‘The Effect of Financial Education and Mental Resources on Financial Performance’
by
ANNE SEMPLONIUS
MSc Marketing Thesis
‘The Effect of Financial Education and Mental Resources on Financial Performance’
by
ANNE SEMPLONIUS
University of Groningen
Faculty of Economics and Business
MSc Marketing Management
June 2016
Anne Semplonius
Achter de Muur 26
9711 PR Groningen
(06) 55 94 77 66
a.g.semplonius@student.rug.nl
Student number: 2588951
Management Summary
People with critical debts are a growing issue for the Dutch government as well as Gemeente Groningen. Not only are the financial consequences problematic for government, but the material and nonmaterial consequences have a big influence on individual lives. This is the reason why different financial education programs have been executed in different forms and in different countries. In this research, we focus on the extent to which financial education programs (in this case, the program ‘Jouw Schuld =Jouw Schuld’ executed through Gemeente Groningen) fit the way their target group processes information. By doing so, we might conclude something about if, and to what extent, these educations programs are effective in this setting and for this target group. This can help increase the programs’ future effectiveness.
To say something about the effectiveness of knowledge on behavior, from a marketing-perspective, it might be interesting to know how message receivers process the message. Research on the Dual Processing Theory suggests that the way people process information depends on the elaboration of the message receiver. Elaboration depends on people’s motivation and ability to process the information in a message. If motivation and ability in information processing are high, receivers are more likely to be influenced by high quality arguments; whereas receivers with low motivation and ability are more likely to be influence by environmental cues.
Because we now know that the way people process information depends on their ability and motivation towards the content, it is interesting to see if and how debtors differ from non-debtors in their motivation and ability. Research turns out that debtors have lower ability and, often, also lower motivation to process messages about financial issues they receive. Several studies link this lower elaboration to the limited mental resources debtors have. Debtors have to make constant trade-offs in their daily decisions because they are much more bound by their restricted financial opportunities. Keeping this in mind, it might be more effective to communicate environmental cues to debtors rather than high quality arguments.
According to this information, the question of whether financial knowledge improves ability to process messages about financial issues appears to be relevant. If so, high quality arguments are more effective after financial education. The common assumption is that financial education programs lead to better financial performance. There is lot of research that proves financial knowledge indeed leads to better financial performance. However, the causal relationship between financial education and financial performance is not so clear in literature. The question of whether financial education increases financial knowledge and so financial performance seems, therefore, relevant.
optimal choices). This phenomenon is called Debt Account Aversion. Researchers also due these inefficiencies to depleted resources where participants lost control over the situation. To regain control, some participants prefer to close one account over repaying the debt with the highest interest rate.
This summary of relevant literature above leads to the following hypotheses we will investigate in this research:
Hypothesis 1: Environmental cues are more likely to improve debtors’ financial performance than high quality arguments.
Hypothesis 2: High quality arguments are more likely to improve financially educated people’s financial performance than environmental cues.
Hypothesis 3: Financial education positively affects financial performance.
This research used a cross-sectional questionnaire method with a 3 (Message framing: High Argument Quality or Environmental Cues or Control Group) x 2 (Resource depletion: Hard Condition or Easy Condition) x 2 (Financial Education: Take Part or Did Not Take Part on Financial Education) factorial design. Participants were randomly assigned to one of the twelve conditions. Data was gathered by means of a questionnaire under 123 pupils of two secondary schools. Of this total sample, 34 pupils participated in the financial education and 89 pupils did not.
When performing tests in order to answer hypothesis 1, we did not find significant results. However, there are some cues indicating that additional messages, in general, positively affect financial performance, compared to the control group with no additional message.
We did find a small amount of evidence supporting hypothesis 2 because non-debtors seem to make better financial decisions. This conclusion is based on participants’ real financial position (if they were in debt or not) and not on manipulation with resource depletion, which is done to simulate debtors’ mental resources.
Results show clear support for hypothesis 3, so we can conclude that financial education indeed positively affects financial performance. Analysis with demographical information showed that participants with a higher education level perform better financially. Combining both, we can conclude that financial education (and knowledge in general) improves financial behavior.
Preface
This research is conducted to finish my MSc-program Marketing Management at the Rijksuniversiteit Groningen. During my studies, I’ve been fascinated by consumer psychology. People’s behavior is often irrational but very predictable. Also, the way mental resources play into the processing of information was very interesting. I decided to do my research on these topics in different financial situations of participants. Because Gemeente Groningen, as a key-player in financial education programs, was interested in our results, practical relevance of this topic was very high. Practical relevance was very high and one of my motivational drivers during this research.
I was hoping in advance for more participants; however, practical issues made this impossible. The research yielded some significance results, which gave pleasure to me as a researcher. A larger sample size might have increased the amount of significant results. In addition, I am convinced that conducting this study on participants who are real debtors and non-debtors will improve practical relevance.
To end, I would like to thank my internal supervisors, especially Dr. Keizer for his practical feedback and no-nonsense way of working during the whole process. I also wish to thank all respondents, since they did not get any advantages by filling in my questionnaire, but were willing to complete it. Without their cooperation I would not have been able to conduct this study and make any conclusions. Besides those people I would also to thank my friends and family. They did not make a real contribution to my paper, but are one of the reasons I felt full of energy and motivated to finish this research.
I hope you enjoy your reading!
Introduction
In 2003, the NVVK received 34,500 debt rescue applications (NVVK, 2005). Ten years later, in 2013, the amount of applications more than doubled to 89,000 (NVVK, 2014). Also, the average debt amount of these applications increased from 30,100 in 2003 to 37,000 in 2013 (NVVK, 2014). In the Netherlands, between 5 and 7.1% of all households have problematic debts. On average, households consist of 2 people, which means that 1 million people in the Netherlands have problematic debts (CBS, 2014).
Debts have some material and nonmaterial consequences. On one hand, household with debts become socially isolated earlier (Dessart & Kuylen, 1986), 53% of people with problematic debts have health problems (Landelijk Platform Schuldhulpverlening, 2004), and many people with problematic debts have bad psychological problems like stress, depression and insomnia (Dessart & Kuylen, 1986). On the other hand, there are materialistic consequences of debts like implementation costs of debt services (Cebeon (2009) estimated these costs for 2010 at 170 million euro) and costs of eviction (between 5.000 and 7.000 euro per eviction (Geuns, Jungmann & Kruis, 2011)). American research also found out that employees with debts are 2.7 days less productive in a month that employees without debt (Kim & Garman, 2004).
