the objective questions with “I don’t know” answers.
Independent variables:

Post A dummy variable that is equal to 1 if it is post-survey and 0 otherwise.
Post # Year2020 An interaction term of Post and Year2020 dummy variables.
Post # Female An interaction term of Post and Female dummy variables.

Sample year:
Year2020 A dummy variable that is equal to 1 if the survey is conducted in 2020-2021.

Gender:
Female A dummy variable that is equal to 1 if the student is a female and 0

otherwise.
Grade level:

Upperclass A dummy variable that is equal to 1 if the student is in grade 11 or 12 and 0
otherwise.

Favorite subject:
English A dummy variable that is equal to 1 if the student’s favorite is English and 0

otherwise.
Math A dummy variable that is equal to 1 if the student’s favorite is Math and 0

otherwise.
Science A dummy variable that is equal to 1 if the student’s favorite is Science, and 0

otherwise.
Favorite learning style:

LBD A dummy variable that is equal to 1 if the student’s favorite learning style is
learning by doing (LBD) and 0 otherwise.

Listening A dummy variable that is equal to 1 if the student’s favorite learning style is
listening and 0 otherwise.

Discussing A dummy variable that is equal to 1 if the student’s favorite learning style is
discussion and 0 otherwise.

Visual A dummy variable that is equal to 1 if the student’s favorite learning style is
visualization and 0 otherwise.

GPA:
GPA A student’s grade point average (GPA).

School district characteristics:
Log (Population) The log of the population in the school district.
Log (Household_Income) The log of the median household income in the school district.
Poverty The poverty rate in the school district.
Pct_College The percentage of residents who have attained bachelor degree or higher.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 191



Appendix E

Table 2A Differences based on student characteristics: Pre-survey results

Average
response

Gender Modality

Male Female In-
person

Online/
hybrid

Panel A. Financial subjective knowledge questions
2. I understand how to establish a financial plan. 3.014 3.083 2.937*** 3.039 2.956*
3. I think financial literacy is important for my future. 4.379 4.361 4.399 4.365 4.412
6. I understand the process by which my parents/

guardians make financial decisions.
3.360 3.408 3.306** 3.358 3.366

7. I know how to determine the appropriate total costs
associated with the colleges/universities I am inter-
ested in attending.

3.044 3.050 3.037 3.086 2.942***

8. I understand the process by which loan repayments
take place including the impact of interest, delin-
quency and default.

2.758 2.879 2.620*** 2.743 2.792

9. I understand the process by which credit card
charges and repayment schedules can impact the
level of financial debt levels.

3.425 3.430 3.419 3.428 3.418

10. When it comes to purchasing a car, I know how to
determine how much of a car I can afford.

3.288 3.403 3.156*** 3.287 3.290

11. I understand how to evaluate the cost-benefit analy-
sis of training for the job I would like to perform
after completing school.

3.022 3.115 2.917*** 3.015 3.038

12. I know what a Roth IRA is and how it works from
a taxation standpoint.

1.986 2.093 1.865*** 1.951 2.072***

13. I know how to create a savings plan based on the
ability to estimate monthly living expenses.

3.127 3.160 3.090 3.148 3.077

14. I know how to plan financially for retirement. 2.629 2.748 2.494*** 2.621 2.647

Total score for financial subjective knowledge questions 33.928 34.571 33.196*** 33.893 34.010

Panel B. Financial behavior questions
1. I like to save money more than I like to spend it. 3.534 3.587 3.473*** 3.505 3.603**
4. I have a checking and/or a savings account. 4.258 4.238 4.280 4.381 4.003***
5. I have conversations with my parents regarding

personal finance.
3.460 3.444 3.479 3.418 3.560***

15. I think it is important to contribute to a retirement
plan (ex: Roth IRA, 401k, etc.)

4.085 4.079 4.092 4.083 4.089

Total score for financial behavior 14.895 14.835 14.964 14.746 15.254***

Panel C. Objective questions (correct answers)
1. Is it safer to put your money into one investment or

put your money into multiple investments?
0.635 0.699 0.569*** 0.622 0.665*

2. If you invest $100 in a Roth IRA and earn 10% per
year for 3 years, how much would it be worth at the
end of three years.

0.298 0.352 0.238*** 0.291 0.315

3. If you use a credit card in January for a total of
$300, which payment option will result in the low-
est amount of overall interest paid.

0.501 0.533 0.469*** 0.496 0.513

(continued on next page)

192 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



Table 2A (Continued)

Average
response

Gender Modality

Male Female In-
person

Online/
hybrid

4. Suppose you decide to buy a BMW for $50,000. If
you take out an auto loan for 5 years with 5% inter-
est, how much total will you pay per year?

