Understanding Validity :Designing, Running, and Analyzing Experiments(Interaction Design Specialization) Answers 2026

Question 1

What is experimental control?

❌ Ensuring nothing happens without the experimenter knowing
❌ Ensuring every subject experiences every condition
❌ Ensuring measures are made correctly and precisely
✅ Ensuring that systematic differences in observed responses can be attributed to systematic changes in manipulated factors
❌ None of the above

Explanation:
Experimental control is about isolating cause and effect so changes in results can be confidently linked to the independent variables.

Question 2

Which are examples of potential confounds? (Select all that apply)

❌ Every visitor was different
✅ Designers saw website A and scientists saw website B
✅ Visitors before noon saw A; after noon saw B
❌ Site A was different from site B
❌ Measuring again with new visitors

Explanation:
Confounds occur when another variable is systematically linked to the condition being tested.

Question 3

Ecological validity and experimental control cannot both be maximized.

✅ True
❌ False

Question 4

Which was not discussed as a way to handle confounds?

❌ Manipulate it
❌ Control for it
❌ Measure it
✅ Hide it
❌ All of the above

Question 5

Which is not another term for the response?

❌ Dependent variable
❌ Measure
❌ Outcome
❌ Y
✅ Factor

Question 6

Assumptions of ANOVA (Select all that apply)

❌ Reliability of residuals
✅ Normality
❌ Homoscedasticity (not the term used)
✅ Independence
✅ Homogeneity of variance

Question 7

Which was not a common data distribution reviewed?

❌ Normal
❌ Lognormal
❌ Bimodal
❌ Exponential
❌ Gamma
❌ Poisson
❌ Binomial
✅ Multinomial

Question 8

A multinomial distribution is relevant when:

✅ The response is categorical with more than two categories
❌ Response is bimodal
❌ Response is scalar
❌ Response is Poisson
❌ None

Question 9

Most precise difference between parametric vs nonparametric analyses:

❌ Operate on ranks
❌ Make assumptions about spread
✅ Make assumptions about the distribution of the response within the population
❌ Easier to use
❌ None

Question 10

Parametric analyses usually have greater statistical power.

✅ True
❌ False

Question 11

Nonparametric analyses must meet ANOVA assumptions.

❌ False
✅ True ❌➡️ Incorrect

Question 12

Nonparametric analyses typically operate on ranks.

✅ True
❌ False

🧾 SUMMARY TABLE