AN EXAMINATION OF THE CONSTRUCT VALIDITY OF EGO INTEGRITY IN LATE ADULTHOOD
This study was a preliminary investigation into the construct validity of Erikson's ego integrity in late adulthood. Empirical definition of ego integrity was approached through correlational assessment of convergent and discriminant validity and through estimation of the developmental progression of the salience of Integrity dimensions in middle and late adulthood. Developmental change was investigated through assessment of self-perceived change in levels of hypothesized Integrity dimensions compared to self-perceived levels of hypothesized Nonintegrity (nonchange-oriented) dimensions among middle-aged and old-aged participants.^ Regarding correlational data, all predictions of convergent and discriminant validity were supported in general by data from the total sample and various sample subsets. Concerning predictions of convergent validity, the hypothesized Integrity variables of Acceptance, Meaning, and Death Anxiety tended to interrelate significantly as hypothesized. Concerning discriminant relationships, again, as predicted, the Integrity variables of Acceptance, Meaning, and Death Anxiety tended to not relate significantly to the hypothesized Nonintegrity dimensions of Order, Abasement, and Social Adroitness. Thus, correlational results of this study contributed in part to the empirical definition and construct validation of ego integrity.^ In relation to the multivariate analysis of variance procedure used to estimate developmental change, the prediction of self-perceived change in the importance of proposed Integrity dimensions was not supported. Specifically, while there were some sex differences found, the proposed Integrity indices of Acceptance, Meaning, and Death Anxiety were not found to be more salient, i.e., were not manifested in a more positive direction for the target age or temporal focus of old age than for the target age or temporal focus of middle age. No self-perceived developmental change was predicted in the salience of the proposed Nonintegrity dimensions of Order, Abasement, and none was found.^ Overall, the data on estimating convergent and discriminant validity and self-perceived change of hypothesized Integrity variables offered only partial support for the validity of the ego integrity construct in late adulthood, as operationalized in this study. Future research was suggested to focus on overcoming possible methodological and conceptual limitations of the area. ^
DEMPSEY, SHEILA S, "AN EXAMINATION OF THE CONSTRUCT VALIDITY OF EGO INTEGRITY IN LATE ADULTHOOD" (1987). ETD Collection for Fordham University. AAI8727843.