Quantitative Research Flashcards
(30 cards)
Participant expectancy
Subjects behave differently because they believe they are receiving a treatment or because they are taking part in an experiment
Single-blind set up
participants don’t know whether they are in the control or experimental group
Double-blind set-up
Researcher and Participant don’t know whether they are in the control or experimental group
Compensatory Rivalry (Avis effect)
Control group is aware of the experimental group and tries to compete with them
Resentful Demoralisation
The reaction of a non-treatment control group or groups receiving less desirable treatment can be associated with resentment and demoralisation
Diffusion or imitation treatment
When treatments involve informational programs and when the various experimental (and control) groups can communicate with each other, responders in one treatment may learn information intended for others
Why single-blind set-up
-Participants expectancy becomes equivalent across groups
- compensatory rivalry is not felt
- resentful demoralisation is not felt
- diffusion or imitation of treatment does not occur with placebo control and less likely to occur with active control
Experimenter expectancy
Experimenters or testers anticipating that certain participants will perform better
Experimental Mortality
Participants withdrawing from a study for any reason
Ways to reduce Experimental Mortality
-Pre-experimental information explaining the importance/ value of the research
- Consideration of design and pilot studies
- Include a control group to see if the loss id different across groups
Ways to reduce Experimenter Expectancy
- ensure experimenter is blinded (doesn’t know who’s in which group)
History (p’s history)
An unintended event that occurs during the treatment period that may affect the study outcome
Examples- illness, not controlling activity during the experimental period
Ways to reduce history as a threat
- Control participants activity
-eg pre-experimental information
Instrumentation
Problems with the validity, reliability, objectivity and calibration of measurement devices (both equipment and observers)
- includes appropriate training of the testers and the consistency of the protocol
Validity of Measurements
The degree to which a test or instrument measures what it purports to measure
Reliability of Measurements
Acceptable agreement between repeated tests made under similar conditions (repeatability, reproducibility, measurement error and precision)
Objectivity of measurements
Acceptable agreement between different observers
- also know as inter-observer reliability or inter-rater reliability
Ways to reduce instrumentation as a threat
- Appropriate calibration, reliability, objectivity and validity of measures
- Appropriate training of testers and good consistency of the protocol
Testing
The effect one test on subsequent administration of the same test
- p will do better on the test the next time
Ways to reduce Testing as a threat
- Familiarisation sessions prior to the start of experiment
Statistical Regression
Groups (or a group) that have been selected on the basis of extreme scores are not as extreme on subsequent testing
- looking at an intervention on poor catches
Ways to reduce statistical regression as a threat
- does the question need extreme group? if not then don’t select on the basis on extreme scores
- An average of a participant scores to get a better classification
Quasi-experimental designs
Researcher tries to fit into a. real world setting whilst try to control as many threats to validity as possible
Ex Post Facto Design
- A comparison of existing groups is made
- Researcher has no control over the IV
- Also known as casual-comparative research
- Considered by some to be descriptive research
- Research seeks cause and effect relationship that explains differences
