The High Cost of Complexity in Experimental Design and Data Analysis: Type I and Type II Error Rates in Multiway ANOVA

The availability of statistical software packages has led to a sharp increase in use of complex research designs and complex statistical analyses in communication research. An informal examination of studies from 2 leading communication journals suggests that the analysis of variance (ANOVA) is often the statistic of choice, and a substantial proportion of published research reports using ANOVA employ complex (k ≥ 3) factorial designs, often involving multiple dependent variables. This article reports a series of Monte Carlo simulations which demonstrate that this complexity may come at a heavier cost than many communication researchers realize. As frequently used, complex factorial ANOVA yield Type I and Type II error rates that many communication scholars would likely consider unacceptable. Consequently, quality of statistical inference in many studies is highly suspect. Communication researchers are warned about problems associated with design and statistical complexity and solutions are suggested.     

The role of expectations in children's experience of novel events

The expectations children bring to interactions, as well as the information they receive prior to them, may be important for children's experiences of new adults. In this study, 148 children (8-13years old) reported on their expectations of adults, received one of three types of information about a new adult (positive, realistic, or control), and then "interacted" with a videotaped "controlling" adult. The effect of information type depended on children's age and prior expectations, with expectancy effects emerging in the context of positive information at the younger end of our age range and in the context of realistic information at the older end of our age range. Furthermore, the more expectations exceeded perceptions (i.e., the more disappointment), the lower children's rapport, affect, and prosocial intentions were and the more internal causal attributions they made. Results are discussed in terms of their theoretical and applied contributions.     

NONCONTINGENT REINFORCEMENT: A FURTHER EXAMINATION OF SCHEDULE EFFECTS DURING TREATMENT

We conducted 2 studies to determine whether dense and thin NCR schedules exert different influences over behavior and whether these influences change as dense schedules are thinned. In Study 1, we observed that thin as well as dense NCR schedules effectively decreased problem behavior exhibited by 3 individuals. In Study 2, we compared the effects of 2 NCR schedules in multielement designs, one with and the other without an extinction (EXT) component, while both schedules were thinned. Problem behavior remained low as the NCR schedule with EXT was thinned, but either (a) did not decrease initially or (b) subsequently increased as the NCR schedule without EXT was thinned. These results suggest that dense schedules of NCR decrease behavior by altering its motivating operation but that extinction occurs as the NCR schedule is thinned. The benefits and limitations of using dense or thin NCR schedules are discussed.