0.439 0.500 0.375*** 0.432 0.456

5. In the future, the cost of things you buy doubles
AND your income also doubles. How much will
you be able to buy in the future in comparison to
today?

0.567 0.602 0.534*** 0.560 0.585

6. Suppose you have $30,000 in student loans. Which
payment option would result in the lowest amount
of overall interest paid?

0.543 0.567 0.521** 0.533 0.567

Total score for objective questions (Correct Answers) 2.982 3.253 2.707*** 2.934 3.100**

Panel D. Objective questions (“I don’t know” Answers)
1. Is it safer to put your money into one investment or

put your money into multiple investments?
0.251 0.191 0.322*** 0.263 0.222**

2. If you invest $100 in a Roth IRA and earn 10% per
year for 3 years, how much would it be worth at the
end of three years.

0.334 0.259 0.424*** 0.345 0.307*

3. If you use a credit card in January for a total of
$300, which payment option will result in the low-
est amount of overall interest paid.

0.337 0.321 0.358* 0.340 0.328

4. Suppose you decide to buy a BMW for $50,000. If
you take out an auto loan for 5 years with 5% inter-
est, how much total will you pay per year?

0.290 0.226 0.366*** 0.295 0.278

5. In the future, the cost of things you buy doubles
AND your income also doubles. How much will
you be able to buy in the future in comparison to
today?

0.176 0.161 0.194** 0.186 0.151**

6. Suppose you have $30,000 in student loans. Which
payment option would result in the lowest amount
of overall interest paid?

0.279 0.252 0.313*** 0.287 0.260

0.000
Total score for objective questions (“I don’t know”

Answers)
1.667 1.411 1.977*** 1.717 1.545**

Note. Table 2A shows the differences of pre-survey student responses on financial knowledge, financial be-
havioral, and objective questions across different gender and GPAs for the whole sample. Gender is identified as
a dummy variable equal to 1 if the student is female and 0 otherwise. Modality is denoted as a dummy variable
equal to one if the sample comes from an online/hybrid modality, the 2020-2021 sample, otherwise, the dummy
variable is zero, indicating an in-person structure based on the sample from 2017 to 2019. We also conduct a t-
test to test the differences in gender (i.e., male vs. female) and modality (in-person vs. online/hybrid). We use
asterisks to show the significance of the t-test result.
***, **, * indicate statistical significance at 0.01, 0.05 and 0.10 level, respectively.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 193



Appendix F

Table 4A t-test Results between pre- and post-survey

Pre Post Diff. t-stat

Panel A. Financial subjective knowledge questions
2. I understand how to establish a financial plan. 3.033 4.029 0.996 27.66***
3. I think financial literacy is important for my future. 4.416 4.660 0.227 9.89***
6. I understand the process by which my parents/guardians make

financial decisions.
3.397 3.888 0.491 14.42***

7. I know how to determine the appropriate total costs associated
with the colleges/universities I am interested in attending.

3.085 3.861 0.776 20.46***

8. I understand the process by which loan repayments take place,
including the impact of interest, delinquency, and default.

2.716 3.864 1.148 28.88***

9. I understand the process by which credit card charges and repay-
ment schedules can impact the level of financial debt levels.

3.406 4.199 0.792 20.53***

10. When it comes to purchasing a car, I know how to determine
how much of a car I can afford.

3.258 4.197 0.940 24.74***

11. I understand how to evaluate the cost-benefit analysis of train-
ing for the job I would like to perform after completing school.

3.010 3.889 0.879 22.70***

12. I know what a Roth IRA is and how it works from a taxation
standpoint.

1.909 3.562 1.653 38.31***

13. I know how to create a savings plan based on the ability to esti-
mate monthly living expenses.

3.087 4.168 1.079 28.21***

14. I know how to plan financially for retirement. 2.581 3.953 1.376 33.45***

Total score for financial subjective knowledge questions 33.839 43.963 10.125 36.38***

Panel B. Financial behavior questions
1. I like to save money more than I like to spend it. 3.540 3.810 0.269 8.09***
4. I have a checking and/or a savings account. 4.290 4.267 0.085 2.41**
5. I have conversations with my parents regarding personal finance. 3.511 3.839 0.328 9.01***
15. I think it is important to contribute to a retirement plan (ex:

Roth IRA, 401k, etc.)
4.064 4.467 0.402 12.02***

Total score for financial behavior 14.944 16.365 1.421 15.11***

Panel C. Objective questions (correct answers)
1. Is it safer to put your money into one investment or put your

money into multiple investments?
0.634 0.691 0.057 3.21***

2. If you invest $100 in a Roth IRA and earn 5% per year for 3
years, how much would it be worth at the end of three years.

