1 Introduction

The first college year is both highly consequential for students’ individual development, and challenging, as students are required to adjust to a novel environment with academically demanding curricula, and unfamiliar peers and instructors at a new institution. Elevated mental health impairment and high drop-out rates during the first college year continue to be a large concern in higher education institutions (Pedrelli et al. 2015; Stolzenberg et al., 2020). A positive adjustment process during the first college year is a key factor for subsequent commitment, retention, and performance in college (Credé and Niehorster 2012; Tinto 1993, 2017), and thus, it is highly relevant for higher education institutions to better understand how students can be supported in their adjustment process in the first college year. Theoretical and empirical work emphasizes that adjustment to both the academic environment and the social environment matters for college retention and success. In response to the COVID-19 pandemic, universities in the United States, as well as in many other countries, shifted all classes to a remote format and university campuses remained closed. Consequently, social interactions with peers and faculty were limited to online settings, and access to campus resources (e.g., computer labs and libraries) was restricted. Students who entered college in fall 2020 started their freshman year under particularly challenging circumstances, which likely jeopardized a smooth adjustment process. Distance learning institutions and remote learning formats in post-secondary education are growing rapidly and hence, understanding college adjustment in remote learning environments remains highly relevant to higher education after the pandemic. In the present study, we used data from two cohorts of students who started their first college year either before or during the COVID-19 pandemic. We investigate the extent to which students’ academic ability beliefs and confidence in getting support from peers and faculty as indicators for academic and social adjustment predict subsequent desirable college outcomes (i.e., performance, psychological distress, and satisfaction).

2 Theoretical background

2.1 College adjustment

College adjustment is often described as a multi-faceted process with several dimensions (Credé and Niehorster 2012). Baker and Siryk (1984), for example, describe a taxonomy of four dimensions of college adjustment including academic, social, personal-emotional, and institutional adjustment. Academic adjustment refers to the extent to which undergraduate students adapt to the academic demands of college and succeed in mastering study activities and coursework. Social adjustment entails establishing positive social relations with peers and faculty, including participation in campus activities or groups, rather than experiencing the alternative of loneliness and isolation. Personal-emotional adjustment refers to the degree to which students experience positive emotions rather than stress and anxiety towards their courses and campus activities, and institutional adjustment describes the extent to which students form an emotional attachment and sense of belonging to their institution (Baker and Siryk 1984; see also Tinto 1993, 2017). In higher education research, many studies have examined the associations between (academic and social) college adjustment and subsequent desirable college outcomes (e.g., Chemers et al. 2021; Pascarella and Terenzini 1983; Tinto 1993, 2017); while others were designed to investigate factors that facilitate or hinder college adjustment (see for an overview Credé and Niehorster 2012). In the present study, we investigate associations between academic and social adjustment indicators and subsequent desirable college outcomes and therefore, we will review studies that conceptualized college adjustment as a predictor of desirable college outcomes, rather than focusing on college adjustment as an outcome itself.

2.2 Ability beliefs and college outcomes

Tinto (1993, 2017) described academic self-efficacy and a sense of belonging towards the institution as indicators of academic and social integration and central drivers of students’ commitment and college success. Self-efficacy beliefs are “people’s beliefs about their capabilities to exercise control over events that affect their lives” (Bandura 1989, p. 1175). And thus, according to Bandura and others, they should be central drivers of agentic behavior in many different situations, including academic settings and social environments. Similarly, according to the Situated Expectancy Value Theory, confidence in one’s academic abilities has a major influence on subsequent academic success (Eccles et al. 1983; Eccles and Wigfield 2020). Literature on academic self-efficacy and—more broadly—on academic ability beliefs has shown a consistent pattern of positive relations of students’ academic self-efficacy and ability beliefs with subsequent performance outcomes, net of prior performances (Eccles and Wigfield 2020; Gore 2006; Honicke and Broadbent 2016; Ramos-Sánchez and Nichols 2007; Schunk 1989). Chemers et al. (2021), for example, found that freshmen with higher academic self-efficacy obtained higher grades and reported higher satisfaction with and commitment to school at the end of the first year in college (see also Gore 2006).

2.3 Social support and college outcomes

Sense of belonging and positive social relations on campus are often described as indicators of social integration (Tinto 2017) and social adjustment (Credé and Niehorster 2012), and as relevant predictors of college success. Sense of belonging at college refers to the perception of being related and integrated into an institution as a whole (Walton and Cohen 2011). Hoffman et al. (2002) pointed out that one central component of a sense of belonging at college is that students find themselves in an environment where they feel comfortable and confident that they can get help and support by peers and faculty if needed. Prior empirical research investigated the relation between social support and sense of belonging with desirable college outcomes, such as performance and persistence (academic outcomes), and with satisfaction and well-being (psychological outcomes). Students with higher perceived social support and sense of belonging during their first college year are more likely to persist in college (Allen et al. 2008; Nicpon et al. 2006). However, social support and social belonging were not predictive of better course grades and GPA in the freshman year in two studies (Allen et al. 2008; Nicpon et al. 2006).

