Chapter 7 Flashcards
Using questionnaires and interviews to gather information about people
survey
All the cases of interest
population
A group from the population
sample
A list of names that the sample can be taken from and represents the operational definition of the population
sampling frame
Order of finding your sample
Population > Sampling frame > Sample
A sample that reflects the population
representative sample
a sample that does not reflect important characteristics of the population and can create confounding variables.
nonrepresentative or biased sample
The average amount of cases in a population who complete the survey
response rate
Limitations of surveys
- Cannot examine cause-effect reactions
- Poor sampling creates poor results
- Everything is self-reported, so there is no guarantee that the responses are their actual beliefs
A tendency to respond in a way that a person feels is socially appropriate, rather than as they truly feel
social desirability bias
Each member of the population has a chance of being selected. The probability of each member can be changed
Probability sampling
Each member of the population has a chance of being selected. The probability of each member being selected cannot be changed
nonprobability sampling
Every member of the sampling frame has an equal probability of being chosen
simple random sampling
Using demographic statistics to increase the probability that certain members will be selected
stratified random sampling
Statistical units that contain certain members of the population are identified to be randomly selected from
cluster sampling
All participants in the randomly selected cluster are chosen to participate
single-stage cluster sampling
A social unit that can be randomly selected or have participants selected from
cluster
The use of two or more stages of sampling to select participants
multistage sampling
Members of a population are selected nonrandomly for inclusion in a sample on the basis of convenience
convenience sampling
A sample is nonrandomly selected to match a specific quota
quota sampling
Occurs when participants who are currently in a specific situation agree to be a part of the sample
self-selection
Researchers select a sample according to a specific goal or purpose
purposive sampling
Researchers identify experts to participate
expert sampling
People contacted to participate are asked to contact more people to participate. (like a pyramid scheme)
snowball sampling
Chance differences in the characteristics of samples that occur when randomly selecting samples from a population
sampling variability
When there is a margin of error in the characteristics of the selected participants
margin of sampling error
The degree of confidence that the sample reflects the true population
confidence level
Causes of sampling error
- Chance fluctuations when statistics are used to estimate population clusters
- Population selection
Advantages of face-to-face interviews
- achieve higher response rates then other forms of getting participants
- can establish a personal rapport to encourage response to sensitive questions
- ensure all participants answer the questions in the same other
- can identify any verbal or facial cues that occur due to uncertainty
Limitations of face-to-face interviews
- high cost
- differences in delivery of questions
- participant cannot remember the question
- general characteristics of the interviewer
Anything that distorts the participant’s response that are characteristics of the interviewer
interviewer effects
Why we don’t use telephone surveys that often
- can’t establish rapport, uses computers
- mentally taxing to listen to a computer
- can’t evaluate facial cues
- Many people don’t have landlines, and cellphone users respond differently then landline users
Why we don’t do mail surveys
- high cost
- not many finish it / low response rate
Advantages of online survey
- low cost
- faster
- can use online social networks to find people
limitations of online survey
- lower response rate
- no interaction with the participant
- participant inattention to answers
When an online survey goes up on the web for people to randomly come across
online convenience sampling
Using Internet services to find participants. So online services like youtube survey their users
crowdsourcing
a collection of people who have consented to be contacted to take survey are contacted
online opt-in survey panels
When people who were selected but didn’t participate in a survey would have provided significantly different answers
nonresponse bias
When people who were selected but didn’t participate in a survey would have provided significantly different answers
nonresponse bias
Steps to make a survey
- understand research goals
- identify variables of interest
- consider limitations
- develop questions
- pretest
- revise
Raters using the same coding system agree on how to score participants
interrater reliability
Questions that lead responsders to an answer
leading questions
Emotionally charged questions
loaded question
Asking two questions at once
double-barreled questions
question phrasing that contains two negatives
double negatives
When the placement of a question changes the results
order effects
when responses are influenced by items around the question
context effects
