Join Aldenham's Quizlet class

Criteria for using each of Inferential Statistical Tests

You need to be able to say the 3 reasons or criteria for each of the 5 tests 

1. test of difference / association
2. nominal / at least ordinal data
3. independent / repeated measures

Mann Whitney

• test of difference
• at least ordinal data
• independent measures

Wilcoxon

• test of difference
• at least ordinal data
• repeated measures

Chi Squared

• test of difference
• nominal data
• independent measures

Binomial

• test of difference
• nominal data
• repeated measures

Spearman's

• test of association
• at least ordinal data
• independent measures

How to calculate the inferential statistical tests

inferential_statistical_tests.pptx
File Size:	308 kb
File Type:	pptx

Download File

---

Type 1 and 2 errors 
Type 1 = false positive, wrongly accepting an incorrect alternative hypothesis when the null should have been accepted. It happens when the probability is set too big (e.g. saying that p = 0.5). It is saying that something works when it really could have happened by chance. 

Type 2 = false negative, wrongly rejecting a correct alternative hypothesis when the null should have been rejected. It happens when the probability is set too small (e.g. saying that p > 0.01). It is saying that something probably happened by chance, when in fact there was a difference / something did happen. 

Levels of Measurement