What does science do The core features of a research study Overall Research approaches Dr David J McKirnan 2015 The University of Illinois Chicago McKirnanUIC gmailcom Do not use or reproduce without permission ID: 488035
Download Presentation The PPT/PDF document "Introductory lectures 2: The Role and St..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Introductory lectures 2: The Role and Structure of Science
What does science do?The core features of a research study.Overall Research approaches.
© Dr
. David J.
McKirnan, 2015
The University of Illinois ChicagoMcKirnanUIC@gmail.comDo not use or reproduce without permissionSlide2
Introductory lectures 2: The Role and Structure of Science
What does science do?The core features of a research study.Overall Research approaches.
Slide3
What does science do?
What does Science do?Describe the worldMade predictions (hypotheses)Develop & test
theoriesApply findings or theories
Section OverviewSlide4
1. Describing the world
Simple observation: what goes on “in the wild”.Anthropology, sociology, ethnography, ethology….
Many quantitative studies begin with qualitative observations.
1. Qualitative Descriptio
n
Typically structured:field notes, checklists,
behavioral counts...Slide5
Describing the world
1. Qualitative Description
Focus groups, structured observations…
… valuable for
formulating hypotheses.
DescriptionPredictionTheoryApplicationSlide6
Science and description
World Music: A Retrospect Across The Globe, Venus Umesh. Link.Mathematics Describing the Real World: Pre-calculus and Trigonometry. Bruce
H. Edwards, link here.
D
escription … key building block in all areas…
DescriptionPredictionTheoryApplicationSlide7
Describing
the world Tracking behavior or systems using numbers
Useful for both developing and testing hypotheses.
Surveys, polls…
Archival data - uniform crime reports.
Physical observationsFrom: Climate Site; How do climate models work? Link.
2
. Quantitative Descriptio
n
Description
Prediction
Theory
ApplicationSlide8
Examples of Descriptive Research
Taxonomies: Classification systemsPaleontology; shared v. distinct evolutionary characteristics
Psychology, Psychiatry; behavioral patterns or disorders.
Demographics: ethnicity, socio-economic status, language group…
Description
PredictionTheoryApplicationSlide9
Taxonomies
: Classification systems
Look on the web site
find out your personality ‘type’.
Description
Prediction
Theory
ApplicationSlide10
Examples of Descriptive Research
Where is AIDS most common in the U.S.?
Epidemiology
; Disease / behavior patterns in a population.
http://
filipspagnoli.wordpress.com/2011/11/17/human-rights-maps-160-gun-crime-in-the-u-s-in-2010/Distribution of gun assaults, by state.
Methods:
D
irect measurement
(e.g., surveys, disease reporting)
Standard records
(e.g., medical visits & diagnoses)
Archival data
(e.g., auto accident reports).
Description
Prediction
Theory
ApplicationSlide11
Descriptive research
Paleontology attempts to accurately describe the predecessors of humans to understand evolution
E X A M P L E
Carefully describing specimens and the conditions where they are
found can:
Produce insights into environmental change and evolutionAllow us to test hypotheses & develop theory.Description
Prediction
Theory
ApplicationSlide12
2. Science and Predictions
Practical applications: What
test score best predicts college success
?
Can I predict which employees will steal?
These applications (should) stem from a theory.V
erbal & quantitative aptitude
academic success
A specific personality constellation
criminality…
That is
(or should be…)
supported by empirical
evidence
.Slide13
Science and Predictions
Practical applications:Testing hypotheses:
An hypothesis is a predictionCorrelational research;
I predict
: cultures undergoing rapid economic change will be more prone to irrational beliefs.
Experimental research; Lab: make Experimental group socially uncertain. Control: simple distraction.
I predict
: experimental group more prone toward superstitious beliefs.
Description
Prediction
Theory
ApplicationSlide14
3. Developing & Testing Theory
Identify basic (social, Ψ, physical…) processes…
that are systematically related…
Theory is the
‘bottom line’ of
science
t
hat show
how
or
why
something works.
i.e., what “causes” a phenomenon in the natural world.Slide15
Predictions and theory development
New data have led us to rethink basic theories of human evolution.From a simple progression……to a widening “bush” of parallel species.
