Caregiver Distress Explains Links between Objective and Subjective SES and Adolescent Internalizing and Externalizing Behaviors

Abstract

The U.S. is amid an unprecedented adolescent mental health crisis, and adolescents who experience socioeconomic disadvantage are at elevated risk for behavioral problems. Low socioeconomic status (SES) is related to parent distress and is one potential pathway through which disadvantage relates to adolescent internalizing and externalizing problems. Empirical support for this pathway is typically based on a single measure of objective SES considered in isolation. Yet, parents’ subjective social status (SSS) also predicts their distress and may, in turn, shape adolescents’ problem behaviors. We examine how multiple objective (annual income, monthly income volatility, and educational attainment) and subjective (SSS) markers of SES independently relate to adolescent internalizing and externalizing problems through parent distress using data from 104 parent-adolescent dyads. In mixed effects models, we observed positive indirect effects of income volatility (within-dyad variation in monthly income) and parent SSS on adolescents’ internalizing and externalizing symptoms through parent distress. Gains in monthly income and greater SSS predicted decreased parent distress, which in turn related to fewer adolescent internalizing and externalizing symptoms. Indirect effects were independent of annual income, educational attainment, and covariates (marital status, adolescent race, sex, SSS). These findings highlight that temporally (income volatility) and psychologically proximal (parent SSS) aspects of SES may serve as points for intervention to reduce parent distress and support adolescent mental health.

Families in the United States (U.S.) are facing increasing economic precarity (Federal Reserve System, 2023; U.S. Census Bureau, 2021). Income and wealth inequality have steadily risen over the past 50 years, with upper-income households bringing in increasingly larger shares of the country’s total wealth and income (Schaeffer, 2020). Moreover, the percentage of Americans experiencing income volatility has grown (Moffitt & Zhang, 2018; Weller, 2018). Low- and middle-income families face substantial swings in monthly income; one report found that these families, on average, experienced five months of income volatility (defined as a monthly income change of ±25%) within a single year (Hannagan & Morduch, 2015). These trends have led to a rise in the prevalence of families experiencing economic stress (Morduch & Schneider, 2017; Schaeffer, 2020). Parents with children may be especially vulnerable to these economic trends due to the cost of raising children, as well as having lower incomes and wealth compared to individuals without children (Horowitz et al., 2020). Moreover, families across the U.S. have seen wage stagnation coupled with a rise in the basic cost of living, with goods and services such as food and housing increasing exponentially (Bureau of Labor Statistics, 2025; Massenkoff & Wilmers, 2023; Petach, 2022).

Growing economic precarity could have downstream consequences for adolescents, including harming their behavioral health. Evidence linking socioeconomic disadvantage to behavioral problems is abundant (Devenish et al., 2017; Peverill et al., 2021). This connection has previously been validated through the Family Stress Model (FSM), which focuses on socioeconomic disadvantage’s effects on child behavior (and development more generally) by way of heightened stress and economic pressure in the home (Masarik & Conger, 2017). Parents’ subjective appraisals of their families’ socioeconomic circumstances fluctuate over time for many reasons including changes in income, employment, or cost of living (e.g., inflation; Conger et al., 2010; Masarik & Conger, 2017). In turn, economic stress can elevate parental distress and lead to inter-familial conflict and negative parenting behaviors like harsh or inconsistent discipline. Increases in parent distress and harsh parenting may impact children’s development, including rises in internalizing (e.g., depressive symptoms, anxiety) and externalizing (e.g., oppositional behavior, aggression) behaviors.

Empirical support for the FSM has considered parent distress as a pathway linking various dimensions of socioeconomic status (SES) to adolescent behavioral problems. Some of the most commonly used indicators of SES in the FSM literature are parental educational attainment, annual income, and occupational status, as these indicators capture different dimensions of SES (e.g., prestige) that may contribute to differences in economic pressure (Conger et al., 2010). However, with the changing economic landscape, there is a need to understand how other dimensions of SES that may be more dynamic and are less commonly studied, like income volatility, wealth, and subjective appraisals of SES – like subjective social status – uniquely and independently contribute to parent distress and adolescent behavioral health to inform policy. For example, if annual income is a principal contributor to adolescent problem behaviors, then cash transfer programs may potentially reduce parent distress and in turn, adolescent problem behavior; however, if income volatility plays a more salient role, then policies aimed at increasing salaried work opportunities may be more fruitful. Moreover, few studies have considered both objective and subjective measures of SES in relation to adolescent problem behaviors, with the exception of a lone study that explored the joint contributions of objective and subjective measures of SES on parental distress and young children’s behavior (child ages 0–3; Roy et al., 2019). It is especially vital to extend this work and explore these links among adolescents as the U.S. is in the midst of an unprecedented rise in mental and behavioral health problems among adolescents (Centers for Disease Control and Prevention [CDC], 2021; Murthy, 2022). Indeed, adolescents are highly vulnerable to mood and emotional issues, substance abuse, and psychiatric disorders, with half of all lifetime mental health issues manifesting by age 14 years (Costello et al., 2011; Kessler et al., 2005). Accordingly, the current study draws data from a recent sample of 104 parent-adolescent dyads to examine how objective and subjective markers of SES independently relate to adolescent internalizing and externalizing problems through parent distress. This has the potential to help researchers, practitioners, and policymakers understand the implications that rising socioeconomic inequality and subsequent caregiver distress may have on America’s adolescent population and identify specific dimensions of SES that may serve as points of intervention to reduce distress and downstream behavioral problems.

Theoretical Foundation

Socioeconomic disadvantage is a significant source of stress. The FSM explained how pressures related to SES and subsequent psychological distress experienced by low-SES families can lead to negative outcomes, including behavioral problems in adolescents (Conger & Elder, 1994; Masarik & Conger, 2017). The model begins by describing how parents in low-SES families face economic pressure, which refers to parents perceived difficulty covering monthly expenses, combating economic shocks, and meeting their families’ basic needs (Conger et al., 2010; Masarik & Conger, 2017). Compared to higher-SES parents, parents with lower levels of education and income often struggle with inconsistent employment, living paycheck-to-paycheck, with insufficient funds to cover basic needs like housing, utilities, and unexpected expenses (Miller et al., 2024; Schenck-Fontaine & Panico, 2019). The economic pressure that comes from socioeconomic disadvantage can exacerbate parent psychological distress (e.g., perceived loss of control), which may undermine family dynamics. For example, parents in distress may experience marital conflict and have limited bandwidth to engage with their adolescents in positive ways (Conger et al., 2002). Furthermore, socioeconomic disadvantages have been associated with harsh and inconsistent parenting that is less nurturing and stimulating, or that is minimally responsive to adolescents’ needs (Conger et al., 1992; McLoyd, 1990). Distress has also been linked to permissive or controlling parenting (Kochanova et al., 2021). The disruption in parenting practices and family relationships that arises from parent distress can then lead to internalizing and externalizing behavioral problems among adolescents (Masarik & Conger, 2017). Adolescents may also perceive their parent’s distress and experience heightened anxiety or depressive symptoms if they are concerned about their parents’ well-being or their family’s economic situation (Conger et al., 1999; McLoyd et al., 1994, 2009). In addition, family arguments about money or not being able to afford things (e.g., going to the movies with friends, buying new clothes) may lead to internalizing or externalizing symptoms, as conflictual parent-child interactions have been tied children’s behavioral problems, including conduct problems, antisocial behavior, and depressive symptoms (Conger et al., 1995; Ge et al. 1994; Sheeber et al. 2001).

