Supporting students beyond the classroom: faculty knowledge and practices in referring students to campus counseling services

Introduction

In the United States, almost 40% of 18–24-year-olds are college students (National Center for Education Statistics, 2023). This group is especially susceptible to developing common mental disorders, such as mood, anxiety, and substance use disorders (Auerbach et al., 2018; Duffy et al., 2019; Halliburton et al., 2021). In addition to understanding and navigating their symptoms, college students are tasked with many novel experiences at university, such as studying for exams, preparing assignments, attending classes, and developing interpersonal relationships (Gibbons et al., 2019; van Rooij et al., 2018). Beyond academic-related pressures, college students also experience social and financial stress, such as finding social support networks, adjusting to inadequate sleep, adapting to reduced family visits, and lacking financial support (Huang et al., 2024; Mofatteh, 2021; Wang et al., 2025).

Some consequences of poor adjustment to university could include poor academic performance, attrition, and increased psychological distress (de Alencar Ribeiro et al., 2025; Mohamad et al., 2025; Sergeyev et al., 2025; Zając et al., 2024). Hence, college years present students with a challenging transition period, during which they need strong support systems to manage this time in their lives.

At the forefront of students’ daily interactions are faculty. Within those encounters, students may signal their distress by submitting late assignments, missing classes, withdrawing socially, or approaching their professors for individual support (Bulan et al., 2025; Coleman, 2022; Kalkbrenner et al., 2021). When students demonstrate these signs, faculty are uniquely positioned to offer various forms of support, such as extended assignment deadlines, exam concessions, one-on-one conversations with students, and referrals to support services to ensure the success and psychological wellbeing of students (Brockelman and Scheyett, 2015; Di Placito-De Rango, 2018). Notably, when faculty support is absent, students report worsened psychological wellbeing (Baik et al., 2019; Choi and Jeon, 2025; Prananto et al., 2025).

Although most faculty members recognize their unique roles, many are inexperienced in referring students to counseling services (Albright and Schwartz, 2017). In fact, previous research indicates that up to 63% of faculty reported feeling unable to refer a student to on-campus counseling services, and over a third of respondents reported not being able to identify whether a student is struggling with a mental illness (Becker et al., 2002). However, changes in institutional culture and the growing trend of faculty going beyond the teaching role to provide holistic care and wraparound services (e.g., psychological support) may inform their likelihood of helping students in distress (Melvin et al., 2025; Woodruff and Boyer, 2024). Several conditions listed below may influence whether a faculty member successfully refers a student to on-campus services.

First, a faculty member's demographics and employment category may impact their referral behaviors. At the demographic level, previous surveys suggest that women faculty members across American colleges are more likely to engage with and reach out to their students than their male counterparts (Guarino and Borden, 2017; Lindecker and Cramer, 2021; Lipson et al., 2021). Additionally, full-time faculty may be more experienced and familiar with the institutional processes for referring students compared to adjuncts or part-time instructors (Kucirka, 2017; Riba, 2025).

Second, faculty navigate the desire to help students while contending with concerns about objectivity and professional boundaries. Referring students to services can be complicated by fears of encroaching on students’ privacy when addressing visible concerns about their well-being (Brockelman and Scheyett, 2015; Kucirka, 2017; Steven and Ada, 2019). Additionally, faculty often worry about offering unwanted care to students or forming unprofessional, overly personal relationships that fall outside the traditional scope of an instructor's role (Goel et al., 2025; Mohamad et al., 2026; Porter et al., 2025). Of note here, is that despite a strong desire to help, instructors frequently feel unqualified to support students and fear that intervening could do more harm than good (White and LaBelle, 2019). This hesitation is compounded by faculty members feeling uncomfortable engaging with students exhibiting symptoms of mental illness (Becker et al., 2002). Interestingly, part-time employment can further hinder the ability to form close relationships with students, thereby preventing the initiation of an intervention (Riba, 2025).

