The hypothesis was tested that subjects fake personality tests by enacting a specific social role, rather than by responding in terms of personality constructs, and that such role faking cannot be detected by validity scales. Results support the hypothesis. (Author)

A method of estimating the parameters of the Rasch Model removing the effect of ramdom guessing is presented. The procedure is an application of the ARRG (Abilities Removing Random Guessing) model recently developed for two parameter latent trait models. Under the Rasch model ARRG provides for estimation of abilities, removing the effects of…

The effects of low and high levels of test-situation stress on creativity test performance were examined. A group of 60 fifth and six graders was randomly assigned to stress situations (high, low, and control) in which verbal subtasks of the Torrance Tests of Creative Thinking (TTCT) were administered. Verbal fluency scores from the TTCT…

Answer changes on tests are more likely to be made on items early in a group of questions, toward the end of a test, and on difficult items. The reasons for this student behavior are examined. (Editor/JD)

Assumptions underlying Cooper's exact probability test for Likert-type scales are examined and found not always to hold true. A modification of his test is derived, differences in results obtained with the two tests are presented, and the significance of the results discussed. (KR)

Three personality measures were administered twice each with an interval of four weeks between administrations, and the response consistency of these tests was analyzed. The evidence is equivocal. The confounding of consistency effects with other sources of variance remains a problem. (Author/CTM)

Either positive or negative test instructions and two item types (attitude or personality) were administered to undergraduates. Students were more likely to respond true when receiving positive instructions. Students receiving attitude test instructions were not more likely to respond true than students receiving personality test instructions.…

Reasons for not returning a questionnaire were studied among Lutheran church members sent a postcard to indicate no intention of replying. The analysis is based on 431 returned postcards. For most nonresponders, the interest value of the questionnaire topic was not high enough. Many who returned the negative reply card would be resistant…

The runs test for random sequences of responding is proposed for application in long inventories with dichotomous items as an index of sterotyped responding. This index is useful for detecting whether the client shifts between response alternatives more or less frequently than would be expected by chance. (LMO)

Statistical methods are presented to facilitate a more complete analysis of results obtained when a scaling model is applied to data from two or more groups. Various kinds of scaling models are considered here in the multiple-group context. (Author/LMO)

Examined group differences in self-reporting anxiety for one hundred chronic pain patients, an equal number of college students, and two smaller comparison samples. Pain patients, relative to nonpatients, acknowledged dramatically fewer total signs of anxiety. Also, pain patients endorsed significantly more somatic than cognitive indicators of…

Discusses ways testing can help students learn if the teacher can remove the aura and function of evaluation from the experience. (GB)

The conclusion of the present experiment was that prior response increases item consistency, but only when the prior experience involves the same item with the same instructions. (Authors)

Sequential effects in 2CRT are due to the structure of the sequence of stimuli, rather than due to local response biasing resulting from the sequence of required responses. (Author)