E X A M P L E
Description
Prediction
Theory
ApplicationSlide16
Description
PredictionTheoryApplicationPredictions and theory development
As paleontologists rethink the shape of the evolutionary tree..They reconsider the basic
processes that shape human evolution.
E X A M P L ESlide17
Theory: Mediating variables
Mediating Variables: The core building blocks of theory,Basic physical, psychological or social processes,
Tell us how or why things work,H
elp us explain
our observations.Slide18
3. Testing theory: Mediating effects
Theory explains an observation or correlation by invoking a mediating variable.When are superstitions or irrational beliefs most common?A. We describe
a simple empirical effect (correlation):
Social & economic uncertainty
Irrational beliefs
Description
Prediction
Theory
Application
Historically,
superstitions are
prevalent during times of social & economic change.Slide19
3. Testing theory: Mediating effects
Need to feel control
Uncertainty
Irrational beliefs
Description
Prediction
Theory
Application
Uncertainty and irrational beliefs
…
A. We
describe
a simple empirical effect (correlation):
B. How does this work? What
Theory
(
ies
)
may explain this?
✓
“Need for control” may
Mediate
the effect of uncertainty on irrational beliefs.
Social & economic uncertainty “causes”
people to seek a greater sense of personal control;
Need for perceived control can “cause” people to adopt superstitious beliefs.Slide20
3. Testing theory: Mediating effects
Need to feel control
Uncertainty
Irrational beliefs
Description
Prediction
Theory
Application
Uncertainty and irrational beliefs
…
How does this work? What
Theory
(
ies
)
may explain this?
✓
“Need for control” is a core
Mediating Variable
in this theory.
It helps us
explain
how social & economic uncertainty may make people vulnerable to
superstitious beliefs.Slide21
3. Testing theory: Mediating effects
Need to feel control
Uncertainty
Irrational beliefs
Description
Prediction
Theory
Application
Uncertainty and irrational beliefs
…
How does this work? What
Theory
(
ies
)
may explain this?
✓
Lessened Critical thinking
✓
We may expand our theory:
Lessened critical thinking may also help
explain
the effect of uncertainty on beliefs.Slide22
3. Testing theory: Mediating effects
Need to feel control
Uncertainty
Irrational beliefs
Description
Prediction
Theory
Application
Uncertainty and irrational beliefs
…
How does this work? What
Theory
(
ies
)
may explain this?
✓
Lessened Critical thinking
✓
This theory is comprised of
two
Mediating
V
ariables
:
The combination of these variables may well
explain
the effect of uncertainty on beliefs.Slide23
Theory and process
Need to feel control
Uncertainty
Lessened Critical thinking
Irrational beliefs
Description
Prediction
Theory
Application
...identifies basic economic / psychological
processes
;
… is built of
mediating variables
;
…specifies how they may be
related
;
…generates testable
hypotheses.
Theory:
✓
✓Slide24
Theory and process
Need to feel control
Uncertainty
Lessened Critical thinking
Irrational beliefs
Description
Prediction
Theory
Application
...identifies basic economic / psychological
processes
;
… is built of
mediating variables
;
…specifies how they may be
related
;
…generates testable
hypotheses
.
Theory:
I hypothesize that systems that introduce critical thinking in 1
o
and 2
o
school will have a lower prevalence of superstitious beliefs
.
I hypothesize that creating uncertainty in a lab will lessen participants’ ability to question their own assumptions and evaluate evidence
…
An hypothesis tested with a
correlational
design.
An hypothesis tested with an
experimental
design.Slide25
Educational theory:
Being active Cognitive “chunking”Multi-media
attention
4.
Applications of theory
Use theory to design interventions…an intervention study can test a theory
Behavioral or biomedical interventions
compare treatments based on differing theories.
Design of this course
Description
Prediction
Theory
ApplicationSlide26
Theories and interventions
Smoking & social networks: people are influenced by their friends’ smoking
E X A M P L E
I
ntervention study
of
smoking
cessation:
quitting with friends may be more effective.
Description
Prediction
Theory
ApplicationSlide27
Theories and interventions
Theory: social networks smoking.Hypothesis: network-based program would work better.
E X A M P L E
Pointed to a practical intervention approach;
S
upported the theory.
