Research texts may describe what should be done, but they are less likely to discuss how to go about doing it, how long it may take, and the problems that may confront the first-time investigator. Key factors include setting realistic goals, securing adequate samples, and obtaining clinical knowledge about the population. (CT)

In tracing the evolution of research methodology in the social sciences, the author notes that the quality of quantitative models outstrips the quality of data used with the models. He believes, however, that refinements in theory building will make prediction, forecasting, and social engineering possible. (Author/KC)

The costs of evaluation are affected by program characteristics and the extent to which it is important to be confident about proof and causality. They are also affected by decisions about issues, resources, evidence, data gathering, data analysis, and reporting. (CT)

Identifies the purpose and need for a vocational evaluation process in the correctional setting. Discusses the five steps in such a process: preliminary intake procedure, primary data gathering, hypothesis formulation phase, hypothesis testing phase, and outcome and/or follow-up. CCT)

Suggests creating a central community registry of private music teachers in a public library or a school music department. Discusses how to solicit listings and provides a sample questionnaire for obtaining listing information from teachers. (SJL)

Survey validity is seen to be dependent on a high rate of response. Practical advice from successful National Geographic Society surveys is provided including presurvey contacts, postsurvey follow-ups, incentives, and letters and questionnaires designed to encourage response. Response bias is not seen as insurmountable. (MLW)

Though hidden from view by being considered "non-existent", the meager evidence indicates that Hispanics have an unusually high arrest and incarceration rate. Hispanic background is rarely asked on the six major sources of criminal justice statistics--statistics of arrests, courts, prisoners, juvenile delinquency, crime victimization, and public…

While preserving the distinctive features of the case study, the case cluster method employs replicable sampling and, if desired, quantitative measurement procedures to complement qualitative analysis. It creates units of analysis that are based on theory and that have meaning for the actors and observers of the case. (Author/IRT)

Lists and discusses 17 "do" and 8 "don't" guidelines for educators and trainers to consider when designing data collection questionnaires. (MF)

Among the problems are disagreement on how much information should be collected and, because of the use of broad-band tests, the difficulty of deciding who is an applicant for particular job. (Author/IRT)

Three different methods of data collection were examined in which subjects judged proximity between object pairs. Significant differences in accuracy were found among the three methods, presumably due to differences in the extent to which subjects are able to describe their perceptions under the various methods. (Author/RC)

This study demonstrates that multiple matrix sampling procedures can be used to collect assessment data efficiently, unabstrusively, and reliably. (MB)

Identifies five methods of phenomenographic research in early work at Gothenburg University (Sweden): discursive; experimental; naturalistic; hermeneutic; and phenomenological. Argues that phenomenography is a productive form of research, even if the degree of methodological reflection applied by phenomenographers to their work is found wanting.…

Presents a model for monitoring school violence locally and linking national, district, and school level data, illustrating its application in a project site in Israel. Examples of reports generated to monitor school violence for the district as a whole and for each site are presented. The article concludes that this is a feasible and useful model…

Examined the differences between compliment responses collected with two different data collection procedures: naturally occurring data analyzed through conversation analytic methodology, and elicited data collected via a discourse completion task (DCT). The DCT was designed to evoke the same discourse context and preceding context observed in the…