Self-rated Mental and Physical Health are Prospectively Associated with Psychosocial and Academic Adjustment to College

Abstract

Objective.

To examine prospective associations between physical and mental self-rated health (SRH), college generation status and college adjustment among first-year college students.

Participants and Methods.

Eighty-seven first-year college students (41 first-generation college students) reported their SRH when starting college and then reported on psychosocial and academic adjustment and health behaviors midway through each semester.

Results.

Better physical and mental SRH were associated with better psychosocial adjustment in both semesters and academic adjustment in the fall but were generally not predictive of health behaviors. Specifically, better physical SRH was associated with less loneliness (fall: B = −.192, p = .048; spring: B = −.233, p = .008) and fewer anxiety symptoms in both semesters (fall: B = −.236, p = .011; spring: B = −.210, p = .014) and fewer depressive symptoms (fall: B = −.134, p = .016) and more fall semester credits (B = .965, p = .002). Better mental SRH was associated with greater sense of belonging (fall: B = .317, p < .001; spring: B = .242, p = .009), less loneliness (fall: −.210, p = .008; spring: B = −.181, p = .012), and fewer anxiety symptoms (fall: −.193, p = .011; spring: −.195, p = .006) in both semesters and higher fall semester grade point average (B = .129, p = .032). Independent effects of physical and mental SRH are also discussed. Largely, college generation status did not matter for college adjustment within this sample.

Conclusions.

Physical and mental SRH when starting college may be important indicators of psychosocial adjustment over the first year and academic adjustment in the fall.

Introduction

For many US emerging adults, the college transition provides students with increased responsibility, autonomy and new social and academic opportunities. Some students thrive, immersing themselves in the campus community and doing well in classes. Others may struggle, experiencing more difficulties connecting with peers and managing their course load.^1,2 Research has focused on identifying pre-college characteristics that either set students up for success or make them more vulnerable to poorer college adjustment, including college generation status.^3,4 Students’ perceived mental and physical health may also affect their adjustment, but these indicators have received less attention.

Self-rated Health and College Adjustment

Self-rated health (SRH) reflects a person’s perceived health, often rated on a scale from “poor” to “excellent” in response to the question, “how would you rate your overall health?” Although a simple single-item measure, SRH has been consistently linked to mortality and morbidity in nationally representative samples of adults.^5–7 The reasons underlying associations between SRH and health-relevant outcomes are not entirely understood;^8,9 however, it is thought that SRH can be informed by many factors, including comparing oneself to same-age peers,⁸ current health conditions,⁸ and, for younger individuals, their family environment (e.g., parents’ health and health behaviors, family socioeconomic status).¹⁰ Nonetheless, SRH appears to be relatively stable among emerging adults¹¹ and better SRH has been cross-sectionally linked to higher self-esteem, school achievement, and fewer risky health behaviors among adolescents.¹² However, few studies have considered whether SRH is prospectively associated with first-year students’ college adjustment and if physical and mental SRH are differentially associated with adjustment¹³ given that single-item ratings of SRH likely reflect both physiological and psychological symptoms.¹² To this end, this study will examine whether physical and mental SRH are separately and independently associated with a comprehensive array of college-adjustment outcomes across psychosocial, academic and health behavior domains. We will further assess whether associations between physical and mental SRH and college adjustment differ as a function of college generation status.

There is reason to believe physical and mental SRH may be predictive of psychosocial and academic adjustment and engagement in certain health behaviors among college students over their first year of college. With respect to psychosocial outcomes, US college students with chronic physical or mental health conditions reported more concurrent loneliness compared to their relatively healthy peers¹⁴ whereas US adolescents who reported better overall SRH simultaneously reported fewer depressive symptoms.¹⁵ Poorer overall SRH has additionally been cross-sectionally associated with greater anxiety symptoms among Canadians.¹⁶ Mental and physical SRH could potentially impact sense of belonging as well, especially if symptoms prevent students from participating in social events (e.g., if students have to miss classes or withdraw from social opportunities because of symptoms). However, it is unclear whether findings generalize to psychosocial adjustment over the first year of college and whether physical and mental SRH independently predict loneliness, sense of belonging, anxiety and depressive symptoms.

Similarly, poorer physical and mental SRH could contribute to academic adjustment¹⁷ as self-reported mental and physical health concerns have been cross-sectionally associated with poorer academic achievement¹⁸ and less motivation to engage in college-related activities among college students in general.¹⁹ Poorer overall SRH has also been prospectively associated with a greater likelihood of failing classes among US adolescents in high school;²⁰ however, neither physical nor mental SRH was predictive of later academic achievement among Swedish college students.²¹ Whether physical and mental SRH are separately and independently associated with academic performance over the first year of college remains unknown.

Finally, physical and mental SRH may be associated with various health behaviors over the first year of college. Rates of certain risky behaviors, such as consuming alcohol, generally increase in college, and can hinder students’ abilities to maintain college-related responsibilities.^22,23 Lower physical SRH has been cross-sectionally associated with increased odds of consuming alcohol, smoking cigarettes and using marijuana among US adolescents¹² and greater self-reported mental health concerns were predictive of future electronic vapor products (EVP) use.²⁴ In contrast, US and Canadian adolescents who reported better physical SRH concurrently reported greater physical activity.²⁵ Less research has examined associations between physical and mental SRH and health behaviors during the college transition, however.

