Abstract
Participant selection is an important step in research on individual differences. If detecting an effect of a personality variable is predicated on the use of extreme groups, then mistakenly including participants who are not in the extremes may weaken the ability to see an effect. In this study, changes in trait worry were evaluated in 68 undergraduate students reporting low or high levels of worry. Participants completed the abbreviated Penn State Worry Questionnaire (PSWQ-A) three times: 1) at the beginning of the se- mester; 2) 3 - 13 weeks later; and 3) 1 hr later, following a psychophysiological assessment session. Test–retest reliability across the three administrations was high, but almost half of the sample no longer met the pre-defined criteria for classification as low or high worriers at the second administration. That is, scores were reliable, but not stable, across time. Instability of self-report worry was significantly greater for high worriers than for low worriers, and this effect was predicted by trait anxiety at the beginning of the semester. These findings suggest that the PSWQ-A is sensitive to factors other than trait worry, which may result in dilution of effects when participants are selected for extreme worry scores. This also sug- gests that screening participants weeks before the actual study should be supplemented by readministra- tion of the screening questionnaire, to identify participants who no longer meet criteria for inclusion.
Highlights
Self-report measures are commonly used to screen for personality traits in both clinical and research settings
The Penn State Worry Questionnaire (PSWQ)-A had strong test–retest reliability across administrations (r = .88, p < .001, from Mass Testing to pre-session, and r = .96, p < .001, from pre- to post-session), further inspection of mean shifts in worry over time indicated that 41% of the sample failed to retain their original classification as members of either the low worry group or high worry group from Mass Testing to pre-session
An independent-samples t-test between proportions indicated that a greater number of high worriers (69%) demonstrated significant drift in self-report worry during this time than low worriers (20%), t(67)
Summary
Self-report measures are commonly used to screen for personality traits in both clinical and research settings. The stability of these questionnaires is crucial for accurate assessment, as well as for determining the effectiveness of a treatment or intervention strategy. Mean shifts in self-report measures of anxiety, for example, can occur with repeated assessment even in the absence of an external variable, such that symptoms appear to improve over time (Knowles, Coker, Scott, Cook, & Neville, 1996; Windle, 1954). The use of questionnaires to recruit participants may, pose a problem for researchers interested in mechanisms underlying anxiety and its associated features, as instability can result in misclassification of participants, weakening of effect sizes, and increased error in the data (Knowles et al, 1996). The purpose of this paper is to examine the stability of repeated administrations of trait questionnaires of worry and anxiety when such measures are used to screen participants for inclusion in a study
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
