Factorial Designs

ANOVA is a powerful tool for the statistical analysis that allows researcher to investigate whether some variable (it is called independent) has any influence on other variable which is called dependent. Moreover, various modifications of ANOVA allow including several independent or several dependent variables into the analysis.In order to find whether familial factor and media factor have influence on eating disorders the best option would be to utilize the independent factorial ANOVA design. The choice of factorial design over the one-way design is explained by the fact that there are two independent factors of interest. In such a case usage of one-way design would require running two independent ANOVA tests (one for each factor) which would have less statistical and explanatory power compared to factorial ANOVA. Moreover, factorial design also tests for interactions, which means that it is investigating whether the two factors of interest have joint effect on eating disorders.The nature of the research question does not allow repeated measures. Indeed, the treatment of eating disorders is beyond the scope of this research, while the factors of interest hardly vary in time. Though, there is no point in collecting data for the same respondents at several points in time and the best option is to stick to the independent groups design.As it was mentioned, two factors of interest (or independent variables) are familial and media. It is proposed that the former will contain information on whether any of respondent’s close relatives has eating disorders, while the latter will categorize the respondents in several groups based on the number of hours they spend watching TV, reading newspapers and surfing the Internet per week. Therefore, both independent variables are between participants factors. The dependent variable in this research is the indicator whether the respondent has any type of eating disorder.The research design described above would give a possibility to investigate the main effect of the two factors of interest and their joint effect. However, such ANOVA design omits other factors, which may happen to be important (such as respondent’s income) and requires independent factors to be categorical. Both these problems may be solved by utilizing multivariate regression analysis.ReferencesGravetter, F. J. & Wallnau, L. B. (2010). Essentials of Statistics for the Behavioral Sciences. Cengage Learning: Print.Triola, M. F. (2010). Essentials of Statistics. Pearson Education: …