Transcript Results PowerPoint

PSYCHOLOGY IA
THE RESULTS
RATIONALE/PURPOSE
•
The results section is where you
report the results that you have
found from your experiment.
•
You must report exactly what
you found, however you must
not explain why you found it
here, this goes in the
discussion.
ASSESSMENT CRITERIA
HL Criterion E - 2 marks
Results are clearly stated and accurate
and reflect the hypotheses and
research. Appropriate descriptive
statistics are applied to the data and
their use is explained. The graph of
results is accurate, clear and directly
relevant to the study. Results are
presented in both words and tabular
form.
HL Criterion F – 3 marks
An appropriate inferential statistics test
has been chosen and explicitly
justified. Results of the inferential
statistical test are accurately stated.
The null hypothesis has been accepted
or rejected appropriately according to
the results of the statistical test. A
statement of statistical significance is
appropriate and clear.
THE ESSENTIALS
•
•
•
•
•
The essential elements of your
results section are as follows:
Descriptive statistics (justification
of and explanation of)
The results table (of descriptive
statistics)
Graph (of descriptive statistics)
Inferential statistics (justification of
and interpretation of)
DESCRIPTIVE STATISTICS
•
Your project must contain two
forms of descriptive statistics
1) measures of central tendency
- (i.e. mean/ median/ mode)
• 2) measure of dispersion
- (i.e. standard deviation)
•
MEASURES OF CENTRAL
TENDENCY


This includes the mean, median
and mode.
You must first of all justify why
you have chosen to use the
mean, median or mode, i.e.
what are the strengths and
weaknesses of them? Which is
the most appropriate? Why?
MEASURES OF CENTRAL
TENDENCY

Then, you must include a table
of the calculated values of
mean, median or mode.

Do NOT include your workings
out here, they go in the
appendix (see ‘additional
important resources’). You can
use a calculator.
MEASURES OF CENTRAL
TENDENCY
Then, you must explain what the
calculated values show. This must
be detailed, i.e. Use actual data in
your description. Are there any
obvious outliers in the results
table which may have distorted
the mean? Etc.
MEASURE OF DISPERSION


The appropriate measure of
dispersion for interval/ratio data is
the standard deviation. This shows
how spread out the data are from the
mean.
You must first justify why you have
chosen this. Hint! Look in the Course
Companion p398 for the conditions
under which the SD is used.
MEASURE OF DISPERSION

Then, you must include the
calculated value of the standard
deviation for conditions A and B in a
table. Put it in the same table as you
did for Mean/median or mode.

Do NOT include your workings out
here, they go in the appendix You
can use a calculator or excel.
MEASURE OF DISPERSION

Then, you must explain what
the calculated value shows.

This must be detailed, i.e. Which
SD is larger? What does this
mean?
THE RESULTS TABLE
•
You must include a table of
results. Not raw data, this goes
in the appendices.
•
Make sure you give a clear and
detailed title.
GRAPH



You must display your results in
an appropriate graph. Only one
graph is required.
Make sure you give a clear and
detailed title.
Label axes correctly
INFERENTIAL STATISTICS





The appropriate measure of statistical
significance depends on:
1) Test of difference or correlation
2) Level of data
3) Whether the data is related (repeated
measures) or unrelated (independent
measures)
It will either be a Wilcoxon signed ranks
test (repeated measures) or a Mann
Whitney U test (independent measures)
depending on whether you have used
repeated measures or independent groups
respectively.
INFERENTIAL STATISTICS

You must first justify why you have chosen
your statistical test.

You need to explain the conditions under which
the test is used (Hint! This is explained on the
previous slide!). Data is at least at the ordinal
level.

Working out can be done on the computer but
a printout must be provided. See booklet for
more detailed instructions.
INFERENTIAL STATISTICS




There are different levels of
significance.
The p<0.05 level is used as it is the
conventional level of significance
that is accepted by psychologists.
P<0.05 means that there is a 95%
chance that the results occur due to
an effect of the IV on the DV, and a
5% chance that the results occur due
to random variability.
The ‘p’ stands for ‘probability’
INFERENTIAL STATISTICS

You must then interpret your stats
test results using a ‘numerical
statement of results’. This should
include:

Calculated value (T (wilcoxon) OR U (Mann
Whitney)
Critical value
No. of Participants (N)
One tailed hypothesis
Level of significance
Whether the null hypothesis is retained or rejected.





NUMERICAL STATEMENT
OF RESULTS EXAMPLE
The calculated value from the
Wilcoxon/Mann Whitney test was __. This
is less than/ more than the critical value of
__ for a directional/ non-directional (1
tailed/ 2 tailed) hypothesis when N = __.
This shows that the risk that this difference
occurred by chance is more than/ less than
0.05 or 5%. The result is therefore
significant/ non-significant and the null
hypothesis can be rejected/ retained.