Supporting First-Generation Undergraduates Through Embedded Writing Tutoring: Emerging Insights from a Pilot Study

Abstract

Writing is essential across disciplines, yet undergraduate programs must balance writing instruction with discipline-specific content. To support writing development, we piloted an embedded writing tutor (WT) in two core public health courses serving primarily first-generation, low-income students of color. In this model, a tutor familiar with course content is integrated into the classroom to supplement traditional writing center support. Our aims were to examine (1) students’ perceptions of the WT compared to the university’s writing center, (2) the WT’s experiences and effective tutoring strategies, and (3) the instructor’s perspective on implementing the program. Using qualitative methods, the WT recorded field observations, the instructor compared course progression to prior semesters without embedded support, and students completed end-of-semester evaluations. Thematic analysis indicated that students valued the tutor’s accessibility, patience, and direct feedback, though perceived usefulness varied by course, likely due to differences in assignment structure. Challenges included role confusion and inconsistent feedback. Suggested improvements included requiring draft submissions, clarifying the tutor’s role, and aligning tutor and instructor feedback. Quantitative ratings of satisfaction were higher for the WT than for the writing center. Although the sample size was moderate (N = 79), these findings suggest embedded tutoring is a promising, equity-focused strategy for discipline-specific writing instruction. In the context of budget constraints in higher education, exploring alternative tutoring and pedagogical support models remains essential, particularly for underserved populations.

1. Introduction

Colleges and universities are increasingly challenged to create equitable educational environments as the number of first-generation (first-gen) and low-income college students continues to grow. In 2020, over half (56%) of U.S. undergraduates identified as first-gen (Hamilton, 2023). Many of these students also come from racially minoritized backgrounds, reflecting broader demographic shifts in the U.S. (Hutchens et al., 2011). However, systemic inequities in K–12 education, such as under-resourced schools and limited academic preparation, place these students at a disadvantage upon entering higher education (Dover, 2009; Avery & Daly, 2010).

These disadvantages are compounded by additional challenges—such as navigating an unwelcoming campus climate, experiencing cultural incongruity, and lacking adequate faculty mentorship—all of which are more acute for students of color (Matos, 2021). These structural and social barriers contribute to higher attrition rates: first-gen students are four times more likely to leave college than their continuing-generation peers, often departing after their first year (Engle & Tinto, 2008). Financial strain, low self-perception, and a diminished sense of belonging are recurring themes in the literature on first-gen and low-income college students (Jehangir, 2010; Gibbons & Borders, 2010; Wilbur & Roscigno, 2016).

Language adds another layer of complexity. Roughly 20% of college students nationally speak a language other than English at home (Doolan & Miller, 2012). For these multilingual students, academic writing poses a significant challenge, especially when the linguistic expectations of higher education are unfamiliar. As writing is a core academic skill that underpins both learning and critical thinking (Gee & Gee, 2007), students who struggle with writing are often at a disadvantage across disciplines.

Effective writing is also deeply connected to academic persistence and students’ sense of agency. Students with stronger writing skills are more likely to feel confident, participate in coursework, and persist toward degree completion (Kamler & Thomson, 2004; Conefrey, 2021). Recognizing this, U.S. universities have long invested in writing support systems. Writing centers, established as early as the 1900s, have played a central role in providing general writing assistance (Molina-Natera, 2017). Yet, these centers often operate independently of course context, which can limit their relevance for students navigating discipline-specific writing demands.

To address these limitations, some institutions have turned to embedded writing tutor models. Unlike traditional writing center tutors, embedded writing tutors are integrated directly into specific courses, providing targeted, contextualized support aligned with course content and expectations. Because the challenges of supporting diverse student populations, fostering academic writing skills, and improving retention are common across higher education systems worldwide, embedded tutoring offers a flexible, low-cost approach adaptable across disciplines and institutional contexts. This is particularly valuable for institutions with growing numbers of international students who may need guidance in navigating unfamiliar academic conventions. The model has shown promise in fields such as law and the social sciences (Drennan & Keyser, 2022). By attending class, building rapport with students, and providing feedback tailored to specific assignments, embedded tutors deliver highly relevant and immediately applicable writing support.

