Statistical Testing

Question 1

● What do statistical tests help psychologists determine?

Answer 1

Likelihood that results are due to chance or the IV/co-variables

Question 2

● What is the accepted significance level in psychology?

Answer 2

Less than 5% (p < 0.05)

Question 3

● What does a p value of <0.05 mean?

Answer 3

Less than a 5% probability that results occurred by chance

Question 4

● What is an alternative hypothesis?

Answer 4

Predicts a difference or correlation

Question 5

● What is a null hypothesis?

Answer 5

Predicts no difference or association

Question 6

● What is the calculated/observed value?

Answer 6

A unique value given by the statistical test for research

Question 7

● What is a critical value?

Answer 7

Value from a table used to judge significance

Question 8

● What happens if the calculated value is significant?

Answer 8

Accept the alternative hypothesis

Question 9

● What happens if the calculated value is not significant?

Answer 9

Accept the null hypothesis

Question 10

● What are the three criteria for justifying a statistical test?

Answer 10

Test of difference/association, experimental design, level of measurement

Question 11

● Which test is appropriate for nominal data with repeated measures?

Answer 11

Sign Test

Question 12

● Which test is appropriate for ordinal data with independent groups?

Answer 12

Mann Whitney U

Question 13

● Which test is used for correlation with ordinal data?

Answer 13

Spearman’s Rho

Question 14

● What is a one-tailed hypothesis?

Answer 14

Directional hypothesis

Question 15

● What is a two-tailed hypothesis?

Answer 15

Non-directional hypothesis

Question 16

● What does N or df represent in statistical tests?

Answer 16

Number of participants or degrees of freedom

Question 17

● What does it mean if the results are significant?

Answer 17

Calculated value meets criteria for significance

Question 18

● For Sign Test, Mann-Whitney, and Wilcoxon, when is the result significant?

Answer 18

Calculated value must be equal to or less than the critical value

Question 19

● For Chi Squared, Spearman’s Rho, and t-tests, when is the result significant?

Answer 19

Calculated value must be equal to or more than the critical value

Question 20

● What is the first factor to find in a scenario to help identify the critical value?

Answer 20

One or two tailed hypothesis

Question 21

● What is the second factor to find in a scenario to help identify the critical value?

Answer 21

Number of participants (N) or degrees of freedom (df)

Question 22

● What is the third factor to find in a scenario to help identify the critical value?

Answer 22

Level of significance (usually p<0.05)

Question 23

● Once you have found the critical value, what value is this then compared to? Where is this found?

Answer 23

Identify the calculated/observed value found in the scenario/stem

Question 24

▲ What three details should you mention in a justification for the Mann Whitney U test?

Answer 24

Test of difference, independent groups, ordinal data

Question 25

▲ Why is it important to link the level of measurement to the DV in your answer?

Answer 25

The level of measurement is always in reference to the DV

Question 26

▲ What do you do if your results are not significant?

Answer 26

Accept the null hypothesis and reject the alternative

Question 27

▲ How do you write up significance using the writing frame provided?

Answer 27

State critical value, p value, hypothesis type, N/df, calculated value, significance, and which hypothesis to accept/reject

Question 28

▲ What does a calculated value of S represent in the Sign Test?

Answer 28

The less frequent sign in the data

Question 29

▲ In the Sign Test, what do you do with participants who show no difference?

Answer 29

Ignore them and adjust N accordingly

Question 30

▲ What must you do after calculating the value of S in the Sign Test?

Answer 30

Compare with critical value for significance

Question 31

▲ What is a directional hypothesis?

Answer 31

One-tailed hypothesis which predicts a direction of results

Question 32

▲ What is a non-directional hypothesis?

Answer 32

Two-tailed hypothesis which predicts there will be a difference in the results

Question 33

▲ What is a worked example of using a statistical table to check significance?

Answer 33

Compare calculated and critical values, apply rules for the test

Question 34

▲ Why do researchers sometimes use a more stringent significance level, e.g. p<0.01?

Answer 34

To reduce risk when people’s lives are at risk (e.g. drugs trials)

Question 35

✪ Critically discuss why psychologists use the 5% significance level.

Answer 35

Psychologists use a P<0.05 as it strikes a balance between the risk of making the type 1 and 2 error. It’s a conventional significance level.

Question 36

✪ Explain the process for accepting or rejecting hypotheses.

Answer 36

If the calculated value is significant we can accept our alternative hypothesis. If it is not significant then we must accept our null hypothesis and reject the alternative hypothesis.

Question 37

✪ Critically discuss why ignoring equal results and adjusting N is necessary in the Sign Test.

Answer 37

We are only interested in participants who have a difference in score.