Nearly half of undergraduate students majoring in science, technology, engineering, and mathematics (STEM) fields leave these majors either by changing their majors or leaving college altogether (Chen, 2014). Researchers have therefore called for inquiries of factors, such as self-efficacy, ability, and interest, that may be related to retention and attrition rates in STEM (Gayles & Ampaw, 2014; Le, Robbins, & Westrick, 2014; Litzler, Samuelson, & Lorah, 2014). Despite such calls, a research gap remains relative to predicting STEM retention and attrition based on career-related factors, such as career readiness and participation in career interventions. To address this gap, we sought to determine whether or not undergraduate student retention in STEM majors from 1st year to 2nd year could be predicted by students' (a) participation in a STEM-focused career planning class, (b) initial major choices, and (c) changes in career readiness assessment scores.
A few studies have examined career-related factors with regard to STEM recruitment or retention. Le et al. (2014) found that ability, as measured by the ACT assessment, and vocational interests, based on Holland's (1997) R1ASEC types (Realistic, Investigative, Artistic, Social, Enterprising, and Conventional), influenced students' decisions to choose STEM majors. However, Le et al. only examined these factors within the context of major selection, without any career intervention. Similarly, Lent, Lopez, Lopez, and Sheu (2008) tested a social cognitive career theory model with students in computing disciplines. Their results indicated that self-efficacy was a strong predictor of outcome expectations, interests, and major choice goals. Lent et al.'s model fit well when gender, ethnicity, education level, and university type (e.g., predominantly White institution or historically Black college/university) were added as grouping variables. However, their study also did not examine the effect of career readiness factors or participation in a career intervention with the participants. To redress this gap in previous studies, we included in our study participation in a career intervention and a measure of career readiness as predictors of retention in STEM majors. Our primary research question was to what extent undergraduate STEM retention (from 1st year to 2nd year) can be predicted by participation in a STEM-focused career planning course, students' initial majors, and changes in scores on a measure of career readiness?
Participants were members of a grant-funded STEM recruitment and retention program at a large university in the southeastern United States between fall 2012 and spring 2015. All participants in the program had a minimum SAT math score of 550 and an expressed interest in STEM disciplines. Participants' initial majors were classified as undeclared (n = 54), STEM (n = 212), or non-STEM (n = 49). Approximately 58% of students (n = 181) took a STEM-focused career planning course. Students taking this course included those without a declared major, STEM-interested students with a declared non-STEM major, and STEM majors who opted into the course. The remaining participants (n = 134) took a STEM seminar course without a career planning focus. Students in both courses responded to the Career Thoughts Inventory (CTI; Sampson, Peterson, Lenz, Reardon, & Saunders, 1996) during the first and last weeks of their first semester of college as a pretest and posttest measure of career readiness. Because we used change in students' CTI scores as an...