Though not as well articulated in FSM literature, parental education is another aspect of SES that may lead to differences in parent psychological distress and negative behavioral outcomes for adolescents (Conger et al., 2010). Conger and colleagues theorize that the effects of parental education operate to influence family stress similarly to income insofar as higher levels of education may reduce financial distress and material deprivation and enable parents to access gainful, stable employment with higher earning potential. Parental education may also have unique effects on parental distress aside from monetary benefits: higher parental education has been linked to better working conditions (e.g., flexibility, benefits, physical demands, regular schedules), which can minimize parent distress and, in turn, positively benefit children’s psychosocial development (Conger et al., 2010; Suiter & Meadows, 2023). In contrast, limited levels of education may cause parents to feel a sense of powerlessness and lack of influence in their formal relationships outside of the home (Parkes et al., 2015). These factors could contribute to elevations in parents’ distress.

The FSM has traditionally not incorporated subjective social status into the model of impacts of SES on parents and children (Roy et al., 2019), but literature has linked it to mental distress (Zell et al., 2018). Subjective social status (SSS) is a measure of SES which reflects how people perceive their standing in their communities relative to others. SSS ratings may be informed by objective measures of SES (e.g., income, education, occupation) as well as feelings of power, privilege, respect, and social acceptance (Anderson et al., 2012). SSS plays an important role in shaping parent mental health over and above objective measures of SES (e.g., Michelson et al., 2016; Scott et al., 2014). SSS may increase mental distress through negative appraisals of oneself, feelings of insecurity, and maladaptive coping strategies such as depressive thinking and rumination (Adler et al., 2000; Schubert et al., 2016). Moreover, SSS has been indirect effects on children’s behavioral problems through parent distress (Scott et al., 2014; Tan et al., 2020; Zell et al., 2018).

Existing Work Linking Objective and Subjective SES to Adolescent Behavioral Problems Through Parent Distress

Extensive empirical work testing the FSM finds support for parent distress linking measures of objective SES (e.g., annual income, parental educational attainment, SES composites) to adolescent outcomes, including behavioral problems (Conger et al., 1992, 1993, 2002, 2010; Masarik & Conger, 2017; Mistry et al., 2008; Oro et al., 2024.). However, only one study has considered how both objective and subjective measures of SES impact adolescent behavioral problems through parent distress (Roy et al., 2019). While Roy and colleagues found that both subjective and objective measures of SES impact child behavior, the study was done with very young children, thus it is not clear whether these processes play out similarly for adolescents. Nonetheless, this study provides empirical evidence suggesting that objective and subjective SES may have indirect links to problem behavior through parent distress (Roy et al., 2019).

Objective Measures of SES

Income.

Income’s role in family stress and adolescent behavior is well-documented (see Devenish et al., 2017; Masarik & Conger, 2017 for reviews). Low levels of income may cause families to experience economic hardship. Economic hardship can threaten family functioning by increasing parent stress and disrupting family interactions, which ultimately impedes children’s behavioral development (Conger et al., 2002; Linver et al., 2002; Mistry et al., 2004). The parent distress mechanism explaining links between income and adolescent problem behavior has been replicated in numerous studies. For example, Mistry and colleagues (2008) observed that low household income (based on income from earnings and public assistance support) led to economic pressure, which contributed to maternal distress, less positive parenting practices, and in turn problem behaviors. Major strengths of this study were the inclusion of maternal education as a covariate, as well as measures of instrumental support (i.e., financial support from friends) and fiscal management strategies (e.g., access to liquid savings, emergency funds, or a line of credit. However, dynamic measures of SES (including SSS and income volatility) were not included. Consistent with these findings, but using household income-to-needs ratios, Morrison Gutman and colleagues (2005) found that lower income-to-needs predicted more parent distress, which in turn resulted in subsequent internalizing symptoms among adolescents. Although the authors found that results did not differ when controlling for parental educational attainment, it is unclear whether associations would remain robust when accounting for other markers of objective and subjective SES (i.e., income volatility, wealth, and parent SSS), as we do in this study.

Income Volatility.

Income volatility is another aspect of SES that could impact adolescent problem behavior through parental distress. Income volatility reflects a change in income from one time period to another (e.g., current minus prior month or current year minus prior year income) and can be operationalized in terms of frequency and magnitude (Holt et al., 2016). Recent literature has considered how income volatility manifests in the daily lives of families. Looking at annual fluctuations in income, previous work has found that when individuals fall below the poverty threshold or when they experience income loss, their mental distress increases (see Thomson et al., 2022 for review). For instance, in a longitudinal study of adults in their 30s, there was a higher incidence of depression over the 10-year period among those who experienced income loss, even when adjusting for their annual household income (Prause et al., 2009). Similarly, a study following adults from 2001 to 2005 found that declines in annual income heightened risk for depression and anxiety (Sareen et al., 2011). Interestingly, Sareen et al. did not find that increases in income predicted a decreased risk of depression or anxiety; although, other studies have shown benefits of income increases on mental health outcomes (Garratt et al., 2016; Pignatti & Parolin, 2024).

There is evidence to support parent distress as a pathway through which volatility relates to adolescent problem behavior (e.g., Conger et al., 1992; Conger et al., 1994; Solantus et al., 2004). Solantaus and colleagues (2004) observed that Finnish parents who experienced a drop in income endorsed greater economic and psychological distress, which culminated in reductions in their adolescents’ behavioral health. Moreover, Conger and colleagues (1992) found that economic stress was indirectly linked to adolescent behavior through reductions in in parental mood and nurturing parenting. Still, research has not substantiated the pathway between income volatility and adolescent behavioral problems through parent distress when controlling for other aspects of SES, such as SSS.

Wealth.

Wealth is a less well-studied facet of SES that may also contribute to differences in parental distress and subsequent problem behaviors among adolescents. Wealth reflects a families’ stock of financial resources (e.g., money in savings accounts, housing equity) that parents can draw upon to support their adolescent children’s development, including smoothing consumption following unexpected financial crises to minimize disruptions to family rhythms and relationships (Rodems & Pfeffer, 2021). For example, parents can draw on their savings to cover the cost of home repairs without needing to delay repairs or pay for them by taking money away from other expenses (e.g., food, extracurricular activities), which could be a source of distress and contention among parents and their adolescent children. In fact, research has shown wealth serves as a safety net against material hardship following unexpected life events (Rodems & Pfeffer, 2021) in part by reducing family members’ perceptions of economic pressure (Rothwell & Han, 2010). In contrast, insufficient levels of wealth appear to confer psychological distress and depressive symptoms among adults even after accounting for other SES indicators (Carter et al., 2009; Ettman et al., 2020).