Third, faculty are often impeded from referring students when they lack structured training on referral procedures. Faculty who receive formal training on mental health services report feeling more confident in intervening and are better equipped to refer students to available resources (Duncan et al., 2025; Goel et al., 2026; Gulliver et al., 2018; Harper et al., 2026; Hughes et al., 2025; Sontag-Padilla et al., 2016). For instance, training can provide familiarity with available campus mental health services, which is significantly related to a faculty member's ability to successfully refer students to the campus counseling center (Becker et al., 2002). Similarly, training can enhance mental health knowledge, and instructors who demonstrate higher mental health knowledge report being more likely to engage with students experiencing mental health issues (Freilich et al., 2020; Gulliver et al., 2018; Halladay et al., 2024).

Current study

Despite a general understanding that faculty play a crucial role in student outcomes, the relationship between the various conditions outlined above remains largely unexamined. The current study aims to examine the association between demographic variables (i.e., gender, age), employment category (i.e., full- or part-time, years at institution), relational concerns (i.e., intrusion on privacy, professional boundaries), and referral training, and how the referral process unfolds in an urban, public college setting. As an additional component, we investigated whether faculty mental health knowledge plays a role in referrals to on-campus counseling services, as a potential reason for the lack of faculty support may be limited mental health knowledge (Becker et al., 2002; Hughes et al., 2018). This knowledge broadly encompasses the ability to identify the early warning signs of psychological distress, understand basic mental health concepts, and recognize the appropriate resources and referral pathways to help students access professional support (Coleman, 2022; Kalkbrenner et al., 2021).

Method

Participants

Participants were full- and part-time faculty members at a large, urban public university system in the northeastern United States. Full-time faculty were those hired by the university for a position on an annual salary. Part-time faculty, also referred to as adjuncts, were categorized as those paid per course rather than receiving an annual salary. For the purpose of this study, part-time faculty were required to have taught at the university within the five years prior to survey dissemination.

Procedure

Faculty members were invited via email to complete an online Qualtrics survey in two waves: full-time faculty first, followed by part-time faculty. We used the university's website to obtain faculty members’ email addresses, which we compiled into two master lists, one for full-time faculty and one for part-time faculty. Every faculty member from both master lists was invited to participate.

The online survey had a cover page that provided an overview of the study, stated that the survey would take approximately 20 min to complete, and that responses were anonymous and confidential. It further stated that participation was voluntary, that participants could discontinue completing the survey at any point, and that for each completed survey, a donation would be made to the National Alliance on Mental Illness (NAMI). The surveys sent to both full-time and part-time faculty were identical, except for an initial question for part-time faculty requiring that they had taught at the university in the last five years before moving on to the remainder of the survey. All recruitment, consent, and administration procedures were conducted in accordance with the Institutional Review Board protocol of our university (anonymized).

Measure

Faculty members completed a questionnaire that was comprised of three sections: 1) faculty experience with student referrals to the college counseling center including their knowledge of the services provided at the college counseling center, their history of referring students to the college counseling center including the number of referrals made and reasons for hesitance in making referrals, their training history, and their confidence in recognizing a student who was struggling with a mental health disorder; 2) faculty knowledge of mental health pertaining to students at the college level including questions about depression, anxiety, panic disorder, suicide, medication, side effects, and where to obtain treatment; and; 3) sociodemographic characteristics (i.e., gender, age, and ethnicity), and academic demographics (i.e., academic rank, academic department, years at the college, and full- and part-time employment).

In the current study, the mental health knowledge items were adapted from the MHLA-c, a validated measure of mental health literacy for college students (Rabin et al., 2021). The survey presented real-world scenarios that inherently captured both acute and chronic distress. For example, items assessed how faculty would respond to a student exhibiting acute panic during an exam (crying, trembling), a student disclosing a prior depressive episode or an eating disorder, and a student expressing active thoughts of self-harm. This vignette-based approach was chosen to measure faculty's practical response tendencies and applied knowledge when confronted with varied student disclosures and behaviors. Specifically, the questions were selected for their relevance and importance to college faculty's knowledge of mental health issues affecting the college population. Each question had three response options and was closed-ended. For the analyses, responses were scored 1 for correct and 0 for incorrect, yielding a maximum composite score of 20 (see Supplementary Materials for sample questions).