Why Might Associations Differ by College Generation Status?

Associations between SRH and college adjustment may also differ between first- and continuing-generation college students. First-generation college students (first-gens) are students whose parents do not have 4-year degrees.²⁶ Generally, they may be more likely to struggle in college compared to their peers who have at least one parent with a 4-year degree (continuing-gens).^3,4 First-gens report more loneliness,²⁷ greater depressive symptoms,²⁸ lower sense of belonging²⁹ and lower college GPAs relative to their continuing-gen peers.³⁰ Research on first-gens’ health behaviors is more limited and findings are mixed. Some studies suggest that young adults whose parents have fewer years of education are more likely to engage in common negative health behaviors (e.g., smoking cigarettes),³¹ whereas other studies suggest they are less likely to engage in such behaviors (e.g., smoking cigarettes, consuming alcohol, smoking marijuana).^31–33

There is reason to believe associations between physical and mental SRH and psychosocial and academic adjustment to college and health behaviors may differ by college generation status. First-gens may rate their health lower than continuing-gens because they are more likely to come from lower income families³ and lower childhood family socioeconomic status has been associated with lower SRH.¹⁰ First-gens are also less likely to seek and receive help from university services,^34,35 perhaps because they are not aware of, cannot afford, or do not feel comfortable contacting university health and counseling services. Thus, first-gens who have lower physical and mental SRH may not be receiving help for these concerns, which could exacerbate college adjustment problems over the first year.

Research Aims and Hypotheses

We used a prospective longitudinal design to examine influences of physical and mental SRH on first-year students’ college adjustment across multiple domains and whether associations differed between first- and continuing-gens. First-year students reported on their college generation status and SRH when starting college. Psychosocial and academic adjustment and health behaviors were then assessed at laboratory visits, midway through the fall and spring semester, respectively. We hypothesized the following:

• H1. Lower physical and mental SRH will be separately associated with more loneliness, less sense of belonging and greater depressive and anxiety symptoms in the fall and spring semester.

• H2. Lower physical and mental SRH will be separately associated with lower fall semester GPAs. We did not have concrete directional hypotheses for how physical and mental SRH would relate to college credits and odds of changing majors given inconsistent findings from previous studies.

• H3. Lower physical and mental SRH will be separately associated with greater odds of consuming alcohol, smoking cigarettes, using marijuana and EVPs, and lower odds of engaging in physical activity.

• H4. First-gens will struggle more adjusting to college compared to continuing-gens such that first-gens will report more psychosocial adjustment concerns (e.g., more loneliness, less sense of belonging) and poorer academic adjustment (e.g., lower fall GPAs). We did not have a clear directional hypothesis for associations between college generation status and health behaviors given mixed findings from prior studies.

• H5. Effects of physical and mental SRH on college adjustment will be moderated by college generation status such that effects of physical and mental SRH on psychosocial and academic adjustment and health behaviors over the first year will be stronger among first- versus continuing-gens.

Because most studies use a single measure of SRH, we also examined independent effects of physical and mental SRH on college adjustment in exploratory models.

Methods

Participants

At a 4-year public university, 87 first-year students were followed over their fall and spring semester. Students were recruited via emails, flyers, and announcements made during new student orientations and in classrooms at the start of the fall semester. Eligibility criteria included being a first-year at the main campus, fluent in English and between 18–21 years of age. Participants needed to be free of serious chronic illnesses and were ineligible if they participated in the summer 6-week Learning Edge Academic Program given the project’s focus on measuring students’ college adjustment at similar points of time in their college transition. See Table 1 for participant characteristics.

TABLE 1.