There is limited evidence suggesting that caregiver distress may serve as a pathway linking wealth to adolescent problem behavior. Research finds that wealth is positively associated with family support and easy and clear communication among adolescents and their parents, which may reduce caregiver distress, and may be protective in reducing adolescents’ risk for experiencing problem behaviors (Ramdahl et al., 2018). Furthermore, a recent study using data from the Panel Study of Income Dynamics found that adolescent children from households with both higher incomes and higher wealth had fewer behavioral problems in part because income and wealth reduced caregivers’ psychological and economic distress (Chen et al., 2023).

Educational Attainment.

Educational attainment is another aspect of SES linked to parent distress. Educational attainment may approximate parents’ labor participation, success, and subsequent earnings, insofar that parents with higher levels of educational attainment have access to a wider array of job opportunities with competitive pay (Conger et al., 2010). For example, in 2021, an estimated 32% of jobs in the U.S. required applicants to have a bachelor’s degree or higher (Georgetown University Center on Education and the Workforce, 2025). In turn, parent educational attainment may reduce parental distress as jobs that require postsecondary education provide greater financial stability, autonomy, and work flexibility that may increase job satisfaction and spillover to benefit family relationships (Yucel, 2019). In line with this notion, existing literature finds that education is positively associated with marital stability and satisfaction (Conger et al., 2010) and negatively associated with parental distress (Brännlund & Hammarström, 2014; Fang et al., 2024). Moreover, the psychological benefits of education appear long-lasting. For example, a 27-year longitudinal study of Swedish participants followed from age 16 to 43 years found that higher education was linked to lower levels of psychological distress (Brännlund et al., 2013). This pattern has been replicated in parents. In a systemic review synthesizing risk factors for parent distress, Fang and colleagues (2024) reported that 4 of 6 studies focused on maternal educational attainment and distress found a negative association, with higher levels of maternal distress observed among mothers with fewer years of education. Moreover, work has shown that higher educational attainment may attenuate maternal psychological distress following stressful life transitions (e.g., divorce; Cooper et al., 2009).

Beyond higher-earning potential, several mechanisms have been proposed to explain links between educational attainment and psychological distress. Parents with relatively higher levels of education may have increased access to resources, including social networks, psychosocial resources, and employment resources, higher quality employment opportunities and work characteristics, and more career success, all of which can reduce mental distress (Conger & Donnelan, 2007; Niemeyer et al., 2019; Parkes et al., 2015). For example, a large-scale study in Canada found that parental education was related negatively to parenting stress (McConnell et al., 2011). This may trickle into the parent-child relationship and negatively influence child behavior (McConnell et al., 2011; NICHD SECCYD et al., 2005). These findings underscore the importance of parent education as a potential driver of adolescent behavioral problems via parent distress.

Subjective Measures of SES: Subjective Social Status

Lastly, SSS, or how a parent perceives their place in society relative to others, is a subjective dimension of SES that may be salient for adolescents’ behavioral health by contributing to parent distress. Correlational and experimental studies alike have shown that SSS is negatively related to perceived stress (Adler et al., 2000; Derry et al., 2013; Gruenewald et al., 2006; Tan et al., 2020). Early seminal work by Adler and colleagues (2000) demonstrated that women who ranked themselves lower in SSS reported higher perceived and chronic stress compared to women who rated their SSS higher. Associations between SSS and perceived stress were stronger than those between objective SES and stress (Adler et al., 2000). More recent experimental work has substantiated correlational findings by demonstrating that lower SSS ratings are related to elevated psychological and physiological stress reactivity (e.g., threat, cortisol inflammation) in response to social-evaluative stressors (e.g., interviewing for job, giving a speech; Derry et al., 2013; Gruenewald et al., 2006; Rahal et al., 2020). Taken together, findings suggest that SSS can contribute to psychological distress and take a toll on mental health. Yet, it remains unclear if parents’ SSS ratings can spillover to negatively impact their adolescent’s behavioral health via parent distress.

Current Study Aims

It is crucial to identify how multiple facets of SES relate to parent distress, and in turn adolescent mental health, so that policies can address ongoing economic and mental health disparities more effectively. Several aspects of objective SES, in isolation, have been linked to adolescent behavioral problems through parent distress (Solantaus et al., 2004; Ponnet, 2014). Less is known about how subjective markers of SES, like SSS, operate to increase parent distress and adolescent internalizing and externalizing problems (Garratt et al., 2016). Using a longitudinal design with monthly data collected over a nine-month period, the current study addressed the following question: how do objective (i.e., annual income, monthly income volatility, educational attainment, wealth) and subjective (i.e., SSS) markers of SES uniquely predict adolescent internalizing and externalizing problems through parent distress? Given prior research, we expected all SES measures to have indirect relations to adolescent internalizing and externalizing problems through parent distress. Given the comparative extensiveness of the literature on annual income, SSS, and development, we expected these markers of SES to have stronger indirect effects when compared to income volatility, educational attainment, and wealth.

Methods

Below, we report how the sample size was determined, all data exclusions, all data manipulations, and all measures included in the analyses.

Participants and Study Procedure

The current study used data from the nine-month, longitudinal Family Income Dynamics Study (FInD) designed to assess the role of family financial circumstances in shaping family processes and adolescent mental and behavioral health. It included 104 parent (92% female) and adolescent (55% female) dyads that were recruited in the greater Pittsburgh area from research registries, community organizations, schools, virtual and physical advertisements, and word-of-mouth referrals. Eligibility was established by a preliminary phone screening based on race/ethnicity and annual household income adjusted for household size, which resulted in a diverse sample. The sample size was determined based on several factors, including funding, power, and timing of the study in relation to the COVID-19 pandemic.

Participants were enrolled on a rolling basis beginning in November 2019. Data collection spanned November 2019 through August 2021. Initial interviews were conducted in person, but in March 2020, we began conducting them over the phone after widespread shutdowns due to COVID-19. The remaining eight waves of data were collected via monthly electronic surveys for all participants. The surveys included questions on sociodemographic background, economic circumstances, psychological functioning, parenting practices, and family functioning. The retention rate was 99% across the 9 waves.

At baseline, parents ranged in age from 31 to 69 years old (M = 43.9, SD = 7.0) and adolescents from 13 to 16 years old (M = 14.8, SD = 0.8). Parents were primarily biological mothers (86%) with a subset of biological fathers (7%), adoptive parents (6%), and grandparents (2%). When it came to family economic circumstances, 51% of the families that participated in the study were classified as low-income based on prior year’s income and family size at baseline (i.e., annual income less than 200% of Federal Poverty Income Level) and 49% were classified as middle class. The majority of parents were working for pay; only 8–15 parents reported being unemployed over the nine waves. Of the parents, 62% identified as White, 32% as Black, and 6% as other or multiracial. FInD was designed to be a study of family income dynamics and adolescent mental health, but it was not a clinical sample; 19% of adolescents exhibited clinical levels of behavioral problems (measured by Strengths and Difficulties Questionnaire) on average across the study period.

