Week 11: Referential alarm calls in yellow warblers

Group project schedule

Lab outline

• Inferential statistics

• T-Tests

• T-Tests in Excel

• Worksheet overview

• To-do for next week

Inferential statistics

“How likely is it that the differences between my results and those expected based on a hypothesis have been produced by chance alone?“

T-Tests

A t-test is an inferential statistics method that tests for a significant difference in means between two groups. There are a variety of t-tests that make different assumptions about the data and your analysis goals, but the core question being asked by a t-test is:

Is the mean of group A statistically significantly different from the mean of group B?

What does statistically significant mean? Assuming the data obey the assumptions of the statistical model, a statistically significant result means that the effect you observe (in this case a difference between two groups) is larger than one expected by chance. This does not mean the results are scientifically significant- that is up to you as the scientist to determine, by observing how large the effect is and whether that effect makes a biological difference in your system.

Put into hypothesis testing language,

Null hypothesis (H₀): The difference in means is due to random chance

Alternative hypothesis (H_A): The difference in means is due to some real factor

The T-Test

Assessing statistical significance

What determines the threshold at which you choose to reject or fail to reject your null hypothesis? The alpha (α) value is a scientifically agreed-upon threshold used across fields. In most scientific papers, you will see an alpha of 0.05 used (although this is argued about a bunch). If your p-value is less than 0.05, then you can reject the null hypothesis and say that the difference in means between the two groups is not due to chance. If your p-value is greater than or equal to 0.05, then you fail to reject the null hypothesis and cannot say that the difference in means between the two groups isn’t due to chance.

T-Tests in Excel

Live demo- detailed overview in the lab worksheet.

Worksheet Overview

To-do

• Upload Lab Worksheet and Excel spreadsheet to Blackboard by next lab (04/10)

• Wrap up data collection- start exploring your data

• Weekly lab survey is your ticket out

• Speaker questions before the end of class on Monday