Surveys, Observations, and Sampling Flashcards
What are the 4 question formats?
Open-ended, forced-choice, likert scale, and semantic differential format.
What are forced-choice questions?
Questions in which there respondents give their opinion by picking the best of two or more options.
What is a likert scale?
A scale containing multiple response options that are anchored by the terms “strongly agree” to “strongly disagree.”
What is semantic differential format?
A self-report response scale whose numbers are anchored with contrasting adjectives (e.g., easy vs. hard).
What is a response set?
A shortcut that respondents may use to answer the items in a self-report measure with multiple items, rather than the content of each item.
When are self-reports especially inaccurate?
When respondents are asked why they did something.
How is one’s confidence in a story related to its accuracy?
There is no correlation between confidence and accuracy.
How can question order influence the validity of the measure?
Earlier questions can influence answers to later questions.
What is a potential threat to the construct validity of observational measures?
Observer bias.
What is observer bias?
When the observer’s expectations influence their interoperation of the subject’s behaviours or the outcome of the study.
What is reactivity?
The change in behaviour of an animal or person due to being watched.
What can be done to reduce reactivity?
Make unobtrusive observations, let the subject become accustomed, or measure the traces that a behaviour leaves behind.
What is a census?
A set of observations that contain all members of the population of interest.
When is a sample biased?
When it is unrepresentative of the population.
What causes biased samples?
Sampling those who are easy to contact, sampling those who are able to be contacted, and sampling only those who initiate themselves.
How does one obtain a representative sample?
Take a random sample in which every member of the population of interest has an equal chance of being selected.
What are 5 variants of probability sampling?
Simple random sampling, cluster sampling, multistage sampling, stratified random sampling, and systematic sampling.
What is simple random sampling?
The most basic form of probability sampling, in which the sample is chosen completely at random.
What is cluster sampling?
When researchers randomly select clusters of participants within the population of interest and then collect data from all the participants in each cluster.
What is multistage sampling?
A random sample of clusters are taken, then a random sample of people from within those clusters are taken.
What is stratified random sampling?
The researcher identifies particular demographic categories of interest and then randomly selects individuals within each of the categories.
What is oversampling?
A variation of stratified random sampling in which one or more groups are intentionally overrepresented.
What is systematic sampling?
The researcher counts off to achieve a sample. E.g., choosing every nth person in a population, where n is a randomly chosen number.
What are two types of non-representative sampling methods?
Convenience sampling and purposive sampling.
What is convenience sampling?
Choosing a sample based on those who are easiest to access.
What is purposive sampling?
The inclusion of only certain kinds of people in a sample.
What is snowball sampling?
A variation on purposive sampling in which participants are asked to recommend acquaintances for a study.
When does external validity matter and not?
It matters in frequency claims, but it amy not be a priority when researchers study association or causal relationships.
What is a margin of error?
A statistical term that quantifies the degree of sampling error in a study’s results.
At one sample size does going bigger become unadvantageous?
After 1000 participants, the decrease in margin of error is too small to be worth the effort of finding a larger sample.