The study was reviewed and approved by the Human Resource Protection Office, IRB, University of Pittsburgh: protocol 18110075. Written informed consent to participate in this study was given by parents for themselves and for their adolescents. This study was not preregistered; data, study materials, and analytic code are not publicly available but can be provided by request by contacting the corresponding author.

Measures

Objective SES

Annual Income.

Annual income was assessed at baseline. Parents were asked a single-item question regarding their household’s prior year income from all sources after taxes, including earnings, wages, child support, alimony, welfare/TANF, Social Security, government assistance, and investments after taxes. The variable was then log transformed to reduce non-normality (West, 2022) and to reflect nonlinear associations between income and outcomes (e.g., Dearing et al., 2001; Miller et al., 2013). Prior research has found that self-reports of annual income are accurate when compared to individual tax records (Sudman & Bradburn, 1974; Marquis, 1981).

Income Volatility.

Each wave of the study, parents were asked detailed questions about income they received from a variety of sources (17 sources in total), including formal/informal work, foster care payments, Social Security benefits, cash assistance programs, and government assistance packages (i.e., food stamps, SNAP, TANF, gifts, stimulus checks etc.). Income from all sources was summed to create a measure of monthly income, which was then log transformed. Log-transformed monthly income was then used in multi-level models as a time-varying variable. This analytical approach allowed us to look at changes in household income from month to month. Self-reports of monthly income have been shown to be accurate (Moore et al., 2000).

Wealth.

At wave 1, parents reported on the value of three of the most common assets that families hold: savings accounts, stocks and bonds, and homes. We summed the value of these three assets to create a measure of household wealth and log transformed the total to address non-linearities.

Parent Educational Attainment.

Parents reported at baseline whether their highest level of educational attainment was less than a high school degree, high school degree/GED, some college or vocational/technical school, bachelor’s degree or higher. We dichotomized education with an indicator of whether the parent had a bachelor’s degree or higher versus less than a bachelor’s degree.

Subjective SES

Subjective Social Status.

The MacArthur Scale of Subjective Social Status – Adult Version was used to assess SSS in the community among parents at wave 1 (Adler et al., 2000). Participants were asked a single-item question which required them to rank themselves on a ladder with the top of the ladder representing individuals who have the highest standing in the community, and the bottom rung representing individuals that have the lowest standing in the community. The measure was scored from 1 to 10 with 1 being the lowest rung and 10 being the highest rung so that higher numbers indicated higher SSS. Adolescents were also given this same item at wave 1 to report their SSS in the community, which we used as a control variable.

Parent Distress

At each wave, parent distress was measured using the Perceived Stress Scale (Cohen et al., 1994). This 10-item survey asked parents to assess how often in the past month they felt or thought certain ways. Example questions included, “In the last month, how often have you felt nervous or stressed?” and “In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?” All items were answered on a 5-point Likert scale (“Never” to “Very often”). Responses were averaged to create a composite measure of parent distress, with higher values indicating more distress (α=0.91).

Adolescent Internalizing and Externalizing

Every wave, adolescents reported on their internalizing and externalizing problem behaviors using the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997). The SDQ consists of 25 items assessing emotional problems, conduct problems, hyperactivity, peer problems, and prosocial behaviors. The adolescents were presented with questions such as “I get very angry and often lose my temper,” “I am often accused of lying or cheating,” “I am often unhappy, depressed or tearful” and responded on a 3-point scale (“not true” to “certainly true”). Consistent with prior approaches, we created internalizing and externalizing composite variables based on validated subscales (Goodman, 1997). The internalizing composite reflected the sum of emotional and peer problems subscales (10 items; α = .66−.74), and the externalizing composite reflected the sum of conduct problems and hyperactivity subscales (10 items; α = .72−.81). Internalizing and externalizing symptoms were measured continuously. Higher scores indicate worse problem behavior. Although this was not a clinical sample, 4.8% of adolescents had externalizing symptoms above the clinical threshold (score ≥ 11), while 21.1% had clinically elevated internalizing symptoms (score ≥ 9).

Covariates

The analyses controlled for potentially confounding factors that might explain variation in adolescent behavioral problems and be related to family SES and/or parent distress. These included adolescents’ sex at birth (male reference) and race/ethnicity (racial/ethnic minority reference) and parent marital status (non-married reference), all of which were coded as categorical variables. Lastly, we controlled for adolescents’ reports of SSS since it may relate to both family SES and their behavior, but is distinct from parent reports of SSS. All covariates were collected at baseline and treated as time invariant.

Data Analysis

All analyses were run on a stacked panel dataset, wherein each wave of data across the nine waves represented a separate observation, to look at variability within and between individuals. We first ran correlations between our indicators of SES to explore interrelations among the measures. To address our main aim of testing whether these objective and subjective measures of SES were related to adolescent internalizing and externalizing problems through parent distress, we ran mixed effects models with random effects for individual dyads to examine repeated measures (level 1) nested within the parent-adolescent dyads (level 2) using a Bayesian estimator (van de Schoot et al., 2014). The use of repeated observations in a mixed effects framework allows us to account for the non-independence of observations within dyads and across repeated measurements, while also modeling the variability between dyads. The random intercept was included to account for differences in baseline values between adolescents, rather than assuming a single, fixed intercept for all groups.

We modeled the indirect effects of objective and subjective measures of SES on internalizing and externalizing in separate models. The path model estimated internalizing or externalizing problems as a function of parent distress (“b” path), which was predicted with monthly income at level 1 (which captures month-to-month income volatility) and annual income, educational attainment, wealth, and SSS at level 2 (“a” paths). All control variables were included as predictors of behavioral problems, and we controlled for all parent characteristics when predicting parent distress (see Figure 1 for the conceptual model). Indirect effects were calculated as the product of the “a” and “b” paths (Mplus code available in supplement) and Bayesian methods were used to test the statistical significance of indirect effects because prior studies have shown that these effects are typically skewed and thus underpowered (see van de Schoot et al., 2014; Zhao et al., 2010 for detailed explanation).

Figure 1.

Conceptual model

We also ran several sensitivity analyses to test the robustness of our results to different specifications. First, because they are often heavily correlated with SES, we dropped the race/ethnicity and marital status control variables to replicate results without potential colinear factors. Second, we dropped the subjective measures of SES (both parent-reported and adolescent-reported SSS) to examine whether the objective markers had stronger associations with distress and behavior problems without SSS in the model. Lastly, we reran models controlling for household size to verify that observed associations between objective and subjective measures of SES and outcomes were robust to adjustments for the number of people within the home.

