Recently Ohnmacht (3) correlated scores on two hidden-figures tests, measures of the field independence (FI) construct described by Witkin, et al. (4), with fourth second-order factor (Q IV) scores from the Sixteen Personality Factor Questionnaire (16 PF). Since Q IV has been described as being . . associated with 'field independence' in the sense used by Witkin (I), Ohnmacht's (3) nonsignificant correlations led him to conclude that Q IV appeared to lack construct validity as a measure of FI. In a rejoinder (2), it was noted that Q IV is only hypothesized as being identical to source trait U.I. 19 (Independence). Ohnmacht's (3) data, at best, were interpreted as perhaps indicating that Q IV and U.I. 19 were different traits; at worst as being irrelevant to the validity of either trait. However, ic seems that, if Q IV is related to Witkin's FI construct as proposed (I), empirical verification with respect to the measures by which the Witkin, et al. (4) construct is defined is required. To evaluate f~~rther the relationship between Q IV and FI, 174 psychiatric inpatients, 61 males and 113 females, representing essentially consecutive admissions, received Series 3 (4) of the Rod-and-frame Test (RFT) and the 16 PF, Forms A and B, within 2 wk. of admission. Correlations between Q IV and RFT scores were computed by sex and were .30 (p < .0l) and .07, for females and males, respectively. The differential results by sex were consistent with earlier findings concerning the relationship berween the RFT and Embedded Figures Test (4) but are apparently not provided for in Cattell and Hundleby's (2) formulation. Although the present results show a stronger relationship berween Q 1V and the Witkin, et al. (4) FI construct than do earlier results (3), both studies clearly indicate that Q IV is not a particularly good questionnaire measure of F1 as defined by Witkin, et al. (4).
Read full abstract