So, financial debts can lead to a wide array of problems which harm individuals personally and society as a whole. Solving financial problems, thus, might lead to less stressful people and lower costs for government. Making right financial decisions can help people to repay debts earlier. Evidence suggests that a lack of adequate financial knowledge and skills leads Americans to make poor financial choices; the implications of those choices extend beyond individual outcomes, affecting the broader society (Mandell and Klein 2009). From this article we can conclude that increasing people's financial knowledge might improve financial decisions. Also, Hilgert, Holgarth & Beverly (2003) found a statistically significant relationship between specific financial knowledge and corresponding behavior. Perry & Morris (2005) admit there has been little empirical study of the relationship between knowledge and behavior. However, their results provide empirical evidence that financial knowledge has a significant effect on financial outcomes, suggesting that devoting resources to consumer financial education may be worthwhile (Perry & Morris 2005). These insights might be very interesting from a government’s perspective, because if information improves people's financial behavior, aforementioned material and nonmaterial consequences can be decreased by providing debtors with information about financial choices. But the questions is to what extent financial education leads to better financial knowledge.
way their target group processes information. By doing so, we might conclude something about if and to what extent these education programs are effective in this setting and for this target group.
Literature review
Dual Information Processing
To improve financial behavior of students on senior secondary vocational education, Gemeente Groningen started with a program titled ‘Jouw Schuld = Jouw Schuld.’ The rationale is to provide student with financial knowledge (information about money, debts and expenditures) and try to improve their financial behavior by increased financial knowledge. So the main goal of ‘Jouw Schuld = Jouw Schuld’ is to give students more financial knowledge in order to improve financial behavior.
But the way in which students process this kind of information might differ in situations. This difference might depend on their knowledge of a specific topic, their involvement and relevance towards the topic, and also distraction from the environment. Participants of the financial education program may handle the provided information in their own way. For evaluating these campaigns, it is interesting to learn which way people process this kind of information. If we know how people in each situation process information, it might be possible to persuade them with the way information is presented. If we can persuade people by the way information is presented, it might be also interesting to use it for communication towards debtors. One important line of research that studies how individual and environmental factors influence information processing focuses on so-called ‘dual information processing models.’ In this research, we will use these model and therefore explain these model further.
Two examples of models based on this dual process theories of information processing are the Elaboration Likelihood Model (ELM) and Heuristic-Systematic Model (HSM). HSM (Chaiken, 1980) assumes that attitude change in response to a persuasive communication is mediated by two modes of information processing (heuristic and systematic) that can occur concurrently (Fennis & Stroebe, 2015). ELM (Petty & Cacioppo & Schumann, 1980) assumes persuasive communications can induce attitude change through two different modes of processing (peripheral and central). Fennis (2016) argues that the ELM and HSM have more similarities than differences. Both models assume that, depending on the recipients’ motivation and ability to engage in message processing, persuasion occurs through either a central/systematic route or a peripheral/heuristic route (Hinyard & Kreuter 2006; Shen, Sheer & Li, 2015). In this paper, the Elaboration Likelihood Model from Petty, Cacioppo & Schumann (1983) will be used because the model is used more often compared to HSM. Carpenter (2015) focused his meta-analysis also on the ELM because of the higher proportion of articles compared to the HSM.
differ in the extent to which individuals engage in message-relevant thought in order to decide on whether to accept message argument” (Fennis & Stroebe, 2015). The mode used depends on the probability that recipients allocate resources to a message (i.e. elaboration likelihood) and is determined by processing motivation and processing ability (O’Keefe, 2002).
Processing motivation can be explained as willingness to receive the message and make an effort to interpret the basic meaning of the message (O’Keefe, 2002). Motivation is high when the outcome of considering a message is important for a person (for example, when purchasing a new television) and low when the outcome is not so important (for example, when purchasing detergent). Processing ability refers to the ability of message receivers to comprehend the intention of the message (O’Keefe, 2002). Ability is high when a receiver is able to handle the technological information in the advertisement. For example, when a receiver with technological knowledge about cars is exposed to a commercial about cars with all its specific features, his or her ability is high. But when another receiver lacks that technological knowledge and is exposed to the same commercial, his or her ability is low. If processing motivation and processing ability are high, elaboration in information processing is high. Likewise, if processing motivation and processing ability are low, elaboration in information processing is low.
In the Elaboration Likelihood Model, two routes of information processing are discussed. Information processing, judgment and decision-making must be viewed as a continuum -- on one end the central route (elaboration is high), and on the other, the peripheral route (elaboration is low). Attitude changes induced via the central route are postulated to be relatively enduring and predictive of behavior (Cialdini, Petty & Cacioppo 1981; Petty & Cacioppo 1980), and attitude changes induced under the peripheral route are postulated to be relatively temporary and un-predictive of behavior (Cialdini, Petty & Cacioppo 1981; Petty & Cacioppo 1980). The route of information processing which is used depends fully on the receiver, and the messenger cannot change the way receivers process information (Fennis, 2016). Which route a participant follows, as explained earlier, depends on the receiver’s motivation and ability. Persuasion can be effective in both routes, although the strength, durability, and resistance of attitudes formed via the two routes may differ (Haugtvedt & Kasmer, 2008). Persuasion via central route is more stable and permanent overtime, whereas persuasion via peripheral route is more temporary (Petty & Krosnick, 1995).
showed more susceptibility to the peripheral cues, such as the credibility of the information source. Two years later, Petty, Cacioppo & Schumann (1983) built further on this outcome. In the study, 160 undergraduates took part in an experiment to measure attitude change based on argument strength. The study had a 2 (involvement: high or low) x 2 (argument quality: strong or weak) x 2 (cue: celebrity or non-celebrity status) factorial design. The results showed that highly involved participants rely more on high quality arguments, whereas minimally involved participants rely more on celebrities as an heuristic cue. Finally, Bhattacherjee & Sanford (2006) studied the ELM Model in an IT context, empirically using a survey study of a document management system acceptance by administrators and staff personnel at L’viv City Hall in Ukraine. Their results indeed confirm that both influence routes (peripheral and central) are moderated by users’ motivation and ability to elaborate or process issue-relevant arguments (Bhattacherjee & Sanford, 2006).
Areni & Lutz (1988) conceptualized the Elaboration Likelihood Model and found out that indeed strong (weak) arguments increase (decrease) attitude toward an object when involvement (ability x motivation to engage in information processing) is high. When there is low involvement, the effect of argument quality on attitude toward an object is less strong. In their conceptualization, Areni & Lutz (1988) decomposed argument quality into two underlying constructs, argument strength and argument
valence. Argument strength is defined as the audience's subjective probability that the attitude object is
associated with some outcome or consequence. Argument valence is the audience's evaluation of that consequence (Ajzen & Fishbein, 2000). This distinction is important because, to alter the argument quality for research purposes, the argument valence or argument strength can be changed.
