Initial approach to a research writing course: a case study of STEM and non-STEM female students under pressure to succeed

Abstract

The present research examined the extent to which the initial approach to a research-writing course by STEM and non-STEM second-language learners may entail the contribution of different dispositions to course performance and ultimately be responsible for dissimilar outcomes. Individual differences in dispositions and behavior were assessed during the first four weeks of the semester. They pertained to an understudied college population of young women from a society (Saudi Arabia) that has only recently begun to address gender inequalities in education and the workforce by placing women at the center stage of its economic development. There were no differences in generic and research-specific writing skills, but performance on the first assignment was higher in non-STEM students, whereas general confidence (i.e., self-efficacy) and confidence specific to research writing were higher in STEM students. Nevertheless, STEM students were more likely to complete the course successfully than non-STEM students. It was concluded that the former treated initial poor performance as a warning call to increase engagement, thereby independently addressing their own difficulties. Instead, the latter required additional instruction and counseling as engagement by itself was insufficient to lead to academic success. These findings underscore the relevance of targeted, evidence-driven interventions that acknowledge the different academic needs of STEM and non-STEM students.

Keywords:

• Academic success
• individual differences
• Middle East

1. Introduction

General education courses devoted to research writing are challenging for most freshmen as they demand the adoption of the rigorous thought processes, methodologies, and technical language used by scientists in their endeavors (Carillo, 2017; Mante-Estacio et al., 2019). Challenges are even higher for students who have English as their second language due to the unique technical prose of such courses compared to everyday English (Hall, 2014; Shukri, 2014). Yet, research writing competency serves as a foundation for major courses, especially in STEM (Science, Technology, Engineering, and Math) majors, and thus is likely to impact college retention and persistence (Lei & Lei, 2019; Morse, 1952; Orillion, 2009).

The present study examines whether freshmen in STEM majors and non-STEM majors differ in their initial approach to a research writing course. It targets an understudied population of female college students of a society (i.e., Saudi Arabia) marked by low participation of women in science (Narasimhan, 2021; Pilotti, 2021), notwithstanding recent substantial financial investments in the education of women (Esmail, 2018). Decrees, declarations, and investments are unmistakable in their intent to move women into the workforce through a reconceptualization of traditional roles. As a result of such top-down influences, women, especially those of college age, are now expected to make a contribution to the economic engine of their society that is equal to that of men (Barry, 2021; Pilotti et al., 2021). As much as men, they are expected to master the English language in recognition that the country’s integration into the global economy requires a workforce that can operate effectively inside and outside the country (Alharbi, 2019). The pressure to succeed, first academically and then professionally, felt by the Saudi youth has led to substantial increases in young women’s enrollment in college (El-Moussa et al., 2021; Labib et al., 2021), thereby fostering a cadre of learners whose underlying motives encompass the allure of agency that a college degree can provide (e.g., financial independence, autonomy, etc.) as well as the need to misspell patriarchal stereotypes. Among young women, the harshest challenges are faced by STEM learners who are asked to counteract stereotypes that are cemented in the very fields that they have chosen to pursue (Makarova et al., 2019), such as the idea that females are bad at math or are unsuited to science-based careers.

In Saudi Arabia, as in other parts of the world, college students in STEM majors and non-STEM majors have quite different backgrounds in math and science. High school education is divided into a scientific and a literal track following one year of common curriculum. The choice of a track is consequential as it determines the students’ exposure to math and science disciplines (e.g., physics, chemistry, and biology) and tends to predetermine the choice of major in college. If culture is understood as an array of norms, values, shared meanings, and behaviors associated with a group of people (Kramsch & Zhu, 2020; Levin & Mamlok, 2021), the experience of research-writing courses for STEM and non-STEM college students can be expected to require different degrees of accommodation to a new culture due to their degree of prior exposure to and predilection for math and science.

Not surprisingly, a review of anonymized records of students’ comments regarding a research-writing course, which has motivated our study, has suggested that during the first few weeks of the semester, although both STEM and non-STEM students tend to expect the course to be intellectually demanding and challenging, they express different profiles of adaptation to the course. For non-STEM students, uncertainties regarding the content of the course and its workload are a tangible source of apprehension. Instead, for STEM students, expectations of difficulties largely reflect time-management concerns based on past experiences in science courses.

