Behavioral Academic Avoidance as an Indirect Partial Mediator of the Association Between Chronic Family Stress and University Grades

Abstract

Background

American university students who self-report higher levels of family stress have lower grade point averages and are less likely to remain enrolled through graduation. Existing research seeking to identify mechanisms that might explain this association has focused primarily on students’ psychological adjustment (eg, levels of depression, anxiety, self-efficacy). In this study, we sought to explicate the family stress – academic grade association by focusing on a behavioral explanation, considering the extent to which students respond to academic difficulties by demonstrating avoidance behaviors specific to their academic experiences.

Methods

Undergraduate students (N = 597) completed self-report surveys during Fall 2021: a measure of chronic family stress at the start of the semester, a measure of behavioral academic avoidance mid-semester. Students provided consent for researchers to access their official grade point averages (GPAs) from the university. Model fit and mediation were tested using structural equation modeling with bias-corrected bootstrapping.

Results

An indirect-only mediation effect was present whereby higher levels of chronic family stress at the start of the academic semester predicted higher levels of behavioral academic avoidance midterm which then predicted lower cumulative GPA.

Conclusion

Students who experience higher levels of family stress may struggle to engage in the behaviors that support their academic success: attending class, completing assignments, and maintaining communication with instructors. Effective support of student success in the postsecondary environment requires not only encouraging these behaviors but also helping students to successfully manage the stress they experience within their family environments.

Keywords

college university undergraduate students family stress academic avoidance grade point average

Students do not experience educational successes and failures within a vacuum. Instead, at all levels, they negotiate academic settings and tasks in the context of variability in the extent to which their families of origin constitute sources of stress or support. Indicators of family stress versus support (eg, family conflict, parental support, quality of relationships with family members) are predictive of trajectories of academic achievement.^1–3 At the postsecondary level (tertiary level, school enrollment after high school graduation; heretofore referred to as “college,” “university,” and “postsecondary enrollment”), students who self-report higher levels of family stress have lower grade point averages (GPAs⁴) and are less likely to remain enrolled through graduation. Students who remain enrolled and attain a four-year postsecondary degree reap economic and psychological benefits, both in the United States⁵ and across the globe.⁶ Moreover, the cost of failing to complete a university degree adversely impacts not just individuals, but society. A single cohort of students who fail to attain their four-year degrees has been estimated to cost the United States $3.8 billion in lost income, $566 million in lost federal income taxes, and $164 million in lost state income taxes.⁷

Experiences of family stress are just one of a myriad of explanations for lower postsecondary GPAs and higher rates of university leaving but are important to consider when thinking about ways to support students. Such consideration includes identifying potential mechanisms that might explain why students who experience higher levels of family stress struggle academically. Current research seeking to identify such mechanisms has focused primarily on students’ psychological adjustment (eg, levels of depression, anxiety, self-efficacy^3,8) as explaining the family stress – postsecondary GPA relationship. In the current project, we elected to focus not on potential psychological mediators, but on behavioral ones. Specifically, we asked whether the association between levels of chronic family stress and postsecondary GPAs might be explained by the extent to which students respond to academic difficulties by demonstrating avoidance behaviors specific to their academic experiences – behaviors that further lower grades and accelerate the pathway to college leaving.⁹

Post-Secondary Educational Success in the Contemporary United States: Benefits and Predictors

Successful completion of a four-year college degree in the contemporary United States predicts a myriad of occupational and economic benefits.¹⁰ Yet despite recognizing potential benefits that a four-year degree offers, nearly 40% of matriculating first-year students fail to obtain a college diploma within six years of enrollment.¹¹ Students who persevere to degree completion have higher grade point averages (GPAs) as early as the first semester in college.¹² Both college GPA and rates of leaving college without graduating (henceforth, “college leaving”) vary systematically based on demographic characteristics. Factors such as socioeconomic disadvantage, racial/ethnic minority status, and being a first-generation college student (defined as a student whose parents have not earned a four-year college degree) are highly intercorrelated and associated with both lower GPAs and higher rates of college leaving. Demographic differences in college GPAs and rates of leaving college are best understood in terms of the wide range of indicators of adversity that disproportionately impact these students. For example, Karimshah et al¹³ reported that 39% of students from economically disadvantaged backgrounds have considered dropping out of college due to adversity experiences that include financial problems, health problems, family issues, and relationship issues. With an eye toward improving postsecondary graduation rates, it is critically important to both identify specific types of adversity experiences that place some American college students at academic risk, as well as specify the mechanisms that link adversity experiences with academic grades.

Conceptual Framework

The focus of the current study is consistent with components of the Conceptual Utility Model for the management of Stress and Psychological Wellbeing (CMMSPW¹⁴) which is itself situated within the Theory of Self vs Externally Regulated Behavior (SR-ER).¹⁵ CMMSPW focuses on explaining the impact of stress experienced by college students and its effects on behavior and academic success. Consistent with the CMMSPW model, we propose that experiences of chronic family stress function as a contextual presage variable that has the potential to support or undermine students’ abilities to engage in the self-regulatory behaviors that are critical to academic success. Specifically, we propose that associations between chronic family stress and GPA are mediated by student academic engagement, a procedural subcompetency in the context of the CMMSPW model. When levels of family stress erode students’ abilities to self-regulate in a manner that supports engagement, they are likely to demonstrate avoidance behaviors within the academic context. In turn, it is this avoidance that predicts lower GPA.

