Date post: | 17-Jul-2015 |
Category: |
Education |
Upload: | aleem-ashraf |
View: | 200 times |
Download: | 1 times |
Aleem Ashraf
Department of Psychology
University of Sindh, Jamshoro.
Experimental Design
• Assignment of subjects to different groups or conditions
• Manipulation of one or more independent variables by experimenter (IV)
• Measurement of effects on one or more dependent variables (DV)
• Control of other variables
• In order to make causal judgments about what causes variation in dependent variable
Ale
em
Ashra
f
2
Manipulation
• Applying certain treatments (levels of independent
variables) to the different conditions
Control
• The researcher’s attempts to make the groups as
similar as possible on most variables other than
independent variable.
• Direct control e.g. focus on certain characteristics
by exclusion
• Age, intelligence, socioeconomic class, previous
experience, good school system etc.
• Most variables cannot be controlled in the direct
fashion, for that randomization is used
Ale
em
Ashra
f
3
Control by Randomization
• Randomized selection of participants from
known population. Required but not often
done in the experiments. Only the willing
volunteers are selected
• Done to give external validity =
generalizability
• Randomized allocation of recruited
participants to different conditions (control or
experimental group)
• Done to give internal validity = the effects in
the DV are really due to IV?
Ale
em
Ashra
f
4
Randomization
• Each subject in the study has an equal
chance of being assigned to the control or
experimental group
• It evenly distributes the important
variables across groups that may effect
the dependent variable
Research Process
R
Random
Selection
R
Random
Assignment
O1
Observation
1
X
Treatment
O2
Observation
2
Ale
em
Ashra
f
5
The Experimental Process
• Flow chart of a classic randomized clinical
trial
Sample is
selected from
the population
Baseline data
are collected
Subjects are
randomized
Experimental
group
Control group
Post test data
are collected
Post test data
are collected
Ale
em
Ashra
f
6
Example
• Does noise effect the amount that people
can remember when learning?
• Participants are selected and randomly
assigned to three conditions:
• Control group = no noise at all
• Treatment group 1 = loud unpredictable
noise
• Treatment group 2 = soft, rhythmic music
• Respondents are given a standard list of
items to remember in a given time
Ale
em
Ashra
f
7
Illustration of the Design
Control group
No noise at all
Treatment group 1
Loud unpredictable
noise
Treatment group 2
Soft, rhythmic music
Ale
em
Ashra
f
8
Procedure
• Randomly allocate participants to the three
groups
• All groups perform the same task (memorize
the list) under different conditions i.e. no noise,
loud noise, soft music
• All participants are given the same test after
the treatment
• Compare the mean score of recall for each
group
• Use statistical tests i.e. t-test or ANOVA to see
if the differences in the mean are significant
Ale
em
Ashra
f
9
Graphical representation of the three groups
0
2
4
6
8
10
12
14
16
18
No noise Loud noise Soft music
Ale
em
Ashra
f
10
Internal Validity
• Internal validity is the degree to which
differences in performance on a
dependent variable can be attributed
clearly and unambiguously to an effect of
an independent variable, as opposed to
some other uncontrolled variable.
Ale
em
Ashra
f
11
Threats to Internal & External Validity
• Is the researcher’s claim about the cause
and effect correct?
• Are the changes in the independent
variable indeed responsible for the
observed variation in the dependent
variable?
• Might the variation in the dependent
variable be attributable to other causes?
Ale
em
Ashra
f
12
Importance of Internal Validity
• We often conduct research in order to determine
cause-and-effect relationships.
• Can we conclude that changes in the independent
variable caused the observed changes in the
dependent variable?
• Is the evidence for such a conclusion good or
poor?
• If a study shows a high degree of internal validity
then we can conclude we have strong evidence of
causality.
• If a study has low internal validity, then we must
conclude we have little or no evidence of causality.
Ale
em
Ashra
f
13
Variables & Internal Validity
• Extraneous variables are variables that may
compete with the independent variable in
explaining the outcome of a study.
• A confounding variable is an extraneous
variable that does indeed influence the dependent
variable.
• A confounding variable systematically varies or
influences the independent variable and also
influences the dependent variable.
• Researchers must always worry about extraneous
variables when they make conclusions about
cause and effect.
Ale
em
Ashra
f
14
Three Criteria of Causality
• Association: The cause and effect must
be associated with each other (Correlation)
• Direction of influence: The effect appears
after the cause. (The effect must respond
to the cause, not vice versa).
• Nonspuriousness: There should be good
reason to believe that there are no hidden
factors that could have created an
accidental relationship between the
variables.
Ale
em
Ashra
f
15
Threats to Internal Validity
1. History
2. Maturation
3. Testing
4. Instrumentation
5. Statistical regression
6. Differential selection of participants
7. Mortality
8. Design contamination
9. Compensatory rivalry
10.Resentful demoralization
Ale
em
Ashra
f
16
History
• Unexpected events occur between the
pre- and posttest, affecting the dependent
variable.
• Example: In a short experiment designed
to investigate the effect of computer-
based instruction, a participant missed
some instruction because of a power
failure at the school.
Ale
em
Ashra
f
17
Maturation
• Changes occur in the participants, from
growing older, wiser, more experienced
etc. during the study.
• Example: The performance of first
graders in a learning experiment begins
decreasing after 45 minutes because of
fatigue.
Ale
em
Ashra
f
18
Testing
• Taking a pretest alters the result of the posttest.
