Skip to content

Doing Tests of Assumptions :Designing, Running, and Analyzing Experiments(Interaction Design Specialization) Answers 2026

Q1. Number of subjects

data <- read.csv("designtime.csv")
nrow(data)

👉 Enter the total number of rows.

Q2. Boxplot & visual comparison

boxplot(Time ~ Tool, data = data,
xlab = "Tool",
ylab = "Time (minutes)",
main = "Task Time by Design Tool")

👉 Look at:

  • Which box has a higher median line

  • Whether the spread (IQR + whiskers) looks similar or different

Choose the option that best matches what you see.

Q3. Shapiro-Wilk test (Illustrator only)

shapiro.test(data$Time[data$Tool == "Illustrator"])

👉 Enter the p-value, rounded to 4 decimal places.

Q4. Shapiro-Wilk test on residuals

model <- aov(Time ~ Tool, data = data)
shapiro.test(residuals(model))

👉 Enter the W value, rounded to 4 decimal places.

Q5. Normality conclusion

Rule:

  • p < 0.05 → ❌ Violates normality

  • p ≥ 0.05 → ✅ Does NOT violate normality

Choose the option accordingly.

Q6. Brown-Forsythe test (Homoscedasticity)

library(car)
leveneTest(Time ~ Tool, data = data, center = median)

👉 Enter the F value, rounded to 2 decimal places.

Q7. Lognormal fit + KS test (Illustrator)

library(MASS)

illus <- data$Time[data$Tool == "Illustrator"]
fit <- fitdistr(illus, "lognormal")

ks.test(
illus,
"plnorm",
meanlog = fit$estimate["meanlog"],
sdlog = fit$estimate["sdlog"],
exact = TRUE
)

👉 Enter the exact p-value, rounded to 4 decimal places.

Q8. Log-transform & mean (InDesign)

data$logTime <- log(data$Time)

mean(data$logTime[data$Tool == "InDesign"])

👉 Round the result to 2 decimal places.

Q9. Welch t-test (log-transformed)

t.test(logTime ~ Tool, data = data, var.equal = FALSE)

👉 Enter the t statistic, rounded to 2 decimal places.

Q10. Exact Mann-Whitney U test

library(coin)

wilcox_test(Time ~ Tool,
data = data,
distribution = "exact")

👉 Enter the Z statistic, rounded to 4 decimal places.

🧾 Summary Table

Question No. Analysis / Test Performed Answer to Enter Key Concept
Q1 Count of observations (Total rows in data) Sample size
Q2 Boxplot comparison (Selected option) Median & variance comparison
Q3 Shapiro-Wilk test (Illustrator) p = ____ Normality of Illustrator data
Q4 Shapiro-Wilk on residuals W = ____ Model residual normality
Q5 Normality decision Do / Do not violate Assumption checking
Q6 Brown–Forsythe test F = ____ Homoscedasticity
Q7 KS test (lognormal, Illustrator) p = ____ Distribution goodness-of-fit
Q8 Mean of log(Time) – InDesign Mean = ____ Log-transformed central tendency
Q9 Welch t-test (log Time) t = ____ Difference in means
Q10 Mann–Whitney U test Z = ____ Nonparametric comparison