Temperament

Based on four-month temperament group status, 278 infants were invited to continue participation, and, of these, 234 participated at nine months. There was no differential attrition based on temperament group. Infants who dropped out were not significantly different from those who remained in the study in terms of negative affect, positive affect, or motor activity.
Of the 234 infants who participated at nine months, 152 (65.8 %) were Caucasian; 31 were African American (13.4 %); 14 Hispanic (6.1 %); 5 Asian (2.2 %), and the remaining 32 children were of other or mixed ethnicity. Eighty-four percent of the children (n = 187) came from intact homes and roughly half of the children (n = 111) had siblings. Mothers averaged 32 years of age (SD = 5.3) and fathers averaged 34 years (SD = 6.1) at the time of the child’s birth.

During home visits, child temperament was assessed using the Lab-TAB–Home Version, a comprehensive home-based temperament assessment that includes behavioral episodes corresponding to dimensions of temperament. This administration of the Lab-TAB included 12 standardized behavioral episodes intended to elicit targeted affective and behavioral reactions. These episodes comprised a revised version of the Preschool Lab-TAB (Goldsmith et al., 1993 ). Preschool Lab-TAB data has previously shown convergent validity with temperament questionnaires (i.e., the Children’s Behavior Questionnaire) as well as continuity across age in several studies (e.g., Pfeifer et al., 2002 (link)). The administration and coding of this home-based version was almost identical to the lab versions. The only significant differences were that the child was in the home, the camera was present as opposed to being in a control room, and the props that the experimenter used were in a bag in the room. Although there is always variance in behavioral observation from place-to-place (including from laboratory to laboratory), our goal was to make the administration as consistent as possible across home administrations. The Lab-TAB assessments typically lasted about 40 minutes and occurred at the midpoint of a two-hour home visit that included other activities such as maternal interviews and play sessions with a sibling. One child tester administered the battery after establishing appropriate rapport with the child. Eight individuals served as child testers for the 4.5 year visits, and each was highly trained and monitored to achieve consistency of administration. During administration of the Lab-TAB episodes, children’s behavior was videotaped and later coded in the laboratory. Thirteen percent of the sample was rated by a second observer and every Lab-TAB episode had a mean Kappa of .90 or higher reflecting chance-corrected inter-rater agreement.
Table 1 lists the 12 Lab-TAB episodes, beginning with a descriptive title that will be used throughout the paper (Bookmark, Box Empty, Dinky Toys, End of the Line, Perpetual Motion, Popping Bubbles, Pop-up Snakes, Snack Delay, Spider, Stranger Approach, Transparent Box, Workbench). Brief descriptions of the episodes as well as the broad domains of temperamental reactions that each episode was intended to elicit are also noted. Within each episode epoch or trial, a number of child responses are coded. Lab-TAB coding involves multiple domains of responses including facial, vocalic, motoric, behavioral and postural modalities (e.g., smiling, reaching, crying, touching, or changes in facial expression). Sometimes the presence or absence of a response is simply noted; however, more often parameters of the response, such as latency, duration, and intensity, are timed or rated. Expressive (e.g., facial and vocal) measures and instrumental or motoric measures often fall into different clusters and can be classified as different episode component scores. For example, the Box Empty episode yields anger, sadness, and approach scores. If episode component scores are intercorrelated, an overall episode summary score is often justified. However, we used many episode-level component scores in the present analyses. The actual process of scoring and scale construction for the Lab-TAB is a key element of our results and will be described in the next section.

Just prior to the infant’s reunion with its mother, the technician who performed the testing rated the overall temperament of each animal during the 25-h test period. A list of 16 adjectives describing affect quality (Table 3) were rated on a Likert-like scale of 1 to 7, with 1 reflecting a total absence of the behavior and 7 reflecting an extremely large amount of the behavior. As with the Holding Cage observations (above), ratings on each adjective were z-scored across all subjects tested within a given birth year, and exploratory and confirmatory factor analyses of the temperament ratings for 1284 infants were conducted by JPC. All procedures were identical (e.g., exploratory factor analysis on one sample, examination of fit criteria, confirmatory factor analysis with other samples, examination of scale reliability), with the exception that maximum likelihood was the method of estimation for these continuously distributed variables. A model with four factors fit the data very well in the exploratory (RMSEA = 0.055) and in the confirmatory models (CFI and TLI ranged from 0.914 to 0.952; RMSEA ranged from 0.078 to 0.084; the standardized root mean square residual (SRMR) ranged from 0.048 to 0.059 [for SRMR, values < .08 indicate a good fit: Muthén and Muthén, 2001 ]). Factor scores were calculated by summing the z-scores for all adjective items loading on a given factor, and then z-scoring each scale. Cronbach’s alpha values for the scales across the five years ranged from 0.6 to 0.9. The four scales, named for the adjective with the highest factor loading, were: Vigilant (vigilant, NOT depressed, NOT tense, NOT timid), Gentle (gentle, calm, flexible, curious), Confident (confident, bold, active, curious, playful), Nervous (nervous, fearful, timid, NOT calm, NOT confident).