methods: inferential statistics

Question 1

what is the purpose of an inferential test of significance?

Answer 1

• determines if the effect is real or due to chance
• used when data differs between conditions
• helps decide whether to accept alternative or null hypothesis
• indicates whether to retain or reject a hypothesis

Question 2

what is the role of probability in inferential testing?

Answer 2

• probability is the likelihood of an event occurring
• tests if data differences are due to chance or something else
• helps decide if we accept the null hypothesis
• inferential tests show if results are likely due to chance

Question 3

what is the significance level in psychology and how is it used?

Answer 3

• significance level is p < 0.05
• means 5% or less chance the results are due to chance
• if p ≤ 0.05, support the alternative hypothesis
• if p > 0.05, retain the null hypothesis
• helps decide if results show a real effect or not

Question 4

what does it mean if an inferential test is significant or not?

Answer 4

• if significant: support experimental hypothesis
• 95% confident prediction is correct
• 5% or less chance results are due to chance
• if not significant: accept null hypothesis
• less than 95% confident prediction is correct
• more than 5% chance results are due to chance

Question 5

what is a type 1 error and when does it happen?

Answer 5

• accepting the alternative hypothesis when it’s wrong
• results were actually not significant
• should have retained the null hypothesis
• happens when significance level is too lenient (e.g. p ≤ 0.1)
• also called a false positive

Question 6

what is a type 2 error and when does it happen?

Answer 6

• retaining the null hypothesis when it’s wrong
• there was actually a real effect
• should have accepted the alternative hypothesis
• happens when significance level is too strict (e.g. p ≤ 0.01)
• also called a false negative

Question 7

what is nominal data?

Answer 7

• most basic type of data
• data is in categories or groups
• only category totals are known
• no info about individual values
• example: number of students with or without pets

Question 8

what is ordinal data?

Answer 8

• data is ranked or ordered
• shows position but not exact differences
• intervals between values aren’t equal
• often from arbitrary scales (e.g. test grades, ratings)
• example: 1st, 2nd, 3rd in house points competition

Question 9

what is interval and ratio data?

Answer 9

• values are measured on a scale with equal intervals
• more precise than nominal or ordinal data
• interval data has no true zero (e.g. temperature in °C)
• ratio data has a true zero (e.g. height, time, distance) bc they start at 0
• often collected using standard scales or instruments