PARAMETRIC Pt.1 Flashcards
2 types of hypothesis Tests
o Parametric Test
o Nonparametric test
PARAMETRIC TESTS
- Assumed distribution is NORMAL
- Assumed variance is homogenous
- Typical data is/are ratio or interval
- Data set relationships are usually independent
- Usual central measure is mean
- Benefits: can draw more conclusions
- Assumed distribution is NORMAL
- Assumed variance is homogenous
- Typical data is/are ratio or interval
- Data set relationships are usually independent
- Usual central measure is mean
- Benefits: can draw more conclusions
PARAMETRIC TESTS
Analysis where we compare against our historical or global value
o t-test
o z-test
ONE SAMPLE: PARAMETRIC TEST
Analysis where we compare a control group versus test group
TWO SAMPLE: PARAMETRIC TEST
2 samples have no relationship
with each other
Independent Samples
Samples are dependent to each
other
Paired Samples
Normal
Homogeneous
Ratio or Interval
Independent
Mean
Can draw more conclusions
Parametric
Any
Homogenous and Heterogeneous
Ordinal or Nominal
Any
Median
Simplicity: Less affected by outliers
Non-parametric
Samples must be independent of each other. There can be no relationship between the subjects in each sample
Z TEST FOR TWO MEANS
Z TEST FOR TWO MEANS: The standard deviations of both populations must be
known
Z TEST FOR TWO MEANS: Samples must be _______________ of each other. There can be no relationship between the subjects in each sample
independent
Z TEST FOR TWO MEANS: If the sample sizes are _____ than 30, the population must be normally or approximately normally distributed.
less
observed value –
mean difference
expected value –
hypothesis