Class 10 — One Simple Test

Quiz 2 opens the class, then you finally run a real statistical test — t.test() on both your real and simulated data — and see last class’s sampling-variability intuition become a p-value.

PART III · CLASS 10

You’ve simulated what sampling variability looks like. Today you run the actual test that turns “looks different” into a number you can report. Quiz 2 (covering classes 6–9) opens the class. You’ll then run a guided t.test() on instructor sample data, interpret what its output actually means, run it again on both your own real HW4 data and your simulated HW5 data, and get a brief look at afex::aov_ez() as an optional ANOVA-framework alternative.

Reading due: — · Due today: HW5 due · Quiz 2

What you’ll be able to do

1. Recall and apply RStudio/tidyverse fundamentals, dplyr wrangling, ggplot2 visualization, and simulation concepts from Classes 6–9.
2. Run a two-sample t.test() in R for a 1-factor/2-level design and correctly read its output (test statistic, degrees of freedom, p-value, confidence interval).
3. Interpret what a p-value, confidence interval, and effect size actually mean for your own design and data — not just recite the statistical definition.
4. Run your own real HW4 data and your own simulated HW5 data through t.test() and describe how the two results compare.
5. Recognize afex::aov_ez() as an ANOVA-framework alternative to t.test() for the same 2-level design, and correctly set its id argument.