COMM3710 - Fall 2011 - Section 001 - Lecture Notes - 09/22

*Please enter 3 lines between the last entry and put in your username using heading 3 then type in your notes.*



riosjosh

Power and Effect Size
Power is defined as the ability to reject the null
  1. As n increases so does power
  2. If variance increases then power decreases
Effect is the magnitude of the difference or relationship. So if your measuring two different things then the difference between them is the effect size. For example if men score 4.2 on happiness and women score 4.4 then the effect size isn’t very great.

Setup the decision logic.

HO is the null hypothesis. H1 is the research hypothesis.

Calculated value is the numerical value you get.

Chi squared is the critical value. This is the point that lies outside of the confidence level. To get this critical value you need alpha and the degrees of freedom. If the calculated value meets the critical value then you reject HO.

DF = degrees of freedom = the number of elements that are free to vary
If there are 5 options then there are 4 elements that are free to vary.

If you repeat the test or measure such as in a pretest and a posttest then degrees of freedom don’t change. Because you can only compare subject 1 pretest to subject 1 posttest.

The null is always hidden. The null always says no difference.

What is the expected value of any data set? The mean [Test question]

A test is either significant or not. There is no more significant.

The math is setup for only one time. You can’t change anything after the fact. One chance and one chance only.

HWK: Bring the theoretical sampling distribution of craps. Sampling distribution is the spread of all possible scores.