Documente Academic
Documente Profesional
Documente Cultură
Comparison of means
Objectives
At the end of this session you will be able to:
Plot a histogram to assess the normality of
the distribution of data
Know when to apply a parametric or nonparametric test
Apply the parametric Independent T Test
Apply the non-parametric Mann-Whitney test
Aims
T-tests
Dependent (aka paired, matched)
Independent
Interpretation
Reporting results
Calculating an Effect Size
T-tests as a GLM
Experiments
The simplest form of experiment that can be
done is one with only one independent
variable that is manipulated in only two ways
and only one outcome is measured.
More often than not the manipulation of the
independent variable involves having an
experimental condition and a control.
E.g., Is the movie Scream 2 scarier than the original
Scream? We could measure heart rates (which
indicate anxiety) during both films and compare
them.
T-test
Dependent t-test
Compares two means based on related data.
E.g., Data from the same people measured at
different times.
Data from matched samples.
Independent t-test
Compares two means based on independent data
E.g., data from different groups of people
Significance testing
Testing the significance of Pearsons correlation
coefficient
Testing the significance of b in regression.
observed difference
expected difference
between sample
between population means
means
(if null hypothesis is true)
t
Realistic distributions
Most data sets found in nature are not
normally distributed
How can you tell and what can you do if
they are not?
6
9
3
4
1
Count
4
0
7
2.00
4.00
6.00
8.00
10.00
Child_age
7
10
6
6
5
8
12
6
7
5
6
7
8
The distribution
is approximately
symmetrical
Independent t-test
Wilcoxon Signed-Rank Test
Robust Tests:
Bootstrapping
Trimmed means
Example
Is arachnophobia (fear of spiders) specific to
real spiders or is a picture enough?
Participants
12 spider phobic individuals
Manipulation
Each participant was exposed to a real spider and a
picture of the same spider at two points in time.
(This is therefore a Dependent t-test)
Outcome
Anxiety
Significance
p< .05 therefore
the results are
significant
Count
Mean
Standard Error
Median
Mode
Standard Deviation
2.708013
Sample Variance
7.333333
Kurtosis
0.888107
Skewness
-0.15826
Range
Minimum
10
Maximum
12
Sum
Count
5
8
2.00
4.00
6.00
Child_age
8.00
10.00
12
6
7
5
6
7
8
6
0.57735
12
132
22
Mann Whitney
Mann-Whitney test
In order to apply the Mann-Whitney test, the
raw data from samples A andB must first be
combined into a set of N=na+nb elements,
These are then ranked from lowest to highest,
including tied rank values where appropriate.
These rankings are then re-sorted into the two
separate samples.
If your data have not yet been rank-ordered in
this fashion, SPSS allows you to import raw
data from a spreadsheet.
Scores
These test scores were
obtained by students taught
by one method in 2003 and
by a new web-based method
in 2004
The research question is:
Is there a difference in the
mean score between the
years?