What is the nominal level of measurement
Data is in separate categories/ groups
What is the ordinal level of measurement
Data is ordered
Eg- rating scale, 1-10, strong agree- strong disagree
What is the interval level of measurement
Data is measured using units of equal increments
Eg number of correct answers, time, test scores
What are descriptive statistics
A way of summarising and describing data but do not allow us to make a conclusion related to our hypothesis
What level of measurement do you use for mean
Interval data
Most sensitive measure- can be distorted by extreme values
What level of measurement do you use for median data
Interval and ordinal data
Not affected by extreme scores
What level of measurement do you use for mode data
Nominal
Ordinal
Interval
Unaffected by extreme values
Wnat are measures of dispersion
How spread out data items are
What is standard deviation
Measure of average distance between each data item above and below the mean
-considers all values
-may hide extreme values
Features of normal distribution
-Bell shaped curve= equal distribution
-Mean median mode are in the exact mid point
Features of negatively skewed distribution
-skews towards high scores, long tail on the left
-mode is greater than mean, median in middle
-few extreme scores
Features of positively skewed distribution
-skews towards low scores, long tail on right
-mean is greater than mode, median in middle
-Few extreme high scores
Features of a bar chart
-Height of bar= frequency
-suitable for categorical and nominal data
Features of histograms
No gaps between bars
Used in ordinal and interval data
Features of scattergrams
Used in correlational analysis
Internal and ordinal data
3 factors that decide a statistical test
1- test of difference/ correlation
2- experimental design
3- level of measurement
Mnemonic for learning statistical data table
Carrots should come mashed with suede under roast potatoes
Chi squared, sign test, chi squared, Mann-Whitney, wilcoxen, spearman’s Rho, unrelated t-test, related t-test, Pearsons R
How do you distinguish a test of difference/ correlation
Difference- looks at difference between 2 groups
Correlation- looks at relationship between 2 co-variables
How to distinguish between experimental designs in a statistical test
Unrelated design- independent measures
Related design- repeated measures or matched pairs
3 pieces of info thst must be used on a critical value table
-One tailed or two tailed hypothesis
- number of participants
- level of significance- written as p value
How do we decide whether a test is significant WITH an R in its name
The calculated value is GREATER than or EQUAL to critical value in order to accept the alternative hypothesis
Otherwise accept null
How do we decide whether a test is significant for test WITHOUT an R in their name
Calculated value is LESS than or EQUAL to critical value in order to accept alternative hypothesis
Otherwise accept null
What is a type 1 error
Researcher is too optimistic and select a p value that is too high (eg 0.1) and leaves too much room for results to be down to chance
False positive
What is a type 2 error
Occurs when a researcher is too stringent and and selects a p value that is too low (eg 0.01) and leaves not enough room for results to be down to chance
False negative
-
Self Report Techniques20
-
Ethical Issues24
-
Experiments17
-
Experimental Design12
-
Observations18
-
Observational Design13
-
Meta Analysis10
-
Correlations8
-
Hypotheses5
-
Sampling Methods20
-
Extraneous Variables9
-
Peer Review And Pilot Studies11
-
Psychology And The Economy10
-
Features Of Science15
-
Content Analysis2
-
Reliability11
-
Statistics27
