Determinants of withdrawal behavior: An exploratory study

Research on college persistence has typically classified nonreturnees as dropouts. Recently, this practice has been criticized by Tinto (1987) who argues that such a practice merges together different types of withdrawal behavior whose determinants may vary as a function of the particular departure behavior under consideration. This paper empirically examines whether the determinants of decisions to withdraw from the institution are similar to those affecting decisions to transfer to other institutions of higher education for the 1984 entering freshman class at a large southern institution. Results provide support for Tinto's proposition of differentiating between different types of voluntary withdrawal behavior. While institutional commitment, academic performance, finance attitudes, and student perceptions of faculty concern for student development and teaching discriminated between persisters and dropouts, only final institutional commitment and final goal commitment discriminated between persisters and transfers.An earlier version of this paper was presented before the 1990 American Educational Research Association annual meeting.     

The determinants of progress to the doctoral degree

The time doctoral students expect to take to complete their degree is investigated in a model which stresses the effects of economic factors. The model also accounts for the interdependency among such variables as employment hours, study hours, household income, and part-time study. We find that field of study, amount of borrowing, household income, and study hours have direct effects on expected time to degree. Indirect effects also occur from employment hours and the amount of fellowship, scholarship, or grant aid. When the sample is separated into foreign and U.S. students, we find that U.S. students underestimate their time to degree. Also, household incomes are higher for U.S. students and increase the time to degree, while the opposite is true for foreign students. Teaching assistantships increase the time to degree for foreign students while research assistantships reduce it.     

Variation in class size,the class size paradox,and some consequences for students

Even where a total allocation of resources is fixed, it is often possible to vary the distribution of the given resources. This distribution can have important, if sometimes hidden, consequences. In the case of universities, considering the mean class size as given, three nonobvious consequences of the amount of variation in class size are examined. (1) The average class size experienced by students is directly related to the amount of variation in class size. (2) Overall student attendance is inversely related to the amount of variation in class size. (3) Overall student participation time is directly related to the amount of variation in class size. The logic of these relationships is explored, and data on 49 departments and interdisciplinary programs at one university are used to illustrate the nature and extent of the class size paradox at one university.     

Gender differences in the academic career

This study examines sources of satisfaction, indicators of stress, academic performance, and reward structure in the academic careers of male and female faculty at a major midwestern state university. Faculty pairs were selected from five academic divisions. A total of 58 matched pairs were identified. Of these, 63 faculty responded, yielding data for 23 matched pairs. Subjects were mailed the Academic Career Development Survey, consisting of 144 items. Results indicated that self-reported physical and mental health, and professional and personal life satisfaction were high for both males and females. Both males and females placed high importance on their career and on marriage/other intimate relationships; however, both reported high levels of dissatisfaction with these factors. Gender differences were found in both satisfaction and type of stress reported; no gender differences were found in composite teaching and research performance indices. Salary data, obtained for 42 matched pairs, indicated similar salary levels for men and women. Results were discussed with respect to prior research on gender differences and the academic career.     

Personal construct psychology as a constructivist approach to learning

This paper proposes that Kelly's Personal Construct Psychology deserves examination as a constructivist basis for science teaching and learning. It argues that because of the explicit nature of the psychology, the clear definition of learning and meaning and the integration of affective, psychomotor and cognitive dimensions of learning, the psychology has much to offer science education.     

The effects on teaching of changes in relative rewards for research and teaching

To the degree that faculty members take into account the relative financial rewards for teaching and research in deciding on the allocation of available time between these two areas, a modification of the reward structure may be expected to produce changes in the amount of time devoted to each. The purpose of this article is to examine with the help of some basic graphic tools of economics the possible results on the quantity of teaching and the quantity of research that may follow from a modification of the reward structure. The possible results are analyzed in terms of displacement, salary, and work effects of the modification.     

THE GENETIC ASPECTS OF HIGH INTELLIGENCE

A cohort of 71 pairs of twins was studied in a project which started within the decade 1965 to 1975. The total level of intelligence (global IQ) and the level of particular intellectual components have been examined annually since the twins were two years of age. The data were analysed by comparing monozygotic and dizygotic twins. The results revealed that high intelligence is influenced much more by hereditary than exogenous factors. Furthermore, there is an irregularity in the development of intellectual level from childhood to adulthood. Consequently, decision making about children cannot be carried out on the basis of a single IQ evaluation. A cluster analysis demonstrated the importance of hereditary factors not only on intellectual level, but also on the pattern of intellectual development.