cream 3 - analysing quantitative data and inferential statistics

why is knowing the level of measurement important

helps you decide what descriptive statistics and which inferential statistical test is appropriate for your data

what is nominal level of measurement

data organised in categories with a measure of frequency and the categories are mutually exclusive - hair colour

what is ordinal level of measurement

data that is ordered in some way by putting things in size order however we don’t know the exact difference between each item - nando’s perimetre

what is interval level of measurement

data is made up of a scale of units that have equal intervals - temperature

why are inferential statistical tests used

to help assess the probability that the difference or correlation observed in the results are statistically significant or not

what are the four main inferential statistical tests

chi - squared, wilcoxon signed ranks, mann whitney u, spearman’s rho correlation coefficient

what does a significance level do

expresses the level of chance they are prepared to accept to be sufficiently satisfied that the results they have obtained show a genuine difference in conditions or correlation between co variables

what is the traditional level of significance chosen by psychologists

5% level or p less than or equal to 0.05

what is the observed value

the statistical tests turn all of the results into one value

what is the critical value

the value that’s in a table of values that have been pre calculated

what does p less than or equal to 0.05 or 5% level mean

the probability of the results having occurred by chance is less than or equal to 5%

what does p less than or equal to 0.1 or the 10% level mean

the probability of the results having occurred by chance is less then 10%

what does p less than or equal to 0.01 or the 1% level mean

the probability of the results having occurred by chance is less then 1%

what is a type one error

accepting the alternative research hypothesis when the results are actually due to chance - false positive

what is a type two error

accepting the null hypothesis when the results were not due to chance - false negative

what are non - parametric tests of significance

statistical tests that can be carried out when dealing with ordinal or nominal levels of measurement

what three things do you need to ask yourself when looking at what inferential statistical test to use

difference or correlation, nominal or ordinal/interval, related or un related

when should you use the chi square non - parametric test

when the test of difference collected nominal, unrelated data

when should you use the mann whitney u test

when the test of difference collected ordinal/interval, unrelated data

when should you use the wilcoxon signed ranks test

when the test of difference collected ordinal/interval, related data

when should you use the spearmans correlation coefficient test

when the test of correlation collected ordinal/interval, related data

under what other circumstances can you use chi square test

when the test of correlation collected nominal, related data