Advising Student-Athletes for Success: Predicting the Academic Success and Persistence of Collegiate Student-Athletes

The brands of many institutions are closely connected to the success of the athletic teams associated with them (Lawlor, 1998; Roy, Graeff, & Harmon, 2008) and consequently to the recruitment, enrollment, retention, and graduation of student-athletes (Joseph, Mullen, & Spake, 2012; Sperber, 2000; Zimbalist, 1999). Successful athletic programs support student-athletes with effective academic support because the academic success and graduation of collegiate student-athletes constitutes a primary component of the success at every level of National Collegiate Athletic Association (NCAA) sport participation.

In October 2016, the NCAA announced that, starting in 2019, it would distribute millions of dollars to institutions with student-athletes who meet or exceed benchmarks established for graduation rates and academic progress (Hosick, 2016). For many years, student-athletes were required to meet criteria that defined eligibility to compete in college athletics. In 1986, the NCAA enacted Proposition 48, which required prospective student-athletes to have attained a minimum 2.00 high school GPA in a core curriculum of at least 11 defined academic courses and to have obtained a minimum ACT composite score of 15 or a minimum SAT total score of 700 to practice, play, and receive athletics-related financial aid as college freshmen. Then in 1992, the NCAA updated the guidelines, in Proposition 16, for assessing the academic progress of admitted student-athletes. These new directives meant that student-athletes took 13 core courses and met minimum thresholds, which are based on a combination of high school GPA and SAT or ACT. Once enrolled, student-athletes were required to complete 25% of the credit requirements for their declared degrees by the end of their second years, 50% by the end of their third years, and 75% by the end of their fourth years. These percentages were increased to 40, 60, and 80%, respectively, in 2003 (Crowley, 2006; Petr & McArdle, 2012).

In 2009, the NCAA published the Graduation Risk Overview (GRO), a set of risk factors, any of which might signal that a student-athlete may not persist to graduation. The five emperical factors were based on academic background and achievement, the role of academics (in the student-athlete's life), transfer status, personal history, and sport-related issues (NCAA, 2009). For example, the GRO suggested that a student-athlete who transfers to an institution faces a greater risk of not graduating than a student-athlete who enrolled directly from high school.

Student-athletes at all levels of competition must meet academic requirements while also balancing their athletic and academic responsibilities (Hendricks & Johnson, 2016). In addition to attending classes, completing homework assignments, and attending scheduled tutoring and study hall sessions, student-athletes participate in a maximum of 20 hours of athletics-related responsibilities each week (Holsendolph, 2006). Although designed to enhance their athletic and academic success, these scheduld activities consume much of a student-athlete's daily life, leaving little time for personal or extracurricular activities. In a national survey of student-athletes enrolled at 18 Division IA institutions, Potuto and O'Hanlon (2006) found that student-athletes are more likely to identify as athletes than as students. Hence, these students may default to spending more time focusing on their athletic performance than on their academic success. However, student-athletes must maintain the academic requirements mandated by the NCAA to continue playing their sports.

Considerable evidence suggests that student-athletes who participate in academic support programs benefit academically. Using focus group data obtained from African American student-athletes enrolled at 11 U.S. institutions, Person and LeNoir (1997) found that African American male student-athletes were “more inclined to persist when advising, research experiences, and career seminars are viewed as effective,” regardless of the student's entering qualifications (p. 86). Butler (1995) discovered that institutions with major football and strong academic support programs retained student-athletes.

The question remains: How can providers of academic support programs identify those student-athletes most in need of the intervention(s) crucial to academic success and persistence? In addition to using traditional cognitive markers, advisors seeking to target students for support can assess noncognitive student factors. Several researchers (e.g., Himelstein, 1992; Kalna, 1986; Pritchard & Wilson, 2003) have identified the noncognitive factors, such as student motivation, study skill proficiency, time management skills, and self-confidence, that prove crucial to retention, academic progress, and ultimately, graduation.

Cognitive Predictive Assessment

Noncognitive Predictive Assessment

Tinto (1993) utilized self-report instruments to study retention. He stated that noncognitive assessments need to be student centered and can be used to “collect information on the attributes, intentions, and activities of each student who enters the institution,” including both precollege characteristics and experiences (academic and social) while enrolled (p. 214). Following Tinto's advice, stakeholders should begin collecting information prior to a student's first year in college to obtain the most accurate data of precollege characteristics and continue harvesting data at various intervals throughout the student's college career.

Noncognitive variables are frequently used when assessing special populations of students, such as student-athletes (Comeaux & Harrison, 2011; Sedlacek, 2004). According to Cunningham (1993), use of noncognitive variables predicts academic success and retention better than use of cognitive variables; therefore, noncognitive variables can be used to enhance the predictive ability of cognitive variables.

