ReviewPersonality in childhood – A longitudinal behavior genetic approach
Highlights
► We examined the structure and etiology of personality in middle childhood. ► We used a sample of MZ and DZ twins. ► Phenotypic results provided evidence for a Five Factor structure. ► Longitudinal results indicated that genetic factors mainly contributed to stability. ► Environmental effects primarily accounted for change.
Introduction
A broad range of studies on personality in adulthood has identified individual differences in the five domains of the Five Factor Model (FFM): Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness. Phenotypic (rank-order) stability and (mean-level) change as well as the factor structure are well researched (Roberts and DelVecchio, 2000, Caspi et al., 2005, Soto et al., 2008). In their meta-analysis, Roberts and DelVecchio (2000) reported increasing stabilities from childhood to early adulthood, and into late adulthood (.31, .54, and .74, respectively). Still, there is no clear taxonomy in early and middle childhood mainly because only a few studies have focused on the interface between the temperamental and personality research traditions and tried to include a common framework of the two research lines. The concept of temperament is widely used to describe individual differences in childhood. Nevertheless, there is a common ground of shared traits between temperament and personality, and it is argued that the two concepts have more in common than differences (see Caspi et al., 2005, Mervielde et al., 2005). Moreover, Deal, Halverson, Havill, and Martin (2005) provided an overview on the longitudinal interrelation between temperament and personality. Based on the idea that adult personality may overlap with childhood personality to a substantial amount, there is initial evidence that the FFM of personality can also serve as a reasonable classification for the structure of personality in childhood (see Mervielde et al., 2005, Caspi et al., 2005).
Studies have used different methods and informants to explore childhood personality (see Soto et al., 2008). Teacher and parent reports were often used for younger children (De Pauw, 2010, Shiner and Caspi, 2003). In their review, Roberts and DelVecchio (2000) showed that of the studies that were carried out with participants from 0 to 12 years of age, less than half of them were classifiable within the FFM. Moreover, only five of them used self-report questionnaires, in part because very young children were not yet able to answer self-report questionnaires. Only one study by Measelle and colleagues investigated self-reported personality in middle childhood. Instead of questionnaires, however, a puppet interview was used. This research yielded a reliable five-factor structure (Measelle, John, Ablow, Cowan, & Cowan, 2005). Overall, parent and teacher reports merely represent an external perspective. Moreover, teachers are only able to rate school-related behavior which may not reflect the entire childhood personality. At the same time, there is evidence that children around that age are able to provide differentiated descriptions of their behavior (Roberts & DelVecchio, 2000). However, there is a discrepancy between the existence of the Five Factor structure in childhood and the availability of questionnaire measures to describe those individual differences in young and middle-aged children. To bridge the gap between childhood and adolescence/adulthood personality structure, a (longitudinal) study of the Five Factor structure based on self-reports is needed.
Another important source of information for a better understanding of the underlying mechanisms and structure of childhood personality comes from a behavior genetic perspective. Studies conducted with adult samples have consistently shown that genetic factors substantially influence interindividual differences in personality (Johnson, Vernon, & Feiler, 2008). Moreover, results from different studies across the lifespan suggest that the same genetic effects influence personality at different points in time (Blonigen et al., 2006, McGue et al., 1993, Pedersen and Reynolds, 1998).
Moreover, most behavior genetic studies covering childhood and early adolescence have focused on the decomposition of individual differences in temperament (Mullineaux et al., 2009, Gagne et al., 2009). Mullineaux et al. (2009) found moderate to substantial (additive) genetic effects and modest effects of shared environment on all temperament traits (effortful control, surgency, and negative affect) and modest effects of shared environment on most traits. Especially surgency showed non-additive genetic effects, which may be due to the fact that the data were based on parent ratings (Mullineaux et al., 2009). A detailed qualitative review on behavior genetic studies on temperament was given by Gagne et al. (2009).
Furthermore, few studies have investigated the etiology of personality traits in adolescence which are similar to patterns in (early) adulthood. Blonigen and others examined the genetic and environmental contributions to personality stability and change in early adulthood. They suggested that stability in the age cohort of 17–24 year olds was mainly due to genetic factors, whereas change was influenced by both genetic and environmental influences (Blonigen, Carlson, Hicks, Krueger, & Iacono, 2008). Bratko and Butkovic (2007) examined a longitudinal sample of adolescents and found similar results.
The purpose of this study was to fill a gap in the literature with a longitudinal behavior genetic study of self-reported personality in childhood based on the FFM. Regarding childhood personality as an early manifestation of adult personality, we expected a similar genetic and environmental pattern as in adulthood which is characterized by roughly equal genetic and nonshared environmental influences.
Section snippets
Sample
Participants were monozygotic (MZ) and dizygotic (DZ) twin pairs from the ongoing German twin study CoSMoS (Cognitive Ability and Self-Perceived Motivation of School Achievement; Spinath & Wolf, 2006). Initial contact with the families was made based on registration data in two German Federal States (Thuringia and North Rhine-Westphalia; see Spinath and Wolf (2006) for details of the sampling procedure and participation rates). The present study included n = 202 twin pairs who took part in the
Phenotypic analyses
Means, standard deviations, factor loadings, Cronbach‘s alpha, and (latent) intercorrelations of the five domains at both measurement occasions are presented in Table 1. We first performed a confirmatory factor analysis (CFA) to test the factorial validity of the FFM. We used item parceling which was based on the item-to-construct balance technique (Little, Cunningham, Shahar, & Widaman, 2002). For each domain we used 3 parcels with three items each. The same item parceling was used at both
Discussion
Many studies have shown that individual differences in personality across the life-span are stable (Caspi et al., 2005). Because those studies mainly focused on adulthood, we intended to examine the usefulness of the Five Factor framework in childhood. The findings from our study suggest that the Five Factor structure can be assessed in childhood with reasonable reliability. The differentiation among the factors was not yet as pronounced as in adulthood as demonstrated by higher
References (30)
- et al.
Temperament in middle childhood: A behavioral genetic analysis of fathers’ and mothers’ reports
Journal of Research in Personality
(2009) Factor analysis and AIC
Psychometrika
(1987)- et al.
Stability and change in personality traits from late adolescence to early adulthood: A longitudinal twin study
Journal of Personality
(2008) - et al.
Continuity and change in psychopathic traits as measured via normal-range personality: A longitudinal–biometric study
Journal of Abnormal Psychology
(2006) - et al.
Stability of genetic and environmental effects from adolescence to young adulthood: Results of Croatian longitudinal twin study of personality
Twin Research and Human Genetics
(2007) - et al.
Personality development: Stability and change
Annual Review Psychology
(2005) - Deal, J. E., Halverson, C. F., Havill, V. L., & Martin, R. P. (2005). Temperament factors as longitudinal predictors of...
- De Pauw, S. S. W., & Mervielde, I. (2010). Temperament, personality and developmental psychopathology: A review based...
Higher-order factors of the Big Five
Journal of Personality and Social Psychology
(1997)- et al.
The genetics of childhood temperament