Chronic Family Stress as a Predictor of Psychological and Academic Adjustment Among College Students

In the current study, we elected to focus on one potential source of chronic stress in the lives of American college students, with chronic stress defined as ongoing, non-acute life strain students face within specific life domains.¹⁶ Chronic stress is distinct from stressful life events, which are acute, typically non-recurring experiences that begin at a specific moment in time.¹⁷ College students experience chronic stress across a range of domains, including financial stress, stress in relationships with peers, and stress associated with health and medical issues. However, for the purposes of this study we chose to focus on family chronic stress, the level of strain versus support in immediate family relationships, characterized by such factors as accessibility, closeness, trust, acceptance, perceived emotional support, frequency of conflict, and quality of conflict resolution between students and their parents or caregivers, as well as siblings.¹⁸ Our decision to focus on chronic family stress was due to the critical role family relationships play in promoting positive adjustment and well-being across the life course beginning well prior to college enrollment and extending across college enrollment and beyond.¹⁹ Interestingly, college students’ experiences related to chronic family stress are less understood than experiences related to specific stressful life events and early life adversity,^16,20 in large part due to challenges associated with the assessment of chronic stress exposure (the life circumstances themselves) as opposed to more subjective perceptions of the chronic stress, which are part of the stress response.^21,22

Although the impact of chronic family stress on college students has been somewhat understudied, there is a large and compelling literature indicating that its components (eg, lack of family support, family conflict, problematic relationships with parents and other family members) are both associated with and predictive of a range of indicators of well-being among young adults enrolled in postsecondary institutions. College students who report lower levels of parental support score higher on measures of depression and anxiety.⁴ Lower scores on measures of family functioning that include indicators of family support and family agreement are associated with a higher incidence of eating disorders among college students.²³ Higher levels of conflict both between parents of college students and between parents and students themselves are associated with higher levels of student anxiety, depression, and suicidal rumination, as documented in countries such as the United States and China.^24–26

Links between stressful family relationships and college GPA have been less frequently assessed but indicate that less stressful relationships between college students and their parents (and between parents of college students) are associated with higher GPAs. Higher levels of retrospectively reported family conflict prior to college enrollment are associated with lower GPAs among first-semester Asian American college students.¹ Cheng et al² documented that higher levels of student-reported family social support predict higher and more stable GPAs across three semesters of college, with the greatest benefits of family support experienced by female students. A composite measure of parent-student relationship quality (including components of parental affection, emotional support, and encouragement of independence) was associated with higher GPAs among both Asian American and European American college students.³ A stronger attachment to parents predicts higher GPAs for White and Hispanic first year college students.²⁷ A large and compelling literature also links an authoritative parenting style (a critical component of which is high levels of parental support/responsiveness) with higher grades in college.²⁸

Behavioral Academic Avoidance as a Potential Mediator of Associations Between Chronic Family Stress and Postsecondary GPA

Researchers seeking to identify mechanisms explaining why indicators of family stress (eg, low levels of support, insecure attachments, high levels of conflict) are linked with lower GPAs in college have focused primarily on students’ levels of internalized stress. Stressful family environments and experiences are linked with unhealthy patterns of stress reactivity²⁹ and lower self-esteem,³⁰ which in turn predict higher levels of depression. Higher levels of depression may then be the more proximal cause of low GPA. This possibility is confirmed by the findings of Deng et al,⁸ who reported that higher levels of both academic and family stress led to higher levels of depression among college students, which in turn negatively impacted academic performance. However, Bahrassa et al¹ reported that among first-semester Asian American college students, negative associations between family conflict and GPA were not mediated by levels of psychological distress.

Other researchers seeking to identify explanatory mechanisms linking family stress with college GPA have focused not on internalized distress, but rather on students’ feelings of efficacy and their overall adjustment to the postsecondary environment. For Asian American college students (but not European American college students), Yuan et al,³ found that the relationship between parent-student relationship quality and GPA was mediated by students’ feelings of self-efficacy; students who reported more positive relationships with parents felt more self-efficacious, and students higher in self-efficacy in turn earned higher grades. Yazedjian et al²⁷ found that the positive association between attachment to parents and college GPA was mediated by scores on a measure of overall adaptation to college; students with more secure attachments to parents adapted better to college, which in turn predicted higher GPAs.

To date, researchers have tended not to focus on the specific behaviors students at risk for low academic grades might engage in, to the detriment of their academic success. We posit that because stress in general is associated with behavioral avoidance,³¹ chronic family stress specifically is likely to prospectively predict behavioral avoidance; in an academic context, this is likely to take the form of academic avoidance. Successful postsecondary students are actively engaged in the learning process, with this engagement reflected in behaviors such as attending class, participating in optional learning activities, turning in assignments on time, and communicating with instructors. Engaging in these specific behaviors is associated with higher grades in college. In contrast, lower grades are predicted by behaviors indicative of academic avoidance (eg, avoiding instructors, not attending class meetings, using cell phones in class, not paying attending when in class).^9,32–34 It is this second group of behaviors that are the focus of the current study.

The Current Study

The purpose of the current study was to examine academic avoidance behaviors as a mediator of associations between college students’ experiences of chronic family stress and their GPAs. We hypothesized that:

College students who report higher levels of chronic family stress at the start of an academic semester will have lower cumulative GPAs during their final semester of attendance.

Associations between levels of chronic family stress and cumulative GPA will be mediated by the extent to which students engage in academic avoidance behaviors. Specifically, students reporting higher levels of family stress will engage in more academic avoidance, and higher levels of academic avoidance will then predict lower GPAs.

