Educational researchers frequently study the impact of treatments or interventions on educational outcomes. However, when observational or quasiexperimental data are used for such investigations, selection bias can adversely impact researchers' abilities to make causal inferences about treatment effects. One way to deal with selection bias is to…

Education research rarely lends itself to large scale experimental research and true randomization, leaving the researcher to quasi-experimental designs. The problem with quasi-experimental research is that underlying factors may impact group selection and lead to potentially biased results. One way to minimize the impact of non-randomization is…

Few research studies have investigated the effectiveness of instructional strategies for students in juvenile corrections. The dearth of research on effective instruction for this population may be due in part to difficulties encountered in carrying out methodologically rigorous studies in these settings. This article reports barriers and…

This article discusses the difficulties in conducting intervention research or evaluating intervention programs in a school setting. In particular, the problems associated with randomization and obtaining control groups are examined. The use of quasi-experimental designs, specifically a paired comparison design using the individual as his or her…

This article discusses the use of propensity scoring in experimental program evaluations to estimate impacts for subgroups defined by program features and participants' program experiences. The authors discuss estimation issues and provide specification tests. They also discuss the use of an overlooked data collection design--obtaining predictions…

Discusses problems of internal and external validity in research in educational computing. Highlights include three stages of modern research in educational computing, the need to describe control groups and instructional methodologies, and the need for a balance between theory and practice. (LRW)

Criteria for suitable research designs for use with large databases are suggested and analyzed. The advantages and disadvantages of several types of quasi-experimental designs are compared. Examples are taken from the authors' research with data from the Manitoba Health Services Commission. (Author/AL)

Develops generalizations for empirical evaluation of software based upon suitability of several research designs--pretest posttest control group, single-group pretest posttest, nonequivalent control group, time series, and regression discontinuity--to type of software being evaluated, and on circumstances under which evaluation is conducted. (MBR)

Discusses an ethical dilemma that emerged in a study with bereaved college students. The instruments used to gather data clearly elicited grief-related distress, and more bereaved students in control groups left the study than did participants in social support groups. Three alternatives to a traditional control-group design are discussed for…

Recruiting schools and kids to participate in drug prevention evaluations is no longer, if it ever was, a "done deal." Since the mid-1980s, schools have been inundated with requirements for drug programs and requests to cooperate with program evaluations. As a consequence, both treatment and control schools are harder to find. In addition, many…

Research efforts have established that the Mantel-Haenszel procedure (MHP) is an effective method for detecting the presence of test items exhibiting differential item functioning (DIF). While the MHP has been advocated for situations where item response theory based methods may not be usable, recent findings have suggested that the performance of…