Variance among Interviewers in Data for the Peabody Picture Vocabulary Test–IIIA

Abstract

Variance among interviewers in prior research reflects that answers provided by respondents and recorded on questionnaires varied depending on which interviewer was assigned to the respondent. For scores from the 36-mo. In-Home Longitudinal Study of Pre-School Aged Children, a special module in the Fragile Families and Child Wellbeing Study, variance among interviewers was evident for the Peabody Picture Vocabulary Test-IIIA and clearly affected inferences made from test data. When variance among interviewers was ignored, mothers' perception of neighborhood closeness was significantly related to childrens' standardized PPVT-IIIA scores; however, when variance among interviewers was accounted for statistically, this effect was not significant.

Gauging the impact of literacy and educational background on receptive vocabulary test scores

“…The tests were administered by the first author in the participants' classroom during class hours in the presence of their tutor (see also Pichette et al, 2019;Puimège & Peters, 2019). To avoid construct-irrelevant variance that may result from changes in the pronunciation of the items during live reading, all prompts were pre-recorded (see also Traxel & Zhang, 2008). There was a 13-second time window for each item (Dunn et al, 2005), resulting in a total administration time of 23 minutes for the nine sets, excluding introduction and practice.…”