Although comparisons across embedded tutoring studies are difficult due to variations in design, tutor training, implementation, and student populations, research consistently shows that courses with embedded tutors have higher retention rates and grade point averages than those without them (Channing, 2019; Kurzer, 2023). In Channing’s 2019 study, retention rates, GPAs, and course completion rates were tracked over several semesters, revealing a consistent advantage for courses with embedded tutors. For example, across three semesters, English sections with an embedded tutor had a 76% retention rate compared to 71% without; in math, retention was 52% versus 48%, and in chemistry, 86% versus 82% (Channing, 2019).

Another study with a student population similar to ours compared the experiences of students in courses with an embedded tutor to those who used a traditional walk-in writing center (Kurzer, 2023). Over five quarters, 16 sections of writing courses incorporated an embedded tutor, involving 280 students in the treatment group and 400 in the control group. Students with embedded tutors reported statistically higher ratings for general course instruction, teacher feedback, overall helpfulness of the course, and tutor feedback (p. 33). Similarly, Divan (2013) found that embedded writing programs can support specific student groups—such as international students—by helping them adapt to new academic writing conventions, further illustrating the broader benefits of the embedded model.

Still, the empirical evidence on the effectiveness of this model—particularly in undergraduate courses serving structurally marginalized students—remains limited (Conefrey, 2021; Perez et al., 2021). Moreover, many studies in this area emphasize student perceptions of tutoring support rather than examining how these supports impact students’ writing development, self-efficacy, or academic integration (Palmquist & Young, 1992; Limpo & Alves, 2014).

Given the importance of writing to student success—and the particular challenges faced by Latinx, first-gen, and low-income students—there is a need to better understand how embedded writing support functions within equity-focused classrooms. This study addresses that gap by examining the implementation of an embedded writing tutor in two undergraduate public health courses with high enrollments of Latinx and first-gen students. By focusing on student evaluations and reflections, the pilot study seeks to investigate how embedded support is experienced in a writing-intensive, discipline-specific context.

The research questions guiding this pilot study are

• How do students perceive the helpfulness of the embedded writing tutor relative to the university’s traditional writing center?
• What benefits and challenges do students identify in working with an embedded writing tutor in a public health course?
• What insights do student evaluations offer for improving embedded writing support in equity-focused classrooms?

2. Methods and Analysis

This study was conducted in Spring 2024 (January–May) at a public, four-year research university in California. The institution is federally designated as a Minority-Serving, Hispanic-Serving, and Asian American and Native American Pacific Islander-Serving Institution. Approximately 62% of students identify as first-generation college students, and 58% are Pell Grant eligible. In order to be eligible for a Pell Grant, students must demonstrate exceptional financial need.

The study was approved as exempt by the university’s Institutional Review Board (#2024-73). All observations and student evaluation data were de-identified, and the course instructor did not access the data until final grades were submitted. All data (i.e., field notes and student evaluations) are stored on the Open Science Framework and available at [https://osf.io/5r8th/?view_only=e8fea824c2b94d96bc5618feae23f2c0] (accessed on 10 August 2025).

2.1. Intervention

An embedded writing tutor (WT) was integrated into two writing-intensive, upper-division public health courses—Health Promotion and Health Communication—taught by the same instructor. In this context, our embedded WT was a graduate student assigned to support two specific courses by providing individualized writing support outside of class, in alignment with course assignments and content. The tutor was not present during class sessions but was integrated into the assignment schedule and communication with the instructor. Each public health course enrolls approximately 60 students and culminates in discipline-specific writing assignments. In prior semesters, students in these courses were required to visit the university’s Writing Center; however, based on inconsistent outcomes and mixed student feedback, the instructor piloted an embedded writing tutor model to provide more accessible and discipline-aligned support. Therefore, during the semester of this study, students were not required to use the Writing Center.

The embedded WT for our pilot study was an advanced graduate student, a native English speaker with strong academic writing experience, and though his graduate training was not in public health, he had work experience in the field. Employed 20 h per week, the WT worked with both courses throughout the semester. Although he was not required to attend class with the students, he did attend in the beginning of the semester to introduce himself and his responsibilities, as well as two additional times throughout the semester. His responsibilities included one-on-one and small group tutoring sessions, preparation and reflection, and collaboration with the course instructor on assignment scaffolding.