Method

Participants

Undergraduate college students enrolled at a medium sized university in the southeastern United States completed surveys at two timepoints during Fall 2021. At the beginning of the semester (Time 1), 336 students completed an online survey, of whom 174 also completed a mid-semester survey (Time 2). An additional 261 students completed only the mid-semester survey, yielding a total N of 597 college students participating in at least one timepoint. Students who participated in the mid-semester survey gave their consent for investigators to access their GPAs from administrative records; 71.69% of the sample (n = 428) had valid data on the GPA outcome. Students self-reported their race/ethnicity as Black or African American (n = 210; 35.2%), Hispanic, Latinx, or Spanish Origin (n = 85; 14.2%), White (n = 203; 34.0%), multiracial or biracial (n = 38; 6.4%), and from another racial/ethnic group (n = 61; 10.2%). Participants self-reported their gender identity as female (n = 448; 75.0%), male (n = 128; 21.4%), and non-binary or another gender identity (n = 19; 3.2%), with two participants not responding to the gender identity question. Students reported a mean age of 19.25 (SD = 3.28). Slightly below half (n = 267; 44.7%) were first-generation students, defined by the United States Higher Education Act (1965/1998) as not having a parent who had completed a four-year college degree.

Measures

Chronic Family Stress

Student experiences of chronic family stress were assessed by student completion of the 15-item family relationship domain subscale of the Chronic Life Stress Questionnaire (CLSQ) at the start of the Fall 2021 semester (Time 1). The CLSQ³⁵ focuses on assessing individuals’ experiences of stressful life circumstances or experiences, distinct from subjective emotional responses to those circumstances. Participants responded to all items on a scale ranging from 1 (lowest levels of family relationship stress) to 5 (highest levels of family relationship stress) over the most recent 6 months, with response options specific to each item. The family relationships domain subscale combined across items assessing closeness to parents/caregivers (7 items; sample item “How much have you felt that you could count on your parent/guardian/caregiver to do the things they say they will do”), conflict with parents/caregivers (3 items; sample item “How often have you argued with your parent/guardian/caregiver? By “argued” we mean that you had an emotional conflict, not just a debate or discussion”), and aspects of relationship quality focused on relationships with family members other than parents/caregivers (5 items; sample item “How honest have you felt that other members of your immediate family have been with you?”). Cronbach's alpha for the three subscales were α = .902, .777, and .714, respectively. Cronbach's alpha for the full scale with 15 items was α = .907. Items comprising the closeness, conflict, and relationships quality with other family members groupings were averaged to yield 3 summary scores with higher values indicating higher levels of family relationships domain stress in these three areas. Chronic family stress was then conceptualized as a latent factor with each of the three summary scores as an indicator. Results of confirmatory factor analysis indicated an acceptable model fit for this subscale: χ² = 223.076, df = 85, χ² /df = 2.624; CFI = 0.933, RMSEA = 0.070 (90% CI [0.059, 0.081]), SRMR = 0.057.

Behavioral Academic Avoidance

Students completed the 10-item avoidance behaviors subscale of the College Student Adaptative and Maladaptive Academic Behaviors Scale (CSAMABS).³⁶ This subscale assesses the extent to which students respond to academic difficulties using avoidance (ie, avoiding thinking about or engaging with course materials or instructors) rather than proactive efforts to address their situation. Participants responded to questions that followed the stem “When I am worried I am not doing well enough in class, I tend to . . .” using a scale ranging from 1 (Not at all true for me) to 5 (Completely true for me). Items were averaged to yield a summary score with higher values indicating more behavioral avoidance. Sample items included “Come up with ways to avoid attending class,” “Don’t make that class a priority in my life,” and “Try to avoid interactions with the teacher.” Cronbach's alpha for the avoidance behaviors subscale was α = .851. Behavioral academic avoidance was conceptualized as a latent factor with each of the 10 individual items as an indicator. Results of confirmatory factor analysis indicated an acceptable model fit of this scale: χ² = 113.015, df = 33, χ² /df = 3.425; CFI = 0.933, RMSEA = 0.075 (90% CI [0.060, 0.090]), SRMR = 0.042.

Academic Grades

Participants provided consent for us to access their official cumulative grade point average (GPA) data at the conclusion of the next four academic semesters (Fall 2021, Spring 2022, Fall 2022, Spring 2023). GPA data at these four timepoints were provided to study investigators by university administrators based on official academic records for all participants enrolled at the university at that timepoint. This means that students who left the university before Spring 2023 had some missing GPA data. In some cases, missing GPA data was due to student graduation, while in other cases it was due to students leaving the university (permanently or temporarily) for reasons other than graduation. For the current study, our measure of GPA was cumulative GPA (reported on a four-point scale ranging from 0.0 to 4.0) at the end of final/most recent semester during which the student was attendance at the university. The semester in which “final GPA” data were collected thus varied across students and could represent their GPA at the close of any semester from Fall 2021 to Spring 2023.

Procedure

Participants were all enrolled in Psychology courses (mostly large introductory level) for which participation in a certain number of research projects (or completion of alternative assignments) was a course requirement. All students enrolled in these courses in Fall 2021 were eligible for participation in the current study. Recruitment of students to participate in these projects was through courses themselves, in which instructors explained both the participation requirements and the “Psychology subject pool” procedures. No additional financial compensation was provided as part of the subject pool. To fulfill research requirements, students could choose to participate in an online mass screening open to all enrolled students at the start of the Fall 2021 semester (Time 1 – measurement of family stress) and/or an online questionnaire follow-up assessment later in the semester (between weeks 6 and 15; Time 2 - measurement of academic avoidance). Cumulative GPA data were collected at later times as described above. Use of the subject pool and project questionnaires completed in Qualtrics were approved by the participating university's Institutional Review Board (IRB# 12-0269, mass screening, and IRB-FY21-268), and informed consent was provided within Qualtrics with students checking a box after reading the consent form. Figure 1 depicts these recruitment procedures.

