Ingemar Wedman

The Influence of Sex, Education and Age on Test Scores on the Swedish Scholastic Aptitude Test.

Abstract This study describes the effects of sex, education and age on the total test score on the Swedish Scholastic Aptitude Test (SweSA T), a test used in the selection process to colleges and universities in Sweden since 1977. Its use has so far been limited to one of four quota groups consisting of applicants 25 years or older and with more than four years of work experience. Statistical methods used in this study are regression models with dummy variables and estimated with a corner‐point parameterization. The results indicate rather genuine differences in every variable studied. Test takers with a higher education obtain a higher mean score than those with a lower education and older test takers obtain a higher mean score on the subtests vocabulary (WORD) and general information (GI) than younger persons. The mean test score for men is higher than the corresponding score for women, even if differences in education and age are controlled for. Finally some statistical problems related to the analysis...

The Significance of Contents for Sex Differences in Test Results.

Abstract Wedman, I. & Stage, C. 1983. The Significance of Contents for Sex Differences in Test Results. Scandinavian Journal of Educational Research 27, 49‐71. A common observation when using tests is that men and women arrive at different results. One of the proposed explanations to this fact is that the contents of the major parts of the tests are more adjusted to mens activities and interests. However, there are still many questions regarding the significance of the contents for obtained sex differences. The main purpose of this study was to examine to what extent it is possible to decide, by looking at the contents of the test items, which items are easy or difficult to judge for expected sex differences. Based on the item pool of the Swedish Scolastic Aptitude Test, two tests were constructed, a synonym test and a general knowledge test. The subjects were instructed to answer the items and to state whether they considered each item to be “male”, “female” or “neutral”. Results of this study show that...

Some Problems in Construction and Evaluation of Criterion-Referenced Tests.

Abstract Henrysson S. & Wedman I. 1974. Some Problems in Construction and Evaluation of Criterion‐Referenced Tests. Scand. J. educ. Res.,18, 1‐12. The interest in criterion‐referenced tests has increased in a very obvious way in recent years. There are, however, many problems associated with a criterion‐referenced testing procedure. In the first part of this study the effect of the aims on the construction of the tests, the homogeneity of the items and the cutting‐scores for what should be regarded as satisfactory knowledge are dealt with. Concerning item‐analysis of criterion‐referenced tests several discrimination measures have been proposed. Some of these are discussed in the second part together with empirical data showing intercorrelations between them. The report ends with some comments on future work with criterion‐referenced tests.

Analysis of the Inter‐Item‐Covariance Matrix

Abstract Henrysson, S. & Wedman, I. 1972. Analysis of the Inter‐Item‐Covariance Matrix. Scand. J. educ. Res. 16, 25‐35. It is shown how some different formulae within test theory can be obtained through different additions in the variance‐covariance matrix. Among other things a simple relation is derived, one between centroid factor loadings and KR 20, as well as one between centroid factor loadings and the standard error of measurement. The first one may be of help when selecting items in order to maximize the homogeneity of a test. With the covariance‐variance matrix as a basis it is finally discussed how to effect the test theoretical teaching, the forming of computer routines, and the further development of test theory.

Selection Based Upon Conventional Data and Value Judgements: A Comparison of Three Methods for Selection.

Abstract Wedman, Ingemar 1978. Selection Based upon Conventional Data and Value Judgements. A Comparison of Three Methods for Selection. Scandinavian Journal of Educational Research 22, 65‐82. Three different selection methods are discussed and compared. One of these, ‘A predictive system of selection based on qualitative judgements’, makes use not only of conventional data but also of subjective judgements. The other two methods are the proportional system of selection and the predictive system of selection based on strict regression analysis. Fictitious data based on the situation of selection to higher education in Sweden are presented to illustrate some of the effects of the three selection methods. These methods may have quite different effects on whom to accept or reject. Finally, some comments are given on selection in general and on the method based on qualitative judgements in particular.

On Test-Wiseness and Some Related Constructs.

Abstract Nilsson, I. & Wedman, I. 1976. On Test‐Wiseness and Some Related Constructs. Scand. J. educ. Res. 20, 25‐40. The present work treats certain concepts which have been used to denote irrelevant variation in observed test scores. The review is mainly limited to objective knowledge and ability tests but also deals to some extent with personality tests. The area has been characterized by a certain conceptual confusion in so far as the same characteristic has been denoted by different concepts while the same concepts have denoted different characteristics. The consequence has been that results from previous studies are often difficult to interpret and consequently today afford little possibility of formulating more precise statements about those errors the concepts represent and their relation to the relevance of the test scores. The concepts treated are ‘test‐wiseness’, ‘coaching’, ‘response‐set’, ‘response‐style’ and ‘response bias’. In conclusion we present a proposal for the systcmatization of the ...

Allocating Individuals into Mastery States on the Basis of Expected Utilities1

Abstract:Wedman, I. 1981. Allocating Individuals into Mastery States on the Basis of Expected Utilities. Scandinavian Journal of Educational Research 25,9‐19. This study describes a procedure for making decisions in connection with criterion‐referenced measurements. The procedure, which may be called a full Bayesian procedure, uses continuous utility functions instead of a dichotomized utility structure. The basic idea is to combine the posterior distribution for a certain person with utility functions for ‘advance’ and ‘retain’ decisions respectively. The decision is then made in favour of the action with the largest expected utility. Fictitious data are presented to illustrate the procedure.