Regarding psychological outcomes in college, empirical findings on the role of social support by peers and faculty appear more consistent. Undergraduate students who perceive higher social support are less likely to suffer from mental health problems, such as depression, anxiety, or somatic symptoms (Harandi et al. 2017; Hefner and Eisenberg 2009; Reid et al. 2016; C.-C. Wang and Castañeda-Sound 2008). Consistently, studies focusing on positive indicators of students’ well-being revealed positive associations between perceived social support and (life) satisfaction among college students (Coffman and Gilligan 2002; Kong et al. 2015; Yalçın 2011). Particularly in stressful situations, social support can be a relevant resource for good mental health (Etzion 1984; Wang et al. 2014). Wang et al. (2014), for example, showed that the buffering effect of social support on depression among college students was substantially larger, for students who experienced high levels of stress compared to students who experienced lower levels of stress. In sum, these findings show that social support is a relevant resource for college students’ mental health and well-being, particularly when individuals experience high levels of stress.

2.4 College during the COVID-19 pandemic

The COVID-19 pandemic brought a plethora of unforeseen challenges for universities and college students. To prevent the virus from spreading, universities took careful measures that were related to major changes in campus operations and undergraduates’ daily lives. Instruction was shifted to a fully remote format and most campus operations, such as club activities, social events, and in-person support programs were paused. Campus resources such as libraries, computer labs, and cafeterias were closed and campus housing was closed in most cases in the USA. In consequence, most undergraduate students moved back to their families and continued their study activities remotely. Students who enrolled during the COVID-19 pandemic did not even move to the campus or close housing areas. This major shift in response to the COVID-19 pandemic affected students in multiple ways. Recent studies show that undergraduate students reported increased levels of stress and mental health impairment (Buizza et al. 2022; Fruehwirth et al. 2021; Gopalan et al. 2022; von Keyserlingk et al. 2022), and significant challenges with adjusting to the remote learning setting (Soria et al. 2020; Soria and Horgos 2020) during the pandemic.

Recent studies indicate that perceived social support was an important resource for mental health and well-being during social isolation in the pandemic among the adult population in general (Bareket-Bojmel et al. 2021; Grey et al. 2020), as well as for college students in particular (Michikyan et al. 2023; Szkody et al. 2021; Zhou and Yu 2021).

2.5 Background characteristics and college adjustment

While transition to college is challenging for all incoming students, a relevant question in higher education research is whether some students might be at higher risk for struggling with the adjustment process than others. Results from the meta-analysis conducted by Credé and Niehorster (2012) revealed no systematic differences in the adjustment process by gender or first-generation college student background. However, the authors point out that effect sizes varied between studies and that differences in college adjustment by demographic background variables might vary substantially between institutions. Other studies, for example, revealed that first-generation college students reported lower academic ability beliefs (Ramos-Sánchez and Nichols 2007; Wang and Castañeda-Sound 2008), obtained lower grades in college (Allen et al. 2008; Ramos-Sánchez and Nichols 2007; Upcraft et al. 2004), higher levels of isolation and psychological disorders (Hefner and Eisenberg 2009; Wang et al. 2020) and lower sense of belonging and social support in college (Gopalan et al. 2022; Stebleton et al. 2014).

3 The present study

In the present study, we built on the college adjustment literature and investigated the associations between students’ mid-quarter ability beliefs and their confidence in getting support by peers and faculty as indicators of academic and social adjustment with subsequent academic outcomes (performance) and psychological well-being outcomes (psychological distress and satisfaction with their adjustment process) in two cohorts that experienced their first college year under different circumstances. The first cohort started their freshman year in the fall of 2019 with regular on-campus operations. The second cohort started their freshman year in the fall of 2020 when instruction was moved to a fully remote setting and campus operations were shut down to a large extent in response to the COVID-19 pandemic. While many universities returned to regular on-campus operations by fall of 2020, many universities continued offering online courses, and fully online universities count increasing numbers of students (National Center for Education Statistics 2022). Investigating the role of academic and social adjustment in face-to-face and remote teaching settings is therefore relevant to understand college success during and beyond the COVID-19 pandemic.