Before beginning tutoring, he reviewed the most recent literature on writing tutoring and developed strategies for tutoring our students. This included, modeling writing and critical thinking, using structured templates for editing, guided questioning to encourage students to think critically about their writing, and scaffolding writing tasks over time (Newell et al., 2011; Rogers & Graham, 2008; Wischgoll, 2016). These strategies align with evidence-based approaches to improve writing fluency and self-efficacy, particularly among underprepared or underserved students.

All sessions were held in English, via Zoom, and scheduled in coordination with assignment milestones. Meetings ranged from 15 to 30 min, depending on student needs and project complexity. All students were required to attend a minimum of three meetings with the WT during the semester, though many scheduled additional sessions as needed. During these mandatory visits, the WT offered support in areas such as topic development, writing organization, clarity, integration of evidence, grammar, and editing. He also took detailed observational field notes after each meeting, documenting student needs, strategies used (e.g., modeling or guided questioning), and students’ responsiveness and engagement. These reflections informed ongoing adjustments to the tutoring approach.

2.2. Course Descriptions and Assignments

Health Promotion focuses on assessing community health needs and designing theory-based interventions. The primary writing assignment required students (in groups of three) to develop a comprehensive health promotion program tailored to a real-world population. This course incorporates evidence-based teaching practices such as flipped classroom instruction, project-based learning, collaborative group work, scaffolded assignments, multiple drafts with a focus on improvement, low-stakes writing tasks, and detailed instructor feedback on writing.

Health Communication introduces students to health communication theories and strategic messaging. Students completed two types of writing assignments: (1) scholarly article summaries using a structured template to develop editing and synthesis skills; and (2) a public health policy brief requiring concise, data-driven argumentation and recommendations. The course is designed using high-impact practices such as a flipped classroom model, repetitive writing assignments with clearly defined structures, in-class writing days, low-stakes practice opportunities, and detailed rubrics to guide student performance. Although we did not collect demographic data from students in the two courses studied, institutional survey data from the 2022–2023 academic year indicate that among broader population of public health majors, approximately 62% speak a language other than English at home, 72% are first-generation college students, and 68% are Pell Grant eligible. In terms of racial and ethnic composition, the public health major student body is 57% Hispanic, 20% Asian, 7% Black, 12% international, and 1% White.

2.3. Data Collection

Data for this study were collected through the following sources:

WT Field Notes: The writing tutor maintained detailed observational notes after each tutoring session, documenting student challenges (e.g., difficulty forming arguments, trouble integrating sources), strategies used (e.g., modeling, scaffolding, probing questions), and reflections on students’ responsiveness, engagement, and progress.

Student Evaluations: On the final day of class, students were invited to complete an anonymous evaluation of the embedded WT. The survey included both closed- and open-ended questions. Those who had previously visited the university’s Writing Center were also asked to compare the two support options. A total of 79 students completed the evaluation (44 from Health Promotion, 35 from Health Communication). Responses were analyzed descriptively and qualitatively to assess trends in student experience. The evaluation was as follows:

• Which course(s) are you enrolled in?
  • Health Promotion
  • Health Communication
• How helpful were your visits with the Writing Tutor, on a scale of 1 (not at all) to 10 (extremely)?
• If you have ever attended the university’s Writing Center before, how helpful were those visits, on a scale of 1 (not at all) to 10 (extremely)?
• If you ever attended the university’s Writing Center before, which visits did you find more helpful and why?
• Please provide any comments or suggestions based on your experiences with the Writing Tutor this semester.

• Professor Observation: The professor documented her reflections throughout the semester, comparing student engagement and writing quality to prior cohorts who did not have access to an embedded tutor. These observations focused on assignment quality and levels of student independence. She documented reflections after each major assignment and discussed her observations with the WT afterwards.

2.4. Data Analysis

All data analyses occurred after the end of the semester. In order to gain an understanding of the WT’s experiences, his field notes were reviewed and recurring patterns were identified and coded deductively based on the factors that led to challenges for the WT and those that were helpful (i.e., helpful strategies). The same approach was implemented for the instructor’s observations. The observational notes were reviewed and thematically coded deductively for strategies that were helpful, challenges faced by the implementation of the WT, and unexpected challenges.

