Can I Help With That? The Role of Goal Setting in a Technology‐Enabled Peer Mentoring Program

ABSTRACT

College students face significant mental health and academic challenges and often struggle to navigate university resources. Students from marginalized backgrounds face additional barriers to help‐seeking and are less likely to utilize campus services. Peer mentoring programs can help bridge this gap and are most effective when they are goal‐focused. Technology can help scale these programs by offering mentee‐mentor pairs a structured approach to setting goals and facilitating mentors' ability to provide targeted support. In this study, we explored a peer‐mentoring platform that enabled incoming first‐year college students to message their mentors and track goals. Goal setting was significantly associated with students' grade point average at the end of their first semester. After controlling for platform engagement, goal setting remained a significant predictor of self‐efficacy and life satisfaction. Goal setting within peer mentoring relationships may serve as a brief, scalable intervention to promote academic success and well‐being during the transition to college.

1. Introduction

First‐year college students face increasing academic and mental health challenges (Duffy et al. 2020). Many struggle to locate and access appropriate supports, and these difficulties fall especially hard on first‐generation and racially minoritized students who underutilize traditional campus services (Lipson et al. 2023). Peer mentoring is a promising strategy for addressing these gaps, yet many programs rely on nonspecific, friendship‐based models that yield only modest effects, in part because they lack structured, goal‐focused components that translate mentees' priorities into concrete actions (Christensen et al. 2020). Although both goal setting and digital tools have been linked to improved academic and well‐being outcomes (Epton et al. 2017; Morisano et al. 2010; Opie et al. 2024), little is known about how goal setting, when embedded in a technology‐enabled peer mentoring program, is taken up by diverse first‐year students or how such goal‐setting relates to their academic performance and well‐being. This study addresses this gap in the literature by examining whether and how first‐year undergraduates in a large, university‐wide peer mentoring program use a technology‐enhanced goal‐setting feature, which students are most likely to set goals, and whether engaging in goal setting is associated with higher grade point average (GPA), academic self‐efficacy, sense of belonging, and life satisfaction after accounting for key demographic characteristics and program engagement.

1.1. College Student Challenges

Many college students are facing significant mental health, academic, social, and financial difficulties (Lipson et al. 2022). Despite the growing need for mental health and other campus services, college students face considerable barriers to accessing needed support (Cohen et al. 2020; Eisenberg et al. 2012; Lipson et al. 2022). Students often find it difficult to navigate the complex web of university resources and seek timely help to meet their needs (Tyton Partners 2023). Systemic and logistical barriers limit students' ability to efficiently access support (Tyton Partners 2023; Cohen et al. 2020). This can be especially challenging for first‐year students who are learning to adjust to the demands of being a college student and are at heightened risk for poor academic and mental health outcomes (Brandy et al. 2015; Wyatt et al. 2017; Duffy et al. 2020). Furthermore, research suggests that first‐generation college students and students from marginalized racial and ethnic backgrounds face unique barriers to mental health and academic help‐seeking (Abelson et al. 2022; Engle and Tinto 2008; Sapiro et al. 2025). These barriers stem in part from cultural stigma around seeking support from non‐family members, differences in skills and knowledge of how to recruit mentors in higher education, and financial and family obligations that often leave minoritized students with less time for support‐seeking on campus (Chang et al. 2020; Richards 2022). Taken together, these obstacles contribute to marginalized students' underutilization of campus resources and high attrition rates. In fact, approximately 20% of first‐time, full‐time undergraduate students enrolled in 4‐year institutions in the United States drop out during their first year (Hanson 2023). This is especially concerning as earning a college degree has been associated with greater economic earning potential, positive health outcomes, and greater civic engagement (Ma et al. 2016).

While research suggests that students' relationships with faculty are associated with positive academic, social, retention, and career‐related outcomes (Baker 2013; Rios‐Aguilar and Deil‐Amen 2012; Schwartz et al. 2016), students' high demand for support often outstrips the availability of highly trained professionals on college campuses, resulting in many students' early concerns going undetected. High caseloads are consistently cited as one of the primary barriers to advising students. There may also be an unnecessary escalation of students' concerns to faculty or specialized staff, further limiting these professionals' ability to help students who need the most support.

Given these challenges, finding ways to connect first‐year students, particularly those from marginalized backgrounds, with timely and effective campus support can be critical for achieving students' goals and increasing retention (Hurd et al. 2016). Peer mentoring is an effective strategy for scaling support, as peers are often viewed as trustworthy, credible role models who share a college student identity (Collier 2017).

1.2. Peer Mentoring

College students often first seek support from their peers when struggling. Peers play a critical role in shaping students' attitudes towards help‐seeking and decision‐making, and also in connecting students to effective resources (Abelson et al. 2022). Peer mentoring interventions have been positively associated with students' well‐being, socioemotional outcomes, and transition to college (Abelson et al. 2022). Furthermore, peer support among racial and ethnic minority students has been shown to improve academic performance and persistence (Baker 2013). Mentoring programs can support students' adjustment to college and help level the playing field for minoritized students by bridging service gaps in higher education. Despite these benefits, there is evidence that these programs are not always effective, in part, because they are insufficiently goal‐focused (Christensen et al. 2020). Indeed, most mentoring programs are based on non‐specific friendship models that do not incorporate skills‐based or goal‐focused approaches. This has resulted in minimal or insignificant improvements in students' academic and mental health outcomes (Christensen et al. 2020; Raposa et al. 2019). While friendship is necessary within a mentoring relationship, it is not the only component of effective student mentoring (Lyons et al. 2019). Recent studies show that goal‐focused approaches are more effective than friendship models of mentoring; in fact, a recent meta‐analysis of 48 studies found that the effect size for goal‐focused mentoring programs was double that of non‐specific friendship models, highlighting the importance of goal setting and focusing on specific challenges within the mentoring relationship (Christensen et al. 2020).

1.3. Goal‐Setting

Goal setting is known to help individuals focus their attention and behavior and target specific performance outcomes (Locke and Latham 2002). Goals are defined as “what an individual is trying to accomplish; it is the object or aim of an action” (Locke et al. 1981, p. 126) and is an effective and commonly used strategy to promote behavior and health change (Locke et al. 1981; Bailey 2017). A meta‐analysis that examined the effects of goal setting in 141 randomized controlled trials (RCTs) found that goal setting was positively associated with behavior change across a wide range of outcomes and determined that “goal setting is one of the building blocks for designing effective behavior change interventions” (Epton et al. 2017, p. 1189). Research suggests that setting goals can reduce cognitive load and improve task performance and learning across various domains (Locke et al. 1981). Goal setting appears most effective when goals are specific and are at an optimal difficulty level (Locke and Latham 2002; Chan et al. 2019).

A growing body of research suggests that goal setting can improve college students' effort, academic and mental health outcomes, and self‐efficacy (Clark et al. 2020; Opie et al. 2024; Midwest Comprehensive Center 2018; Hsieh et al. 2007). In a randomized control trial, researchers found that a goal‐setting intervention improved academic performance among struggling undergraduate students (Morisano et al. 2010). Another study that examined the effects of a goal‐setting intervention on undergraduate students' academic outcomes showed that goal‐setters had a 22% improvement in academic performance compared to the control group (Schippers et al. 2020). Similarly, Van Lent and Souverijn (2020) found that first‐year undergraduate students who set an academic goal within a peer mentoring relationship earned higher grades than students in a control group.

