Sampling Flashcards
Target Population
The total collection of things that we seek information about.
Sampling Frame
The largest sample you have access to from this target population.
Sample
A collection of those things drawn from the population.
What is a question to ask about the sample in reference to the target population?
Is the sample representative of the target population?
What is a question to ask in regards to the individuals of the sample?
How did the participants get into the study.
What should representative sampling be held to?
The highest standard for frequency and association claims.
Biased Sample
An unrepresented sample where some members of the target population have a higher probability of being included compared to other members.
What are the methods of biased sampling referred to as?
Nonprobability Sampling
What are the 4 types of Nonprobability Sampling?
- Convenience
- Purposive
- Snowball
- Self-Selected / Quota
Convenience Sampling
Researcher requests volunteers from a group who meet general criteria but are recruited in a variety of “non-random” ways.
Purposive Sample
Researchers target or hand pick participants that meet specific criteria based on knowledge of the research question.
Snow Ball Sample
Variant of purposive, but now participants are asked to recommend others they know that would also meet criteria.
Self Selected Sample
Sampling only those who invite themselves to participate in the study.
What type of biased sampling is this situation:
A psychology professor wants to study college students’ stress levels during finals week. Instead of randomly selecting students from the entire university, the professor surveys the students in her own classes because they are easy to access and willing to participate.
Convenience
What type of biased sampling is this situation:
A researcher is studying the experiences of undocumented immigrants in a specific city. Because this population is hard to reach and may not want to disclose their identity, the researcher starts by interviewing one participant who is willing to talk. That participant then refers the researcher to a few other undocumented immigrants they know, who in turn refer others.
Snowball
What type of biased sampling is this situation:
A public health researcher wants to understand how female community health workers in rural areas manage limited medical resources. Instead of surveying all health workers, they specifically select women working in rural clinics who have at least 5 years of experience, because these participants are most likely to provide rich, relevant insights for the study.
Purposive
What type of biased sampling is this situation:
A university researcher posts an online survey about sleep habits on social media and invites anyone interested to participate. Only people who see the post and choose to complete the survey are included in the study.
Self Selected
Representative Sample
Every member of the population has an equal chance of being chosen for the sample
What is representative sampling also known as?
Probability Sampling
or
Random Sampling
What are the 5 types of Probability Sampling?
- Simple Random
- Systematic
- Multi Stage Cluster
- Cluster
- Stratified Random
Simple Random Sampling
There is no pattern or bias to choosing participants, just random drawing from a sample frame.
Systematic Sampling
Pick the first participant at random and then selected every nth person on the list after that.
Cluster Sampling
The population is divided into groups, and ENTIRE clusters are randomly selected to be included in the study.
Multi Stage Cluster Sampling
The population is divided into groups, and the participants are randomly selected from each subgroup created.
Stratified Random Sampling
The population is divided into subgroups based on a characteristic (age, gender, race ,etc.) and a random sample is taken from each subgroup.
What type of representative sampling is this situation:
A teacher wants to select 5 students from a class of 30 to participate in a special project. To do so, she writes down the names of all 30 students on pieces of paper, puts them in a hat, and randomly draws 5 names. Each student has an equal chance of being chosen.
Simple Random
What type of representative sampling is this situation:
A researcher wants to study the academic performance of students in a large city. Instead of selecting individual students from all over the city, the researcher divides the city into districts (clusters). They randomly select 3 districts and then survey all the students within those 3 chosen districts.
Cluster
What type of representative sampling is this situation:
A researcher wants to select 10 students from a list of 100 students for a survey. First, they randomly select a starting point on the list (say, student number 4). Then, they choose every 10th student after that.
Systematic
What type of representative sampling is this situation:
A researcher wants to study the study habits of college students and believes that the habits may differ by year of study (freshman, sophomore, junior, senior). They divide the student population into four strata based on their year (freshman, sophomore, junior, senior), then randomly select 50 students from each group to participate in the study.
Stratified Random
What type of representative sampling is this situation:
A researcher wants to study the health behaviors of people in a large country. The country is divided into regions, and the regions are divided into districts, and the districts are divided into neighborhoods. 5 groups from each category are chosen, until you have the ideal number of participants.
Multistage Cluster
Oversampling
Researcher intentionally over-represents one or more groups to get an accurate estimate from them.
Is oversampling a sample method?
No its an intentional limitation
Why is oversampling not a sampling method?
A variable of interest might have a low proportion in the population, so you would have to oversample to get enough participants for a meaningful analysis.
What is this situation an example of:
Interested in the effects of a bilingual household on when kids learn to read. You would want to compare an equal number of bilingual families to monolinguals, even though there are far more monolingual families in the population.
Oversampling
How do you determine which sample method to use based on these terms of accurate representation preference:
“Not Very”
“Sort of”
“Very”
“Not Very” - Convenience
“Sort of” - Simple Random
“Very” - Stratified Random
