Upload
domenic-dale-lindsey
View
216
Download
1
Embed Size (px)
DESCRIPTION
Descriptive methods Naturalistic observation Watching behavior in normal environment Watching behavior in normal environment Major Advantage: Realistic data Realistic data
Citation preview
Why is psychology a science?Why is psychology a science?
Scientific Method:Scientific Method: System of gathering data so that System of gathering data so that
bias and error in measurement are bias and error in measurement are reducedreduced
Why is psychology a science?Why is psychology a science?
Scientific Method:Scientific Method:1. Perceive question1. Perceive question2. Form hypothesis 2. Form hypothesis 3. Test hypothesis3. Test hypothesis4. Draw conclusions4. Draw conclusions5. Report results5. Report results
-Replication-Replication
Descriptive methodsDescriptive methods
NaturalisticNaturalistic observation observation Watching behavior in normal environmentWatching behavior in normal environment
Major Advantage:Major Advantage: Realistic dataRealistic data
Descriptive methodsDescriptive methods
NaturalisticNaturalistic observation observation
Disadvantages:Disadvantages: Observer effectObserver effect Observer biasObserver bias Lack generalizeabilityLack generalizeability
Descriptive methodsDescriptive methods
LaboratoryLaboratory observation observation Watching behavior in laboratory environmentWatching behavior in laboratory environment
Advantages:Advantages: ControlControl Specialized equipmentSpecialized equipment
Descriptive methodsDescriptive methods
LaboratoryLaboratory observation observation
Disadvantage:Disadvantage: Artificial situation = artificial behaviorArtificial situation = artificial behavior
Descriptive methodsDescriptive methods
Case studyCase study One individual, great detailOne individual, great detail
Advantage:Advantage:• Tremendous detailTremendous detail
Disadvantages:Disadvantages:• Lack of generealizeabilityLack of generealizeability• Biases in self-reportBiases in self-report
• Phineas GagePhineas Gage
Descriptive methodsDescriptive methods SurveysSurveys
series of questions (e.g., Mate Preferences)series of questions (e.g., Mate Preferences)
Representative sample - Representative sample - randomlyrandomly selected selected sample of from a larger populationsample of from a larger population
Population - entire group of people/animals Population - entire group of people/animals of interestof interest
SamplingSampling
Descriptive methodsDescriptive methods SurveysSurveysAdvantages:Advantages:
Data from large numbers (large N)Data from large numbers (large N) Private informationPrivate information
Disadvantages:Disadvantages: Ensure representative sampleEnsure representative sample Inaccurate data (courtesy bias, memory)Inaccurate data (courtesy bias, memory) Order, wording effectsOrder, wording effects
MeasurementMeasurement
Scientific measurement requires:Scientific measurement requires:
Reliability = consistencyReliability = consistency
Validity = accuracyValidity = accuracy
Measure must be Measure must be reliablereliable: Coke : Coke vs. Pepsivs. Pepsi
Reliability:Reliability: Interrater reliabilityInterrater reliability-Do -Do observersobservers agree? agree?
Test-retest reliabilityTest-retest reliability--Are measures consistent Are measures consistent over over timetime??
BUT people might self report liking Coke very consistently simplybecause they reliably like the advertising.
Advertising preferencesAdvertising preferences
Measure must be Measure must be validvalid: : Coke vs. PepsiCoke vs. Pepsi
Internal validityInternal validity-Are-Are effects due to conditionseffects due to conditions manipulated?manipulated?
e.g., preference may comefrom product advertising orfrom product taste (e.g., initial sweetness)
Measure must be Measure must be validvalid: : Coke vs. PepsiCoke vs. Pepsi
Internal validityInternal validity-Are-Are effects due to conditionseffects due to conditions manipulated?manipulated?
... A blind taste-test would be a... A blind taste-test would be abetter measure…it wouldbetter measure…it wouldeliminate brand loyalty as aeliminate brand loyalty as avariable ( = higher internal validity)variable ( = higher internal validity)
Types of research designs: Types of research designs: relationshipsrelationships between variables between variables
CorrelationalCorrelational
ExperimentalExperimental
Quasi-experimentalQuasi-experimental
Types of research designs: Types of research designs: relationshipsrelationships between variables between variables
Correlational:Correlational: Association between variablesAssociation between variables No manipulationNo manipulation Correlation coefficient: r (range: -1.0-+1.0)Correlation coefficient: r (range: -1.0-+1.0) Correlations does not equal causation!Correlations does not equal causation!