As the student evaluation forms provided both qualitative and quantitative data, these analyses were completed separately. Quantitative items from the student evaluation survey (e.g., Likert scale ratings) were analyzed descriptively, calculating means and frequency distributions to assess overall perceptions of the WT and the university writing center. Open-ended responses were reviewed multiple times then coded inductively for patterns, including perceived benefits of embedded support, comparisons to the writing center, and suggestions for improvement. Codes were refined and collapsed into themes that highlighted students’ evaluations of the WT’s accessibility, course-specific feedback, and personalized guidance.

3. Results

3.1. Writing Tutor’s Field Notes

3.1.1. Challenges

The WT observed recurring writing challenges, including clarity, grammar, paragraph organization, sentence structure, citations, and spelling. Given the 15 to 30 min limits on tutoring sessions, the WT prioritized one or two major issues per session rather than addressing every concern. The broad range of writing difficulties made it difficult to provide quick, targeted guidance.

Many students came to the tutoring sessions without drafts prepared, making it difficult to offer substantive feedback. In these cases, the WT accessed previous writing assignments on the learning management system (LMS) and asked students to reflect on their strengths and areas for improvement. However, these sessions felt less productive and more like the students were merely checking off a requirement for the assignment. Fortunately, due to the embedded tutor design, the WT had access to the course LMS and could access students’ assignments and grades. If he did not have access, it would have been more challenging to adapt in these situations.

Group tutoring sessions introduced additional challenges. Since the Health Promotion students met with the WT in their project groups, it was uncomfortable and awkward when one student’s writing skills were significantly weaker than their peers’. Additionally, students sometimes received conflicting writing advice from other sources (e.g., the teaching assistant or university Writing Center) that conflicted with the professor’s and WT’s, leading to confusion.

Another challenge involved how students processed feedback. For instance, the health promotion program prompt explicitly stated “Do not use long quotes.” The WT reviewed this guideline during the first session. However, one student persisted in using long quotes, despite losing points and receiving professor comments reiterating the rule. When asked why he had not removed them, the student explained that he believed that he had shortened them enough and that the Writing Center had approved the approach—despite the professor and WT’s clear instructions.

3.1.2. Helpful Strategies

With the challenges described above, a few helpful strategies emerged. First, was reflective questioning, or encouraging students to assess their own writing and problem solve ways to improve. The other helpful strategy was modeling writing, where the WT shared examples of academic writing to compare and contrast with. For example, demonstrating how to make a sentence more concise by cutting the unnecessary language. This was especially helpful due to the limited amount of time available.

Additionally, access to the LMS allowed the WT to track patterns and progress in students’ work. While students could share their feedback with others, the WT could directly review grades and professor comments on assignments, making it easier to identify key areas for improvement. The tutoring sessions ran smoothly when students had a draft prepared for the WT to review.

3.2. Student Evaluations

On the final day of class, students were invited to anonymously evaluate their experiences with the embedded Writing Tutor (WT). Of the 119 students enrolled across both courses, 79 students completed the evaluation (44 from Health Promotion and 35 from Health Communication).

When asked to rate the helpfulness of their visits with the WT on a scale from 1 (not at all) to 10 (extremely), the overall mean rating was 7.5 (SD = 2.1, N = 79). Health Communication students reported slightly higher average ratings (M = 7.9, SD = 2.3, n = 35) than Health Promotion students (M = 7.6, SD = 1.9, n = 44). See

Table 1. Helpfulness Ratings for Embedded Writing Tutor (WT) and Writing Center by Course.

Course	Support Type	n	M (SD)	t	df	p	Cohen’s d
Health Communication	Embedded WT	35	7.9 (2.3)	2.10	46	0.049	0.73
Health Communication	Writing Center	13	7.0 (2.6)	-	-	-	-
Health Promotion	Embedded WT	44	7.6 (1.9)	0.94	10	0.362	0.25
Health Promotion	Writing Center	8	5.9 (1.7)	-	-	-	-

Note. WT = Writing Tutor. Mean (M) and standard deviation (SD) reflect student ratings of helpfulness on a 1–10 scale. t, df, p, and Cohen’s d are reported for independent samples t-tests comparing WT and Writing Center ratings within each course.

below for a summary of the results.

A subset of 31 students reported having previously used the university’s Writing Center. Among them, the average helpfulness rating was 6.7 (n = 31). Health Communication students rated the Writing Center more positively (M = 7.0, SD = 2.6, n = 13) than Health Promotion students (M = 5.9, SD = 1.7, n = 8).

