Research methods - Inferential Statistics

Question 1

Step 1

Answer 1

• write out null hypothesis

Question 2

Step 2

Answer 2

• choose a significance level

Question 3

Step 3

Answer 3

• turn experimental results into a single test statistic
• find out what the probability is that this test statistic

Question 4

Step 4

Answer 4

• assume alternative hypothesis is true instead

Question 5

Step 5

Answer 5

• if null hypothesis is rejected, results are statistically significant

Question 6

Step 6

Answer 6

• if you don’t reject the null hypothesis, it means that your results could have occurred by chance
• if this happens, you’ve proved nothing

Question 7

Step 7

Answer 7

• significance level of 0.05 is okay for more tests

Question 8

Type I error

Answer 8

• when you reject the null hypothesis when it was actually true
• significance levels gives you the probability of this happening
• this is why significance levels are small

Question 9

Type II error

Answer 9

• when you don’t reject the null hypothesis when it was actually false
• can happen if your significance level is too small

Question 10

1. Must decide if it’s a one-tailed or two-tailed test

Answer 10

one-tailed test: when the researcher has predicted an association and has also stated which way the results will go
two-tailed test: when the researcher has predicted an association, but hasn’t stated which way the results will go

Question 11

2. The observed value is then looked up in a critical value table

Answer 11

• some of the critical values for a Spearmen’s rho correlation coefficient are shown here
• as the test is two-tailed, the researcher needs to look at the row of significance levels for two-tailed tests

Question 12

3. Read off the table

Answer 12

• if the psychologist wants to see if his results are significant at the p=0.05 level, he should use the appropriate column

Question 13

4. Make a conclusion about significance

Answer 13

• for the data to be significant, the observed value from the Spearman’s rho calculation must be greater than or equal to the relevant critical value

Question 14

Experimental design

Answer 14

• related measures or unrelated measures

Question 15

Research aims

Answer 15

• whether there is a significance difference between 2 groups of scores
• see if there is a significant association between 2 variables

Question 16

Level of Measurement / Type of Data

Answer 16

-