Testing the
hypothesis:
Description
Prediction
Theory
ApplicationSlide28
The values of science & empiricism
Empirical descriptionQualitative vs. quantitative…Developing hypotheses.Predictions
Core feature of a hypothesis;
Force us to clearly test our ideas.
Developing and testing theoryCentral to science; explain how a natural process works.
Applications of theoriesExpand and elaborate a theory.Advances in, e.g., technology, behavioral science
SUMMARYSlide29
Elements of science, review 1
Epidemiology is…A way of classifying people or behaviorsThe study of how frequent a behavior or “condition” is across a population.The identification of basic psychological principles that underlie important behaviorsA core feature of a hypothesis.Slide30
Elements of science, review 2
A Theory is…A way of classifying people or behaviorsThe study of how frequent a behavior or “condition” is across a population.The identification of basic psychological principles that underlie important behaviorsA core feature of an hypothesis.Slide31
Elements of science, review 2
A mediating variable is…A way of describing the world.An observed correlation.A psychological process that is the building block of a theory.A way of testing an hypothesis.Slide32
Elements of science, review 3
A Prediction is…A way of classifying people or behaviorsThe study of how frequent a behavior or “condition” is across a population.The identification of basic psychological principles that underlie important behaviorsA core feature of a hypothesis.Slide33
Elements of science, review 4
A Taxonomy is…A way of classifying people or behaviorsThe study of how frequent a behavior or “condition” is across a population.The identification of basic psychological principles that underlie important behaviorsA core feature of a hypothesis.Slide34
Introduction to science, 5
What does science do?Core features of a research study.Overall Research approaches.
Slide35
Section Overview
The overall “flow” of research
Phenomenon
- Larger
question the research addresses
Theory
-
E
xplanatory
processes & how they are related
Hypothesis
Concrete variables
Specific
prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise analysis
of
outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studiesSlide36
Phenomenon
- Larger
question the research addresses
Theory
-
E
xplanatory
processes & how they are related
Hypothesis
Concrete variables
Specific
prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise analysis
of
outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studies
Research Flow
What needs explaining?
Why is it important
?
How / why do I think it works?
What is my specific prediction?
What concrete evidence or data will I collect?
What was the outcome?
H
ypothesis supported?
What do the results mean for the theory?
What is unanswered?Slide37
Research Flow
Phenomenon
- Larger
question the research addresses
Theory
-
E
xplanatory
processes & how they are related
Hypothesis
Concrete variables
Specific
prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise analysis
of
outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studies
The research flow begins with broad, abstract questions
G
ets more specific & concrete
Then back to a more general discussion.Slide38
Core features of research: Theory.
Phenomenon
- Larger
question the research addresses
Theory
-
E
xplanatory
processes & how they are related
Hypothesis
Concrete variables
Specific
prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise analysis
of
outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studies
Abstract
statement of how
basic processes
relate to each other…
…
how
or
why
the
phenomenon “works.”
Theory:Slide39
B
asic physical or behavioral processes are the building blocks of theories.
Theory
Economic uncertainty
Cognitive style
StressGravityLearning…Slide40
Economic uncertainty
Cognitive styleStressGravityLearning…
They are abstract:
We cannot actually “see” gravity.
We just see what it does (stuff falls).
TheorySlide41
Economic uncertainty
Cognitive styleStressGravityLearning…
Gravity is a
Hypothetical Construct
:
Central to our understanding of nature.Abstract; we cannot measure it directly.
We measure and understand it by observing its effects on the world.
TheorySlide42
TheorySlide43
How do we use theory in research?
Test a theory:
Compare theories:
Extend an established theory:
Apply a theory:
Do
women who feel strong stereotype threat about math actually do worse
?
Which best explains women’s statistics performance: stereotype threat or social role learning?
Can
stereotype threat help us explain athletic as well as academic performance?
Can
I create instructions that relieve stereotype threat for women during
statistics
?Slide44
Core features of research: Hypotheses.
Phenomenon
- Larger
question the research addresses
Theory
- Basic
Ψ
processes we think explain the phenomenon
Hypothesis
Concrete variables
that express the
Ψ
processes
Specific prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise
statistical analysis of study outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studies
A prediction
Derived from / testing the theory
That is potentially falsifiable.
