The statistical techniques used in American Educational Research Journal (AERJ) research articles between 1979 and 1983 were coded by (a) type of technique, and (b) "major" or "minor" importance for comprehending the research findings. Twenty-seven types of technique were identified and further classified as "basic," "intermediate," or "advanced" level, in terms of typical statistics course progression. Of the "major" techniques, 33%, 37%, and 17% were basic, intermediate, and advanced level,, respectively. A review of statistics texts currently in use revert led that basic texts predominate and that the content coverage in both the basic-and intermediate-level texts generally matched well the techniques found in the AERJ articles. The findings suggest that students with both a basic-and intermediate-level knowledge of statistics would understand most of the techniques encountered in AERJ. Other topics addressed included an examination of trends in use of statistical techniques over time, and a comparison of AERJ techniques with those used in a sample of Journal of Educational Psychology (JEP) articles for the same time period.
This study presents some frequently encountered measurement misconceptions and various measurement "rules." Included are four misconceptions pertaining to the estimation of validity and reliability, and six rules often used during instrument development. Of these six, two pertain to the estimation of internal consistency reliability and item analysis, and four pertain to factor analysis. Whenever possible, the origins of the misconceptions and rules are described, along with the reasons why they are problematic. Alternate approaches or considerations also are given. When estimating validity and reliability, researchers are encouraged to be "ruled" by logical and critical evaluation of data rather than by arbitrary rules.
Interest in and use of qualitative methodological strategies in evaluating research have increased considerably in the last few years. Many of the recent evaluation frameworks or models are entirely or partly oriented toward use of qualitative methods. A number of methodological issues and concerns have been raised, including the appropriateness of validity and reliability estimation for the measurement strategies employed in qualitative evaluations becoming more common in health and other fields. In this article, the views of prominent qualitative methodologists on this topic are briefly summarized; a case is made for the relevance of validity and reliability estimation; definitions of validity and reliability for qualitative measurement are presented; and appropriate estimation techniques are suggested. It is hoped that discussions such as this will promote increased attention to validity and reliability concerns in qualitative evaluations and thus help improve the quality of those evaluations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.