Retirement preparedness is a major concern in the United States, with over 40% of people saying they lack sufficient funds for retirement (Bogan, 2025). Many attempts have been made to understand the gap in retirement preparedness between women and men and to provide solutions (Leopold et al. 2015; Richardson 2014). One solution that can help people adequately prepare for retirement is to seek help from financial advisors. Understanding gender differences in the determinants of financial help-seeking is essential for addressing these long-standing inequities. Women face additional unique challenges that can result in further under-preparedness for retirement (Gornick & Sierminska 2021; Hasler & Lusardi 2019). For example, women live longer on average and experience more career breaks and income loss due to caregiving (Auerbach et al. 2017; Grigoryeva 2017). They also tend to have lower average financial knowledge and greater risk aversion, which further depresses wealth accumulation and widens retirement gaps (Hasler & Lusardi 2017; Bannier & Neubert 2016; Fisher & Yao 2017). These factors culminate in the striking reality that women, on average, retire with approximately 30% less in savings than men (Richardson, 2014). Women experiencing inadequate financial resources in retirement can result in extended poverty in older age (Richardson 2014), trouble accessing medical care (Zhu 2016), loneliness and depression (Khan, Addo & Findlay 2024), inability to leave abusive situations (Betz-Hamilton & Zurlo 2019), and other negative effects.
Financial professionals help clients prepare for retirement in many ways, such as calculating retirement needs, setting retirement goals, and ensuring diversified retirement accounts (Marsden, Zick & Mayer 2011). The benefits of using a financial advisor are well-documented; they help clients set goals, save for emergencies, and encourage other positive financial behaviors (Marsden, Zick & Mayer 2011). Working with a financial advisor helped clients increase their financial knowledge and demonstrate better financial management behaviors like paying off their credit cards on time (Moreland 2018). With financial advice being connected to positive financial outcomes, it is troubling if men and women pursue financial advice at different rates, as research has suggested that men tend to work more with financial planners (The American College 2021).
Research has not fully examined how financial help-seeking differs between men and women and what factors explain these differences (Leopold et al. 2015; Richardson 2014). Some have indicated women are more likely to seek help generally (Juvrud & Rennels 2017; Liddon, Kingerlee & Barry 2018). Others point to social norms that discourage women from engaging in financial activity, such as investing or other business pursuits that create differences in financial behaviors (Ke 2021; Sekścińska et al. 2023). This paper assesses who is more likely to seek help and uses a Blinder-Oaxaca decomposition (specifically a multivariate decomposition for nonlinear response models) to analyse whether individual differences or group characteristics explain the gap in help-seeking behaviour or whether it is due to unexplained factors such as social norms, discrimination, or other factors not measured (Powers, Yoshioka & Yun 2011). Individual differences, such as retirement preparedness behaviours, risk tolerance, and financial knowledge, can potentially be addressed through targeted interventions to reduce gender disparities. However, if men and women with similar characteristics experience different outcomes, this may point to social norms, gender biases, or unequal treatment, which then must also be considered in efforts to close the financial help-seeking gap, and, by extension, the retirement preparedness gap.
Rather than solely documenting whether gender differences in financial help-seeking exist, this study uses data from the 2022 Survey of Consumer Finances (SCF) to examine why such differences persist by distinguishing among competing explanations. Specifically, we assess whether gender differences in help-seeking are primarily attributable to group differences in predisposing characteristics—such as risk tolerance, saving behaviour, and financial knowledge—or to differences in how these characteristics translate into help-seeking behavior. By applying a logit-based multivariate decomposition, this study provides insight into whether observed gender gaps reflect compositional differences or differential processes, advancing existing research that has largely focused on gender effects conditional on controls. In sum, this study’s research question is whether the gap in help-seeking is due to gender-based differences in treatment or to group differences in individual characteristics that explain the gap.
To comprehensively understand the factors influencing individuals’ engagement in retirement planning and their decisions to seek financial assistance, this paper employs Grable and Joo’s (1999) help-seeking model, which is based on the earlier health belief model (Rosenstock 1974). Four perceptions drive behaviour: perceived susceptibility, perceived severity, perceived benefits, and perceived barriers. Grable and Joo’s (1999) help-seeking framework offers insights into the determinants that guide the pursuit of financial help, highlighting the roles of financial stressors, risk tolerance, and demographic factors in the decision-making process. When applied to retirement preparedness, a person’s perceived vulnerability in retirement, the severity of their financial condition, the effectiveness of engaging in retirement planning behaviours, and the barriers to engaging in retirement planning behaviours could influence whether someone takes action (Joo & Grable 2001; Kim & Hanna 2015; Mayer, Zick & Marsden 2011).
Grable and Joo (1999) describe five stages within the financial help-seeking framework. These stages include (1) exhibiting a financial behaviour, (2) analysing the consequences of the behaviour, (3) determining why the behaviour occurred, (4) deciding to seek help, (5) deciding whom or what to seek help from. Within these five stages, Grable & Joo (1999) argue that predisposing factors such as financial knowledge, risk tolerance, and perceived financial stress and anxiety will impact one’s decision at each stage. Higher rates of financial knowledge, risk tolerance, saving generally, saving for retirement, planning for retirement, and perceived financial stress and anxiety should lead to higher rates of financial help-seeking (Grable & Joo 1999; 2001). For an illustration of the conceptual framework, see Joo and Grable (2001, p. 44). In this framework, variables can help explain differences in behaviours for multiple stages. For example, financial knowledge influences the decision to start managing personal finances, the ability to analyse, and the comfort with seeking help. Individuals with lower financial knowledge may avoid attempts to manage their finances themselves. They may also be reticent to analyse the outcome of their decisions and then be uncomfortable talking to a professional about their finances.
