statistical testing
Cards (6)
- Statistical test
- Used to determine whether a different or association/correlation found in a particular investigation is statistically significant (ie whether the result could have occurred by chance or there is a real effect)
- criteria for statistical (inferential) tests
- Looking for a difference or a correlation/association?
- Is experimental design related (repeated measures/matched pairs) or unrelated (independent groups)?
- What is the level of measurement (nominal/ordinal/interval data)?
- levels of measurement
- Nominal data:
- Categories
- Each item can only appear in one category, there is no order
- Eg = people naming their favourite football team
- Ordinal data:
- placed in order, intervals are subjective
- Whole numbers that can be ordered, but are not necessarily precise measurements
- Eg = ranking your mood on a scale of 1-10
- Eg = 1st, 2nd and 3rd place in a race
- Interval data:
- Units of equal, precisely defined size
- Based on numerical scales
- Eg = weights in kg
- Eg = heights in cm
- Also includes counting observations in an observational study (eg 8 tallies is twice as much as 4 tallies)
- choosing a statistical test:
- How to remember = Carrots Should Come Mashed With Swede Under Roast Potatoes
- Parametric and non-parametric tests
- Parametric tests = unrelated t-test, related t-test, pearson’s r
- Non-parametric tests = chi squared, sign test, mann whitney, wilcoxon, spearmans rho
- Sign test
- The score of condition B is subtracted from condition A to produce the sign of difference (either a plus or minus)
- Add up the number of pluses and the number of minuses
- Participants who achieved the same score in condition A and condition B should be disregarded and deducted from the N value (number of participants)
- The S value is the total of the less frequent sign
- If S is equal to or less than the critical value = then S is significant and the null hypothesis is rejected and the alternative hypothesis accepted at the 5% level of certainty