An independent samples t-test was conducted to compare students’ ratings of the Embedded Writing Tutor (WT) and the Writing Center within each course. Although the t-test assumes normally distributed data and homogeneity of variance, these assumptions were examined through visual inspection of the distributions; given the small sample sizes, particularly for students who had previously used the Writing Center, results should be interpreted with caution. In the Health Communication course, WT ratings were significantly higher than Writing Center ratings, t(46) = 2.10, p = 0.049, Cohen’s d = 0.73, indicating a medium to large effect. In the Health Promotion course, WT ratings were also higher, but the difference was not statistically significant, t(10) = 0.94, p = 0.362, Cohen’s d = 0.25, suggesting a small effect size. These results are presented as exploratory and descriptive, intended to complement the qualitative findings regarding students’ perceptions of the WT.

Open-ended responses provided additional insight—see

Table 2. Themes and Representative Quotes from Student Evaluations of the Embedded Writing Tutor.

Theme	Description	Representative Quote
Tutor Approachability & Support	Students highlighted the WT’s relaxed, patient, and kind demeanor, and his willingness to help without judgment.	“I enjoyed the meetings with [the WT]; he truly is very helpful with your writing. He doesn’t make you feel dumb and truly wants to understand your writing to help you succeed.”
Writing & Course Support	Feedback on grammar, clarity, focus, and alignment with course content was valued.	“[The WT] was always at our disposal and was a very valuable asset. He was better than the Writing Center because the Writing Center doesn’t specialize in topics like public health.”
Accessibility & Meeting Format	Students appreciated ease of scheduling, multiple meeting formats (Zoom/in-person), and convenient access to the tutor.	“The most helpful thing was that the meetings were over Zoom. It made meeting with him quick and easy.” “I would highly recommend that the meetings could be in person because I feel communication is easier that way or more can get done quicker.”
Mixed Experiences & Limitations	Some students faced scheduling difficulties, preferred the Writing Center, or noted inconsistent feedback.	“I never had the chance to meet with [the WT]. Our schedules never worked out. I don’t think you should hire another writing tutor. Utilizing the Writing Center is more helpful.”
Required Meetings	Some students suggested fewer required sessions while still valuing the support.	“Maybe in the future, less meetings could be ‘required’ although they were very helpful.”

for a thematic table of the qualitative results. The majority of students described the WT as helpful, highlighting his relaxed demeanor, patience, kindness, and detailed feedback. One student shared, “I enjoyed the meetings with [the WT]; he truly is very helpful with your writing. He doesn’t make you feel dumb and truly wants to understand your writing to help you succeed.” Another student echoed this: “Overall, meeting [the WT] was very helpful. [He] was good at making corrections. Aside from that, he was someone who we could easily ask questions about our writing without making us feel bad about what we wrote.”

Students particularly valued the WT’s support with grammar, focus, and clarity, as well as his familiarity with course content and assignment expectations. One student noted, “[The WT] was always at our disposal and was a very valuable asset. He was better than the Writing Center because the Writing Center doesn’t specialize in topics like public health. [The WT] assessed more than just the writing itself.”

Accessibility was another commonly praised feature: “Accessibility to meet [the WT] was great—my team never had an issue with meeting [him]. Most of his feedback and explanations made sense and were helpful to our team as well.”

However, a few students reported mixed experiences. Some described difficulty scheduling meetings or expressed that they did not benefit from the tutoring. One student wrote, “I found the tutor unnecessary and the requirement of going was frustrating,” while another shared, “I never had the chance to meet with [the WT]. Our schedules never worked out. I don’t think you should hire another writing tutor. Utilizing the Writing Center is more helpful.” A few also raised concerns about inconsistent feedback, such as being told a project “looked good” by the WT, only to later receive critical comments from the professor.

Preferences regarding meeting format were also mixed. Many appreciated the convenience of Zoom: “The most helpful thing was that the meetings were over Zoom. It made meeting with him quick and easy. [I] prefer this over making an appointment to go to the Writing Center.” However, some students preferred in-person meetings, suggesting they could lead to more efficient communication: “I would highly recommend that the meetings could be in person because I feel communication is easier that way or more can get done quicker.”