Summarizing the ELM model, information processing via the central route means people focus on argument strength (high quality arguments: issue-relevant information that says something about the true merits of a choice). Information processing via the peripheral route means people focus on heuristic cues (heuristic cues: issue-irrelevant information that is easier to process and allows the audience to more quickly establish an opinion). If it is known which route of information processing a person uses, they can be persuaded with either high quality arguments (ad content, message quality and / or theoretical background of a message) or heuristic cues (source attractiveness, mood and / or appeal of the ad). In this paper, we investigate if and how debtors differ from non-debtors in the way they process information when making financial choices.
Cognitive Resources
Mani, Mullainahtan, Shafir & Zhao (2013) stated that poor people are worse managers of their finances because poor people must manage sporadic income, juggle expenses, and make difficult tradeoffs. This takes many mental processes, and the human cognitive system has limited capacity (Baddeley, Hitch & Bower, 1974). By constantly making these tradeoffs, cognitive resources get depleted. If cognitive resources are depleted, people will rely more on low-effort decisions (e.g. intuition, heuristics), and so take the peripheral route of information processing (Bolderdijk, 2015). If this relationship really exists, it might be interesting for policy-makers to influence debtors with information based on environmental cues rather than high quality arguments.
The article of Mani et al. (2013) finds evidence for the relationship between people’s financial possibilities and mental resources in two experiments where two groups (rich and poor) were compared. In the first experiment, shoppers at a New Jersey mall (divided into a rich and poor group) were invited into a lab study; in the second experiment, Indian sugarcane farmers before harvest (poor) and after harvest (rich) were compared. In this second experiment, participants were first shown four scenarios describing a financial problem the participants might experience. The financial problems were either relatively high (a hard condition) or relatively low (an easy condition). After answering those four scenarios, participants took part in a test to measure cognitive function. The study showed that poor participants in the hard condition performed less accurately on the cognition test than rich people in the same condition. Mani et all. (2013) attribute this difference to the depleted cognitive resources of poor people.
To conclude, people with limited/depleted mental resources perform less accurately and make more low-effort decisions based on heuristic cues and irrelevant information. People with full mental resources are able to base their decision on high quality arguments and relevant information about a specific topic. Since we know that debtors are more likely to have limited/depleted mental resources, persuasion based on heuristic cues seems to be more efficient than high quality arguments when influencing them in the decision of repaying debts.
Financial Education
As a result, communication has to be more based on environmental cues before financial education, and more based on high quality arguments after financial education.
These so-called financial education interventions (Fernandes, Lynch & Netemeyer, 2014) have as their main goal to increase participants’ knowledge about financial behavior. In America, financial education programs spread rapidly throughout the 1990s as concern grew over the feasibility of long-term financial security and the viability of Social Security, Medicare, and retirement savings vehicles (Vitt et al. 2001). This proliferation of programs has signaled an acknowledgment of education’s role in promoting financial foresight. From Mandell and Klein (2009), we know that the lack of adequate financial knowledge and skills leads Americans to make poor financial choices.
Research about financial education has shown striking results about the effect of this education on financial performance. In this research, we use Fernandes et al.’s (2014) meta-analysis whereby they analyzed 168 papers, covering 201 non-redundant studies. The most striking finding of this meta-analysis was that financial education interventions have statistically significant but miniscule effects: r2 = 0.0011, implying that interventions explained about 0.1% of the variance in downstream financial behaviors studied. One specific article of the effects of financial education was conducted through Mandell & Klein. Mandell & Klein (2009) did a study where they compared a group of 400 high schools students, half of which took a financial education course, and a group which do not take the course. A comparison did not find any positive impact of the financial education course. These findings suggest that financial education have a miniscule, if any effect on financial behavior.
Bernheim, Garrett, and Maki (2001) found that high school students participating in a mandatory financial education courses subsequently saved at higher rates than did nonparticipating students, which can be seen as good financial behavior. Cole, Paulson, and Shastry (2014) found no such correlation in their examination of the same data, and pointed to exogenous factors that contributed to the association observed by Bernheim and colleagues Grinstein-Weis, Guo, Reinertson & Russell (2005). This discrepancy is illustrative of the lack of clarity with which financial education interventions can be assessed.
Based on Mandel and Klein (2009), it can be concluded that financial knowledge indeed leads to better financial behavior, but, at the same time, Fernandes, et al. (2014) indicated the miniscule effect of financial education on financial behavior. Based on this conclusion, the question if financial education interventions indeed improve financial knowledge (and so financial behavior) can be asked.
al. (2010), participants’ ability to make the right financial decisions will increase. So, based on these outcomes, we expect that financial education increases financial knowledge. Better financial knowledge, in this research, leads to higher ability to process information and make the right decision.
However, while Mandell & Klein (2007) found out that motivational variables positively influence financial literacy, the opposite (financial education leads to higher motivation / involvement) is not proven in literature. But, based on Martenson (2005), the level of motivation may be increased by increasing perceived ability and/or opportunity to understand and make decisions (Poiesz and de Bont, 1995). When we apply this study to financial literacy, financial literacy may increase perceived ability and/or opportunity to understand and make decisions, and therefore increase the level of motivation. Based on these outcomes, we suggest that higher information processing ability leads to higher information processing motivation. Overall, we expect that financial education leads to higher information processing ability and this leads to higher information processing motivation.
In the previous subsection, we proposed that financial education increases participants’ motivation and ability to make the right financial choices. These two variables are the main drivers of elaboration in the Elaboration Likelihood Model. According to Petty and Cacioppo (1983), the central route of information processing will be used when elaboration is high. The two drivers (involvement and ability) of the central route of information processing are likely to be increased by financial education. In other words, financial education increases the use of the central route of information processing which helps to form an attitude towards the subject and so predict subsequent behavior. If this is really the case, Gemeente Groningen can educate debtors with financial knowledge. By doing so, these debtors will be more open to high quality arguments in persuasion. When Gemeente Groningen communicates to debtors and gives advice about repaying debts, this communication must be based on high quality arguments.
Financial performance
One of the main topics of this paper is to measure if financial education influences financial performance, and if message framing and mental resources influence the way participants process this education. To measure financial performance, a dependent variable where participants have to make choices which can be judged as good or bad financial behavior is desirable. The article from Amar, Ariely, Ayal, Cryder & Rick (2011) is very interesting in this case. Some examples of financial behavior are cash-flow management, credit management, saving, and investment (Hilgert, et all. 2003). In order to make the best financial choice in managing debts, financial knowledge is needed. As Block-Lieb, Gross & Wiener (2001) stated: “Without adequate financial literacy skills, consumers cannot effectively compare and contrast credit offers; nor can they distinguish between legitimate and predatory lending practices.”
decisions chose to repay a debt with the highest interest rate first. But, as Thaler & Sunstein (2008) investigated, individuals’ decision making is less rational than usually assumed. In that case, participants repaid a debt account with the lowest amount of money in order to minimize the amount of debt accounts. This phenomenon is called debt account aversion (Amar et al. 2011). In such a case, a psychological phenomenon is taking place and decision making is not rational.