Sample descriptives of first-year college students

	Full sample (n = 87§)		First-gen students (n = 41; 47.1%§)		Continuing-gen students (n = 46; 52.9%§)
	n (%)	M (SD)	n (%)	M (SD)	n (%)	M (SD)
Covariates
Men	18 (20.7)		3 (7.3)*		15 (32.6)*
Racial/ethnic minority	32 (36.8)		19 (46.3)		13 (28.3)
Taking psychotropic medication	6 (6.9)		2 (4.9)		4 (8.7)
Baseline survey completed prior to fall semester	35 (40.2)		8 (19.5)*		27 (58.7)*
Baseline self-rated health
Physical health		3.67 (0.76)		3.51 (0.81)		3.80 (0.69)
Mental health		3.18 (1.01)		3.02 (0.99)		3.33 (1.01)
Psychosocial adjustment (fall)
Sense of belonging		3.46 (0.73)		3.32 (0.71)		3.59 (0.73)
Loneliness		0.96 (0.69)		1.09 (0.73)		0.84 (0.64)
Depressive symptoms		1.37 (0.39)		1.46 (0.41)*		1.29 (0.36)*
Anxiety symptoms		1.09 (0.66)		1.22 (0.71)		0.96 (0.60)
Psychosocial adjustment (spring)
Sense of belonging		3.49 (0.78)		3.40 (0.82)		3.56 (0.74)
Loneliness		0.84 (0.63)		0.92 (0.58)		0.78 (0.67)
Depressive symptoms		1.32 (0.46)		1.48 (0.46)*		1.20 (0.44)*
Anxiety symptoms		1.09 (0.63)		1.18 (0.65)		1.01 (0.61)
Academic adjustment
Fall GPA		3.46 (0.50)		3.39 (0.51)		3.52 (0.49)
Fall credits		15.71 (2.12)		15.45 (2.08)		15.92 (2.14)
Spring credits		16.44 (1.86)		16.24 (1.72)		16.60 (1.98)
Changed major	27 (32.9)		11 (29.7)		16 (35.6)
Health behaviors (fall)
Smoked cigarettes at least once over past 30 days	7 (8.0)		5 (12.2)		2 (4.3)
Used EVP at least once over past 30 days	28 (32.6)		18 (43.9)*		10 (22.2)*
Used marijuana at least once over past 30 days	24 (27.6)		16 (39.0)*		8 (17.4)*
Consumed alcohol at least once over past 30 days	64 (73.6)		29 (70.7)		35 (77.8)
Engaged in physical activity for 20 min on 1 or more days over past week	84 (96.6)		39 (95.1)		45 (97.8)
Engaged in physical activity for 60 min on 1 or more days over past week	73 (83.9)		32 (78.0)		41 (89.1)
Health behaviors (spring)
Smoked cigarettes at least once over past 30 days	4 (4.9)		3 (8.1)		1 (2.2)
Used EVP at least once over past 30 days	21 (25.9)		12 (32.4)		9 (20.0)
Used marijuana at least once over past 30 days	26 (32.1)		17 (37.8)*		9 (20.0)*
Consumed alcohol at least once over past 30 days	62 (75.6)		28 (62.2)		34 (75.6)
Engaged in physical activity for 20 min on 1 or more days over past week	79 (96.3)		35 (94.6)		44 (97.8)
Engaged in physical activity for 60 min on 1 or more days over past week	67 (81.7)		27 (73.0)		40 (88.9)

^§82 (37 first-gens and 45 continuing-gens) participants were retained for the spring semester visit. The five participants who only completed the fall semester visit did not significantly differ from participants who completed both visits with respect to mental SRH (t(85)=−.878, p=.383), college generation status (χ²(1, N=87)=2.30, p=.129), gender (χ²(1, N=87)=1.38, p=.239), race/ethnicity (χ²(1)=.024, p=.878), psychotropic medication use (χ²(1, N=87)=.393, p=.531), and baseline survey completion (χ²(1, N=87)<.001, p=.991). They did, however, rate their physical health as being higher on average compared to the 82 participants who completed both visits (5 participants lost to follow up: M = 4.00, SD = .00; 82 retained participants: 3.65, SD = .78; t(81) = 4.128, p < .001)

^*p < .05 when conducting independent sample t-tests and chi-square tests to compare first- and continuing-gen college students. More first-gens were women (χ² (1) = 8.451, p = .004), completed the baseline survey after the fall semester commenced (χ² (1) = 13.842, p < .001), had higher mean depressive symptoms (fall: t(85) = 2.008, p = .048; spring: t(79) = 2.717; p = .008) and more reported using EVP (fall: χ² (1) = 4.592, p = .032) and marijuana (fall: χ² (1) = 5.079, p = .024; spring: χ² (1) = 5.992, p = .014) in the 30 days leading up to their in-person visits. First- and continuing-gens did not differ in terms of physical SRH (first-gen: M=3.51, SD=.81; continuing-gen: M=3.80, SD=.69, t(85)=-1.820, p=.072), mental SRH (first-gen: M=3.02, SD=.99; continuing-gen: M=3.33, SD=1.01, t(85)=-1.404, p=.164), race/ethnicity (χ²(1, N=87)=3.048, p=.081), or psychotropic medication use (χ²(1, N=87)=.492, p=.483).

Note. First-gen students = students whose parents do not have 4-year degrees; Continuing-gen students = students who have at least one parent with a 4-year degree; EVP = electronic vapor products

Procedure

Following an online eligibility survey, participants completed an online baseline survey at the start of the fall semester (± two weeks) and then attended a 70-minute laboratory visit at the university’s Clinical Research Center in October 2019 and again in February 2020. Participants provided implied consent before the online surveys and informed (written) consent in October. Visits followed the same format with participants completing questionnaires and a semi-structured interview after completing other tasks not specific to the present analyses. Participants were compensated $60 and The Pennsylvania State University’s Institutional Review Board approved the study.

Measures

SRH

As part of the baseline survey, participants separately rated their physical and mental health on a 5-point Likert scale (1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent) in response to the following: Overall, how would you rate your physical health? Overall, how would you rate your mental health? SRH questions were modeled off of the SRH measure used in the National Longitudinal Adolescent to Adult Health study.³⁶

College Generation Status

Participants reported their parent(s)’ educational attainment as part of the eligibility survey. Forty-one participants (47.1%) were considered first-gen, meaning neither parent had a 4-year degree and 46 (52.9%) were continuing-gen (at least one parent had a 4-year degree). College generation status was coded as 1 for first-gen and 0 for continuing-gen.

Psychosocial Adjustment

Sense of belonging.