Figure 1.

Study design and recruitment procedures.

Our intent to test a mediational model meant that family stress data needed to be collected temporally prior to academic avoidance behaviors data. However, students could elect to participate in the mass screening, follow-up questionnaire, or both. To ensure an adequate sample size, a “planned missingness” approach was adopted for the project.³⁷

In Fall 2021, students had returned to campus and in-person instruction following the COVID-19 lockdown and associated move to online instruction in the United States and globally. Some course sections from which recruiting occurred were taught online, but the majority were in-person. Since both the mass screening and the follow-up questionnaire were administered online, there were no meaningful differences in participation experiences of students enrolled in online versus in-person sections of courses.

Data Analytic Procedures

We calculated descriptive statistics and bivariate correlation coefficients using SPSS³⁸ before estimating the focal mediation model. We utilized Mardia's test for multivariate normality (skewness and kurtosis) to examine the assumption of normality for the model variables in Mplus 8.11,^39–41 p. 818). Significant results of the test for multivariate skewness and kurtosis indicate a violation of the normality assumption.

Based on our planned missingness design and recruitment procedures, the reasons for data missingness were as described below. Students who only participated in the mass screening at Time 1 (n = 162) had missing data for behavioral academic avoidance (measured only at Time 2). Participants who only chose to respond to the follow-up survey at Time 2 (n = 261) had missing data on the chronic family stress variable (measured only at Time 1). Since consent to access GPA was only sought at Time 2, those who only participated at Time 1 had missing data for GPA (n = 162). Some sporadic missingness was observed for family stress (n = 3) and GPA (n = 7). To identify missingness patterns, we utilized logistic regression models to examine whether other observed variables, including focal variables and/or demographics (ie, gender and race/ethnicity), served as predictors of data missingness. Results indicated that race/ethnicity was associated with all three types of missingness; One item measuring behavioral academic avoidance was associated with missingness at Time 1. These findings support a “missing at random” (MAR) pattern (see detailed results in the Supplemental Material).

All model estimation was conducted in Mplus.⁴¹ Confirmatory factor analyses (CFA) was utilized to assess the structural validity of measurements (see these results in the Measures section). To test our hypotheses, we estimated a structural equation model (SEM) to examine the total effect of chronic family stress on GPA (Hypothesis 1) and the mediation effect through behavioral academic avoidance (Hypothesis 2) simultaneously. Following the guidelines of Hayes et al,⁴² we utilized a Bias-corrected bootstrapping approach with 5000 resamples to test the significance of the mediation effect, considered significant if the 95% confidence intervals (CIs) do not include zero. In this model, chronic family stress was conceptualized as a latent construct indicated by mean scores for the three dimensional subscales within the family relationship domain of the CLSQ. Behavioral academic avoidance was conceptualized as a latent variable indicated by its ten individual items. GPA was an observed variable in the model. We controlled for effects of both gender and race/ethnicity on GPA in the model. Gender was coded with two dummy variables (male, non-binary/other) with female as the reference group; race/ethnicity was coded with four dummy variables (Black or African American; Hispanic, Latinx, or Spanish Origin; multiracial or biracial; another racial/ethnic group) with White as the reference group.

For robustness of model estimation, we utilized Maximum Likelihood estimation with robust standard errors (MLR option in Mplus) to address potential violations of multivariate normality in the study variables. Full-information maximum likelihood (FIML) estimation was applied to handle missing data based on an assumed MAR pattern. Model fit was evaluated using the Comparative Fit Index (CFI), the Root Mean Square Error of Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMR). We considered model fit to be acceptable if CFI > 0.90, SRMR < 0.08, and RMSEA < 0.08 (including the upper limit of the 90% confidence interval).^43,44

Results

Descriptive statistics and bivariate correlation coefficients are presented in Table 1. Results indicated the three indicators of chronic family stress were positively and significantly intercorrelated and all items comprising the behavioral academic avoidance scale were positively and significantly intercorrelated. All indicators of behavioral academic avoidance were associated with lower GPA. Only one indicator of chronic family stress (higher levels of family conflict) was associated with lower levels of GPA. The correlation coefficient between the latent chronic family stress variable and GPA indicated this association was not significant (r = -0.033, p = 0.713). The results of Mardia's test for multivariate normality indicated a violation of the normality assumption for chronic family stress, behavioral academic avoidance, and GPA, with tests for both skewness and kurtosis significant (ps < .001).

Table 1.

Descriptive and Bi-Variate Correlations for Model Variables.

	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Closeness	1
2. Conflict	.622***	1
3. Other	.630***	.429***	1
4. Avoid attending class	.128	.240**	.021	1
5. Avoid thoughts	.105	.145	.040	.320***	1
6. Avoid teacher interaction	.191*	.240**	.095	.476***	.369***	1
7. Keep mind off	.142	.230**	.055	.315***	.562***	.459***	1
8. Drop classes	.036	.152*	.169*	.347***	.155**	.341***	.220***	1
9. Use cell phone	.021	.127	.034	.280***	.170***	.274***	.239***	.265***	1
10. Avoid asking for help	.223**	.238**	.172*	.391***	.341***	.566***	.388***	.205***	.376***	1
11. Deprioritize classes	.206**	.187*	.167*	.449***	.289***	.403***	.398***	.389***	.357***	.484***	1
12. Pay less attention	.133	.220**	.140	.396***	.287***	.441***	.260***	.243***	.446***	.342***	.409***	1
13. Study less	.124	.212**	.147	.447***	.318***	.422***	.332***	.311***	.424***	.392***	.434***	.710***	1
14. Final GPA	−.096	−.237**	−.034	−.307***	−.165**	−.192***	−.103*	−.184***	−.157**	−.204***	−.216***	−.121*	−.230***	1
n	333	333	333	435	435	435	435	435	435	435	435	435	434	428
M	2.397	2.457	2.414	1.811	3.074	2.340	2.979	1.786	2.154	2.428	1.986	2.175	2.113	2.772
SD	0.912	0.884	0.765	1.028	1.225	1.204	1.213	1.004	1.085	1.164	1.078	1.050	1.050	0.892

