Allen L. Edwards

An internal consistency check for scale values determined by the method of successive intervals

The method of successive intervals is a psychological scaling procedure in which stimuli are classified into successive intervals according to the degree of some defined attribute which they are judged to possess. A psychological continuum is defined and the scale values are then taken as the medians of the distributions of judgments on the psychological continuum. It is assumed that the distributions of judgments for each stimulus are normal on the psychological continuum as defined.An internal consistency check indicates that the cumulative distributions of empirical judgments for the various stimuli can be reproduced by means of a limited number of parameters with an average error that compares favorably with that usually reported for paired comparison data. Furthermore, the scale values obtained by successive interval scaling, for the data reported, are shown to be linearly related to those obtained by the method of paired comparisons.

Note on the correction for continuity in testing the significance of the difference between correlated proportions.

Methods of correcting for continuity in tests of significance of the difference between correlated proportions are presented. These corrections should increase the range of usefulness of the formulas developed by McNemar (1).

Social Desirability and the Factorial Interpretation of the MMPI

MESSICK and Jackson (1961) have reviewed eight factor analyses of Minnesota Multiphasic Personality Inventory (MMPI) scales. For each factor analysis, they calculated Spearman rank correlations between the proportion of items keyed True in each scale and the loadings of the scale on the largest factor. The rank correlations ranged in magnitude from approximately .52 to .91, and Messick and Jackson (p. 300) state: &dquo;These strikingly consistent findings indicate that in most of these studies the largest factor on the MMPI is interpretable in terms of acquiescence.&dquo; Six of the eight studies reviewed by Messick and Jackson involved a limited number, 11 to 15, of the MMPI scales. Only two studies,

Response Sets in Standard and Experimental Personality Scales

In a factor analysis of 58 Minnesota Multiphasic Personality Inventory (MMPI) scales and 3 other personality scales (Edwards, Diers, and Walker, 1962), it was found that the first factor could be interpreted as the tendency to give socially desirable responses and the second as the tendency to give acquiescent responses. The third factor was tentatively identified as the tendency to falsify answers to personality items. The present study was undertaken to determine whether the same three factors could be identified in a new battery consisting primarily of non-MMPI scales.

A Factor Analysis of the EPPS and PRF Personality Inventories.

Two multi-scale personality inventories have followed Murray’s need structure theory of personality in the formulation of their scale definitions and item domains. Edwards (1957a), on the basis of the list of manifest needs presented by Murray and others (1938), developed the Edwards Personal Preference Schedule (EPPS) which measures the strength of 15 needs: Achievement (ach), Deference (def ), Order (ord), Exhibition (exh), Autonomy (aut), Affiliation (ajy), Intraception (int), Succorance (suc), Dominance (dom), Abasement (aba), Nurturance (nur), Change (chg), Endurance (end), Heterosexuality (het), and Aggression (agg). The EPPS uses a forced-choice item format in which two state-

Relationships Between Various Psychometric Properties of Personality Items

EDWARDS (1953) originally reported a correlation of .87 between the probability of endorsement of a personality item and the social desirability scale value of the item for a set of 140 personality items. This finding has been confirmed in various other studies (Cowen & Tongas, 1959; Edwards, 1957a, 1957b, 1959; Hanley, 1956; Hillmer, 1958; Kenny, 1956; Taylor, 1959; Wright, 1957). Both the probability of endorsement of a personality item and the social desirability scale value of an item may be regarded as item characteristics or parameters. Additional item parameters may also be specified. One of these is the dispersion or standard deviation of the distribution of social desirability ratings assigned to an item. Another is the probability that an item will be marked doubtful if subjects are given the opportunity to so mark an item when they are not sure as to whether it does or does not describe them. A

On Guttman's Scale Analysis

present stage of development. One of these difficulties concerns the location of cutting points for various items. With perfect reproducibility there would be, of course, no problem. The cutting points for an item would simply be the dividing points in the rank-order scores where the response shifts from the more to the less favorable category. Items are perfectly reproducible when from rank-order scores we can reproduce all the responses to the individual items. All subjects with rank-order scores above the cutting point would respond by checking the more favorable alternative and all those below would respond by checking the less favorable alternative. There would be, in the case of perfect reproducibility, no overlap in the responses of the subjects below and above the cutting point. But perfect reproducibility is an ideal case and is not

A principal-components analysis of the Minnesota multiphasic personality inventory factor scales

Johnson, Butcher, Null, and Johnson (1984) reported on factor scales developed from the Minnesota Multiphasic Personality Inventory (MMPI). A principal-components analysis of the 21 factor scales and Edwardss (1957) Social Desirability (SD) scale gave 5 factors. The SD scale had the highest absolute loading on the 1st MMPI factor. Also, the 1st-factor loadings of the scales were linearly related to the SD intensity intensity index of the scales. A 2nd principal-components analysis of 16 of the Johnson et al. scales along with marker scales for the first 3 MMPI factors also resulted in 5 factors; the SD scale again had the highest absolute loading on the 1st factor, and the 1st-factor loadings were again linearly related to the SD intensity index. On only 2 of the 5 factors did any of the factor scales have higher absolute loadings than 1 of the marker scales, suggesting that Johnson et al. were overly optimistic that their scales may serve as marker scales

Social Desirability as a Variable in 2 Technique Studies

Stephenson (4) has described the use of &dquo;structured&dquo; and &dquo;unstructured&dquo; samples of items in ~ technique. By an unstructured sample he means a set of propositions which have not been subdivided in any way by the experimenter into smaller subsets. Thus a set of items in a personality inventory, designed to yield but a single score, might be taken as an example of an unstructured sample. By a structured sample, Stephenson means that we have at least two kinds of items

Scale analysis and the measurement of social attitudes

This paper discusses and compares the methods of attitude scale construction of Thurstone (method of equal-appearing intervals), Likert (method of summated ratings), and Guttman (method of scale analysis), with special emphasis on the latter as one of the most recent and significant contributions to the field. Despite a certain lack of methodological precision, scale analysis provides a means of evaluating the uni-dimensionality of a set of items. If the criteria for uni-dimensionality are met, the interpretation of rank-order scores is made unambiguous, and efficiency of prediction from the set of items is maximized. The Guttman technique, however, provides no satisfactory means of selecting the original set of items for scale analysis. Preliminary studies indicated that both the Likert and the Thurstone methods tend to select scalable sets of items and that their functions in this respect are complementary. A method of combining the Likert and Thurstone methods in order to yield a highly scalable set of items is outlined. Sets of 14 items selected by the method have, in the two cases where the technique has been tried, yielded very satisfactory scalability.