Chapter 10: Research Designs for Special Circumstances Flashcards
Program evaluation
research designed to evaluate programs (e.g. social reforms, innovations) that are designed to produce changes or certain outcomes in a target population
Questions that guide program evaluations
needs assessment, program theory assessment, process evaluation, outcome evaluation, efficiency assessment
2 key elements of a good/true experiment
manipulation (IV) and random assignment
Quasi-experimental designs
study design that has many features of an experiment, but lacks some aspects of a true experimental design due to necessity e.g. control conditions and random assignment (and can’t support causal inferences)
Types of quasi-experimental designs
one-group posttest only, one-group pretest-posttest, non-equivalent control group (posttest only), non-equivalent control group (pretest-posttest), interrupted time series, control series
One-group posttest only deisgn
has no control group and no pretest comparison; very poor in terms of internal validity
One-group pretest-posttest design
the effect of an independent variable is inferred from the pretest-posttest difference in a single group; fail to consider alternative explanations
Threats to internal validity
history, maturation, testing, instrument decay, regression toward the mean, attrition, selection effects, cohort effects
History effects
outside events not part of the manipulation (that occur after/around it) influence the DV, providing an alternative explanation for results
Maturation effects
any naturally occurring change within individuals that occurs over time could provide an alternative explanation for results
Testing effects
simply taking the pretest changes behavior, without any effect of the IV
Instrument decay
characteristics of the measurement instrument changes over time, providing an alternative explanation for results
Regression toward the mean
statistical phenomenon where extreme scores on a variable tend to be closer to the mean when a measurement is repeated
Non-equivalent control group design
groups of participants in the different conditions are not equivalent (e.g. naturally occurring/pre-existing groups) and there is no pretest
Seelection differences
pre-existing differences in the type of participants who make up each group in a between-subjects experimental design; happens when there is no random assignment
Non-equivalent control group pretest-posttest design
non-equivalent groups are used, but a pretest allows assessment of equivalency and pretest-posttest changes
Interrupted time series design
a treatment (often a natural manipulation) is investigated by examining a series of measurements made over an extended time period, both before and after the treatment is introduced
Control series design
an extension of the interrupted time series design in which there is a non-equivalent comparison or control group
Single case experimental design
the effect of the IV is assessed using data from a single participant
Baseline
form of control condition in which participant behavior is measured during a control period before introduction of the manipulation; used in single case experimental design
Reversal design
treatment is introduced after a baseline period then withdrawn during a second baseline period; to increase certainty in the effect of an intervention
Multiple baseline design
observing behavior before and after a manipulation under multiple circumstances (across different individuals, behaviors, or settings)
Developmental research
study the way people change as a function of age; good internal validity but expensive and high attrition
Longitudinal method
the same people are observed repeatedly as they grow older; similar to within-subjects design
Developmental research methods
longitudinal method, cross-sectional method, sequential method
Cross-sectional method
people of different ages are studied at a single point in time; similar to between-subjects design
Cohort
group of people born around the same time
Cohort effects
attitudes and beliefs influenced by popular discourse at formative times in people’s lives; confounded with age in cross-sectional study
Sequential method
combination of cross-sectional and longitudinal designs to study developmental research
Potential outcomes of sequential design
(1) All cohorts show the same pattern no matter their current age (2) All ages show the same pattern in a particular year no matter what cohort