We addressed three questions. First, we investigated if mid-quarter ability beliefs and confidence in getting support by peers and faculty as indicators of academic and social college adjustment differed between the two cohorts. Second, we investigated the extent to which mid-term ability beliefs and confidence in getting support by peers and faculty predicted end-of-term performance, psychological distress, and students’ satisfaction with their adjustment process after their first term in college. Third, we examined whether associations of confidence in getting support with the outcome variables of interest differed between the two cohorts of the study. The specific research questions and corresponding hypotheses were as follows:

RQ1

To what extent do mid-term ability beliefs and confidence in getting support by peers and faculty as indicators of academic and social college adjustment differ between students who experienced their first college term under regular campus operations (cohort 1) and students who experienced their first college term with fully remote instructions during the COVID-19 pandemic (cohort 2)?

Based on the college adjustment literature, we used mid-term ability beliefs and confidence in getting support as initial indicators of academic and social adjustment to college (Baker and Siryk 1984; Tinto 2017). Students who entered college in fall 2020 experienced their first term under substantially changed campus operations with no in-person interactions with peers or faculty on campus during remote instructions. Such limited possibilities to interact with new peers and faculty shortly after the transition most likely made it difficult to establish new social relationships for these students. Thus, we expected that students from the second cohort would report less confidence in getting support by peers and faculty in the middle of their first college quarter. To our knowledge, no longitudinal study examined changes in social support or social belonging across multiple cohorts of freshman students since the outbreak of the COVID-19 pandemic. Gopalan et al. (2022) investigated undergraduates’ sense of belonging with a cohort that started their freshman year under regular campus operations in fall 2019 and reported no changes in students’ sense of belonging before and shortly after the shift to remote instructions in spring 2020. However, we predicted that confidence in getting support by peers would be lower among students’ who did not get the chance to meet their peers and faculty in person in the first months of their college experience. We had no specific hypotheses about how the shift to remote instruction would affect students’ academic ability beliefs. Therefore, our analysis on potential cohort differences in academic ability beliefs had an exploratory character.

RQ2

To what extent do academic ability beliefs and confidence in getting support by peers and faculty predict students’ end-of-term performance (academic outcome), psychological distress, and satisfaction with their adjustment process after their first term in college (psychological outcomes)?

Based on the findings on the relevance of a positive academic adjustment for subsequent college outcomes (Chemers et al. 2021; Eccles and Wigfield 2020; Gore 2006; Honicke and Broadbent 2016), we expected that students with higher mid-term ability beliefs would attain higher GPAs in their first term at college and would report less psychological distress after their first term at college. Because literature is not consistent about associations between social support and performance (Allen et al. 2008; Nicpon et al. 2006), we had no directed hypotheses regarding the relation of social support and students GPA. Consistent with prior findings on social support and psychological well-being outcomes (Coffman and Gilligan 2002; Hefner and Eisenberg 2009; Reid et al. 2016), we predicted that students who reported higher confidence in getting support by peers and faculty would report less psychological distress. Lastly, we predicted that students who perceived themselves as capable to succeed in their classes (ability beliefs) and who were confident that they could rely on their new social network to get help if needed (confidence in getting support from peers and faculty) in the middle of their first term report higher retrospective satisfaction with their personal adjustment process after the first term in college (see e.g., Chemers et al. 2021; Coffman and Gilligan 2002).

RQ3

Do relations between confidence in getting support by peers and faculty with performance, psychological distress, and students’ overall satisfaction with the adjustment process in the first college year differ between students who started their first college year during regular campus operations and students who started their first college year during fully remote instructions in response to the COVID-19 pandemic?

Although social support is associated with better mental health outcomes in general, empirical studies have shown that the protective effect of social support is particularly relevant in stressful situations (Etzion 1984, Wang et al. 2014). Recent empirical studies showed that college students’ psychological distress increased during the COVID-19 pandemic (Szkody et al. 2021; Zhou and Yu 2021). Furthermore, recent findings indicate that social support can be a relevant resource to buffer against distress during social isolation in the COVID-19 pandemic (Fruehwirth et al. 2021; Gopalan et al. 2022). Based on these recent findings and theoretical models that emphasize the importance of social adjustment and integration for students’ college success (Baker and Siryk 1984; Tinto 2017), we predicted that confidence in getting support by peers and faculty would be particularly relevant for students’ psychological distress, and satisfaction during the challenging transition to college in the midst of the pandemic—a time where many students reported elevated stress and mental health impairment.

