Quantitative Research Design Flashcards
What are the important quality criteria in quant research?
Internal validity, external validity, construct validity, statistical conclusion validity, and ecological validity
What is meant by triangulation?
Triangulation can be applied for the purposes of measurement, data collection, or research strategy.
Increased triangulation should improve the ability of the researcher to draw conclusions from their studies.
o result in a more robust and generalizable findings (higher external validity).
o can enable the researcher to make the recommendations for managers with more clarity and confidence.
What is the “three-horned dilemma”?
Various research designs result in more (or lesss)
- Generalizability to the population that supports the issue of external validity
- Precision in measurement and control of the behavioral variables, affecting internal and construct validity.
- Realism of context
Generalizability- external validity, statistical conclusion validity
Precision- internal validity, construct validity
Realism- ecological validity, external validity
What is internal validity?
concerns causality: A cause-and-effect relationship can only be asserted if there is true covariation between the variables under investigation
the procedures used to gather the data demonstrate that the cause preceded the effect, and alternative explanations have been discarded.
Can we establish causality between the treatment and effect (IV and DV)?
What is external validity?
External validity: refers to generalizing across times, settings, and individuals.
Selection of subjects, settings, and time matters
What are the nine types of research strategy?
o Formal theory
o Litterateur review
o Sample surveys
o Laboratory experiments
o Judgement tasks
o Computer simulations
o Experimental simulations
o Field studies (primary data)
o Field studies (secondary data)
o Field experiments
What is research design?
Research design is the overall strategy that you will pursue to answer your research question. It ensures coherence between the different steps of the research process, so the study flows in logical manner.
Why is experiments the gold standard when it comes to internal validity?
Controlled environment and manipulation of treatment
Randomization- selecting random samples from population should produce statistically equal groups, equal probability distributions of all potential confounders
- So, if they are assigned to different treatments, the variance in dependent variable should be caused by treatment since everything else is equal
Why is causality so important?
It is the difference between descriptive and explanatory research (testing theory)
What is construct validity?
Construct validity concerns how well the measures employed fit the theories for which a test is designed
Techniques to test for construct validity:
o Confirmatory factor analysis (CFA)
o Exploratory analysis (EFA)
What is statistical conclusion validity?
Statistical Conclusion Validity refers to the ability to draw conclusions based on statical evidence of covariation as well as prediction.
Is the design precise and powerful enough to detect the relationship between variables if it indeed exists?
Violations of assumptions of statistical tests
Power-related issues
- This is usually a problem of big enough sample size
- In case of experiments careful monitoring of treatment is also important
- Unreliable measures and other sources of error
What is ecological validity?
measures how generalizable experimental findings are to the real world, such as situations or settings typical of everyday life. It is a subtype of external validity.
what is meant by MAX MIN CON?
The main function of a good research design is to explain and control variance- MAXMINCON
To maximize systematic variance. in all research designs we want to make sure we have as much variance caused by/associated with the interesting independent variables as possible.
To control extraneous systematic variance. there will always be variables that cause variance in the dependent variable, but we are not interested in their effects- they need to be minimized, nullified, or at least isolated so we can pull them apart.
To minimize the error or random variance.