Note. “Closeness”, “Conflict” and “Other” are three composite scores for the dimensions of chronic family stress. Ten items from “Avoid attending class” to “Study less” measure the construct of behavioral academic avoidance. *p < .05, **p < .01, ***p < .001 (2-tailed). N = sample size, M = Mean, SD = Standard Deviation.

SEM analyses indicated an acceptable model fit: χ² = 321.148, df = 151, χ² /df = 2.127; CFI = 0.908, RMSEA = 0.044 (90% CI [0.037, 0.050]), SRMR = 0.058. Regarding the first hypothesis, chronic family stress was not significantly associated with GPA (β = -0.113, SE = 0.082, p = .168). However, results of bias-corrected bootstrapping indicated a significant mediation effect through academic avoidance behaviors (β = -0.083; 95% CI [-0.157, −0.026]). Analyses related to testing measurement models preparatory to conducting final hypothesis testing in SEM are detailed in the Supplemental Materials. In terms of final hypothesis testing, higher levels of chronic family stress at the start of the academic semester were associated with higher levels of behavioral academic avoidance at midterm (β = 0.244, SE = 0.087, p = .005). In turn, higher reported midterm behavioral academic avoidance was associated with lower cumulative GPA during the final semester of enrollment (β = -0.338, SE = 0.051, p < .001). These results are illustrated in Figure 2.

Figure 2.

Structural mediational model examining behavioral academic avoidance as a mediator of associations between chronic family stress and GPA. Note. “Closeness”, “Conflict” and “Other” are three composite scores for the dimensions of chronic family stress. Ten items from “Avoid attending class” to “Study less” measure the construct of behavioral academic avoidance. **p < .01, ***p < .001 (2-tailed). All the factor loadings were statistically significant at a 0.001 confidence level.

Discussion

Findings from this study indicated that mid-semester behavioral academic avoidance behaviors among undergraduate college students mediated the association between their reports of chronic family stress at the start of the semester and cumulative GPA up to 3 full semesters later. Inconsistent with hypotheses, the latent chronic family stress construct was not directly associated with cumulative GPA. Higher levels of chronic family stress during the six months prior to the start of the academic semester predicted higher levels of self-reported behavioral academic avoidance midterm. In turn, students who engaged in more behavioral avoidance midterm had lower cumulative GPAs during their final semesters of enrollment. This effect constituted indirect-only mediation, in that the association between the latent chronic family stress variable and GPA was not significant, either before or after taking levels of behavioral academic avoidance into account.

Existing research has documented that a range of indicators of family stress and dysfunction are associated with lower GPAs among undergraduate college students. These include higher levels of family conflict, lower levels of family support, and relationships with family members indicative of less closeness and insecure attachment.^1,2 Indeed, our first hypothesis was that there would be a direct association between our latent family stress variable and cumulative GPA. However, we found no significant association between our latent chronic stress variable and GPA. In fact, our data indicated that only a single component of family stress - conflict with parents - was significantly correlated with college students’ GPAs. This was surprising, given that the three components of our family stress latent factor (conflict with parents/caregivers, less supportive relationships with parents/caregivers, problematic relationships with other family members) were positively and significantly intercorrelated. We suspect that these null findings may be explained by our inclusion in the latent family stress variable an indicator of the quality of relationships with non-parental family members. In terms of bivariate correlations, the association between the other family relationships indicator and GPA was particularly weak, and relationships with other family members (siblings, extended family members) likely are not as important as relationships with parents for supporting academic success. The lack of direct associations between family stress and GPA can be understood with reference to the CMMSPW model¹⁴ which posits that there will be weaker associations between contextual and person factors more distal from academic outcomes. Within our model, the relation between family stress and GPA was more distal than the relationship between family stress and academic avoidance, perhaps making it more difficult to detect longer term associations.

However, the absence of a direct association between our latent family stress variable and GPA did not mean these variables could not be indirectly associated, by way of student engagement in behavioral academic avoidance. Our second hypothesis was that levels of student academic avoidance behaviors would mediate associations between chronic family stress and GPA, and this hypothesis was supported. Chronic family stress experienced by undergraduate students plays an important role in predicting cumulative GPA over time – potentially until graduation. However, its impact must be understood in terms of its relation to the academic behaviors students engage in. Students experiencing higher levels of family stress are more likely to disengage from their academic work when it becomes difficult for them. Taken together, findings that did not support our first hypothesis (direct associations between family stress and GPA) but did support our second hypothesis (academic avoidance behaviors mediate associations between family stress and GPA) are consistent with tenets of the CMMSPW model.¹⁴ This is because the CMMSPW model proposes that more distal predictors (here, family chronic stress) will be more weakly associated with indicators of student adjustment than procedural subcompetencies (academic avoidance behaviors).

