When psychologists wish to make inferences about collected data in order to estimate the likelihood that the data collected occurred due to chance or to draw conclusions about a larger population from which the sample was collected
What are parametric tests?
More powerful and robust methods of analysis
More accurate estimates of probabilities are collected, making type 1 and 2 errors less likely to occur
In order for data to be parametrically tested, what criteria must be met?
The data must be precise, interval level of measurement
Scores must be drawn from a normal distribution of the population with few scores at both ends and more in the middle
The variability of the two sets of the scores must be similar
How could a researcher test the distribution of data collected?
The mean, median and mode are compared
If the data is of a normal distribution, these figures should be similar
What is the homogeneity of variance?
The standard deviation2
How might one measure the variability of the data sets collected?
The standard deviation
Dot plot or box plots
What requirements must be met in order for non parametric tests to be employed?
The data must be nominal or ordinal
Results must be of a skewed distribution
There must be a wide variance in scores between two sets
What questions should you ask yourself before choosing a statistical test?
Am I looking for a difference between scores or a correlation?