Hypothesis
:Slide45
An hypothesis is a Prediction
Relating variables derived from the theory.Specifies cause and effect.…that is potentially falsifiable (see text for discussion)Can be conceivably / logically shown to be untrueSpecific enough to be tested
HypothesisSlide46
H
ypotheses
are expressed in
control
terms:
IF I make people relaxed
then
their fear and loathing of statistics will decrease.
Measurement studies
do not manipulate anything.
H
ypotheses are expressed as a
relation.
People who are high on a measurement of relaxation will tend to fear statistics less;
I predict a
correlation
between relaxation & fear of stats.
Hypotheses: Measurement v. Experimental studies
Hypothesis
In
experiments
we
manipulate
the
Independent Variable
.Slide47
Core features of research: Methods.
Phenomenon
- Larger
question the research addresses
Theory
- Basic
Ψ
processes we think explain the phenomenon
Hypothesis
Concrete variables
that express the
Ψ
processes
Specific prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise
statistical analysis of study outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studies
We attempt to make each term of the study as objective as possible.
Both measurement & experimental procedures rely on
Operational Definitions
Methods:Slide48
Core element of scientific approachObjective
; designed to separate data from personPublic: multiple people participate in, challenge, or use scientific findingsReplicable: others can repeat or expand the study
T
urn our hypotheses into concrete
variables
That we examine via specific procedures
MethodsSlide49
We turn our hypotheses into variables via an
operation definition.Depression
“vegetative”; sleep, eating
Verbal behavior
Appearance
Suicidality, drug use, work…Survey / questionnaire answers…
What is “depression”?
What
operations
could assess it?
MethodsSlide50
Our
Operation definition determines what a variable “means” in our study.
What is “economic uncertainty”?
The unemployment rate?
Subjective attitude ratings?
What is “Stress”?Heart rate & cortisol levels?Speech patterns?MethodsSlide51
This is the
independent variable To test the hypothesis that stress impairs memory I may
create stress in the lab via…
Shock
.Requiring a difficult performance in front of others.
shock (a physical threat) and performance (a social threat)
Equivalent ‘Stressors”?
Are
MethodsSlide52
I may
measure stress levels in this class via:A questionnaire scaleHeart rate
Anxious behavior, sleep loss, appetite change…
Heart rate
and sleep lossEquivalent measures of Stress?
Are
MethodsSlide53
Operationalizing “Stress”
Does stress lead to illness?
E X A M P L E
“Stress”
“Illness”
Experimental designMeasurement study
Manipulate
the independent variable
Threatening information.
Shock.
Require difficult public performance.
Measure
the predictor
Questionnaire
scale on life events
Assess cardio-vascular reactivity
Measure
the dependent variable
Infection post-exposure
Observed respiratory infections
Measure
the outcome variable
Self-reported respiratory
infections
Morbidity & mortalitySlide54
Any theory must be operationalized to be heuristically useful
Operational definitions real observable world. What does “stress” mean?
How will you measure it?
What does “health” mean?
How will you measure it?
T
hink critically
about what each variable means.
Methods: Why use Operational Definitions?
I think stress makes people less healthySlide55
Some variables are easy to operationalize; e.g., the effect of a drug dose on hypertension.
IV = drug dose1 v. drug dose 2 v. Placebo…DV = blood pressure.Some constructs can only be roughly operationalized.
“Pro-social attitude”, “self-concept”…
Some constructs have diverging operational definitions.
How do you operationally define “stress”?
…learning? Some domains may not be operationalizable.String theory… “Spirituality”? “Happiness”?Behavior?
Self-perception?
Physiological?
Methods: The limits of Operational DefinitionsSlide56
When am I dead?
Middle ages: the soul departs the body – weighs 21 gramsThe name:17th Century: Cordelia’s daughter in King Lear shows no breath on a mirror held to her nose
19th Century: Development of the stethoscope and “
heart death”.
Mid-20th Century: Development of respirators / life support and “brain death”.21
st Century: fMRI images show responsiveness even in some “vegetative” patients E X A M P L E
Physical death
Your body is consigned to the grave
Someone speaks your name the last time
When am I dead?
