MPSI Mid Term Quiz

Which control technique is best when you want to make groups similar on a known participant variable (e.g., IQ)?

• Randomization

• Matching

• Holding constant

• Counterbalancing

What is the main purpose of holding a variable constant in an experiment?

• Increase external validity

• Eliminate variance caused by that variable

• Reduce sample size requirements

• Create factorial designs

Random assignment primarily helps reduce which threat?

• Instrumentation

• Individual differences as confounds

• History effects

• Low power

Which of these is a time-related threat in within-subjects designs?

• Matching

• Regression to the mean

• Maturation

• Random sampling

Counterbalancing is used to control:

• Differential attrition

• Order effects

• Instrumentation bias

• Selection bias

Which design uses different participants in each treatment condition?

• Within-subjects design

• Between-subjects design

• Mixed design

• Repeated measures design

In an independent-measures t-test, df is computed as:

• n – 1

• n1 + n2 – 2

• n1 – 1

• k – 1

What does Cohen’s d measure?

• Significance level

• Standardized mean difference (effect size)

• Sample variance

• Proportion of variance explained

When comparing three independent groups, which test is most appropriate?

• Independent t-test

• Paired t-test

• One-way ANOVA

• Chi-square

If F = t² for the comparison of only two groups, what does that imply?

• ANOVA and t test are unrelated

• The F statistic equals the square of the t statistic when comparing two means

• Use chi-square instead

• Effect size must be zero