Sample size determination for health psychology interventions with binomially distributed outcomes.

Health intervention outcomes are often assessed as binomially distributed variables. In designing such interventions it is important to model the pre-intervention rate of the target behavior when performing sample size calculations. Unfortunately, the majority of sample size programs model post-intervention outcomes only, which results in exaggerated sample size estimates. An exception is Yoo and Spoth's (1993) conditional binomial method of sample size determination. This approach explicitly models pre-intervention behavior by focusing on baserate-adjusted post-intervention outcomes, and always results in smaller sample size estimates than conventional approaches. Advantages of the conditional binomial method are discussed and user-friendly software is presented.

Development and initial psychometric evaluation of the sport interference checklist.

The Sport Interference Checklist (SIC) was developed in 141 athletes to assist in the concurrent assessment of cognitive and behavioral problems experienced by athletes in both training (Problems in Sports Training Scale, PSTS) and competition (Problems in Sports Competition Scale, PSCS). An additional scale (Desire for Sport Psychology Scale, DSPS) was developed to assess the degree to which athletes desire sport psychology assistance in areas that are determined to be problematic. Factor analysis of PSCS items reveals six factors (Dysfunctional Thoughts and Stress, Academic and Adjustment Problems, Injury Concerns, Lack of Motivation, Overly Confident/Critical, Pain Intolerance), accounting for 64% of the total variance. PSTS and DSPS items yield four factors (Dysfunctional Thoughts and Stress, Academic Problems, Injury Concerns, Poor Team Relationships), accounting for 59% and 63% of total variance, respectively. Scores from these scales demonstrate acceptable internal consistency and convergent and discriminative validity. Response patterns of SIC scales are not influenced by gender or athlete type.

Psychometric evaluation of self- and collateral timeline follow-back reports of drug and alcohol use in a sample of drug-abusing and conduct-disordered adolescents and their parents.

One hundred eighty-eight drug-abusing and conduct-disordered adolescents and their parents provided retrospective reports of the youths' frequency of alcohol and illicit drug use for each of the 6 months preceding their initial session in an outpatient treatment program. Youths' and parent reports of youths' drug and alcohol use for each month were similar. For marijuana and alcohol, frequency reports were related for each month, whereas hard drug reports were related only for the 4 months preceding intake. Relationships among alcohol and hard drug reports were strong during Months 1, 2, and 4 preceding intake, whereas strength of relationship among marijuana reports was similar throughout the 6 months.

A Monte Carlo evaluation of tests for comparing dependent correlations.

The authors conducted a Monte Carlo simulation of 8 statistical tests for comparing dependent zero-order correlations. In particular, they evaluated the Type I error rates and power of a number of test statistics for sample sizes (Ns) of 20, 50, 100, and 300 under 3 different population distributions (normal, uniform, and exponential). For the Type I error rate analyses, the authors evaluated 3 different magnitudes of the predictor-criterion correlations (rho(y,x1) = rho(y,x2) = .1, .4, and .7). For the power analyses, they examined 3 different effect sizes or magnitudes of discrepancy between rho(y,x1) and rho(y,x2) (values of .1, .3, and .6). They conducted all of the simulations at 3 different levels of predictor intercorrelation (rho(x1,x2) = .1, .3, and .6). The results indicated that both Type I error rate and power depend not only on sample size and population distribution, but also on (a) the predictor intercorrelation and (b) the effect size (for power) or the magnitude of the predictor-criterion correlations (for Type I error rate). When the authors considered Type I error rate and power simultaneously, the findings suggested that O. J. Dunn and V. A. Clark's (1969) z and E. J. Williams's (1959) t have the best overall statistical properties. The findings extend and refine previous simulation research and as such, should have greater utility for applied researchers.