In educational research, rigid distinctions between the relative merits of quantitative research often rest on positivistic dogma. In contrast, post-positivistic thought eliminates the intractable problem engendered by positivistic epistemology of a forced choice between value-laden and value-free research methods. (KH)

This study's interests dictate which statements researchers attend to during interviews and which ones researchers consider to constitute data during later analysis. The paper shows that hearing all of what students are saying requires careful effort. Explicit consideration of possible research agendas is suggested to increase opportunities for…

This book outlines the basic elements of longitudinal qualitative data, focusing on micro-levels of change observed within individual cases and groups of participants, and presents 16 questions through which researchers can approach the analysis of longitudinal qualitative data. The chapters are: (1) "Longitudinal Qualitative Studies, Time, and…

Data can help identify solutions to some of the biggest problems in schools. Yet few schools use data effectively. The purpose of this book is to help schools become aware of the advantages of collecting and using data (data analysis) for overall school improvement. The book begins with an overview of the current state of education and the…

Data analysis has become a topic of increasing emphasis within middle school mathematics in the last few years, especially in the recent recommendations by the National Council of Teachers of Mathematics (NCTM 2000). In order to better inform efforts to expand data analysis's role in middle school curricula, we have begun the development of a…

Reports how students use raw data to create a number of narratives integrating group observations in the laboratory with individual responses to humanities readings. (Contains 21 references.) (DDR)

Found that in an introductory biology course at the University of Massachusetts Amherst, adoption of the Data Aggregation Project (free software that allows students in a course to aggregate a large body of data for some particular problem) resulted in an increase in the number of data points students analyzed per semester by a factor of almost…

Notes two faulty aspects of the Reznick et al., twin study (PS 526 688): the expressive language measure at 14 months, which has practically no spread of item difficulty, as well as measures included to assess specific cognitive characteristics; and the notion of infant transition as it affects interpretation of the results. (HTH)

A meta-analysis combining expert evaluation with naturalistic case study methods indicates that such investigations must use special methods to render evaluative judgments of worth. It is demonstrated that descriptive, interpretive, and evaluative aspects of such a study must be combined to yield justifiable conclusions. (TJH)

Canonical weights and structure correlations are used to construct low dimensional views of the relationships between two sets of variables. These views, in the form of biplots, display familiar statistics: correlations between pairs of variables, and regression coefficients. (SLD)

Discusses three topics related to interpretation of discriminant analyses (DA's): (1) partial F ratios and partial Wilks's lambdas for predictor variables in standard, step-down, and stepwise DA's; (2) relation of goals of classification to definition/evaluation of classification rules; and (3) significance tests for total hit rates in internal…

Questions the adequacy with which the giftedness construct is defined in applied educational settings, and the adequacy of the validation data available for these gifted identification procedures. (Author/BJV)

One hundred thirteen elementary students were tested using four standardized instruments; their teachers rated each student on eight bipolar constructs and predicted scores. Results indicate that students who are perceived positively receive higher score predictions, even when academic competence is controlled. Boys seemed to be especially subject…

The entire thesis of "The Bell Curve" disintegrates due to biased use of data, misrepresentations, and logical inconsistencies. Five basic flaws are: inferring causality from correlation, use of dubious racial categories, contradictory arguments concerning the immutability of cognitive ability and the relative contributions of heredity…

Comments on some methodological limitations of the research base of "The Bell Curve": blind use of the normal distribution (bell curve); avoidance of nonnormal statistical distributions, which are more appropriate for some social and economic characteristics; copious use of percentiles and quintiles, inappropriate with nonnormal…