Test-Retest Effect on Personality Questionnaires

Windle, Charles

doi:10.1177/001316445401400404

Cited by 128 publications

(62 citation statements)

References 80 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For over 50 years, researchers have noticed that reports of levels and severity of certain symptoms and maladaptive experience diminish over repeated assessments (1)(2)(3)(4).…”

Section: Conflicts Of Interestmentioning

confidence: 99%

Initial elevation bias in subjective reports

Shrout

Stadler

Lane

et al. 2017

Proc. Natl. Acad. Sci. U.S.A.

173

194

View full text Add to dashboard Cite

Initial Elevation of Reports 2 SignificancePeople's reports of their own thoughts, feelings and behaviors are essential assessment tools in biomedical and social science. They be used to take a snapshot of how people are doing and to track change and the effects of interventions. When subjective states have been studied over time, researchers have often observed an unpredicted and puzzling decrease with repeated assessments. Our results across multiple outcomes in four field experiments suggest that this pattern is due to an initial elevation bias. This effect is larger for reports of internal states rather than behaviors and for negative mental states and physical symptoms than for positive states. This initial elevation bias needs to be considered in all types of research using subjective reports. AbstractPeople's reports of their thoughts, feelings and behaviors are used in many fields of biomedical and social science. When these states have been studied over time, researchers have often observed an unpredicted and puzzling decrease with repeated assessment. When noted, this pattern has been called an "attenuation effect", suggesting that the effect is due to bias in later reports. However, the pattern could also be consistent with an initial elevation bias. We present the first systematic, experimental bias, rather than a later decline. This bias is larger for reports of internal states than behaviors, and for negative mental states and physical symptoms than positive states.

show abstract

“…For over 50 years, researchers have noticed that reports of levels and severity of certain symptoms and maladaptive experience diminish over repeated assessments (1)(2)(3)(4).…”

Section: Conflicts Of Interestmentioning

confidence: 99%

Initial elevation bias in subjective reports

Shrout

Stadler

Lane

et al. 2017

Proc. Natl. Acad. Sci. U.S.A.

173

194

View full text Add to dashboard Cite

show abstract

“…The link between item location and criticalness response style uncovered in this study represents a special instance of "secular trends ll (Loevinger, 1957) on psychological tests--changes in score over time--which have been reviewed by Fiske and Rice (1955) and Windle (1954). These reviews described a number of studies that found changes between two or more administrations of the same test when no experimental treatment intervened between the two administrations, and even when one administration immediately followed the other.…”

Section: Discussionmentioning

confidence: 81%

“…Evidence already exists about social desirability response style. Windle's (1954) review indicated that more socially desirable scores on the standard content scales were generally obtained when personality inventories were administered a second time, and Gordon (1952) found that more socially desirable responses were made to the later items in a personality inventory. The role of the other response styles is less clear.…”

Section: Discussionmentioning

confidence: 99%

Difficulty and Other Correlates of Criticalness Response Style at the Item Level1