Goal‐focused approaches that target particular areas of growth and facilitate skill development may be particularly beneficial in university settings, helping students navigate the difficult transition of being a college student while focusing on their emotional, social, academic, and career development. Within this context, a growing body of work suggests that both the number and the content of goals may matter for behavior change and academic success. Research suggests that while students can benefit from holding several well‐aligned goals, pursuing too many or overly diffuse goals may dilute effort and undermine follow‐through (Zhang et al. 2007). Similarly, studies of college students have shown that task‐based goals tied to specific behaviors (e.g., making flash cards based on class notes) are more effective than general performance‐based goals (Clark et al. 2020).

In recognition of the importance of goal setting, universities are structuring programs, such as advising services, to be more goal‐focused, with researchers recommending “that academic advising programs should give greater prominence to goal setting, and that students should be encouraged to set task‐based goals for activities that are important for educational success” (Clark et al. 2020, p.4). While goal setting may be a promising, low‐cost, and scalable intervention, the literature suggests it is most effective when individuals are provided with regular feedback and opportunities for self‐reflection, stay committed to their goals, and have confidence in their ability to achieve them (for reviews, see Locke and Latham 2002; Midwest Comprehensive Center 2018). Mentors may be particularly well suited to support students in setting and achieving their goals by providing support, monitoring progress, delivering feedback, and troubleshooting barriers, a model referred to as supportive accountability (Mohr et al. 2011; Rhodes 2020; Werntz et al. 2026). Despite these benefits, mentoring programs are often challenging to scale, as mentors require consistent training and oversight, which can limit the potential reach of this approach (Burton et al. 2022). Additionally, while Christensen et al. (2020) stresses the importance of goal‐focused mentoring programs, it remains unclear how mentees and mentors should optimally set and track goals within the context of a mentoring relationship.

1.4. Role of Technology

Technology‐assisted tools can help address challenges associated with scaling peer mentoring programs and offer mentee and mentor pairs a structured approach to operationalize, track, and discuss their goals. Moreover, online platforms could help colleges identify student goals and provide tailored and timely support to meet students' needs. Goal setting is frequently used to promote behavior change and positive health outcomes within digital health interventions (Martinez et al. 2021; Suffoletto 2024; Consolvo et al. 2009). For example, popular apps, such as My Fitness Pal, use goal setting to help over a million users track their progress and reach their weight loss goals (Gordon et al. 2019). Goal setting has also been associated with reductions in mental health symptoms across various digital health interventions (Linardon et al. 2019). More recent work has shown the effects of technology‐assisted goal‐setting with a coaching or human component. A systematic review found that goal setting was most often found in effective digital mental interventions for youth and young adults that included human support (Opie et al. 2024). Additionally, Alberts et al. (2023) examined the acceptability of a virtual coach who helped users set SMART goals to promote increased physical activity and provided either generalized or personalized examples of how other users accomplished their goals. Users generally held favorable attitudes towards the virtual coach and found it more motivating when a coach sent personalized messages. Research shows that individuals are most likely to use technology‐delivered interventions and engage in behavior change when a supportive person holds them accountable and provides encouragement to achieve their goals (Mohr et al. 2011; Schueller et al. 2017; Werntz et al. 2023b). Studies within the mentoring literature also suggest that optimal outcomes occur when programs are goal‐focused and attend to the mentee‐mentor relationship (Werntz et al. 2023c; Lyons et al. 2019). A study of 1360 mentee and mentor pairs showed that youths' positive relationships with their mentor and use of goal setting and feedback activities together led to the most significant changes in youth outcomes (Lyons et al. 2019). Taken together, this highlights the need for a blended approach that integrates goal setting in the context of a strong mentoring relationship (Werntz et al. 2023c).

To date, little is known about how first‐year college students use technology‐enabled goal‐setting tools when these are embedded in peer mentoring programs, or whether such tools are differentially taken up across student groups and linked to academic and adjustment outcomes. Building on evidence that goal setting can support student motivation and performance, and that peer mentoring can ease the transition to college, the present study examines a peer‐supported, technology‐assisted platform that invites first‐year students to set and track goals with the support of trained mentors. Specifically, this work asks whether students find it acceptable and feasible to use an online goal‐setting interface; how goal‐setting engagement varies by sex, generational status, race and ethnicity, and age; and whether using the goal feature is associated with students' GPA, academic self‐efficacy, sense of belonging, and life satisfaction. Within this context, we also explored whether the number of goals students selected and whether they set goals in particular domains (e.g., academics, connections, career, health and well‐being) were differentially associated with academic and well‐being outcomes. Of particular interest is whether engagement with the goal‐setting feature varies across student groups. In addition, students' age and pathways into college (e.g., traditional‐age students entering directly from high school vs. older or nontraditional entrants) may shape both their comfort with technology‐enabled tools and their expectations about seeking help from peers. Examining these patterns can clarify whether a technology‐enabled goal‐setting feature extends structured support to students who have historically been underserved, or whether additional design and implementation strategies are needed to ensure that engagement is equitable across diverse first‐year students.

1.5. Present Study

Our team developed MentorPRO, a novel evidence‐based online platform that pairs trained, paid peer mentors with first‐year university students. The platform allows students to securely message their assigned peer mentors, identify weekly challenges, set and track goals, access university resources and event information, and complete in‐app questionnaires (Werntz et al. 2026, 2026a). In this study, first‐year students were matched with paid, older peers from the same university as part of a formal university‐wide mentoring program. Peer mentors were advanced students, in their second year of study or higher, who were trained to use MentorPRO to answer students' questions, provide tailored support and recommendations based on mentees' challenges and goals, and refer them to relevant university resources. Mentors provided ongoing support to first‐year students throughout the Fall and Spring semesters. More information about mentor training and the program can be found in Werntz et al. (2023a).

We evaluated whether first‐year students' use of MentorPRO's goal feature was associated with (1) higher GPA at the end of their first semester of enrollment and (2) well‐being (e.g., sense of belonging, confidence in being a successful student, and life satisfaction). We hypothesized that students who set goals would have higher GPAs, a greater self‐reported sense of belonging, academic self‐efficacy, and overall well‐being. Additionally, given disparities in students' access to financial and social capital, and the underutilization of university services by racial and ethnic minority and first‐generation students, we explored whether there were demographic differences between students who set goals and students who did not (Schwartz et al. 2023; Lipson et al. 2018). This analysis aimed to determine whether goal setting was both acceptable and beneficial for a diverse sample of first‐year students.