The theory suggests that people are likely to adopt a given behaviour (financial advice-seeking) when they perceive a threat to their personal situation, and they believe that adopting the behaviour will mitigate that threat; in this case, the lack of retirement preparedness (Carpenter 2010). Because women generally report higher levels of general stress and anxiety than men (Jalnapurkar, Allen, & Pigott 2018), the theory predicts they should seek advice more often.
Research confirms this pattern: women are more likely than men to turn to a financial advisor, but only when analyses control for income, education, and wealth (Fan et al. 2021; Ludwig, Heckman, & McCoy 2023; Reiter & Qing 2024). Yet industry statistics still show men receiving advice at higher rates (Baghai et al. 2020). Importantly, Grable and Joo’s (1999) framework implies two distinct pathways through which gender differences in help-seeking may arise. Differences may emerge because individuals enter the help-seeking process with different predisposing characteristics (e.g., financial knowledge, risk tolerance, saving behaviour) or because individuals with similar characteristics experience different perceived barriers or benefits when deciding to seek help. This study leverages this theoretical distinction by empirically separating differences attributable to group characteristics from differences attributable to differential returns to those characteristics.
Planning for retirement is critical to a successful retirement. Planning may include participating in retirement pensions or savings plans and actively saving through private savings programs (Berger & Denton 2004; Kim & Hanna 2015). A recent survey by Morgan Stanley (2024) found that women report feeling less confident in their ability to comfortably retire. This sentiment may reflect an accurate assessment of their objective condition, as women, on average, save 32% less than their male peers (Bolton & Concepcion 2024). In fact, multiple studies have found a retirement savings gap for women, including lower rates of retirement account ownership, lower estimated account balances, and fewer overall assets than men (Chang, Hernández Kent & McCulloch 2021; Enda & Gale 2020; National Women’s Law Center and National Institute on Retirement Security 2021; U.S. Census Bureau 2022).
This retirement savings gap is due to multiple factors. Not only do women live longer than men, but women also experience disrupted life-course work patterns resulting from family care needs (e.g., caring for their children or ageing parents). Discontinuous work results in several negative financial consequences for women (Curtis & Rybczynski 2015; Evers & Sieverding 2014). For example, women may be eligible for reduced government retirement benefits, ineligible for participation in pension plans, and income from part-time work may be insufficient to set aside adequate savings (Tomar et al. 2021). These persistent economic disadvantages (interrupted work and part-time employment) may negatively impact women’s financial preparedness for retirement as they often have more debt and less savings, which can cause retirement delays (Yakoboski, Lusardi & Hasler 2023).
One factor that positively influences behaviours towards retirement preparedness is financial help-seeking (Fan 2021). While women are more likely to seek financial help from non-professional sources, they are less likely to seek help from professionals (Glenn & Heckman 2020). However, studies that have controlled for more factors (e.g., knowledge, risk tolerance), find that when those factors are controlled, women become more likely to seek professional financial advice (Hanna 2011; Robb, Babiarz & Woodyard 2012; White & Heckman, 2016). This may seem confusing, given that men work with financial planners at much higher rates (Catania et al. 2025). The key insight is that the results should be interpreted as indicating that if men and women had equivalent levels of risk tolerance, financial knowledge, and related characteristics, they would seek financial advice or assistance at higher and more comparable rates. Another possible explanation is that there is a structural bias against women that prevents them from seeking financial help at the same rate. This analysis uses a Blinder-Oaxaca decomposition to better understand the multiple factors associated with the gender gap in financial-help seeking.
In addition to retirement preparedness, women generally report lower levels of financial risk tolerance than men, a pattern observed in both everyday financial decisions and long-term investment strategies (Bannier & Neubert 2016; Fisher & Yao 2017; Yao & Hanna 2005). One explanation for this disparity draws on social role theory, which posits that gender norms and socialisation shape financial behaviours (Eagly 1987). Men are traditionally socialised toward agentic characteristics such as assertiveness and competitiveness, and generally report higher risk tolerance and are more inclined to choose higher-risk, higher-return assets. In contrast, women’s socialisation often emphasises communal or cautious orientations, resulting in lower self-reported risk tolerance and more conservative investment choices (Eagly 1987).
The consequences of this gender gap in risk tolerance are significant for retirement preparedness. Conservative investment strategies often yield lower returns over time, and women’s comparatively lower risk tolerance may, therefore, contribute to the persistent gender wealth gap (Bannier & Neubert 2016; Fisher & Yao 2017). Conversely, men’s higher propensity for risk-taking can amplify potential gains, though it also increases exposure to market volatility and the possibility of larger losses.
