Experiments

Definition:

“a method of testing a hypothesis through manipulating an independent variable (the variable being changed) and measuring the outcome of the changes (dependent variable)”

Experiments are a primary research methods but can use secondary techniques for data collection such as self-reports and observations.

Features of an Experiment:

• hypothesis

• IV and DV

• Design

• Controls → extraneous variables

Which inferential statistical test should be used?

Mann-Whitney:

• test for difference

• independent groups/quasi

• ordinal data

Wilcoxon:

• test for difference

• repeated measures/matched groups

• ordinal data

How to conduct a Mann-Whitney U Test:

1. Rank whole data set

2. Sum the ranks for each condition

3. Apply U₁ = R₁ - (n₁ (n₁+1))/2 and U₂ = R₂ - (n₂ (n₂+1))/2

4. The U_calc is then the smaller of U₁ or U₂

5. If U_calc is < than the U_crit then the alternate can be accepted at p<0.05

n = number of pieces of data that were present in each data set

R = rank total per condition

How to conduct a Wilcoxon Test:

1. Rank the pairs by absolute differences from smallest to largest, discounting any 0 results

2. Find the sum of the positive ranks and then the sum of the negative ranks

3. W_calc is the smaller of the rank totals

4. If W_calc is ≤ W_crit then the alternate can be accepted at p<0.05