Some students also commented on the required nature of meetings. One wrote, “I found the meetings with [the WT] to be helpful… Maybe in the future, less meetings could be ‘required’ although they were very helpful.” Despite some constructive feedback, the consensus was overwhelmingly positive. As one student summarized, “All of the meetings were helpful. He would help out with whatever we were struggling with and gave us good feedback.”

3.3. Professor’s Experiences

After teaching several rounds of writing-intensive public health courses, the professor was motivated to test the efficacy of an embedded writing tutor. Since many students struggle with foundational writing skills (e.g., spelling, grammar, paragraph organization), managing these issues alone had become overwhelming. Although not a professor of writing, the instructor recognized the quality of student writing necessitated additional support. In collaboration with the university’s writing center director, tutoring visits were integrated into course assignments. However, students often reported that the writing tutors lacked the content knowledge needed to provide meaningful feedback. Thus, the embedded tutor format seemed like the perfect solution. The instructor hoped that a WT familiar with course content and assignments would address some of the challenges students previously faced with general tutoring services.

Several key insights emerged from this pilot—the first being role confusion. Students often confused the WT’s role with that of the undergraduate Learning Assistant, the Teaching Assistant, and even the professor. Despite multiple clarifications in person and via email, the confusion persisted and was evident in the WT evaluations. Similarly, some students struggled to understand the hierarchy of these roles and whose feedback they should prioritize. She noted that this may be due to the high percent of first-gen students who may be unfamiliar with the hierarchy of power within higher education.

Having an embedded WT alleviated some of the burden of providing detailed writing feedback. It was reassuring for the instructor to direct students to him, knowing he was well prepared because the two conferred in advance. Debriefing tutoring sessions with the WT also provided opportunities to reinforce key skills and content. For example, if the WT noticed students struggling with citations, the instructor would address it in class and review the relevant material.

One unexpected challenge was the number of students who attended tutoring sessions without prepared drafts. The instructor was unaware of this issue until gaining access to the tutor’s notes after the course had ended. If this approach were to be implemented again, the instructor would require students to bring a draft to make the sessions more productive. While managing an additional team member alongside a Teaching Assistant and Learning Assistant required more effort, the WT significantly lightened the instructor’s workload. For example, she spent significantly less time reviewing students’ work before they submitted it, since the WT reviewed their assignments with them in their tutoring sessions.

Lastly, while Institutional Review Board restrictions prevented the collection of course or assignment grades for formal analysis, the course instructor observed notable improvements in student writing performance compared to previous semesters. These improvements were most evident in the Health Communication course, where students demonstrated stronger grammar, greater concision, and fewer overall errors. In prior semesters, students were required to summarize assigned scholarly articles but frequently selected the incorrect article. In contrast, during the semester with the embedded writing tutor, all students correctly selected and summarized the appropriate articles—an outcome the instructor noted as unprecedented in her experience.

4. Discussion

The embedded WT was generally well received by both courses, with students appreciating the accessibility, patience, direct feedback, and non-judgmental approach. Health Communication students rated the WT more helpful than the Health Promotion students, likely due to the differences in assignment structure. The Health Communication course is based primarily on brief, repetitive writing assignments that encourage students to focus on brevity, accuracy, and grammar. In this class, students have multiple opportunities to edit their drafts to earn points back and they complete the same assignment several times which allows for targeted practice. On the other hand, Health Promotion is based around a semester-long, group-based project with evolving components. Although this class is also writing-intensive, each writing assignments is unique and requires students to learn new skills. We believe the repetition in Health Communication may have led to a more positive perceived experiences as there is usually less confusion and frustration surrounding the writing assignments.

Since the major writing assignment in Health Promotion was also group-based, another difference that arose was that the tutoring sessions in Health Promotion occasionally heightened anxiety, especially when skill levels varied within groups. The Health Promotion students rated the university writing center lower than Health Communication students, likely because their discipline-specific writing needs were not well addressed by generalist tutors.

The WT’s kind demeanor and supportive approach were frequently described in student comments, suggesting the importance of affective dimensions of tutoring. First-generation and minoritized students often report anxiety around writing and fear of being judged—factors that can inhibit their engagement and development. The WT served as a bridge between students and the professor, providing a culturally responsive, low-pressure entry point into the writing process. This aligns with literature on inclusive pedagogy, which highlights the importance of care, patience, and relationship-building in supporting underrepresented students (Conefrey, 2021; Matos, 2021).