Data management and analyses

We used R (version 4.5.2) for all data analyses [MCA was conducted using the TExPosition (Beaton et al., 2014) and the ggplot2 (Wickham and Sievert, 2009) packages], with the threshold for statistical significance set at α = .05 unless otherwise stated. Before data analysis, we generated a complete set of descriptive statistics to assess the normality of the data distributions and test the assumptions underlying subsequent inferential analyses.

Chi-square tests of independence were used to examine the association between the history of referring students to the college counseling center (responses recorded as yes or no) and sociodemographic characteristics such as gender, age, and academic demographics, such as years at the college, full-time/part-time employment, and referral training, with p-values adjusted for multiple comparisons.

Multiple correspondence analysis (MCA) was used to examine whether faculty knowledge of mental health was associated with sociodemographic factors (gender and age) and academic demographics (years at the college, full-time/part-time employment, and referral training). Knowledge of mental health was assessed based on responses to twenty questions related to depression, anxiety, panic disorder, suicide, medication, side effects, and where to obtain treatment. Because these responses were scored as correct or incorrect (i.e., categorical), we used MCA, a method that examines variance across observations measured from multiple categorical variables (Greenacre and Blasius, 2006). Our goal was to understand overall response patterns across the twenty questions and to identify which specific questions contributed most to the differences among participants. MCA extracts multiple dimensions, each representing a part of the variance in the data (denoted by λ), with the first dimension representing the largest variance, the second dimension representing the second largest variance, and so on. The total variance of the dataset (i.e., the sum of all the λ) is called inertia (denoted by I). MCA identifies the variables that drive the variation across observations in each dimension, measured as loadings, which represent the relationship between the variable and a given dimension. The importance of each variable to a dimension is expressed through its contribution, calculated as the ratio of variance of that variable to the total variance of that dimension (Abdi and Valentin, 2007). MCA dimensions are displayed in the form of bi-dimensional maps that depict observations as points, where the distance between observations represents the extent of their differences (i.e., observations that are closer together are more similar in their response patterns, whereas those that are farther apart are more dissimilar).

Inference procedures

To determine whether the extracted dimensions reliably contribute to the variance in the dataset, MCA uses permutation tests, which are akin to a null hypothesis test and evaluate whether the variance (i.e., eigenvalue) for each dimension is greater than would be expected by chance (Berry et al., 2011). To determine the stability of loadings, MCA uses a bootstrap procedure, which generates many (e.g., 1,000) bootstrapped samples that, in turn, create distributions of loadings. These distributions are used to compute a bootstrap ratio (BSR; the ratio of the mean of the bootstrapped loadings to the standard deviation of the bootstrapped loadings) for each variable. The ratios are equivalent to a z-statistic, such that variables with BSRs exceeding ±1.96 (p < .05) are considered reliable for a given dimension (Greenacre and Blasius, 2006). A threshold of ±3.10 [p_boot <.001] was used here to account for multiple comparisons. In addition, bootstrapped confidence intervals are used to examine post hoc group differences (e.g., gender, age, rank), where group means are computed across the 1,000 bootstrapped samples and a confidence ellipsoid is drawn around a given proportion (e.g., 95%) of the bootstrapped means for each group. These ellipsoids play the role of confidence intervals for the means, such that if two ellipsoids do not overlap, it implies that the group means reliably differ in that space (Hesterberg, 1999).

Results

Sociodemographic characteristics

Our final sample consisted of N = 105 respondents. As Table 1 shows, most respondents identified as women (n = 63), heterosexual (n = 80), and White (n = 66). The age distribution was relatively balanced between the two brackets, indicating an even spread across ages 20–80.

Table 1. Sociodemographic characteristics (N = 105).

Other includes non-binary and prefer not to answer.

Other includes prefer not to answer.

Other includes people who selected more than one race/ethnicity without selecting the multiracial option.