Results

Descriptive statistics for the sample are reported in Table 1. Table 2 illustrates the correlations among markers of SES. Overall, correlations ranged from very small to moderate (rs = |.02, .70|). The objective measures of SES were most strongly intercorrelated, with household annual income and wealth showing the highest correlation among our variables at .70 and the other markers having correlations of .50 or lower (rows and columns 3–6). Our subjective SES measure, SSS, was mostly weakly associated with objective SES (column 7). It was not correlated with parental education and correlated with the other measures at around .20–.30, which suggests that objective and subjective SES capture distinct dimensions of socioeconomic context. However, weak to non-significant correlations may also reflect the dichotomous nature of the parental education variable and additional factors, such as household consumption habits and household size. Lastly, Tables 3 and 4 report the coefficients from the path models and indirect effects calculations, which are discussed in depth below.

Table 1.

Descriptive Statistics

	Mean (SD) or %	Range
Adolescent Behavior Problems
Internalizing	3.07 (1.62)	0–8
Externalizing	2.62 (1.66)	0–8.5
Objective SES:
Annual Income	$58,480 ($48,422)	$1,200–$300,000
Income Volatility	$0 ($1,702)	−$7,336–$25,533
Household Wealth	$40,580 ($65,236)	$0–$250,001
Parent Ed.: Bachelor’s +	54.9%	0–1
Subjective SES:
Parent-Report SSS	5.33 (2.31)	1–10
Parent Mental Distress	1.65 (0.71)	0–3.6
Covariates
Parent Relationship Status: Married	47.1%	0–1
Adolescent Race/Ethnicity: White	44.1%	0–1
Adolescent Sex: Female	54.9%	0–1
Adolescent-Report SSS	5.25 (2.29)	1–10

Note: SD = standard deviation; SSS=subjective social status

Table 2.

Bivariate Correlations Among Variables

Variable	1	2	3	4	5	6	7	8	9	10	11	12
1. Annual Income (log)	—	.48***	.70***	−.03	.22***	−.03	−.22***	−.06	−.05	.45***	.29***	.10***
2. Income volatility (log)		—	.41***	.28***	.17**	.00	−.22***	−.01	−.03	.29	.29***	.12**
3. Wealth (log)			—	.55***	.33***	.02	−.16***	−.01	.04	.46***	.24***	.21***
4. Parent Ed.				—	.03	.00	−.14***	−.20***	−.−.04	.59***	.22***	.05
5. Parent-Report SSS					—	.00	−.25***	.02	−.07*	.20**	−.08**	.18***
6. Adolescent-Report SSS						—	−.15***	−.19***	−.20***	.01	−.02	.08*
7. Parent Distress							—	.24***	.24***	−.19***	.07*	−.19***
8.Externalizing								—	.54***	−.13***	−.07*	−.15***
9. Internalizing									—	.01	.09**	.07*
10. Parent Marital Status										—	.22***	.05
11. Adolescent Race											—	.01
12. Adolescent Sex												—

^***p < .001

^**p < .01.

^*p < .05.

^†p < .10

Table 3.

Paths and Indirect Effects from SEM Predicting Adolescent Internalizing Symptoms

	Coeff. (95% CI)
IV to Mediator (a Paths)
Annual Income → Parent Distress	−0.15 [−0.33, 0.04]†
Income Volatility → Parent Distress	−0.04 [−0.08, −0.01] ***
Household Wealth → Parent Distress	−0.02 [−0.02, 0.05]
Parent Ed. → Parent Distress	−0.18 [−0.47, 0.13]
Parent-Report SSS → Parent Distress	−0.07 [−0.14, −0.01] *
Mediator to DV (b Path)
Parent Distress → Internalizing	0.93 [0.44, 1.47] ***
Covariates
Parent Marital Status → Parent Distress	0.11 [−0.18, 0.39]
Adolescent Race → Parent Distress	0.16 [−0.08, 0.39]†
Adolescent Sex → Parent Distress	−0.25 [−0.47, −0.01] *
Annual Income → Internalizing	−0.07[−0.52, 0.39]
Income Volatility → Internalizing	0.01 [−0.05, 0.06]
Household Wealth → Internalizing	0.03 [−0.06, 0.13]
Parent Ed. → Internalizing	−0.20 [−0.94, 0.50]
Parent-Report SSS → Internalizing	0.01 [−0.16, 0.17]
Parent Marital Status → Internalizing	0.06 [−0.59, 0.73]
Adolescent Race → Internalizing	0.16 [−0.44, 0.70]
Adolescent Sex → Internalizing	0.56 [−0.03, 1.10] *
Adolescent-Report SSS→ Internalizing	−0.15[−0.31, 0.01] *
Indirect Effects
Annual Income → Parent Distress→ Internalizing	−0.13 [−0.35, 0.04]†
Income Volatility → Parent Distress → Internalizing	−0.04 [−0.09, −0.01] ***
Household Wealth → Parent Distress → Internalizing	0.01 [−0.03, 0.05]
Parent Ed.→ Parent Distress → Internalizing	−0.15 [−0.51, 0.13]
Parent-Report SSS → Parent Distress → Internalizing	−0.07 [−0.15, −0.01] *
Total Effects
Annual Income → Internalizing (ab + c′)	−0.20 [−0.69, 0.29]
Income Volatility → Internalizing (ab + c′)	−0.03 [−0.10, 0.04]
Household Wealth → Internalizing (ab + c′)	0.04 [−0.06, 0.14]
Parent Ed. → Internalizing (ab + c′)	−0.35 [−1.13, 0.43]
Parent-Report SSS → Internalizing (ab + c′)	−0.06 [−0.24, 0.12]

Note: N=907; covariates were included in all paths;

^***p < .001

^**p < .01.

^*p < .05.

^†p < .10.

CI=confidence interval; DV=dependent variable; IV=independent variable; SSS= subjective social status.

Table 4.

Paths and Indirect Effects from SEM Predicting Adolescent Externalizing Symptoms

	Coeff. [95% CI]
IV to Mediator (a Paths)
Yearly Income → Parent Distress	−0.15 [−0.33, 0.41]†
Income Volatility → Parent Distress	−0.04 [−0.08, −0.01] ***
Household Wealth → Parent Distress	−0.02 [−0.02, 0.05]
Parent Ed. → Parent Distress	−0.18 [−0.48, 0.13]
Parent-Report SSS → Parent Distress	−0.07 [−0.14, −0.01] *
Mediator to DV (b Path)
Parent Distress → Externalizing	0.70 [0.17, 1.28] **
Covariates
Parent Marital Status → Parent Distress	0.11 [−0.18, 0.40]
Adolescent Race → Parent Distress	0.16 [−0.08, 0.39]†
Adolescent Sex → Parent Distress	−0.25 [−0.47, −0.01] *
Annual Income → Externalizing	0.07 [−0.42, 0.57]
Income Volatility → Externalizing	−0.03 [−0.09, 0.01]
Household Wealth → Externalizing	0.06 [−0.04, 0.16]
Parent Ed. → Externalizing	−0.62 [−1.43, 0.13]†
Parent-Report SSS→ Externalizing	0.10 [−0.08, 0.27]
Parent Marital Status → Externalizing	−0.20 [−0.90, 0.53]
Adolescent Race → Externalizing	−0.25 [−0.89, 0.34]
Adolescent Sex → Externalizing	−0.37 [−1.03, 0.22]
Adolescent -Report SSS → Externalizing	−0.16 [−0.33, 0.01] *
Indirect Effects
Annual Income → Parent Distress → Externalizing	−0.10 [−0.28, 0.03]†
Income Volatility → Parent Distress → Externalizing	−0.03 [−0.07, −0.004] **
Household Wealth → Parent Distress → Externalizing	0.01 [−0.02, 0.04]
Parent Ed. → Parent Distress → Externalizing	−0.11 [−0.43, 0.10]
Parent-Report SSS → Parent Distress → Externalizing	−0.05 [−0.13, −0.003] *
Total Effects
Annual Income → Externalizing (ab + c′)	−0.03 [−0.54, 0.48]
Income Volatility → Externalizing (ab + c′)	−0.06 [−0.12, 0.003]†
Household Wealth → Externalizing (ab + c′)	0.07 [−0.03, 0.17]
Parent Ed. → Externalizing (ab + c′)	−0.73 [−1.54, 0.08]†
Parent-Report SSS → Externalizing (ab + c′)	0.05 [−0.13, 0.23]

