TPC-Journal-V6-Issue2

The Professional Counselor /Volume 6, Issue 2 195 Table 1 gives the parameter estimates from the MLR that analyzed the effects of the predictor variables on postsecondary education status and presents two nonredundant logits since our criterion variable (postsecondary status) has three possible values: leaver, still enrolled in a two- year institution, and still enrolled in a four-year institution. When comparing those still enrolled in a two-year institution to those no longer enrolled, the only parameter estimate that was significantly different from zero was time spent in extracurricular activities. Those students with no extracurricular activities (β=-.25) compared to those with a high number extracurricular activities (15 or more hours per week) were less likely to still be enrolled in a two-year institution. When examining the second logit (those still enrolled in a four-year institution compared to those no longer enrolled in any postsecondary institution), two predictors were significant: talks to the math teacher outside of class and time spent in extracurricular activities. Those students who spoke with their math teacher outside of class increased their chances of still being enrolled in a four-year institution rather than being in the leaver group by a factor of 1.24. The parameters for homework were not significant. In regards to the number of weekly hours in extracurricular activities, the parameters for none and low (1–4) hours were significant. Those students who spent either no or a low number of hours in extracurricular activities compared to those with a high number of hours (15 or more) were less likely to still be enrolled in a four-year institution. The difference between a moderate number (5–14) and a high number (15+) of hours spent in extracurricular activities was not significant. Discussion Based on previous research about factors in college related to persistence, this study hypothesized nine criterion variables on the high school level to predict college persistence. The hypothetical question guiding this study was: Would the same variables on the college level known to influence persistence predict persistence when measured at the high school level? Three of these nine variables were significant in the overall model: talks with math teacher outside of class, number of hours spent weekly on homework, and number of hours spent weekly on extracurricular activities. Six of the nine variables were not significant: academic support, social support, talks with English teacher outside of class, has gone to counselor for college entrance information, performed community/volunteer service, and number of hours worked. As a result, our original model was replaced with a more parsimonious model of three predictor variables. Furthermore, number of hours spent weekly on homework, while significant in the overall model, was not a strong enough predictor to distinguish those who persisted in two-year colleges from those who left or to distinguish those who persisted in four-year colleges from those who left. In the end, the two predictors strong enough to differentiate among the three groups were: talks with math teacher outside of class and number of hours spent in extracurricular activities. Some of the predictor variables, like academic support and social support, were composite variables of just three Likert-scaled student-reported items. Thus, the reliability of these is questionable and may explain their lack of predictive value. Previous research (Kuh et al., 2008; Pascarella & Terenzini, 2005) has shown that college students with both academic and social support have a greater chance of persisting. Related to academic support, however, is seeking out and talking with professors outside of class. College students who interact with professors outside of class have a greater chance of persisting. The results of the present study indicate that high school students who spoke with their math teacher (not the English teacher) outside of class had a greater chance of persisting in a four-year college, but not necessarily in a two-year college. This result is not surprising as it was hypothesized that high school students who speak with their teachers outside of class would have a greater likelihood of doing so on the college level and, in turn, a greater likelihood of persisting in college. What may be surprising is that the predictive value lies