EPPP Research Methods and Statistics questions and answers 100% correct

EPPP Research Methods and Statistics questions and answers 100% correctTheories and Hypotheses scientific theories must result in a set of propositions, or hypotheses, that can be refuted or disconfirmed internal validity extent to which we can draw cause-and-effect inferences from a study 3 criteria for internal validity 1. statistical association-the hypothesized cause and its effect covary 2. temporal precedence- the putative cause precedes its effect in time IE. symptoms remission could not predate exposure to therapy if therapy is the cause of symptom remission 3. nonspuriousness- the hypothesized cause, not another factor, is responsible for the effect spurious causes are threats to internal validity external validity the extent to which the results of a study can be generalized to or across other situations statistical conclusion validity whether there is a statistical association and the magnitude of this association Construct validity -does this go along with existing theory - does it correlate with other measures it should and not correlate with other measures it shouldnt -convergent and discriminant -extent inferences can be made from particular study instances tot he higher-order construct from which they derive Campbell's Validity typology internal validity external validity statistical conclusion validity construct validity Group-based randomized experiments randomized are goal standard for assessing causality chance process-probability equated on all measured and unmeasured characteristics randomized control condition generated hypothetical counterfactual IR what would happen if the control group had gotten the treatment Efficacy trials group based experimental study design intervention effects examined under ideal circumstances; require significant clinical training, supervision, and close monitoring effectiveness trials group based experimental study design intervention effects examines under real-world conditions; outside of academic settings; clients less homogeneous diagnostically Intent-to-treat Analyses group based experimental study design analyze data from randomized experiments involving participant attrition; analyze outcome as function of the original group assignment; provide a conservative (real-world) estimate of treatment effects drop outs usually fail to provide outcome data creating missing data Single-case experiments designed to increase internal validity, assess causal influence of intervention ABAB study design A=baseline, B=treatment multiple baseline design study design looking for replication of effect across multiple baseline (IE behaviors, setting, individual etc) ways to evaluate results in single-case mean changes- averaged of outcome across phases level shifts-compare last data point in phase to first point in next phase slope changes- rate of behavior in different phases quasi-experimental studies experiments that lack random assignment of units to conditions 1. participants self-select into treatment conditions 2. various intact groups (like clients at different community health clinics) are differentially exposed to treatment conditions researchers must identify and render implausible alternative explanations for an apparent intervention effect (rule our various threats to internal validity) Correlational studies aka passive observational studies; researcher not actively manipulating anything case-control design correlational study compare a group with certain characteristics with a group who does not possess those characteristics EX ADHD (case) vs no ADHD (control) Cohort design correlational study intact group (cohort) followed over time to examine emergence or change in some outcome of interest longitudinal multiple cohort studies is different from case-control because in multiple the group differs on exposure, not a central characteristic uncontrolled case studies one or a few cases studied intensively; hope to understand case in comprehensive and nuances way Maturation threat to internal validity naturally occurring changes (passage of time) mistaken for intervention effects history threat to internal validity event/s during study that impact results and mistaken for intervention effects statistical regression threat to internal validity extreme scores revert back to the mean; more plausible in single-group studies where extreme performers (ex severely depressed) comprised study sample attrition threat to internal validity threat when the pattern of drop-out impacts the results and interpreted as an effect testing (threat to internal validity) exposure to pre-test changes the individual EX questions about alcohol use inspires person to lower consumption instrumentation threat to internal validity measurement tool changes EX campaign to reduce alcohol-related injuries operationalized as "alcohol-related ER visits" selection and interactions with selection threat to internal validity