4 Methods

4.1 Data and procedure

We used data from the UCI-MUST project, a longitudinal multi-cohort study at a public university in California (Arum et al. 2021). In summer 2019 (cohort 1) and in summer 2020 (cohort 2), all undergraduate students beginning their freshman or junior year were invited via email to participate in the longitudinal study. In cohort 1, 1275 undergraduate students consented to participate in the study. A subsample of 359 undergraduates consented to participate in an in-depth version of the study that included the completion of weekly surveys throughout the academic year. 239 of these students were freshmen and formed the cohort 1 sample of our study (65% female, 58% first-generation college students, 40% belonged to a historically underrepresented minority at college (URM—students who declared their ethnicity as Hispanic, Black, Pacific-Islander, or American Indian/Alaskan Native)). In cohort two, 1323 undergraduate students consented to participate in the study, of which 498 students consented to participate in the in-depth version of the study. 279 of those students were freshmen in fall 2020 and formed the sample of cohort 2 in our study (69% female, 58% first-generation college students, 34% URM students). We used data from survey instruments that were administered from September to February in the academic year 2019/20 (cohort 1) and 2020/21 (cohort 2).

We chose a September to February observation window to have comparable exposures to either 2019/20 in-person college experience or 2020/21 remote college experiences. Indicators for students’ academic adjustment (ability beliefs) and social adjustment (confidence in getting support by peers and faculty) were assessed in the middle of students’ first college term (T1). Hence, students were able to consider their experiences in class and the interactions with their peers and faculty when rating their abilities to succeed in classes and their confidence in getting help form peers or faculty when it is needed. Term GPA was obtained from the student registrar shortly after the fall term has ended (T2). Students’ satisfaction with their adjustment process was assessed shortly after the fall term (T2). Psychological distress was assessed shortly before the start of the fall term (T0) and in week 3 of the winter term (T3). Figure 1 provides an overview about the time points of data collection in the fall and winter term.

Fig. 1
figure 1

Overview abut the measurement points i the fall and winter quarter

Full size image

4.2 Measures

An overview about all items is given in Table 1.

Table 1 Overview of all survey items used in the present study
Full size table

Cohort

We generated a dichotomous indicator for cohort membership. Students who enrolled at university in fall 2019 prior to the outbreak of the COVID-19 pandemic were assigned the value 1 and students who enrolled in fall 2020 during the COVID-19 pandemic were assigned the value 2.

Academic ability beliefs

Students’ academic ability beliefs were measured with three items in the middle of the fall quarter in a course they perceived as their most difficult course in the quarter. Items were based on scales developed by Eccles and colleagues (see Eccles and Wigfield 2020). One example item is: “Over the past weeks, how good have you been in learning new material in your most difficult course [course name]?” Students responded to the items on a 7-point Likert scale form 1—not at all good to 7—extremely good. Internal consistency of the items was satisfactory (cohort 1: α = 0.80, cohort 2: α = 0.90).

Confidence in getting support from peers and faculty

Confidence in getting support from peers and faculty was assessed in the middle of fall quarter with three items for each. Items were based on social belonging scales by (Hoffman et al. 2002). Items on confidence in getting peer support asked about support in course- and study-related matters. One example item is: “How confident are you that you could call another student if you had a question about an assignment?” One example item for confidence in getting faculty support is “How confident are you that a faculty member would take the time to talk to you if you needed help?” Students responded to the items on a slider scale from 0—not at all confident to 100—extremely confident. We used linear interpolation to rescale the responses from 0–100 to a scale from 0–10. Internal consistency of the items was high (confidence in getting peer support—cohort 1: α = 0.87, cohort 2: α = 0.86; confidence in getting faculty support: cohort 1: α = 0.92, cohort 2: α = 0.94).

Performance

We used information on students’ term grade point average (GPA) of their first term at university to measure performance. Term GPA from fall 2019 (cohort 1) and fall 2020 (cohort 2) were obtained from student registrar data. Term GPA combines the obtained GPAs of students across all their courses in a term and can have values between 0 and 4, where 4 stands for the highest possible score.

Non-specific psychological distress

We assessed students’ non-specific psychological distress with the K10 screening instrument from Kessler et al. (2002) in the week before instruction started (T0), and again in February after the fall term ended (T3). This instrument includes ten items on self-reported symptoms that indicate the presence of psychological disorders. An example item is: “In the past 30 days, how often did you feel so sad that nothing could cheer you up?” Students responded to this item on a 5-point Likert scale from 0—never to 4—all the time. The internal consistency of the items was high (T0—cohort 1: α = 0.90, cohort 2: α = 0.91; T3—cohort 1: α = 0.93, cohort 2: α = 0.93).

Adjustment satisfaction

Students’ overall satisfaction with their adjustment to the university was assessed with one item after their first term at the university. The item was: “In the past quarter, how satisfied have you been with your adjustment process to [university name]?” Students responded to the item on a slider scale from 0—not at all satisfied to 100—extremely satisfied. We used linear interpolation to rescale the responses from 0–100 to a scale from 0–10. We used this approach to obtain a scale that has a comparable range to most of the other Likert-scales used in the study.

