Basic stats knowledge Flashcards
Correlational research: pros and cons
Pros: Ecological validity
Cons: Not good for causality
Experimental research: pros and cons
Pros: Good causality (because IV manipulated)
Cons: Not always ecologically valid
Parameter
Value that describes the population (e.g. SD, mean…)
Describe the place of the median vs mean if: left skewed, right skewed or symmetrical
Left skewed (-): mean < median
Symmetrical: mean = median.
Right skewed (+): mean > median
Range formula
X largest - X lowest
What’s the fundamental difference between z-test and t-test?
z-test: SD pop known.
t-test: SD pop not known.
-> But sample SD & mean + population mean.
p-value
Proportion of data sets that would yield a result AS extreme or MORE extreme than the observed result IF H0 IS TRUE.
if p < alpha, what do we do?
Reject H0
if p > alpha, what do we do?
Retain H0
Confidence interval definition
If we repeat our experiment many times, 95% of the time, our CI will contain the population mean.
As alpha decrease, what happens to our confidence interval?
CI becomes larger
Is the sample size increase, our CI becomes wider of narrower?
Narrower (more precise).
Can F be negative?
No.
What are the 3 measures of central tendency?
Mean, median, mode
What’s the problem with the mean?
Very vulnerable to extreme values (outliers).
What’s good with the median
Less affected by extreme scores.
What’s the mode and its positive aspects?
Mode = The value that occurs most frequently.
Pros: not affected by extreme values
Categorical variables (2)
Nominal (objects, unordered), Ordinal (degree, age)
Continuous variables (2)
Interval (T°), Ratio (time, years…)
Family of tests (definition)
Set of related hypotheses.
Familywise Type I error rate (definition)
Probability of making at least one Type I error in the family of tests if the null hypotheses are true.
Formula for Familywise Type I error rate
= 1 - (1- alpha)**number of family test
Type 1 error definition
Reject the null when it’s true.