The present study asks four separate questions regarding STEM and non-STEM students. First, do these expectations reflect differences in students’ initial performance? Consider the greater scientific background (including practice with the scientific method) that STEM students have acquired in high school. One may predict that research writing courses will be particularly challenging for those who have less familiarity with the rigorous thought processes, methodologies, and technical language demanded by research reports (i.e., non-STEM majors). Alternatively, one may predict that prior familiarity (as per STEM majors) will lead to students’ over-confidence in research writing courses, reducing effort and ultimately being counterproductive.

Second, do these expectations reflect other individual differences? To this end, the study specifically asks whether particular individual differences in behavior and disposition may be used as indicators of early academic difficulties. Included factors are a measure of behavioral engagement (i.e., attendance during the first four weeks of the semester), and dispositions, such as confidence in one’s abilities (i.e., self-efficacy; Usher et al., 2019), confidence specific to research writing, and generic writing skills (i.e., prior grades in a general writing course). Chronotype (preference for morning versus evening activities; Ferguson et al., 2018) is included to measure whether there is any advantage resulting from a preference for activities that match the time of class delivery. Findings in the extant literature link each factor to academic performance, thereby justifying its selection, although the available evidence rarely includes the understudied population chosen for the present study. For instance, research illustrates that it is advantageous for a student to attend class (Nieuwoudt, 2020), to have confidence in one’s abilities to complete a variety of tasks (Choi, 2005; Freire et al., 2020; Usher et al., 2019; i.e., self-efficacy) as well as confidence specific to research writing (McBride et al., 2021), to possess generic writing skills (Lei & Lei, 2019), and to be a morning types, especially if classes are offered in the morning (Ferguson et al., 2018). Based on such evidence, one may predict the probability of learning failures to be higher when engagement is minimal, overall confidence (i.e., self-efficacy) and research-specific confidence are low, and generic writing skills are poor. The mismatch between the time at which a course is offered and a student’s chronotype may further aggravate the likelihood of failure.

The preceding research questions logically lead to inquiries about sources and consequences. Namely, do individual differences (if present) contribute to the early course performance of STEM and non-STEM students differently? If differences are uncovered in the early performance of STEM and non-STEM majors, are they to be treated mostly as harbingers of academic failures or merely as a wake-up call for the recipients to spring into action? The answers to these questions can inform interventions for at-risk learners when their academic journey is beginning, and thus improve college retention and persistence.

2. Method

2.1. Participants

The participants were 256 female students enrolled in a course of the general education curriculum in which learning how to write research reports is a key aspect of performance. The course is offered to freshmen by a Saudi university that conforms to a US curriculum and pedagogy (i.e., an inquiry-based learning model). A generic course in written communication serves as a prerequisite. At the selected university, all courses of the general education curriculum have been developed and approved by the Texas International Consortium, an organization that exercises quality control over the curriculum and instruction of higher educational institutions outside the US. The university is organized into a female campus and a male campus, each offering the same courses and degree programs. Although gender segregation is being slowly dismantled, courses in the general education curriculum, including the research writing course chosen for our study, still enroll students of one gender on each campus.

Students’ age ranged from 18 to 29 (M = 19.22; SD = 1.78). Their majors encompassed all those offered by the selected university, including business, engineering, computer science, interior design, and law. Among the participants, 45.31% were non-STEM majors (e.g., business, interior design, information technology, and law), whereas 54.69% were STEM majors (e.g., mechanical, civil, software, and computer engineering, and computer science). Instruction, which relied on textbooks imported from the US, was delivered entirely in English, albeit students were Arabic-English bilingual speakers. For each participant, English language competency was determined to be equivalent to that of a “competent user” (as per IELTS) prior to enrollment in courses of the general education curriculum. Sections of the research-writing course included between 19–25 students and were offered in the morning over a period of four semesters. They were taught face-to-face before and after the pandemic by one educator to ensure consistency of instruction.

3. Materials and procedure

The following behavioral measures were collected from the participants: attendance records in the first four weeks of the semester (%), which served as an index of students’ behavioral engagement, grades in a prerequisite writing course (%), which were treated as indices of generic writing skills, and self-reported research writing skills (%), which were intended to offer an appraisal of students’ confidence in their writing abilities specific to research reports. In addition, individual differences in dispositions included the morningness scale developed by Smith et al. (1989) and the self-efficacy scale of Chen et al. (2001). The morningness scale, which contains 13 items, assessed individual differences in circadian rhythms. Scores may range from 13 to 55, including evening types (13–22), intermediate types (23–43), and morning types (44–55). Cronbach’s alpha, a measure of internal consistency, was .78. The self-efficacy scale of Chen et al. (2001) measured students’ confidence in their abilities to complete a variety of tasks. The scale consists of 8 generic statements of self-efficacy (e.g., “When facing difficult tasks, I am certain that I will accomplish them”), which students were to rate on a continuum from strongly disagree (1) to strongly agree (5). Cronbach’s alpha was .93.