After more than 30 years of research related to noncognitive-based assessment of student success variables, Sedlacek (2004) concluded that the currently available instruments for gathering student data used to predict academic success, typically high school GPA and SAT or ACT performance, were not providing adequate results for diverse racial and cultural groups. The results of the assessments did not typically correlate to student outcomes (grades or retention) and were not easily utilized by college administrators. As a result, Sedlacek developed the Noncognitive Questionnaire (NCQ) on the basis of Sternberg's (1986) experiential and contextual domains.

Other researchers found that self-concept and self-appraisal predict academic success for collegiate students. Lounsbury, Huffstetler, Leong, and Gibson (2005) studied a first-year student's sense of identity and the extent to which this perception influences academic success. The authors found that sense of identity and emotional stability were positively correlated with collegiate GPA, with identity showing a stronger relationship with academic performance for African American students than for other groups studied by ethnicity. Tracey and Sedlacek (1985) also looked at predictors of academic success using the NCQ. They reported positive self-concept and realistic self-appraisal as predictors of a student's academic performance throughout his or her collegiate career. An example of a recent study that seems to reveal problematic self-appraisals, the research of Beron and Piquero (2016), who examined the GPAs of more than 19,000 student-athletes across all NCAA divisions, revealed that student-athlete participants with the lowest GPAs viewed themselves as primarily athletes and considered their roles as students less salient.

Researchers also have reported motivation as a predictor of academic success. Dunham (1973) found that achievement motivation increased the ability to predict college GPA when combined with high school GPA and gender. Dennis, Phinney, and Chuateco (2005) conducted a longitudinal study of students with minority status to examine the relationship between motivation and college GPA. They found career or personal motivation as a strong predictor of GPA and adjustment in college.

Several scholars have analyzed the persistence and time management of students. Schmelzer, Schmelzer, Figler, and Brozo (1987) surveyed 117 students about the factors that affected success and failure in college. They reported that successful students identified persistence and active study as reasons for student success and cited poor time management as a reason for failure. In a study of first-semester students, McCausland and Stewart (1974) identified delayed avoidance (procrastinating on study tasks) and work methods as factors that contributed to college success. Multiple researchers have ascertained that more conscientious students earn higher college GPAs (Bauer & Liang, 2003; Lounsbury et al., 2005). In addition, Farruggia, Han, Watson, Moss, and Bottoms (2016), after talking to participants enrolled as first-year students at a large, ethnically diverse, urban university, established a correlation of the noncognitive variables of academic mindsets, academic perseverance, learning strategies, social skills, and academic behaviors with academic success. They also found that academic performance was most strongly related to an academic mindset followed by the will to persevere (grit). They found few differences when analyzing the results by the ethnicity of the participants.

The Transition to College Inventory (TCI)

After extensive research, Pickering, Calliotte, and McAuliffe (1992) developed the TCI, which was influenced by the works of Astin, Tinto, and Sedlacek (as cited in Pickering, Calliotte, Macera, & Zerwas, 2005; Pickering et al., 1992). The noncognitive data from entering students obtained by the TCI is used to supplement and improve the predictive ability of cognitive and demographic variables. Through the TCI, students self-report their precollege characteristics and make predictions about their performance and involvement in college. The nine-factor model includes college involvement, influences on college choice, student role commitment, athletic orientation, personal and academic concerns, self-confidence, institutional commitment, socializing orientation, and independent activity focus. The TCI has been used to assess students at 4-year institutions and at community colleges (Freeze, 2000). It also has been modified to study both transfer students and student-athletes (Cunningham, 1993; Duggan, 2002).

Richardson and Sullivan (1994) examined the noncognitive factors that influenced the success of 62 academically underprepared college freshmen who were among 199 college freshman at a small, private, liberal arts college. Using the College Student Inventory, they found that noncognitive variables predicted freshman GPA and influenced persistence more, in a statistically significant way, than cognitive variables did. The TCI was normalized as a predictor of student success and utilized for both entering first-year students and those transferring to 4-year institutions (Banta, Lund, Black, & Oblander, 1996). The TCI can be used to predict which students are most at risk for attrition and to guide advisors in the use of appropriate interventions for these students early in the students' collegiate careers (Pickering & Calliotte, 1996).

Factors Affecting the Academic Success of Student-Athletes

Sedlacek and Adams-Gaston (1992) found that the NCQ was correlated with first-semester grades of first-semester student-athletes; the SAT was not so correlated. Three noncognitive variables combined to predict first-semester grades: a strong support person, community involvement, and positive self-concept. These factors relate to confidence and support, which seem to be important for a student-athlete's academic success. However, Cunningham (1993) also found that too much self-confidence (top 10%) could exert a negative influence on the academic success of a student-athlete.

A student-athlete's perceived academic ability, as well as long-range goals, remains a valid predictor of academic success (Garrett, 2000). Student-athletes with “inadequate academic backgrounds, poor study habits, tight schedules, peer distractions, and waning motivation” often become frustrated, which contributes to low grades in college (Adler & Adler, 1985, p. 247). In a study using the Student Adaptation to College Questionnaire, Melendez (2006) found that student-athletes reported higher academic adjustment scores than their nonathletic peers did. Melendez concluded that this finding may have resulted from enhanced educational support programs, psychological support systems, mentor programs, and the increased admissions requirements mandated by the NCAA.