The 26-item Sense of Belonging Scale³⁷ measured the extent to which participants felt they belonged on campus. On a 5-point Likert scale (1=completely untrue; 5=completely true), participants indicated how true they felt statements captured their current college experiences. Items included statements about socializing with peers to comfort reaching out to professors. Social isolation items (e.g., “I know very few people in my class”) were reverse-scored and all items averaged so higher mean scores reflected greater sense of belonging. This measure has strong internal consistency based on prior college samples (Cronbach’s α = .89).³⁸

Loneliness.

The Robert’s 8-item version of the UCLA Loneliness Scale³⁹ assessed students’ feelings of loneliness. Participants indicated on a 4-point Likert scale (0=never; 3=often) the extent to which they felt left out or lacked closeness with others. Four positively-phrased items were reverse-scored and all responses averaged so higher mean scores reflected more loneliness. This scale is acceptable for use with adolescents and emerging adults and internal consistency was good within a prior adolescent sample (Cronback’s α = .78).³⁹

Depressive symptoms.

Participants completed the 10-item Center for Epidemiological Studies of Depression (CESD) scale,⁴⁰ a commonly used measure of depressive symptoms that has been validated for use with adolescents and emerging adults.⁴¹ On a 4-point Likert scale (0=rarely; 3=most of the time), participants reported how often they experienced certain feelings and behaviors. Two positively-phrased items were reverse-scored and all items averaged so higher mean scores reflected more depressive symptoms. The CESD is appropriate for use with adolescent and emerging adult samples and had good internal consistency in a sample of adolescents (Cronbach’s α = .89).⁴¹

Anxiety symptoms.

The Generalized Anxiety Disorder 7 Scale⁴² measured anxiety symptoms. Participants indicated on a 4-point Likert scale (0=not at all; 3=nearly every day) how frequently they experienced certain emotions, thoughts and sensations. All items were averaged to approximate each participant’s anxiety symptoms. This scale has strong internal consistency (Cronbach’s α = .92) and test-retest reliability, and is recommended for use in research settings.⁴²

Academic Adjustment

As part of the spring visit, participants reported their fall GPA, the number of credits they completed in the fall and how many credits they were currently taking in the spring. The sample average of 16 credits per semester (about 5 classes) is considered full-time. To reduce recall bias, participants were encouraged to look up their GPA and enrollment history on their mobile devices. Participants also reported whether they changed their intended major since starting college.

Health Behaviors

Based on selected items from the Youth Risk Behavior Surveillance Survey.⁴³ participants self-reported whether they smoked cigarettes, used EVP or marijuana or consumed alcohol over the previous 30 days and whether they engaged in physical activity for 20 or 60 minutes on one or more days the previous week. Participants were asked about additional substances (e.g., cocaine), but so few participants endorsed using them (range=0–2 participants) so we were not powered to examine associations between SRH, college generation status, and these substances.

Covariates

Participants reported their gender, race and ethnicity. The date of the baseline survey about SRH was recorded. At the visits, participants indicated whether they were currently taking any medications regularly and the reasons for taking them and six (6.9%) reported taking medication for anxiety or mood disorders.

Statistical Analyses

Participants were categorized as White non-Hispanic (63.2%) or of a racial/ethnic minority (36.8%). Because half of the baseline surveys were completed shortly after the fall semester began (59.8%), responses to SRH items may have varied based on survey completion time. A dichotomous variable was created to adjust for participants who completed the survey before or after classes started. Models adjusted for gender (1 = female, 0 = male), race/ethnicity (1 = racial/ethnic minority, 0 = White non-Hispanic), survey completion (1 = before fall semester started, 0 = after fall semester started), and whether participants were taking psychotropic medications to account for pre-existing mental health disorders (1 = taking medication, 0 = not taking medication). All statistical analyses were performed using SPSS version 26.0 (IBM, New York, NY).

We examined the separate main effects of physical and mental SRH and college generation status on psychosocial and academic adjustment and health behaviors each semester. Multiple linear regression was used to predict continuous outcomes (e.g., loneliness) and logistic regression was used to model dichotomous outcomes (e.g., odds of smoking cigarettes). We also tested the independent effects of physical and mental SRH on these adjustment outcomes by including both physical and mental SRH as predictor variables in the same models.

To assess whether associations between SRH and college adjustment differed by college generation status, we examined interactions between physical SRH and college generation status and then between mental SRH and college generation status on each psychosocial, academic, and health behavior outcome. Following recommendations from Aiken, West and Reno,⁴⁴ all continuous dependent variables (i.e., physical SRH, mental SRH) were first sample mean-centered at zero. College generation status remained coded as 1 = first-gen and 0 = continuing-gen. The interaction term was created by multiplying sample mean-centered physical SRH by college generation status and sample mean-centered mental SRH by college generation status. Covariates and sample mean-centered dependent variables were entered in Step 1 and the interaction term in Step 2. Two sets of interaction models were fit for each outcome: one for mean-centered physical SRH by college generation status and one for mean-centered mental SRH by college generation status.

Prior to fitting regression models, we examined distributions of continuous outcome variables and conducted chi-square tests and independent-sample t-tests to compare first- and continuing-gens on relevant study variables. We also assessed associations between SRH, college generation status and college adjustment in empty models and then adjusted for covariates. All results reported below are from fully adjusted models only. Similarly, we also conducted a series of sensitivity analyses excluding participants who reported taking psychotropic medication instead of adjusting for medication use and results did not substantively change, with a few exceptions noted below.