Background information

We included information about students’ gender, first-generation college student status, and high school GPA in our analyses. This information was retrieved from the university’s administrative records. We used a binary variable (0 = female, 1 = male) as indicator for students’ biological gender. We used a binary variable (0 = continuing generation college student status, 1 = first-generation college student status) as an indicator for the educational background of students’ parents. In the university’s administrative data, students were assigned a 1, if none of their parents had a college degree. Students’ high school GPA was used as an indicator of prior achievement in our analysis. High school GPA was available on a weighted 5.0 scale for all study participants that accounted for the difficulty of high school courses (e.g., when advanced placement (AP) courses were taken; courses with a level of introductory college classes).

4.3 Statistical analysis

We analyzed our data with the program Mplus, version 8.4 (Muthen and Muthen 1998–2017). We used a latent modelling approach and conducted confirmatory factor analyses in a first step to identify latent factors for confidence in getting support by peers, confidence in getting support by faculty, and academic ability beliefs. All models had a satisfactory model fit following the criteria of good model fit described by (Hu and Bentler 1999), with a root mean square error of approximation (RMSEA) of less than 0.08, a comparative fit index (CFI) of more than 0.95, a Tucker-Lewis index (TLI) of more than 0.95, and a standardized root mean square residual (SRMR) of less than 0.06. We tested for metric (i.e., same factor loadings) and scalar (i.e., same factor loadings and intercepts) measurement invariance in all latent variables across the two cohorts. We concluded that models were invariant, when they did not differ by more than |0.01| on the CFI and 0.015 on the RMSEA (Chen 2007). Models reached partial scalar invariance with same factor loadings and same intercepts across both cohorts, except varying intercepts in the third items of ability beliefs and confidence in getting peer support (see Table 2). Models also reached measurement invariance across students’ self-identified gender and first-generation college student status. We did not include tables on these additional invariance tests in the manuscript because the research questions of the study did not focus on group differences by gender or first-generation college student background.

Table 2 Measurement invariance test on mid-quarter ability beliefs, confidence in getting peer support, and confidence in getting faculty support across both cohorts
Full size table

Second, we used regression models in Mplus to address our research questions. In Model 1, we investigated potential cohort differences in students’ mid-term ability beliefs and confidence in getting support by peers and faculty as indicators of students’ academic and social adjustment (RQ1). In Models 2a to 4a, we investigated the extent to which mid-term ability beliefs and confidence in getting support by peers and faculty predicted students’ term GPA, students’ psychological distress after the fall term, and students’ overall satisfaction with their adjustment process after the first term at college (RQ2).

In Models 2b to 4b, we added interaction terms of cohort-membership and confidence in getting support by peers and faculty using the XWITH function in Mplus. These interaction terms allowed us to investigate if the association between confidence in getting support by peers and faculty with students’ term GPA, psychological distress, and satisfaction with the adjustment process varied between cohort 1 and cohort 2 (RQ3). In Models 2c, d to 4c, d (Table 7), we added prior psychological distress from T0 (shortly before instruction started in the fall quarter) as a predictor to the regression analyses. In these analysis we were particularly interested in Models 3c and 3d, which allowed us to investigate if ability beliefs and confidence in getting support by peers and faculty was related to change in psychological distress in the first college year.

In all models, we report standardized regression coefficients. Supplement A provides the Mplus code for the regression analysis to predict fall term GPA (Models 2a–b), which is exemplary for all regression analyses used in the study. The full information maximum likelihood (FIML) approach was used to deal with missing data.

5 Results

Table 3 provides a correlation matrix among all variables added to the models for cohort 1 (below the diagonal) and cohort 2 (above the diagonal). Descriptive statistics in Table 4 show means and standard deviations in students’ academic ability beliefs, confidence in getting support by peers and faculty, fall term GPA, psychological distress, and adjustment satisfaction after the first college term for both cohorts. Students in cohort 1 reported higher mid-quarter confidence in getting support by peers, higher satisfaction with the adjustment process, and lower psychological distress after the first college quarter, compared to students in cohort 2.

Table 3 Correlation matrix
Full size table
Table 4 Descriptive statistics and mean differences in cohort 1 and 2
Full size table

5.1 Cohort differences in mid-quarter ability beliefs and confidence in getting support (RQ1)

Results from regression analyses can be found in Table 5, 6 and 7. Table 5 provides findings regarding RQ1 on cohort differences in mid-term ability beliefs and confidence in getting peer and faculty support. Students of the second cohort (fall 2020) reported statistically significant lower confidence in getting support by peers (β = −0.42, SE = 0.05, p < 0.001). Students’ ability beliefs and confidence in getting support by faculty were comparable in both cohorts. Furthermore, results showed no differences in mid-term ability beliefs and confidence in getting support by gender or first-generation college student status.

