According to Gore and Altman (1992), Chi-square test is used in situations where the data table has two or more rows and columns and where there is no cell with less than 5 as a value. In this case, Chi-square tests can be performed since the table obeys the two conditions. If otherwise, the Chi-square exact goodness of fit test is used since it takes care of cells with less than value 2. In case of less than two rows, then independent t-test will be used.
The ANOVA test compares the variations due to regression (SSR) and un-explained variation (SSE). Further, the analysis computes the F statistics (F computed) using the formula; F = MSE/MSR and compares it with the F tabulated value; qf(.95,1,198) = 3.89 and then compares the two values. According to Gore and Altman (1992), the regression model is significant if F (computed) > F (tabulated).
The analyzed data further assures that there is no significant difference between the two groups with F =1.929 (p=.166; p>.05). This is so because, F (computed) = 1.926 < F (tabulated), since p = .166 > .05 we reject the null hypothesis that the impact of the two groups is statistically significant at 95% level of significance. The two groups explain about 4.805 (sum of squares due to regression) of the variations in back pain improvements. The difference between participating in group discussions and group exercises is about 4.805 regression sum of square with. The un-explained variation (residual sums of squares) is about 493.995.
Statistically, the interaction between certain factors may have a negative impact on the data being modelled. Considering each factor alone is encouraged to avoid the negative effects of interactions. Adjusting for extra factors changes the outcome results. For example, in this analysis, fully adjusting the explanatory variables results into a different p-value although statistically significant. The interaction between age, race and sex adjusted and