Correlational researchCorrelational research
CorrelationCorrelation relationship between two variablesrelationship between two variables
VariableVariable• Anything that can change or varyAnything that can change or vary
Types of research designs: Types of research designs: relationshipsrelationships between variables between variables
Experimental:Experimental: CausalCausal relationship between variables relationship between variables Experimental manipulationExperimental manipulation
Types of research designs: Types of research designs: relationshipsrelationships between variables between variables
Quasi-experimental:Quasi-experimental: No manipulation possibleNo manipulation possible
Correlational research Correlational research
Correlational researchCorrelational research
CorrelationCorrelation Measures of two variablesMeasures of two variables Mathematical formulaMathematical formula
Correlation coefficient (r)Correlation coefficient (r), , representsrepresents1. direction of relationship (+/-)1. direction of relationship (+/-)2. strength of relationship 2. strength of relationship
Correlational researchCorrelational research
Correlation coefficient (r):Correlation coefficient (r):––1.00 to +1.001.00 to +1.00
Closer to 1.00 or -1.00, the stronger the Closer to 1.00 or -1.00, the stronger the relationshiprelationship
No correlation = 0.0No correlation = 0.0 Perfect correlation = -1.00 OR +1.00Perfect correlation = -1.00 OR +1.00
Correlational researchCorrelational research Positive correlation – Positive correlation – samesame direction direction
One increases, the other increases; one One increases, the other increases; one decreases, the other decreasesdecreases, the other decreases
Negative correlation – opposite directionNegative correlation – opposite direction One increases, other decreases.One increases, other decreases.
CORRELATION DOES CORRELATION DOES NOTNOT EQUAL EQUAL CAUSATION!!!CAUSATION!!!
Experimental researchExperimental research
manipulation of a variable manipulation of a variable
to see if changes in behavior resultto see if changes in behavior result
determination of cause-and-effect determination of cause-and-effect relationshipsrelationships
Experimental researchExperimental research
Operational definitionOperational definition Definition variable of interest Definition variable of interest Allows precise, direct measurementAllows precise, direct measurementExample: Aggression Example: Aggression
Variables in experimental designVariables in experimental design
Independent variableIndependent variable-what is manipulated-what is manipulated e.g., Prozac dosagee.g., Prozac dosage e.g., Violent TVe.g., Violent TV
Dependent variableDependent variable-what is measured, the -what is measured, the outcomeoutcome e.g., Depressione.g., Depression e.g., aggressive playe.g., aggressive play
ParticipantsParticipants in experimental design in experimental design
Experimental groupExperimental group: group whose environment : group whose environment or experience has changedor experience has changed e.g., Prozace.g., Prozac e.g., TVe.g., TV
Control groupControl group: group treated as much like : group treated as much like experimental group as possibleexperimental group as possible Does NOT participate in experimental manipulationDoes NOT participate in experimental manipulation
• e.g., No Prozac, sugar pille.g., No Prozac, sugar pill• No TVNo TV
Random assignmentRandom assignment Assignment to experimental or control groups Assignment to experimental or control groups
randomlyrandomly Participants have an equal chance of being in Participants have an equal chance of being in
either groupeither group
Controls for confounding variablesControls for confounding variables
Experimental designExperimental design
Placebo effectPlacebo effect expectations of the participants influence expectations of the participants influence
behavior (e.g., Viagra vs. sugar pill)behavior (e.g., Viagra vs. sugar pill)
Single-blind studySingle-blind study Participants do NOT know if they are in the Participants do NOT know if they are in the
experimental or the control group (reduces experimental or the control group (reduces placebo effect)placebo effect)
Experimental designExperimental design
Experimenter effectExperimenter effect Experimenter’s expectations influence resultsExperimenter’s expectations influence results
Double-blind studyDouble-blind study NeitherNeither the experimenter nor the participants the experimenter nor the participants
know group assignment (reduces placebo and know group assignment (reduces placebo and experimenter effect)experimenter effect)
Correlational designCorrelational design
Correlational designCorrelational design
ResultsResults::
Correlational designCorrelational design
Conclusions:Conclusions: NONO causal relationship can be inferred causal relationship can be inferred
Experimental designExperimental design
QuestionQuestion: Does close physical contact : Does close physical contact promote secure infant-mother attachment?promote secure infant-mother attachment?
MethodMethod: Women : Women randomlyrandomly assigned to assigned to use “Snuglis” (experimental) or standard use “Snuglis” (experimental) or standard infant seats (control) to carry their infantsinfant seats (control) to carry their infants
Experimental designExperimental design ResultsResults: Carrier effect on attachment: Carrier effect on attachment
SnuglisSnuglis: more responsive moms, infants more : more responsive moms, infants more likely to develop secure attachmentlikely to develop secure attachment
Standard carrierStandard carrier: less responsive moms, : less responsive moms, infants less likely to develop secure attachmentinfants less likely to develop secure attachment
ConclusionConclusion: : causalcausal relationship between relationship between type of carrier and attachmenttype of carrier and attachment