• A pre-test may sensitize participant in unanticipated ways and their performance on the post-test may be due to the pre-test, not to the treatment.
• Example: In an experiment in which performance on a logical reasoning test is the dependent variable, a pre-test cues the subjects about the post-test.
Ale
em
Ashra
f
19
Instrumentation
• Did any change occur during the study in
the way the dependent variable was
measured?
• Example: Two examiners for an
instructional experiment administered the
post-test with different instructions and
procedures.
Ale
em
Ashra
f
20
Statistical Regression
• Extremely high or extremely low scorers
tend to regress to the mean on retesting.
• Example: In an experiment involving
reading instruction, subjects grouped
because of poor pre-test reading scores
show considerably greater gain than do
the groups who scored average and high
on the pre-test.
Ale
em
Ashra
f
21
Differential Selection of Participants
• Participants in the experimental and
control groups have different
characteristics that affect the dependent
variable differently.
• Random assignment of participants into
different groups usually resolve this
threat.
Ale
em
Ashra
f
22
Mortality
• Different participants drop out of the study
in different numbers, altering the
composition of the treatment groups.
• Example: In a health experiment
designed to determine the effect of
various exercises, those subjects who
find the exercise most difficult stop
participating.
Ale
em
Ashra
f
23
Design Contamination
• Did the comparison group know (or find
out) about the experimental group? Did
either group have a reason to want to
make the research succeed or fail? Often,
investigators must interview subjects after
the experiment concludes in order to find
out if design contamination occurred.
• Example: In an expectancy experiment,
students in the experimental and
comparison groups “compare notes” about
what they were told to expect.
Ale
em
Ashra
f
24
Compensatory Rivalry
• When subjects in some treatments receive goods or services believed to be desirable and
this becomes known to subjects in other groups,
social competition may motivate the latter to attempt to reverse or reduce the anticipated
effects of the desirable treatment levels.
• Known as “John Henry” effect in honor of the steel driver who, upon learning that his output
was being compared with that of a steam drill, worked so hard that he outperformed the drill
and died of overexertion.
Ale
em
Ashra
f
25
Resentful Demoralization
• If subjects learn that their group receives
less desirable goods or services, they
may experience feelings of resentment
and demoralization.
• Their response may be to perform at an
abnormally low level, thereby increasing
the magnitude of the difference between
their performance and that of groups that
receive the desirable goods or services.
Ale
em
Ashra
f
26
External Validity
• External validity refers to the extent to
which findings from a research study can
be generalized to individuals, settings,
and conditions beyond the scope of the
specific study.
Ale
em
Ashra
f
27
Threats to External Validity
1. Pretest–treatment interaction
2. Selection–treatment interaction
3. Multiple-treatment interference
4. Specificity of variables
5. Treatment diffusion
6. Experimenter effects
7. Reactive arrangements
Ale
em
Ashra
f
28
Pretest–treatment Interaction
• The pretest sensitizes participants to
aspects of the treatment and thus
influences posttest scores.
• Example: In a physical performance
experiment, the pre-test clues the
subjects to respond in a certain way to
the experimental treatment that would not
be the case if there were no pre-test.
Ale
em
Ashra
f
29
Selection–treatment Interaction
• An effect of some selection factor of intact
groups interacting with the experimental
treatment that would not be the case if the
groups were randomly selected.
• Example: The results of an experiment in
which teaching method is the
experimental treatment, used with a class
of low achievers, do not generalize to
heterogeneous ability students.
Ale
em
Ashra
f
30
Multiple-treatment Interference
• When participants receive more than one treatment, the effect of prior treatment can affect or interact with later treatment, limiting generalizability.
• Example: In a drug experiment the same animals are administered four different drug doses in some sequence. The effects of the second through fourth doses cannot be separated from the possible delayed effects of preceding doses.
Ale
em
Ashra
f
31
Specificity of Variables
• Poorly operationalized variables make it
difficult to identify the setting and
procedures to which the variables can be
generalized.
Ale
em
Ashra
f
32
Experimenter Effects
• Conscious or unconscious actions of the
researchers affect participants’
performance and responses.
• Example: Differences in which the
different researchers give different
instructions to the different groups.
• Age, gender, attractiveness also make a
difference
• Solution: Double blind experiments
Ale
em
Ashra
f
33
Double-blind Experiment
• The experimental group receives the real
drug, the control group gets the placebo
• There are two experimenters but the one
who administers the drug is unaware if he
is distributing the placebo or the drug
• Neither the experimenter nor the subjects
know if they getting the drug or placebo
• Eliminates experimenter bias and placebo
effects.
Ale
em
Ashra
f
34
Reactive Arrangements
• The fact of being in a study affects
participants so that they act in ways
different from their normal behavior. The
Hawthorne and John Henry effects are
reactive responses to being in a study.
Ale
em
Ashra
f
35
Acknowledgements
• Prof. Graham R. Gibbs, University of
Huddersfield.
• Prof. Robert S Michael, Indiana University. (For giving baseline ideas for the slides)
Reference
• Lorraine R. Gay, 2011. Educational
Research: Competencies for Analysis and
Applications (10th Edition). 10 Edition.
Pearson.
• John J. Shaughnessy, 2011. Research
Methods In Psychology, 9th Edition. 9th
Edition. McGraw-Hill.
Ale
em
Ashra
f
36
Ale
em
Ashra
f
37