“…Because our own previous work and that of others has indicated that a measure of cumulative social-environmental risk is a more powerful predictor of child functioning than individual measures of such risk, we computed a social-environmental risk score. Following from the work of Sameroff and colleagues (Sameroff, Seifer, Barocas, Zax, & Greenspan, 1987), we did this by dichotomizing and tallying eight social-environmental risk variables: maternal education, maternal IQ, family socioeconomic status, maternal adaptive functioning, psychosocial stressors, and maternal behavior in interaction with the infants at 4, 12, and 24 months (see Bernstein & Hans, 1994, for further details of risk score calculation). A number of the variables that went into this risk score were assessed during prenatal interviews as follows: maternal education (measured in terms of years of school completed, less than 11 years counted as a risk), family socioeconomic status (measured on the Hollingshead TwoFactor Index of Social Position; Hollingshead & Redlich, 1958; Level 5 counted as a risk), maternal adaptive functioning (following psychiatric interviewing rated on DSM-III Axis 5, American Psychiatric Association, 1987, with poor or worse counted as a risk), maternal stressors (following psychiatric interviewing scaled on DSM-III Axis 4; more than moderate psychosocial stressors counted as a risk), and maternal intelligence (measured by the full-scale score from the Wechsler Adult Intelligence Scale; Wechsler, 1955; with IQ less than 85 counted as a risk).…”