T - Tests II Flashcards
What is a t - test used for?
Used to determine whether the mean for a sample differs significantly from:
- Mean of another sample drawn from the same population.
- Estimated population mean. (hypothesised mean).
What is the purpose of a spirometry?
A measure of flow through the airways.
What is COPD?
Constriction of the airways due to smoking.
What is a hypothesis?
It is a statement and IS NOT the same as a scientific question.
How do you make a comparison between two data sets?
You propose a relationship between them.
What does H0 mean?
There is NO difference between the two data sets.
What does H1 mean?
There is A difference between the two data sets.
What does the null hypothesis mean?
There is NO DIFFERENCE between their means.
What does the alternative hypothesis mean?
That there IS A DIFFERENCE between their means.
What does a T score do?
Converts raw data into a standardised score when you do not know the population mean.
What does the position of the standardised T score on the T distribution link to?
Relates raw data to likelihood of the occurence.
Therefore, allows the comparison of the data.
What are the 3 main types of t - tests?
One Sample t - test
Paired t -test
Two sample t -test
What is a one sample t - test?
Compares mean of a single sample to known or hypothesised population mean.
What is a paired t - test?
Compares difference between two samples to hypothesised difference.
What is a two sample t - test?
Compares means of two samples.
How can you tell if a data set is independent?
It is independent if the probability of A occurring DOES NOT influence the probability of B occurring.
How can you tell if a data set is paired?
It is paired if there is a link between the two observations.
What is standard error?
Standard deviation of the sample means where multiple samples are taken from the sample population ๐/โ๐
What does variance mean?
How far a data set is spread out (average of squared distances from the mean).
What does the Leveneโs test measure?
Determines whether the samples have equal variance.
Alongside this, it tests whether the two conditions have the same or different amounts of variability between scores.
What happens if the samples have equal variance?
A standard t - test can be used.
At what probability level is equal variance considered?
P > 0.05
What is the p value?
Probability of obtaining an effect at least as extreme and the one in your sample data.
When is the null hypothesis considered true?
When the sample difference is greater than the P = 0.05 (or whatever it is)
This assumes that there is NO difference.
When is a two tailed test used?
There is NO difference between the groups.
What is confidence intervals (CI)?
Gives measure of uncertainty of the estimated difference.
What value is assumed a normal distribution?
Value of 1.96
What happens if the 95% CI for the difference contains the value 0?
P value > 0.05
What happens if <0.05 occurs?
Reject the null hypothesis and thus, there is a STATISTICALLY significant difference.
