probability and significance

Question 1

what is probability?

Answer 1

a measure of the likelihood that a particular event will occur where 0 indicates statistical impossibility and 1 statistical certainty

Question 2

what does a statistical test determine?

Answer 2

whether the alternate or null hypothesis is true and therefore whether we accept or reject the null hypothesis

Question 3

what is significance?

Answer 3

a statistical term that tell us how sure we are that a different or correlation exists

Question 4

what is a significance level?

Answer 4

• the point at which the researcher can reject the null hypothesis and accept the alternative hypothesis

Question 5

what is the usual level of significance in psychology?

Answer 5

• 0.05
• properly written as p ≤ 0.05

Question 6

what does p ≤ 0.05 mean?

Answer 6

• probability that the observed effect occurred when there is no effect in the population is equal to or less than 5%
• even when a researcher claim to have found a significant difference / correlation, there is still up to 5% chance that it isn’t true for the target population it was drawn from

Question 7

why can pyschologists never be 100% certain about a particular result?

Answer 7

• they have not tested all members of the population under all possible circumstances
• therefore, psychologists have settled on a conventional level of probability where they are prepared to accept that results may have occurred by chance

Question 8

how is the calculated value checked for statistical significance?

Answer 8

must be compared with a critical value

Question 9

what is a critical value?

Answer 9

numerical boundary that tells us whether or not we can reject the null hypothesis and accept the alternative hypothesis

Question 10

what 3 criteria can determine what critical value to use from a table of critical values?

Answer 10

• one-tailed or two-tailed test?
• number of participants in the study (N or df)
• level of significance (p value, 0.05)

Question 11

which tailed test should you use for different hypotheses?

Answer 11

• one-tailed = directional hypothesis
• two-tailed = non-directional hypothesis
• probability levels double when two-tailed tests are used as they a more conservative prediction

Question 12

what is a type I error?

Answer 12

• false positive
• incorrect rejection of a true null hypothesis
• finding a significant difference or correlation when one does not exist

Question 13

what is a type II error?

Answer 13

• false negative
• failure to reject a false null hypothesis
• not finding a significant difference or correlation when one does exist

Question 14

which type of error is most likely based on the level of significance?

Answer 14

• more likely to make a type I error is significance level is too high eg. 0.1
• more likely to make a type II error if significance level is too low eg. 0.01
• 5% level of significance best balances risk of making a type I or II error

Question 15

what is the rule of R?

Answer 15

• tests with ‘R’ in their name should have a calculated value that is equal or more than the critical value
• tests without ‘R’ in their name should have a calculated value that is equal to or less than the critical value

Question 16

which statistical tests need a calculated value that is equal to or more than the critical value?

Answer 16

• chi-squared
• related t-test
• unrelated t-test
• pearson’s r
• spearman’s rho

Question 17

which statistical tests need a calculated value that is equal to or less than the critical value?

Answer 17

• mann-whitney U
• sign test
• wilcoxon

Question 18

why might drug testing have more stringent levels of significance?

Answer 18

• impact of research may have far-reaching consequences
• psychologists want to be more sure the effects are due to drug rather than due to chance, especially if the drug has unpleasant side effects