statistics Flashcards
what are the 3 factors to choose a statistical test ?
- difference or correlation
- Design measure
- Levels of measurement
What is nominal data ?
Data is represented in the form of categories
What is ordinal data?
Data is ordered in some way eg a rating scale where scores are converted into ranks
What is interval data?
Based on numerical scales that include precise units eg time, temperature, weight
What central tendency + dispersion is used for nominal data?
mode
n/a
what central tendency + dispersion is used for ordinal data?
median
range
what central tendency + dispersion is used for interval data?
mean
sd
When do you use chi squared?
Nominal
Unrelated
Test of association or correlation OR difference
When do you use Mann Whitney?
Test of difference
Ordinal
Unrelated
When do you use unrelated t test
test of difference
Interval
unrelated
When do you use the sign test?
Related design
Nominal data
Test of difference
When do you use wilcoxon?
Test of difference
Related
Ordinal
When do you use spearman’s rho?
Test of association or correlation
Ordinal data
When do you use related t test
Test of difference
Related design
Interval
When do you use Pearsons r ?
Test of association or correlation
Interval
what does the null hypothesis state?
no difference between conditions
what is the level of significance - definition and value?
point at which the researcher can claim to have discovered a large enough difference or correlation within the data so they can accept alternative
0.05
what is the difference between calculated and critical value ?
calculated - observed value
critical - tells us whether or nor we can reject the null hypothesis and accept alternative
what are the 3 criteria to use tables of critical values?
One tailed or two tailed
No of ppts in the study
Significance level
What is the N value on the table of critical vales?
no of ppts
what is a type I error?
false positive as a correlation does not exist so they should have accepted null hypothesis instead of reject
what is a type II error?
false negative is we should reject null hypothesis and accept alternative
