Differences Among Community College Students in Anxiety-Coping Typologies

• In Event: Roundtable Session 3
  In Roundtable Session: Enhancing Community College Student Success: Influence of Psychosocial Factors in Achievement, Persistence, and Degree Pursuit

Abstract

High levels of test anxiety have long been associated with poorer academic outcomes in high school and college students (e.g., Chapell et al., 2005; Hembree, 1988; Keogh et al., 2004). Test anxiety has also been positively associated with different coping behaviors such as emotion-oriented coping and avoidance (Cohen, Ben-Zur, & Rosenfeld, 2008; Zeidner, 2007), although such coping strategies appear to exhibit differential relationships with academic achievement (Nett, Goetz, & Daniels, 2010). Coping via avoidance behaviors has further been observed to exhibit a moderating effect of test anxiety and maladaptive outcomes such as negative affect (Stowell, Tumminaro, & Attarwala, 2008).

Gender differences have also consistently been observed with regard to test anxiety. Females tend to exhibit higher levels of general test anxiety than males (Putwain, 2008), and there is also some evidence that gender differences are more prominent with regard to specific facets (Stober, 2004). Gender may also moderate the relationship between test anxiety and academic performance (McCarthy & Goffin, 2005).

General Test Anxiety and Coping are broad, complex constructs appropriately recognized in the literature as being comprised of multiple facets (Nes & Segerstrom, 2006; Cohen et al., 2005). Zeidner’s (2007) review of the test anxiety literature highlighted the need for more methodologically advanced studies and the incorporation of other constructs such as coping responses. Although multivariate methods have been previously used to group students based on test anxiety and learning strategies (Birenbaum, 2007; Davis, DiStefano, & Schutz, 2008), the development of typologies referencing the combination of test anxiety and coping behaviors (and the relationship of such typologies to academic outcomes) remains a gap in the literature. Moreover, the psychological well-being of two-year college students is an understudied area despite having more severe psychological concerns and fewer resources than four-year students (Katz & Davison, 2014). The current study addresses this gap in a community college sample using modern analytic techniques.

Using a person-centered approach, we employ latent profile analysis (LPA) to identify underlying heterogeneity accounting for student test anxiety and coping mechanisms. A large (N = 2,941) sample of entering two-year college students completed a broad psychosocial assessment where two focal constructs were Test Anxiety (24 items) and Coping (27 items). Each item group targeted three facets. Closely aligned to Stober’s (2004) review, Test Anxiety was manifested by physical symptoms, interference, and worrying behaviors. General Coping, in keeping with previous meta-analytic work (Nes & Segerstrom, 2006), was composed of problem-approach, problem-avoidance, and emotion-avoidance coping behaviors.

LPA identifies the number of profiles and probabilistically assigns students to typologies. Models with different numbers of profiles are compared with solutions evaluated via global goodness of fit (Pearson chi-squared, parametric bootstrap), local fit (average class assignment probabilities), likelihood ratio tests, and Bayesian Information Criterion. Results indicated distinct Anxiety-Coping profiles differentiated with regard to academic outcomes. Multiple-group analyses were conducted to examine gender differences in expressed typologies and profile membership. Females appeared to demonstrate qualitatively distinct latent profiles and associations with academic outcomes versus males. Implications for future classroom interventions and college advising are discussed.

Authors

• Samuel Rikoon, Educational Testing Service

• Margarita Olivera Aguilar, Educational Testing Service