Chapter 7: Sampling - Estimating the Frequency of Behaviors and Beliefs Flashcards
Two most important validities for frequency claims
- Construct validity: What survey questions were used?
How did they operationalize their variables? - External validity: What sampling technique was used? (how did they get their sample)
In what setting was the data collected?
Population
The entire set of people or products in which you are interested.
Sample
A smaller set taken from the population.
Census
A set of observations that contains all members of the population of interest.
Biased sample
Unrepresentative sample.
Some members of the population have a much higher probability of being included in the sample.
Not every participant had equal chance to make it to the sample stage!
Representative sample
Unbiased sample.
All members of the population have an equal chance of being included in the sample (randomized).
Allow us to make inferences about the population of interest.
Determining sample bias
- Specify the population to which you want to generalize.
- Look at how the sample was obtained - was it random or nonrandom? Random samples are representative, while nonrandom samples are unrepresentative of the population of interest.
Convenience sampling
Sampling only those who are easy to contact and readily available to participate. Creates biased samples.
Self-selection
Sample only those who volunteer. Creates biased samples.
Probability sampling
Draw the sample at random from the population, every member has an equal chance of being in the sample, high external validity. AKA random/representative sampling.
Simple random sampling
The most basic form of probability sampling, in which the sample is chosen completely at random from the population of interest. Examples include:
- The names of every single member of the population are put into a hat and someone who is blind-folded pulls out names.
- Random number generator.
Systematic sampling
A probability sampling technique in which the researcher uses a randomly chosen number N (through RNG), and counts off every Nth member of a population to achieve a sample. Example: Start with the 4th person and sample every 7th person.
Cluster sampling
Start with clusters, take a random sample of those clusters, and ALL members of those selected clusters are in your sample.
Example: Interested in college athletes in Virginia (population), start with a list of clusters (all colleges in Virginia), select a random sample of clusters (VCU, Roanoke College, William & Mary, UMW, JMU), and include all athletes from these identified colleges.
Multistage sampling
Start with clusters, take a random sample of those clusters, then RANDOMLY sample from those selected clusters to get your sample. Example: Interested in college athletes in Virginia (population), start with a list of clusters (all colleges in Virginia), select a random sample of clusters (VCU, Roanoke College, William & Mary, UMW, JMU), and then randomly select athletes from these identified colleges.
Stratified random sampling
Select particular demographic categories and randomly select people within each of these categories. Example: Interested in college athletes in Virginia (population), want to make sure to include equal male and female athletes, stratify the population based on gender, then randomly sample from each category.