H
ow do we operationally define “death”?Slide57
Core features of research: Results.
Phenomenon
- Larger
question the research addresses
Theory
- Basic
Ψ
processes we think explain the phenomenon
Hypothesis
Concrete variables
that express the
Ψ
processes
Specific prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise
statistical analysis of study outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studies
Qualitative
Quantitative
; statistical
reasoning
Results:Slide58
Two major streams:
Quantitative
research
hypothesis tests
Numerical scales
Statistical reasoningQualitative research rich descriptionText, Images, Video…
Results:Slide59
Quantitative
research.Numerical representation of realityDescriptive statistics
Simple characterization: “who / what / when?”Inferential statistics
Generalize to a larger population.
“Statistical reasoning”:
Probability judgments using the Normal distribution.Results:Slide60
Core features of research
Phenomenon
- Larger
question the research addresses
Theory
- Basic
Ψ
processes we think explain the phenomenon
Hypothesis
Concrete variables
that express the
Ψ
processes
Specific prediction
Methods / Data
Operational definitions
Study procedures
Results
- Hypothesis-wise
statistical analysis of study outcomes
Discussion & Conclusion
Relate results back to theory
Study limitations & Future studies
Implications for theory?
Study limitations?
Where now?
Discussion & Conclusions:Slide61
What does it mean that the hypothesis was (was not) supported?Change / modify theory?
Other interpretations / hypotheses?Applications?Study implications.
Discussion
Critical Thought
about the meaning – and alternate interpretations – of our results.Slide62
Study
limitations.Boundaries on what this study can tell us?Internal validity:Well did we represent the hypothetical constructs…?
Quality / nature of operationalization & design?External validity:
Our sample?
Manipulation / measurement of the independent variable(s)?Assessment of the dependent variable(s)?The research setting itself.
How representative was…DiscussionSlide63
What does science do?
Hypothetical constructs
In important relationship
S
pecific
variablesFalsifiable
prediction
Operational definition
Internal & external validity
Meaning of
results
for the
theory
Alternate interpretations
Study Limitations.
Qualitative
/ Quantitative
Descriptive question
or exploration
Hypothesis test
Phenomenon & Theory
Hypothesis
Methods
Results
Discussion
SUMMARYSlide64
Basic Elements of a Research Project
Methods
Measurement v.
experimental
Conclusions
Future research?PhenomenonBig picture / question
Theory
Hypothetical Constructs
Causal explanation
Hypothesis
Operational definition
Specific prediction
Data / Results
Descriptive data
Test hypothesis
Discussion
Implications for theory
S
pecific methods
& operational definitions
…derive concrete hypotheses.
A
ctual data & results…
… articulate a clear theory
Begin with the
“big question
”
…and larger
issues.
… implications for the theorySlide65
Basic Elements of a Research Project
MethodsMeasurement v.experimental
Each element of the project corresponds to a later / earlier issue…
Conclusions
Future research?
Phenomenon
Big picture /
question
Theory
Hypothetical Constructs
Causal explanation
Hypothesis
Operational definition
Specific prediction
Data / Results
Descriptive data
Test hypothesis
Discussion
Implications for theorySlide66
Basic Elements of a Research Project
Methods
Measurement v.
experimental
1
. Observation or hunch
Conclusions
Future research?
Phenomenon
Big picture /
question
Theory
Hypothetical Constructs
Causal explanation
Hypothesis
Operational definition
Specific prediction
Data
Descriptive data or observation
Discussion
Implications for theory
Study results often lead to the next experiment…
3. …explanation?
4.
T
heory
hypothesis
5. …how do we test or measure it?
2. …what don’t we understand?
Then we run the rest of the process
Data / Results
Descriptive data
Test hypothesis
Data / Results
Test
hypothesisSlide67
Basic Elements of a Research Project
Methods
Measurement v.
experimental
1. Alternate hypotheses?
Unanswered questions?
Conclusions
Future research?
Phenomenon
Big picture /
question
Theory
Hypothetical Constructs
Causal explanation
Hypothesis
Operational definition
Specific prediction
Results
Was the hypothesis supported?
Discussion
Implications for theory
R
esearch results
new experiment or study.
