This methodological note illustrates how a commonly used calculation of the Delta-p statistic is inappropriate for categorical independent variables, and this note provides users of logistic regression with a revised calculation of the Delta-p statistic that is more meaningful when studying the differences in the predicted probability of an…

This article presents a methodology for obtaining information about some of the significant variables relevant to the integration of graduates into the work force. After revisiting the main issues dealt with in previous research on the professional integration of graduates, research was carried out to identify the major issues and a survey was…

Utilizing data from the national administration of a Cooperative Institutional Research Program (CIRP) baseline and follow-up survey of the first-year experience during the 2002-2003 academic year, this study explores the relationship between three curricular interventions--first-year seminars, service-learning, and learning communities--and the…

In the wake of the civil rights movement, new organizations formed which were based on the collective interests and identities of their constituencies. Some of these organizations brought together national origin groups who often differed by ethnicity, language, culture, religion and immigration history. In this paper, I focus on the conditions…

Governments now expect universities to behave entrepreneurially generating revenues from the commercial exploitation of their intellectual assets ("third stream" income). This study questions whether this is possible for all universities in the UK and for all disciplines, or whether there are differences between institutions, between…

A study used a multivariate approach to assessing college departmental excellence. Factor analysis identified three potentially useful factor clusters ("faculty research,""input," and "student") and seven indicators based on those clusters. Indicators combining faculty research and input and combining all three clusters were found to be the best…

A study using discriminant analysis to correct bias in predicting faculty rank is reported. Findings concerning the relationship of prediction error and gender and suggestions for use of the results to study salary-bias problems are also discussed. (Author/MSE)

Research that has traditionally supported program accreditation has tended to be primarily descriptive in nature, focusing on outcomes such as the satisfaction of graduates and passing rates on licensure exams. The study reported here illustrates how research employing multivariate methodologies can go beyond such descriptive reporting to address…

Focusing on student retention and time to degree completion, this study illustrates how institutional researchers may benefit from the power of predictive analyses associated with data-mining tools. The following are appended: (1) Predictors; and (2) Variable Definitions. (Contains 5 figures.)

This paper describes a methodology used in an on-going retention study at New York University (NYU) to identify a series of easily measured factors affecting student departure decisions. Three logistic regression models for predicting student retention were developed, each containing data available at three distinct times during the first…

The use of of the Automatic Interaction Detector (program AID3 of the OSIRIS statistical package) to study a university program is discussed. The performance of students who took general physics lecture and laboratory concurrently is compared to the performance of those who took them separately. Five years of data are analyzed, covering 1,997…

Two quantitative enrollment projection techniques and the methods used by researchers at the Ohio Board of Regents (OBR) are discussed. Two quantitative approaches that are typically used for enrollment projections are curve-fitting techniques and causal models. Many state forecasters use curve-fitting techniques, a popular approach because only…

Two studies illustrate how omitting faculty rank as a predictor variable from gender equity studies of faculty salaries can lead to incorrect conclusions about gender discrimination. One uses hypothetical data constructed so there is no gender difference in salary, but omission of academic rank skews the results. The second uses data from a…

Quantitative and qualitative methods were used to study retention and first-semester grade point average at Boise State University (Idaho), a metropolitan university with many commuter students. The quantitative portion of the study was based on questionnaires completed by 235 freshmen; the qualitative part was based on interviews with a…

Topics concerning the adult learner that are relevant to institutional researchers are addressed in four articles: marketing, predicting success for adult students, enrollment projection, and follow-up studies of adult learners. In "Institutional Research in Support of Marketing the Adult Student," Lydia Jurand notes the importance of…