Despite the overall positive feedback, several challenges emerged. Students expressed confusion about the WT’s role relative to the TA, Learning Assistant, and professor. This role confusion sometimes undermined the effectiveness of feedback and may reflect a broader issue—many first-gen students are unfamiliar with academic hierarchies and the division of instructional labor in higher education.

Additionally, both the WT and the professor noted that students often arrived unprepared for their sessions, lacking drafts to review. This limited the usefulness of the tutoring time. In future implementations, requiring students to bring a draft—however rough—could improve session productivity and better scaffold the revision process.

Our findings align with Caron’s (2008) argument that many students do not have “writing problems” per se but rather “thinking problems.” Caron found that when students wrote about their own experiences, their writing seemed competent—but when asked to respond to a new reading, their difficulties became evident. This suggests that student writing struggles often stem from challenges in processing and articulating unfamiliar content, rather than with language mechanics alone (Caron, 2008). Bizzell (1986) similarly emphasized that learning to write in college requires learning the epistemologies and “languages” of specific disciplines.

In our pilot study, the structured repetitive assignments in Health Communication likely allowed students to internalize both content and writing expectations, leading to stronger writing performance and higher satisfaction. Conversely, the dynamic, multi-part assignment in Health Promotion may have made it harder for students to achieve fluency before being asked to write in new formats. These differences underscore the importance of aligning writing instruction not only with discipline-specific conventions, but also with students’ readiness to process complex content.

The students’ emphasis on the WT’s kindness and clarity reinforces the value of embedding support within a pedagogical model that is both discipline-specific and empathetic. As educators, we must encourage students to move beyond transactional thinking about writing (i.e., “just tell me what you want”) toward a developmental mindset that embraces struggle as part of the learning process. This shift is especially crucial for students from under-resourced educational backgrounds, for whom failure may feel high-stakes or stigmatizing (Gay, 2002; Doolan & Miller, 2012; Doolan, 2013).

The embedded WT model helped create a safer space for students to take academic risks. He served as a less intimidating intermediary between students and the “scary professor,” providing consistent, course-aligned guidance and reinforcing key skills throughout the semester. As the professor observed, collaboration with the WT also created a feedback loop: patterns identified in tutoring sessions informed in-class instruction, allowing the instructor to proactively address common writing issues.

A limitation of this study is the small sample size, particularly among students who had previously used the Writing Center, which limits statistical power and generalizability. While independent samples t-tests provide exploratory insight into differences between ratings of the Embedded Writing Tutor and the Writing Center, assumptions of normality and homogeneity of variance should be considered, and results interpreted cautiously. The tutor’s intermittent presence in class may also have limited student access and the full integration of the embedded model. Future research with larger samples and more consistently embedded tutors is needed to more rigorously evaluate the model’s impact on student perceptions and outcomes.

Lastly, from an institutional perspective, effective implementation of an embedded tutoring model requires careful alignment with course objectives, clear communication of the tutor’s role, and coordination between the instructor and tutor to ensure consistent feedback. Logistical and pedagogical barriers, such as scheduling constraints, intermittent tutor presence, and variability in students’ prior experience with academic writing, may limit the model’s effectiveness. For long-term sustainability, institutions should consider resource allocation, training protocols for tutors, and mechanisms for ongoing evaluation, ensuring that embedded tutoring can be maintained as an equity-focused support strategy. By addressing these conditions, the embedded model can be a scalable approach to strengthen writing instruction across disciplines while enhancing student engagement and success.

5. Conclusions

With growing challenges in higher education, including budget constraints and increasing diversity among students, embedded Writing Tutors (WTs) offer a cost-effective, discipline-specific approach to supporting student writing and engagement. This pilot study provides preliminary evidence that embedded WTs can enhance writing confidence, provide targeted feedback aligned with course content, and support academic agency and a sense of belonging, particularly for first-generation, low-income students and students of color. While these findings are specific to two public health courses, they suggest that similar models may benefit other courses serving structurally marginalized students. Institutions can promote inclusive teaching by integrating embedded WTs across courses, clarifying their role, and ensuring coordination with faculty to provide consistent, equity-focused support. Future research should examine long-term impacts, scalability, and which student populations benefit most, refining best practices for sustainable implementation.