Data Flashcards
What is standard deviation?
How scores vary from the mean
What is the p-value?
Probability level, under 0.05 is statistically significance
What is Pearson’s correlation coefficient?
Strength of linear relationships (parametric)
What is Spearman’s correlation coefficient
Strength of relationships that are in the same direction but does not have to be linear (non-parametric)
Reliability
Consistency of results
Range
Difference between lowest and highest number
Variance
On average how much the scores differ from the mean
Validity
The extent to which a test measures or predicts what it is supposed to
What is ordinal data?
Categories which are ordered (ranks, scales) but has irregular gaps i.e. body mass
What is nominal data?
Data is qualitative with no rank order i.e. hair colour
What is interval data?
Steps in the scale are evenly placed but zero does not mean zero (nothing) i.e. temperature
What is ratio data?
Steps in the scale are evenly placed and zero means zero (nothing) i.e. length
Parametric has these qualities:
- Ratio or interval data
- Mean
- Pearson correlation coefficient for correlation test
Non-parametric has these qualities:
- Ordinal or nominal data
- Median
- Spearman correlation coefficient for correlation test
Parametric is used for these tests:
- Independent measure with two groups = independent measure t-test
- Independent measure with more than two groups = one way independent measure ANOVA
- Repeated measure with two conditions = Matched pair t-test
- Repeated measure with more than two conditions = one way repeated measure ANOVA
Non-parametric is used for these tests:
- Independent measure with two groups = Mann Whitney-test
- Independent measure with more than two groups = Kruskal-wallis test
- Repeated measure with two conditions = Wilcoxon test
- Repeated measure with more than two conditions = Friedman’s test
Why is counterbalancing needed for repeated measures?
To avoid order effect as it can confound experiment results
When is the mean the most appropriate measure of central tendency?
When there is little sample variance and few outliers present
When is the median the most appropriate measure of central tendency?
When there is a higher variance as demonstrated by the SD
When is the mode the most appropriate measure of central tendency?
When dealing with nominal data
What is IQR?
Find the median and determine Q1 and Q3
i.e. 5, 9, 11, 13
5, 9, | 11, 13
Q1 = 7 Q3 = 12
IQR = Q3 - Q1
IQR = 12 - 7
IQR = 5
What are confidence intervals?
After 100 tests, it would end up in this margin with 95% confidence
Larger confidence intervals = less precise, more variance in the data
Overlapping = not statistically significant
What is a type I error?
A false positive
What is a type II error?
A false negative
What are some ethical considerations?
- Confidentiality
- Voluntary participation
- Withdrawal rights
- Informed consent
- Deception in research
What is an independent variable (IV)?
A variable changed by the experimenter i.e. music present
What is a dependent variable (DV)?
A variable measured to test the effect of the IV i.e. time, correct answers
What is an alternative hypothesis?
Predicts that there is a relationship between IV and DV
What is a null hypothesis?
Predicts that there is no relationship between IV and DV
