P&B Chapter 12: Quant Research: Sampling, Data Collection, Measurement and Data Quality Flashcards
N =
N = Population: entire aggregation of cases in which the researcher is interested
n =
n = sample: subset of population elements
2 key considerations of representative quantitative sample
1) representativeness
2) size
Quantitative sample size can be determined by _____ _______.
power analysis
What specifies population criteria?
eligibility criteria
examples: TBI with at least 5 days inpatient rehab;
diagnosis of Diabetes;
participant of Rebuilding Together
**sometimes a population is also defined by characteristics that people must NOT possess (exclusion criteria)
List at least 4 considerations for eligibility?
1) cost
2) practical constraints
3) people’s ability to participate
4) design considerations (remember designs from Quiz 1)
Which type of sample? Probability or Non-Probability
* elements selected by non-random methods
non-probability
Which type of sample? Probability or Non-Probability
*involves random sampling
probability
Which type of sample? Probability or Non-Probability
*no way to establish probability that each element will be included in sample
non-probability
Which type of sample? Probability or Non-Probability
*estimates probability that an element will be included in sample
probability
4 types of non-probability sampling
1) Convenience sampling**
2) Quota sampling
3) Consecutive sampling
4) Purposive sampling
**Snowball sampling is a TYPE of convenience sampling (pg. 276 in the book lists convenience, quota, consecutive, & purposive as the 4 types though :) )
Which type of non-probability sampling?
1) Convenience sampling
2) Quota sampling
3) Consecutive sampling
4) Purposive sampling
5) Snowball sampling
* network or chain sampling by referral
Snowball sampling
Which type of non-probability sampling?
1) Convenience sampling
2) Quota sampling
3) Consecutive sampling
4) Purposive sampling
* recruiting all from accessible population over specified time and size
Consecutive sampling (can be selected retrospectively or prospectively)
Which type of non-probability sampling?
1) Convenience sampling
2) Quota sampling
3) Consecutive sampling
4) Purposive sampling
* researcher uses knowledge about population to select sample
purposive sampling
Which type of non-probability sampling?
1) Convenience sampling
2) Quota sampling
3) Consecutive sampling
4) Purposive sampling
* using those who are most available as participants
convenience sampling
pretty much the hallmark of non-probability sampling
4 types of probability sampling
I. simple random sampling
II. stratified random sampling
III. multicluster sampling
IV. systematic sampling
Which type of probability sampling? I. simple random sampling II. stratified random sampling III. multicluster sampling IV. systematic sampling * selecting every \_\_\_th case from a list; sample interval
systematic sampling
Which type of probability sampling? I. simple random sampling II. stratified random sampling III. multicluster sampling IV. systematic sampling * selecting broad groups in stages than randomly selecting sample
multistage cluster sampling
ex: OT students in US, then create regions, then random selection
Which type of probability sampling? I. simple random sampling II. stratified random sampling III. multicluster sampling IV. systematic sampling *subdivide population into homogeneous subtests then randomly select sample; proportionate vs. disproportionate
stratified random sampling
need large population to start with so that you can random sample from that
In Probability sampling, what error/bias is described?
*difference between population values and sample’s value
sampling error
In Probability sampling, what error/bias is described?
*systematic over-representation or under-representing of population segment; based on population’s homogeneity
sampling bias
Place the steps in sampling quantitative studies in order:
A. recruit the sample
B. identify the population
C. specify sampling plan (to include method of drawing sampling and power analysis)
D. specify eligibility criteria
- B. identify the population
- D. specify eligibility criteria
- C. specify sampling plan (to include method of drawing sampling and power analysis)
- A. recruit the sample
“Researchers usually sample from an _______ population, but should identify the ______ population to which they want to generalize their results”
accessible, target
Can you name one positive/strength of nonprobability sampling? What about one negative?
Positive: practical
Negative: strong potential for bias
Can you name one positive/strength of probability sampling? What about one negative?
Positive: best method for obtaining a representative sample (if the sample isn’t representative of the population, the study’s external validity and construct validity are at risk)
Negative: impractical/laborious
What does this describe: it specifies in advance how participants are to be selected and how many participants to include
Sampling plan
True or False - samples are sometimes selected in multiple stages
True (multistage sampling) - For example, in the first stage, large units are selected (nursing homes); then in a later stage, individual people are selected
the first stage can involved deliberate selection of the study site (nursing homes) and then people within the nursing homes can be randomly selected = a way to combine probability/nonprobability sampling together
True or False - “The probability of selecting a deviant sample size INCREASES as the size of the sample DECREASES”
False - the deviant sample size DECREASES as the sample size gets bigger and bigger (becomes more representative)
True or False - “The LARGER the sample, the SMALLER the sampling error”
True
Fill in the blank - “Power analysis builds on the concept of _______ _____, which expresses the strength of relationships among research variables”
effect size (so, if there is reason to believe that the IV and the DV will be strongly related, then a relatively small sample may be adequate to reveal the relationship statistically)
True or False - “If the population is relatively HOMOGENEOUS, a small sample size may be adequate”
True
When is attrition in a study most likely to occur (when people drop out of the study)?
1) if the time lag between data collection points is great
2) if the population is mobile
3) if the population is at risk of death or disability
