Inferential Testing Flashcards
What are degrees of freedom and how do you find them for a chi-squared test?
The number of values that are free to vary given that the overall total values are known
Calculation: (rows - 1) x (columns - 1) in your contingency table
What is the mnemonic to see if chi-squared test is appropriate?
C - hi-squared
A - ssociation
N - ominal data
D - ifference
I - ndependent groups design
What are the 3 levels of measurement?
Nominal data
Ordinal data
Interval data
What is nominal data?
Data that is classified into distinct categories without any order or ranking and can be recorded in terms of frequencies. Each piece of data can only fit into one category
What is ordinal data?
Data that can be placed in a specific order or rank, but the intervals between the data points aren’t necessarily equal e.g a scale from 1-10
What is important to remember to identify the right level of measurement?
Focus on what is being measured, ignore any other noise
What is interval data?
Data that can be measured using fixed units of equal intervals and an extension to ordinal data
- will have a standardised measure of units
What is the mnemonic for related t-test?
D-ifference between two variables
I-nterval data
R-epeated measures design
T-test
What is the mnemonic for unrelated t-test?
U-nrelated
N-ominal data
I-ndependent groups
T-test
D-ifference
What is a level of significance?
The probability that there is a statistical difference or correlation that would make results statistically significant
What is a p-value?
The percentage probability that the findings have happened by chance
What does it means that psychological research usually uses a p-value of <= 0.05
Probability that results were due to chance are equal or less than 5%
- if the results are found to be statistically significant, there would be a 5% chance that it was a fluke e.g willing to accept a 5% margin for error
What is a lenient and stringent p-value?
A lenient p-value is one where there is a large margin for error e.g accept high probability that results occurred by chance
Stringent p-value is the opposite
Why do psychologists typically use a p-value of less than or equal to 0.05?
. This degree of uncertainty is acceptable as psychological tests aren’t a matter of life and death
. In clinical trials, a 1% p-value is needed as we need to be very careful about taking chances
. In psychology there are so many individual differences that a more stringent p-value would rarely ever produce statistical significance
What is a type 1 error?
When the p-value is very lenient so we may decide results are statistically significant when they’re not as we are taking a heavy proportion of chance results (false positive)
What is a type two error?
When you accept the null hypothesis when you should have rejected as the significance level is too stringent (false negative)
