Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
‹date/time›
=
‹fo=
oter›
=
‹#&=
#8250;
=
Click to edit Master title style=
Click to edit Master subtitle st=
yle
‹date/time›
=
‹fo=
oter›
=
‹#&=
#8250;
=
------=_NextPart_01CB1450.96832F30
Content-Location: file:///C:/94649A89/ResearchMethods(Ch3)_files/master136.xml
Content-Transfer-Encoding: quoted-printable
Content-Type: text/xml; charset="utf-8"
![](3D"master135_image001.gif")
![](3D"master135_image002.gif")
![](3D"master135_image003.gif")
Research
Methods in Psychology
(Pp 20-24)
(Pp 20-24)
![](3D"master136_image004.gif")
Principles of Experimental
Design
![](3D"slide0001_image005.gif")
![](3D"slide0001_image007.gif")
Overview of the Scientific Method=
![](3D"slide0004_image011.jpg")
Research Groups
p<=
b>Experimental
Group
p Group of subjects exp=
osed
to the independent variable
p<=
b>Control
Group
=
p Subjects not exposed =
to
some or all of the independent variable; used to compar=
e with
the experimental group.
Experimental Variables
Independent Variable (IV)
pis the variable being manipulated a=
nd controlled
by the experimenter so that the effect can be measured.
(Cause)
Dependent Variable (DV)
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.15%;top:.3em;font-family:Wingdings;font-size:75%'>qis
the behavior to be measured that is “dependent” =
on
the independent variable. (Effect)
Hypothesis
Research Hypothesis
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.07%;top:.3em;font-family:Wingdings;font-size:75%'>qIs
a clear, concise prediction of what is expected to be demonstra=
ted
in the experiment.
Null Hypothesis
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.19%;top:.3em;font-family:Wingdings;font-size:75%'>qStates
that no significant difference is expected to be found bet=
ween
the results as represented by the dependent variable and
the general population is due only to a random variable or due=
to
chance.
![](3D"slide0002_image013.jpg")
=
Types of
Experimental Hypothesis
pOne-tailed (Directional) Hypothesis:
predicts the direct effect o=
f the
Independent Varia=
ble
has on the Dependent Variable. For
example: an
authority figure observing a practiced
behavior will reduce performance.
<=
span
style=3D'font-size:86%;color:#999900;mso-color-index:3'>qTwo-tailed
(Non-Directional) Hypothesis: =
predicts that the Independent
Variable will have an ef=
fect
(of some sort) on the Dependent Variable,
but the direction is not specified. For example: an authority figure observing a practiced behavior will have an affect on performance.
Typ=
es of
Errors in Statistical Testing
=
pType
I Error
we
may reject the null hypothesis if favor of the experimental hypothesis even though the find=
ings
are actually du=
e to
chance; the probability of this happening is given by the level of statistical significan=
ce
selected.
Type II =
Error
we may retain the null hypothesis even though the <=
span
style=3D'position:absolute;top:63.74%;left:15.91%;width:69.85%;height:4.25=
%'>experimental hypothesis is actually correct.=
Research
Methods in Psychology
(Pp 25-32)
(Pp 25-32)
Designing
Experiments
nControl Variables
n are variables that are controlled in order to secure
the independent varia=
ble.
These are controlled to reduce the possible
impact these variables may have on the dependent variable.
Controlling too many variables may reduce external validity; but may produce higher internal validity.
nConfounding Variables
n is any circumstance or change that occurs as you change the independent variable; these variables c=
an unintentionally confound the dependent variable. T=
hese need to be limited or controlled as much as possib=
le
nCounterbalancing
n an attempt to control confounding variables by distributing an equal amount of an confounding var=
iable to each level of the independent variable.<=
span
style=3D'font-size:83%;mso-special-format:lastCR;display:none'>
=
Experimental Validity<=
/span>
External Validity
To what degree the results o=
f an
experiment can be gener=
alized
to the population; how well a causal relationship
can be generalized across populations, environments, and time.
Internal Validity
Refers to whether the manipu=
lated
change in the indep=
endent
variable caused the change in the dependent variable or whether something else caused the chan=
ge.
This can also be prod=
uced
by systematic errors or proced=
ural errors done by
the experimenter.