In terms of academic demographics, most of the sample held advanced degrees: two-thirds had a doctoral degree (n = 70) and nearly a third held a master's degree (n = 32). Respondents were distributed across a variety of academic fields. The largest segment of the sample belonged to the humanities (n = 35), followed by the social sciences (n = 25). Furthermore, many of the faculty were at their current institution for more than 10 years (n = 60) and nearly two-thirds were employed full-time (n = 66). The distribution of academic ranks was characterized primarily by full professors (n = 42) and lecturers (n = 36; see Table 2).

Table 2. Academic demographics (N = 105).

Other includes PhD students and Master of Fine Arts (MFA).

STEM, science, technology, engineering, mathematics.

Other includes instructor, emeritus professor, adjunct associate professor, adjunct assistant professor, and clinical supervisor.

Knowledge of campus counseling services

Participants’ institutional knowledge about the campus counseling center, self-reported confidence and comfort making referrals, and historical behavioral practices and experiences regarding student referrals to the campus counseling center are presented in Table 3.

Table 3. Referral knowledge, experiences, and practices (N = 105).

Each correct option is out of N = 105.

Although more than half of the sample (n = 56) could not correctly identify the center's physical location, all respondents (N = 105) were aware that the services are free. Furthermore, when asked about the services offered at the campus counseling center, almost all respondents correctly identified counseling for general anxiety (n = 102), depression (n = 101), and grief issues (n = 96).

Regarding their individual capacity to intervene, the majority of faculty expressed being very comfortable making a referral (n = 67) and felt somewhat confident in identifying student distress (n = 62). More than two-thirds of the sample had a history of actively using the referral process (n = 69). Among those who had initiated referrals, referral frequency varied, with almost a third having made 1–2 referrals. Formal preparation was notably rare, and only 15 respondents reported receiving formal referral training.

Respondents were also asked about their referral approach to situations when they wanted to make a referral but did not. In such situations, the most cited reasons included not wanting to intrude on a student's privacy, not believing it is part of the instructor's job, not knowing how to make a referral, and juggling other responsibilities. Table 4 shows the responses for those who encountered such a situation, including samples for open-ended responses.

Table 4. Reasons for referral hesitance (N = 105).

Additional analyses

Chi-square tests of independence examined whether faculty sociodemographic characteristics (gender and age) and academic demographics (referral training and years employed) were associated with a history of making student referrals. Of those variables, a significant association was found between gender and the history of making a referral, χ²(2) = 5.97, p = .047. An examination of the frequencies indicates that women were notably more likely to have a history of making a referral (n = 47) than not (n = 16), whereas men reported an equal likelihood of having made a referral (n = 17) or not (n = 17). However, when applying Bonferroni's adjustment for multiple comparisons (p = alpha/5 [number of tests] = .01), this significant result disappeared and should therefore be interpreted with caution. No other variable was significantly associated with whether a faculty member made a referral. See Table 5.

Table 5. Differences in referral patterns by group (N = 105).

Fisher's exact test was applied because the expected value was below 5.

The level of significance using Bonferroni's correction for the multiple tests is alpha = .05/5 = .01.

Regarding the mental health knowledge items, most faculty scored an average of 12.94 (SD = 2.13) on a scale ranging from 8 to 18. Table 6 displays the subscale descriptive statistics and the measure's internal consistency.

Table 6. Mental health knowledge descriptive statistics (N = 105).

Multiple correspondence analysis (MCA)

We ran a multiple correspondence analysis to examine the variance across participants based on their responses to the twenty mental health knowledge items, scored as correct or incorrect. MCA identified eight dimensions, of which the first two accounted for 72% of the variance. Based on permutation tests, the overall variance was statistically significant (I = 0.012, p_perm <.001), as was the variance explained by Dimension 1 (55%; λ₁ = 0.006, p_perm <.001) and Dimension 2 (17%; i = 0.002, p_perm <.019).

