Researchers have shown that the five major dimensions of personality (extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience) and two additional factors (irritability and positive activity) are evident from adolescence. This study attempted to replicate and extend these results in a longitudinal study of 102 Swedish children, followed from 2.3 to 15.2 years of age. Item analyses revealed consistently reliable irritability, conscientiousness, and positive activity factors, whereas the internal reliability of the extraversion, agreeableness, neuroticism, and openness to experience factors increased over time. Irritability and positive activity were not independent of the other factors. Scores on most of the personality factors were fairly stable over time. Over time, children became less extraverted, more agreeable, and more conscientious. Neuroticism and openness to experience increased in Phase III, although openness then decreased in Phase V. Validity of the original factors was demonstrated by correlations with independent assessments of the children's cognitive performance and adjustment to school.

No abstract

Two sets of quality measures of group care were used to assess their predictive power for two sets of measures of the development of infant and toddlers in group day care. One of the quality measures we investigated was the Early Childhood Environment Rating Scale (ECERS). We replicated the findings of Scarr, Eisenberg, & DealerDeckard (1994) which were that the total score of the ECERS represents a global index and that the 37 items making up the scale are redundant and could be shortened considerably without loss of the scale as a quality criterion of group care for young children. Neither Scarr, Eisenberg, and Dealer-Deckard (1994) nor our own Factor Analyses of the 37 items supported the a priori distinction of seven Subscales. However further findings indicate that regardless of the redundancy within ECERS, two Subscales, dealing mainly with adultchild, child-child and adult-adult interactions, predict the development of infants and toddlers, but only when the measures of development were based on participant observations of the children over a longer period oftime and in the broad context ofthe natural daycare environment. None of the Subscales, nor the total ECERS predicted social development when it was based on precise time sampling observations, assessed by non participant observers, in interactions between a child and a specific caregiver. Such measures ofdevelopment were well predicted in our study by caregiver behaviors assessed through Time Sampling Observations.

In this longitudinal study, 122 Swedish children were followed from an initial assessment at 16 months until they were 8.5 years of age. Parental ratings of field independence, ego-resilience, and ego-control were obtained using the California Child Q-set when the children averaged 28, 40, 80, and 101 months of age. These assessments of personality were then examined in the context of contrasting child care experiences. There was virtually no difference between the developmental trajectories of the children in home care and in centre-based day care. In contrast, ego-undercontrol decreased less, whereas ego-resilience and field independence increased less, in children who attended family day care than in the children in either centre care or exclusive parental care. The quality of home and out-of-home care, as well as socioeconomic status and family background, did not moderate or qualify these effects of the type of care experienced. The results are discussed in relation to previous work on the effects of early out-of-home care on child development.

One ofthe most important goals ofchild care research has been to determine whether or not nonparental care has adverse effec ts on child development. Answering this question involves making causal attributions about the origins of differences between groups of children who have experienced divergent nonparental care arrangements. Some of the pr oblems researchers f ace when trying to demonstrate causal relationships are illustrated in this paper using data from the Goteborg Child Care Study. a comprehensive. pro spective. longitudinal study comparing the developmental trajectories ofchildren in exclusive home care. family daycare, and center daycare. It is argued that the discovery of significant differences between groups must initiate extensive efforts to validate and interpret the findings . Focusing on the specific characteristics of prospective longitudi nal studies. the article shows what measures can be taken to avoid the misinterpretation ofgroup differ ences that are actually attributable to pre-selection effects or confo und ing vari ables. By rig or ously id entif ying suc h confo unds. researchers can accumulate support fo r hypothesized causal relationships, but they cannot 'prove ' them because unknown and unmeasured conf ounding variables may always have impor tant effects.