Stats Flashcards
What is nominal data?
Data that is categorical e.g. gender, colour
What is ordinal data?
Data that has a natural order e.g. position in a race
What is continuous data?
Data that is on a scale where differences between intervals are always the same e.g. height
What is the independent variable?
Variable manipulated by the experimenter - the one you use to test effects on the dependent variable
What is the dependent variable?
Variable that you measure
Dependent on the independent variable
What is the confounding variable?
Variable that has an effect on the dependent variable but that you are not interested in testing.
What is a between subjects design?
Independent samples where each subject is presented in a different condition
What is a within subjects design?
Related samples where each participant receives all conditions
What is inter-observer reliability?
Do 2 people get the same result when examining the sample?
What is intra-observer reliability?
Does one person get the same result when examining the sample at different times?
From basic tests how do you work out if the data is normally distributed?
Do a mean, median and mode test.
If all values are relatively similar data is normally distributed.
What tests determine whether data is normally distributed?
Kolmogorov-Smirmoff Test
Shapiro-Wilk Test
What tests do you use to test the means of two sets of data for differences?
Two independent sample t-test
Two paired smaple t-test
What test do you use to test the variance of two sets of data?
One sample t-test
What test do you use to test the means of more than two sets of data for differences?
One way ANOVA
Two way ANOVA
What do the terms one sided/two sided (tailed) mean when describing hypotheses?
One tailed - can be answered with yes or no e.g. “a happy dog will eat more” - tells you what type of effect there will be
Two tailed - can be answered with different outcomes e.g. “mood affects the appetite of dogs” - tells you there will be an effect
What is a type 1 error?
Rejecting the null hypothesis when it is true
aka diagnosing a male as pregnant
What is a type 2 error?
Failing to reject the null hypothesis when it is false
aka telling a pregnant lady she is not pregnant
If your data was non-parametric what test would you do instead of an independent samples t-test?
Mann-Whitney U test
If your data was non-parametric what test would you do instead of a paired samples t-test?
Wilcoxon Signed-rank test
What does it mean if a study is under-powered?
Too little data collected so nothing looks statistically significant
What does it mean if a study is over-powered?
Too much data collected so everything looks statistically significant
What two forms do you need to give the participants of your study?
Ethics consent form
Participant information sheet
How do u calculate the mean?
Add all numbers divide by sample size
How do u calculate mode?
Most common number
How do u calculate median?
Middle number
How do u calculate standard deviation?
Work out the Mean (the simple average of the numbers)
Then for each number: subtract the Mean and square the result.
Then work out the mean of those squared differences.
Take the square root of that and we are done!
What does standard deviation show?
How spread out the data is from the mean
High SD = very spread out
Low SD = little spread
