Psychology as a Science zIn this lecture we will discuss: yscience - a method for understanding ylimits of common sense ymethods of science xdescription xcorrelation xexperimentation yevaluating data with statistics ysources of error and bias in research

Science vs. Common Sense zCommon sense and intuition often tell us about psychology ye.g., suppose a study tells us that ‘separation weakens romantic attraction’ ycommon sense may tell us - “out of sight, out of mind” yor common sense may say the opposite - “absence makes the heart grow fonder” zCommon sense can be inconsistent and based on hindsight

Science vs. Common Sense zScience helps build explanations that are consistent and predictive rather than conflicting and postdictive (hindsight) zScience is based on yknowledge of facts ydeveloping theories ytesting hypotheses ypublic and repeatable procedures

Scientific Inquiry zFacts are what need to be explained yobjective - viewable by others ybased on direct observation yreasonable observers agree are true zTheory is a set of ideas that yexplains facts ymakes predictions about new facts zHypothesis yprediction about new facts ycan be verified or falsified

Methods in Psychology zSetting - field vs. laboratory zMethods of data collection yself-report vs. observational zResearch plan or design ydescriptive ycorrelational yexperimental

Descriptive Study zDescribes a set of facts zDoes not look for relationships between facts zDoes not predict what may influence the facts zMay or may not include numerical data zExample: measure the % of new students from out-of-state each year since 1980

Correlational Study zCollects a set of facts organized into two or more categories ymeasure parents disciplinary style ymeasure children’s behavior zExamine the relation between categories zCorrelation reveals relationships among facts ye.g., more democratic parents have children who behave better

Correlational Study zCorrelation cannot prove causation yDo democratic parents produce better behaved children? yDo better behaved children encourage parents to be democratic? zMay be an unmeasured common factor ye.g., good neighborhoods produce democratic adults and well behaved children

Experiments zDirect way to test a hypothesis about a cause-effect relationship between factors zFactors are called variables zOne variable is controlled by the experimenter ye.g., democratic vs. authoritarian classroom zThe other is observed and measured ye.g., cooperative behavior among students

Experimental Variables zIndependent variable ythe controlled factor in an experiment yhypothesized to cause an effect on another variable zDependent variable ythe measured facts yhypothesized to be affected

Independent Variable zMust have at least two levels ycategories - male vs. female ynumeric - ages 10, 12, 14 zSimplest is experimental vs. control yexperimental gets treatment ycontrol does not

Experimental Design zLevels may differ between or within people zWithin-subject experiment - different levels of the independent variable are applied to the same subject zBetween-groups experiment - different levels of the independent variable are applied to different groups of subjects

Experimental Design zRandom sample - every member of the population being studied should have an equal chance of being selected for the study zRandom assignment - every subject in the study should have an equal chance of being placed in either the experimental or control group zRandomization helps avoid false results

Research Settings zLaboratory ya setting designed for research yprovide uniform conditions for all subjects ypermits elimination of irrelevant factors ymay seem artificial zField research ybehavior observed in real-world setting ypoor control over conditions ymeasures may be more representative of reality

Data-Collection Methods zSelf-report - procedures in which people rate or describe their own behavior or mental state yquestionnaires yrating scales xon a scale from 1 to 7 rate your opinion of … yjudgements about perceptions xon a scale from 1 to 100 how hot is...

Data-Collection Methods zObservational methods - researchers directly observe and record behavior rather than relying on subject descriptions ynaturalistic observation - researcher records behavior as it occurs naturally ytests - researcher presents stimuli or problems and records responses

Descriptive Statistics zVariable - something that can vary or change zDependent variable - something we measure zData - a collection of measurements zStatistics - summary descriptions of data (i.e., mean, medium, range)

Descriptive Statistics zUsed to describe or summarize sets of data to make them more understandable ymeasures of central tendency xmean, median, mode ymeasures of variability xrange, standard deviation ymeasures of association xcorrelation coefficient

Measures of Central Tendency zWhat is the average family income above? zMean - the arithmetic average zMedian - the center score zMode - the score that occurs the most

Measures of Variability zRange - the difference between the highest and lowest score in a set of data zStandard deviation - reflects the average distance between every score and the mean

Correlation Coefficient zOften we measure more than one variable zGrade point and SAT score zAre they related? zCorrelation statistic is a way to find out

Correlation Coefficient zMeasures whether two variables change in a related way yCan be positive (max +1.00) zNegative (min -1.00) zOr not related! (~ 0.0)

Inferential Statistics zDescriptive statistics summarize a data set zWe often want to go beyond the data zIs the world at large like my sample? zAre my descriptive statistics misleading? zInferential statistics give probability that the sample is like the world at large

Statistics and Probability zProbability means how likely something is zHow likely are results like mine to occur by chance? zStatistical inferences ysignificant result - reflects the real world rather than chance, with high probability (e.g., >.95 ) ynot significant - results reflect chance

Measurement Errors zWhy is inference based on probability instead of certainty? zData can be misleading because of variability ylow variability yhigh variability

Measurement Errors zWhy is inference based on probability instead of certainty? ylow bias yhigh bias zData can be misleading because of bias

Measurement Error zVariability and bias can combine Variability & Bias VariabilityBias

Sources of Bias zBiased sample - when the members of a sample differ in a systematic way from the larger population the researcher is interested in zExample yinterested in all voters ycontact by telephone ybiased sample - lower economic groups may not own telephones

Sources of Bias zObserver-expectancy effect yresearcher has expectations that influence measurements zSubject-expectancy effect ysubject knows design and tries to produce expected result zBlinding yminimize expectancy by removing knowledge about experimental conditions

Blinding zSingle-blind study - when subjects are kept uninformed as to the treatment they are receiving zDouble-blind study - when both subjects and experimenter are kept uninformed about aspects of the study that could lead to differential expectations

Ethical Issues in Psychological Research zRight to privacy zInformed consent yuse of deception zAnimal rights yIs there justification for discomfort or harm a research procedure may produce? zAPA publishes ethical guidelines