Lecture 7 Current Lecture Chapter 4: Research Methods in the Study PsychopathologyI. Science and Scientific MethodsII. Approaches to Research on PsychopathologyIII. Integrating the Findings of Multiple StudiesStatistical and Clinical Significance Statistical significance• Probability ≤ .05• Can be influenced by number of participantsLarger samples increase likelihood of significance• Clinical significance• Is the association meaningful as well as statistically significant?• Reducing purges from 4 times a week to 3 times a week may not be meaningfulThreats to external validity: Setting Limitations of experimental settingWas there something odd about the time period duing which the data was collected? Unique history accompanied treatment (e.g. Surgeon general’s report on smoking issued during course of smoking cessation treatment)Is there something unique about the experimental settingContext effects (rural vs. urban site; doctor’s office vs. experimental lab)• PROBLEM: Will Tx only generalize to sites with same history or same setting/context?• SOLUTION: conduct replications at different sites, at different timesThreats to external validity: Limitations of sampling Selection factors limiting generalizabilityWill results generalize to other groups?• College studentsHigher SES, IQ, more psychologically minded, personality differences (ambition, Type A, openness)• VolunteersPsych 380 1st EditionMore likely to be female, compliant, in need of incentive, interested in topic (Should SONA include titles when recruiting subjects?)• Treatment seeking vs. self careTreatment seeking tend to be female, have more intractable problem, less natural support, have more $, more favorable attitudes toward seeking helpExperimental Research: Treatment Outcomes Research designed to answer the question: “Does treatment work?” Empirically supported treatments• Efficacy and Effectiveness• Established protocol à Treatment Manuals• Need for dissemination Evidence-based practice: Definition An EBP is an intervention for which there is strong research demonstrating effectiveness in improving client outcomes  Effectiveness refers to demonstration of helpfulness in the applied setting, not justin strictly controlled research settings Problem: No single agency like FDA is for drugs to decide what is evidence basedAnalogue Experiment Experiments not always possible in psychopathology• Ethical or practical constraints Examine related or similar behavior in the lab• Induce temporary symptoms• Recruit participants with similarities to diagnosable disordersCollege students who tend to be anxious or depressed• Animal researchSingle-Case Experimental Research Examine how individual participants respond to changes in the independent variable. Reversal (ABAB) Design• The reversal technique not always possible Initial state may not be recoverableIntegrating Findings from Multiple Studies Meta-analysis• Identify relevant studies• Compute effect sizeTransforms results to a common scale Smith et al. (1980)• Meta-analyzed 475 outcome studiesInvolved 25,000 subjects• Results: Psychotherapy is