It is also important to consider both objective and subjective financial knowledge when examining how women and men seek help. Objective financial knowledge refers to what individuals actually know, measured by standardised assessments of financial concepts (Lusardi & Mitchell, 2014). Subjective financial knowledge involves individuals’ self-perceived understanding of financial matters (Robb & Woodyard 2011). Although both forms of knowledge influence positive financial behaviours, studies indicate that women often demonstrate lower levels of objective knowledge on financial assessments than men (Allgood & Walstad, 2016). Women’s self-perceived financial knowledge may also differ from men’s, which can affect decision-making processes and ultimately their retirement security (Lind et al. 2020; Robb et al. 2015).
Higher objective financial knowledge is linked to activities such as making retirement contributions, performing calculations to estimate retirement needs, and accumulating greater overall savings (Lusardi & Mitchell 2017; van Rooij, Lusardi & Alessie 2011). However, how researchers measure objective knowledge varies considerably, often resulting in inconsistent reliability and validity (Kaiser & Menkhoff 2020; Knoll & Houts 2012). Subjective financial knowledge also influences significant retirement behaviours, including efforts to plan for future income (Allgood & Walstad, 2016; Tokar & Asaad, 2015). Errors in perceiving how much one truly knows can lead to suboptimal outcomes; for instance, overestimation may lead to uninformed risk-taking (McCoy, White & Love 2019), while underestimation may cause individuals to forgo beneficial opportunities (Heo et al. 2021).
Gender differences arise from factors like varying exposure to complex financial products, traditional roles in household financial management, and social norms that affect financial choices (Kim, Lee, & Kim 2022; Robb & Woodyard 2011). As a result, women may enter the retirement planning process with less familiarity and may sometimes underestimate their abilities (Heo et al. 2021; Lind et al. 2020). Measuring both objective and subjective financial knowledge provides a clearer picture of how women approach retirement planning (Allgood & Walstad 2016; Xiao, Serido & Shim 2011). This approach clarifies discrepancies between what women genuinely know and what they believe they know, thereby helping researchers and practitioners develop targeted programs and planning tools that address specific gaps (Magwegwe & Lim 2021; Mayer et al. 2011).
This study focuses on stage four of the financial help-seeking framework to determine how retirement preparedness influences the likelihood of help-seeking. Differences in retirement preparedness (as measured by perception of retirement preparedness and having retirement accounts and savings), financial attitudes (as measured by risk tolerance), and objective and subjective financial knowledge may help explain why men and women have demonstrated different financial help-seeking behaviours in prior literature. Based on the original framework and the accompanying literature, the following hypotheses are presented:
H1. Men and women will exhibit different help-seeking behaviours, with men more likely to decide to seek help. H2. Gender group mean differences in perceived retirement preparedness will partially explain gender differences in financial help-seeking. H3. Gender group mean differences in intentional savings behaviours will partially explain gender differences in financial help-seeking. H4. Gender group mean differences in retirement account ownership will partially explain gender differences in financial help-seeking. H5. Gender group mean differences in risk tolerance will partially explain gender differences in financial help-seeking. H6. Gender group mean differences in subjective financial knowledge will partially explain gender differences in financial help-seeking. H7. Gender group mean differences in objective financial knowledge will partially explain gender differences in financial help-seeking. H8-H13. Differences in the unexplained portion for perceived retirement preparedness, intentional savings behaviours, retirement account ownership, risk tolerance, and subjective and objective financial knowledge will partially explain gender differences in financial help-seeking.
This study used the Survey of Consumer Finances (SCF) 2022 to test these hypotheses. The SCF is a triennial survey that interviews households using a dual-frame probability-based design to gather financial and demographic data. The SCF also includes weighting files, which are provided to make the sample representative of the population of the United States of America. The SCF uses five imputations to account for missing data and to protect privacy, which results in each respondent having five sets of data. Multiple imputations are used to handle nonresponse items. All respondents were included in the analysis. The desired sample for this study was 4,595 respondents.
The dependent variable for this study examined stage four of the help-seeking framework. The related survey question is, “When making saving and investment decisions, some people search for the very best terms while others don’t. On a scale from zero to ten, where zero is no searching, and ten is a great deal of searching, what number would you be on the scale?” This question about investment and saving decisions was most relevant to this paper’s inquiry into retirement preparedness. This was treated as a continuous variable to capture variation in help-seeking intensity.
The three variables used to measure retirement attitudes and stressors were self-assessed retirement preparedness, saving habits, and retirement account ownership. Similar financial behaviours have been highlighted as supportive of multiple stages of the financial help-seeking (Grable & Joo, 2001). Self-assessed retirement preparedness was measured by respondents rating the adequacy of the retirement income they expect to receive from 1 inadequate to 5 very satisfactory. Saving habits were defined as those who were intentional savers. Those who saved regularly or spent one income and saved the other were considered intentional savers and coded as 1. Those who do not save, or who save only whatever is left over each month, were coded 0 for not intentionally saving. Finally, whether the respondent owns a retirement account, which included Keoghs, IRAs, pensions, or other tax-deferred savings plans, was included as a binary variable where 0 = no accounts and 1 = at least one account.
Risk tolerance was a continuous variable that was measured by the question, “Some people are fully prepared to take financial risks when they save or make investments, while others try to avoid taking financial risks. On a scale from zero to ten, where zero is not at all willing to take risks, and ten is very willing to take risks, what number would you be on the scale?
