Anxiety, repression, and denial on Byrne's repression-sensitization scale

The purpose of the study was to develop and evaluate a rationale for the repeatedly occurring empirical finding that the Byrne (1963) Repression-Sensitization (R-S) scale is moderately negatively related to indices of, or related to, psychological adjustment. The R-s scale (high scoring indicating sensitization) was postulated to essentially be a self-report measure of anxiety, and the concepts 'repression' and 'denial' were advanced as alternative defense mechanisms which would account for the apparent tendency of subjects not manifesting an extremely high degree of sensitization to have high interpersonal variability in level of psychological adjustment. Anxiety was defined as the combination of a rapidly increasing and/or relatively high level of physiological arousal and the arising to awareness of the concept 'threatening' as a response to an environmental demand. Repression (Rp) was defined as the process of automatically expunging from awareness the concept 'threatening,' with the result that the physiological arousal elicited by the perception of the environmental demand remains available for coping purposes. Denial (Dn) was considered to be a result of the inability to effectively exercise Rp, and was defined as the abandoning of a concept and/or percept of an environmental demand from conscious awareness in order to eliminate the accompanying concept 'threatening,' with the result that the physiological arousal originally stimulated by awareness of the environmental demand is left without a causative source and, consequently, is dissipated. It was postulated, therefore, that there may be two distinctively different ego defensive styles which both would tend to make it less likely that a person using them would score as an extreme sensitizer (i.e., as highly anxious), Rp, which would lessen the need to rely on rigid, less adaptive modes of ego defense, and Dn, which ignores the existence and/or importance of coping with environmental demands which otherwise would sharply increase the intensity and frequency of the experiencing of anxiety. The two research questions investigated were whether the R-S scale functions essentially similarly to anxiety as measured using the same method (i.e., a self-report questionnaire), and whether Rp and Dn are useful in accounting for variance in pattern of responding to the R-S scale? The predictions related to the first question were: (a) the r between the R-s scale and Sixteen Personality Factor Questionnaire (16 PF) anxiety (Ax) will exceed .pl, (b) in a multi-factor space factor analysis the R-S scale will load significantly only on the anxiety factor, and (c) the R-S scale will be significantly moderately negatively correlated with 16 PF extraversion (Ex), poise (Ps), and independence (Ind), and significantly moderately positively correlated with 16 PF neuroticism (without its anxiety components) (Nwa), and Rotter's measure of tendency to believe in external control of rewards (1-E). The predictions related to the second question were: (d) the R-s scale will be significantly correlated with tendency to have an R-s scale response pattern similar to each of the factor-analytically defined Rp and Dn grouping tendencies, and (ej two grouping tendencies (factors) will emerge in a factor analysis of R-S scale response patterns with one being significantly positively correlated with Ex and negatively correlated with Nwa and 1-E, and the other being significantly negatively correlated with Ex and positively correlated with Nwa and 1-E. A sample of 107 undergraduates completed the revised R-S scale, the 16 PF, and the 1-E scale. The first factor analysis was performed on 29 variables taken from the questionnaires using the responses of all 10? subjects. The second factor analysis was performed on the R-s scale response patterns (using each subject's response pattern as one variable) of the 50 subjects who satisfied the twin criteria of not being extreme sensitizers (i.e., extremely anxious) and not scoring sufficiently near to the scale's lower limit to force their response patterns to be highly similar. The respective results in terms of the five predictions were as follows: (a) R-S correlated .73 (p < .001) with Ax, (b) R-S loaded significantly (p<.05) on the anxiety (-.77) and intelligence (.23) factors out of seven factors extracted, with the former clearly showing predominance, (c) R-S correlated -.21 with Ex (p<.05), -.08 with Ps (p>.10), .02 with Nwa (p>.lO), and .40 with 1-E (p<.001), (d) R-S correlated -.34 (p<.05) with the Dn and .15 (p>.10) with the Rp grouping tendencies, and (e) the first grouping tendency to emerge correlated -.34 with Ex (p < .05), .20 with Nwa (p>.lO), and .04 with 1-E (p>.lO), and the second grouping tendency to emerge correlated .43 with Ex (p<.01), -.47 with Nwa (p<.001), and .03 with 1-E (p>.lO). In light of the pattern of the partial confirmation of the predictions it was concluded that: (a) the R-S scale is greatly similar to self-report measures of anxiety, (b) the R-S scale is probably no more than moderately related to self-report measures thought to bear some relation to adjustment, but which are not strongly related to anxiety, (c) Dn shows a negative relation to R-S, but Rp appears to be relatively independent of it, and (d) Rp and Dn appear to both be operating in R-S scale responding.