Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
18 Cards in this Set
- Front
- Back
Populations
|
entire aggregation of cases in which a researcher is interested
|
|
Sampling
|
Process of selecting a portion of the pop represents the entire pop
|
|
Sample
|
subset of pop elements
|
|
Representative sample
|
a sample that is closely represents the population, achieved by randomization
|
|
Sampling bias
|
systematic overrepresentation or underrepresentation of segment of pop in terms of characteristics researched
|
|
strata
|
subpopulations
|
|
sampling designs in quantitative study
|
Nonprobability sampling (no randomization)
Probability sampling ( randomization |
|
Non probability sampling
|
1/ Convenience: most conveniently available participants. -> snowball sampling: ask for reference. This is the weakest type of sampling
2/ Quota: identify strata and determine how many participants are needed from each. An upgrade from convenient however rarely used. 3/ Consecutive: recruit all people from accessible pop with eligibility. Advantage: a better approach. Disadvantage: sampling period is long, potential bias. 4/ Purposive: (judgmental) research handpick sample. ex. a sample of experts. used in qualitative |
|
Evaluation of nonprobability sampling
|
rarely represent pop, however convenient and economical
|
|
Probability sampling
|
random selection of elements from pop.
1/ simple random 2/ stratified random 3/cluster 4/ systematic |
|
Random selection of sample
|
each element of pop has equal, independent chance of being selected. There is no guarantee that sample is representative of pop with any type of sampling design
|
|
Simple random sampling
|
most basic. research establish a sampling frame which is the name for the list of pop elements. then roster pop.
|
|
Stratified random sampling
|
pop is divided in two or more strata based on demographic attributes as age and gender.
This sharpen the representativeness of sample. however, this sampling design is impossible if information about stratifying variables is unavailable. |
|
Cluster sampling
|
random sampling of units, then random sampling of members in units. also called MULTISTAGE SAMPLING.
Less accurate than simple and stratified random, but more economical and practical. |
|
Systematic
|
selection of every kth case from al ist.
Sampling interval: standard distance between selected elements. Advantage: preferable than simple random, and more convenient. |
|
Sampling error
|
differences b/t pop values and sample values
|
|
Sample size
|
larger sample is more representative
|
|
Power analysis
|
estimate sample size based of group differences from a previous research. if differences are large, large samples are not necessary and vice versa.
Large sample are not assurance of accuracy, esp with nonprobability sampling |