Subjective financial knowledge was a continuous variable that asked respondents how knowledgeable they felt about personal finance on a scale from 0, not at all knowledgeable, to 10, very knowledgeable. Subjective financial knowledge and risk tolerance (as previously described) were modelled as continuous predictors, consistent with prior research showing that Likert-type scales with more than five categories can be treated as approximately continuous without inflating Type I error rates or reducing statistical power (Robitzsch 2020; Huh & Gim 2025).
Objective financial knowledge was measured by three financial knowledge questions from Lusardi and Mitchell (2011) covering diversification, compound interest, and inflation. These three questions were used to make a summative scale (Lusardi and Streeter 2023; O’Connor and Ivory 2025).
Age, income, net worth, race, marital status, employment status, and education were controlled for in this study. Age was a continuous variable. Income was log-transformed to account for the skewness of the data. The inverse hyperbolic sine of net worth was used to account for zero and negative values. Race was a binary variable where respondents were grouped as White or non-White. Marital status was a binary variable, which included those who were married or living with a partner and those who were single (including those who were divorced, widowed, separated, or never married). Employment status was collapsed into those who were working full- or part-time and those who were not employed. Lastly, education was categorised as follows: less than high school, some college, associate’s degree, bachelor’s or master’s degree, and a doctorate or professional degree. Whether a household expects to pay for a major foreseeable expense in the next five to ten years was a binary yes/no variable.
To test how retirement attitudes and stressors, financial attitudes, and financial knowledge influence help-seeking behaviour, t-tests were first used to examine gender differences in stage four of the help-seeking model. A regression-based multivariate decomposition for a non-linear response model was then used to examine gender differences, with the Stata command mvdcmp and population weights applied. A multivariate decomposition estimates how much of the mean differences between genders is due to observed characteristics (i.e., the explained portion), which include the independent variables of interest in this study, and how much is attributable to differences in treatment for those same or other unidentified factors (i.e., unexplained portion; Powers, Yoshioka & Yun 2011).
The portion of the gap due to differences in observed characteristics is the explained portion and reflects the amount of disparity in help-seeking that could be reduced if men and women had similar attributes (risk tolerance, financial knowledge). The remaining portion, or unexplained portion, is often attributed to differences in treatment, in this case, a difference in treatment based on gender, or on factors not included in the analysis. The portion explained by differences in groups suggests equal returns to the characteristics measured, while the remaining portion may be attributable to discrimination or differences in treatment. The unexplained portion is the estimated amount of inequity that would remain even if characteristics were equalised (Rahimi and Nazari 2021). This approach allows us to assess whether gender differences in financial help-seeking are driven primarily by differences in observable characteristics or by differences in how those characteristics translate into help-seeking behaviour, a distinction not captured by standard regression-based analyses. As there are no methods available to account for multiple imputations in decomposition analysis, point estimates are used as is common convention (e.g., Shin & Hanna 2015).
To ensure no multicollinearity or construct overlap, we ran factor analyses of financial knowledge and retirement preparedness, which yielded eigenvalues within the acceptable range (<1).
Additionally, we tested the relationship between account ownership and saving habits, and the VIF results were within the acceptable range (1–4). These tables are available upon request.
The weighted descriptive statistics are reported in Table 1. When considering the degree to which respondents felt that their retirement income was satisfactory, the majority reported it as at least ‘adequate,’ which relates to about a 3 out of 5 on the scale. About 55% of respondents reported intentionally saving, and 42% reported owning a retirement account. Respondents appeared to be more risk-averse, with a mean risk score of 4.24. Respondents averaged 2.23 (SD = .81) on the financial knowledge scale and assessed their financial knowledge at 7.24 (SD = 2.16) on average.
Descriptive statistics for the total sample, men, and women in the 2022 SCF
| Variables | Total Sample n = 4,595 | Men n = 2,593 | Women n = 2,002 | |||||
|---|---|---|---|---|---|---|---|---|
| Prop (%) | M (SD) | Prop (%) | M (SD) | Prop (%) | M (SD) | Min | Max | |
| Help-seeker (Stage four) | 6.20 (3.17) | 6.42 (2.98) | 5.98 (3.33) | 0 | 10 | |||
| Perceived retirement preparedness | 3.05 (1.32) | 3.23 (1.31) | 2.89 (1.31) | 1 | 5 | |||
| Intentional saver | 54.68 | 58.48 | 50.97 | 0 | 1 | |||
| Owns retirement account | 41.55 | 45.86 | 37.35 | 0 | 1 | |||
| Risk tolerance | 4.24 (2.74) | 4.81 (2.65) | 3.68 (2.70) | 0 | 10 | |||
| Objective financial knowledge | 2.23 (.81) | 2.41 (.73) | 2.05 (.85) | 0 | 3 | |||
| Subjective financial knowledge | 7.25 (2.16) | 7.36 (2.02) | 7.16 (2.29) | 0 | 10 | |||
| Age | 52.44 (17.61) | 51.84 (17.41) | 53.04 (17.79) | 18 | 95 | |||
| Log income | 11.14 (1.25) | 11.33 (1.39) | 10.95 (1.05) | $0 | $458,000,000 | |||
| IHS Net worth | 10.95 (6.51) | 11.82 (5.67) | 10.11 (7.13) | $ (555,500.00) | $2,390,000,000 | |||
| White/Caucasian | 68.89 | 72.58 | 67.26 | 0 | 1 | |||
| Married | 63.22 | 67.28 | 55.22 | 0 | 1 | |||
| Employed | 61.03 | 0 | 1 | |||||
| Education | ||||||||
| Less than high school | 8.37 | 7.34 | 9.38 | 0 | 1 | |||
| Some college | 20.89 | 20.55 | 21.22 | 0 | 1 | |||
| Associate’s degree | 27.99 | 23.63 | 32.25 | 0 | 1 | |||
| Bachelor’s or master’s degree | 24.46 | 29.06 | 19.98 | 0 | 1 | |||
| Doctorate or professional degree | 18.28 | 19.42 | 17.17 | 0 | 1 | |||
Note. Statistics are population weighted. Min and max values reported for income and net worth are not transformed for descriptive purposes.