Table 5 Regression analysis on cohort differences in mid-quarter ability beliefs, confidence in getting peer support, and confidence in getting faculty support
Full size table
Table 6 Regression analysis to predict term GPA, T3 psychological distress, and adjustment satisfaction after the first college term by cohort membership, background characteristics, and adjustment indicators (standardized regression coefficients with standard errors in parenthesis)
Full size table
Table 7 Regression analysis to predict term GPA, T3 psychological distress, and adjustment satisfaction after the first college term by prior psychological distress, cohort membership, background characteristics, and adjustment indicators (standardized regression coefficients with standard errors in parenthesis)
Full size table

5.2 Mid-quarter ability beliefs and confidence in getting support and subsequent college outcomes

Academic outcomes

Table 6 (Models 2a–b) provides results from regression analysis to predict fall term GPA. Students in cohort 2 received higher grades compared to students in cohort 1 (β = 0.18, SE = 0.05, p < 0.001). First generation college students obtained lower term GPAs than continuing generation (β = −0.13, SE = 0.04, p < 0.05), and students with higher high school GPA obtained higher term GPAs (β = 0.16, SE = 0.05, p < 0.001). Consistent with our hypothesis, students with higher ability beliefs in the middle of the fall term obtained better end-of-term GPAs (β = 0.41, SE = 0.06, p < 0.001), even after controlling for pre-college achievement (high school GPA). Mid-term confidence in getting support by peers or faculty were not related to students’ end-of-term GPA.

In Model 2b, we added interaction terms to investigate if confidence in getting support by peers and by faculty played a more important role for students’ end-of-term GPA in cohort 2, with students who experienced their first college quarter during the COVID-19 pandemic. Although the interaction term of cohort membership and confidence in getting faculty support scratched the 5% significance level, no cohort differences in the role of confidence in getting support with subsequent performance were found.

Psychological outcomes

Results from Model 3a (Table 6) show that male students reported lower psychological distress compared to female students (β = −0.12, SE = 0.05, p = 0.023). Although the regression coefficients of students’ mid-term ability beliefs and confidence in getting peer support on psychological distress were negative, the coefficients were not statistically significant. Students with higher mid-term confidence in getting support by faculty, however, reported lower psychological distress after the fall quarter (β = −0.20, SE = 0.06, p = 0.002). Interaction effects of cohort membership and confidence in getting support by peers and faculty were not statistically significant (see Model 3b).

In Models 4a–b, we investigated the extent to which academic and social indicators of a positive college adjustment in the middle of the first college term predicted students’ overall satisfaction with their adjustment process after the first college term. Students in cohort 2 reported less satisfaction with their adjustment process compared to students in the cohort 1 (β = −0.14, SE = 0.04, p = 0.013). Results further show that higher mid-term ability beliefs (β = 0.16, SE = 0.06, p = 0.012), greater confidence in getting peer support (β = 0.15, SE = 0.07, p = 0.038), and greater confidence in getting faculty support (β = 0.22, SE = 0.07, p = 0.001) predicted higher end-of-term satisfaction with the adjustment process overall.

The statistically significant interaction term of cohort membership and confidence in getting peer support (Model 4b) indicates that peer support was more relevant for adjustment satisfaction among students of the second cohort (β = 0.13, SE = 0.06, p = 0.050). The interaction effect of cohort membership and confidence in getting faculty support was not statistically significant.

Lastly, in Models 2c–d to 4c–d (Table 7) we added prior psychological distress as a control variable to the models. We were particularly interested in Models 3c–d to investigate whether confidence in getting faculty support had a buffering effect on changes in psychological distress across students’ first college term. Prior psychological distress positively predicted psychological distress after the first college quarter (β = 0.61, SE = 0.04, p < 0.001). Furthermore, students in cohort 2 reported higher levels of psychological distress after the first quarter compared to students from cohort 1 (β = 0.16, SE = 0.05, p < 0.001). Hence, psychological distress increased more strongly among students who experienced the first college term with remote teaching and off-campus. When prior psychological distress was controlled, confidence in getting faculty support was no longer predictive for psychological distress after the fall quarter. Hence, although students with higher confidence in getting faculty support reported lower psychological distress after the first college quarter (see Model 3a–b), confidence in getting support did not serve as a buffer against increases in psychological distress in the first college year. The results and proportion of explained variance in students’ term GPA and satisfaction with their adjustment process and the pattern of regression results did not change when prior psychological distress was added to the models (see Models 2a, b vs. 2c, d and Models 4a, b vs. 4c, d).

