Inferential Statistics

1. What are inferential statistics based on?

2. What makes results significant?

1. The probability that results are due to chance or not (Is there a genuine difference b/w the conditions, or was it chance? Is there a real effect?)

2. If findings are unlikely to be due to chance = significant

What level of probability is needed in psych to be considered significant?

p=0.05 (5% chance they are due to chance/the null hypothesis is correct)

What does the letter p stand for in the statement p<0.05?

Probability

Sometimes the statement p≤0.05 is used instead of p<0.05. Explain the difference.

The first statement means that if the probability of the null hypothesis is true is equal to, or less than 5% the results are significant, whereas the second one means that it must be less than 5% to be significant

Explain what is meant by the phrase ‘significant at p ≤0.05’. You must use the term null hypothesis in your answer.

The likelihood/probability of the results occurring even if null hypothesis being true must be equal to or less than 5% in order for the results to be considered significant, and so the null hypothesis can be rejected.

What are observed values (OV)?

The number produced by the inferential test using the data collected in a study

What are critical values (CV)

Numbers in a table that you compare the OV w/ to see if it’s significant (+ reject the null hypothesis)

What must you need to know for finding a CV to compare w/ the OV?

1. Degrees of freedom - no. of Ps (N)

2. One-tailed or two-tailed test - directional/non-directional hypothesis

3. Significance level - assume p≤0.05, unless told otherwise

4. Whether OV needs to be </> than CV (see rule of R) to be significant

What is the rule of R?

If there’s an R in the name of the test, the OV must be GREATER than the CV to be significant

If there’s no R in the name, then the OV should be LESS than the CV

EXAM Q:

Use the info below to write a statement of significance based on your you values:

Level of significance for a one tailed test	0.05	0.01
Level of significance for a two tailed test	0.10	0.02
N	T≤
19	53	46
20	60	52
21	67	58
22	75	65

• Degrees of freedom: 20 people in your study so df/N 20

• You used a directional hypothesis so a one tailed test is required.

• You have selected a 5% significance level.

• Your observed value is 61. For Wilcoxon test the table of critical values says that the value of T must be equal to or less than the critical value for significance.

The OV (T=61) is greater than the CV (60) at p≤0.05, for a one tailed test where N=20. The Wilcoxon Signed Ranks test states the CV must be equal to or less than T. So, we must accept the null hypothesis as there is a greater than 5% probability that the results are due to chance and would still occur even if the null hypothesis is true.