Results

Five participants (5.7%) who completed the fall visit were not able to participate in the spring. Two were no longer enrolled in the university, two had prior work or sports commitments that prevented them from participating and one did not respond to scheduling messages. These five participants did not differ from the 82 retained participants except these five reported slightly better physical SRH (see Table 1).

More first-gens were women and completed the baseline survey after the semester commenced. First- and continuing-gens did not differ on SRH or other demographics (Table 1). There was only a low to moderate correlation between physical and mental SRH (r(87) = .371; p < .01), suggesting they are similar constructs that capture distinct aspects of health and may be differentially associated with college adjustment. Internal consistency was strong for our measures of sense of belonging (Cronbach’s α = .934 and .946 each semester, respectively), loneliness (Cronbach’s α = .870 and .879 each semester, respectively) and anxiety symptoms (Cronbach’s α = .881 in both semesters). Internal consistency for the CESD was low (Cronbach’s α = .518 and .661 each semester, respectively) because the two reverse-scored items were not highly correlated with the other items. Although internal consistency would have been acceptable if these items were removed, we kept them for their face validity in terms of depressive symptomatology and to allow for comparisons with other emerging adult samples.

Continuous variables were relatively normally distributed and did not need to be transformed prior to fitting regression models (skewness = |.075| - |1.183|; kurtosis = |.083| - |1.130|). Smoking cigarettes was endorsed among only a small number of students (n = 7 in fall, 4 in spring) whereas other health behaviors (e.g., consuming alcohol) were more common (n = 64 in fall, 62 in spring; see Table 1).

Main Effects on Psychosocial Adjustment (Table 2)

TABLE 2.

Multiple linear regression analyses of the main and independent effects of physical SRH, mental SRH, and college generation status on psychosocial adjustment during the fall and spring semester

	College generation status				Physical SRH				Mental SRH				Independent effect of physical SRH				Independent effect of mental SRH
	B	SE	β	R2	B	SE	β	R2	B	SE	β	R2	B	SE	β	R2	B	SE	β	R2
Fall semester
Sense of belonging	−.164	.176	−.113	.122	.195	.103	.203	.150	.317	.079	.437	.259	.056	.105	.058	.262	.299	.086	.413	.262
Loneliness	.043	.165	.031	.146	−.192	.096	−.211	.186	−.210	.077	−.306	.217	−.111	.102	−.122	.228	−.175	.084	−.255	.228
Depressive symptoms	.065	.094	.083	.132	−.134	.054	−.259	.188	−.076	.046	−.195	.156	−.116	.059	−.224	.134	−.040	.048	−.102	.134
Anxiety symptoms	.133	.158	.101	.146	−.236	.091	−.270	.205	−.193	.075	−.294	.205	−.171	.097	−.195	.234	−.140	.080	−.212	.234
Spring semester
Sense of belonging	−.058	.193	−.038	.121	.190	.111	.191	.153	.242	.089	.317	.197	.090	.118	.091	.203	.213	.097	.279	.203
Loneliness	−.020	.152	−.016	.182	−.233	.085	−.286	.255	−.181	.071	−.292	.247	−.174	.091	−.214	.281	−.125	.075	−.202	.281
Depressive symptoms	.174	.106	.188	.263	−.093	.062	−.156	.259	−.097	.051	−.215	.272	−.055	.067	−.093	.279	−.079	.056	−.175	.279
Anxiety symptoms	.128	.148	.101	.234	−.210	.084	−.256	.286	−.195	.068	−.313	.302	−.139	.089	−.170	.324	−.150	.074	−.241	.324

Significant effects (p < .05) are in bold

Note. College generation status (1 = first-generation college student, 0 = continuing-generation college student). SRH = self-rated health, B = unstandardized beta coefficient, SE = standard error, β = standardized beta coefficient, R² = coefficient of determination. Models adjusted for gender (1 = female, 0 = male), race/ethnicity (racial/ethnic minority = 1, White non-Hispanic = 0), psychotropic medication use (1 = taking medication, 0 = not taking medication), and whether the baseline survey was completed before or after the fall semester started (1 = before, 0 = after).

Physical SRH

Participants with better physical SRH reported less loneliness (B = −.192, SE = .092, p = .048), fewer depressive symptoms (B = −.134, SE = .054, p = .016) and fewer anxiety symptoms in the fall (B = −.236, SE = .091, p = .011) and less loneliness (B = −.233, SE = .085, p = .008) and fewer anxiety symptoms in the spring (B = −.210, SE = .084, p = .014). Physical SRH was not associated with sense of belonging in either semester or depressive symptoms in the spring (ps >.06).

Mental SRH

Participants with better mental SRH reported greater sense of belonging (fall: B = .317, SE = .079, p < .001; spring: B = .242, SE = .089, p = .009), less loneliness (fall: B = −.210, SE = .077, p = .008; spring: B = −.181, SE = .071, p = .012), and fewer anxiety symptoms (fall: B = −.193, SE = .075, p = .011; spring: B = −.195, SE = .068, p = .006). Mental SRH was not associated with depressive symptoms in either semester (ps >.06).

