Comparing means part 2
Part 1
Paired samples t test
A version of the t test for within subjects designs
Aka
o Related samples t test
o Paired means t test
o Within subjects t test
o Repeated measures t test
More powerful than a between subjects t test (i.e. we are more likely to find a
significant difference/effect
Equation
o As with any t test: size of effect divided by the standard error of the effect
o T = mean of the differences for the pairs, divided by the standard error of the
differences of the pairs
Assumptions
o Data must be on an interval or ratio scale
o The sample of pairs is a random sample from the population
o The difference between scores is normally distributed (not the raw score)
Example
o Research question
o Do we spend more time using social media on the computer, or on our
mobile phones?
o Null hypothesis: there will be no difference in the amount of time spent using
social media.
o Alternative hypothesis: people will spend more time using social media on
their phones than on the computer
Part 2
Paired samples t test – output
o Paired samples statistics – descriptive and dispersive data describing the two
conditions of analysis. Nice, basic descriptive data – used for interpreting significance
of t test itself
o Paired samples t test – 3 important things to work on in order to figure out whether t
test has a significant difference. (t, df and p value (sig.))
o Knowing how to interpret the basic output of a test in spss is transferable skill across
different statistical software packages e.g. spss to jamovi.
Part 1
Paired samples t test
A version of the t test for within subjects designs
Aka
o Related samples t test
o Paired means t test
o Within subjects t test
o Repeated measures t test
More powerful than a between subjects t test (i.e. we are more likely to find a
significant difference/effect
Equation
o As with any t test: size of effect divided by the standard error of the effect
o T = mean of the differences for the pairs, divided by the standard error of the
differences of the pairs
Assumptions
o Data must be on an interval or ratio scale
o The sample of pairs is a random sample from the population
o The difference between scores is normally distributed (not the raw score)
Example
o Research question
o Do we spend more time using social media on the computer, or on our
mobile phones?
o Null hypothesis: there will be no difference in the amount of time spent using
social media.
o Alternative hypothesis: people will spend more time using social media on
their phones than on the computer
Part 2
Paired samples t test – output
o Paired samples statistics – descriptive and dispersive data describing the two
conditions of analysis. Nice, basic descriptive data – used for interpreting significance
of t test itself
o Paired samples t test – 3 important things to work on in order to figure out whether t
test has a significant difference. (t, df and p value (sig.))
o Knowing how to interpret the basic output of a test in spss is transferable skill across
different statistical software packages e.g. spss to jamovi.