Amar et al. (2011) ascribe this effect because that superordinate goal is too big and may overwhelm debtors. This explanation is comparable with Newell & Simon (1972), who attribute Debt Account Aversion to the fact that the superordinate task (repay all debts) is too big, so participants break the overall task into smaller, more manageable tasks. By closing a bank account, a borrower feels that he regains control over his situation. Loewenstein & O’Donoghue (2006) argue that debtors with multiple loans lose control over the situation and find it difficult to fully attend –or even recognize- the broader consequences of their financial behavior. The assumptions of these three studies are in line with the conclusion of the Mani et al. (2013) article explained earlier: poor people perform less accurately on a task compared to rich people due to depleted cognitive resources.
Based on previous research, we expect that debtors rely more on environmental cues in order to make these financial decisions because they have limited cognitive resources and therefore rely on heuristics. Debtors lack the motivation and ability to rely on strong arguments during decision making. In our research we had no real debtors, so we first let participants make financial trade-offs in order to deplete their cognitive resources. Participants without debts or, in this case, participants with fully available cognitive resources, will rely less on heuristics and likely more on the central route of processing. In making the right financial choices, persuasion can be based on this knowledge.
We expect that financially educated participants rely more on high quality arguments, and this is likely to be the case for both groups (participants with full cognitive resources and participants with depleted cognitive resources). Due to financial education, participants’ ability and motivation to make the right financial choices will increase. Increase in ability and motivation increases participants’ elaboration of message content. Therefore, participants are more susceptible to high quality arguments instead of environmental cues in making financial decisions.
Conceptualization
Hypothesis 1: Environmental cues are more likely to improve debtors’ financial performance rather than high quality arguments.
As a potential moderator, ‘financial education’ is included in the model. The effect of financial education on the financial behavior of two groups (one group of participants who joined a financial education group, another group of participants who did not join a financial education program) can be compared. The two drivers (involvement and ability) of the central route of information processing are likely to increase due to financial education. In other words, financial education increases the use of the central route of information processing which helps to form an attitude towards a subject and so predict subsequent behavior. If financial education indeed positively influences the relationship between high quality arguments and financially optimal behavior, financially educated people are more susceptible to high quality arguments.
Hypothesis 2: High quality arguments are more likely to improve financially educated people’s financial performance rather than environmental cues.
A direct relationship between financial education and financial behavior can also occur. In this case, financial education positively influences financial behavior, regardless of the kind of information processing being used. However, previous research did not reach an unambiguous conclusion (especially on the long-term effects) as to whether financial education improves financial performance; it is clear that financial knowledge improves financial behavior. Fernandes et al. (2014) found a decay in the effect of a financial education program. Because the effect measurement is relatively short (two weeks) after the financial education, a positive effect is expected.
Hypothesis 3: Financial education positively affects financial performance
The variables and hypotheses are presented in the conceptual model below. In this research, we investigate the influence of argument quality (independent variable) on people with or without mental resources (independent variable), in the choice of repaying debts (dependent variable). We also want to find out if financial education (moderator) moderates this relationship and/or if financial education has a direct effect on financial behavior. In the conceptual model below the relationships are displayed.
Methodology Operationalization
This research focuses on the direct relationship between information processing (environmental cues, high quality arguments, and the control group) and person’s financial behavior. One part of the sample got easy questions, so they retained full mental resources. The other part was asked difficult questions to deplete cognitive resources. The independent variable consisted of three kinds of information processing, and the moderator was defined by two groups (one of which had taken the financial education program ‘Jouw Schuld = Jouw schuld’, and the other of which had not). The dependent variable measured financial performance in the different conditions.
The different levels of every variable are depicted in table 1 below.
Table 1: Research design
To be practically relevant, it was desirable to measure the effects of the independent variable on financial behavior for people in debt / with multiple loans. Because it was not possible to base our sample on people’s real financial situation, an alternative manner had to be found. By depleting participants’ cognitive resources, we tried to awaken the same feelings people have with multiple debts. In line with Mani et al. (2013), we confronted participants with situations in which they had to make financial choices and decisions to deplete their resources. Participants in the ‘limited cognitive resources’ condition received four difficult financially-related questions, where they had to make difficult tradeoffs. Participants in the ‘full cognitive resources’ condition received the same set of questions, but the presented solutions were much clearer and easy to answer. Examples of these questions can be found in the materials section.
(respondent makes a decision to repay the smallest debt 3 or 4 times), Stuck in the Middle (respondent makes a decision which is neither financial optimal, nor debt account averse 3 or 4 times), and No Category (respondent’s decisions are not covered by the first three terms). In the accumulated form, the previous terms are valued (Debt Account Aversion = 0, Stuck In the Middle = 1, and Financially Optimal Player =2) for every single decision a respondent makes. The scores of respondents’ every single decision are added, which gives a total score between 0 (Bad Financial Performance) and 8 (Excellent Financial Behavior). Financial Performance as the nominal dependent variable will be called ‘Financial Categorical Performance.’ Financial Performance as the accumulated dependent variable will be called ‘Financial Accumulated Performance.’
The accumulated dependent variable is included because by only looking at categories, information (due to categorization) might be lost. For example, in the nominal situation, a respondent who has two ‘financially optimal’ decisions and two ‘stuck in the middle’ decisions and another respondent who has two ‘stuck in the middle’ decisions and two ‘debt account averse’ decisions will be classified in the same category (No Category). However, the earlier respondent might be a better financial player. In the case of the accumulated dependent variable, the first respondent will earn 6 points, whereas the latter will earn only 2 points. This might better indicate their differences in financial behavior.
Participants
The sample consisted of scholars of two secondary schools (ROC Menso Alting and Alpha College) where the financial education program ‘Jouw Schuld = Jouw schuld’ was conducted. At both schools, a few classes who participated in the program were compared with similar classes who did not participate. The sample was separated by a question in the questionnaire which asked if subjects participated in the financial education program or not. The questionnaire was presented in Dutch because the education program was presented in Dutch and not all participants were able to understand English questions. The full questionnaire is presented in appendix B.
Design
This research used a cross-sectional questionnaire method with a 3 (Message framing: High Argument Quality or Environmental Cues or Control Group) x 2 (Resource depletion: Hard Condition or Easy Condition) x 2 (Financial Education: Take Part or Did not Take Part in Financial Education) factorial design. The 123 participants were randomly assigned to the twelve conditions. Distribution is visualized in table 2 below.
