The possible bias resulting from excluding nonresponders in tabulating data from mail survey questionnaires in a dentist population was investigated. The differences between the responders and the nonresponders were analyzed with respect to the dentists' demographic data, attitudes, and knowledge. Nonresponse bias did not affect the results of…

Misleading data reports concerning faculty pay and benefits have resulted from several errors: missing data, misleading data, misinterpreted data, and mistakes in data. Data will be understood only if examined in the context of institutional research, which takes into account the insights of the behavioral sciences. (CT)

A matrix approach can solve the problem of identifying quality technical education programs. This method focuses on the end products of the program rather than on the educational process itself. Educators must use their professional competencies to judge program quality based on follow-up data and costs per pupil per program. (CT)

To shed light on the prevalance of families choosing one residential area over another on the basis of the quality of schools in each area, the article explores whether a geography of intelligence exists in a small Kentucky town of approximately 20,000 people. Findings indicated significant spatial associations among IQ, achievement, and…

The use of multiple regression analysis was compared to the use of discriminant function analysis in the prediction of college faculty rank. The multiple regression technique was shown to be generally superior in this instance. (JKS)

When statistics are the focus, consumer/consultant relationships are often strained. Therefore, decision points in the research and consulting processes are examined for ways to clarify and improve the consulting relationship. A model for analyzing statistical techniques to provide a graphic basis for negotiation is presented. (Author/GDC)

To determine if an association exists between research methods and findings in attitude-behavior research, a review of the literature was conducted. More laboratory studies reported a positive relationship between attitudes and behavior, while more field studies found no such relationship. Extensive bibliography included. (Author)

Emphasizes the importance of computer instruction and statistical methods and suggests ways in which instructors of geography on the college level can alleviate student fears. Tips include discussing geography as a science, explaining the relevance of the scientific method to a specific research problem in geography, and introducing problem…

Examines the relationship between language universals and language acquisition, comparing data from studies on first and second language acquisition and universals of relative clause formation. It is concluded that second language acquisition data provide a clearer window for the investigation and verification of language universals than first…

Reports on a faculty evaluation system used at the College of Business Administration of the University of Georgia. Data analyzed included effects of publications, teaching experience, departmental affiliation, school from which Ph.D. was granted, and teaching effectiveness on faculty member's salaries. (Author/DB)

Describes the use of content analysis by a journalism class in studying television news. Indicates that the method is flexible, generates familiarity with quantitative approaches to the analysis of broadcast journalism, can result in increased awareness of the complexity of the broadcast news medium, and increases student motivation. (TJ)

A survey done properly by accepted and scientifically justifiable techniques can provide information on behalf of the goals of institutional advancement in academe. Deficiencies in institutional research are seen as the result of the lack of guidance on survey design, questionnaire construction, or problem definition in education research texts.…

Appropriate methods are described for research workers in sociology to use as they perform causal analyses of data from surveys. Particular emphasis is placed on the causal analysis of qualitative variables. (Author/DB)

The estimation of the "size of effect" of educational programs is a difficult problem in program evaluation. It is argued that the intentions of the program and the nature of the measures must be known in order to estimate program effects. (JKS)

The usual interpretation of suppressor effects in a multiple regression equation assumes that the correlations among variables have been generated by a particular structural model. How such a regression equation is interpreted is shown to be dependent on the structural model deemed appropriate. (Author/JKS)