Choosing a statistical test

Question 1

what criteria should be considered when picking a statistical test?

Answer 1

• if it is a correlation or difference
• whether level of measurement is nominal, interval or ordinal
• if it is independent groups or repeated measures

Question 2

explain how to decide if it is a correlation or difference?

Answer 2

• check if the researcher within the aims is looking for a difference or correlation
• look at hypothesis

Question 3

explain how to pick the experimental design?

Answer 3

• repeated measures= the same pp’s are used in all conditions + matched pairs as pp’s have been matched even if pp’s in both conditions aren’t the same
• independent groups= pp’s are different

Question 4

diagram for ‘choosing a statistical test’

Answer 4

-look at flashcards

Question 5

explain how to identify Nominal data for ‘levels of measurement’

Answer 5

• counting things and putting them into categories eg. male and female
• no scale of measurement
• can’t be organised into rank
• categorical

Question 6

explain how to identify Interval data for ‘levels of measurement’

Answer 6

• measuring things on a scale with equal units
• eg. seconds on a stopwatch, kg…
• standardised

Question 7

explain how to identify Ordinal data for ‘levels of measurement’

Answer 7

• measuring things that can be put into rank
• but no scale of measurement
• height= tall, average, short
• subjective

Question 8

provide the paragraph on how to write up inferential findings?

Answer 8

• the results are significant/not significant
• suggesting there is/is no correlation/difference between
• the observed value(…) as it is less/more than the critical value (…)
• when p≤ 0.05 and is one/two tailed
• therefore we can accept/reject the null hypothesis and accept/reject the research hypothesis
• we conclude that (and relay the hypothesis of study)