Statistical Tests and Experiments

These flashcards cover key concepts from the lecture on statistical tests, specifically focusing on the Wilcoxon and Mann-Whitney tests, data types, and hypothesis formulation.

What is the purpose of a test of difference in experiments?

To determine if one set of data is significantly different from another set of data.

What type of design uses related data from the same person?

Repeated measures design.

Which design involves pairs of matched data?

Matched pairs design.

What does independent groups design imply about the data sets?

The data sets are unrelated and come from separate groups.

What test is used for related samples?

Wilcoxon test.

Which test is appropriate for unrelated (independent) samples?

Mann-Whitney test.

What was Zajonc's (1968) primary hypothesis about familiar faces?

People rate the more frequently seen face as more likeable than the less frequently seen face.

What are the two types of hypotheses stated in Zajonc's study?

Alternative hypothesis and Null hypothesis.

How are the scores from Zajonc's study categorized?

They are categorized as ordinal data.

What should you do if the differences in scores are zero when ranking?

Omit it from the ranking and reduce N accordingly.

In Zajonc's study, how many different men were shown to participants?

12 different men.

What does T represent in the Wilcoxon test calculation?

The figure that is the smallest when the ranks are totalled.

What is the significance level typically used for a one-tailed Wilcoxon test?

0.05 or 5%.

What must the calculated value of T be in order to show significance?

It must be equal to or less than the critical value of T.

What was the outcome of Zajonc's study regarding the hypothesis?

The result was not significant, so the null hypothesis was accepted.

What type of test was suggested by Zajonc's directional hypothesis?

One-tailed test.

What happens when there are tied ranks in the Wilcoxon test?

Calculate the final rank by finding the mean of the ranks that would have been given.

What does N represent in the context of statistical tests?

The total number of scores, ignoring any zero values.

If the calculated value of T is greater than the critical value, what should be concluded?

Accept the null hypothesis.

How is the ranking of differences performed in the Wilcoxon test?

Rank differences from low to high, ignoring the signs.

What did Zajonc et al. conclude based on the significance of their findings?

There is no difference in the likeability score for more and less familiar faces.