what do parametric tests assume?
- data is normally distributed
- groups have equal variance
- interval/ratio data
homogeneity of variance
equal variance
when can parametric tests be abused?
when data does not meet the underlying assumptions of tests
difference between non-parametric tests
- they do not make assumptions about population distributions
- less powerful and flexible than parametric tests
when is the kolmogrov-smirnov test used?
to determine the likelihood of data belonging to a normal distribution
k-s test: if the observed value is less than 0.05…
- this is significant
- not normally distributed
- parametric test cannot be used
k-s test: if the observed value is above 0.05…
- this is not significant
- data is normally distributed
- parametric test may be used
when is the levenes test used?
to test for homogeneity of variance, and determine data sets are from the same population
levenes test: if the probability is below 0.05…
- this is significant
- accept experimental hypothesis
- conclude there is not equal variance
levenes test: if the probability is above 0.05…
- not significant
- can assume there is equal variance
what is needed to perform a parametric test?
- normally distributed data
- equal variance between samples
report writing
- title
- abstract
- introduction
- method
- results
- discussion
- references
- appendices
title
provides a succinct statement of the core purpose of the research
abstract
short overview of the research (150-250 words) which states key variables, participants, and results
introduction
must introduce the general topic, theories relating to the topic, relevant research, and the specific hypothesis
method
specific information regarding how the study was conducted
- participants
- design
- materials
- procedure
results
details of all data produced, and relevant descriptive and inferential statistics
references
provide information about the source of all the research
appendices
extra information, e.g., raw data, consent form, instructions