Ecological Validity<=
span
style=3D'font-size:71%;color:#CCCC66;mso-color-index:5;display:none'>
=
To what extent an experiment
realistically represents a “real
life” situation.
Types of Experimental Desig=
n
pIndependent Sampl=
es
(Between Group Designs) Some subjects are only exposed
to part of the independent variable. Utilizes control groups.&#=
13;
=
p
pRepeated Measures
(Within Group Designs) All subjects are exposed to all <=
/span>parts
of the independent variable
=
p
pMatched Pairs
Same kinds of subjects are assigned to each level of the independent
variable.
Independent Samples/Between Groups
![3D"MPj03143710000[1]"](3D"slide0019_image015.gif")
![](3D"slide0019_image017.gif")
![](3D"slide0019_image018.gif")
![](3D"slide0019_image019.gif")
Recruit a group of participants
Divide them into two
This group does the experimental task with the IV set for condition =
1
This group does the experimental task with the IV set for condition 2=
Measure the DV for each group
=
Compare
the results for the two groups
Group
1 Group 2
![](3D"slide0006_image021.jpg")
Independent
Samples/Between Groups
=
pWhat sorts of problem are likely to arise from using two different groups of people?
<=
span
style=3D'visibility:hidden'>p=
Problem:
=
The natu=
ral
variation between the individuals (participant var=
iables) in =
each gro=
up may
affect the DV measurements, making it look as if =
the IV
has had an effect when it actually hasn’t=
Control:
=
After the
subjects have been recruited, they =
should b=
e ra=
ndomly =
as=
signed =
to their
groups. =
This sho=
uld
ensure the groups are similar, on =
average.=
Repeated Measures/Within Groups
![3D"MPj03143710000[1]"](3D"slide0020_image023.gif")
![](3D"slide0020_image024.gif")
Recruit a group of participants
Condition 1 Condition 2
The group does the experimental task with the IV set for condition =
1
The group repeats the experimental task wit=
h the IV set for condition =
2
![](3D"slide0020_image025.gif")
=
Compare
the results for the two conditions
Repeated
Measures/Within Groups
=
pWhat sorts of problem are likely to arise from using the same group of people =
twice?
<=
span
style=3D'visibility:hidden'>p=
Problem:
<=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>Doing bo=
th
conditions may (1) give Subjects practice on the task;=
(2) <=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>make the=
m bored
or tired; (3) allow them to work out <=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>the aim =
of the
study, all of which might affect the DV <=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>measurem=
ent. (4) Reuse <=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>of stimu=
li is
not possible.
Control:
<=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>Divide t=
he
Subjects into two groups.
Half does condition A first, then <=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>condition
B. The rest do <=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>conditio=
n B then
condition A. DV measur=
ements
for the conditions A and B are then compared <=
span
lang=3DEN-GB style=3D'mso-hansi-font-family:Arial;font-size:133%'>(=
co=
unterbalancing).
Matched Pairs=
![3D"MPj03143710000[1]"](3D"slide0021_image027.gif")
![3Dpersonye=](3D"slide0021_image029.gif")
=
![3Dpersongr=](3D"slide0021_image031.gif")
=
![3Dpersongr=](3D"slide0021_image033.gif")
![3Dpersonred](3D"slide0021_image035.gif")
<=
![endif]>![3Dpersonye=](3D"slide0021_image036.gif")
![3Dpersong=](3D"slide0021_image037.gif")
![3Dpersong=](3D"slide0021_image038.gif")
![3Dpersonr=](3D"slide0021_image039.gif")
=
Recruit a group of participants
Find out what sorts of people you have in the group
Recruit another group that matches them one for one
Condition 1 Condition 2
![](3D"slide0021_image040.gif")
=
Compare
the d=
ifferences between the matched =
pairs
Treat the experiment as independent measures
Matched Pairs
=
pWhat sorts of problem are likely to arise from using matched pairs of PPs?
<=
span
style=3D'visibility:hidden'>p=
Problem:
=
Several
problems: (1) time consuming; (2) an =
exact ma=
tch is
rarely possible; (3) if one PP =
drops ou=
t you
lose 2 PPs’ data.
Control:
=
Members =
of each
pair should be randomly assigned to
conditions. =
However,=
this
does not solve all these problems.