2
. Negative results:
Reframe hypothesis…
…operational
definitions
Data / Results
Descriptive data
Test hypothesis
Data / Results
Test new hypothesis
3
. Run follow-up study.Slide68
Basic Elements of a Research Project
Methods
Measurement v.experimental
1. Findings may lead us to rethink our theory.
Conclusions
Future research?PhenomenonBig picture / question
Theory
Hypothetical Constructs
Causal explanation
Hypothesis
Operational definition
Specific prediction
Discussion
Implications for theory
2
. We:
Formulate different hypotheses
D
ifferent study design & variables
New results
Results
Our initial findings
Data / Results
Descriptive data
Test hypothesis
Data / Results
Test new hypothesis
Results
Other findingsSlide69
Basic Elements of a Research Project
Methods
Measurement v.experimental
Key elements of research:
N
ot a simple linear process.All elements interact.ConclusionsFuture research?PhenomenonBig picture /
question
Theory
Hypothetical Constructs
Causal explanation
Hypothesis
Operational definition
Specific prediction
Discussion
Implications for theory
Data / Results
Descriptive data
Test hypothesisSlide70
Elements of science, review 1
A hypothetical construct is:A = A concrete description of a variableB = An abstract statement about a ψ process that cannot be seen directly.C = An excuse you construct to explain why you are late.D = An abstract use of statistical theory to test a hypothesis.Slide71
Elements of science, review 2
A theory is:A = Wild-eyed speculation about some topic that most people are not interested in.B = An authoritative statement of how something works: truth. C = Always tentative or provisional. D = A statement about how two (or more) hypothetical constructs are related.Slide72
Elements of science, review 3
An operational definition is:A = The specific way we manipulate an independent variable.B = A surgical procedure we use to test a hypothesis.C = The particular procedures we use to measure a study variable. D = An abstract statistical statement using probability theory to test hypotheses.Slide73
Elements of science, review 3
In science the numbers are what count most.Yes, if I have enough of themYes, if they have been operationally definedNo, the measures are most importantNo, the Theory and hypothesis is most important.Slide74
What does science do?
The core features of a research study.
Overall Research approaches.
Introduction to science, 6
Slide75
Overall research strategies
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
Measurement studies
vs.
Experiments
Slide76
Overall research strategies
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
A
ssess
nature
Describe behavior
T
est hypotheses
E
xert
control
over nature
T
est
experimental
predictions (hypotheses).Slide77
Overall research strategies
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
Rich / detailed
description.
Often targeted samples.
Count behaviors
e.g., by age
, gender,
ethnicity...
Archival
data
Measured variables
T
est or generate hypotheses.
N
atural events
Experiments:
N
o
control
over
IV
Non
-
equivalen
t
Groups
Manipulate
Independent Variabl
e
Measure
Dependent
Variable
.
Control
all aspects of experiment
Explore
behavior
.
Describe
trends.
Relate
variables
Field studies
C
ontrolled
(“lab”)
conditions.Slide78
Overall research strategies: Drug use
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
Methods:
Surveys, interviews
, archival
data
Show simple frequencies across, e.g., groups.
Methods
:
S
urveys
or
interviews
Archival data
Test hypotheses by correlating specific variables.
Methods
:
Experimental design.
Groups are
non-
equivalent
not
blind
not randomly
assigned…
Methods
:
Stimulate
contrasting brain
areas (IV
).
Assess drug
-
seeking (DV).
Methods:
Directly observe
drug
markets
,
In
-depth interviews
drug users,
Police…
Research Question
:
Mechanics of drug use.
Research Question
:
E
pidemiology
of drug
use.
Research Question:
S
ocial /
ψ
variables
in
drug use.
Research Question:
Compare drug treatments.
Research Question:
What brain centers control “drug craving”?Slide79
Experiments
Participant Selection
Group Assignment
Experimental Procedures
Experimental Condition
Results
Sample
Experimental group
Procedure
Experimental /“Treatment” condition
Outcome
Control
group
Procedure
Control / placebo condition
Outcome
Random
sample
of the target population.
Randomly assign
participants to groups.
E
xactly equal study procedures X groups.
Assess
Dependent Variable.
Controlled experiments:
“
G
old
standard” for testing
hypotheses.
