Chapter 5 Flashcards
sampling
Process of selecting people (or element) for inclusion in research study
external validity
Extent to which studies findings hold true outside of particular context of research
voluntary sample
Ppl. volunteer to be in study
Less representative and generalizable
conventional sample
Most readily available participants
Less generalizable
replication
Repeating study with different samples, different locations, etc.
Enhances generalizability of small experimental studies done on nonrandom samples
generalizability of experiments
more generalizable b/c designed to produce evidence of true causal relationships
qualitative research generalizability
typically not generalizable in terms of statistical representativeness b/c small samples, but can produce generalizable theories of how and why
census
Data collected on entire population, not just sample
sampling frame
Identify some kind of list, mao or other operational representation of population from which sample can be selected
Then select a sublet of units or elements to be included
how large a sample should be
Depends on how much statistical precision you want
Precision
Amount of random error in stats computed (larger = better)
2. How much you plan to subgroup for separate analysis
3. Does Not depend on size of population
sampling bias
Sample that gives systematically different results b/c shortcomings in same process
coverage bias
Members of sampling frame are systematically different from target population in way that influence results
Telephone surveys - 607 area code, people move, or live here but not with that area code
nonresponse rate
Nonresponders differe average in variable of interest therefore, underrepresented in study creating a bias
purposive sampling
Choosing people who have a unique perspective or occupy important roles, or selecting individuals or artifacts to represent theoretical categories or considerations
systematic sampling
Using intervals like every 20th person
