| Term | Definition |
|---|
| attrition | loss of participants over course of study, which can create bias by changing composition of sample initially drawn |
| baseline data | data collected before intervention, including pretreatment measures of outcomes |
| between-subjects design | research design in which there are separate groups of ppl being compared (e.g. smokers and nonsmokers) |
| blinding | process of preventing those involved in study (participants, intervention agents, or data collectors) from having information that could lead to bias (e.g. knowledge of which treatment group of participant is in) also called masking |
| case-control design | nonexperimental research design involving comparison of "cases" (ie ppl with condition under scrutiny, such as having had fall) and matched controls (similar ppl without condition) |
| comparison group | group of study paticipants whose scores on dependent variable are used to evaluate outcomes of group of primary interest (eg nonsmokers) as comparison group for smokers); term often used in lieu of control group when study design is not true experiment |
| control group | subjects in experiment who do not receive experimental treatment and whose performance provides baseline against which effects of treatment can be measured |
| correlational research | research that interrelationships among variables of interest, without researcher intervention |
| crossover design | experimental design in which one group of subjects is exposed to more than one condition or treatment, in random order |
| cross-sectional design | study design in which data are collected at one point in time; sometimes used to infer change over time when data are collected from different age or developmental group |
| counterfactual | condition or group used as basis of comparison in study, embodying what would have happened to same ppl exposed to casual factor if they simultaneously were not exposed to causal factor |
| experiment | study in which researcher controls (manipulates) independent variable and randomly assigns subjects to different conditions |
| external validity | degree to which study results can be generalized to settings or samples other than one studied |
| factorial design | experimental design in which two or more independent variables are simultaneously manipulated, permitting separate analysis of main effects of independent variables and their interaction |
| history threat | occurrence of events external to intervention but concurrent with it, which can affect dependent variable and threaten study's internal validity |
| homogeneity | 1) in terms of reliability of instrument, degree to which its subparts are internally consistent (ie are measuring same critical attribute); 2) more generally, degree to which objects are similar (ie characterized by low variability) |
| internal validity | degree to which it can be inferred that experimental treatment (independent variable), rather than uncontrolled, confounding factors, caused observed effects |
| intervention fidelity | extent to which implementation of treatment is faithful to its plan |
| longitudinal study | study designed to collect data at more than one point in time, in contrast to cross-sectional study |
| manipulation | introduction of intervention or treatment in experimental or quasi-experimental study to assess its impact on dependent variable |
| matching | pairing of subjects in one group with those in comparison group based on their similarity on one or more dimension, to enhance group comparibility |
| maturation threat | threat to internal validity of study that results when changes to outcome (dependent) variable result from passage of time |
| mortality threat | threat to internal validity of study, referring to differential loss of participants (attrition)no from different groups) |
| nonequivalent control group design | quasiexperimental design involving comparison group that was not created through random assignment |
| nonexperimental research | studies in which researcher collects data without introducing intervention |
| posttest-only design | experimental design in which data are collected from subjects only after intervention has been introduced |
| pretest-posttest design | experimental design in which data are collected from research subjects both before and after introducing intervention |
| prospective design | study design that begins by measuring presumed cause and then goes forward in time to measure presumed effects (ie lung cancer); also called cohort design |
| quasi-experimental design | design for testing intervention in which participants are not randomly assigned to treatment conditions; also called nonrandomized trial or controlled trial without randomization |
| random assignment | assignment of participants to treatment conditions in random manner (ie in manner determined by chance alone); also called randomization |
| retrospective design | study design that begins with manifestation of dependent variable in present (eg lung cancer), followed by search for presumed cause occurring in past (eg cigarette smoking) |
| selection threat | threat to study internal validity resulting from preexisting differences between groups under study; differences affect dependent variable in way extraneous to effect of independent variable |
| time series design | quasi-experimental design involving collection of data over extended time period, with multiple data collection points both before and after intervention |
| within-subjects design | research design in which single group of subjects is compared under different conditions or at different points in time (eg before and after surgery) |
Combine with other sets
Share on Facebook
Share on MySpace