This article is designed to give the reader a conceptual, nontechnical overview of estimation and model fit issues in multilevel modeling (MLM). The process of MLM generally involves fitting a series of multilevel models that increase in complexity. When conducting multilevel analyses, it is important to balance the need for complexity and the…

Prior research studies associated the employment of part-time faculty with student degree and/or certificate completion (Benjamin, 2002; Ehrenberg & Zhang, 2005; Jacoby, 2006; Leslie & Gappa, 2002; Umbach, 2008; Umbach & Wawrzynski, 2005). To date, institutional-level data have been utilized to investigate whether such employment…

The purpose of this chapter is to discuss the conceptual, methodological, and statistical challenges in selecting appropriate peer institutions for comparative purposes. The authors' approach embraces a Western scientific tradition that physical things and phenomena can be "reduced into a set of key variables--identifiable parts--that make key…

This study uses a regression model to determine if a significant difference exists between the actual budget allocation that an academic department received and the model's predicted budget allocation for that same department. Budget data from a Southeastern Master's/Comprehensive state university were used as the dependent variable, and the…

The influence of financial aid on college student persistence was studied using causal modeling. The study population was 227 full-time freshmen, who enrolled in fall 1982 at a liberal arts college at a large commuter institution. All the students received financial aid. The model, which included background and academic variables, accounted for…

This paper argues that a typical use of regression models to target student recruitment efforts is theoretically unsound and may therefore be operationally inefficient. It presents results from a study using a predictive model to identify the prospective students on whom recruitment efforts have the greatest impact. The model uses four kinds of…

A comprehensive faculty flow model developed to forecast a "committed resources index" analyzes the future flexibility of a university. The model's construction, implementation, and assessment are described. (Editor/LBH)

Linear regression models that predict faculty salaries from professional and affirmative action variables were developed to validate promotion and merit criteria, set a starting salary scale, and identify possible systematic and individual salary inequities. All 377 full-time teaching faculty and librarians (309 men and 68 women) employed by a…

The development of a predictive equation capable of explaining a significant percentage of enrollment variability at Florida International University is described. A model utilizing trend analysis and a multiple regression approach to enrollment forecasting was adapted to investigate enrollment dynamics at the university. Four independent…

This study applied the methodology of competing risks survival analysis to determine the probability that a student's first enrollment in the university will end in graduation, transfer, or withdrawal. The risk factors associated with each mode of exit were assessed, with attention to factors such as admission status, full-time or part-time…

A model to distinguish persisters and nonpersisters at the University of North Carolina, Wilmington, was developed. Persisters and nonpersisters differed in background characteristics, attitudes toward and interaction with the university, opportunities to leave, and intentions. Responses to a mail survey were returned by 86 first-time freshmen…

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…

A two-dimensional model proposing three predictor dimensions (cognitive, noncognitive, and combined) and four criterion dimensions (grade point average (GPA), enrolled or not, graduate or not, and less traditional criteria) is presented. Preliminary results of a study using the model to examine both white and minority student predictors of…

A model of administrator development was tested in the 21 colleges, graduate and professional schools of a major state university system. Data were gathered from 494 administrators. Administrator development was measured in three ways: (1) individual rate of advancement, (2) perceptions of mobility and "success," and (3) perceptions of…

The relationship between student satisfaction and performance and persistence were studied at a large midwestern, urban commuter university. A student satisfaction survey was completed by 1,643 students (out of 3,004 students), who rated their level of satisfaction with 48 specific and 5 general aspects of their college experiences. Performance…