There were some notable differences between men and women in the sample. More men reported being help-seekers than women (92% versus 86%). Based on the descriptives, women in the sample may experience more retirement and financial stressors. For example, women did not feel their income sources in retirement would be as adequate (M = 2.89, SD = 1.31) compared to men (M = 3.23, SD = 1.31). They were less likely to report being an intentional saver than men (51% and 58%, respectively), and 38% of women said they owned a retirement account compared to 46% of men. Women reported lower risk tolerances than men; however, the percentage of men and women who expected a major financial expense was about the same. Finally, women had lower objective and subjective financial knowledge scores than men.
The mean age of the sample was about 52 years old. The transformed results for income and net worth are presented in Table 1. The median income was $94,039, and the median net worth of the sample was $384,500. The median income for men was $143,761. The median income for women was substantially less, at $60,531.06. Men’s median net worth in the sample was $878,300, while the median net worth for women was $144,256.
The sample was predominantly White (69%). Most of the respondents were married (63%), and most of the respondents were employed (61%). However, 55% of women were employed compared to 67% of men. Nearly 71% of the sample had at least an associate’s degree, with 72% of men having at least an associate’s degree and 69% of women having at least an associate’s degree.
In addition to these descriptives, a t-test was conducted to examine whether there were significant gender differences in help-seeking behaviour. T-tests revealed that there were significant gender differences in help-seeking, with men reporting higher levels of help-seeking on average than women (t(4593) = −3.61, p < .001, 95% CI [−.69, −.20]). Men (M = 6.67, SD = 2.92) reported higher levels of help-seeking on average than women (M = 6.07, SD = 3.33).
Table 2 explains to what extent men and women differ for each of the predictors, as well as what percentage the financial help-seeking gap between genders would reduce if the factors were equalised. In this table, help-seeking refers to anyone who sought any level of financial help. For example, men (M = 4.81) and women (M = 3.68) in this sample differ significantly in underlying risk tolerance, with a larger proportion of females having risk tolerance scores of 0 to 4 and males having risk tolerance scores of 5 to 10. The largest female (21%) to male (10%) ratio was at 0 (no willingness to take risks), while the largest male (12%) to female (8%) ratio was at 7. In the total sample, only 66% sought financial help when they rated themselves 0 on risk, whereas 97% of those who rated themselves 7 did so. If men and women were represented in equal proportions at each level of risk, the gap would shrink from nearly 6% to 2% based on this one factor alone. The following table presents the same analysis for each predictor, and this paragraph can be used as an example.
Results of gender differences between individual predictors
| Men | Women | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Most prominent | Biggest advantage | Level of advantage | % help-seeker | Most prominent | Biggest advantage | Level of advantage | % help-seeker | Gap if equal | |
| Perceived retirement preparedness | 4 to 5 | 22%/15% | 5 | 92% | 1 to 3 | 22%/15% | 1 | 79% | 4.4% |
| Intentional saver | 1 | 58%/41% | Yes | 94% | 0 | 49%/42% | No | 83% | 4.7% |
| Owns retirement account | 1 | 46%/37% | Yes | 95% | 0 | 63%/54% | No | 85% | 4.4% |
| Risk tolerance | 5 to 10 | 12%/8% | 7 | 97% | 0 to 4 | 21%/10% | 0 | 66% | 2.0% |
| Objective financial knowledge | 3 | 54%/35% | 3 | 94% | 0 to 2 | 25%/12% | 0,1 | 81% | 0.3% |
| Subjective financial knowledge | 7 to 9 | 22%/16% | 7 | 93% | 1 to 6, 10 | 14%/9% | 5 | 82% | 2.3% |
| Education | Bachelor’s or more | 29%/20% | 4 | 0.96 | Associate’s or less | 32%/24% | 3 | 0.88 | 0.042 |
The results of the OLS regression are reported in Table 3 along with model fit statistics. Perceived retirement behaviour was positively associated with help-seeking behaviour (B = .11, p = .036). Compared to those who were not intentional savers, those who were intentional savers were also more likely to engage in more help-seeking behaviours (B = .69, p < .001). Risk tolerance was positively associated with help-seeking behaviour (B = .19, p < .001). Moreover, higher objective (B = .20, p = .016) and subjective (B = .30, p < .001) financial knowledge were associated with more help-seeking behaviour.