Stricker¹

1964

ETS Research Bulletin Series

View full text Add to dashboard Cite

This study examined the.relationship between the difficulty of items on an achievement-type test of report writing and (a) the items' correlations with the test's criticalness response style score, and (b) the items' correlations with the test's content score. Item readability and item format variables were also analyzed. The major findings were that (a) the items' difficulty was related to their correlations with the content score and unrelated to the items' correlations with the response style score; (b) the items' location in the test was generally related to their correlations with the response style score, and unrelated to the other variables; and (c) the readability indexes were not related to the three basic item variables. DIFFICULTY AND OTHER CORRELATES OF CRITICALNESS RESPONSE STYLE AT THE ITEM LEVEL 1Largely as a result of Cronbach's (1946; reviews of the early literature on response sets--or response styles--on tests, the view is widely held that tests and test items that are difficult or ambiguous are most affected by those response styles, such as acquiescence, evasiveness, and extremeness, that are linked to the response format.Despite the prevalence of this view, the results of relevant studies are not altogether convincing, and are largely limited to two response styles--acquiescence, and the position of the chosen alternative on multiple-choice tests.One major group of relevant studies examined acquiescence on tests or items of varying difficulty or ambiguity. Only one of the studies (Gage, Leavitt, & Stone, 1957) concerns the difficulty issue. In a comparison of difficult and easy general information items, the reliability of the number of true responses was .68 for the 50 difficult items, and .09 for the 40 easy ones. The two kinds of items, however, did not represent exactly the same content areas.The other studies of this kind concern various facets of item ambiguity. Two of these studies directly concern ambiguity. In one study (Bass, 1955), the original and reversed version of the California F scale items that were rated most opposite in meaning (and, hence, assumed to be clear in meaning) appeared to measure content (i.e., the correlation between the two items was relatively high and in the content direction), while those rated less opposite in meaning (and, hence, assumed to be -2-relatively ambiguous in meaning) did not, evidently because of the operation of acquiescence on the latter items. Moreover, an analysis of variance found significantly (£ < .01) greater acquiescence to the pairs of items that were least opposite in meaning. The assumption in this study that the rated level of opposition measures ambiguity is questionable, for it is possible for a pair of items that are rated as similar in meaning to be very upambiguous. In a second study (l3a.nta, 1961), the number of agree responses on attitude scales increased and the number of extreme responses decreased with the ambiguity of the scales' referents. The referents were "President Eisenhower," "College Fraternities,"...

show abstract

“…Social desirability does not seem to offer a reasonable alternative explanation for the reliability shifts, response polarization, or construct learning effects that we have found. However, we believe that social desirability plays a Part in the mean shifts that have been found for tests of adjustment (e.g., Windle, 1954). Social desirability is likely to influence (a) the content of the thoughts that are retrieved in response to each question, (b) the arguing and .counterarguing that occurs during and after considering a question, and (c) the ease with which questions and thoughts are assimilated into the self-concept.…”

Section: Alternative Explanationsmentioning

confidence: 90%

“…Test-Retest Mean Answer Shifts Windle (1954) compiled test-retest data from numerous personality inventories and concluded that a variety of "adjustment" scores showed significantly better adjustment on retest, especially with retest intervals of less than two months. This intriguing mean answer shift on retest continues to be observed for various tests (Chance, 1955;Goldberg, 1978;Payne, 1974;Perkins & Goldberg, 1964;Windle, 1955).…”

Section: Test-retest Reliability Shiftsmentioning

confidence: 99%

Context Effects in Social and Psychological Research

Schwarz¹,

Sudman²

1992

210

View full text Add to dashboard Cite

), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use of general descriptive names, trade names, trademarks, etc., in this publication, even if the former are not especially identified, is not to be taken as a sign that such names, as understood by the Trade Marks and Merchandise Marks Act, may accordingly be used freely by anyone.Camera-ready copy supplied by the authors. Printed and bound by Braun-Brumfield, Inc., Ann Arbor, MI. Bibb Latane and Deborah Richardson hosted the conference at the Nags Head Conference Center in Kill Devil Hills, North Carolina. Their hospitality, the joys of southern cooking, and the outer banks of North Carolina greatly contributed to an enjoyable and productive meeting. Special thanks are due to Mary A. Spaeth, who edited all contributions to this volume and turned a pile of manuscripts into a readable book. Finally, we thank all conference participants for lively and controversial discussions from breakfast to midnight and for the fine chapters that they contributed.

show abstract

Test-Retest Effect on Personality Questionnaires

Cited by 128 publications

References 80 publications

Initial elevation bias in subjective reports

Initial elevation bias in subjective reports

Difficulty and Other Correlates of Criticalness Response Style at the Item Level1

Context Effects in Social and Psychological Research

Contact Info

Product

Resources

About