2. Methods

2.1. Data Collection

Our analytic sample included first‐year undergraduate students (n = 5782) from a large private university in the northeastern United States who used MentorPRO during Fall 2022 or Spring 2023. Students were excluded if they did not complete a MentorPRO questionnaire (n = 232), use the Check‐In or message feature (n = 139), or had missing baseline demographic information (n = 12; 9 missing sex and 3 missing race/ethnicity). The data flow process is presented in the Supplementary Materials (Supplementary Table 6). Of the final sample, 58.9% identified as female, with the remainder of the sample identifying as male. 35.2% (n = 2034) of students identified as White, 20.0% (n = 1157) as Asian, 10.6% (n = 612) as Hispanic/Latino, 4.3% (n = 246) as Black, and 8.0% (n = 464) as another race. The university did not collect race or ethnicity data for international students (n = 1269). Additionally, 16.9% (n = 980) of the sample was identified by the university as belonging to an underrepresented minority group (e.g., Black, Hispanic, or American Indian), and 8.2% (n = 475) were identified as first‐generation students. Mentors were undergraduate students in their second year or higher, as well as graduate students from the university. Each mentor was assigned to approximately 30 mentees. Mentors were primarily matched to mentees based on their major or academic interests and were assigned to first‐year students who attended the same college. This study was deemed exempt from the university's Institutional Review Board as it was an evaluation of a university‐wide initiative for first‐year students.

2.2. Measures

MentorPRO prompted all mentees to complete a brief seven‐item questionnaire within the app upon logging in for the first time and every 3 months thereafter. No additional eligibility or selection criteria were used to determine which students received the questionnaire. Students could complete the questionnaire up to four times. For this study, a subset of three items was selected a priori for the present analyses, and data from the most recent questionnaire completed by each student were used. Each item was presented on a 5‐point Likert scale from (1) strongly disagree to (5) strongly agree.

2.2.1. Student Well‐Being

Three brief indicators of student well‐being were used in this study: sense of belonging, academic self‐efficacy, and overall life satisfaction. Students' sense of belonging was measured with the item, “I feel part of the [university] community,” adapted from Goodenow (1993) Psychological Sense of School Membership scale. The full scale has demonstrated strong internal consistency across diverse samples (Cronbach's α = 0.77–0.88) and evidence of construct validity. Students' academic self‐efficacy was assessed with the item, “I am confident that I will be a successful student at [the university],” adapted from prior measures of college academic self‐efficacy (Bedewy and Gabriel 2015), which have shown acceptable internal consistency (Cronbach's α = 0.70) and good content and convergent validity. Overall well‐being was measured with the item, “All things considered, I am satisfied with my life as a whole,” adapted from commonly used single‐item life satisfaction measures. Prior research has demonstrated that single‐item life satisfaction measures show good criterion and construct validity when compared with longer, established life satisfaction scales (Cheung and Lucas 2014).

These items were administered as single indicators rather than full multi‐item scales to minimize respondent burden within the MentorPRO platform and maintain feasibility in a high‐volume, university setting. Consistent with prior work that has used single‐item measures of well‐being and related constructs in large applied samples (Abdel‐Khalek 2006; Cheung and Lucas 2014; Hoeppner et al. 2011), items were selected a priori based on their face validity and established use in college student and mental health research.

2.2.2. Student Experiences

The questionnaire also included four additional items assessing mentees' perceptions of the MentorPRO application (e.g., ease of use, perceived helpfulness, and comfort seeking support from a peer mentor). Example items included, “MentorPRO is an easy‐to‐use app” and “When I need suggestions on how to deal with a challenge, I can turn to my peer mentor.”

These items were not included in the present analyses and were used to inform ongoing application development and improvement.

2.2.3. MentorPRO Engagement

2.2.3.1. Check‐In

Students used an optional Check‐In feature to identify their weekly challenges, which was developed based on Weisz et al.'s (2011) Youth Top Problems assessment. Students could select challenges from six categories: Academic Habits, Academic Planning, Career, Connectedness, Finances, and Health & Well‐Being. Students were asked to rate their top challenges and dragged a slider between 0 (not a challenge) and 5 (most challenging) to indicate how challenging they found each domain. Students could complete this Check‐In daily; however, they were sent a reminder to complete this assessment weekly. Engagement was defined by the number of Check‐Ins each student completed.

2.2.3.2. Number and Type of Messages

Mentees were able to exchange secure messages with their mentors through the MentorPRO platform. The number of messages was calculated based on the total messages exchanged between mentee and mentor pairs. We conducted a targeted analysis of MentorPRO messages to identify instances of the words “goal” or “goals,” to assess whether students explicitly discussed their goals with their mentors.

2.2.4. Goal Setting

When students first logged into MentorPRO, they were asked to identify goals for the semester. While setting up their account, students were presented with the following message: “Almost there! But first, let's set some goals so I can best help you this semester. This will take < 1 min. What do you hope to accomplish this semester?” Students chose from a list of goals that were developed using an iterative participatory research approach and were organized into five broad categories (academics, career, health and well‐being, connections, and finances). Students could select multiple goals across categories and could also write their own custom goals. Prior to August 2022, users could set any number of goals; however, after this date, they could only select up to five goals. Students could indicate that their goal was a priority and update their goals through the app at any time. Selecting and tracking goals was an optional feature for first‐year students. For our analyses, a goal‐setter was defined as a user who set at least one goal. We also examined whether the number of goals students selected or the category of goal (e.g., academics, career) was associated with students' GPA and well‐being. The number of goals was defined as the total number of goals a user set in the app, while the type of goal was determined based on whether a student had ever set an academic, career, health and well‐being, or connections goal. Financial goals were not included in the main analyses due to relatively few students setting goals within this category.

2.2.5. Student Characteristics

Demographic and academic variables, such as race, ethnicity, first‐generation student status, age, sex, and high‐school GPA, were obtained from university records and were used as covariates in the main analyses. Sex was recorded using two categories (male and female), consistent with institutional data collection procedures. GPA at the end of each student's first semester (Fall 2022 or Spring 2023, depending on the semester of entry) was also collected from university records.

2.3. Procedures

2.3.1. Description of the Peer Mentoring Platform

MentorPRO is an evidence‐based smartphone and web‐based application that was made available to all incoming first‐year students and their mentors at a large private university. Previous evaluations of MentorPRO have demonstrated the platform's feasibility and acceptability among first‐year college students (Werntz et al. 2023a). Furthermore, greater engagement with MentorPRO has been associated with higher end‐of‐first‐year GPA, as well as greater academic self‐efficacy, sense of belonging, and overall well‐being (Werntz et al. 2026).

In this study, we assessed the impact of a novel approach to goal setting within the context of a peer mentoring relationship. This allowed us to examine the patterns and potential benefits of goal setting across a diverse sample of first‐year undergraduate students. First‐year students were assigned to a paid, trained, and supervised peer mentor as part of an existing mentoring program at the university. Mentees were matched to their mentor based on declared majors and were given access to the platform approximately 1 month before matriculating to help facilitate a successful transition to college. Information about MentorPRO was disseminated through multiple channels. Emails were sent to students during the summer as they prepared to matriculate. Additionally, information about the program was shared with parents and posted on the university's social media platforms. Mentees could opt in to the program by downloading and logging into the app. During the summer, students received messages from their peer mentors who were trained to provide support, answer questions, and share helpful resources. Students were also encouraged to meet their peer mentors in person once they began college.

