Sampling Procedures

(Part 2 of Blog Prompt 3)

Sampling Procedures for Quantitative Studies

Procedure:	Definition:	Pros:	Cons:
Simple Random Sampling	Each member of the population has the same probability of being selected.	Decreases any chance of sampling bias.	It is difficult to find a list of all the members of the population being studied. Also, it generalizes.
Systematic Sampling	Every nth element is elected from a list of all elements in the sampling frame, beginning with a randomly selected element.	Also decreases sampling bias.	Same as for Simple Random Sampling.
Stratified Sampling	Modification of simple random or systematic sampling in which the population is divided into homogeneous subgroups, and then subjects are selected from each subgroup.	This is better because it is more representative of the population that is being studied because each subgroup is represented.	Just like with the previous sampling styles, it is really difficult to get information on these students due to confidentiality issues.
Proportional Stratified Sample	When the number of subjects selected from each subgroup represents the percentage of subjects in the population have the characteristics of the subgroup.	If there was a list of the population being researched, then this would be great because we could use a proportional sample compared to the total.	However, there is not a list, and just as above, this would be very difficult to do. Also, the difference between subgroups may not make much difference in outcome.
Disproportional Stratified Sample	The number of subjects from each subgroup does not represent the percentage compared to the population.	Easier to figure out than the proportional.	Does not accurately represent the population.
Cluster Sampling	Subjects are selected from naturally occurring groups (clusters) like universities, school districts, etc.	Easier to find participants.	Excludes those that are not within the cluster.
Convenience Sample	Subjects are selected based on availability and ease.	It's super easy.	Most likely will not give you accurate representative information.
Quota Sampling	Subjects are selected nonrandomly to represent composite profiles of major groups in the population.	Easier to do because the sample is smaller.	Generalizes to the population. Could have sample bias.

Sampling Procedures for Qualitative Studies

Procedure:	Definition:	Pros:	Cons:
Purposive Sampling	Subjects are selected because they will be particularly informative about the topic.	You only have participants that are useful to you. I could use this method for my question within my school.	It is typically a very small sample and may not represent the theoretical population.
Criterion Sampling	Participants are selected based on identified characteristics that provide needed information.	This is the type of Purposive Sampling that I could use by checking through school records for students that fit the homeless/transient criteria.	Same as above, this does not tell you anything about the theoretical population.
Typical Case Sampling	Subjects are selected because they are “typical” or “representative” of many.	Would give you an idea of  "typical" homeless students.	Very generalized, and has room for sampling bias and error.
Extreme Case Sampling	The opposite of typical case, subjects are selected because they are unique participants.	This is helpful for showing just how much of an effect being homeless/transient can have on behavior, to the extreme.	The extreme may not be relevant to the majority. Also, the extreme may not be available for studying.
Maximum Variation Sampling	Subjects are selected to represent both extremes of characteristics of interest.	The sample would give you comparative information.	This is twice as much work and the comparative info may be irrelevant.
Snowball Sampling	Subjects are selected and then asked to nominate or recommend others who are qualified.	I could reach students that I may not have known were qualified.	There is no system to who is chosen and the subjects may not be who you want to study.
Critical Case Sampling	Subjects are selected based on how they illustrate the phenomenon of interest.	I would be able to select students that I know are homeless/transient.	I would not reach students that do not have it recorded that they were homeless/transient.

This was difficult when trying to think of pro's and con's in relation to my question, considering most of them would not work or be possible because of the limited access to student records.

As far as I can tell, I would say that Criterion Sampling would be the best bet for my research question. As the school counselor, I would have access to the school records which I could use to search for students that fit the criteria of homeless/transient.