APA 2180 Lecture Notes - Lecture 16: Effect Size, Null Hypothesis, Alternative Hypothesis
20 Mar 2019
School
Department
Course
Professor
Document Summary
Without collecting any data, how can we determine how many participants are. Ratio: money is an example, absolute zero. Mean can be impacted by a high score or a low score. Median is best used in skewed distributions. Alternative hypothesis: has a direction, there is differences between groups. The value you give the p-value is called the alpha. If p-value is greater than alpha you do not reject the hypothesis and say that there is a difference between the group. 01 is more conservative and the data needs to be stronger to reach that number. Type i and type 2 error (must know the difference between these. Type 1 is a false positive, null hypothesis is true, but you reject it. Type 2 is a false negative, hypothesis is false, but you accept it. Being more conservative can sometimes be better. Alpha is the risk you are willing to take. Choosing a smaller alpha reduces the power.