Students primarily communicated with their mentor via the secure messaging feature and could schedule videoconferencing meetings directly in the app. Mentors were specifically trained to answer students' questions, track students' challenges and goals, provide referrals to university resources and events, and escalate students' challenges when necessary. Mentors were directed to notify a staff member if students rated a challenge as either a ‘4’ or ‘5.’ Mentors were supervised by university staff, who could review messages that were exchanged between mentees and mentors. For a more detailed description of MentorPRO's features, the intervention, and mentor training and supervision, please refer to Werntz et al. (2026).

2.4. Hypotheses

We hypothesized that students' use of the goal‐setting feature (n = 3601) would be significantly associated with their GPA and well‐being. Specifically, we predicted that students who set goals would exhibit higher GPAs and improved well‐being scores compared to MentorPRO users who do not set goals. We tested this hypothesis using regression analysis, where goal setting served as the independent variable, and GPA and well‐being were the dependent variables. We expected to see positive regression coefficients for the association between goal setting and each outcome, controlling for relevant academic and demographic covariates. We also conducted additional analyses to explore the effects of specific goal‐setting characteristics and examined whether the number and type of goals set are significantly associated with GPA and well‐being. Sensitivity analyses were performed to evaluate the robustness of the primary findings and included adjustments for overall MentorPRO engagement (e.g., number of Check‐Ins and messages). We also assessed whether there were significant demographic differences between students who discussed goals with their mentor and those who did not.

2.5. Data Analyses

All analyses were conducted in SAS. We used descriptive statistics to explore demographic and academic differences among groups of students based on their goal‐setting behaviors. This included comparisons between students who used the goal‐setting feature and those who did not, as well as differences related to the number and type of goals set, and whether the words “goal or goals” were mentioned in messages. To test for statistical differences, Student t‐tests and ANOVAs were used for continuous variables, and Chi‐Square tests were used for categorical variables. Statistical significance was determined at the 95% confidence level (p < 0.05).

OLS (Ordinary Least Square) regression was conducted to determine whether the independent variables (i.e., goal setting, number of goals, and type of goals) explained a statistically significant amount of variance in the dependent variables (GPA and well‐being), after controlling for key academic and demographic variables that are associated with academic performance and well‐being. Statistical significance was determined at the 95% confidence level. Lastly, post‐hoc power analyses were performed for each model, and overall results showed adequate power (i.e., more than 80%) to detect medium‐sized effects.

To evaluate the appropriateness of the OLS regression models, diagnostic tests were conducted to assess normality, multicollinearity, and homoscedasticity. The distributional assessment of the dependent variables was conducted using Kolmogorov–Smirnov tests, supplemented by visual inspection of the corresponding histograms. All outcome variables—GPA, sense of belonging to the university, academic self‐efficacy, and overall well‐being, exhibited statistically significant departures from normality (ps < 0.01). These findings suggest that the empirical distributions are non‐normal, which is consistent with the skewed and bounded nature commonly observed in educational and psychosocial measures that utilize Likert‐type scales. However, OLS regression does not require dependent variables or predictors to be normally distributed; rather, the normality assumption pertains to the residuals. Thus, even when variables deviate from normality, OLS estimates remain unbiased and consistent, and statistical inference is appropriate so long as residual diagnostics indicate approximate normality and homoscedasticity.

Multicollinearity among predictors was evaluated using the variance inflation factor (VIF) across all OLS models. VIF values quantify the extent to which each predictor's variance is inflated due to linear dependence with other predictors in the model. Across all models, the observed VIF values fell below commonly accepted thresholds for problematic multicollinearity. Specifically, VIF values greater than 5 are often regarded as indicative of moderate multicollinearity, while values exceeding 10 suggest severe multicollinearity warranting corrective action. In the present analyses, none of the predictors approached these thresholds, indicating that the predictors were not excessively correlated with one another and that the estimated regression coefficients are stable and interpretable.

Finally, heteroscedasticity was evaluated for each OLS regression model using the White test, which assesses whether the variance of the residuals is systematically related to fitted values or nonlinear combinations of the predictors. Across all dependent variables, White test statistics were nonsignificant, indicating no detectable evidence of heteroscedasticity. This suggests that the variance of the error terms remains relatively constant across the range of observed predictor values—an important condition for ensuring that OLS standard errors are unbiased and that hypothesis tests and confidence intervals remain valid. Taken together, these diagnostics support the appropriateness of OLS regression for the present analyses.

3. Results

3.1. Goal Setting

Sixty‐three percent (n = 3601) of students used MentorPRO's goal feature. Students set goals across all five domains: connections, academics, career, health & well‐being, and finances. The most frequently selected goals were related to connecting with peers, academic planning and success, and exploring internships and work opportunities. The five most common goals included joining clubs and organizations (n = 2730), planning course schedules (n = 2347), building time management skills (n = 2300), preparing for and finding relevant work experience (n = 2288), and improving grades (n = 2153) (see Table 1). While slightly less common, students also selected health‐related goals such as improving well‐being (n = 1807), managing stress (n = 1723), and improving health and fitness (n = 1714). Additionally, some students wrote custom goals (n = 5), such as being financially independent, building self‐discipline, and reading more books.

Table 1.

Goal setting by category.

Goal selected	Goal category	N (% of goals set)
Join clubs/orgs	Connections	2730 (7.25)
Plan course schedule	Academics	2347 (6.23)
Build time management skills	Academics	2300 (6.11)
Prepare for and find relevant work experience	Career	2288 (6.07)
Improve grades	Academics	2153 (5.72)
Improve study/organization skills	Academics	1898 (5.04)
Learn about internships	Career	1869 (4.96)
Choose major/minor	Academics	1866 (4.95)
Improve well‐being	Health & well‐being	1807 (4.80)
Build professional network	Career	1770 (4.70)
Manage stress	Academics	1723 (4.57)
Improve health and fitness	Health & well‐being	1714 (4.55)
Find scholarships	Finances	1543 (4.10)
Find my people	Connections	1525 (4.05)
Maintain work‐life balance	Health & well‐being	1421 (3.77)
Find student employment	Finances	1381 (3.67)
Attend campus events	Connections	1354 (3.60)
Volunteer	Connections	1086 (2.88)
Find a future job	Career	1042 (2.77)
Prioritize self‐care	Health & well‐being	1030 (2.73)
Improve financial literacy	Finances	948 (2.52)
Dealing with billing/tuition	Finances	796 (2.11)
Find career resources	Career	508 (1.35)
Get/manage/pay student loans	Finances	422 (1.12)
Explore global experiences	Academics	95 (0.25)
Find research opportunities	Academics	41 (0.11)

3.1.1. Demographic Comparisons

The descriptive statistics (see Table 2) illustrated significant demographic differences between students who set goals (n = 3601) and those who did not (n = 2181). Students who set goals were significantly younger, with a mean age of 19.56 years (SD = 1.69), compared to 21.42 years (SD = 5.03) for non‐goal‐setting students (t = 20.41, p < 0.001). Sex differences were also notable, with a higher proportion of females (66.8%) setting goals compared to males (55.9%; χ² = 70.61, p < 0.001). Significant racial and ethnic disparities were observed (χ² = 420.36, p < 0.001), with a larger percentage of Asian students (74.2%) and White students (71.0%) engaging in goal setting compared to other racial groups. Additionally, first‐generation students were significantly more likely to set goals than continuing‐generation students (χ² = 49.46, p < 0.001). Lastly, students who set goals had a slightly higher mean high school GPA (M = 4.30, SD = 0.29) than those who did not (M = 4.21, SD = 0.33), with this difference being statistically significant (t = 7.38, p < 0.001).