At the end of the fourth week of the course, students were expected to submit the first assignment. Namely, the introduction of a research report of a correlational study conducted in class. The assignment was mostly evaluated for its content and clarity of exposition to include the research question/purpose of the study, definitions of key terms, the significance of the research, and a brief and generic overview of how the research question would be answered. Students were told peer-reviewed articles had to be used to define key terms and illustrate the significance of the research. Instructions and demonstrations about searching for suitable articles were covered in class meetings.

At the end of the semester, final grades were obtained from the instructor. Final grades comprised the grades obtained on (a) a research report organized into 5 parts (i.e., assignments) and written in APA style, (b) in-class summaries and critical analyses of peer-reviewed research articles of empirical studies mostly in the social and behavioral sciences (20%), a midterm examination (15%), and a final examination (15%). The writing of the research report included the introduction (10%), literature review (10%), method and result sections (10%), discussion and abstract (10%), and revision of the completed report to ensure coherence (10%). The latter assignment was intended to reinforce the habit of revising one own’s work under the assumption that good writing heavily relies on careful reviewing and rewriting. Midterm and final examinations covered basic quantitative research methodologies in the social and behavioral sciences. Questions required students to identify particular concepts in summaries of studies (e.g., research methodology, hypotheses, independent and dependent variables, etc.). The final examination also included a few questions that required students to detect APA-related errors. Each examination was preceded by a review of the materials covered and practice sessions.

Qualitative information collected during office hours by the instructor in an anonymized format, simply listing the major of a student, was used to clarify differences in disposition and performance that might emerge from the collected data. Comments gathered during debriefings, also anonymized, were added to the comments made by students during office hours. A broader set of comments was gathered from the instructor who had collected anonymized comments made by students at the end of each semester from 2015 to 2021. The instructor had labeled the source of each comment as STEM or non-STEM student in response to an administrative directive to determine whether the course satisfied the needs of both groups of students. The most frequent comments were selected to illustrate students’ reactions to the content of the curriculum and instruction of the course (e.g., openly verbalized reflections on assigned readings). The guidelines of thematic analysis were used (Braun & Clarke, 2012, 2022). Categorization of comments was approached as a descriptive enterprise to remain as close as possible to both the denotation and connotation of the participants’ reactions (inductive coding). Two independent raters yielded inter-rater reliability of 89%. When disagreement was detected a third rater was utilized.

The percentage of students who had all selected measures was 89.20% (256 out of 287). All information was coded to protect participants’ identities so that no names or other identifying information would appear on data files after all information was collected. The guidelines for educational research of the Office for Human Research Protections of the US Department of Health and Human Services as well as the American Psychological Association’s ethical standards were followed. The study was conducted under the purview of the Deanship of Research of the selected institution.

4. Results

Table 1 displays descriptive statistics of the variables of the present study. Inferential statistics were intended to answer research questions. Results were considered significant at the .05 level.

Table 1. Mean (M) and Standard Error of the Mean (SEM) of the selected variables

Variables	Non-STEM M	SEM	STEM M	SEM
Attendance (0–100)	78.31	1.96	78.30	1.79
Generic writing skills (0–100)	79.76	0.81	81.13	0.73
Confidence in research writing*(0–100)	50.69	1.60	61.09	1.46
Self-efficacy (1–5)*	3.59	0.09	3.93	0.08
Chronotype (13–55)	31.52	0.58	32.65	0.53
First assignment* (0–100)	77.88	2.99	69.09	2.72

Significant differences between STEM and non-STEM students are marked by an asterisk.