Table 2: Distribution participants over conditions
Procedure
The first part of the questionnaire consisted of some general questions to form an image about the participants. The general part consisted of demographic questions (about age, gender, education, and income level), attitudinal questions (opinions towards money / spending and loans), and also questions about whether they participate in ‘Jouw Schuld = Jouw Schuld. After that, the second part started. The aim of the second part was to deplete (or not deplete) participants’ cognitive resources. Depleting resources was done by questions whereby participants had to make trade-offs about financial issues. These questions were based on Mani et al. (2013), who did comparable research. A manipulation check, whereby participants had to indicate if the previous decisions were either easy or hard, determined if it could be assumed resources were indeed limited or not. The last section consisted of four scenarios in which participants had to make financial choices; by doing so, financial performance can be judged.
Materials and analysis plan
The whole questionnaire is in Appendix B, but only the parts which are used in the results section are discussed in this paragraph. The first part of central questions aimed to gather demographical information. Participants’ age gave interval data; gender gave nominal data; type of education also gave nominal data; level of education gave nominal data; and year of education gave ratio data. We performed an analysis of all demographic information on financial performance. Only education level turned out to be significant (p=0,002). Analyzing the effects of education level (which level of education do you follow? Four answers: level 1, 2, 3, 4) on financial performance might be interesting.
limited mental resources and are therefore worse managers of financial decisions. Through this question we received nominal data from participants who are in debt and those who are not (Do you have debts in
your own situation? Yes/No). We can compare both these groups on their financial performance. Because
we did not know beforehand if and how many participants face real debts, we included four financial questions to deplete participants’ resources. Participants were randomly assigned to either the hard (leading to depleted mental resources) or easy condition (consolidating mental resources). The form of the questions and answer options in both conditions (hard and easy condition) are the same; however, the answers in the hard condition are more difficult to calculate and more comparable. This way of depleting resources is based on the article of Mani (2013), Schmeichel (2007) and Otgaar (2012). One example of those questions was ‘If you want to borrow €1.000,-, which debt account is most beneficial?’ The possible answer options are in table 3 below.
Table 3: Example question for mental resource depletion
In the next section participants received the information based on their specific condition followed by four financial choices. In the operationalization part of this paper, the way the dependent variable was classified as either categorical or accumulated data is explained. These questions (In table 4 below, we gave an example of one of the four questions, which are further explained in appendix A) were preceded with a general explanation of the question and a manipulation of message framing.
The general explanation is: ‘Imagine, in the next four questions, that you have multiple debts. In
all three questions, the debt amount and interest rate differ. In all four questions, you get some financial bonus, which you can use to repay your debts. Which of the debts do you decide to repay first?’
Participants in the ‘High Quality Arguments’ condition got the additional message ‘first repay the debt
with the highest interest rate’ in their explanation. Participants in the ‘environmental cues’ condition got
the Dutch saying ‘Money does not grow on a tree’ added to their explanation, and the participants in the control group did not receive an additional message. Additional messages were based on Petty et al. (1983), who framed messages with either environmental cues or high quality arguments in an advertisement of razors.
Of all four choices, one can be assessed as the best decision (indicating a financially optimal player) and the worst decision (indicating a debt account averse player). The other two options are classified as ‘Stuck in the Middle’ options.
Results
Further in this section, results per hypothesis will be discussed. But first, Table 5 shows how both measurements of the dependent variables relate to each other.
Table 5: Relation between dependent variable in categorical and accumulated form
As expected, the financially optimal player scores highest on financial performance (7,2222) whereas the Debt Account Averse Player scores lowest (0,8235). The scores of ‘Stuck in the Middle’ and ‘No Category’ players are comparable, but the scores of ‘No Category’ players have a higher standard deviation. This higher standard deviation seems logical because players who made three different choices (Debt Account Averse option, Stuck in the Middle option, and the Financially Optimal option) are in this category.
Manipulation check:
Table 6: Outcome One-way-ANOVA; Depletion condition on Financial accumulated Performance
In spite of the fact that the results showed low significance for the effect of resource depletion, we proceeded with manipulation of resource depletion. Because on the one hand, the small sample size might be the reason for the low significance. On the other hand, previous research proves for resource depletion by asking difficult (or easy) questions. Due to these two arguments, financial knowledge questions are still used as manipulation to deplete participant’s resources.
From here, the analyses for the three hypotheses will be elaborated per measurement type of the dependent variable. After that, the analyses for possible side-effects will be mentioned.
Hypothesis 1: Environmental cues are more likely to improve debtors’ financial performance than
high quality arguments.
I. Financial Categorical Performance
Table 7: Chi-Squared Goodness-Of-Fit-Test; Message framing, combined with mental resources on financial categorical performance
II. Financial Accumulated Performance
In order to analyze the main and combined effects of message framing and mental resources on financial performance, we performed a 3 (message framing: high argument quality vs. environmental cues vs. control) x 2 (resource depletion: hard vs. easy) ANOVA on financial accumulated performance. Both main effects were not proved. The main effect of message framing on financial performance
F(2,117)=0,85 p=0,43 and the main effect of resource condition F(1,117)=0,55, p=0,46 are not
significant. Also, the interaction effect between both is not significant F(2,117)=1,02, p=0,36, see table 8 below. So we cannot conclude that environmental cues (Mean scenario ‘Hard condition and Environmental cues’=4,82) positively influences financial accumulated performance of participants with limited mental resources. We also cannot conclude that high quality arguments (Mean scenario ‘Easy condition and High argument quality’=5,60) positively influences financial accumulated performance of participants with full mental resources.
In table 8 below, the results of the ANOVA are presented. Although participants framed with high quality arguments performed better, on average; and participants in the easy conditions performed better, on average, than participants in the hard condition, there is no pattern in the conditions similar to the hypothesis. Additional messages (either high quality arguments or environmental cues) seem to have had a positive effect on financial behavior, compared to the control group. But we cannot make any conclusions out of it because the test is not significant.
Table 8: Outcome 3 (message framing: high argument quality vs. environmental cues vs. control) x 2 (Resource depletion: hard vs. easy)
If our expectation was correct, the financial performance of people in the resource depletion-condition should be improved by framing messages with environmental cues (scenario 2 & 8; hard condition and environmental cues), rather than high argument quality (scenario 1 & 7; hard condition and high quality arguments). In order to analyze whether or not the financial accumulated performance of participants in scenario 2&8 was better than the financial accumulated performance of participants in scenario 1&7, an independent sample t-test with scenario and financial accumulated performance was performed. The independent sample t-test was not significant t(48) =0,010, p=0,921. The accumulated financial performance of scenario 1 & 7 (M=5,1538, SD=2,36123) did not differ from the accumulated financial performance of scenario 2 & 8 (M=4,8182, SD=2,44241). Outcomes are presented in table 9 below.
