Data Flashcards
what is nominal, ordinal, and interval data?
nominal: data produced as named categories
ordinal: data that can be ranked in order (however the size of each increase does not need to be equal) - e.g. rating scales (discrete data)
interval: data measured using units of equal intervals e.g. height (continuous data)
how do you carry out the binomial sign test?
- state the alternative hypothesis and place raw data in a table
- record a plus sign if the number in column A is larger than the number in column B, and a minus sign if it’s not
- add up the number of pluses and minuses and select the smaller value - this is the observed value of S
- find the critical region by looking at the critical values table
- if the observed value is less than or equal to the critical value, findings are significant
how do you carry out the mann whitney u test?
- rank scores from both groups all together
- add up the ranks for both groups separately
- use the smaller sum of the ranks for the R value in the formula
- use the table to find the critical value for the test, through finding the intersection between the no. of participants in each group
- the research is significant if the observed value of U is equal to or less than the critical value
how do you carry out the wilcoxon signed ranks test?
- find the differences between each set of scores
- rank the differences, ignoring any positive/negative signs and any 0 values (if ranks are the same, calculate the mean)
- find how many positive/negative differences there are and add the rankings of the less frequent sign - this is the observed value
- the n value is the no. of differences (excluding 0s) - use this to find the critical value
- research is significant if the observed value is less than or equal to the critical value
what is the observed value?
the value produced after applying an inferential test formula
what is the significance level?
the level of probability at which it has been agreed to reject the null hypothesis
what is a table of critical values?
a list of numbers that inform you whether an observed value is significant or not
what is the criteria for using a parametric test?
- data must be interval
- data must be drawn from a population with a normal distribution
- data must be from groups with similar variances