5. Were there differences between non-STEM and STEM students?

One-way ANOVAs were conducted with major as the independent variable (non-STEM and STEM) and attendance, generic writing skills, general confidence (i.e., self-efficacy), confidence specific to research writing skills, chronotype, and performance on the first assignment as the dependent variables. Initial performance in the course (as measured by the first assignment) differentiated between STEM and non-STEM majors [F(1, 254) = 4.74, MSE = 1035.20, p = .030, ηp² = .018]. Although there were no differences in generic writing skills [F = 1.58, ns], STEM majors reported greater confidence in their research writing skills [F(1, 254) = 23.05, MSE = 297.67, p < .001, ηp² = .083], as well as greater self-efficacy [F(1, 254) = 8.43, MSE = .876, p = .004, ηp² = .032] than non-STEM majors. The partial eta square values, which illustrated the proportion of variance uniquely explained by students’ major, were rather modest though. STEM and non-STEM majors did not differ in any of the other measures, including attendance and chronotype [Fs ≤ 2.08, ns].

The greater general self-efficacy and confidence in research-specific writing skills reported by STEM majors matched qualitative information collected during debriefing as well as anonymized comments collected by the instructor during office hours. Qualitative records suggested that the difference in performance between STEM and non-STEM majors might originate from these two groups of students approaching the course differently. An examination by two independent raters of students’ anonymized comments confirmed that STEM majors started the course with greater overall confidence in their abilities to overcome challenges and greater confidence specifically in their research writing skills than non-STEM students. For instance, during the initial weeks of the semester, although both groups anticipated that the course would be challenging and demanding, non-STEM students tended to perceive it as full of uncertainties and novelties towards which they felt apprehension (e.g., “this type of writing is different”; “there are many new words that I have not seen before”; “words have a different meaning here”). Instead, the apprehension of STEM students was mostly linked to time management concerns and was supported by the conviction that the course would ultimately replicate activities carried out in high school (e.g., “I hope I have enough time to complete all assignments”; “research reports are not easy to write and require a lot of time and effort, but I have written research reports before and I can do it again”; “I just have to figure out a few new things and find the time”; “peer-reviewed articles may be difficult to read and take time, but I have some experience”). Taken together, quantitative and qualitative findings suggested that the greater general and research-related self-confidence of STEM students might have reduced the effort they invested in the initial assignment of the course, thereby undercutting their performance.

6. What were the sources of initial performance?

To understand non-STEM and STEM students’ initial performance, a linear regression analysis was utilized to assess the contribution of attendance, generic writing skills, self-efficacy, research-related confidence, and chronotype (i.e., predictors) to performance on the first assignment (i.e., outcome variable). To highlight the pattern of the selected variables’ contribution in each group, regression analyses were conducted separately for non-STEM and STEM students. Regression analyses did not produce evidence of multi-collinearity (tolerance values greater than .79; mean VIF for STEM = 1.13; mean VIF for non-STEM = 1.13; Field, 2013). In Table 2, the b-values indicate the degree to which each predictor contributes to the outcome variable when the other predictors are held constant. In the last two columns, structure correlations (i.e., r_s = r _Y–X/R; the relationship between each predictor and the outcome variable when variables are taken two at a time) and squared structure correlations (i.e., the amount of variance in the outcome variable that is explained by each predictor) are displayed (Thompson & Borrello, 1985).

Table 2. Regression analyses

Predictor Variables	B	SE	Beta	t	Sign.	rs	rs^2
Non-STEM majors
Constant	−10.704	23.237
Attendance	.595	.103	.487	5.769	.000	.90	.81
Generic writing skills	.605	.271	.185	2.233	.028	.46	.21
Research writing skills	.022	.111	.017	.203	ns	.32	.10
Self-efficacy	2.986	2.293	.110	1.302	ns	.19	.04
Chronotype	−.574	.377	−.130	−1.521	ns	.00	.00
STEM majors
Constant	−23.590	28.154
Attendance	1.205	.114	.687	10.546	.000	.99	.98
Generic writing skills	−.189	.267	−.048	−.710	ns	.09	.01
Research writing skills	−.017	.170	−.007	−.101	ns	.02	.00
Self-efficacy	.794	2.520	.020	.315	ns	.02	.00
Chronotype	.356	.390	.064	.912	ns	.28	.08

Note: Non-STEM: R = .547; STEM: R = .695. b = unstandardized regression coefficient; SE = standard error of the unstandardized regression coefficients; Beta = standardized regression coefficients; r_s = structure coefficients; r_s² = squared structure coefficients.

If indeed STEM majors suffered from overconfidence, writing skills would be more likely to be engaged by non-STEM students, thereby contributing more significantly to their performance in Assignment 1. Non-STEM majors’ attendance (serving as a measure of behavioral engagement) might also be more likely to contribute to performance in Assignment 1.