College Generation Status

College generation status was not associated with any psychosocial outcomes (ps >.10).

Independent Effects of Physical and Mental SRH

Mental SRH remained an independent predictor of psychosocial outcomes when both mental and physical SRH were simultaneously included as predictors. Better mental SRH continued to be associated with greater sense of belonging in both semesters (fall: B = .299, SE = .086, p = .001; spring: B = .213, SE = .097, p = .032), less loneliness in the fall (fall: B = −.175, SE = .084, p = .039; spring: p >.10) and fewer anxiety symptoms in the spring (fall: p > .08; spring: B = −.150, SE = .074, p = .044). Physical SRH was not independently associated with loneliness, depressive or anxiety symptoms in either semester (ps >.05).

Main Effects on Academic Adjustment (Table 3)

TABLE 3.

Multiple linear and logistic regression analyses of the main and independent effects of physical SRH, mental SRH, and college generation status on academic adjustment during the fall and spring semester

	College generation status				Physical SRH				Mental SRH				Independent effect of physical SRH				Independent effect of mental SRH
	B	SE	β / exp. B	R2	B	SE	β / exp. B	R2	B	SE	β / exp. B	R2	B	SE	β / exp. B	R2	B	SE	β / exp. B	R2
Fall semester GPA	−.170	.124	−.170	.136	.043	.074	.066	.118	.129	.059	.262	.167	−.024	.079	−.037	.168	.137	.065	.278	.168
Fall semester credits	−.337	.550	−.080	.035	.965	.306	.354	.143	.231	.269	.111	.039	1.015	.335	.372	.144	−.105	.279	−.050	.144
Spring semester credits	−.543	.463	−.146	.146	−.076	.290	−.032	.131	.122	.227	.067	.133	−.020	.296	−.008	.125	.116	.247	.063	.125
Odds of changing majors	−.253	.527	.776§	.083†	.385	.334	1.469§	.091†	−.309	.271	.734§	.087†	.668	.393	1.950§	.076†	−.526	.325	.591§	.076†

Significant effects (p < .05) are in bold

Note. College generation status (1 = first-generation college student, 0 = continuing-generation college student). SRH = self-rated health, B = unstandardized beta coefficient, SE = standard error, β = standardized beta coefficient, exp. B = exponentiated beta/odds ratio, R² = coefficient of determination. Models adjusted for gender (1 = female, 0 = male), race/ethnicity (racial/ethnic minority = 1, White non-Hispanic = 0), psychotropic medication use (1 = taking medication, 0 = not taking medication), and whether the baseline survey was completed before or after the fall semester started (1 = before, 0 = after).

^§Exponentiated Beta/odds ratio

^†= pseudo R² based on Cox and Snell’s ratio of likelihoods approach for dichotomous outcomes

Physical SRH

Better physical SRH was associated with taking more fall credits (B = .965, SE = .306, p = .002), but did not predict fall GPA, spring credits or odds of changing majors (ps >.20).

Mental SRH

Better mental SRH was associated with higher fall GPA (B = .129, SE = .059, p = .032), but did not predict fall or spring credits or odds of changing majors (all ps >.20).

College Generation Status

College generation status was not associated with any academic outcomes (all ps >.10)

Independent Effects of Physical and Mental SRH

Physical SRH continued to be associated with fall credits (physical SRH: B = 1.015, SE = .335, p = .003; mental SRH: p > .70) and mental SRH continued to be associated with fall GPA (physical SRH: p > .70; mental SRH: B = .137, SE = .065, p = .038) when considering both SRH measures simultaneously, but neither independently predicted spring credits or odds of changing majors (ps >.08).

Main Effects on Health Behaviors (Table 4)

TABLE 4.

Multiple logistic regression analyses of the main and independent effects of physical SRH, mental SRH, and college generation status on health behaviors during the fall and spring semester