Note: N=907; covariates were included in all paths;

^***p < .001

^**p < .01.

^*p < .05.

^†p < .10.

CI=confidence interval; DV=dependent variable; IV=independent variable; SSS= subjective social status.

Indirect Effects of SES on Adolescents’ Internalizing Via Parent Distress

Results showed some significant indirect effects of objective and subjective SES on adolescent internalizing symptoms operating through parent distress (Figure 2; Table 3). First, there was a negative indirect effect of income volatility (i.e., within-dyad variation in monthly income) on adolescents’ internalizing symptoms through parent distress (ab path; coeff. = −.04, p < .001; Table 3). Specifically, when parents reported a higher monthly income than was typical for their household, they reported less distress (a path; coeff. = −.05, p < .001). A log unit increase in monthly income (roughly 2.7 times prior month’s income) was related to a decrease in parental distress of .07 SD. Distress was in turn positively related to adolescent internalizing symptoms (b path; coeff. = .96, p < .001), which equated to a .83 SD increase in internalizing per SD increase in distress. Thus, increases in monthly income predicted less parental distress, and less distress predicted fewer internalizing problems. Parents’ SSS was also negatively related to adolescent internalizing symptoms through parent distress (ab path; coeff. = −.06, p = .021), as parents’ SSS was negatively related to their level of distress (a path; coeff. = −.06, p = .021). For every SD increase in -SSS, distress decreased by .04 SD. There were no indirect effects of household annual income, wealth, or parent educational attainment on adolescents’ internalizing symptoms through distress. These SES indicators were not significantly predictive of parent distress (Table 3).

Figure 2.

Path Model Linking SES to Internalizing Via Parent Distress

As shown in Table 3, there were no direct effects of objective SES indicators and subjective SES on adolescent internalizing symptoms. The only covariate associated with parent distress was sex at birth; female adolescents had parents that were, on average, less distressed than parents of male adolescents. Female adolescents tended to have higher levels of internalizing symptoms compared to males. There was also a negative association between adolescent SSS and their self-reported internalizing symptoms (see Table 3).

Indirect Effects of SES on Adolescents’ Externalizing Via Parent Distress

Results for externalizing symptoms generally mirrored those for internalizing problems (Figure 3; Table 4). Income volatility had a negative indirect effect on adolescents’ self-reported externalizing symptoms through parent distress (ab path; coeff. = −.03, p = .007). As stated above, parents reported less distress in months when their income was higher than typical for their household (a path; coeff. = −.05, p < .001). Distress was positively related to adolescent externalizing symptoms (b path; coeff. = .73, p = .007). For every SD increase in parental distress, adolescent externalizing was .62 SD higher. Thus, higher monthly income related to fewer externalizing problems due to its link to lowered parental distress. There was also a negative indirect effect of SSS on externalizing problems through parent distress (ab path; coeff. = −.04, p = .028). As discussed above, SSS was negatively related to distress (a path; coeff. = −.06, p = .021), and distress predicted more adolescent externalizing symptoms (b path; coeff. = .73, p = .007). As with internalizing, there were no indirect effects of household annual income, wealth, or parent educational attainment on adolescent externalizing behaviors via distress.

Figure 3.

Path Model Linking SES to Externalizing Via Parent Distress

Even with income volatility and parent SSS having significant indirect effects on adolescent externalizing through parent distress, both maintained positive direct effects on adolescents’ externalizing behaviors (Table 4). In addition, after accounting for the role of parent SSS on parent distress, there was also a positive direct effect between parents’ SSS and adolescent externalizing symptoms. As with internalizing symptoms, the only covariate to be associated with parent distress was sex at birth, such that parents of female adolescents reported comparatively less distress; though, adolescents’ sex at birth was not related to their externalizing symptoms. There was also a negative association between adolescent SSS and externalizing symptoms. No other markers of SES or covariates were significant (Table 4).

Sensitivity Analyses

Results of sensitivity analyses are included in the supplemental material (Tables A–F). Our results were replicated across all models. Importantly, the associations between objective markers of SES—annual income, wealth, and parental education—and parent and adolescent functioning that were non-significant when SSS was in the models remained nonsignificant (Tables B and D). Thus, it does not appear that objective SES predicts adolescent problem behavior through SSS.

Discussion

The present study adds to prior literature by demonstrating that both objective and subjective SES uniquely and independently relate to adolescent children’s behavioral health by contributing to differences in parent distress. Namely, we found that month-to-month income loss was related to increases in parent distress and greater SSS reduced parent distress, both of which were associated with adolescent problem behaviors. Moreover, indirect effects were independent of annual income, household wealth, and parent educational attainment, as well as related sociodemographic factors of adolescent sex, age, race, and parent marital status. In contrast to prior studies and our hypotheses, however, we failed to find support for parent distress as a pathway between household annual income, household wealth, parents’ educational attainment, and adolescents’ internalizing and externalizing symptoms.

Given the theoretical tenets and extensive empirical support for the FSM described and reviewed by Masarik and Conger (2017), we had hypothesized that parent distress would serve as an important pathway linking multiple objective (i.e. parental income, income volatility, education and wealth) and subjective measures of SES (i.e., parent SSS) to adolescent problem behaviors. While our exact hypothesis was not supported, findings highlight the importance of considering the psychological and temporal proximity of different aspects of SES as they relate to parent distress. Specifically, SSS and income volatility were independently related to parent distress, while parent educational attainment, household wealth, and annual household income were not; although, the coefficients for annual income and parent educational attainment were in the expected direction. Although not entirely clear, it is possible that SSS may be more psychologically proximal, and income volatility more temporally proximal to parents’ distress levels compared to annual income, household wealth, or parent educational attainment (Hoebel & Lampert, 2020). As SSS is thought to represent parents’ own internalization of objective measures of their SES as well as their perceived power, privilege, and social acceptance relative to others in their community (Anderson et al., 2012), parents’ distress levels may be intimately tied to their SSS and their perceived ability to provide for their child. For example, among mothers of young children, Michelson and colleagues (2016) observed that mothers’ SSS was negatively related to psychological distress (e.g., depressive symptoms, lack of social support) independently of maternal education.

