Week 12 Homework
1. Imagine a dataset with a continuous outcome variable, one continuous predictor
variable, and one categorical predictor variable with two levels. Imagine you run a
regression model and look at the coefficients table, and the estimate for the categorical
predictor is positive and the estimates for the continuous predictor and the interaction
are negative. All three of those coefficients are significantly different from 0. You then
plot two best-fit regression lines, one for each level of the categorical predictor. What do
the lines look like approximately? You can draw or describe them.
2. Imagine you are trying to learn a new language, and you see that 3 language-learning
apps have done a regression analysis to perform a statistical test of the difference in
comprehension scores between learners before and after using the app (the reference
level is before using the app). Below are the relevant results (estimates and p-values)
from the three tests:
App A App B App C
Estimate p-value Estimate p-value Estimate p-value
Intercept 10.62 < .0001 *** 15.43 < .0001 *** 8.00 .00244 **
Group – afterAppUse 15.78 < .0001 *** 17.25 .00135 ** 20.24 .03255 *
Based on this, which app would you choose to use? Why?
apps have done a regression analysis to perform a statistical test of the difference in
comprehension scores between learners before and after using the app (the reference
level is before using the app). Below are the relevant results (estimates and p-values)
from the three tests:
App A App B App C
Estimate p-value Estimate p-value Estimate p-value
Intercept 10.62 < .0001 *** 15.43 < .0001 *** 8.00 .00244 **
Group – afterAppUse 15.78 < .0001 *** 17.25 .00135 ** 20.24 .03255 *
Based on this, which app would you choose to use? Why?
