The t-test is very handy when you have two groups to statistically compare. Analysis of Variance or ANOVA is the statistical technique that is analogous to the t-test.
The ANOVA is an inferential statistic, a parametric statistic and is very powerful.
This means that interval or ratio scales (where units are exactly the same) are used.
The statistic you obtain to determine statistical significance is the F ratio or F statistic.
This statistical technique answers the null hypothesis: There is no difference among three or more (3 ) groups on their respective mean scores.
There is one Independent variable with three or more (3 ) categories. There is one Dependent variable that is continuous in its numeric range.The descriptive statistics may be presented numerically, graphically, or both.The results of the analysis of variance should be discussed with reference to a graph of the group means. Describe the relevant outcomes and back up any claims with the results of statistical tests.It can reject the null or find differences among groups - if indeed they exist.The assumptions of homogeneity of variance, equal group sizes and normal distribution of scores should be adhered to - just as the t-test should meet these assumptions.Do not let the statistical analysis become the focus of the discussion.Instead, focus the discussion on the graph of the means and use the statistical analysis as a way to substantiate the effects you point out in the graph.Half children at each age level were presented with word stimuli; the other half were presented with pictorial stimuli.The percentage correct on the test was recorded for each child.For example, consider the following hypothetical experiment on age differences in memory for words and pictures.Sixteen 8-year-old children and 16 12-year-old children were shown a set of stimuli and later given a test to see how well they could recognize the stimuli that had been presented.