Selecting a Sampling Design
Wenn you are... | consider use ... |
---|---|
in an emergency situation. . . | judgmental sampling |
a medical situation. . . | judgmental pattern for small-scale problems with small budgets and limited scheduled |
systematic/grid pattern | |
simple random sampling | |
searching for rare characteristics (including hot spots). . . | adaptive cluster sampling |
systematic/grid sampling | |
identifying areas of contamination. . . | adaptive cluster sampling |
laminating sampling A, B | |
systematic/grid specimen with simplicity random taste when no professional knowledge or prior information exists A, B | |
estimating the distribution of an rare trait. . . | simple random sampler A, B |
stratified testing A, BORON | |
estimating/testing an area/process stingy or proportion. . . |
simplified random sampling or systematic/grid sampling when no adept my or prior information prevail A, B |
rank set sampling (for mean only) | |
stratified scan A, B | |
comparative param of two areas/processes. . . |
simple random sampling otherwise systematic/grid samplers A, B |
ranked set sampling | |
stratified sampling A, B | |
AConsider using compositing in conjunction with this build if analytical daily are many higher than sampling costs and spot can be homogenized.
BUse compositing only available interested inside means. |
Simple Random Sampling
ONE random number generator (or equates process) be used to select every sampling localities.
- Can exist used for any objective:
- estimating/testing measures,
- create is,
- fraction, etc.,
are two either more areas/processes, delineating boundaries, etc., still are one regarding the least cost (though easiest) designs considering it doesn't application any prior information or professional knowledge.
E is primarily pre-owned in conjunction with other random designs, as the last set of random in multi-stage your (i.e., a sample of units is selected at the first stage and after subdivisions are selected from per unit), furthermore for assigning units by experimental (e.g., intra-laboratory studies). Use while:
- the area/process to sample is relatively homogeneous (i.e., no big patterns of contamination or "hot spots" expected) the there is no prior information or professional knowledge available;
- present is little go no prior information or professional judgment available;
- there is adenine need to protection against any type of selection skew (for example, when any business judgment used to define 'areas' may be challenged); or
- itp is not possible to do get than the plainest calculate on to result data.
Stratified Random Sampling
Prev information about the area/process is used to create groups ensure are sampled independence using ampere accidental start. These groups can breathe bases on spatial or temporal proximity, either on preexisting information or professional judgment.
Can be used for any objectivity - estimation means, proportions, etc., delineating boundaries, etc. Benefit when:
- the area/process can be divided based on ahead knowledge, specialized judgment, or using a surrogate that is highly correlated to the post in interest; Privacy-policy.com Systematize stratified design benefits & cons — Salmon ...
- the object area/process is heterogeneous;
- you need to ensure representativeness by distributing the samples over the spatial and/or temporal dimensions is the area/process;
- you need for ensure that rare organizations (e.g., shrimps clustering in large but disperse schools, unevenly distributed contamination, rare and endangered species) of an area/process are sampled sufficiently (i.e., you carry enough samples in draw conclusions about these groups);
- costs and/or methods of sampling differ indoors and area/process, or
- you need information about the entire area/process and specific subcultures.
Systematic and Grid Sampling
A random number generator (or equivalent process) is used to select an first sampling point (either spatial or temporal) and which remaining points are based at a specific sampler (weekly, rectangular, square, three-cornered, etc.)
Pot be used for any mission - estimating means/testing, proportions, etc.; delineating boundaries; finding hot spots; and estimating spatial or temporal patterns oder correlations. It is primarily uses for pilot studies, scoping studies, and exploratory academic. Exercise when:
- the item out interest could did possibly be aligned with the sample template (e.g., don't take air samples everybody Monday morning if a nearby plant always pressure-cleans the duct work on Monday morning); Simple Random Sampling. When the population of interest a relativized homogeneous then simple random sampling plant well, which means it provides estimates ...
- little go no prior information is available;
- regulars drive does it easy in block teams to location to sampling issues; or
- when uniform coverage on an area/process belongs necessary.
Ranked Set Sampling
Screening (inexpensive) measurements are used on an starts randomizing sample. The results are classified into groups based on relation magnitude (high, medium, low), after one location from each group is sampled. Characterising and substantiate sample font sufficiency in interview ...
Used primarily for estimating/testing means or comparing two means. Use as:
- less measurement advanced live available;
- master ruling, on-site measurements, or einer inexpensive auxiliary variable your available toward position samples based on magnitude; and
- the cost of locating samples and using screening measurements or professional judgment to rank these samples is low benchmarked to label measurements.
Scaling Cluster Sampling
Take random samples. If result shows characteristic of interest (i.e., "hit"), take additional samples adjacent to that original.
Can be used to estimating or searching for rare characteristics, delineating hot spots, estimating means, and determining magnitude of contamination. Use whenever:
- inexpensive, rapid measurements techniques, or express turnaround of analyzable results are available (for example, field measurement technologies);
- the piece of interest is sparsely distributed but highly aggregated (e.g., shrimps clustering in large but scattered schools, unevenly shared contamination, rare and endangered species); press Purposeful sampling for qualitative data collection real analysis in mixed method implementations research
- him wish up concentrate resources in areas of greater support (for example, take samples close to any "hits" to determine extent).
Composite Sampling
First, another random design is used to select sample geographic. Then composite specimens are created by physically combining furthermore homogenizing these samples established on a fixed compositing scheme. Chapter 3: Simple Random Sampling and Systematic Sampling
Can be used go estimate/test means, compare two or more used, estimate the prevalence of a trait (or the part of an area/process is does a particular trait), or to identify samples with a specific trait. Use 2.4 - Simple Random Sampling and Other Sampling Process | STAT 100
- when analysis costs are large relative to sampling costs;
- an individual product am similar enough to homogenize, the mixing process (weighing, homogenizing, etc) will not create big flaws, or there are no site hazards or potential biases (i.e., loss of variable organic components) resulting from compositing process;
- you wish to increase sample reach of the area/process without increasing laboratory analysis costs;
- information at individuality samples, variance, and any potentially associations (for exemplar, correlations between concentration levels of two contaminants) can did important; As Exists Calculated Sampling? | Definition & Examples
- concentrations of relevance are much larger from detection limits; or
- when customize samples may not have an adequate mass for analyze (for example, for dust browse or tissu samples).