A range of factors might explain the association between family stress and academic disengagement/avoidance. It might be that simultaneously negotiating stressful family relationships and stressful academic experiences overwhelms students. In such cases, students might choose to prioritize managing family relationships over academic endeavors – or find that they have fewer cognitive or psychological resources available to effectively manage demands within both the family and the academic environments. This possibility is consistent with Conservation of Resources theory which proposes that personal relationships (“conditions”) constitute a resource individuals work to protect but can also serve as a source of demands when they are problematic.⁴⁵ Excessive demands among college students can lead to feelings of burnout that have been linked to lower levels of academic engagement.⁴⁶ Alternatively, students who have histories of stressful family relationships may not have internalized the emotional regulation strategies that would enable them to persevere in the face of academic struggles. Consistent with this possibility, both lower levels of perceived parental support⁴⁷ and overly intrusive parenting⁴⁸ have been linked with emotion regulation difficulties among college students. Finally, students with more stressful family relationships may not receive the types of encouragement and guidance from parents/caregivers that would help them to remain focused on coursework when experiencing academic challenges. College students credit their parents’ support as a critical factor that helps them to remain academically engaged through graduation.⁴⁹ Not surprisingly, our findings also documented the potentially detrimental role behavioral avoidance plays with respect to academic success at the postsecondary level. Students likely find college-level work more time-intensive and challenging than high school work. College students are expected to manage expectations for more advanced use of critical thinking skills, heavier reading and writing loads, and requirements that they process more advanced subject matter – all within a context that requires a level of responsibility and self-motivation not typically observed within the high school environment. College students can simply not attend class. Failure to submit assignments may not be addressed by college instructors who believe college students need to make their own choices. College instructors are easier to avoid than high school teachers. In short, it is easier to engage in avoidant behaviors in college than it is in high school, and the consequences of such avoidance are likely to be lower academic grades.

Implications for Application

The link between behavioral academic avoidance and lower grades is both intuitive and well-documented.^9,32–34 What is new in terms of the findings reported in the current study is that students who report experiencing higher levels of family stress are particularly likely to engage in these avoidance behaviors. This finding suggests that effective interventions to improve postsecondary grades should target both avoidance behaviors and the contextual factors that contribute to them. Existing interventions to support low-performing students at the postsecondary level typically focus on encouraging students to engage in behaviors such as attending class, turning in assignments on time, building connections with instructors, and accessing supports such as tutoring when experiencing difficulties (eg,.⁵⁰ Our findings suggest the importance of also acknowledging the sources of stress that make it less likely students will engage these behaviors and offering struggling students access to counseling services that will help them to effectively manage this stress. Students would likely also benefit from access to non-family support systems, such as faculty mentors or intensive advising structures – supports that might compensate for a lack of support from family members. These resources should be widely available on college campuses, targeting not just students who are already engaging in avoidance behaviors, but also endeavoring to connect with students who are experiencing family-based stress but have not yet encountered the types of academic challenges that might lead them to disengage.

Strengths and Limitations

The current study relied on self-report data to assess both levels of chronic family stress and academic avoidance behaviors. Common source and method variance may partially explain the positive association between these two variables. However, they were still valid operationalizations of these constructs. Experiences of stress exposure are subjective. If a student perceives their relationships and interactions with family members to be stressful, it likely matters little whether other family members – or objective observers - would reach similar conclusions. Ideally, we would have included objective indicators of academic avoidance behaviors – such as class attendance records and missing assignments. However, social desirability bias would have made it less likely that students would report engaging in avoidance behaviors. This increases our confidence that students accurately reported on their use of less socially desirable responses to academic challenges. Another limitation of this study was its focus on the role of chronic family stress as a risk factor for postsecondary GPA, while ignoring the numerous additional contexts and factors that play a role in shaping student success. Among these would be characteristics of academic institutions, majors, and instructors, which are of critical importance in relation to students’ academic success. A particular strength of our study was its reliance on GPA from official school records, eliminating any impact of student bias that might have clouded students’ own reports of their grades. Our measure of GPA pulled data from different time points for different students, because students left the participating university at different times and for different reasons. This was an inevitable consequence of our design and may be viewed as a potential limitation. However, cumulative GPA, which averages across multiple classes and semesters, is an inherently better measure of overall achievement than GPA data from a single class or semester. On balance, strengths associated with our operationalization of GPA likely outweigh weaknesses. Finally, the temporal ordering of variables within our design is a strength of this study, in that perceptions of family stress were assessed at the start of an academic semester and referenced the prior six months, while measurement of avoidance behaviors occurred mid-semester, and GPA data were collected weeks or semesters after measurement of avoidance behaviors. This temporal ordering is critically important within any design focused on identifying mediation or indirect effects.

Conclusion

The transition to postsecondary education constitutes not only an academic transition, but also a psychosocial one, in that college students are expected to demonstrate levels of autonomy not required during high school enrollment. Parents of college students are no longer responsible for monitoring their children's class attendance, homework completion, or involvement in activities that may support or threaten academic success. This is true regardless of whether students continue to live within the parental home. However, it would be an error to believe that parents and other family members do not continue to play a critical role with respect to academic success at the postsecondary level. Academic success in college depends on students engaging in a known set of behaviors: attending class, completing assignments, and maintaining open lines of communication with instructors. Students who struggle to manage stress within their relationships with family members can be derailed from engaging in these behaviors, and the academic consequences of this derailment can be serious. Effectively supporting students in the postsecondary environment requires a broad understanding of the factors that may explain students’ disengagement from academic tasks.