Single vs. Double Blind
Studies
![](3D"slide0009_image042.jpg")
Types of Studies
Single
Blind Study
If the subjects don’t know if they are in the
experimental group or not, this is a single blind design (ie.,
subjects do not know if they are given a placebo or the real
drug)
=
p
pDouble
Blind Study
If both subjects and experimenter are blind
(they do not know who is in the experimental group), thi=
s is
known as a double-blind design.
Systematic
and Random Errors
pSystematic errors=
are
made by the researcher
<=
span
style=3D'visibility:hidden'>p
p<=
/span>Random
errors are characteristics subjects bring with them to the stu=
dyTypes=
of
Systematic and Random Errors
(Goodwin, 1998;
Schumacher and McMillan, 1984)
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>pHistory
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>pSelection
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>pMortality
=
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>pMaturation
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>pDiffusion
of Treatment
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>pResponse
Style
=
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>pOrder
Effect
=
<=
span
style=3D'mso-special-format:bullet;color:#666600;mso-color-index:2;positio=
n:
absolute;left:-4.43%;top:.3em;font-family:Wingdings;font-size:75%'>qHawthorne
Effect
Placebo Effect
=
Filler Activity
=
q
Types=
of
Systematic and Random Errors
(Goodwin, 1998;
Schumacher and McMillan, 1984)
pExperimenter Bias
pInstrumentation
<=
/div>
pConstancy of Condition
<=
/span>
pTesting and Progressive Errors
Experimental
Ethics
&=
#13;
=
p
p
&=
#13;
=
p
&=
#13;
&=
#13;
pAPA requirements – “first do no
harm”
pInformed Consent
pDebriefing Process
pIB Ethical Guidelines
![](3D"slide0030_image044.jpg")
Research
Methods in Psychology
(Pp 32-39)
(Pp 32-39)
Choosing Participants<=
/div>
------=_NextPart_01CB1450.96832F30
Content-Location: file:///C:/94649A89/ResearchMethods(Ch3)_files/slide0040.htm
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html; charset="us-ascii"
pRandom Sample:=
selection of subjects that is randomly selected from a larg=
er population
based on an unbiased assignment or selection of subjects&#=
13;pOppo=
rtunity
Sample: used when random sampling is impossible
pRan=
dom
Allocation: subjects are randomly assigned to a control group =
or =
experimental
group or to levels of the independent variable.
Ways to Collect Data<=
/div>
------=_NextPart_01CB1450.96832F30
Content-Location: file:///C:/94649A89/ResearchMethods(Ch3)_files/master01.htm
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html; charset="us-ascii"
p Based
on your hypotheses, choose a method to collect data which can
calculate means (averages) or proportions (percents)
<=
span
class=3DBB style=3D'position:absolute;left:-4.0%;top:.3em'>pDevelop=
a task
for subjects to complete
<=
span
class=3DBB style=3D'position:absolute;left:-4.43%;top:.3em'>pUse an
observation grid
pCare=
fully
construct a questionnaire or use a standardized psychometric
test‹header›
‹date/time&=
#8250;
Click to edit Master text styles
Second level
Third level
Fourth level
Fifth level
‹footer›
‹#›
=
=
=
15
=
=
=
17
=
=
=
19
' + ENDSHOW_MESG + '
![](3D"notes_flag.gif")
No=
tes![](3D"buttons.gif")
![](3D"slide0001_image006.png"){kind=link}
![](3D"slide0001_image008.jpg"){kind=link}
![](3D"slide0001_image009.jpg"){kind=link}
![](3D"slide0004_image010.png"){kind=link}
![](3D"slide0002_image012.jpg"){kind=link}
![](3D"slide0019_image014.jpg"){kind=link}
![](3D"slide0019_image016.png"){kind=link}
![](3D"slide0006_image020.jpg"){kind=link}
![](3D"slide0020_image022.jpg"){kind=link}
![](3D"slide0021_image026.jpg"){kind=link}
![](3D"slide0021_image028.png"){kind=link}
![](3D"slide0021_image030.png"){kind=link}
![](3D"slide0021_image032.png"){kind=link}
![](3D"slide0021_image034.png"){kind=link}
![](3D"slide0009_image041.png"){kind=link}
![](3D"slide0030_image043.jpg"){kind=link}