Results of OLS regression results
| Variable | Coefficient | SE | p |
|---|---|---|---|
| Perceived retirement preparedness | 0.11 | 0.05 | 0.036 |
| Intentional saver (Is not an intentional saver) | 0.69 | 0.13 | < .001 |
| Owns retirement account (Does not own) | 0.13 | 0.14 | 0.337 |
| Risk tolerance | 0.19 | 0.03 | < .001 |
| Expects future expense (Does not expect) | 0.16 | 0.12 | 0.185 |
| Objective financial knowledge | 0.20 | 0.08 | 0.016 |
| Subjective financial knowledge | 0.30 | 0.03 | < .001 |
| Gender (Male) | 0.06 | 0.12 | 0.638 |
| Age | −0.02 | 0.00 | < .001 |
| Log income | −0.08 | 0.04 | 0.077 |
| IHS Net worth | 0.01 | 0.01 | 0.206 |
| White (Non-White) | −0.22 | 0.13 | 0.088 |
| Single (Married) | −0.08 | 0.13 | 0.536 |
| Employed (Unemployed) | 0.01 | 0.14 | 0.92 |
| Education (Less than high school) | |||
| Some college | 0.25 | 0.27 | 0.356 |
| Associate’s degree | 0.53 | 0.27 | 0.05 |
| Bachelor’s or master’s degree | 0.78 | 0.27 | 0.005 |
| Doctorate or professional degree | 0.63 | 0.28 | 0.027 |
| Intercept | 3.06 | 0.57 | < .001 |
Note. Source: 2022 Survey of Consumer Finances, weighted analysis using all five implicates and RII technique.
F(18, 4533.0) = 28.44, p < .001.
R2 = .17.
When examining the control variables, age was negatively associated with help-seeking behaviour, with older respondents engaging in less help-seeking than younger respondents (B = −.02, p < .001). Education was also significantly related to help-seeking behaviour. Compared to those with less than a high school degree, those with an associate’s degree (B = .53, p = .05), those with a bachelor’s degree (B = .78, p = .005), and those with a doctorate or professional degree (B = .63, p = .027) were more likely to engage in greater help-seeking behaviour.
The decomposition results are reported in Table 4. As previously stated, men sought help at a higher rate than women. The decomposition analysis seeks to explain what portion of the gap was due to statistically associated returns to variables. This should not be confused with causal effects, though discussions of decomposition results typically use language similar to that of causal effects (Fonseca et al. 2012; Preston, Qiu & Wright 2024). Most of this difference was attributable to observed differences between men and women, including perceived retirement preparedness, intentional savings behaviours, ownership of retirement accounts, risk tolerance, objective and subjective financial knowledge, and a variety of demographic variables. This endowment portion accounted for .514 (115.5% of the total difference) of the observed gender differences in help-seeking behaviour. The explained component exceeds 100% of the observed gap because the unexplained component is negative, indicating that differences in unexplained components partially offset the predicted gap, though this offset is not statistically significant.
Difference due to explained and unexplained components in financial help-seeking
| Financial help-seeking | Coefficient | p-value | 95% CI | Percent |
|---|---|---|---|---|
| Explained Components | 0.514 | <0.001 | (.46, .57) | 115.5 |
| Unexplained Components | −0.069 | 0.259 | (−.19, .05) | −15.5 |
| Total Difference | 0.445 | <0.001 | (.34, .54) |
Table 5 reports the multivariate decomposition results of gender differences and financial help-seeking. Self-assessed retirement preparedness accounted for an estimated 6% of the help-seeking gap between men and women, while being an intentional saver accounted for 11% of the financial help-seeking gap. Owning a retirement account reduced the gap in help-seeking behaviour by 3%. The most significant individual factor was risk tolerance, accounting for 46% of the difference in financial help-seeking. Objective and subjective financial knowledge were also significant, accounting for 24% and 17% of the gap, respectively. Group mean differences in education levels, age (5%), and net worth (11%) were also significant factors in explaining the gap. Income (−4%) and race (−5%) reduced the gender gap in help-seeking behaviour.
Multivariate decomposition results of gender differences and financial help-seeking
| Group 1 = male (reference group) | ||||||
|---|---|---|---|---|---|---|
| Group 2 = female (comparison group) | ||||||
| Difference due to explained components | Difference due to unexplained components | |||||
| Variable | Coef | SE | Percent shared | Coef | SE | Percent shared |
| Perceived retirement preparedness | 0.027* | 0.010 | 5.97 | −0.127 | 0.134 | −28.61 |
| Intentional saver (Is not an intentional saver) | 0.051*** | 0.005 | 11.46 | −0.029 | 0.057 | −6.42 |
| Owns retirement account (Does not own) | −0.013* | 0.007 | −3.00 | −0.217*** | 0.047 | −48.74 |
| Risk tolerance | 0.206*** | 0.018 | 46.44 | −0.017 | 0.083 | −3.82 |
| Expects future expense (Does not expect) | −0.001 | 0.001 | −0.18 | −0.221*** | 0.057 | −49.67 |
| Objective financial knowledge | 0.109*** | 0.018 | 24.46 | 0.355* | 0.149 | 79.84 |
| Subjective financial knowledge | 0.075*** | 0.004 | 16.85 | 0.773*** | 0.207 | 173.81 |
| Age | 0.022*** | 0.003 | 5.05 | −0.279 | 0.206 | −62.66 |
| Log income | −0.019* | 0.008 | −4.37 | 0.890 | 0.470 | 200.27 |
| IHS Net worth | 0.049*** | 0.011 | 11.05 | 0.255** | 0.088 | 57.27 |
| White (Non-White) | −0.022*** | 0.004 | −5.06 | −0.263** | 0.076 | −59.1 |
| Single (Married) | 0.011 | 0.016 | 2.51 | −0.021 | 0.062 | −4.73 |
| Employed (Not employed) | −0.019 | 0.011 | −4.24 | −0.191** | 0.068 | −42.99 |
| Education (Less than high school) | ||||||
| Some college | −0.002* | 0.001 | −0.54 | 0.035 | 0.052 | 7.87 |
| Associate’s degree | −0.077*** | 0.015 | −17.36 | 0.213** | 0.077 | 47.95 |
| Bachelor’s or master’s degree | 0.092*** | 0.016 | 20.74 | 0.088 | 0.050 | 19.7 |
| Doctorate or professional degree | 0.025*** | 0.004 | 5.71 | 0.175*** | 0.044 | 39.34 |
Note. Reference groups are in parentheses. Continuous variables do not have a reference group.