Supplemental Material

sj-docx-1-css-10.1177_24705470261426467 - Supplemental material for Behavioral Academic Avoidance as an Indirect Partial Mediator of the Association Between Chronic Family Stress and University Grades

Supplemental material, sj-docx-1-css-10.1177_24705470261426467 for Behavioral Academic Avoidance as an Indirect Partial Mediator of the Association Between Chronic Family Stress and University Grades by Anne Fletcher, Xiaochen Xie, Michaeline Jensen and Suzanne Vrshek-Schallhorn in Chronic Stress

Footnotes

ORCID iDs

Anne Fletcher

Xiaochen Xie

Michaeline Jensen

Suzanne Vrshek-Schallhorn

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Bahrassa

Syed

Lee

. Family conflict and academic performance of first-year Asian American undergraduates. Cultur Divers Ethnic Minor Psychol. 2011;17(4):415–426. 10.1037/a0024731

Cheng

Ickes

Verhofstadt

. How is family support related to students’ GPA scores? A longitudinal study. High Educ. 2012;64(3):399–420. 10.1007/s10734-011-9501-4

Yuan

Weiser

Fischer

. Self-efficacy, parent–child relationships, and academic performance: A comparison of European American and Asian American college students. Soc Psychol Educ. 2016;19(2):261–280. 10.1007/s11218-015-9330-x

Hunter

Baumgardner

Kelner

, et al. An examination of maternal parenting profiles and concurrent effects on college student symptoms of anxiety and depression. J Child Fam Stud. 2025;34(3):824–838. 10.1007/s10826-025-03023-0

Rivera

. College Graduate Salaries: 2025 Projections. Bankrate; 2025, June 2, https://www.bankrate.com/loans/student-loans/average-college-graduate-salary/ .

Clark

Etilé

Postel-Vinay

Senik

Van der Straeten

. Heterogeneity in reported well-being: Evidence from twelve European countries. The Economic Journal. 2005;115(502):C118–C132. 10.1111/j.0013-0133.2005.00983.x

Schneider

Yin

. 2011. The High Cost of Low Graduation Rates: How Much Does Dropping Out of College Really Cost? https://www.air.org/sites/default/files/downloads/report/AIR_High_Cost_of_Low_Graduation_Aug2011_0.pdf.

Deng

Cherian

Khan

NUN

, et al. Family and academic stress and their impact on students’ depression level and academic performance. Front Psychiatry. 2022;15. 10.3389/fpsyt.2022.869337

Leino

Gardner

Cartwright

Döring

. Engagement in a virtual learning environment predicts academic achievement in research methods modules: A longitudinal study combining behavioral and self-reported data. Scholarship of Teaching and Learning in Psychol. 2024;10(2):149–162. 10.1037/stl0000281

10.

Baum

Bell

Elliott

. Trends in college pricing. Trends in Higher Education Series. The College Board; 2014. https://files.eric.ed.gov/fulltext/ED572551.pdf

11.

McFarland

Hussar

de Brey

, et al. 2017. The Condition of Education 2017 (No. NCES 2017144). U.S. Department of Education. https://nces.ed.gov/use-work/resource-library/report/compendium/condition-education-2017?pubid=2017144.

12.

Gershenfeld

Ward Hood

Zhan

. The role of first-semester GPA in predicting graduation rates of underrepresented students. J College Student Retention. 2016;17(4):469–488. 10.1177/1521025115579251

13.

Karimshah

Wyder

Henman

Tay

Capelin

Short

. Overcoming adversity among low SES students. A study of strategies for retention. Australian Universities’ Rev. 2013;55(2):5–14.

14.

de la Fuente

Martinez-Vicente

. Conceptual utility model for the management of stress and psychological wellbeing, CMMSPW^TM in a university environment: Theoretical basis, structure, and functionality. Front Psychol. 2024;14:1299224. 10.3389/fpsyg.2023.1299224

15.

de la Fuente-Arias

. Theory of self- vs. Externally-regulated learning^TM: Fundamentals, evidence, and applicability. Front Psychol. 2017;8:1675. https://doi:10.3389/fpsyg.2017.01675

16.

Hammen

. Stress and depression. Annu Rev Clin Psychol. 2005;1(1):293–319. 10.1146/annurev.clinpsy.1.102803.143938

17.

Monroe

. Modern approaches to conceptualizing and measuring human life stress. Annu Rev Clin Psychol. 2008;4(1):33–52. 10.1146/annurev.clinpsy.4.022007.141207

18.

Jones

Lam

Hoffer

Chen

Schreier

HMC

. Chronic family stress and adolescent health: The moderating role of emotion regulation. Psychosom Med. 2018;80(8):764–773. 10.1097/PSY.0000000000000624

19.

Thomas

Liu

Umberson

. Family relationships and well-being. Innovation in Aging. 2017;1(3):1–11. 10.1093/geroni/igx025

20.

Mazure

. Life stressors as risk factors in depression. Clinical Psychology. 1998;5(3):291–313. 10.1111/j.1468-2850.1998.tb00151.x

21.

Harkness

Monroe

. The assessment and measurement of adult life stress: Basic premises, operational principles, and design requirements. J Abnorm Psychol. 2016;125(5):727–745. 10.1037/abn0000178

22.

Vrshek-Schallhorn

Ditcheva

Corneau

. Stress in depression. In: The Oxford Handbook of Stress and Mental Health. Oxford University Press; 2020:97–126.

23.

Murga

Cabezas

Mora

Campos

Núñez

. Examining associations between symptoms of eating disorders and symptoms of anxiety, depression, suicidal ideation, and perceived family functioning in university students: A brief report. Int J Eating Disord. 2023;56(4):783–789. 10.1002/eat.23787

24.

Adare

Zhang

Wang

. Relationship between parental marital conflict and social anxiety symptoms of Chinese college students: mediation effect of attachment. Front Psychol. 2021;12. 10.3389/fpsyg.2021.640770

25.