	College generation status				Physical SRH				Mental SRH				Independent effect of physical SRH				Independent effect of mental SRH
	B	SE	exp. B	${\mathit{R}}_{\mathit{C}\mathit{S}}^2$	B	SE	exp. B	${\mathit{R}}_{\mathit{C}\mathit{S}}^2$	B	SE	exp. B	${\mathit{R}}_{\mathit{C}\mathit{S}}^2$	B	SE	exp. B	${\mathit{R}}_{\mathit{C}\mathit{S}}^2$	B	SE	exp. B	${\mathit{R}}_{\mathit{C}\mathit{S}}^2$
Fall semester
Smoking cigarettes in last 30 days	2.188	1.163	8.914	.084	−1.981	.808	.138	.131	−1.691	.697	.184	.128	−1.551	.841	.212	.172	−1.367	.729	.255	.172
Using EVP in last 30 days	1.472	.598	4.359	.093	−.290	.331	.748	.062	−.532	.279	.587	.027	−.065	.356	.937	.062	−.513	.297	.598	.062
Using marijuana in last 30 days	1.195	.596	3.303	.062	−.369	.336	.691	.040	−.419	.285	.658	.029	−.206	.364	.814	.043	−.353	.308	.703	.043
Consuming alcohol in last 30 days	.020	.596	1.020	.070	−.228	.340	.796	.107	−.559	.311	.572	.075	.036	.379	1.037	.107	−.572	.340	.564	.107
Physical activity for 20 min on 1 or more of last 7 days	−.319	1.498	.727	.056	−1.040	.858	.354	.100	−1.574	.874	.207	.073	−.303	1.156	.739	.100	−1.425	1.019	.240	.100
Physical activity for 60 min on 1 or more of last 7 days	−1.164	.827	.312	.134	.567	.421	1.763	.113	−.073	.369	.930	.132	.810	.511	2.247	.141	−.433	.457	.648	.141
Spring semester
Smoking cigarettes in last 30 days	2.718	1.594	15.150	.120	−2.700	1.410	.067	.153	−2.316	1.139	.099	.137	−1.819	1.540	.162	.169	−1.660	1.130	.190	.169
Using EVP in last 30 days	.912	.638	2.490	.090	−.688	.383	.503	.102	−.535	.307	.586	.105	−.517	.406	.596	.120	−.392	.330	.676	.120
Using marijuana in last 30 days	1.882	.678	6.568	.137	−1.213	.399	.297	.157	−.441	.285	.643	.062	−1.156	.421	.315	.158	−.130	.322	.878	.158
Consuming alcohol in last 30 days	.288	.615	1.333	.027	−.083	.347	.920	.025	−.136	.294	.873	.027	−.022	.380	.978	.027	−.129	.322	.879	.027
Physical activity for 20 min on 1 or more of last 7 days	−1.531	1.779	.216	.114	.202	.988	1.223	.106	−.351	.853	.704	.107	.853	1.537	2.348	.111	−.838	1.283	.433	.111
Physical activity for 60 min on 1 or more of last 7 days	−1.088	.726	.337	.064	.802	.418	2.230	.083	.020	.322	1.020	.037	.982	.474	2.670	.092	−.335	.381	.715	.092

Significant effects (p ≤ .05) are in bold

Note. College generation status (1 = first-generation college student, 0 = continuing-generation college student). SRH = self-rated health, B = unstandardized beta coefficient, SE = standard error, β = standardized beta coefficient, exp. B = exponentiated beta/odds ratio, $R_{CS}^2$ = pseudo R² based on Cox and Snell’s ratio of likelihoods approach for dichotomous outcomes. Models adjusted for gender (1 = female, 0 = male), race/ethnicity (racial/ethnic minority = 1, White non-Hispanic = 0), psychotropic medication use (1 = taking medication, 0 = not taking medication), and whether the baseline survey was completed before or after the fall semester started (1 = before, 0 = after).

Physical SRH

Participants with better physical SRH were less likely to smoke cigarettes in the fall (fall: B = −1.981, SE = .808, p = .014; spring: p >.05) and to use marijuana in the spring (fall: p >.20; spring: B = −1.213, SE = .399, p = .002). Physical SRH was not associated with the likelihood of using EVP, consuming alcohol, or engaging in physical activity in either semester (ps >.05)

Mental SRH

Participants with better mental SRH were less likely to smoke cigarettes (fall: B = −1.691, SE = .697, p = .015; spring: B = −2.316, SE = 1.139, p = .042). Mental SRH was not associated with using EVP or marijuana, consuming alcohol, or engaging in physical activity in either semester (ps >.05)

College Generation Status

First-gens were more likely to use EVP in the fall (fall: B = 1.472, SE = .598, p = .014; spring: p >.10) and marijuana in both semesters (fall: B = 1.195, SE = .596, p = .045; spring: B = 1.882, SE = .678, p = .005). College generation status was not associated with smoking cigarettes, consuming alcohol, or engaging in physical activity in either semester (ps> .06).

Independent Effects of Physical and Mental SRH

Better physical SRH continued to be associated with a decreased likelihood of using marijuana in the spring (B = −1.156, SE = .421, p = .006) and was independently associated with greater odds of engaging in physical activity for 60 minutes on one or more of the last seven days in the spring (fall: p > .10; spring: B = .982, SE = .472, p = .038). There were no other independent effects (ps >.06).

Moderation by College Generation Status

College generation status did not moderate associations between physical or mental SRH and any college adjustment outcomes in either semester (all ps >.05).

Sensitivity analyses

When excluding participants who reported using psychotropic medication instead of adjusting for psychotropic medication use, 25 of 28 significant effects were substantively unchanged. The following three associations were no longer significant: the independent association between mental SRH and anxiety symptoms in the spring (B = −.152, SE = .077, p = .051), the association between college generation status and odds of using marijuana in the fall (B = .901, SE = .604, p = .135) and the independent effect of physical SRH on the odds of engaging in physical activity for 60 minutes on one or more days in the spring (B = .866, SE = .479, p = .070). An additional four new associations emerged: an independent effect of mental SRH on loneliness in the spring (B = −.143, SE = .069, p = .043), an independent effect of physical SRH on anxiety symptoms in the fall (B = −.194, SE = .074, p = .041) and depressive symptoms in the spring (B = −.201, SE = .086, p = .023), and an independent effect of mental SRH on odds of smoking cigarettes in the fall (B = −1.564, SE = .724, p = .031).