As an important noncognitive predictor of student persistence or nonpersistence, the motivation of student-athletes for both academics and athletics should be assessed. Gaston (2002) discovered that male athletes were more motivated toward athletics than their female peers were. In an interesting finding, low-profile male athletes earned the highest ACT scores but showed the lowest motivation scores toward academics. In general, Gaston found no significant difference between high-profile and low-profile athletes and that female athletes showed higher academic motivation then their male peers.

Johnson (2013) found the NCAA GRO model effective in predicting academic risk as measured by semester GPA. As a critical determinant of athletic eligibility, GPA as a marker of the GRO can help advisors seeking to determine athletes at risk for nonpersistence to graduation.

Factors Affecting Retention of Student-Athletes

Rivera (2004) studied the key factors student-athletes believe are important in their decisions to remain in college by analyzing the results from the Understanding College Student-Athlete Retention questionnaire data from 330 student-athletes at the Division I level. Rivera reported the following factors as most important: timing of courses, a variety of course offerings, institutional fit, academic performance, and ease of declaring a major. The least important factors emerged as involvement in special interest groups and extracurricular activities, academic support from teammates, informal student–faculty member interactions, and individual athletic achievement. Regarding intent to persist, the student-athletes who had noted a prior intent to leave the institution rated their athletic experience as most important, whereas student-athletes without a prior intent to leave rated the quality of their academic experience as most important.

Using data derived from the 1990 survey of Beginning Postsecondary Students, Leppel (2005) determined that female athletes were more likely than male athletes to persist at the institution where they first enrolled. Both male and female student-athletes, irrespective of the division or institutional athletic membership, were more likely to be retained from the first year to the second year. However, males were more likely to change institutions than females because of their athletic participation.

Gaston-Gayles (2003) interviewed seven directors of academic support services for student-athletes with relatively high graduation rates in an attempt to identify those institutional factors that might account for the higher-than-average success rate of student-athletes. Themes that emerged included reporting lines of athletic support services to academic affairs, a relatively small institutional size that provides ease of communication between students and faculty members, nonadmittance of students deemed unlikely to be successful, institutional support and culture, athletic department support, and intentional advising.

The research reviewed for this article revealed that traditional variables typically utilized in admissions decisions, such as high school GPA and SAT or ACT scores, along with other nontraditional variables (race, gender, SES, and sport revenue status), can be used collectively to predict the academic success of college students. We examined the feasibility of enhancing this predictive ability by utilizing data from a noncognitive assessment tool, the TCI, to establish a model for the most accurate predictors of academic success for the student-athlete population.

Methods

We examined the factors that might effectively predict the academic success of student-athletes at the end of their first year of enrollment and their retention into the second year. We analyzed cognitive factors including high school GPA, standardized test scores (SAT or ACT), and college GPA. Demographic variables included gender, race, SES, and status in a revenue sport. We analyzed data from the TCI, used to assess precollege characteristics, skills, and attitudes, to determine the predictive ability of the instrument for both academic success and retention. For this study, we considered a cumulative GPA of 2.0 or greater representative of academic success and a GPA of less than 2.0 as a sign of academic difficulty. Retention was defined as a student-athlete's reenrollment after the first year.

We sought to answer the following questions:

• RQ1.

  Which noncognitive variables, as measured by the nine factors of the TCI, predict academic success for student-athletes at a large, public, moderately selective, research-extensive, Division I university?

• RQ2.

  What combination of demographic, cognitive, and noncognitive variables (nine factors of the TCI) predicts academic success for student-athletes at a large, public, moderately selective, research-extensive, Division I university?

• RQ3.

  Which noncognitive variables, as measured by the nine factors of the TCI, predict retention for student-athletes at a large, public, moderately selective, research-extensive, Division I university?

• RQ4.

  What combination of demographic, cognitive, and noncognitive variables (nine factors of the TCI) predicts retention for student-athletes at a large, public, moderately selective, research-extensive Division I university?

Findings and Summary

We examined the use of noncognitive, cognitive, and demographic variables as predictors of college success and retention of Division I first-year student-athletes. We used stepwise logistic regression in the analyses.

According to the analysis of the academic success of student-athletes, both null hypotheses were rejected, and both the TCI and BioQ models were found to predict the outcome at a level of statistical significance. When the nine factors of the TCI were entered, we found three factors were significant predictors of academic success: self-confidence, institutional commitment, and independent activity focus. When demographic and cognitive predictors were added to the model, we found high school GPA was a significant predictor.

In the analysis of the models for retention of student-athletes from their first years to their second years, only one variable was deemed a significant predictor. The nine factors of the TCI did not provide any predictive ability of first- to second-year retention. When the demographic and cognitive variables were added to the models, we discovered only cumulative GPA after the students' first years as predictive of retention at a level of significance.

Discussion