0.294 0.481 0.187 10.58***

3. If you use a credit card in January for a total of $500, which
payment option will result in the lowest amount of overall in-
terest paid.

0.489 0.740 0.251 14.90***

4. Suppose you decide to buy an Audi for $50,000. If you take
out an auto loan for 5 years with 5% interest, how much total
will you pay per year?

0.438 0.578 0.140 7.65***

5. In the future, the cost of things you buy doubles BUT your
income remains the same. How much will you be able to buy
in the future in comparison to today?

0.573 0.773 0.200 12.07***

6. Suppose you have $40,000 in student debt. Which payment
option will result in the lowest amount of overall interest paid?

0.542 0.716 0.174 10.05***

(continued on next page)

194 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196



References

Amagir, A., Groot, W., Maassen van den Brink, H., & Wilschut, A. (2018). A review of financial-literacy educa-
tion programs for children and adolescents. Citizenship, Social and Economics Education, 17, 56–80. https://
doi.org/10.1177/2047173417719555

Brown, M., Grigsby, J., Van Der Klaauw, W., Wen, J., & Zafar, B. (2016). Financial education and the debt

behavior of the young. The Review of Financial Studies, 29, 2490–2522. https://doi.org/10.1093/rfs/hhw006
Bucher-Koenen, T., Lusardi, A., Alessie, R., & van Rooij, M. (2017). How financially literate are women? An

overview and new insights. Journal of Consumer Affairs, 51, 255–283. https://doi.org/10.1111/joca.12121
Cameron, M. P., Calderwood, R., Cox, A., Lim, S., & Yamaoka, M. (2014). Factors associated with financial lit-

eracy among high school students in New Zealand. International Review of Economics Education, 16, 12–21.

https://doi.org/10.1016/j.iree.2014.07.006
Cupák, A., Fessler, P., Schneebaum, A., & Silgoner, M. (2018). Decomposing gender gaps in financial literacy:

New international evidence. Economics Letters, 168, 102–106. https://doi.org/10.1016/j.econlet.2018.04.004
De Beckker, K., De Witte, K., & Van Campenhout, G. (2021). The effect of financial education on students’ con-

sumer choices: Evidence from a randomized experiment. Journal of Economic Behavior & Organization,

188, 962–976. https://doi.org/10.1016/j.jebo.2021.06.022
Faulkner, A. (2017). Financial literacy education in the United States: Exploring popular personal finance literature.

Journal of Librarianship and Information Science, 49, 287–298. https://doi.org/10.1177/0961000615616106
Filbeck, G., Zhao, X., & Pettner, J. (2020). Financial literacy: Profiling a successful high school outreach pro-

gram. Financial Services Review, 28, 315–340.
Gerrans, P., & Heaney, R. (2019). The impact of undergraduate personal finance education on individual financial

literacy, attitudes, and intentions. Accounting & Finance, 59, 177–217. https://doi.org/10.1111/acfi.12247
Henager, R., & Cude, B. (2016). Financial literacy and long-and short-term financial behavior in different age

groups. Journal of Financial Counseling and Planning, 27, 3–19. https://doi.org/10.1891/1052-3073.27.1.3

Table 4A (Continued)

Pre Post Diff. t-stat

Total score for objective questions (correct answers) 2.970 3.979 1.009 18.55***
Panel D. Objective questions (“I don’t know” answers)

1. Is it safer to put your money into one investment or put your
money into multiple investments?

0.246 0.045 "0.201 "14.87***

2. If you invest $100 in a Roth IRA and earn 5% per year for 3
years, how much would it be worth at the end of three years.

0.346 0.117 "0.230 "15.03***

3. If you use a credit card in January for a total of $500, which
payment option will result in the lowest amount of overall in-
terest paid.

0.339 0.068 "0.270 "18.50***

4. Suppose you decide to buy an Audi for $50,000. If you take
out an auto loan for 5 years with 5% interest, how much total
will you pay per year?

0.297 0.096 "0.201 "13.15***

5. In the future, the cost of things you buy doubles BUT your
income remains the same. How much will you be able to buy
in the future in comparison to today?

0.172 0.066 "0.106 "8.68***

6. Suppose you have $40,000 in student debt. Which payment
option will result in the lowest amount of overall interest paid?

0.290 0.089 "0.201 "13.96***

Total score for objective questions (“I don’t know” answers) 1.690 0.482 "1.209 "22.43***

Note. Table 4A shows the t-test results of student responses to financial behavior and knowledge questions
before and after the financial literacy educational efforts for the test sample.
***, **, * indicate statistical significance at 0.01, 0.05, and 0.10 level, respectively.