6 Discussion

6.1 College adjustment under varying circumstances

One aim of this study was to investigate college adjustment across cohorts, who experienced their first college year under different circumstances. First, we asked if students’ adjustment indicators (i.e., mid-quarter ability beliefs and confidence in getting support by peers and faculty) differed among students with different college experiences in their first college term. Students in cohort 1 experienced their first college term under regular face-to-face operations on campus, whereas students in cohort 2 experienced their first college term under fully remote settings during the COVID-19 pandemic. All classes were moved to an online format and instructors offered office hours online, rather than face-to-face on campus. Campus housing closed and most campus operations were shut down. Hence, peer-to-peer interactions were limited to online interactions. As predicted, students of cohort 2 reported substantially lower confidence in getting peer support, whereas, contrary to our predictions, confidence in getting faculty support was comparable between both cohorts (hypotheses on RQ1). This indicates that instructors and students still managed to establish supportive interactions in remote settings. The university strongly encouraged and supported instructors to be responsive to students’ needs during remote instructions. Our findings suggest that instructors were able to provide support during remote teaching and with online office hours and students’ confidence in getting support by faculty remained on a medium level before and during the COVID-19 pandemic.

A possible explanation for cohort differences in confidence in getting peer support could be that peer-to-peer interactions inside and outside the classroom were much more affected by the shift to remote instructions off campus. Most interactions outside the classroom, e.g., through face-to-face study groups, joined lunch breaks, joined housing, or extra-curricular on-campus activities, were unavailable for students in the second cohort. However, such interactions are important factors for positive social integration and adjustment at college (Baker and Siryk 1984; Credé and Niehorster 2012; Tinto 2017). The lack of those face-to-face interactions during the pandemic could explain lower perceived confidence in getting support by peers in the second cohort.

6.2 The role of academic ability beliefs and confidence in getting support for academic and psychological college outcomes

We predicted that students’ mid-quarter academic ability beliefs and confidence in getting support by peers and faculty would positively predict subsequent academic outcomes and psychological well-being outcomes (hypotheses on RQ2). Overall, our findings provide evidence of positive associations between adjustment indicators and subsequent academic and psychological college outcomes.

Academic outcomes

As hypothesized, students with higher mid-quarter academic ability beliefs attained higher term GPAs in their first quarter at college, even after controlling for their pre-college achievement (high school GPAs). Consistent with existing literature and theory, students who felt confident in their academic abilities in challenging learning environments, mastered demanding curricula and coursework and attained higher grades (Bandura 1989; Chemers et al. 2021; Eccles and Wigfield 2020; Gore 2006; Honicke and Broadbent 2016). Consistent to prior studies (Allen et al. 2008; Nicpon et al. 2006), confidence in getting support by peers and faculty was not related to subsequent performance.

Contrary to what we might expect given the stress of the pandemic, students in cohort 2 obtained better GPAs in their first college quarter compared to students in cohort 1. Perhaps students in cohort 2 dedicated more time to their studying because most competing activities, such as leisure activities and work were limited due to social distancing during the COVID-19 pandemic. Furthermore, many universities encouraged instructors to be responsive to students’ hardships and needs during the pandemic, as well as to adopt a flexible grading policy (Chan 2022), which could also contribute to better grades of students in cohort 2.

Psychological outcomes

As predicted, our indicators of social adjustment were particularly relevant for students’ psychological outcomes (see Allen et al. 2008; Hefner and Eisenberg 2009; Nicpon et al. 2006). Higher confidence in getting support by faculty was associated with lower psychological distress in both cohorts.

However, when prior psychological distress was controlled in the models, two interesting findings emerged. First, in the analysis controlling for prior psychological distress, students in cohort 2 reported an increase in psychological distress across the first college term. This finding supports results from recent studies on aggravated mental health impairment among college students during the outbreak of the COVID-19 pandemic (Fruehwirth et al. 2021; Gopalan et al. 2022).

Second, when prior psychological distress was controlled, the associations between mid-term confidence in getting support with psychological distress after the first college term were no longer statically significant. These findings indicate that confidence in getting support by faculty did not have a buffering effect on increases in psychological distress across the first college quarter. Thus, the directionality of associations between higher confidence in getting support by faculty with lower psychological distress of students remains unclear. The results could also indicate that students with elevated psychological distress felt less confident they could obtain support from peers and faculty, which in turn could have hindered their social adjustment in the freshman year. Further research with repeated measures of perceived social support and mental health impairment in the first college year is needed to improve our understanding of the directionality and potential reciprocal effects of both variables.

Besides associations of college adjustment with a negative indicator of students’ mental health (i.e., psychological distress), we investigated the association between literature-based indicators of college adjustment (i.e., academic and social) with students’ personal satisfaction with their overall adjustment process after the first college quarter (Baker and Siryk 1984; Credé and Niehorster 2012). Supporting our hypotheses (RQ2), our findings show that both were central predictors of students’ personal satisfaction with their adjustment. In other words, students who perceived that they could master their coursework and students who were confident they could rely on their new social network to get help if needed, reported higher satisfaction with their adjustment after their first term at college.