MCA results are displayed as bi-dimensional maps, with each participant represented as a point and colored according to total score [e.g., scores <12 (red—low), 12–14 (orange—medium), and >14 (green—high)]. The x-axis corresponds to the first dimension and differentiates participants based on their overall performance on the mental health knowledge items. Participants with more correct responses appear on the left side of Dimension 1, whereas those with more incorrect responses appear on the right. A post-hoc group comparison was conducted to determine whether these differences in performance levels were reliable. MCA assesses group differences by generating 95% bootstrapped confidence intervals, represented as ellipsoids drawn around the bootstrapped means for each group (e.g., low, medium, high). Ellipsoids that do not overlap indicate that group means differ within this bi-dimensional MCA space. In the current study, differences in performance along the first dimension were found to be reliable (p_boot < 0.001; see Figure 1).

MCA also identifies the important variables that contribute to the differences across participants for each dimension. Specifically, five questions that related to medication (i.e., the danger present when taking anti-anxiety medication while engaging in certain behaviors) significantly contributed to the difference between high and low mental health knowledge scores on the first dimension. The second dimension, represented on the y-axis, differentiated faculty based on their responses to questions about the appropriateness of calling campus security when a student is experiencing severe psychological symptoms. See Table 7.

Table 7. MCA bootstrap ratios for recoded variables (N = 105).

Statistical significance for all items was determined via bootstrap ratio tests (BSR). Values over 3.10 are considered significant in their contributions to the dimensions (i.e., the BSR value must exceed the cutoff value ±3.10, corresponding to p < .001).

Significant variables are bolded.

We conducted additional post hoc group comparisons to examine whether response patterns across participants were driven by personal and academic demographic factors. Differences in response patterns on mental health knowledge items were not driven by a history of referral training and were also unrelated to gender, age, years at the college, and full-time/part-time employment status.

Discussion

This study sought to identify the variables related to faculty members’ history of referrals to the college counseling center for their undergraduate students with mental health struggles. Data gathered were based on responses of full- and part-time faculty members at a large, urban college and included personal demographics such as gender, age, and ethnicity, and academic variables such as employment status, faculty rank, and years at the college. Importantly, as one of the first academic studies on faculty referrals to do so, we incorporated an assessment of a faculty member's mental health knowledge to determine if this was associated with higher rates of reported referrals.

Additionally, we used Multiple Correspondence Analysis (MCA) to examine response patterns across multiple mental health knowledge questions within a single model. In using MCA we were able to: 1) determine that there were reliable differences in mental health knowledge across faculty; and 2) identify which specific areas of mental health knowledge determined a faculty member's overall score on this assessment.

The descriptive data highlighted faculty awareness and willingness to refer students for counseling services. All respondents correctly identified that campus counseling services were free of charge. Additionally, faculty demonstrated a high awareness of the specific treatments offered by the center, with the vast majority correctly identifying that the center provided counseling for general anxiety, depression, grief issues, panic attacks, and relationship issues. This strong foundational knowledge was paired with a proactive willingness to act: more than two-thirds of the sample had a history of actively utilizing the referral process, and nearly 95% of faculty reported feeling either very or somewhat comfortable making a referral. Despite these high levels of awareness and comfort with the services provided, some notable gaps remained and are discussed below.

Factors influencing student referrals

Our study investigated the relationship between student referrals and faculty gender, age, employment status, longevity at the college, and training history.

Though previous literature shows that women are more likely to have a positive attitude regarding professional help for mental health issues, we did not find a significant association between faculty gender and history of referral (Kalkbrenner and Carlisle, 2021; Wendt and Shafer, 2016). The lack of significant findings may be due to the smaller sample size in our study. Similarly, our study found no significant association between faculty age and history of referral. The research on age and referrals shows mixed findings, with some research showing a relationship between age and referrals (i.e., older faculty being more likely to make referrals), while others highlight a multidimensional relationship, where the fixed demographic of age is less valuable in drawing conclusions about referral patterns than faculty's perceived readiness (Kalkbrenner et al., 2021; Lott et al., 1999).

