Week 8 T-tests Flashcards
What does a t-test tell us?
Is there a statistically significant difference between two means
What does an independent t-test measure?
If there if a difference between two different groups of people
What does a paired t-test measure?
If there is a difference between one group were they provide two bits of data
What are the assumptions for a paired t-test?
Data is continuous/scale (interval or ratio)
Difference are normally distrbuted
No significant outliers in the differences
In paired t-test how do you calculate the degrees of freedom?
n-1
What do we do if our t-statistic is further away from 0 than the critical ‘t’ value?
We reject the null hypothesis
What happens if the 95% confidence interval contains 0?
We fail to reject the null hypothesis
What does it mean if in the ‘Paired samples test’ under ‘Sig (2-tailed)’ the value is <0.05?
Reject the null hypothesis
What are the assumptions of an independent t-test?
Data is continuous (ratio or interval) Data is normally distributed Both groups are a random sample of a population Homogeneity of variance
How do you calculate the degrees of freedom in an independent t-test?
n-2
If the value is above 0.05 in the ‘independent samples test’ table under ‘levene’s test’ under ‘sig’ what does it mean?
There is not a significant difference in the standard deviations
Annotate
Yellow- the difference between the two means Pink- t-statistic and degrees of freedom Blue- strong evidence to reject null hypothtesis Green- 95% confidence that this interval captures the true underlying difference.
What do you need to conclude in a report?
Introduction to the research State the hypothesis What test was used Assumptions were/were not met Mean difference or mean change (95% CI) T-statistic and p-value Reject, or fail to reject the null hypothesis Conclusion in laymen terms
How many participant are there in the study?
40
If a paired t-test is used in a study with 20 people what is the degrees of freedom?
19