In the same vein, income volatility may be more temporally proximal than comparatively more distal measures of objective SES, like annual income, household wealth, or parent educational attainment. That is, how much income parents bring in each month (and how much this amount deviates from their typical monthly income) may be more salient to their level of distress compared to how much they bring in over the course of the entire year or how much money they have in savings, stocks, or home equity (Bania & Leete, 2022). For example, Prause and colleagues (2009) found that income loss (but not annual income) independently related to the onset of depression among adults. This is not to say that parents’ educational attainment, annual income, or household wealth do not contribute to distress, but rather that these indicators of objective SES may not be at the forefront for parents relative to income volatility or SSS unless they are, for example, applying for a job where level of education is part of the application, they are applying for a loan and are asked about the value of their assets, or they are negotiating their salary. However, it is also possible that different dimensions of SES may interact with one another to magnify or attenuate parent distress and subsequent problem behaviors among adolescent children. For example, in addition to finding a main effect of income loss on depression onset, Prause and colleagues (2009) found that annual income moderated the extent to which income loss increased risk for depression. Specifically, a loss in income increased the risk for depression among low- versus high-income adults. This may suggest that a loss of a few hundred dollars from month-to-month may be negligible or only modestly stressful for an upper-income family but may be a major source of economic strain for a low-income family already struggling to cover expenses. Similarly, Conger and colleagues (2010) theorize that high parental educational attainment may serve as a buffer against parent strain following economic shocks, such that more highly educated parents may feel a sense of reassurance that they will be able to find another well-paying job given their level of education. Future research should therefore examine whether associations between income volatility, parent distress, and adolescent internalizing and externalizing symptoms are magnified among low-income or low-wealth households or among parents with low levels of educational attainment, which are tests of moderated mediation that we were not powered to assess here given our sample size.

When considering objective measures of SES, parent distress partly mediated the relation between income volatility and adolescent problem behaviors, but not for annual income, household wealth, or educational attainment. Findings extend the current work linking income volatility to adolescent behavior (e.g., Cheng et al., 2020; Gennetian et al., 2015; Lee et al., 2019; Miller et al., 2021; Miller & Votruba-Drzal, 2017) by demonstrating that heightened parent distress may be a proximal pathway explaining the association. In a household struggling to make ends meet, a decrease in monthly income may increase parents’ distress as they are forced to cut back on spending (e.g., on leisure activities, groceries, or any non-essentials), choose which bills to pay and not pay, or pick up additional work. In turn, month-to-month losses in income may give rise to disagreements about money and hostile/harsh interactions between parents and their adolescent children that add to parent distress (Conger et al., 1994). Results are especially timely given today’s gig economy. According to a recent McKinsey American Opportunity Survey, as many as 36% of U.S. adults work non-standard, gig jobs like food delivery and rideshare services (Dua et al., 2022). Although gig jobs offer the benefit of flexibility and autonomy with setting one’s schedule, which could plausibly alleviate stress (Yucel, 2019), income volatility is also often inherent to these types of positions (Auguste et al., 2024; Reynolds & Kincaid, 2023). While the current study did not inquire specifically about gig work, 23% of employed participants reported weekly changes in work schedules. Such financial unpredictability and volatility in earnings may explain why gig workers have reported greater psychological and financial distress and poorer mental health compared to salaried workers (Freni-Sterrantino & Salerno, 2021; Glavin & Schieman, 2022). However, future research is needed to examine whether the distress of working in a financially volatile job like gig work trickles down to negatively impact employees’ children.

Interestingly, we observed a direct, negative association between income volatility and adolescent externalizing behaviors in our additional analyses excluding race and marital status (c path; coeff = −.07, p <.001) and parents’ and adolescents’ SSS (c path; coeff = −.06, p = .024). This direct association may indicate that income volatility impacts adolescent externalizing behaviors through other mechanisms beyond parent distress. One such pathway could be through parental investments of time and money (Zhang, 2021). For example, in addition to contributing to less parent distress, when a family’s monthly income is higher than usual for them, parents may have more disposable income to put towards activities with their adolescent children. Being able to invest more time and money into these activities may, in turn, reduce adolescents’ frustration and subsequent externalizing symptoms (Jiang & Shi, 2024). In contrast, when families experience a drop in their monthly income, parents may need to respond by cutting back on non-essential activities (e.g., family outings, giving adolescent children spending money for activities with friends), which could give rise to aggressive or deviant behavior or internalizing symptoms among adolescents who are not able to engage in the activities that they may have been able to do previously (Plenty & Mood, 2016). Indeed, recent work has shown that in months when adolescents report more financial hardship, they endorse more externalizing symptoms (Miller et al., 2024). Then again, this direct effect did not reach conventional significance in the race- and marital status-adjusted model, so direct associations should be interpreted cautiously.

With respect to subjective SES, this is one of the first studies to show that parents’ distress partly explains why parents’ SSS is related to their adolescent children’s internalizing and, to a lesser extent, externalizing symptoms. In this study, parents who rated themselves lower in SSS compared to others in their communities endorsed more distress, which was in turn related to higher internalizing and externalizing symptoms among their adolescent children. Findings extend prior work that has shown separate associations between SSS, distress (Derry et al., 2013; Michelson et al., 2016; Zell et al., 2018), and stress-related conditions (Adler et al., 2000; Hoebel & Lampert, 2020; Zell et al., 2018), and between parents’ distress and their adolescent children’s anxiety and depressive symptoms (Masarik & Conger, 2017) by testing the full mediated pathway. Findings also extend work by Roy and colleagues (2019), which showed that lower parental SSS relates to greater behavioral problems among young children by way of parent distress even when accounting for objective SES measures of household income-to-needs ratio, and parents’ educational attainment and employment. Our collective findings show that subjective and objective SES indicators, although related, appear to capture distinct experiences of status that are uniquely related to parent distress and subsequent behavioral problems among children and adolescence. For example, in our study, parent SSS was weakly correlated with parental educational attainment and household annual income (rs = .026 and .216, respectively). Interestingly, Roy and colleagues (2019) also failed to find support for an indirect effect of income-to-needs ratio on children’s behavior through distress, which they attributed to maternal education mattering more for young children’s behavioral development. In accordance with their reasoning, Roy and colleagues found that parental educational attainment had an indirect effect on young children’s behavior through stress, which we did not find in our adolescent sample. However, this is not completely surprising as some research suggests that associations between maternal education and adolescent behavioral health are the strongest in early childhood and wane as children age (Meyrose et al., 2018).

