T-Tests Explained With Mean Girls

Hypothesis testing explained through a 21st century classic.

T-tests are a statistical tool that allow for the comparison of two groups on a particular attribute.

But first we need to decide on a research question that compares two groups: Are the Plastics more evil than the threesome of Cady, Janis, and Damien?

First, we start by formulating the Null Hypothesis (that there is no difference between the two groups). The Plastics are not more evil than Cady, Janis, and Damien.

Then, we formulate our Experimental Hypothesis (what relationship you believe exists). The Plastics are more evil than Cady, Janis, and Damien.

Now that we have our hypotheses, we can start the 5 steps of completing a t-test!

Step 1. Get the means for each group. The average evilness of Cady, Janis, and Damien vs. the average evilness of the Plastics.

Step 2. Determine the Standard Error of the difference between the means (how much of the difference is due to chance, not actual group differences). Was Cady giving Regina the Kalteen bar to make her gain weight due to evilness or just a coincidence?

Step 3. Calculate your t statistic! (Don't worry we won't go into all the math here)

Step 4. Find critical t score (threshold where you can say there is a difference between the groups). At what point, exactly, are the Plastics deemed more evil?

Step 5. Compare your t-statistic to the critical t score and decide whether you reject or fail to reject your null hypothesis! The crucial moment that decides whether the Plastics truly are more evil.

But, we're not done yet!

If your t statistic > critical t score, then you reject the null hypothesis, that means there was a difference between the groups. The Plastics are more evil!

If your t statistic < critical t score, then you fail to reject the null hypothesis, this means there is not a difference between the two groups. The Plastics and Cady, Janis, and Damien are on the same level of evilness.

And those are the basic steps of conducting a t-test!