2022 Introduction to Statistics in Research Mitchell 2nd ed

I N T R O T O R E S E A R C H : D A T A V I S U A L I Z A T I O N & C O M M O N S T A T T E S T S

This table is similar to Table 16.3 in the Adams & McGuire textbook. Starting with Chapter 9, the Adams & McGuire textbook includes the statistical analysis approach, when to use it and discusses the effect size and post hoc (if any) that should be ran. According to Adams & McGruie (2022), effect size i s the “strength or magnitude of the effect of a variable or strength of the relationship between two variables” (p. 630). Effect si ze is discussed in detail because researchers don’t always agree on the effect size statistic that should be ran. A definition for a post hoc test is “Additional analysis when you find statistically significant results when comparing three or more groups (sometimes also performed when you find a statistically significant interaction ” (Ibid, p. 635).

When to use?

Dependent (outcome variable)

Independent (Explanatory variable)

Parametric Test (data is normally distributed)

Non-Parametric test

The average of TWO INDEPENDENT groups

Scale

Nominal (binary)

Independent t-test Mann-Whitney U

r 2 or Cohen’s d

Effect size :

Rank sum test

Scale

Nominal

One-way ANOVA

Kruskal-Wallis Test Use a form of η2

The averages of 3+ independent groups

Effect size : η2

Run Post hoc? Yes, if F is statistically

Yes, is H is statically

significant, use Fisher’s LSD

significant- run Rank Sums test

The average difference between paired (matched) i.e. weight before and after diet

Scale

Time/ condition variable

Paired t-test (also called dependent samples t test)

Wilcoxon signed rank test

Effect size:

Cohen’s d

η2

Run Post hoc?

Dunn’s test (McNemar)

Scale

Time / condition variable

Repeated measures ANOVA (one way within subjects ANOVA)

Friedman chi- squared test

The 3+ measurements on the same subject

Effect size: η2

Use a form of η2 Yes, if X 2 is statistically significant, use Nemenyi’s procedure (this is similar to Tukey’s HSD)

Post Hoc? Yes, if F is statistically

significant, use Fisher’s LSD, Tukey or Sidak

116