An assessment of the association between college attendance and changes in students' values

Thomas Oliver Kelly, University of Pennsylvania

Abstract

Subjects are those who participated as high school seniors in the National Longitudinal Study of the High School of 1972 (NLS) and who completed follow-up surveys in 1973–74, 1975, and 1977. 23,451 subjects participated in the NLS. The 7,080 subjects included in this study completed all surveys and components of the 327 survey questions/items contained within the initial and follow-up surveys used in the data analyses. A longitudinal observational study design is utilized and includes a control group (those who did not attend college) and a treatment group (those who attended college). Subjects' responses within the initial NLS survey identify the degree to which ten values-related statements are important. The follow-up surveys provide information on activity states at the time of response, including college attendance and degree received, if any. In the 1977 survey, subjects identify the degree to which the ten values-related statements are important. These responses are compared to initial responses to identify changes in each of the ten values-related statements. Logistic regression models are used to assess the association between college attendance and changes in values. Initial differences among subjects, in terms of personal characteristics and demographic data, are included in the models, to provide appropriate controls. The NLS data are flawed and assessment is limited. Measurement errors, resulting primarily from floor and ceiling effects because of the use of three-point Likert Scales and the failure of the NLS team to validate the values-related statements, are cited as the primary limiting factors. Results for three of the ten values-related statements are disregarded, due to the poor fit of the data. Results for five are considered with caution, due to the low proportion of the variance in the dependent variable explained by each model. However, college attendance is found to be a significant factor in students' changes in the importance of (1) finding the right person to marry and having a happy family life, and (2) getting away from this area of the country. Additional studies are needed to assess the association between college attendance and changes in students' values, based on a properly designed survey/data collection instrument.

Subject Area

Higher education|Academic guidance counseling|Developmental psychology

Recommended Citation

Kelly, Thomas Oliver, "An assessment of the association between college attendance and changes in students' values" (2002). Dissertations available from ProQuest. AAI3043896.
https://repository.upenn.edu/dissertations/AAI3043896

Share

COinS