Table 9: Outcome Independent samples t-test; Scenario 1&7 vs. 2&8 on financial accumulated performance
As stated earlier, manipulation by resource depletion was done because it was unknown beforehand how many participants were in debt, and asking those questions simulated the same effect as people in real debt (depleted mental resources); however, 17 participants (13,8% of total sample) admitted they were in real debt. Because this was in line with the real meaning of this research, we analyzed if participants who were in debt performed worse financially compared with people who were not in debt. We performed a Chi-Squared Goodness-Of-Fit-Test with debt and financial categorical performance. The Chi-Squared Goodness-Of-Fit-Test was not significant, Chi-square (3) = 5,59, p=0,11. Based on these outcomes, we cannot conclude non-debtors perform financially better than debtors. However, results suggest that non-debtors have a higher percentage of financial optimal players (40,6%) than debtors (11,8). A larger sample may increase significance of this conclusion. Outcomes are presented in table 10 below.
Table 10: Chi-Squared Goodness-Of-Fit-Test; Debt situation on financial categorical performance
participants was skewed and results are marginally significant, we are cautious in making hard conclusions about the results.
Table 11: Independent samples T-test; Debt situation on financial accumulated performance
Based on the manipulation check, we cannot support hypothesis 1. Even though the easy condition is marginally significant, the results are in the opposite direction. Sample size was relatively small and participants were skewed, divided over scenarios. When financial performance of debtors was compared with non-debtors, we found some evidence suggesting that non-debtors perform better financially compared to debtors.
Hypothesis 2: High quality arguments are more likely to improve financially educated people’s financial
performance than environmental cues.
I. Financial Categorical Performance
The combined effect of financial education (nominal data) and message framing (nominal data) on financial categorical performance (nominal data) was measured in hypothesis 2. A Chi-Squared Goodness-Of-Fit-Test analysis with message framing (high argument quality, heuristic cues or control message) on financial behavior (Debt Account Aversion, Financial Optimal Player, Stuck In the Middle, and No Category) was performed, where the groups had been split by participation in financial education. The Chi-Squared Goodness-Of-Fit-Test was not significant (6) =3,35, p= 0,76 for participants who had taken financial education. There was also no significant effect found for participants who did not take financial education. So we cannot conclude any effect of message framing, combined with financial education, on financial behavior. Results are in table 12 below.
II. Financial Accumulated Performance
In order to analyze the main and combined effects of message framing and financial education on financial performance, we performed a 3 (message framing: high argument quality vs. environmental cues vs. control) x 2 (financial education: yes vs. no) ANOVA on financial accumulated performance. The main effect of message framing on financial performance F(2,117)=0,21, p=0,81 was not significant, but the main effect of financial education on financial performance was significant F(1,117)= 4,42, p=0,04; the interaction effect between both was not significant F(2,117)=0,37, p=0,69. So we cannot conclude that high quality arguments (Mean scenario ‘Financial education and High quality arguments’ =6,00) positively influences financial accumulated performance of financially educated participants, and we can also not conclude that environmental cues (Mean scenario ‘No financial Education and Environmental cues=5,11) positively influence financial accumulated performance of participants who did not take part in the financial education.
Although the effect of message framing and the interaction effect are not significant, the effect of financial education is. So we can conclude that education has a positive effect on financial performance. In table 13 below, we can see that the results are in the expected direction. Education combined with high quality arguments leads to better financial performance versus environmental cues combined with no financial education. For people who did not take part in the financial education, environmental cues were the most effective message framing. Again, as in the ANOVA-test in hypothesis 1, a positive effect of additional messages (either high quality arguments or environmental cues) on financial performance, compared to the control group, is visible. It might be possible that when the sample size is bigger, both indications become significant.
Table 13: Outcome 3 (message framing: high argument quality vs. environmental cues vs. control) x 2 (Education take part vs. did not take part)
ANOVA on financial accumulated performance
heuristic cues (scenario 2 & 5; environmental cues and financial education). In order to analyze whether or not the financial accumulated performance of participants in scenario 1&4 was better than the financial accumulated performance of participants in scenario 2&5, an independent sample t-test with scenario and financial accumulated performance was performed. The independent sample t-test was not significant t(21) =0,003, p=0,95. The accumulated financial performance of scenario 1 & 4 (M=6,00, SD=1,61) did not differ from the accumulated financial performance of scenario 2 & 5 (M=6,50, SD=1,84). Outcomes are presented in table 14 below.
Table 14: Outcome Independent samples t-test; Scenario 1&4 vs. 2&5 financial accumulated performance
We also expected that participants who were not financially educated would score higher when relying on heuristic cues (scenario 8 & 11; Environmental cues and financial education) rather than high argument quality (scenario 7 & 10). In order to analyze whether or not the financial accumulated performance of participants in scenario 8&11 was better than the financial accumulated performance of participants in scenario 7&10, an independent sample t-test with scenario and financial accumulated performance was performed. The independent samples t-test was not significant t(61)=0,49, p=0,49. The accumulated financial performance of scenario 7 & 10 (M=5,07, SD=2,18) did not differ from the accumulated financial performance of scenario 8 & 11 (M=4,87, SD=2,51). Outcomes are presented in table 15 below.
Table 15: Outcome Independent samples t-test; Scenario 7&10 vs.8&11 on financial accumulated performance
We cannot support hypothesis 2 based on all the performed tests above. However, some results are in line towards the hypothesis. A larger sample size might be the solution in order to get significant results supporting this hypothesis.
Hypothesis 3: Financial education positively affects financial performance. I. Financial Categorical Performance
There are more financially optimal players in the group of participants who took part in the financial education program (61,8%) compared to participants who did not (27%). Outcomes are presented in table 16 below.
Table 16: Chi-Squared Goodness-Of-Fit-Test; Financial education on financial categorical performance
II. Financial Accumulated Performance
An independent sample T-test was conducted to analyze if financial education influences financial accumulated performance. This analysis was significant F(1,122)=4,50, p=0,04. Based on this analysis, we can also conclude financial education influences financial performance. Financially educated participants’ scores were higher on average (5,82) compared with people who were not financially educated (4,85). Outcomes are presented in table 17 below.
Table 17: Independent samples T-test; Financial education on financial accumulated performance
Based on both these tests, hypothesis 3 is supported.
