Dunnett's procedure for comparing K-1 treatments with a control is discussed within the context of three nonparametric models: those of Kruskal-Wallis, Friedman, and Cochran. (Author/MH)

Purpose: The primary aim of the research reported here is to provide strategic guidance for the development of a national communication strategy to improve sun protection practice amongst young people. Design/methodology/approach: The research adopted an exploratory approach, employing qualitative focus groups to represent three population groups,…

This paper describes seven studies (all that could be found on the subject) comparing nine respondent selection procedures with regard to their potential to reduce refusal rates and to increase representativeness of samples; particular concerns are underrepresentation of men and multiple-adult households. The paper states that (1) the seven…

A pilot study to ascertain the most cost-efficient and statistically reliable way of obtaining data on drug use by means of the telephone was conducted. Typically, surveys of sensitive or stigmatizing information have been contaminated by an unknown amount of error due to evasive or incorrect responses. The randomized response technique, which…

It is sometimes desirable to obtain an estimated total-test score for an individual who was administered only a subset of the items in a total test. The present study compared six methods, two of which utilize the content structure of items, to estimate total-test scores using 450 students in grades 3-5 and 60 items of the ll0-item Stanford Mental…

The notion of comparing in curriculum evaluation is discussed, and some approaches to making comparative judgments about program effectiveness during the formative and summative evaluation stages are examined. Sources of information provided by external and internal program comparisons are described, and a model in which a policy capturing…

Details of a reasonably precise normalization technique for coefficient alpha are outlined, along with methods for estimating the variance of the normalized statistic. These procedures lead to the K-sample significance test. (RC)

The values and life-style preferences of 1978-79 university students are compared with those studied 30 years ago to determine whether university students' preferential rankings of C. Morris's "Ways to Live" (self-sufficient, carefree, etc.) have changed significantly. Possible reasons for differences between the generations are…

An alternative method for analyzing and evaluating prototype and revised instructional material is compared to traditional methods. The procedure and three approaches of sequential analysis are presented and advantages are discussed. References are included. (BK)

Two estimators of effect size that are based on the sample quartiles are proposed and studied. One is for the situation where a treatment effect is evaluated against a control group, and the other is for a situation in which two parallel treatments are compared. Both are illustrated and evaluated. (SLD)

Recently, C. E. McCulloch (1987) suggested a modification of the Morgan-Pitman test for comparing the variances of two dependent groups. This paper demonstrates that there are situations where the procedure is not robust. A subsample approach, similar to the Box-Scheffe test, and the Sandvik-Olsson procedure are also assessed. (TJH)

This research examined 2 elements of multiple-choice test construction, balancing the key and optimal number of options. In Experiment 1 the 3 conditions included a balanced key, overrepresentation of a and b responses, and overrepresentation of c and d responses. The results showed that error-patterns were independent of the key, reflecting…

We compared partial-interval recording (PIR) and momentary time sampling (MTS) estimates against continuous measures of the actual durations of stereotypic behavior in young children with autism or pervasive developmental disorder-not otherwise specified. Twenty-two videotaped samples of stereotypy were scored using a low-tech duration recording…

Not infrequently, investigators assume that reliability for groups is greater than reliability for persons, or that the error variance for groups is less than that for persons. Using generalizability theory, it is shown that this "conventional wisdom" is not necessarily true. Examples are provided from the course-evaluation and the…