p < .05,
p <.01,
p < .001.
Although the unexplained portion was not statistically significant, several individual factors were statistically significant, reflecting variation in how specific factors relate to financial help-seeking between genders. Differences in the returns to subjective financial knowledge and objective financial knowledge contributed positively to the unexplained portion, while differences in returns to retirement account ownership, future expenses, age, race, and employment status offset the predicted gender gap. Overall, the results indicate that gender differences in financial help-seeking behaviour may be associated with differences in observed characteristics such as retirement attitudes and stressors, risk tolerance, financial knowledge, and demographics, rather than with differential returns to those characteristics.
Before examining why financial help-seeking differed between men and women, it was critical to determine whether gender differences existed. The results from the t-tests and descriptives provide support for H1, that men and women exhibit different help-seeking behaviours, with men seeking financial advice at higher rates. Then, decomposition analysis was used to delineate sources of differences between groups (Powers, Yoshioka & Yun 2011). A common application is to determine whether groups are receiving equal treatment for the same prerequisites. In this study, the question is whether men and women with identical individual characteristics would seek financial help at the same rate. If not, factors outside of the variables here, including potential discrimination or the effects of gender norms, should be considered. Gender differences in the decision to seek help were attributable to differing retirement attitudes and stressors, partially supporting H2–7. For example, men’s intentional saving behaviours and risk tolerance explain a significant portion of differences in financial help-seeking (H5).
The decomposition results also suggest that gender differences in subjective and objective financial knowledge are important factors in understanding men’s and women’s financial help-seeking behaviours, supporting H6 and H7. Men’s higher levels of subjective financial knowledge, which is often referred to as a measure of financial confidence, and objective financial knowledge were positively associated with help-seeking behaviour. Hypotheses eight through fourteen were not supported, as the unexplained portion of the decomposition did not significantly explain the gender gap in help-seeking behaviour. This suggests that differences in behaviour are more attributable to group differences in observed characteristics (or factors not included in this study) than to differential treatment. Instead, women’s and men’s financial help-seeking behaviours were explained by differences in intentional saving behaviours, retirement account ownership, risk tolerance, subjective and objective financial knowledge, and other control variables (age, income, education, and net worth). In other words, the finding that the unexplained portion of the decomposition is not statistically significant is theoretically informative. It suggests that, within the constructs measured, women and men with similar levels of risk tolerance, financial knowledge, and saving behaviour do not differ meaningfully in how these characteristics translate into help-seeking decisions. Thus, observed gender differences are more consistent with compositional differences than with differential treatment or unequal returns.
This analysis differed from other decompositions, as the differences between men and women are understood to be internal or individual factors, such as different perceptions of challenges, differences in retirement needs, and willingness to seek advice or help, rather than external factors like unequal treatment, discrimination, or sexism. Other possible factors could be at work in explaining the difference in financial help-seeking behaviour. For example, debt and denial, guilt, or embarrassment about a financial situation can prevent seeking financial help (Fan 2021; Uhl et al. 2022). Women experience more stress around debt than men, yet are less likely to be responsible for debt decision-making (Callegari, Liedgren, & Kullberg 2017; Dunn & Mirzaie 2022). Moreover, anticipation of future financial challenges and opportunities may prompt financial help-seeking. Future-oriented mindsets drive present behaviours (Gjesme 1983; Raynor 1970). However, when women feel less capable and responsible for planning, they are less likely to act (Hitczenko 2024). Similar to how women’s motivation for academic achievement is dampened by social pressures of academic success as inappropriate for their gender, social pressures may also discourage striving toward financial goals or seeking financial help (Horner 1974). Stereotype threat and social norms remain barriers that impede women’s attainment of financial knowledge (Tinghög et al. 2021). Future research should continue to explore these variables to provide a complete picture of gender differences in financial help-seeking behaviour.
