Ch 8 Part II- Statistics: In Search of Answers

What tasks do we need to accomplish during our data analysis?

1. make sure our groups were similar on variables that may provide alternate explanations

2. make sure our manipulation was effective

3. test our hypothesis (do the two groups differ on the outcome?")

What are the key features of our design?

between subjects design (two levels of IV)

What statistic should we use?

t test of independent means (why needed?)

What does an independent samples t test consist of?

• compare means of two groups (if two groups/conditions are indeed different)

• a statistical test that examines the difference between two independent means against the backdrop of the within-group variability (the noise)

• the larger the difference between the means and/or the smaller the variability, the greater the value of t will be

• the larger the t value, the more likely the difference between the mean of the two groups will be considered to be “statistically significant”

What is the statistical significance of t values?

computing a t test only provides a t score, which, by itself, does nothing to clarify whether the test is statistically significant

What is a note on p-values?

• p-value or level of significance tells us the maximum probability that the results could have been obtained by chance

• a p value of .05 means that the test demands that there is less than a 5% (.05) probability that the results are caused by chance

What is an effect size?

• quantifies the magnitude of a relationship between variables or the difference between groups

• while statistical significance tells us if an effect exists, effect size tells us how meaningful that effect is in the real world

• think of it as the practical significance of a research finding