Predicting Success at School and Work
As each of these constructs appears to have different but complementary functions, it would follow that each provides important but somewhat different information for predicting success at school and at work. Because a more complete review of this issue is included in Mattern et al. (2014) and earlier in this report, we provide only a brief summary here. Where these constructs are correlated, they can be expected to predict some of the same outcomes (e.g., investigative interests and science skills may both predict success in a science class) but are less likely to provide incremental validity. Where they are less correlated (e.g., skills related to conscientiousness and core academic skills), they may tend to predict different outcomes but also have the potential for providing incremental validity for the same outcome. This can help us understand results found in the literature to date but also to anticipate how these constructs can be used together to increase readiness in the future.
Success at School
Meta-analyses have demonstrated that cognitive ability and academic achievement, interests, and personality all predict grades in school (e.g., Nye, Su, Rounds, & Drasgow, 2012; Poropat, 2009; Richardson, Abraham, & Bond, 2012; Robbins et al., 2004). For example, Poropat (2009) found that cognitive ability tests strongly predicted grades at each level of schooling. He also found that personality predicted grades at the earlier levels of school (primary, secondary); with conscientiousness predicting grades just as well as cognitive ability in college. Conscientiousness also predicts college GPA incrementally over ACT or SAT test scores and HSGPA (Richardson et al., 2012). Further, in their meta-analysis, Nye et al. (2012) found that interests predicted college grades and were even stronger predictors of grades when they were congruent with students’ majors. Finally, one longitudinal study of over 400,000 high school students found that, of the variance accounted for in college GPA, 50% was explained by ability, 23% by personality, and 27% by interests (Su, 2012, as cited in Rounds & Su, 2014).
In addition to predicting grades, cognitive ability and academic achievement have been shown to predict college persistence and graduation. For example, ACT and SAT test scores are positively correlated with persistence and graduation at both two- and four-year colleges (Mattern & Patterson, 2014; Radunzel & Noble, 2012). As with the prediction of grades, interests and personality-related variables also predict persistence, perhaps more so. The Nye et al. (2012) meta-analysis found that interests predict persistence and do so more strongly when they are congruent with a student’s major. Su (2012) found that ability, personality, and interests explained 66%, 12%, and 22%, respectively, of the variance accounted for in persistence. Furthermore, several variables related to personality and interests were related to persistence in the Robbins et al. (2004) meta-analysis, with academic goals and academic self-efficacy predicting incrementally over cognitive predictors.
ACT research has shown that personality, behaviors, and interests can provide incremental validity in the prediction of school performance. Casillas et al. (2012) found that personality and academic behaviors as measured in middle school students by ACT Engage predicted HSGPA above and beyond academic achievement as measured by ACT Explore (see Figure 11). Furthermore, Allen and Robbins (2010) found that academic discipline as measured by ACT Engage and interest–major congruence predicted timely degree attainment in college incrementally over prior test scores and college GPA.