One-Within, One-Between ANOVA

Statistics Solutions provides a data analysis plan template for the One-Within, One-Between ANOVA analysis.  You can use this template to develop the data analysis section of your dissertation or research proposal.

The template includes research questions stated in statistical language, analysis justification and assumptions of the analysis.  Simply edit the blue text to reflect your research information and you will have the data analysis plan for your dissertation or research proposal.

Data Analysis Plan: One-Within, One-Between ANOVA

Copy and paste the following into a word document to use as your data analysis plan template.

 

Research Question:

Do pretest and posttest scores differ by independent variable (group 1 vs. group 2)?

Ho: Pretest and Posttest scores do not differ by group (independent variable (group 1 vs. group 2).

Ha: Pretest and Posttest scores differ by independent variable (group 1 vs. group 2).

Data Analysis

To examine the research question, a one-within, one-between analysis of variance (ANOVA) will be conducted to assess if mean differences exist.  The one-within, one-between ANOVA is an appropriate statistical analysis when the purpose of research is to compare two or more discrete groups on a continuous dependent variable that is measured more than once (Tabachnick & Fidell, 2006).   For this research, the continuous dependent variables are pretest and posttest; the independent variable has the following groups (group 1 vs. group 2 vs. group 3…). The mixed model ANOVA uses the F test, which researchers to make the overall comparison on whether group means differ. If the obtained F is larger than the critical F, the null hypothesis is rejected. The results of the mixed model ANOVA will present findings for the main effect and will evaluate the overall differences by time (within-subjects) and also separately by group (between-subjects). The interaction of group and time will evaluate if differences exist among group and time simultaneously (Pagano, 2010).  The assumptions of normality and homogeneity of variance/covariance matrices will be assessed.  The dependent variable should be approximately normally distributed for each level of the independent variable.  This will be assessed using the Kolmogorov-Smirnov statistic and the Normal Q-Q Plot (Pallant, 2007).  Homogeneity of variance assumes that both groups have equal error variances and the Levene’s test will be used to assess this assumption.  The Mauchley’s Test of Sphericity will be examined to assess the equality of variances and covariances for each level of the within-subject variable (Leech, Barrett, & Morgan, 2008).

Reference

Statistics Solutions. (2016). Data analysis plan: One-Within, One-Between ANOVA [WWW Document]. Retrieved from http://www.statisticssolutions.com/data-analysis-plan-one-within-one-between-anova/