Table 2.

Descriptive statistics by goal setting.

Parameters	Total	Goal setting		t/χ2	p value
		Yes	No
Number of students	N = 5782	N = 3601	N = 2181	—	—
Age, mean (STD)	20.26 (3.48)	19.56 (1.69)	21.42 (5.03)	20.41	< 0.001
Sex, N (%)				70.61	< 0.001
Male	2377	1328 (55.9)	1049 (44.1)
Female	3405	2273 (66.8)	1132 (33.2)
Race/Ethnicity, N (%)				420.36	< 0.001
White	2034	1445 (71.0)	589 (29.0)
Black	246	142 (57.7)	104 (42.3)
H/L	612	384 (62.7)	228 (37.3)
Asian	1157	858 (74.2)	299 (25.8)
International	1269	502 (39.6)	767 (60.4)
Other	464	270 (58.2)	194 (41.8)
URM, N (%)				0.001	0.962
Yes	980	611 (62.3)	369 (37.7)
No	4802	2990 (62.3)	1812 (37.7)
First‐Generation, N (%)				49.46	< 0.001
Yes	475	367 (77.3)	108 (22.7)
No	5307	3234 (60.9)	2073 (39.1)
HS GPA, mean (STD)	4.28 (0.30)	4.30 (0.29)	4.21 (0.33)	7.38	< 0.001

Note: Bold values indicate statistically significant differences, p < 0.05.

Abbreviations: H/L, Hispanic/Latino; HS, high school; STD, standard deviation; URM, under‐represented minority.

We examined the number of goals set by students, revealing a mean of 10.45 goals (SD = 4.59) and 2.73 priority goals (SD = 3.17) (see Table 3). Significant sex differences were found, with females setting a greater number of goals on average (M = 11.05, SD = 4.48) compared to males (M = 9.42, SD = 4.58; F = 108.34, p < 0.001). Race and ethnicity were also significant factors, with Black (M = 11.36, SD = 4.48), Hispanic/Latino (M = 11.29, SD = 4.53), and Asian students (M = 11.22, SD = 4.58) setting more goals compared to White students (M = 10.10, SD = 4.44; F = 16.38, p < 0.001). First‐generation students also set more goals (M = 11.92, SD = 4.54) compared to non‐first‐generation students (M = 10.28, SD = 4.56; F = 42.71, p < 0.001).

Table 3.

Descriptive statistics by number of goals.

Parameters	N	Number of goals mean (STD)	F value	p value
Overall	3601	10.45 (4.59)	—	—
Priority goals	2274	2.73 (3.17)	—	—
Sex			108.34	< 0.001
Male	1328	9.42 (4.58)
Female	2273	11.05 (4.48)
Race/Ethnicity			16.38	< 0.001
White	1445	10.10 (4.44)
Black	142	11.36 (4.48)
H/L	384	11.29 (4.53)
Asian	858	11.22 (4.58)
International	502	9.37 (4.74)
Other	270	10.17 (4.58)
URM			25.31	< 0.001
Yes	611	11.29 (4.51)
No	2990	10.27 (4.58)
First‐Generation			42.71	< 0.001
Yes	367	11.92 (4.54)
No	3234	10.28 (4.56)

Note: Bold values indicate statistically significant differences, p < 0.05.

Abbreviations: H/L, Hispanic/Latino; HS, high school; STD, standard deviation; URM, under‐represented minority.

We also explored whether there were any demographic differences based on the types of goals set (connections, academics, career, and health and well‐being) (see Supplemental Table 1). Students who set connections goals were significantly younger (M = 19.4 years, SD = 0.82; t = 9.49, p < 0.001) than those setting other types of goals. Sex differences were significant for all types of goals, with females being more likely than males to set goals across all categories except for academics (χ² = 3.75, p = 0.05). Racial differences were also observed. Asian and Hispanic identifying students were more likely to set connections (χ² = 26.72, p < 0.001) and career goals (χ² = 17.26, p < 0.001), while Black and Hispanic students were more likely to set health & well‐being goals (χ² = 40.17, p < 0.001). First‐generation students were also more likely to set health & well‐being goals (85.8%, χ² = 18.27, p < 0.001), highlighting the influence of background characteristics on goal type.

We found that among students who used the goal feature, 524 (14.6%) students sent messages to their mentors explicitly related to their goals (see Table 4). A significant difference in age was found between these groups, with those who sent messages about their goals being younger (M = 19.40 years, SD = 0.75) than those who did not (M = 19.59 years, SD = 1.80; t = 2.36, p = 0.019). No significant differences were found for sex, race/ethnicity, or first‐generation status.

Table 4.

Sensitivity analyses I: Descriptive statistics by messages with “Goals”.

Parameter	Total	Message with “goals”		t/χ2	p value
		YES	NO
Number of students	N = 3601	N = 524	N = 3077	—	—
Age, mean (STD)	19.56 (1.69)	19.40 (0.75)	19.59 (1.80)	2.36	0.019
Sex, N (%)				0.01	0.903
Male	1328	192 (14.5%)	1136 (85.5%)
Female	2273	332 (14.6%)	1941 (85.4%)
Race/Ethnicity, N (%)				9.51	0.091
White	1445	180 (12.5%)	1265 (87.5%)
Black	142	24 (16.9%)	118 (83.1%)
H/L	384	57 (14.8%)	327 (85.2%)
Asian	858	135 (15.7%)	723 (84.3%)
International	502	81 (16.1%)	421 (83.9%)
Other	270	47 (17.4%)	223 (82.6%)
URM, N (%)				0.59	0.443
Yes	611	95 (15.6%)	516 (84.5%)
No	2990	429 (14.4%)	2561 (85.6%)
First‐Generation, N (%)				0.76	0.382
Yes	367	59 (16.1%)	308 (83.9%)
No	3234	465 (14.4%)	2769 (85.6%)
HS GPA, mean (STD)	4.30 (0.29)	4.32 (0.28)	4.29 (0.29)	1.55	0.121

Note: Bold value indicates statistically significant differences, p < 0.05.

Abbreviations: H/L, Hispanic/Latino; HS, high school; STD, standard deviation; URM, under‐represented minority.

3.1.2. Regression Analyses

The main findings for GPA are presented in Table 5. In Model I, goal setting was significantly associated with GPA (B = 0.05, SE = 0.02, p = 0.03), suggesting that students who engaged in goal setting had higher GPAs than those who did not. However, when controlling for MentorPRO engagement in Model II, the effect of goal setting was no longer statistically significant (B = 0.04, SE = 0.02, p > 0.05). Interestingly, the number and type of goals set were not significantly associated with GPA (p > 0.05). These findings suggest that while goal setting may have an overall positive effect on GPA, the impact of specific goal types on academic outcomes requires further investigation.