Our findings indicated that full-time versus part-time employment status was not significantly associated with the number of student mental health referrals reported by faculty. This finding was somewhat unexpected, as full-time faculty generally have more opportunities to interact with students through teaching, advising, and other campus activities. Research suggests that greater contact with students should increase opportunities to recognize students in distress and make referrals (Di Placito-De Rango, 2018; Farrer et al., 2015). One possible explanation is that simply spending more time on campus does not translate into greater confidence or preparedness to identify and respond to student mental health concerns. Future research should continue to examine other factors that may better explain faculty referral behaviors.

Years of employment at the college were not significantly associated with the number of referrals made. Faculty with more years of experience may have developed stronger relationships with students or greater familiarity with campus resources, making this finding somewhat surprising. Although previous research has identified years of teaching experience as a potential predictor of mental health knowledge (Khan et al., 2018), others have similarly found that years of teaching were not significantly related to mental health knowledge among university faculty (Raji et al., 2022). It is possible that years of employment alone are insufficient to influence referral behaviors without additional knowledge or experience related to student mental health.

Only a small percentage of participants reported having received formal training on how and when to refer a student experiencing a mental health concern. Surprisingly, prior training was not significantly associated with the number of referrals made, suggesting that faculty who had received formal training did not refer students more frequently than those without training. This finding contrasts with previous research indicating that mental health training can improve faculty confidence and preparedness to respond to students experiencing mental health concerns (Farrer et al., 2015; Freilich et al., 2020; Gulliver et al., 2018; Halladay et al., 2024; Hughes and Byrom, 2019). It is possible that the training participants received varied in content, quality, or duration, making it difficult to observe a measurable effect. Future research should evaluate the effectiveness of standardized, multi-component training programs that emphasize recognizing signs of distress, understanding referral procedures, and increasing familiarity with campus mental health resources (Gulliver et al., 2018; Halladay et al., 2024).

The discrepancy between comfort and confidence

The results of this study illustrate a discrepancy between faculty's comfort and confidence levels with regard to supporting students experiencing mental health challenges: 63.8% reported feeling very comfortable with referring a student to the college counseling center, while only 17.1% reported feeling very confident in their ability to identify mental health challenges. This discrepancy suggests that faculty want to support students experiencing mental health challenges, but they may lack adequate mental health knowledge to effectively do so. We investigated situations where a faculty member wanted to refer a student but did not do so. In response, the primary reason reported was that they did not want to impose on the student's privacy with a mental health issue (e.g., “I felt the student would not want to take the referral or would feel embarrassed by it.”). This finding aligns with previous literature showing that faculty experience uncertainty on how to approach mental health issues with their students, particularly regarding the appropriate level of involvement with a student's personal life (Marais, 2024; Price et al., 2021). Similarly, a systematic review by Kreuziger and Snethen (2025) found that faculty experiences with student mental health issues often involved struggling with maintaining student-instructor boundaries so that they did not become counseling relationships.

Another interpretation of the results pattern could be a lack of personal responsibility when a student expresses psychological distress. Indeed, 53.3% of surveyed faculty did not know where the campus counseling center was located. Drawing on social exchange theory, individuals weigh the benefits of providing help in a stressful situation, and in this case, faculty who lacked the competence to make referrals may have perceived encounters as sources of stress and helplessness (Alvis et al., 2023; Homans, 1958). Therefore, when faculty lack the resources to support students, their prosocial and altruistic behavior is thwarted. The gap between faculty comfort and confidence therefore underscores the need for targeted, multifaceted interventions that equip faculty with the knowledge and confidence needed to support students effectively.

The role of mental health knowledge

Scores on the mental health knowledge questions indicated the majority of faculty members scored within the average range, supporting previous findings that there is a need to enhance faculty mental health knowledge (Miles et al., 2020). MCA identified two dimensions that distinguished high scorers from low scorers in the mental health knowledge questions. Those who scored high on the overall measure were able to correctly identify the level of danger present in activities carried out when taking anti-anxiety medication and the appropriateness of calling campus security when a student was either having a panic attack or experiencing suicidal ideation. These findings suggest that faculty who understand these nuanced areas of mental health knowledge have a better grasp of the mental health challenges students experience.