We found that for both STEM and non-STEM majors, attendance positively contributed to performance, thereby suggesting that awareness of the demands of the course matters. However, the greater the attendance rates and the writing skills of non-STEM majors, the higher were their grades on the first assignment. Thus, if attendance is understood as merely indicating students’ knowledge of course demands, generic writing skills could be said to have been engaged more intensely by non-STEM students than by STEM students, under largely equivalent conditions of knowledge of course demands.

7. Were the consequences of overconfidence long-lasting?

Initial poor performance could be treated by STEM students as a wake-up call to increase engagement, thereby leading to performance equal to or greater than that of non-STEM students at the end of the semester. Alternatively, it could be treated as an abnormal event to ignore. No behavioral change would most likely lead to STEM students’ lower performance at the end of the semester. To test these contrasting predictions, we examined the final course grades of STEM and non-STEM students across several semesters of the selected course (2015–2021) as well as comments spontaneously made by students regarding their performance towards the end of the course. Students’ course grades (including their scores obtained on the different parts of a research report, analyses of peer-reviewed empirical studies as well as a midterm test and a final test devoted to research methodology) were arranged into one of two categories (see, Table 3): Poor performance (from F to C; below 76%), and satisfactory performance (from C+ to A+; 76–100%). This broader viewpoint, which combined qualitative and quantitative information, supported the notion that most STEM students treated the first-assignment grade as a wake-up call for their effort to be increased in the course. Both groups tended to describe the course as challenging, but STEM students saw it as manageable, whereas non-STEM students persisted in seeing it as overwhelming and often questioned the utility of the course in their future academic endeavors. These views were reflected in the overall performance of the two groups of students. Namely, non-STEM students were twice as likely to exhibit poor performance in the course [Χ² (1, n = 3491) = 49.39, p < .001].

Table 3. Final course performance (2015–2021)

Course performance	Non-STEM	STEM
Poor	18.02%	9.44%
Satisfactory Performance	81.98%	90.56%

8. Discussion

The role that attendance plays in students’ academic performance has an intuitive appeal. Namely, if pursuing an undergraduate degree stands for opportunities to learn interesting and valuable knowledge and skills, attendance is less a responsibility and more of a necessity to which students reliably comply. Not surprisingly, a popular piece of advice often disseminated by educators to their students is that those who attend class regularly make good grades. However, attendance in undergraduate courses tends to be poor (Friedman et al., 2001; Marburger, 2001; Woodfield et al., 2006), including courses that comprise the general education curriculum (Moore, 2003; Summers et al., 2015). The latter courses aim to ensure the acquisition of skills and knowledge upon which students’ performance in courses of their selected majors relies. The findings of the present study are consistent with those of studies that have uncovered an association between students’ performance (as measured by grades) and attendance (Dey, 2018; Kassarnig et al., 2017; Lukkarinen et al., 2016; Marburger, 2001; Moore, 2003; Thomas & Higbee, 2000). Our findings are also consistent with the interpretation of this association offered by Credé et al. (2010), according to which class attendance offers distributed practice and fosters overlearning, through multiple reiterations of materials and skills that are then assessed in the students’ courses. Both distributed practice and overlearning are known to be beneficial to academic success (Cepeda et al., 2006; Donovan & Radosevich, 1999; Peladeau et al., 2003).

The findings of the present investigation suggest that initial indicators of performance can be used independently by students as warning calls regarding their approach to the demands of a course. They also indicate that the interpretations that are given to the perceived challenges of a course matter. If challenges are seen as manageable, they can motivate students to independently enhance their engagement (as per STEM majors), thereby addressing the deficiencies that may be experienced at the start of a course. Instead, if they are framed as overwhelming, and additional instructional support is not seen as forthcoming, challenges are more likely to translate into less than desirable performance.

These findings are consistent with those that highlight the relevance of attitudes towards courses and related coping responses (i.e., cognitive and behavioral responses to demands that are deemed to be threats) to students’ academic success (Freire et al., 2020). Our findings contribute to the existing literature by illustrating the responses to academic challenges of a sample of female students from an understudied population. These responses are critical to the success of women in a society that has only recently begun to address gender inequalities (Varshney, 2019).

