Few empirical investigations have evaluated learning disabilities (LD) identification methods based on a pattern of cognitive strengths and weaknesses (PSW). This study investigated the reliability of LD classification decisions of the concordance/discordance method (C/DM) across different psychoeducational assessment batteries. C/DM criteria were…

Although advances have improved our ability to describe the measurement precision of a test, it often remains challenging to summarize how well a test is performing overall. Reliability, for example, provides an overall summary of measurement precision, but it is sample-specific and might not reflect the potential usefulness of a test if the…

A differential item functioning (DIF) simulation study was conducted to explore the type and level of impact that contamination had on type I error and power rates in DIF analyses when the suspect item favored the same or opposite group as the DIF items in the matching subtest. Type I error and power rates were displayed separately for the…

This study considers the problem of deciding when a selection test is invalid for members of a minority group. There is both a strong empirical and theoretical basis for rejecting the choice of zero correlation between test and criterion as an appropriate null hypothesis. (Author)

Discusses several statistical and substantive criteria that can be used to choose between parametric and nonparametric tests. Presents a non-parametric test capable of testing a number of statistical hypotheses using existing computer packages. Provides recommendations encouraging researchers to routinely use nonparametric tests in their data…

The conventional approach for testing the equality of two normal mean vectors is to test first the equality of covariance matrices, and if the equality assumption is tenable, then use the two-sample Hotelling T[superscript 2] test. Otherwise one can use one of the approximate tests for the multivariate Behrens-Fisher problem. In this article, we…

Cross-validation in relation to choosing the best tests and selecting the best items in tests is discussed. Cross-validation demonstrated whether a decision derived from one set of data is truly effective when this decision is applied to another independent, but relevant, sample of people. Cross-validation is particularly important after…

Some aspects of test reliability are discussed. Topics covered are: (1) how high should a reliability coefficient be?; (2) two factors affecting the interpretation of reliability coefficients--range of talent and interval between testings; (3) some common misconceptions--reliability of speed tests, part vs. total reliability, reliability for what…

Outlines the purposes of statistical analysis and types of research questions that can be answered. Defines data types as a guide for choosing statistical tools. Describes a hypothetical research study on teaching computer software to illustrate the use of analytical tools. (SK)

Ten item discrimination indices, across a variety of item analysis situations, were compared, based on the validities of tests constructed by using each of the indices to select 40 items from a 100-item pool. Item score data were generated by a computer program and included a simulation of guessing. (Author/CTM)

The results of the present study suggested that PPVT and SIT IQs are not directly comparable to the WISC-R IQs when the moderate to low correlation coefficents are considered. In addition, the SIT appears to be more comparable to the WISC-R than does either Form A or Form B of the PPVT. (Author)

The data collected in higher education research are not always quantitative or continuous. Statistical methods using the log-linear model provide the institutional researcher with a powerful set of tools for addressing research questions when data are categorical. (Author/MSE)

Regression analysis is one of the most frequently used statistical techniques in institutional research. Principles of least squares, model building, residual analysis, influence statistics, and multi-collinearity are described and illustrated. (Author/MSE)

Path analysis and linear structural relations (LISREL) provide the institutional researcher with some extremely powerful statistical tools. However, they must be applied and interpreted carefully with a full understanding of their limitations and the statistical assumptions on which they are based. (Author)

The relationship between total scores and interrelationships among certain similarly named subtests of the Metropolitan and Stanford Achievement Tests are considered. The correlation of total scores for the two batteries was .89, indicating that the batteries measure essentially the same things. If one were interested in selecting a battery to…