The direct implications of the analysis are that most of the gender gap is explained by the association between individual characteristics (financial knowledge, risk tolerance, and savings intentions) and financial help-seeking. Programs and other efforts that specifically target helping women increase their financial knowledge, define their risk tolerance, and improve their savings behaviours may help narrow the gap. These factors are perceived barriers from the individual’s perspective that lead to self-exclusion from financial matters. Initiatives may increase objective financial knowledge and improve women’s confidence in their financial knowledge through workshops, support groups, and online seminars and materials. This may include directly confronting gender norms and discrimination that result in deficits in these specific measures.
While the original financial help-seeking framework by Grable and Joo (1999) suggests that stressors can prompt help-seeking behaviour, research has found that women can face significant financial stressors that discourage both informal and formal help-seeking (Uhl et al. 2022). This may account for the results that men sought more help than women. If men decide to seek financial help because they perceive they are more ready for retirement, are willing to take on more risk in their investments, have greater confidence in their financial abilities, and have more financial knowledge, it may be that women feel more overwhelmed by their future financial outlook and are discouraged from seeking financial help.
There are also some important implications for practitioners. First, prior research has found that women are more likely to experience internalising symptoms (i.e., anxiety and depression) in light of stressor events. In contrast, men are more likely to experience more action-oriented or aggressive behaviours when faced with stressors (Farhane-Medina et al. 2022). Anxiety and depression are more associated with decision paralysis and may lead to difficulties motivating oneself to engage in help-seeking behaviours or may result in women seeking help for their mental well-being over their financial well-being, as there is a lower stigma for mental health services for women than men (Schnyder et al. 2017). In fact, a study at a clinic that offered both mental health services and financial services found that men were more likely to engage in financial therapy services than mental health services despite presenting with needs in both areas (Ford et al. 2020). Future studies would benefit from examining gender differences in the role financial anxiety plays in help-seeking behaviours.
The findings of this study also have implications for future research design. A book called Invisible women: Data bias in a world designed for men describes how women are still under-represented in research and how the gender gap has caused findings that may counter best practices for women. The book cites examples of how medication dosages are often determined for a standard man’s body and may not be appropriate for many women (Perez 2019). Another example is that medical research on heart attacks has caused public health campaigns to focus on chest pains, which is the most common symptom for men, but not for women. Women’s symptoms are shortness of breath, nausea, or jaw pain, which are not discussed (Al Hamid et al. 2024). While financial stressors may be more motivating for men, women may be more likely to seek help when they are financially secure. The finding that women seek help less than men in this study may be a result of increased anxiety from these stressors that inhibit their help-seeking (Grable & Joo 2001). Since the financial planning field (both practitioners and their clients) remains disproportionately skewed toward men, the help-seeking framework may need to be revisited to assess its applicability to women and men. It is possible that the very measures used in and the design of this study are tailored to a framework more applicable to men than to women. Even the typical decomposition solution of addressing the gap by equalising underlying factors assumes that the solution is to make women more like the men in this sample. Rather, a better solution may be to consider different factors that are more motivational for women but not measured here. Incorporating these additional factors into the help-seeking framework may improve its applicability.
The results also have implications for financial advising professionals and counsellors on how they market to their clients. Men and women appear to have different reactions to their retirement stressors, financial attitudes, and financial knowledge that explain their help-seeking behaviour. The results can help practitioners identify factors that may prevent women from seeking help and better understand major motivational factors. Recognising that anxiety can act as a significant barrier to seeking financial help, marketing strategies should emphasise safety, trust, and empowerment. Women who perceive themselves as risk-averse and not financially knowledgeable are less likely to seek financial advice. Positioning women as capable financial managers who take smart risks may be more motivational than highlighting future challenges and personal shortcomings.
Campaigns should highlight the collaborative and judgment-free nature of financial planning services, reassuring prospective clients that their unique concerns will be met with understanding and respect. Messaging should avoid technical jargon or overly complex financial language, which could heighten anxiety, and instead focus on relatable, solution-oriented narratives. For example, sharing testimonials or case studies of women who successfully navigated their financial challenges with the support of a planner can foster relatability and inspire confidence. Visuals and tone should reflect inclusivity and warmth, creating an approachable image of financial planning. Moreover, emphasising the accessibility of services through flexible formats, such as virtual meetings or asynchronous support, can appeal to women managing anxiety or busy schedules. Highlighting the availability of female financial planners or professionals trained in financial therapy could also resonate with women seeking an empathetic and understanding approach (Reiter, Seay, & Loving 2022).
In addition to the multiple implications discussed previously, some limitations and areas for future research should be addressed. First, the command used for the decomposition analysis did not support multiple imputation. This is a common methodological limitation with decomposition analysis, and future research will ideally continue efforts to find workarounds to this issue. Second, this study focused solely on help-seeking behaviours (stage four). Future studies may wish to examine other stages of the help-seeking framework. Moreover, there were some data limitations in this study. Retirement preparedness, for example, is a complex concept that has been measured in many ways. While this study used a self-assessed measure, future research may employ other datasets or primary data collection to capture more detail about retirement preparedness. Another area for future research is regarding the question used to examine help-seeking behaviour, as it was specific to saving and investment decisions. For example, the SCF also asks whether the respondent seeks help with borrowing money or obtaining credit. The results may be different if a different help-seeking question is used. While the SCF has detailed financial data, there may be other financial and retirement factors that influence gender differences in help-seeking behaviour. Future research could also explore this topic with other datasets to capture more explanatory variables.