Furthermore, our findings showed that the indirect effects of parent SSS on adolescent problem behaviors through parent distress held even when accounting for adolescent children’s own SSS, which is an important predictor of adolescent problem behaviors (Steen et al., 2020). Interestingly, parents and their children’s SSS ratings were only modestly correlated with one another (r = .219), which could suggest that these ratings are informed by different factors. For example, it is possible that parent SSS is more closely tied to feelings of power, prestige, and respect, whereas adolescents’ SSS may be more closely linked to a sense of belonging and acceptance (Destin et al., 2012). Thus, future work should interrogate what social and economic factors inform parent and adolescents’ SSS ratings, and how the level of concordance, or lack thereof, between parent and adolescent’ SSS ratings relates to adolescents’ problem behaviors. Nevertheless, both adolescent and parent SSS were negatively and independently related to adolescent internalizing and externalizing symptoms, which highlights the need for additional work investigating adolescents’ own perspectives on their family’s social status and financial circumstances, and the impact that this perspective has on their behavioral health (Miller et al., 2024).

Strengths, Limitations, and Future Directions

The present study has several strengths, including the inclusion of multiple measures of objective (i.e., income volatility, yearly income, parent educational attainment) and subjective SES that were included simultaneously in path models to examine their independent contributions to parent distress and, in turn, adolescent problem behaviors. Empirical support for the FSM has often used household annual income or a composite based on annual income and educational attainment to measure SES. However, by looking at the relative contributions of each objective SES measure separately (versus creating a composite) and considering subjective experiences of SES, we were able to investigate how multiple aspects of SES come to be associated with adolescent problem behaviors by contributing to parents’ stress levels.

Results may provide important information for interventions aimed at reducing socioeconomic inequalities in adolescent problem behaviors. Findings are further strengthened by the inclusion of various demographic controls that could contribute to links between parent distress and adolescent problem behaviors, such as parents’ relationship status and adolescents’ age, sex, race, and their own SSS. Another strength is that analyses were conducted on a sample of socioeconomically diverse parent-adolescent dyads, several of which are not part of traditional two-parent households but rather multigenerational households (e.g., grandparents raising grandchildren), which may be more common in lower SES households. Analyses also capitalized on parent and adolescent reports. Parents reported on objective and subjective measures of SES and their own distress while adolescents reported on internalizing and externalizing symptoms. By leveraging both individuals’ perspectives, we reduced the potential for reporting bias that may have occurred if parents projected their own mental health symptoms onto their adolescents when reporting on their child’s problem behaviors (e.g., Fjermestad et al., 2017).

Nonetheless, this study is not without limitations. Foremost, results are correlational, so causation cannot be determined. For example, the association between parent distress and adolescent problem behaviors has been shown to be bidirectional where parent distress may lead to adolescent problem behaviors, which in turn could be a source of added distress among concerned parents (Kochanova et al., 2021; Solantaus et al., 2004). Future research could employ a cross-lagged model design to rule out potential bidirectionality. Additionally, we did not find significant total effects between objective and subjective measures of SES and adolescent externalizing and internalizing symptoms. Although indirect effects may be observed in the absence of significant total effects (O’Rourke & MacKinnon, 2018), it is possible that our results may reflect the influence of a possible unmeasured confound, such as parent’s negativity bias.

There were also some limitations to the measures captured by our data. A general measure of perceived stress was used to assess parent distress, rather than examining socioeconomic- or parenting-specific stress, or parent psychopathology. However, the Perceived Stress Scale has been widely used in the literature and has been shown to be moderately correlated with financial strain as well as depressive and anxiety symptoms among adults (Lee, 2012; Park et al., 2017). We also focused on parental distress as one pathway linking various measures of SES to adolescent internalizing and externalizing symptoms; however, more proximal substantiations of economic disadvantage, like material deprivation, financial hardship, and food insecurity could be other pathways through which SES comes to be associated with adolescent problem behaviors (Duncan & Magnuson, 2002; Gershoff et al., 2007). While data on these constructs are available in the FiND sample, we are limited in our power to be able to adequately test these pathways simultaneously or in a serial path model with parent distress. Second, there are other indicators of SES that were not measured (e.g., occupational prestige, accumulated debts) in the current study that could contribute to parental distress and subsequent problem behaviors among adolescents. We also measured SSS once. While prior research suggests that SSS ratings are relatively stable over time (Goodman et al., 2015; Niu et al., 2023, Operario et al., 2003), decreases in SSS have been linked to concomitant increases in negative affect among midlife adults (Weiss & Kunzmann, 2020). Therefore, future research should consider how changes in SSS may relate to changes in parental distress, especially for parents who experience decreases in their SSS following the loss of their income or employment. Also, due to limited variability in our parental education measures resulting from FInD’s sampling frame of low- and middle-income families, we had to dichotomize parental education. Lastly, although we were able to draw on repeated measures of monthly income collected over multiple months to assess income volatility, future work could consider how income volatility over longer periods of time or at earlier stages of development may prospectively relate to adolescent problem behaviors (Zhang et al., 2024).

A final limitation is that much of our data collection overlapped with the COVID-19 pandemic. For context, approximately 40% of participants completed the baseline interview prior to March 2020 (specifically 12 dyads in November 2019, 11 in December, 10 in January, and 10 in February), Thus, most participants had five waves of data collection pre-COVID (data were collected every month for 9 months), while 60% of our sample completed all surveys during the widespread shutdown. Due to the unique economic and psychosocial circumstances created amidst the pandemic, it is possible that findings may differ for data collected pre- vs. post-pandemic (Browne et al., 2021; Donker et al., 2021; Patel et al., 2020). For example, household income may have been negatively impacted by job loss/reductions in hours, closures, or reductions in economic activity. On the other hand, COVID-19 safety policies may have lessened the burden of financial hardship for many families through stimulus checks, better unemployment benefits, and eviction moratoriums (Leifheit et al., 2021; Ruffini & Wozniak, 2021). Furthermore, families’ perceptions of hardship and security were most likely remolded by pandemic living. However, it is important to highlight that by collecting and analyzing longitudinal data monthly, our study was able to measure and control for variation in our key constructs before and during the height of the pandemic. Moreover, the timing of data collection did not predict to our SES indicators (including income volatility) or change patterns of findings (Miller et al., 2024). Although we were able to control for variation, it is still important to highlight potential impacts a global pandemic had on families both financially and emotionally.

Conclusion

Socioeconomic disadvantage increases the risk for problem behaviors among adolescents. Support for this pathway has typically centered on parent distress as a mediator between family socioeconomic circumstances. The present study adds to this literature by considering the importance of parent distress in explaining how other objective SES measures beyond annual income (like household income volatility) as well as subjective SES (i.e., parent SSS) shape adolescent problem behaviors. Indeed, parent distress helped explain how household income volatility and parent SSS are related to adolescent problem behaviors. Findings highlight that temporally and psychologically proximal socioeconomic factors (namely, having more or less money at the end of the month or how parents see themselves in relation to others in their communities) may impact parent distress levels to a greater degree compared to more distal objective measures of SES like educational attainment and annual income. Once replicated in other samples, findings could inform stress-reduction interventions aimed at reducing risk for adolescent problem behaviors focused on financial stability or making parents feel valued, respected, and supported by their communities.