Side effects:
Table 18: Means; Education level on financial performance
To test if the difference in financial accumulated score is only caused by education level and is not influenced by an interaction effect with financial education, a 3x2 ANOVA test was conducted (outcomes in table 16 below). The main effect of financial education on financial performance F(1,118)=2,27,
p=0,135 was not significant, but the main effect of education level on financial performance was
significant F(2,118)= 4,24, p=0,02. This is in line with the significant outcomes of the ANOVA before. However, the interaction effect between both is not significant F(1,118)=1,75, p=0,19. So we can conclude that education level explains a big part of the variance in assessing the effect of financial education on financial accumulated performance. Results suggest education level, rather than financial education, as a more important predictor of financial accumulated performance. Therefore, we can interpret the outcomes of table 19.
Table 19: Outcome 3 (Education Level: 2 vs. 3 vs. 4) x 2 (Education: take part vs. did not take part) ANOVA on financial accumulated
performance
Discussion Main findings
We focused on to what extent financial education programs fit the way their target group processes information. By doing so, we might conclude something about if and to what extent these educations programs are effective in this setting and for this target group. We discuss our results for every hypothesis.
Hypothesis 1: Environmental cues are more likely to improve debtors’ financial performance than high quality arguments.
because the sample size is so small and the categorization is skewed, we cannot make any hard conclusion about the results. From the results, we see a difference in effect between two kinds of message framing and the control group. Both kinds of message framing seem to improve financial performance, but because the results are not significant we cannot make conclusions about them.
When financial performance of debtors was compared with non-debtors, we found evidence suggesting that non-debtors perform financially better compared to debtors.
Hypothesis 2: High quality arguments are more likely to improve financially educated people’s financial performance than environmental cues.
We did not find significant results for an interaction effect between message framing and financial education on financial performance. Some results were in the expected direction (high argument quality and education had the highest score on financial performance), but as said, the results were not significant. Again, in this test we also found some evidence of a positive effect from both kinds of message framing compared to the control group. Both kinds of message framing led to higher financial performance, compared to the control group. A larger sample size might increase the significance of the results, which is recommended for future research.
Hypothesis 3: Financial education positively affects financial performance.
This hypothesis is supported by both tests (financial categorical performance and financial accumulated performance), which showed significant results. Participants who took part in ‘Jouw Schuld = Jouw Schuld’ scored on average 5,82 on financial accumulated performance, whereas participants who did not took part scored on average 4,85. There were also more financially optimal players among the educated participants (61,8%) compared with non-educated participants (27%).
Side-effect
As a side-effect, we found that education level positively influences financial performance. The higher participants’ education level, the better the financial accumulated performance. Participants of education level 4 scored higher (5,78 on financial accumulated performance) compared to participants of education level 2 (3,82 on financial accumulated performance). Participants of education level 3 scored 5,40 on financial accumulated performance.
Theoretical and Practical Implications
financial performance by non-debtors compared to debtors. This is in line with studies like Mani et al. (2013), who have evidence that non-debtors have more mental resources and therefore make better financial decision.
Unfortunately, we did not find significant effects of message framing on debtors’ financial performance which we could add to existing literature. However, outcomes indicate that additional messages (either high quality arguments or environmental cues) improve financial performance, compared to the control category.
Outcomes can be used by governmental parties. If financial education and knowledge in general improve financial performance, it might worthwhile to invest in this knowledge. From this research, financial education programs seem to be efficient. Commercial parties can also use these outcomes for their accounts receivable policy. They can include additional messages to their bills or urgent requests to stimulate their debtors to repay earlier and more efficiently. Additional messages, rather than no additional message, are likely to improve debtors’ financial performance. Commercial parties will receive their money earlier by doing so.
Limitations
The sample size (N=123) is the biggest limitation of this research. As part of the ‘Jouw Schuld = Jouw Schuld’ project, different comprehensive schools promised to assist and facilitate data gathering from their students. However, communication was difficult and relatively few participants were made available. The sample size was big enough to run analyses, but for more reliable conclusions, a bigger sample size is needed. Due to the small sample size, it was hard to find significant results. On the other hand, the significance results we found are extra meaningful.
The different message frames (high quality argument, environmental cues, control group) were based on Petty, et al. (1983), who turned out to very effective. In our research, however, the effects of message framing were not significant. The reason might be that engagement towards our research was low. Students had to fill in our questionnaire during regular lessons and they did not get any incentive for filling it in. Engagement could maybe improve by filling in the questionnaire during a separate setting, or by giving an incentive for participations. However, an incentive might also influence involvement as one of the drivers of elaboration, as Petty, et al. (1983) investigated.
That might be the reason why the manipulation check did not work out. However, because there was such strong evidence from previous research, we kept the results in our research. A bigger sample size might also have a positive effect on the significance of the manipulation of the resource depletion. In further research, it would be very interesting to compare real debtors with non-debtors, because that was the real idea behind this research. Even though the sample was very skewed (17 debtors and 106 non-debtors), results turned out to be significant. This suggests that, indeed, non-debtors perform better than debtors, which might be due to more cognitive resources available.
The last limitation is about the scenarios in which participants had to choose which debt they want to repay first. In these scenarios, participants had to imagine they were in that particular situation. As stated before, participants were not really engaged to the questionnaire and perhaps answered carelessly. If participants were really in this scenario, engagement would have been higher, and their choices more reliable. Because, in real life, people will be punished or rewarded directly through their choices, this pressure will also influence people’s mental resources, which are important in these choices.
Conclusions
References:
Amar, M., Ariely, D., Ayal, S., Cryder, C.E., & Rick, S.I. (2011). Winning the Battle But Losing the War: The Psychology of Debt Management. Journal of Marketing Research, 98, 38-50.
Areni, C.S., & Lutz, R.J. (1988). The Role of Argument Quality in the Elaboration Likelihood Model.
Advances In Consumer Research, 15(1), 197-203
Ajzen, I., & Fishbein, M. (2000). Attitudes and the attitude-behavior relation: Reasoned and Automatic Processes. European review of Social Psychology, 11(1), 1-33.
Baddeley, A.D., Hitch, G.J., & Bower, G.H. (1974). The Psychology of Learning and Motivation. The
psychology of learning and motivation, 8, 47-90.
Bassili, J.N., & Fletcher, J.F. (1991). Response-Time Measurement in Survey Research a Method for CATI and a New Look at Nonattitudes. Public Opinion Quarterly, 55(3), 331-346.
Bernheim, B.D., Garrett, D.M., & Maki, D.M. (2001). Education and Saving: The Long-Term Effects of High School Financial Curriculum Mandates. Journal of Public Economics, 80(3), 435-465.
Bhattacherjee, A., & Sanford, C. (2006). Influence Processes for Information Technology Acceptance: An Elaboration Likelihood Model. MIS quarterly, 805-825.
