Selecting Research Participants: Chapter 5 Flashcards
population
group sharing some common characteristics
sample population
- subset of the population
- people selected for the study
representative sample goal
- to select samples that are similar to the populations
- if the sample represents the population, the results of the study can be generalized to the population
- refer to PowerPoint 5, slide 4 for diagram
target population
group defined by the researcherβs specific interests
accessible population
- easily available segment of a target population
- researchers typically select their samples from this type of population
representative sample
sample which has the same characteristics as the population
bias in regard to representativeness
- bias is a major threat to representativeness
- biased samples characteristics are very different from the population
- bias arises from sampling bias
ways to get a biased sample
- Sampling only those who are easy to contact, like a convenience sampling
- sampling only those who volunteer, like self-selection (volunteering) sampling
sample size
- large sample will probably be more representative than a small one
- minimum of 10 participants is required for statistical purposes
power analysis
- to determine the sample size needed to obtain the expected results with a given degree of confidence
Law of large numbers
the larger the sample size, the more likely it is that values obtained from the sample are similar to the actual values for the population
categories of sampling
- non-probability sampling
- probability sampling
non-probability sampling issues
- exact size of the population is NOT known, and it is NOT possible to list all the individuals in the population
- probability each individual has to be selected in the sample is UNKNOWN
- selection process is NOT unbiased
- greater risk of producing a biased sample than probability sampling
probability sampling
- simple random sampling
- systematic random sampling
- stratified random sampling
- proportionate stratified random sampling
- cluster random sampling
- multistage random sampling
simple random sampling
- equality: each individual has an equal chance of selection.
- independence: choice of one individual does not influence the probability of choosing another individual.
- E.g., Draw names out of a hat, use a
random number table
systematic random sampling
sample members from a larger population are selected according to a random starting point and a fixed, periodic interval
* Entire population is enumerated in a list
* Random starting point
* Every nth person
stratified random sampling
- population is divided into subgroups (strata); equal numbers are then randomly selected from each of the subgroups.
- guarantees that each subgroup will have adequate representation
- overall sample is usually not representative of the population
- useful when the goal is to make comparisons among subgroups
proportionate stratified random sampling
- population is subdivided into strata.
- number of participants from each stratum is selected randomly.
- proportions in the sample correspond
to the proportions in the population. - ensures the sample will be representative of the
population
cluster random sampling
- clusters (preexisting groups) instead of individuals are randomly selected from a list of all the clusters that exist within the population
- all members of the selected clusters comprise the sample
- easy method for obtaining a large, relatively random sample
multistage random sampling
- random sampling at multiple stages
- effective in choosing a sample that is representative of a widely dispersed population
E.g.:
Canadian university professorsβ
opinions on student literacy - Stage 1: Randomly select universities
across Canada - Stage 2: Randomly select departments
within universities - Stage 3: Randomly select professors
within departments
methods of simple random sampling
- sampling with replacement
- sampling without replacement
issues with simple random sampling
It is possible (although usually unlikely) to obtain a very distorted sample because of chance
systematic random sampling: how to figure out n?
Select a random sample of 100 participants from a
population of 50,000.
* Step 1: Target population must be placed in a list
* Step 2: Select the interval: ππππ’πππ‘πππ π ππ§π /
ππ πππππ π ππ§π, π
= 500
* Step 3: Randomly pick a number from 1 to 500, e.g.,
342
* Step 4: Start with 342 and pick every 500th person
on the list after that number
issues with systematic random sampling
ensures a high degree of representativeness, though it may violate the principle of independence
issues with stratified random sampling
- does not adequately represent proportions found in the population
- individuals in the population have different probabilities of being selected
issues with proportionate stratified random sampling
- requires a lot of work
- may make it difficult or impossible to compare subgroups within strata
issues with cluster random sampling
- selections are not independent
- E.g., students in one classroom are more likely to be similar to each other than students in another classroom)
issues with multistage random sampling
- cost- and time-effective
- could be issues with nonindependence
probability random sampling issues
- methods can be unrealistic
- too much time and effort
- may not be practical or possible
- need to know the population (i.e., have a list of all members of the population)
non-probability random sampling
- Convenience Sampling
- Quota Sampling
- Purposive Sampling
- Snowball Sampling
convenience sampling
- individual participants are obtained by
selecting those who are available and willing - easy method
- less expensive, time efficient than probability
sampling methods - select a reasonably representative sample and
clearly describe the selection process to help
correct the problems with this form of sampling
issues with convenience sampling
- a weak form of sampling
- probably biased, especially if participants volunteer to participate
quota sampling
- subgroups are identified to be included
- quotas are established for individuals to be selected through convenience from each subgroup
- reflect proportions in population
- allows controlling the composition of a convenience sample
- easy, economical, and time-efficient
issues with quota sampling
- sample is probably biased
- not randomly selected
purposive sampling
- known as judgmental, selective, or subjective sampling
- requires prior knowledge about the purpose of their study in order to know who to select and how to approach eligible participants
snowball sampling
- begin with someone who meets the criteria for inclusion in your study
- then ask them to recommend others who they may know who also meets the criteria
- most used in hard-to-reach
populations
cluster random sampling example
looking at a school, and each cluster is a classroom
we pick among the clusters
