Inferential testing Flashcards
(23 cards)
What is a statistical test?
a calculation on data to determine which hypothesis to accept
Which hypothesis is accepted is results are significant
The experimental hypothesis
Which hypothesis is accepted is results aren’t significant
The null hypothesis
What factors are needed for a parametric test
- interval level of measurement
- data must have a normal distribution
What is nominal data?
- when ppts are placed into
categories - represented on a bar chart
What is ordinal data?
when ppts have a rough numerical score but proper units arent used
What is interval data?
info is scores using recognisable equal units
What is a strength of nominal data
- easy to generate from closed questions
- can generate alot of data quickly
What are limitations of nominal data?
- data is unsophisticated as it doesn’t yield a numerical result
- ppts are unable to express different degrees to a response
What is a strength of ordinal data
ppts are able to express different degrees to s response so data is more sensitive
What are limitations of ordinal data?
- ## data lacks precision as its based on subjective opinion
What is a stregth of inteval data?
- more precise than nominal data or ordinal data as its based on numerical scales that include units
When would you use the sign test
When level of measurement is nominal
and when experimental design is repeated measures or matched pairs
When would you use the chi squared test?
When data is nominal and if experimental design is independent measures or a correlation
When would you use a Wilcoxon test?
When data is ordinal and experimental design it repeated measures or matched pairs
When would you use a Mann whitney test?
What data is ordinal and experimental design is independent measures
When would you use a spearmans rho test?
when data is ordinal and correlational
When would you use a related t test?
When data is interval and the experimental design is repeated measures or matched pairs design
When would you use an unrelated t test?
when data is interval and experimental design is independent measures
When would you use a pearsons r test?
When data is interval and
correlational
What is a one tailed test
used when the hypothesis is directional
What is a two tailed test
used when hypothesis is non directional
How do you use the sign test
- use ‘-‘ and ‘+’ next to data which does/doesn’t support the hypothesis
- if there’s no difference use 0 (omitted value)
- add un the ‘-‘ and this is the S value
- Calculate N by Counting the number of ppts and misusing the omitted values
- find critical value on contingency table