G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196 195



Kabala, M., & Natali, G., Jr., (2012). The missing semester. Harriman House: Petersfield, Hampshire, UK.
Kaiser, T., & Menkhoff, L. (2017). Does financial education impact financial literacy and financial behavior, and

if so, when? The World Bank Economic Review, 31, 611–630. https://doi.org/10.1093/wber/lhx018
Kim, J., Russell, M. B., & Schroeder, A. (2017). Online financial education programs: Theory, research, and rec-

ommendations. Journal of Human Sciences and Extension, 5. https://doi.org/10.54718/LWDE8654
Lusardi, A., & Mitchell, O. (2008). Planning and financial literacy: How do women fare? American Economic

Review, 98, 413–417. https://doi.org/10.1257/aer.98.2.413
Lusardi, A., & Mitchell, O. (2011). Financial literacy and planning: Implications for retirement wellbeing. In

O. S. Mitchell & A. Lusardi (Eds.), Financial literacy: Implications for retirement security and the financial
marketplace (pp. 17–39). Oxford: Oxford University Press.

Lusardi, A., & Mitchell, O. (2014). The economic importance of financial literacy: Theory and evidence. Journal
of Economic Literature, 52, 5–44. https://doi.org/10.1257/jel.52.1.5

Lusardi, A., Mitchell, O., & Curto, V. (2010). Financial literacy among the young. Journal of Consumer Affairs,
44, 358–380. https://doi.org/10.1111/j.1745-6606.2010.01173.x

Morgan, P., Huang, B., & Trinh, L. Q. (2019). The need to promote digital financial literacy for the digital age,
in ADBI and JICA. In K. Yoshida & A. Cardini (Eds.), Realizing Education for All in the Digital Age (pp.
40–47). Asian Development Bank Institute, ADBI Press: Tokyo.

Natali, G. (2020). The missing second semester. Harriman House: Petersfield, Hampshire, UK.
Noetel, M., Griffith, S., Delaney, O., Sanders, T., Parker, P., del Pozo Cruz, B., & Lonsdale, C. (2020). Video

improves learning in higher education: A systematic review. Available at https://psyarxiv.com/kynez
Pangestu, S., & Karnadi, E. (2020). The effects of financial literacy and materialism on the savings decision of genera-

tion Z Indonesians. Cogent Business & Management, 7, 1743618. https://doi.org/10.1080/23311975.2020.1743618
Panos, G., & Wilson, J. (2020). Financial literacy and responsible finance in the FinTech era: Capabilities and chal-

lenges. The European Journal of Finance, 26, 297–301. https://doi.org/10.1080/1351847X.2020.1717569
Preston, A., & Wright, R. (2019). Understanding the gender gap in financial literacy: Evidence from Australia.

Economic Record, 95, 1–29. https://doi.org/10.1111/1475-4932.12472
Riener, G., & Wagner, V. (2017). Shying away from demanding tasks? Experimental evidence on gender differ-

ences in answering multiple-choice questions. Economics of Education Review, 59, 43–62. https://doi.org/
10.1016/j.econedurev.2017.06.005

Saygin, P. O., & Atwater, A. (2021). Gender differences in leaving questions blank on high-stakes standardized
tests. Economics of Education Review, 84, 102162. https://doi.org/10.1016/j.econedurev.2021.102162

Semercioglu, M., & Akcay, A. (2016). High school student financial literacy according to school type. Economic
and Social Development: Book of Proceedings, Proceedings of the 15th International Scientific Conference
on Economic and Social Development – Human Resources Development, Varazdin 9-10, 2016, 88–92.

Stopler, O., & Walter, A. (2017). Financial literacy, financial advice, and financial behavior. Journal of Business
Economics, 87, 581–643. https://link.springer.com/article/10.1007/s11573-017-0853-9

Tang, N., & Baker, A. (2016). Self-esteem, financial knowledge, and financial behavior. Journal of Economic
Psychology, 54, 164–176. https://doi.org/10.1016/j.joep.2016.04.005

van Alten, D. C. D., Phielix, C., Janssen, J., & Kester, L. (2019). Effects of flipping the classroom on learning
outcomes and satisfaction: A meta-analysis. Educational Research Review, 28, 100281. https://doi.org/
10.1016/j.edurev.2019.05.003

Wolla, S. (2017). Evaluating the effectiveness of an online module for increasing financial literacy. Social
Studies Research and Practice, 12, 154–167. https://doi.org/10.1108/SSRP-04-2017-0014

196 G. Filbeck and X. Zhao / Financial Services Review 31 (2023) 169–196