Really Simple Statistics: T-Tests #MRX

Welcome to the first in a series of Really Simple Statistics (RSS). There are lots of places online where you can ponder for hours over the minute details of complicated equations but very few places that make statistics understandable to everyone. I won’t be explaining exceptions to the rule or special cases here. Let’s just get comfortable with the fundamentals.

What is a t-test?

T-tests are used to determine if two average numbers are different from each other. For example, is the average height of men different from the average height of women? Is the average temperature this winter different from the average temperature last winter. Is the average spend at Starbucks in April different from the average spend at Starbucks in July.

Independent t-test

There are two basic kinds of t-tests. Independent t-tests compare the average scores for two different groups. Average scores of men vs average scores of women. Average scores of adults vs average scores of children. Average yum of Skittles vs average scores of Smarties. Average cost of roses saverage cost of tTulips. Buyers vs NonBuyers. If there are 100 people in the study, 50 of them belong to the buyer group and 50 of them belong to the NonBuyer group.

duboix from morguefile

Dependent t-test

The second basic type of t-test is the dependent t-test. It might also be called a paired t-test or a matched t-test. This test compares the average scores for one group on two occasions. For example, our average scores today vs our average scores yesterday. Our average scores in the daytime vs our average scores in the nighttime. Our average purchase spend last week vs our average purchase spend this week. If there are 100 people in the study, 100 of them shared their spend from last week and all 100 of them shared their spend from this week.

Easy peasy!

5 responses

  1. Hey LoveStats, I will be doing a taxometric analysis study any thoughts on appropriate statistical tools?

    1. You might want to ask @annmariastat, president of The Julia Group, statistical consulting. I love stats but she IS stats.

  2. Hi Annie,
    Only one word to say….. Chapeau

  3. Annie, this is great! I can’t wait to learn more.


  4. You’re right LoveStats: there just aren’t enough places that make statistics accessible and easy to understand. We think this is great, and look forward to the next in the series!

%d bloggers like this: