Chapter 13- Sampling in Quantitative Research Flashcards
Sampling Plan
quantitative researchers seek to select samples that will allow them to achieve statistical conclusion validity and to generalize their results beyond the sample used.
specifies in advance how participants are to be selected and how many to include
Population
the entire aggregation of cases in which a researcher is interested
accessible population
the aggregate of cases taht conform to designnated criteria and that are accessible for a study
target population
the aggregate of cases about which the researcher would like to generalize
Researchers usually sample an accessible population and hope to generalise to a target population
eligibility criteria
criteria that specify the population characteristics
aka inclusion criteria
a study’s construct validity is enhanced when there is a good match between the eligibility criteria and the population construct
eligibility criteria may also reflect some of the following: costs practical constraints people's ability to participate in study design considerations
exclusion criteria
characteristic that people must NOT possess
exclusion criteria may reflect a desire to strenthen internal validity, at the expense of external validity
sampling
the process of selecting cases to represent an entire population, to permit inferences about the population.
sample
a subset of population elements, which are the most basic units about which data are collected.
elements
the most basic units about which data are collected
in nursing research, elements most often human
representative sample
one whose key characteristics closely approximate those of the population
if sample is not representative of population, the study’s external validity and constuct validity are at risk
probability sampling
involves random selection of elements
researchers can specify the probability that an element of the population will be included in the sample.
greater confidence can be placed in the representativeness of probability samples
nonprobability samples
elements are selected by nonrandom methods- there is no way to estimate the probability that each element has of being included in nonprobability sample, and every element usually does not have a chance of inclusion
rarely representative of the population
will continue to predominate because of their practicality
strata
sometimes useful to think of populations as consisting of subpopulations or strata
a stratum is a mutually exclusive segment of the population defined by one or more characteristics
Often used in sampling selection to enhance the sample’s representativeness.
Using strata in sampling designs can facilitate the analysis of data for subgroups, to see if results are different for people with different characteristics.
multistage sampling
samples sometimes selected in multiple phases
first stage- larger units are selected (hospitals)
next stage- smaller units (individuals)
it is posisble to combine probability and nonprobability sampling.
For ex:: the first stage could involve the deliberate selection of study sites.
Then, people within the selected sites could be selected through random procedure.
Sampling bias
refers to the systematic overrepresentation or underrepresentation of a population subgroup on a characteristic relevant to the reserach question
sampling bias often occure unconsciously.
partly a function of population homogeneity
when variations occur in the population, then similar variations should be reflected, to the extent possible, in a cample
convenience sampling
entails using the most conveniently available people as participants
the problem with convenience sampling is that those who participate might be atypical of the population with regard to critical variables
the weakest for of sampling
in herterogenoeous populations, no other sampling approach faces a greater risk of sampling bias.
Yet, convenience sampling is the most commonly used method in many disciplines.
snowball sampling
variant of convenience sampling- early sample members (seeds) are asked to refer other people who might be eligible
Approach is often used when the population involves people who might otherwise be difficult to identify.
quota sampling
one in which the researcher identifies population strata and determines how many participants are needed from each startum
By using information about population characteristics, researchers can ensure that diverse segments are represented in the sample, in the proportions in which they occur in the population
does a better job of representing views of the population
consecutive sampling
involves recruiting all the people from an accessible population who meet the eligibility criteria over a specific time interval or for a specified sample size.
far better approach than convenience, especially if the sampling period is sufficiently long to deal with potential biases that reflect seasonal or other time-related fluctuations.
Greatly reduces the risk of bias.
good sampling design when there is a rolling enrollment into a contained accessible population.
purposive sampling
uses the researcher’s knowledge about the population to make selections
drawback is that this approach may not result in a typical or representative sample.
Ex is a delphi surveys- developed as a took for short-term forecasting. the technique involves a panel of experts who are asked to complete several rounds of questionnaires focusing on their judgment about a topic of interest.
sometimes used to good advantage in two-staged sampling
probability sampling
involves the random selection of elements from the population
the best method of obtaining representative samples. if all the elements in a population have an equal probabiltiy of being selected, then the resulting sample is likely to do a good job of representing the population.
Another advtanges in that probability sampling allows researchers to estiamt the magnutude of sampling errors.
impractical- typically not possible to select a probability sample unless the population is narrowly defined.
random sampling
involves a selection process in which each element in the population has an equal, independent chance of being selected.
Probability sampling is complex, technical topic,
Simple random sampling
researchers use simple random sampling to etablish a sampling frame- the technical name for the list of elemetns from which the sample will be randomly choses.
once sampling frame has been developed elements are numbered consecutively.
random selection ensure that differences in the attributes of the sample and the population are purely a function of chance
Stratified random sampling
population is first divided into two or more homogeneous strata (based on gender) from which elements are selected at random.
Unlike quota sampling, stratified random samping requires that a person’s status in a stratum be known before making selections, which can be problematic.
Can guarantee appropriate representation of different population segments
enable researchers to sharpen a sample;s representativeness.
May be impossible if information on the critical variables is unavailable.
a stratified sample requires even more labor than simple random- sample must be drawn from multiple enumerated listings.
proportionate stratified sampling
participants are selected in proportion to the size of the population stratrums.
disproportionate samplings
comparisons are sought between strata of greatly unequal sizes
when using disproportionate sampling it is important to make adjustments to arrive at the best estimate of OVERALL population values. These adjustments, called WEIGHTING, is a simple mathematic computation described in textbooks on sampling.
cluster sampling- multistage
involves selecting broad groups (clusters) rather than selecting individuals, as the first stage of a multistage approach
described in the number of stages
clusters can be a simple or stratified method.
systematic sampling
involves selecting kth case from a list, such as every 10th person on a patient listing or every 25th person on a student roster.
when this sampling method is applied to a sampling frame as essentially random sample can be drawn, using the following proceduer
sampling interval
the standard distance between sampled elements.
the desired sample size is established at some number (n)
The size of the population must be known or estimated (N)
By diving N by n, a sampling interval (k) is established.
yields essentially the same results as simple random sampling but involves less work.
sampling error
Refers to the difference between sample values (eg average age of the sample) and population values (the average age of the population)
Another advantage in that probability sampling allows researchers to estimate the magnitude of sampling errors.
power analysis
procedure can be used to estimate sample size needed.
effect size
power analysis builds on the concept of an effect size, which expresses the strength of relationships among research variables
If there is a reason to believe the IV and DV are strongly correlated, then a relatively small sample may be adequate to reveal the relationship statistically.
typically, however,r nursing interventions have a moderate effect.
homogeneity of a population-= small sample size may be adequate
sample recruitment
two major tasks: identifying eligible candidates and persuading them to participate.
may be necessary to create a screening instrument- which is brief for that allows researchers to determine whether a prospective participant meets the study’s eligible criteria.
response rates
even with a rigorous sampling plan, the sample may be biased if not all people invited to participate in a study agree to do so.
research reports should provide info about response rates (i.e. the # of people participating in a study relative to the number of people sampled) and about possible nonresponse bias, which reflects the difference between participants and those who decline to participate.
in longitudinal studies, attrition bias should be reported.