Kalibatseva

Leong

FTL

Ham

Lannert

Chen

. Loss of face, intergenerational family conflict, and depression among Asian American and European American college students. Asian Am J Psychol. 2017;8(2):126–133. 10.1037/aap0000067

26.

Lamis

Jahn

. Parent–child conflict and suicide rumination in college students: The mediating roles of depressive symptoms and anxiety sensitivity. J Am Coll Health. 2013;61(2):106–113. 10.1080/07448481.2012.754758

27.

Yazedjian

Toews

Navarro

. Exploring parental factors, adjustment, and academic achievement among white and hispanic college students. J Coll Stud Dev. 2009;50(4):458–467. 10.1353/csd.0.0080

28.

Love

Thomas

. Parenting styles and adjustment outcomes among college students. J Coll Stud Dev. 2014;55(2):139–150. 10.1353/csd.2014.0013

29.

Levens

Elrahal

Sagui

. The role of family support and perceived stress reactivity in predicting depression in college freshman. J Soc Clin Psychol. 2016;35(4):342–355. 10.1521/jscp.2016.35.4.342

30.

Restifo

Akse

Guzman

Benjamins

Dick

. A pilot study of self-esteem as a mediator between family factors and depressive symptoms in young adult university students. J Nervous & Mental Disease. 2009;197(3):166–171. 10.1097/NMD.0b013e318199f790

31.

Vogel

Schwabe

. Stress, aggression, and the balance of approach and avoidance. Psychoneuroendocrinology. 2019; 103:137–146. 10.1016/j.psyneuen.2019.01.020

32.

Credé

Roch

Kieszczynka

. Class attendance in college: A meta-analytic review of the relationship of class attendance with grades and student characteristics. Rev Educ Res. 2010;80(2):272–295. 10.3102/0034654310362998

33.

Cung

Eichhorn

. Increasing interpersonal interactions in an online course: Does increased instructor email activity and voluntary meeting time in a physical classroom facilitate student learning? Online Learning. 2018;22(3):193–215. 10.24059/olj.v22i3.1322

34.

Moore

Jensen

. Do policies that encourage better attendance in lab change students’ academic behaviors and performances in introductory science courses. Science Educator. 2008;17:64–71.

35.

Corneau

. 2022. Development and Psychometric Validation of the Chronic Life Stress Questionnaire: A Multidomain Measure of Chronic Stress Exposure (Publication No. 29256900) [Doctoral dissertation, The University of North Carolina at Greensboro]. ProQuest Dissertations & Theses. https://www.proquest.com/docview/2719410135?pq-origsite=gscholar&fromopenview=true&sourcetype=Dissertations%20&%20Theses.

36.

Fletcher

Alligood

Jensen

Vrshek-Schallhorn

. The college student adaptive and maladaptive academic behaviors scale: development, reliability, and validity. J Coll Stud Retent. 2024. Advance online publication. 10.1177/15210251241261318

37.

Rhemtulla

Little

. Tools of the trade: Planned missing data designs for research in cognitive development. J Cogn Dev. 2012;13(4):425–438. 10.1080/15248372.2012.717340

38.

IBM Corp. 2025. IBM SPSS Statistics for Windows (Version 31.0) [Computer software]. IBM Corp. https://www.ibm.com/products/spss-statistics.

39.

Kline

. Principles and practice of structural equation modeling. Fifth edition. The Guilford Press; 2023.

40.

Mardia

. Measures of multivariate skewness and kurtosis with applications. Biometrika. 1970;57(3):519–530. 10.1093/biomet/57.3.519

41.

Muthén

. Mplus user’s guide. 8th ed. Muthén & Muthén; 1998–2017.

42.

Hayes

. Beyond baron and kenny: Statistical mediation analysis in the new millennium. Commun Monogr. 2009;76(4):408–420. 10.1080/03637750903310360

43.

Bentler

. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary J. 1999;6(1):1–55. 10.1080/10705519909540118

44.

MacCallum

Browne

Sugawara

. Power analysis and determination of sample size for covariance structure modeling. Psychol Methods. 1996;1(2):130–149.

45.

Hobfoll

. Conservation of resources: A new attempt at conceptualizing stress. American Psychologist. 1989;44(3):513–524. 10.1037/0003-066X.44.3.513

46.

Alarcon

Edwards

Menke

. Student burnout and engagement: A test of the conservation of resources theory. J Psychol. 2011;145(3):211–227. 10.1080/00223980.2011.555432

47.

Zhao

Zeng

Chen

Zhang

. Perceived parental support and college students’ depressive symptoms during the COVID-19 pandemic: The mediating roles of emotion regulation strategies and resilience. Curr Psychol. 2023;42:20275–20286. 10.1007/s12144-022-03049-3

48.

Wenze

Pohoryles

Decicco

. Helicopter parenting and emotion regulation in U.S. College students. Psi Chi J Psychol Res. 2019;24(4):274–283. 10.24839/2325-7342.JN24.4.274

49.

Benito-Gomez

Lee

McCurdy

Fletcher

. If I hadn’t had that support system, I think I would have dropped out by now”: Parental support in college and its implications for student adjustment. J Fam Issues. 2022;43(12):3373–3394. 10.1177/0192513X211044490

50.

Melnyk

Kelly

Jacobson

Arcoleo

Shaibi

. Improving physical activity, mental health outcomes, and academic retention in college students with freshman 5 to thrive: COPE/healthy lifestyles. J Am Assoc Nurse Pract. 2014;26(6):314–322. 10.1002/2327-6924.12037

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB