researcher anything statistically relevant H u m a n i t i e s

Nahil (peer 1)

Unit 7

**What is the point of an ANOVA? It is overly complicated with a dozen little steps and, in the end, all it tells you is that at least one of the sample means differs from the rest. ANOVA doesnâ€™t tell you how many of the sample means are different or the degree of that difference. Seems like ANOVA is just a time suck; why not just do a series of t-tests in the first place, and save yourself a lot of grief? Further, since ANOVA only tells us that there is a difference, but it doesn’t tell us which groups are different, what do researchers need to do to figure out where the difference lies?**

While the t-test determines whether or not two populations are statistically different from each other, the ANOVA test does pretty much the same, just with 3 or more populations instead. Though an ANOVA test is quite consuming, it is way more reliable than performing multiple t-test. As we have learnt so far, every time we conduct a t-test, there is a chance for error. For example, if a researcher sets the alpha level at 0.05, he is allowing up to a 5% chance of error in his statistical analysis. If he happens to run two t-test on his data, that then increases his chances of error to 10%. An ANOVA test controls the amount error one can make and therefore, helps to produce more reliable findings. Though the ANOVO only tells us that there is a difference and not where the difference lies, post-hoc comparison tests such as the Scheffe or Turkey are done to find those answers.

Aron, A., Coups, E. J., & Aron, E. (2013). *Statistics for psychology*. Pearson.

ANOVA (Analysis of Variance) is relevant because it can show a researcher anything statistically relevant on a seemingly unlimited combination of differences in means. T-tests are limited by the comparing of two groups or hypothesis, whereas if multiple tests are needed for comparison, ANOVA is the best used process. Also T-tests are prone to type errors where, ANOVA test do not have these issues. The downfall to an ANOVA test is that it does not tell you where the difference in the means lies. To mitigate this issue researchers can use a post hoc test to make these determinations.