Block-Lieb, S., Gross, K., & Wiener, R.L. (2001). Lessons From the Trenches: Debtor Education in Theory and Practice. Fordham J. Corp. & Fin. L., 7, 503.
Bolderdijk, J.W. (2015). Lecture 3: “Limited Mental Resources”, Rijksuniversiteit Groningen, 23 november 2015
Cacioppo, J.T., & Petty, R.E. (1984). The Elaboration Likelihood Model of Persuasion. Advances in
Consumer Research, 11(1), 673-675.
Carpenter, C.J. (2015). A Meta-Analysis of the ELM's Argument Quality × Processing Type Predictions.
Centraal Bureau voor de Statistiek.(2014). Statline: Bevolking, huishoudens en bevolkingsontwikkeling;
vanaf 1899. Den Haag: Centraal Bureau voor de Statistiek.
Chaiken, S. (1980). Heuristic versus Systematic Information Processing and the Use of Source Versus Message Cues in Persuasion. Journal of Personality and Social Psychology, 39, 752–766.
Chaiken, S. & Maheswaran, D. (1994). Heuristic Processing Can Bias Systematic Processing: Effects of Source Credibility, Argument Ambiguity, and Task Importance on Attitude Judgment. Journal of
Personality and Social Psychology, 66, 460–473
Cialdini, R.B. Petty, R.E., & Cacioppo, J.T. (1981). ‘Attitude and Attitude Change’. Annual Review of
Psychology. 32, 357-404.
Cole, S., Paulson, A., & Shastry, G.K. (2014). Smart Money? The Effect of Education on Financial Outcomes. Review of Financial Studies, 27(7), 2022-2051.
Dessart, W.C.A.M., & Kuylen, A.A.A. (1986). The Nature, Extent, Causes, and Consequences of Problematic Debt Situations. Journal of Consumer Policy, 9(3), 311–334.
Fennis, B.M., & Stroebe, W. (2015). The psychology of Advertising. Psychology Press.
Fennis, B.M. (2016). Lecture 5: “Marketing communications”, Rijksuniversiteit Groningen, 7 march 2016
Fernandes, D., Lynch Jr, J.G., & Netemeyer, R.G. (2014). Financial literacy, Financial Education, and Downstream Financial Behaviors. Management Science, 60(8), 1861-1883.
Geuns, R. van, Jungmann, N., Kruis, G., Calkoen, P., & Anderson, M. (2011). Schuldhulpverlening loont!
Een onderzoek naar de kosten en baten van schuldhulpverlening. Amsterdam: Regioplan.
Grinstein-Weiss, M., Guo, S., Reinertson, V., & Russell, B. (2015). Financial Education and Savings Outcomes for Low-Income IDA Participants: Does Age Make a Difference? Journal Of Consumer
Haugtvedt, C.., & Kasmer, J.A. (2008). Attitude Change and Persuasion. In C. P. Haugtvedt, P. Herr, F. R. Kardes (Eds.), Handbook of consumer psychology (pp. 419-436). New York, NY: Lawrence Erlbaum Associates.
Hilgert, M.A., Hogarth, J.M., & Beverly, S.G. (2003). Household Financial Management: The Connection Between Knowledge and Behavior. Federal Reserve Bulletin, 89, 309.
Hinyard, L.J., & Kreuter, M.W. (2006). Using Narrative Communication as a Tool for Health Behavior Change: a Conceptual, Theoretical, and Empirical Overview. Health Education & Behavior.
Kim, J., & Garman, E.T. (2004). Financial Stress, Pay Satisfaction and Workplace Performance.
Compensation& Benefits Review, 36(1), 69–76.
Loewenstein, G., & O’Donoghue, T. (2006). ‘We Can Do This the Easy Way or the Hard Way’: Negative Emotions, Self-Regulation, and the Law. University of Chicago Law Review, 73, 183–206.
Mandell, L., & Klein, L.S. (2007). Motivation and Financial Literacy. Financial Services Review, 16(2), 105-116.
Mandell, L., & Klein, L.S. (2009). The Impact of Financial Literacy Education on Subsequent Financial Behavior. Journal of Financial Counseling and Planning, 20(1).
Mani, A., Mullainathan, S., Shafir, E., & Zhao, J. (2013). Poverty Impedes Cognitive Function. science,
341(6149), 976-980.
Martenson, R. (2005). Success in Complex Decision Contexts: The Impact of Consumer Knowledge, Involvement, and Risk Willingness on Return on Investments in Mutual Funds and Stocks. The
International Review of Retail, Distribution and Consumer Research, 15(4), 449-469.
Newell, A. & Simon, H. (1972), Human Problem Solving. Englewood Cliffs, NJ: Prentice Hall.
NVVK. (2005). Heldere Lijnen naar Professionele Hulp. Jaarverslag 2004. Den Haag: NVVK.
O'Keefe, D.J. (2002). Persuasion Theory & Research (2nd ed.). Thousand Oaks, CA: Sage Publications.
Otgaar, H., Alberts, H., and Cuppens, L (2012). ‘How Cognitive Resources Alter our Perception of the past: Ego Depletion Enhances the Susceptibility to Suggestion.’ Applied Cognitive Psychology 26.1: 159-163.
Perry, V. G., & Morris, M.D. (2005). Who is in Control? The Role of Self‐perception, Knowledge, and Income in Explaining Consumer Financial Behavior. Journal of Consumer Affairs, 39(2), 299-313.
Petty, R.E., & Cacioppo, J.T. (1980). Effects of Issue Involvement on Attitudes in an Advertising Context.
Proceedings of the Division, 23, 75-79.
Petty, R.E., Cacioppo, J.T., & Goldman, R. (1981). Personal Involvement as a Determinant of Argument-Based Persuasion. Journal of Personality and Social Psychology, 41(5), 847.
Petty, R.E., Cacioppo, J.T., & Schumann, D. (1983). Central and Peripheral Routes to Advertising Effectiveness: The moderating Role of Involvement. Journal of Consumer Research, 135-146.
Petty, R.E., & Krosnick, J.A. (Eds). (1995). Attitude strength: Antecedents and consequences. Hillsdale,
NJ: Lawrence Erlbaum Associates.
Poiesz, T.B.C. & Bont, de, C.J.P.M. (1995) Do we Need Involvement to Understand Consumer Behavior?, Advances in Consumer Research, 22, pp. 448–452.
Schors, van der, A (2016). Nibud presentation during kick-off Jouw Schuld = Jouw Schuld, Martiniplaza
Groningen, 13 march 2016
Schmeichel, B.J. (2007). Attention Control, Memory Updating, and Emotion Regulation Temporarily Reduce the Capacity for Executive Control. Journal of Experimental Psychology. General, 136, 241–255.
