Cognitive Data Analysis Flashcards
Strengths of mean
Most sensitive and powerful as it uses all the scores
Weaknesses of mean
Extreme scores can skew the score - may not be representative
May not be an actual score (eg 0.5 of something that can’t be divided)
Strengths of median
Not affected by extreme score
Weaknesses of median
If there’s a small data set it may not be useful - may not give accurate representation
Can take a long time to figure out with large sets of data
Strengths of mode
Easy to calculate
Can be used on non-numerical data
Weaknesses of mode
If there are more than 2 modes it becomes meaningless
Strengths of range
Easy to calculate
Weaknesses of range
Can be affected by extreme scores
Does not tell us if the spread of data is normal or skewed
Strengths of standard deviation
More accurate as it uses all scores
Weaknesses of standard deviation
Hard to calculate manually
What is the observed value?
The value that has been worked out from the ppts’ results that is compared to the critical value (on the table at front of paper)
What is a Type 1 error?
If you reject the null hypothesis and it is actually true
Happens when the level of significance is too lenient
What is a Type 2 error?
If you keep the null hypothesis when your experiment did work. (when there is a significant difference)
Happen when the level of significance is too harsh
What level of probability do we start at?
p <= 0.05
The probability the results are due to chance is equal to or less than 5%
(unless we’re trying to disprove someone else’s results. Then we start at p <= 0.01)
What are the 4 levels of measurement/types of data?
Nominal
Ordinal
Interval
Ratio
What is Nominal data?
Most basic level of measurement
Used when data is put into tally charts/categories
Only tells us how many ppts picked each option
Doesn’t give an individual score for each ppts
What is Ordinal data?
Used when data can be ordered eg 1st, 2nd, 3rd
Can’t tell the gap between 1st and 2nd etc
Always used with questionnaires
Gives us an individual score for each ppts
What is Interval data?
The distance between each score has a meaning and equal value.
eg gap between 5-10cm is same as 20-25cm
0 is just another number on the scale, it does not mean there is nothing
eg 0 degrees c doesn’t mean there’s no temperature
What is Ratio data?
The distance between each score has meaning and an equal value.
However, 0 does mean nothing.
eg 0 seconds to do something means it took no time
Which test should you use? :
It is NOT an experiment
It uses ORDINAL data
Spearman’s Rho
Which test should you use? :
It is an experiment
It uses REPEATED MEASURES
It uses ORDINAL data
Wilcoxon Test
Which test should you use? :
It is an experiment
It uses INDEPENDANT GROUPS
It uses ORDINAL data
Mann Whitney U Test
Which test should you use? :
It is an experiment
It uses INDEPENDANT GROUPS
It uses NOMINAL data
Chi Square