Interaction effects

Prior studies have shown that social support is a particularly relevant resource for mental health, when individuals experience high levels of stress (Etzion 1984, Wang et al. 2014). Thus, we hypothesized that social support would be particularly important for students’ well-being in cohort 2 because they experienced their first college quarter during the challenging COVID-19 pandemic. Findings regarding this hypothesis were mixed. As predicted, increases in students’ psychological distress and lower adjustment satisfaction among students in cohort 2 show that the transition to college was particularly challenging during the COVID-19 pandemic. However, contrary to our hypothesis findings showed no statistically significant interaction effects of cohort membership and students’ confidence in getting support on reported psychological distress. Hence, mid-quarter confidence in getting faculty support was similarly related to lower psychological distress after the first college quarter in both cohorts.

Analysis regarding students’ retrospectively reported satisfaction with their adjustment process during the first college quarter revealed small interaction effects of cohort membership and confidence in getting support by peers. Particularly in cohort 2, students who reported higher confidence in getting support by their peers were more satisfied with their adjustment after the first college quarter. This is noteworthy, because students in cohort 2 on average reported lower confidence in getting peer support and lower satisfaction with their adjustment process. Thus, even though average levels of confidence in getting support by peers was lower in cohort 2, associations between mid-quarter confidence in getting peer support and end-of-quarter adjustment satisfaction were stronger in cohort 2. In summary, as predicted, perceived social support during their fall term predicts students’ well-being in stressful situations (Fruehwirth et al. 2021; Gopalan et al. 2022; Wang et al. 2014).

6.3 Limitations

We used data from two cohorts of freshmen students who started college either before or during the COVID-19 pandemic. We consider this multi-cohort design a strength of our study. Nevertheless, we used data from only one university in the United States. Given that universities used different approaches and measures to respond to the COVID-19 pandemic, our findings cannot be generalized to undergraduate students in general. This is particularly true given the vary unique nature of the student at the university in which this study was run: near 50% of the students are first-generation college students, around 40% are Asian and Asian/American, and around 35% belong to a historically underrepresented minority at college. Further research is needed to replicate our findings with samples from other universities and other countries.

To investigate the role of confidence in getting support for college outcomes we used items based on Hoffman et al. (2002). The wording of these items captures two aspects of confidence: a) the confidence of students that they can reach out for help, which can be understood as a facet of self-efficacy for help-seeking, and b) the confidence that peers or faculty would be responsive to the students’ request for help. Hence, we measured the extent to which students perceived support as an available resource for them, rather than the extent to which they used this resource or whether the resource was actually available to them. To deepen our understanding about what aspects of peer and faculty support was most helpful for a positive transition and adjustment to college, further research should also assess the support that students actually received.

We used students’ GPA to operationalize an academic college outcome. GPA is a commonly used outcome variable in the literature on college adjustment (Credé and Niehorster 2012). Nevertheless, when used as an indicator for achievement, this variable has some limitations because it depends on current grading policies. It is known that the grading policies in fall 2019 differed from grading policies in fall 2020, because the instructors at this university were encouraged to be responsive to students’ needs and hardships during the pandemic. This should be taken into account when interpreting the results.

All of our findings share the usual limitations of analyses conducted between persons (for overview, see e.g., Moeller 2021; Molenaar 2004). This includes the concern that psychological studies tend to look for linear regression effects only, which may risk overlooking non-linear relationships. We also would like to remind the reader that in a set of regressions, each path may be driven by different individuals, implying that not all paths in a model may describe the same individuals (Reitzle 2013). To find out how many individuals in a population are described by findings such as ours, intra-individual research designs and analyses should be conducted in follow-up studies. Finally, all effects are net of the influence of the other predictors. Thus, our findings are likely to vary across studies using a different set of predictors in each of the statistical analyses.

7 Implications and outlook

A central finding of this study is that academic and social adjustment are relevant for positive academic and psychological college outcomes during regular face-to-face instructions as well as during remote instructions and off-campus operations during the pandemic. Even though students overall reported lower confidence in getting social support by peers during remote instructions in fall 2020 compared to students in fall 2019, the support was still a significant predictor of students’ adjustment satisfaction after the end of their first fall term in college. Higher education institutions should thus create environments that actively promote social interaction with both peers and faculty, particularly under challenging circumstances and in online environments. Further research is needed to identify specific aspects of social interactions and social support that facilitate students’ thriving in college. Additionally, longitudinal research should investigate reciprocal effects of students’ social interactions and mental health impairment to examine if social support could serve as a targeted resource to facilitate positive college adjustment and buffer against increasing psychological distress.