Discussion

We sought to examine whether physical and mental SRH were prospectively associated with college adjustment across psychosocial, academic and health behavior domains and whether associations differed as a function of college generation status. Overall, we found that physical and mental SRH at the time of transitioning to college were associated with first-year students’ psychosocial adjustment in the fall and spring semesters and academic adjustment in the fall, but not their health behaviors. College generation status was largely unassociated with college adjustment, except for EVP and marijuana use, and the impact of physical and mental SRH on college adjustment did not differ between first- and continuing-gen students. Our findings collectively suggest that physical and mental SRH may be important indicators of psychosocial adjustment over the first year of college and students’ initial academic adjustment.

Physical and mental SRH were most consistently associated with psychosocial adjustment. Aligning with previous research,^14,16 better physical and mental SRH were separately associated with less loneliness and fewer anxiety symptoms in each semester. Similar to Mechanic and Hansell,¹⁵ we found that students with better physical SRH reported fewer depressive symptoms in the fall. Better mental SRH was additionally associated with greater sense of belonging in both semesters, suggesting that students with lower mental SRH may experience more difficulties connecting with others on campus. When examining independent effects, only mental SRH continued to be associated with sense of belonging and loneliness in the fall and sense of belonging and anxiety symptoms in the spring. This suggests that mental SRH is a stronger predictor of psychosocial adjustment and that physical and mental SRH represent two similar but distinct constructs.¹³

Physical and mental SRH were also prospectively associated with first-year students’ academic adjustment, but only in the fall. Independent of mental SRH, students with better physical SRH completed more fall credits, implying that poorer physical SRH may contribute to absenteeism and smaller courseloads.²⁰ Conversely, better mental SRH was independently associated with higher fall GPA. Aligning with Stallman,¹⁸ findings suggest that students with lower mental SRH may perhaps benefit from academic accommodations (e.g., extending assignment deadlines) if symptoms impede on their academic performance. Neither physical nor mental SRH predicted spring credits or odds of changing majors. However, future research should examine how SRH may relate to changes in course enrollment given that the number of credits reported by students midway through the spring semester may not reflect the number of credits ultimately completed.

In this sample, SRH did not predict health behaviors except for smoking cigarettes and using marijuana. Students with better physical SRH were less likely to smoke cigarettes in the fall and, independently of mental SRH, were less likely to use marijuana in the spring. Better mental SRH was only associated with a decreased likelihood of smoking cigarettes in the fall, but neither physical nor mental SRH independently predicted odds of smoking. This may be partly because so few students endorsed smoking cigarettes. Physical and mental SRH were not associated with consuming alcohol, using EVP or engaging in physical activity. However, information on when health behaviors were initiated (before or after coming to college) and why (to socialize with others or to cope with health concerns) was unavailable.

Counter to our hypotheses, college generation status was not associated with adjustment, except for EVP and marijuana use. First-gens were more likely to use EVP in the fall and marijuana in both semesters, although this should be interpreted with caution until replicated in larger samples. These findings contrast Patrick and colleagues’³¹ findings of more marijuana use among young adults with more educated parents. Importantly, however, when looking at group differences, first-gens reported poorer SRH and college adjustment across all outcomes (e.g., lower sense of belonging), but group differences generally did not reach statistical significance. Counter to previous research on first-gens as an at-risk group,^4,27,28 we did not find that associations between SRH and college adjustment differed by college generation status.

The present study has several strengths including the prospective assessment of students over their first year of college, with SRH assessments at the time of transitioning to college and repeat assessments of adjustment midway through each semester. Nearly 94% of participants completed both visits and all visits occurred during the month of October and again in February, prior to mid-semester breaks and before final exams began. This reduces the likelihood that differences in college adjustment across students were due to when participants completed their individual visit. Extending previous studies that employed a single rating of SRH, students reported on both their perceived physical and mental health and we considered how each was independently associated with college adjustment. Finally, a number of psychosocial and academic adjustment and health behavior measures were included to better understand the widespread impacts of SRH on college adjustment.

Limitations include the broad assessment of students’ college adjustment, which may have increased risk for Type I error. Nonetheless, the consistent pattern of results across multiple measures suggests that findings are not due to chance. The smaller sample size may have contributed to non-significant findings, especially for interaction effects and predicting less common health behaviors. Further, first-gens were defined as students whose parents did not have 4-year degrees, similar to previous research;²⁷ however, it is possible that SRH and college adjustment differ between first-gens whose parents have some versus no higher education experience. Most participants were women and the majority White non-Hispanic, so future research examining whether physical and mental SRH prospectively predict college adjustment among students of color, especially from underrepresented groups in higher education and at different types of institutions is essential.

Conclusions

SRH may prospectively predict college adjustment. More specifically, first-year students with lower mental SRH may face more psychosocial adjustment problems, which could negatively affect their ability to thrive in college.²⁹ Physical SRH could also be important for academic adjustment, at least in terms of fall credits. College generation status generally did not play a statistically significant role in college adjustment; however, it is notable that first-gens in this study consistently reported more adjustment concerns and lower SRH compared to continuing-gens. Findings underscore the importance of delving more deeply into the influence of pre-college characteristics on adjustment so that universities can better allocate resources towards students who may struggle during the college transition.