One possible explanation for faculty knowledge of medication use and available resources may stem from their personal or professional lived experiences (Gumus and Avci, 2024). Retrospectively, the current study was limited by the omission of a question regarding faculty members’ personal experiences with mental health. We excluded this to respect participant privacy and minimize response burden. Although our data do not reveal a distinct pattern of referrals, the literature demonstrates that personal lived experience with mental illness can influence faculty mental health knowledge and shed light on why some faculty are more likely to make referrals (Kalkbrenner et al., 2021; Miles et al., 2020). This experience may be further likely given the growing use of medication (from as low as 15% in 2013 to over 20%–30% in 2019) to treat depression and anxiety among American college students (Hoeflich et al., 2023).

Limitations and future directions

The present study has some notable limitations. First, our sample size was relatively small, which may have reduced the statistical power to detect significant differences across certain variables, such as the relationship between receiving formal training, sociodemographic and demographic variables, and the frequency of student referrals. This limitation, combined with the sample's broad heterogeneity in terms of disciplines or rank, may have masked more nuanced demographic differences. The current sample was drawn exclusively from full- and part-time faculty at a single large, urban college in the northeastern United States, which may have limited the range of perspectives and experiences of faculty making referrals to counseling centers. Moreover, we obtained faculty email addresses from the publicly available institutional websites and may have included individuals who were no longer affiliated with the institution or were unavailable because of sabbatical, retirement, medical, or other leave. Thus, the size of the eligible and reachable population could not be reliably determined. To address these limitations and improve the generalizability of the findings, future studies should recruit larger stratified samples, attempt to obtain confirmation of currently affiliated faculty, and expand data collection to include multiple colleges and universities.

Second, the questions used to evaluate faculty mental health knowledge, though based on a validated measure of mental health literacy (Rabin et al., 2021), may not have accurately captured the constructs underlying faculty mental health knowledge. Consequently, faculty mental health knowledge should be assessed more comprehensively in future studies to better capture how faculty understand and respond to student distress.

Recommendations for instructor training

We propose a two-pronged approach for instructor training that focuses on 1) knowledge of the referral process to the college counseling center and 2) recognition of various psychological disorders and their symptoms. Regarding the former, instructors should be able to answer the following questions: Where is the Counseling Center at my university? How do I refer a student to the center? What services are provided there? If I need help making a referral, who is the contact person? Regarding the latter, instructors should be able to answer the following questions: What are the most common disorders that tend to surface during the college years? What symptoms are associated with these disorders? What behaviors should I be most concerned about in a student? How can I gain language to articulate what I’m seeing in students?

To answer these questions and provide multi-faceted, specific training to instructors, we propose integrating training into formal faculty seminars, much as instructors at many universities are required to participate in annual trainings on topics such as workplace violence, anti-discrimination, cybersecurity, and syllabus development. Using digital platforms that combine lectures, readings, and case studies, such trainings would be eye-opening, empowering, and actionable, and, most importantly, would contribute to the psychological well-being of students across the college population.

Conclusion

The present study provides insight into factors associated with faculty referrals of undergraduate students to campus counseling centers. Faculty members had considerable diversity across age, gender, academic department, employment type, and faculty rank and were found to engage with the referral process for students experiencing mental health concerns. Results identified a gap between faculty members’ comfort with making referrals and their confidence in recognizing students experiencing mental health challenges. Although demographic characteristics, employment variables, and prior referral training were not significantly associated with self-reported referral history, specific aspects of faculty mental health knowledge emerged as an important area for further investigation. Strengthening faculty mental health knowledge through targeted training may improve early identification and referral of students in distress. Future research with larger faculty samples across multiple colleges is needed to better understand the myriad of factors that promote effective faculty engagement in supporting student mental health.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/feduc.2026.1971647/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Melanie Jayne Beckett, Arden University, United Kingdom

Reviewed by: Maura Pilotti, Prince Mohammad bin Fahd University, Saudi Arabia

Helen Meldrum, Bentley University, United States