A meta-analysis by Tran, Sanchez, Arellano, and Swanson (2011) of the published RTI literature found that the magnitude of effect size (ES) between responders and low responders at posttest was significantly moderated by the pretest ES and the type of dependent measure administered, whereas no significant moderating effects were found in the mixed…

Semipartial correlation is one of several ways of determining the relative importance of independent variables in a multiple regression analysis. A veteran teacher of statistics and related topics explains his reasons for avoiding semipartial correlations. (SV)

Hierarchial causal models are described as pictorial representations of multiple regression equations. These models are particularly helpful for three reasons: (1) the formulation of problems in a path analytic framework forces a degree of explicitness that is often not present in research reports that rely solely on regression; (2) they provide a…

Suggests a critical methodological flaw in a study done about the relationship between the Bem Sex-Role Inventory and certain indices of self-disclosure (Stokes, et al.). Notes that multiple regression analysis was not performed in appropriate hierarchical fashion. Includes Stokes reply to the critique. (PAS)

This paper encourages the use of multiple logistic analysis in early intervention research, to assess the degree of association of multiple factors (such as subject or situational characteristics) with a dichotomous outcome (such as benefitting or not benefitting from an intervention) and to estimate the probability of each outcome. (JDD)

Examines the credibility of inferences based on cross-group comparisons of regression coefficients using Black-White differences in the educational attainment process as an example. Shows that reanalyses of data from previous studies fail to produce consistent results even when major differences in samples, models, and methods are held constant.…

Addresses three issues concerning the assessment of the overrating and underrating of academic competence: (1) the impossibility of separating effects of overrating and underrating from effects of perceived and actual competence; (2) the questionable validity of Connell and Ilardi's method; and (3) the proposal of a new method and its implications…

In a research note to a nine-volume, six-state study of the impact of school finance reform on minorities and the poor, the author examines methodological issues involved in using statistics to measure the distribution of local fiscal capacity among school districts with different income and ethnic compositions. Using Texas as an example, the…

Argues that perception studies and ranking are lacking in validity and reliability because of small sample size, simplistic statistical analyses, and the evaluation of quality by educators. It is asserted that quality control is more valid when evaluated by the customers of education, i.e., the students. (11 references) (EA)