Table 5.

Main output of multivariable OLS regression on outcome of GPA by different primary predictors.

Model	N	Parameters	EST (se)	95% CI	t value	p value
I	5782	Goal Setting	0.05 (0.02)	[0.01–0.09]	2.19	0.028
II	5782	Goal setting (controlling for app engagement)	0.04 (0.02)	[−0.01–0.08]	1.65	0.099
III	3601	Number of goals	0.01 (0.01)	[−0.01–0.01]	−0.23	0.820
IV	3601	Type of goals: Connections	0.01 (0.02)	[−0.05–0.05]	0.04	0.967
		Type of goals: Academics	0.10 (0.06)	[−0.02–0.21]	1.67	0.095
		Type of goals: Career	0.01 (0.03)	[−0.06–0.07]	0.05	0.961
		Type of goals: Health & wellbeing	0.01 (0.03)	[−0.05–0.05]	−0.09	0.931

Note: Bold value indicates statistically significant differences, p < 0.05.

Table 6 includes the main outputs from the OLS regression for the outcome of students' well‐being. The regression results indicate no significant association between goal setting and sense of belonging to the university (B = 0.01, SE = 0.04, p < 0.05). Even when controlling for app engagement, goal setting did not significantly predict students' sense of belonging (p > 0.05). Neither the number of goals set nor the specific goal types were significantly related to this outcome (p > 0.05). These results suggest that goal setting may not play a substantial role in enhancing students' sense of belonging.

Table 6.

Main Output of Multivariable OLS regression on outcome of students’ well‐being by different primary predictors.

Model	N	Parameters	EST (se)	95% CI	t value	p value
Sense of belonging to university
I	5782	Goals setting	0.01 (0.04)	[−0.07–0.07]	0.07	0.945
II	5782	Goal setting (controlling for app engagement)	0.01 (0.04)	[−0.07–0.07]	−0.06	0.955
III	3601	Number of goals	0.01 (0.01)	[−0.01–0.01]	−0.22	0.829
IV	3601	Type of goals: Connections	0.01 (0.03)	[−0.06–0.06]	0.03	0.978
		Type of goals: Academics	−0.11 (0.08)	[−0.26–0.05]	−1.31	0.191
		Type of goals: Career	−0.02 (0.05)	[−0.1–0.07]	−0.34	0.734
		Type of goals: Health & wellbeing	−0.05 (0.04)	[−0.12–0.02]	−1.49	0.136
Academic self‐efficacy
I	5782	Goals setting	0.06 (0.03)	[−0.01–0.13]	1.86	0.063
II	5782	Goal setting (controlling for app engagement)	0.07 (0.03)	[0.01–0.13]	2.07	0.039
III	3601	Number of goals	0.01 (0.01)	[−0.01–0.01]	−0.18	0.858
IV	3601	Type of goals: Connections	0.01 (0.03)	[−0.05–0.07]	0.26	0.798
		Type of goals: Academics	−0.02 (0.08)	[−0.16–0.13]	−0.23	0.820
		Type of goals: Career	0.05 (0.04)	[−0.03–0.14]	1.26	0.207
		Type of goals: Health & wellbeing	−0.01 (0.03)	[−0.07–0.06]	−0.18	0.861
Life satisfaction (Overall Wellbeing)
I	5782	Goals setting	0.09 (0.04)	[0.01–0.18]	2.12	0.034
II	5782	Goal setting (controlling for app engagement)	0.10 (0.04)	[0.01–0.19]	2.17	0.030
III	3601	Number of goals	0.01 (0.01)	[−0.01–0.01]	0.43	0.667
IV	3601	Type of goals: Connections	0.03 (0.04)	[−0.05–0.11]	0.76	0.446
		Type of goals: Academics	0.01 (0.10)	[−0.19–0.21]	0.09	0.930
		Type of goals: Career	0.02 (0.06)	[−0.10–0.13]	0.30	0.765
		Type of goals: Health & wellbeing	0.01 (0.04)	[−0.08–0.09]	0.15	0.881

Note: Bold values indicate statistically significant differences, p < 0.05.

No significant association between goal setting and self‐efficacy was found (B = 0.06, SE = 0.03, p > 0.05. However, in Model II, after controlling for app engagement, goal setting was significantly associated with increased academic self‐efficacy (B = 0.07, SE = 0.03, p = 0.04). The number of goals set and the type of goals did not significantly influence self‐efficacy (p > 0.05). These findings partially support the hypothesis that goal setting can enhance academic self‐efficacy but indicate that the impact is not significantly driven by the number or type of goals set.

We found a positive association between goal setting and life satisfaction (B = 0.09, SE = 0.04, p = 0.03), even after adjusting for MentorPRO engagement and demographic factors. This finding supports the primary hypothesis that goal setting is associated with overall well‐being. However, neither the number of goals nor the specific goal types significantly predicted overall well‐being (p > 0.05), indicating that while goal setting generally benefits well‐being, the content of the goals may be less relevant than the act of setting goals itself.

In summary, even after accounting for demographic factors and MentorPRO engagement, use of the goal feature was significantly associated with greater academic self‐efficacy and life satisfaction. However, the impact of the number and type of goals on specific outcomes varied. Please see Supplementary Tables 2–5 for the complete statistical models.

4. Discussion

In this study, we explored first‐year undergraduate students' goal‐setting behavior within MentorPRO, a novel online mentoring platform that connects peer mentors with college students. The naturalistic study aimed to examine patterns of goal setting among a diverse cohort of students and evaluate the feasibility and acceptability of using a technology‐assisted platform to facilitate structured goal setting within peer mentoring relationships. Given known differences in students' help‐seeking behaviors and utilization of campus services based on racial identity and other demographic characteristics (Schwartz et al. 2023; Lipson et al. 2022), we also explored demographic differences between students based on their goal‐setting behavior.

The optional goal‐setting feature was widely used by first‐year college students, providing initial evidence of its acceptability within the context of a supportive mentoring relationship. Approximately 63% of the study sample engaged with this feature, highlighting its potential to facilitate goal setting among this population. Significant demographic differences were found based on students' goal‐setting patterns. Goal setters were more likely to be younger, female, and first‐generation students. Some significant racial, ethnic, and sex differences were also observed based on the number of goals set. Female, Black, Hispanic/Latino, Asian, and first‐generation students set more goals than male, White, and continuing‐generation students, respectively; however, these differences were small and are unlikely to have practical significance.

The finding that younger students are more likely to use the goal feature aligns with a previous study among students at risk for early school dropout, which showed that younger students may be more likely to set goals within a youth school‐based mentoring program (Martins et al. 2024). Younger college students may have less external support and may take greater advantage of this structured goal‐setting approach. Nonetheless, more research is needed to understand this effect. The observed sex differences are consistent with a growing body of research that suggests that females have greater academic and psychological help‐seeking attitudes and self‐efficacy than males (Brown et al. 2021; Sagar‐Ouriaghli et al. 2019). Research shows a notable gender gap in academic achievement, in which women typically perform better academically (Conger and Long 2010). This underscores the importance of developing targeted strategies to encourage help‐seeking and engagement among male college students.

