Authors
Background
Establishing a firm identity is one of the key developmental tasks in adolescence (Erikson, 1968). It has been increasingly suggested to study identity processes on the short-term, in order to advance our knowledge on how short-term identity formation unfolds in adolescence (Lichtwarck-Aschoff et al., 2008). Following this suggestion, studies have started using daily diary methods to study identity development in adolescence. However, for valid comparisons of identity scores over time or across sex, it is important to assess whether these daily reports assess the same construct over time as well as across boys and girls. That is, such daily diary reports must show measurement invariance, implying that they measure an identical construct across time and groups (van de Schoot, Lugtig, & Hox, 2012). Therefore, the aim of the current study was to assess measurement invariance of adolescents’ daily identity reports across time (i.e., from early to late adolescence) and across sex.
Methods
Dutch adolescents (N = 494; Mage = 13.32 years at T1, 56.7% male) from the general population reported on their identity commitments, exploration in depth, and reconsideration on a daily basis for a week every four months, across 5 successive years. We used the single-item version of the Utrecht Management of Identity Commitments Scale (U-MICS; Klimstra et al., 2010), a measure of identity formation processes covering both the interpersonal and educational identity domains. Using confirmatory factor analysis (CFA), this study tested configural invariance (i.e., equal pattern of factors across time and sex), metric invariance (i.e., equal factor loadings across time and sex), scalar invariance (i.e., equal item intercepts across time and sex) and strict invariance (i.e., equal item residual variances across time and sex). These tests for measurement invariance were conducted across days within weeks, across sex, across weeks within years, and across years.
Results & Conclusion
Findings based on the CFAs indicated that daily identity reports showed configural, metric, scalar and strict measurement invariance (1) across days within weeks; (2) across days within weeks for boys and girls; (3) across weeks within years; (4) and across 5 years in both the educational and interpersonal identity domains. Our findings that daily diary methods consistently show measurement invariance across various time intervals (i.e., from days to years) and across sex suggests that daily diary reports of identity formation processes can be used to measure identity across adolescence as well as across boys and girls.