Impose
experimental manipulation
Independent
Variable
.Slide80
Experiments
Participant Selection
Group Assignment
Experimental Procedures
Experimental Condition
Results
Sample
Experimental group
Procedure
Experimental /“Treatment” condition
Outcome
Control
group
Procedure
Control / placebo condition
Outcome
Random
sample
of the target population.
Randomly assign
participants to groups.
E
xactly equal study procedures X groups.
Assess
Dependent Variable.
Controlled experiments:
“
G
old
standard” for testing
hypotheses.
Impose
experimental manipulation
Independent
Variable
.
Random assignment
ensures
groups are equivalent at baseline
.
(Rather than, e.g., using existing groups
…)Slide81
Experiments
Participant Selection
Group Assignment
Experimental Procedures
Experimental Condition
Results
Sample
Experimental group
Procedure
Experimental /“Treatment” condition
Outcome
Control
group
Procedure
Control / placebo condition
Outcome
Random
sample
of the target population.
Randomly assign
participants to groups.
E
xactly equal study procedures X groups.
Assess
Dependent Variable.
Controlled experiments:
“
G
old
standard” for testing
hypotheses.
Impose
experimental manipulation
Independent
Variable
.
Experimental Control
ensures groups have the same procedures. Slide82
Experiments
Participant Selection
Group Assignment
Experimental Procedures
Experimental Condition
Results
Sample
Experimental group
Procedure
Experimental /“Treatment” condition
Outcome
Control
group
Procedure
Control / placebo condition
Outcome
Random
sample
of the target population.
Randomly assign
participants to groups.
E
xactly equal study procedures X groups.
Assess
Dependent Variable.
Controlled experiments:
“
G
old
standard” for testing
hypotheses.
Impose
experimental manipulation
Independent Variable
.
Manipulating the Independent Variable
ensures that we (the experimenter) cause the experimental effect. Slide83Slide84
Often true experiments are not possible:
We must use existing groups
We cannot control all the procedures
We cannot manipulate the Independent Variable
….
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
Naturally
occurring
events;
Education
research;
Public health
research.
Examples:Slide85
Measurement v. experiments
ExperimentHigh control / ‘lab’ conditions
Internal
validity
Determine “cause and effect”: validly interpret data
Measurement
Less control; ‘
research in nature’
External
validity
Data can generalize to “real world” & capture more complexity
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
Explore
Describe
Relate measured variables
Field studies
.
Test specific hypotheses
Slide86
Less control:
Observe / test phenomenon under natural conditions.More accurate portrayal of:“how it works in nature”complexity of phenomenonLess able to interpret cause & effect
Overall Research strategies:
Validity
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
Explore
Describe
Relate measured variables
Field studies
.
Test specific hypotheses
External validity Internal validity
More control:
Isolate (or
create
)
the phenomenon
in a
controlled
environment
Addresses
specific
questions or hypotheses
More
able
to interpret cause & effectSlide87
OverviewSlide88
Core course topics
How do we know things?What does scientific method tell us that other methods (political, religious thought) do not?
What does science do?
Describe the world
TaxonomiesEpidemiology
Qualitative researchPredict eventsSimple predictionsCorrelational studiesExperiments
Test theories
Cause & effect
Identify basic processes
Show how processes are related
Test applications of theories
E.g., behavioral interventionsSlide89
key terms
Features of research: Key termsTheoryHypothetical constructHypothesisVariable
Operational definition
Experimental control
Measurement v. experimental studiesInternal & external validitySlide90
Basic Elements of a Research Project
MethodsMeasurement v.experimental
ConclusionsFuture research?
Phenomenon
Big picture /
question
Theory
Hypothetical Constructs
Causal explanation
Hypothesis
Operational definition
Specific prediction
Data / Results
Descriptive data
Test hypothesis
Discussion
Implications for theorySlide91
Basics of major forms of research.
Observation or Measurement
Experiments
Simple Description
Correlational Studies
Quasi-experiments
“True” experiments
Qualitative
Quantitative
Explore
the actual process of a behavior.
Describe
a behavioral or social trend.
Relate measured variables
to each other to test hypotheses.
Test hypotheses in
naturally occurring events
or
field studies
.
Test specific hypotheses
via controlled “lab” conditions.
External validity Internal validity