The common approach to studies of predictive bias is analyzed within the context of a conceptual model in which predictors and criterion measures are viewed as fallible indicators of idealized qualifications. (Author/PN)

Item bias, when present in a multiple-choice test, can be detected by appropriate analyses of the persons x items scoring matrix. Five related schemes for the statistical analysis of bias were applied to a widely used, primary skills multiple-choice test which was administered in either its English- or Spanish-language version at each of the two…

Fairness or unfairness may be an attribute of a test per se, or of its use, or of its statistical treatment. An hypothetical situation designed to be intrinsically fair and unbiased is used to show that analysis of covariance as a statistical method may introduce bias to the treatment of test scores. In contrast, equipercentile equating methods…

The predictive relationship between the WISC-R and the PIAT and WRAT scores of 125 Mexican-Americans and 125 non-minority students (K-8) was analyzed using four methods. Conventional methods were found to be insensitive to and inappropriate for the detection of predictive bias. An alternative means of detecting predictive bias is proposed.…

The purpose of this investigation was to establish the effects of repeaters on test equating. Since consideration was not given to repeaters in test equating, such as in the derivation of equations by Angoff (1971), the hypothetical effect needed to be established. A case study was examined which showed results on a test as expected; overall mean…

An intelligence test for use with rural Filipinos was developed using 25 general information questions. The questions bear on experiences common to most rural Filipinos. The methods of development and analyses are described. It is suggested that this strategy could be used with other peasant groups where an assessment of intelligence is needed in…

The purpose of this paper is to describe a set of procedures, that, when carried out, permit the conclusion that a test is a fair measure from the standpoint of specific sub-groups within a test population. A fair test is defined as a test for which a set of data-collection procedures have been carried out and the results reported. (Author/JM)

Differential item performance (DIP) is discussed as a concept that does not necessarily imply item bias or unfairness to subgroups of examinees. With curriculum-based achievement tests, DIP is presented as a valid reflection of group differences in requisite skills and instruction. Using data from a national testing of the ACT Assessment, this…

Several methods of assessing test item bias are described, and the concept of fair use of tests is examined. A test item is biased if individuals of equal ability have different probabilities of attaining the item correct. The following seven general procedures used to examine test items for bias are summarized and discussed: (1) analysis of…

Noting that the practitioner is confronted with conflicting definitions of "fairness" that claim to avoid "bias" in test usage, this article asserts that "mean fairness" can be achieved by use of multiple regression with the dichotomous variate, group membership, included as a predictor; mean fairness also holds for sub-groupings, based on test…

A method is proposed for the detection of item bias with respect to observed or unobserved subgroups. The method uses quasi-loglinear models for the incomplete subgroup x test score x item 1 x ... x item k contingency table. If the subgroup membership is unknown, the models are the incomplete-latent-class models of S. J. Haberman (1979). The…

The use of loglinear latent class models to detect item bias was studied. Purposes of the study were to: (1) develop procedures for use in assessing item bias when the grouping variable with respect to which bias occurs is not observed; (2) develop bias detection procedures that relate to a conceptually different assessed trait--a categorical…

Responses on both the state and trait scales of the State-Trait Anxiety (STAI) Inventory were examined under two conditions. The first condition presented a simulated real-life situation containing competitive and evaluative cues without directly suggesting faking and asked subjects to complete the STAI. After an intervening task, the STAI was…

During the past few years the problem of bias in testing has become an increasingly important issue. In most research, bias refers to the fair use of tests and has thus been defined in terms of an outside criterion measure of the performance being predicted by the test. Recently however, there has been growing interest in assessing bias when such…