Question 6

Now consider an appropriate kind of follow-up test (i.e., that adjusts for multiple comparisons using

the Holm correction) to compare mean confidence ratings in each pair of conditions: true-consensus

vs. no-consensus, true-consensus vs. false-consensus, and false-consensus vs. no-consensus.

Report the results of each test in one or two sentences, including the

p

-value for each comparison. You

don't need to include descriptive statistics or effect sizes here:

p

-values and their interpretation are

sufficient. Refer to Figure 1 above when discussing which group is higher/lower.[3]