4 Error in survey research Sampling error - Are there differences in your sample compared to the population as a whole?Response rateWhat proportion of the sample actually responded to the survey?Hidden costs here - what can you do to increase response ratesNon-response error (bias)Is there something special about the data that you’re missing? From the people who didn’t respondMeasurement error (validity and reliability)Are your questions really measuring what you want them to?
5 Quasi-experiments What are they? Almost “true” experiments, but with an inherent confounding variableDesign includes a quasi-independent variableExamplesAn event occurs that the experimenter doesn’t manipulateInterested in subject variablesTime is used as a variable
6 Quasi-experiments What are they? Almost “true” experiments, but with an inherent confounding variableAdvantagesAllows applied research when experiments not possibleThreats to internal validity can (sometimes) be assessed
7 Quasi-experiments What are they? Disadvantages Almost “true” experiments, but with an inherent confounding variableDisadvantagesThreats to internal validity may exist (which can not be addressed)Be careful when making causal claimsStatistical analysis can be difficultMost statistical analyses assume randomness
8 Quasi-experiments What are they? Common types Almost “true” experiments, but with an inherent confounding variableCommon typesNonequivalent control group designsProgram evaluationInterrupted time series designs
9 Quasi-experiments Nonequivalent control group designs With pretest and posttest (most common)participantsExperimentalgroupControlMeasureNon-RandomAssignmentIndependent VariableDependent VariableBut remember that the results may be compromised because of the nonequivalent control group
10 Quasi-experiments Program evaluation Research on programs that is implemented to achieve some positive effect on a group of individuals.e.g., does abstinence from sex program work in schoolsSteps in program evaluation:Needs assessment - is there a problem?Program theory assessment - does program address the needs?Process evaluation - does it reach the target population? Is it being run correctly?Outcome evaluation - are the intended outcomes being realized?Efficiency assessment- was it “worth” it? The the benefits worth the costs?
11 Quasi-experiments Time series designs treatment Basic method: Observe a single group multiple times prior to and after a treatmenttreatmentObsObsObsObsObsObsThe pretest observations allow the researcher to look for pre-existing trendsThe posttest observations allow the researcher to look for changes in the trendsIs it a temporary change, does it last, etc.?
12 Quasi-experiments Time series designs treatment A variation of basic time series designAddition of a nonequivalent no-treatment control group time seriesObstreatmentObs
13 Developmental designs Used to study changes in behavior that occur as a function of age changesAge typically serves as a quasi-independent variableThree major typesCross-sectionalLongitudinalCohort-sequential
14 Developmental designs Cross-sectional designGroups are pre-defined on the basis of a pre-existing variableStudy groups of individuals of different ages at the same timeUse age to assign participants to groupAge is subject variable treated as a between-subjects variable
15 Developmental designs Cross-sectional designAdvantages:Can gather data about different groups (i.e., ages) at the same timeParticipants are not required to commit for an extended period of time
16 Developmental designs Cross-sectional designDisavantages:Individuals are not followed over timeCohort (or generation) effect: individuals of different ages may be inherently different due to factors in the environmentExample: are 5 year old different from 13 year olds just because of age, or can factors present in their environment contribute to the differences?Cannot infer causality due to lack of control
17 Developmental designs Longitudinal designFollow the same individual or group over timeAge is treated as a within-subjects variableRather than comparing groups, the same individuals are compared to themselves at different timesRepeated measurements over extended period of timeChanges in dependent variable reflect changes due to aging processChanges in performance are compared on an individual basis and overall
18 Developmental designs Longitudinal designAdvantages:Can see developmental changes clearlyAvoid some cohort effects (participants are all from same generation, so changes are more likely to be due to aging)Can measure differences within individuals
19 Developmental designs Longitudinal designDisadvantagesCan be very time-consumingCan have cross-generational effects:Conclusions based on members of one generation may not apply to other generationsNumerous threats to internal validity:Attrition/mortalityHistoryPractice effectsImproved performance over multiple tests may be due to practice taking the testCannot determine causality
20 Developmental designs Cohort-sequential designMeasure groups of participants as they ageExample: measure a group of 5 year olds, then the same group 5 years later, as well as another group of 5 year oldsAge is both between and within subjects variableCombines elements of cross-sectional and longitudinal designsAddresses some of the concerns raised by other designsFor example, allows to evaluate the contribution of generation effects
21 Developmental designs Cohort-sequential designAdvantages:Can measure generation effectLess time-consuming than longitudinalDisadvantages:Still time-consumingStill cannot make causal claims
22 Small N designs What are they? Historically, these were the typical kind of design used until 1920’s when there was a shift to using larger sample sizesEven today, in some sub-areas, using small N designs is common place(e.g., psychophysics, clinical settings, expertise, etc.)
23 Small N designs One or a few participants Data are not analyzed statistically; rather rely on visual interpretation of the dataObservations begin in the absence of treatment (BASELINE)Then treatment is implemented and changes in frequency, magnitude, or intensity of behavior are recorded
24 Small N designs Baseline experiments – the basic idea is to show: when the IV occurs, you get the effectwhen the IV doesn’t occur, you don’t get the effect (reversibility)Before introducing treatment (IV), baseline needs to be stableMeasure level and trend
25 Small N designs Level – how frequent (how intense) is behavior? Are all the data points high or low?Trend – does behavior seem to increase (or decrease)Are data points “flat” or on a slope?
26 ABA design ABA design (baseline, treatment, baseline) The reversibility is necessary, otherwisesomething else may have caused the effectother than the IV (e.g., history, maturation, etc.)
27 Small N designs Advantages Focus on individual performance, not fooled by group averaging effectsFocus is on big effects (small effects typically can’t be seen without using large groups)Avoid some ethical problems – e.g., with non-treatmentsAllows to look at unusual (and rare) types of subjects (e.g., case studies of amnesics, experts vs. novices)Often used to supplement large N studies, with more observations on fewer subjects
28 Small N designs Disadvantages Effects may be small relative to variability of situation so NEED more observationSome effects are by definition between subjectsTreatment leads to a lasting change, so you don’t get reversalsDifficult to determine how generalizable the effects are
29 Small N designsSome researchers have argued that Small N designs are the best way to go.The goal of psychology is to describe behavior of an individualLooking at data collapsed over groups “looks” in the wrong placeNeed to look at the data at the level of the individual