Summer SAVY 2016 (Session 1, Day 2) – Behavioral Science in Action
Dear Behavioral Science in Action parents,
We had a wonderful second day exploring inferential statistics and their use in behavioral science.
To teach the class the different types of relationships that can exist between variables, we introduced them to correlations and t-tests. To understand correlations, the class created scatterplots and discussed how the variables they examined were related or unrelated, as well as what inferences we can make using scatterplots. Once it was clear we couldn’t use correlations to conclude that one variable causes a change in the other, we moved on to t-tests. As a class we explored how the t-test can help us determine whether differences between groups of participants are likely to occur, more so than chance.
To explore causation, the class participated in a social experiment meant to demonstrate how canned laugh-tracks after insults on TV shows can change how likely we are to step in when we see bullying. Half of the class saw video clips from a popular tween tv show and the other half say the same clips, but with the canned laugh tracks removed. At the end, the students watched a mock bullying encounter and were asked to write down when they would step in and stop the bullying. As expected, our no-laugh track group indicated they would step in sooner than their peers in the laugh track group. The class was even surprised by the results!
Tomorrow we move on to using the statistical procedures we have learned about on computers with much larger data sets!
Emily and Israel