Furthermore, it was particularly encouraging that first‐generation students were more likely to set goals than continuing‐generation students, as first‐generation students often face additional barriers that limit their access to and use of university programs, advising, and resources, such as limited awareness of campus resources, academic preparedness, marginalization, and stigma (Nguyen et al. 2024; Stebleton and Soria 2012; Schuyler et al. 2020). We also found no significant differences between underrepresented minority students' (e.g., Black, Hispanic, or American Indian) use of the goal feature compared to other students, suggesting that this feature could reduce barriers to accessing goal‐focused support among traditionally underserved student populations. This is a promising finding as both first‐generation and racial and ethnic minority students are more likely to drop out of college and are at higher risk for poor academic and mental health outcomes. This has led to greater calls for stronger academic advising, utilization of pedagogical approaches such as peer instruction and support, and greater transitional adjustment to support first‐generation and other marginalized students' transition to college (Schuyler et al. 2021), which have all been linked with improved college retention and academic achievement.

Platforms like MentorPRO address these objectives by helping to triage students' concerns so that highly trained professionals can assist students with the greatest and most complex academic and socio‐emotional needs. Additionally, the use of peer mentors to support first‐year students' transition to college could help reduce the stigma associated with more formal help‐seeking, making support more accessible for marginalized student groups (Lapon and Buddington 2023). This is especially salient given the importance of students' social integration within the college environment during their first year (Tinto 1993). Peer support has been shown to be particularly helpful for students navigating the transition to college and has been associated with greater academic self‐efficacy, academic performance, satisfaction, and retention among first‐generation and students of color (Schuyler et al. 2021).

Goal‐focused mentoring platforms offer additional benefits, such as providing mentors and colleges with rich information about students' real‐time challenges and needs. College students set goals across all five domains, which suggests that the goal categories (connections, academics, career, health & well‐being, and finances) were relevant for first‐year students. Surprisingly, joining clubs and organizations was the most frequently reported goal, highlighting the importance that first‐year college students place on forming connections with their peers through shared interests and activities. Students could use additional support from both peer mentors and the university on how to select and join student organizations. Many students also set academic‐focused goals, highlighting first‐year students' desire to learn skills to adjust to academic demands and succeed academically. Students would likely benefit from additional education and targeted skills practice, which could be delivered directly through the mentoring platform to enhance time management and organization skills. Peer mentors could then provide supportive accountability for students to complete these exercises and apply the skills to their own context and academic goals (Mohr et al. 2011; Schueller et al. 2017). Most students also selected goals related to improving their well‐being. This is consistent with extant literature, which highlights the current college student mental health crisis (Lipson et al. 2022; Lee et al. 2021; The Healthy Minds Network 2020; Active Minds 2020; Son et al. 2020; Wang et al. 2020). In light of these challenges, it is encouraging to see that students hope to prioritize their mental health during their first year. Mentors could connect interested students with personalized well‐being resources, digital mental health apps, and campus mental health services. Universities could also continue to fund and support programs that promote students' well‐being, such as digital mental health programs, and train mentors to support students' use of these resources (Lattie et al. 2019).

Demographic differences also emerged based on the type of goals students set, highlighting the diverse needs and priorities of first‐year college students. For example, female students set more goals in all categories except academics. It is possible that females may have less of a need to set academic goals because they tend to experience fewer academic challenges compared to their male peers (Verbree et al. 2023). Moreover, younger students were more likely to set connection goals. Younger students may prioritize developing social connections as they transition into the college environment. This aligns with youths' developmental needs for belonging and social integration in college (Tinto 1993). Additionally, Asian and Hispanic students had an increased likelihood of setting career and connection goals, which could partially reflect cultural values that prioritize community, relationships, and career advancement. This finding also suggests that some students from these groups are seeking additional social and career support on campus. Lastly, Black, Hispanic, and first‐generation students were more likely to set health & well‐being goals. This is especially promising given that psychological help‐seeking among marginalized students is disproportionately low on college campuses, despite their high mental health needs (Lipson et al. 2022). Given these students' interest in focusing on their health and well‐being, universities should develop culturally relevant, targeted mental health support that is tailored to first‐year students from underserved backgrounds. Peer mentors could play a key role in sharing and engaging with these resources with students, potentially reducing the stigma associated with mental health help‐seeking. These findings underscore that the types of goals students set vary across demographic groups and highlight how students have unique goals and priorities. Addressing these needs effectively requires collaboration between peer mentors and the university at large.

We also investigated the effects of setting goals on student academic and well‐being outcomes, as well as program engagement. Consistent with prior studies, we found that goal setting was significantly associated with students' GPA at the end of their first semester (Midwest Comprehensive Center 2018). However, after controlling for app engagement, this finding was no longer significant. Previous trials of MentorPRO have associated program use with GPA, which suggests that the use of other MentorPRO features, such as having access to academic support and resources, may contribute more to academic performance than goal‐setting specifically (Werntz et al. 2026). Another study evaluated the effects of two goal‐setting interventions among first‐year Canadian college students and found no effect on academic performance (Dobronyi et al. 2019). Moreover, a recent systematic review of goal‐setting studies in higher education found mixed evidence for the effects of goal‐setting interventions on academic outcomes (Van Jaarsveld et al. 2025), highlighting the diversity of goal‐setting interventions and institutional contexts. For example, some studies demonstrated significant effects on academic performance only for task‐based, rather than performance‐based goals (Clark et al. 2020; Roy and Saha 2019). Greater standardization in reporting goal‐based interventions and increased use of RCTs could help clarify the relationship between goal setting and GPA among college students, particularly in the context of peer mentoring programs.

Given that college students are experiencing significant mental health concerns and high rates of attrition (Lipson et al. 2022), we evaluated the impact of goal setting on multiple aspects of well‐being that have been extensively linked to student success (e.g., sense of belonging, academic self‐efficacy, and life satisfaction). In the present study, goal setting was not associated with a sense of belonging. Although relatively few studies have explicitly examined the association between goal setting and belonging, research consistently indicates that feeling connected to one's university community is linked to greater persistence, more positive help‐seeking behaviors, and better mental health outcomes (Tinto 1993; Won et al. 2021; Haim‐Litevsky et al. 2023; Allen et al. 2024; Gopalan and Brady 2020). Prior work also suggests that peer mentors and faculty play a critical role in fostering students' sense of belonging (Moschetti et al. 2018; Murray et al. 2022). Together, these findings highlight the need for future research to examine how mentoring relationships and goal‐setting interventions could be leveraged to support students' integration into the university community. In contrast, goal setting was significantly associated with college students' academic self‐efficacy and life satisfaction, even after controlling for app engagement. This aligns with existing research that demonstrates the role of goal setting in supporting self‐efficacy and well‐being among adult and youth populations (Midwest Comprehensive Center 2018; Ye 2021; Schunk 1990; Yourell et al. 2025; Opie et al. 2024).

An analysis of messaging data showed that 15% of students explicitly discussed their goals with their mentors. This indicates that goal setting could have been a driver of conversation and may have enhanced the effectiveness of the mentoring relationship (Christensen et al. 2020). Overall, these findings suggest that, aided by technology and a supportive relationship, goal setting can act as a brief, scalable intervention for students to promote positive academic outcomes and overall well‐being. These positive findings are consistent with previous research demonstrating that goal setting can be beneficial in both clinical contexts and mentoring relationships (Jacob et al. 2022; Yourell et al. 2025; Christensen et al. 2020). Although prior studies have shown the benefits of blending social support with goal setting (Chen et al. 2022), more research is needed to understand how mentors can optimally enhance the effects of goal‐setting interventions in higher education.

4.1. Study Strengths & Limitations

To the best of our knowledge, this is the first study to examine first‐year college students' goal‐setting behaviors within an online peer support platform. Our results provide preliminary evidence that this feature was feasible and acceptable for a large and diverse sample of first‐year undergraduate students. Uptake of the goal feature in a real‐world college setting was high, especially given that students did not receive any incentives for using the mentoring platform. This highlights the study's high external validity. Despite the strengths of this study, there are important limitations to consider. This was a cross‐sectional observational study; we were not able to conduct a randomized controlled trial to experimentally manipulate whether students had access to the goal feature. Thus, we are unable to draw causal conclusions regarding the impact of goal setting on students' academic functioning and well‐being. While we controlled for overall engagement with MentorPRO, we cannot isolate the effects of goal setting from the impact of the overall intervention or mentoring relationship. Additionally, this study only looked at the effects associated with setting goals. We did not examine whether students accomplished their goals and how goal completion may be linked to greater academic performance, self‐efficacy, sense of belonging, or overall well‐being. Furthermore, this study was conducted within a private university that had a strong emphasis on career development. Additionally, MentorPRO was deployed as part of an existing mentoring program with support from university staff, which means that the results of this study may not generalize to all universities.

In addition, several well‐being constructs were assessed using single items adapted from validated scales rather than full multi‐item measures. Although this approach is consistent with prior work using brief indicators in applied settings, it departs from the original validation of these instruments at the scale level and likely reduces reliability and attenuates observed effect sizes. Lastly, all demographic data were obtained from university records, which limited how variables were operationalized. Sex was recorded using binary categories (male or female), and additional response options were not provided. As a result, we were unable to assess students' gender identity, which may limit the representativeness of the sample. Accurately measuring demographic characteristics is increasingly important in higher education research to ensure that findings are generalizable and reflect the diversity of student experiences.

4.2. Future Directions and Recommendations for University Mentoring Programs

In future work, researchers could conduct a randomized controlled trial to examine the effects of goal setting on students' academic performance, well‐being, and satisfaction with peer mentors compared to students who did not have access to the goal‐setting feature. Given the optimistic findings that goal setting was associated with positive student outcomes, future studies may wish to conduct a more thorough analysis of messaging data using natural language models to explore the impact of discussing goals with a mentor on student academic and socio‐emotional outcomes. Lastly, only a relatively small percentage of students talked to their mentor about their goals. We might infer this could be improved by providing additional scaffolding for mentors. Mentors could be trained to collaborate with students in developing SMART goals and utilize motivational interviewing techniques alongside supportive accountability to encourage goal completion. Mentors could help students explore and connect their personal values to their goals (Chase et al. 2013), potentially enhancing students' academic and well‐being outcomes.

5. Conclusion

The present findings suggest that embedding structured goal setting within a digital peer mentoring platform is a feasible, scalable strategy for supporting first‐year students' motivation and well‐being. Consistent with prior work showing that brief, structured goal‐setting activities can improve academic performance and self‐regulation, and that high‐quality mentoring relationships promote adjustment and persistence, this study indicates that even an optional, low‐intensity goal feature, when nested in a supportive peer relationship, is associated with higher academic self‐efficacy and life satisfaction during the transition to college. By giving students a simple way to articulate what they hope to accomplish and equipping peer mentors and institutions with timely information about those goals, goal‐focused mentoring programs can help move the field beyond diffuse, friendship‐only models toward approaches that deliberately connect students' priorities to concrete actions and support.

The results have implications for mentoring and goal‐setting programs. First, universities can use brief, technology‐enabled goal tools at scale to invite students, particularly new students, to indicate academic, social, career, and well‐being goals early in their college journey and to revisit these goals over time with a trained peer. Second, integrating goal setting into mentor training and supervision can help mentors normalize help‐seeking, identify when mentees' goals signal elevated risk or unmet needs, and make more targeted referrals to campus resources, thereby increasing the precision and timeliness of support. Third, aggregating goal information across cohorts can provide program staff and institutional leaders with information about first‐year students' most pressing goals, informing the design of student services.

This study also begins to address several gaps in the existing research base. Whereas most prior goal‐setting interventions with college students have been standalone exercises delivered outside of ongoing mentoring relationships, this work examines goal setting within a large, real‐world peer mentoring program, documenting both uptake and associations with academic and well‐being outcomes. In addition, by exploring demographic patterns in goal‐setting engagement, the study sheds light on who is most and least likely to take up a technology‐assisted goal tool, offering early insight into whether such platforms may help extend structured support to students who are often underserved by traditional services. At the same time, these findings underscore the need for subsequent experimental and mixed‐methods work to clarify the mechanisms through which goal setting, in the context of mentoring relationships, drives students' academic success and well‐being (e.g., task shifting, targeted support). In summary, the present findings suggest that embedding structured online goal setting within peer mentoring may be a scalable strategy for supporting first‐year students' self‐efficacy and well‐being.

Author Contributions

Emily Hersch, Jean Rhodes, and Alexandra Werntz contributed to the study's conception and design. Material preparation and data acquisition were performed by Megyn Jasman. Data analysis and interpretation were performed by Yangyang Deng and Emily Hersch. All authors contributed to drafting the manuscript, and all read and approved of the final version.

Conflicts of Interest

Authors Werntz, Deng, Jasman, and Rhodes received limited research support from the university at which the study was implemented. Rhodes is the co‐founder of MentorPRO; Werntz and Jasman are employees of MentorPRO. There are no other relevant or non‐financial interests to disclose.

IRB Statement

This study was deemed exempt from the university's Institutional Review Board, as it was an evaluation of a university‐wide initiative for first‐year students.

Supporting information

Table S1: Descriptive Statistics by Type of Goals Set.

Table S2: Full Output of Multivariable OLS Regression Examining Effects of Goal Setting on Student Outcomes.

Table S3: Full Output of Multivariable OLS Regression Examining Effects of Number of Goals Set on Student Outcomes.

Table S4: Full Output of Multivariable OLS Regression Examining Effects of Types of Goals Set on Student Outcomes.

Table S5: Sensitivity Analysis II: Full Output of Multivariable OLS Regression Examining the Effects of Goal Setting on Student Outcomes, adjusted by MentorPRO Engagement.

Table S6: Data Flow Diagram.

Data Availability Statement

The data that support the findings of this study are available from the first author, EH, upon request.