Data from infants unable to complete experimental sessions were compared to those for whom there were complete data. Results suggest that the elimination of large numbers of infants may have a potentially biasing effect on reported data. (Authors)

Much of educational administration research suffers from lack of conformity between data analysis and research design, inattention to the test/retest stability of measurement instruments, and the use of statistics for nominal or interval data with ordinal data, random-sample statistics with nonrandom samples, and unnecessarily complex statistical…

Sternberg (1981) presented correlations based on independent samples of about 25 cases each with a statistical comparison in a population parameter of zero and interpreted his results as if with appropriate tests. He needed to use a nonzero population parameter and compare for the difference between the z transformations. (Author/CM)

Sternberg's reply to paper Humphrey criticizes (see TM 507 045) includes: (1) invited position paper and detailed statistical testing would have been inappropriate; (2) Humphrey commits some of "misdeeds" of which he accuses Sternberg; (3) number of cases in the three nonentrenchment experiments are more than adequate for drawing conclusions; and…

Explored possible response bias in a needs assessment survey for a hospice program. Results showed differences between the 22 caregivers who responded to the survey and the 38 who did not respond. Bias in the study may have eliminated clients with the most difficult experiences. (Author/JAC)

While major test publishers randomly select school districts for their national norming studies, a survey of "accepting" and "declining" districts supports the hypothesis that self-selection bias results in overrepresentation of districts which already use a specific publisher's tests or instructional materials. (Author/BW)

Examined the hypothesis that the higher the divorce rate, the more likely that unhappy couples will terminate their marriages. Shows that when type of sample, dependent measure, and method of analysis are held constant, the increased divorce rate is not an important factor in explaining marital satisfaction. (Author)

A case study of a survey of new students made at a postsecondary vocational-technical school illustrates several sampling problems, including sample size, target population definition, determination of research purposes, costs, error tolerance, accuracy, and interpretation of research results for use by policymakers. (RW)

This research note examines respondent-nonrespondent bias in a statewide population survey. The findings indicate general support for using master lists in the selection of respondents for participation in a large mail questionnaire survey. Two papers follow which react to this information. (Author/CT)

This study investigated the anorectic's experience in daily living using the Experience Sampling Method. Results suggest that anorectics spend more time alone and experience lower average affect than other young single women. (Author/GK)

Brunswik's concept of representative design is reviewed with special reference to studies of clinical bias. Limitations of single-stimulus, actor-script, and serial replications are discussed. No satisfactory alternatives exist to adequate sampling of stimulus persons. Various attributes that are found in population must be properly represented in…

Examines parental sex role attitudes in relation to parental child rearing values of independence granting and pressure for achievement. Subjects of the study were natural mothers and fathers of White, middle class, two-parent families. (Author/DA)

Results indicate the Marital Interaction Coding System is a dependable and valid measure of marital interaction. Generalizability theory is cost effective and provides quality data. It presents a comprehensive view of reliability and validity. (JAC)

The study compared the productions of 35 articulation impaired children (ages 3 to 10) obtained by picture elicited single word and connected speech sampling methods. Productions were compared on the basis of both the number and type of errors made under the two testing conditions. (Author/PHR)