Week 4/5- Descriptive statistics and evaluation of assumptions Flashcards
what is a positive skew?
when we have a lot of high results
what is a negative skew?
when we have a lot of low results
when we have symmetrical data where are the mean mode and median placed?
They are all the same value and they are in the middle of the range of data
what are we looking for when we look at the distributional normality using the shappy-wilks test?
How close our results are to the symmetrical distribution
give two examples of interval data
temperature scales or hours of the day
what is ratio data?
ordinary numbers which range from minus infinity to plus infinity
which type of test should we try to use if possible and why?
Parametric tests because they have higher statistical power
what must we do before performing a parametric test?
Check the data meets the assumptions
how do you know if the data you have is normally distributed?
The p value should be greater than 0.05
what should you do if your shappy Wilks test shows that your data is not normally distributed?
Use the non-parametric equivalent ( either mann-whitney or Kruskal- wallis)
what does the homogeneity of variance assumption check?
How similar the dispersion around the mean is for both variables that your using
what does it mean if your p-value is less than 0.05 when working out the homogeneity of variance
Your data is not homogenous
does the result of the Levene’s test affect whether you can go ahead with your parametric test?
no
what should we do if our data is not normal?
Check for outliers using boxplots to see if outliers are contributing to your non- normality