9. Implications

The present study highlights how information collected in the classroom can inform the selection of instructional interventions for different groups of students. The study rests on the assumption that challenging first assignments or tests offer students a window on how they are approaching courses. Early performance information can be useful to both students and instructors to guide effective remedial actions intended to avoid disengagement, unnecessary withdrawals, failures, and course repetitions. Yet, a poor initial score may be sufficient to induce behavioral change in STEM students, whereas supplemental instructional support and counseling may be needed to obtain academic success at the end of the semester in non-STEM students (as illustrated by overall course performance in the current study). The findings of our study are intended to be used to develop an early-warning system to direct proactive counseling interventions devoted to (a) informing students about the impact of specific dispositions and behaviors on academic success, as well as (b) training them to acquire desirable behaviors and modify undesirable dispositions. The guiding assumption of interventions is that behavioral change can lead to dispositional chance (Olson & Zanna, 1993).

The implications of our study are defined by the broader societal context in which its findings exist. To this end, it is important to bear in mind Dator’s conceptual framework for understanding social change (Dator, 2002), which defines societal transformation as entailing the replacement of norms, beliefs, and acts with new norms, beliefs, and acts, to address current and future challenges. The findings of the present study refer to a society in transition from a strict code of conduct that lionized stereotypical gender inequalities to one that is slowly fostering gender equity (Ménoret, 2005; Varshney, 2019). Le Renard (2014, p. 3) notes that “institutional actions, official declarations, lectures, decrees, regulations, reports, and measures” have redefined the “possibilities, opportunities, and spaces accessible to Saudi women”. Thus, how female college students in STEM and non-STEM programs approach the demands of the courses in which they are enrolled, and the actions taken to foster academic success, may be consequential for gender-equity progress not only inside educational institutions but also in the workplace and at home where stereotypical gender-role expectations may more forcefully resist replacement. For instance, although women can now pursue careers and occupations before prohibited, and enjoy greater freedom of movement and self-determination, working outside the confines of the home is still a challenge, thereby keeping women’s employment markedly low (Bursztyn et al., 2020). Such challenges are likely to extend to other female students in societies of the Gulf region that have pursued institutional changes similar to those of Saudi Arabia guided by their desire to join the global marketplace.

10. Limitations and suggestions for future research

The current study has limitations to be addressed in future research. First, it focuses on female freshmen, thereby leaving unanswered the question of how male freshmen approach the same research-writing course. In the society targeted for the present research, only recently female and male students have experienced equal treatment. The pressure to succeed academically and professionally may be a novelty for young women, but for young men, it has been a ubiquitous expectation since birth. Due to the patriarchal fabric of the society of which they are citizens, however, men have also benefited from a sense of entitlement that is now been challenged by the top-down restructuring of their society intended to promote meritocracy and gender equity in education and the workplace (Alhazmi & Kamarudin, 2021; Barry, 2021). Whether STEM and non-STEM male students show a pattern of responses to the demands of a research-writing course similar to that of female students is a matter to be investigated. Second, by and large, the students of the present research, irrespective of their major, possessed an intermediate (neutral) chronotype, which underscored their lack of an unambiguous preference for morning or evening activities and thus the ability to adapt to both day and evening schedules. In the extant literature, the intermediate type is generally the most prevalent. For instance, BaHammam et al. (2011) and Pilotti et al. (2022) reported that Saudi college students did not exhibit a stark preference for morningness or eveningness. If the prevalence of the intermediate type is conceptualized as limited variability in a data set, it is not surprising that the results of our regression analyses indicated that students’ chronotypes did not significantly contribute to their initial performance in a course with classes scheduled in the morning. Yet, one may ask whether the inclusion of classes scheduled across the entire day may make even the intermediate chronotype more relevant. Third, our study focuses on learners who possess English as their second language. Although all met the “competent user” proficiency criterion to enroll in general education courses via standardized tests (e.g., IELTS) for reading, writing, speaking, and listening, proficiency differences among students may make the research-writing course more or less challenging, irrespective of the STEM/non-STEM classification. Such differences, which were not made available to the researchers, may serve as factors in a hierarchical regression analysis to determine the impact of the selected dispositions on performance uncontaminated by variability in linguistic competencies. Fourth, introductory courses, such as those devoted to competencies in math and natural sciences, may need to be examined to assess whether the differences between STEM and non-STEM students uncovered in a research-writing course generalize to other domains in the sciences (Pilotti et al., 2022). Lastly, additional dispositions may be included to examine the extent to which they may offer predictions of academic attainment.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.