Chapter 7: Survey Research & Sampling* Flashcards
Panel study or longitudinal design
administering survey questions to the same people at 2 or more points in time
Response set/bias
a pattern or tendency of responding to questions in a certain way rather than providing answers directly related to the questions e.g. social desirability, yea-saying, nay-saying, fence-sitting
Social desirability or “faking good”
an individual answers in the most socially acceptable way or the way he/she thinks most people would respond
3 general types of content measured by survey questions
attitudes and beliefs, facts and demographics, behaviors
What are some problems that can arise when writing questions?
unnecessary complexity, negative wording, double-barreled questions, loaded questions
Double-barreled questions
questions that ask two things at once, making any one answer ambiguous and hard to interpret
Loaded questions
written in such a way as to try to bias people’s response toward a particular answer
“Yea-saying” (acquiescence) or “nay-saying” response sets
tendency for some survey respondents to agree (yea) or disagree (nay) with the vast majority of questions being asked regardless of question content
Inattentive responding
responding to survey questions without reading the item content, providing answers that are not thoughtful
What are the 2 main types of questions?
closed-ended and open-ended questions
Closed-ended questions
questions that offer respondents a limited number of response options e.g. multiple choice, T/F, Likert scale
Open-ended questions
questions that allow respondents to answer in any way they like with no restrictions e.g. short/long essay
Pros and cons of open-ended questions
Pros: freedom for participants to respond how they like, good ecological validity; Cons: tough to quantify and code
Pros and cons of closed-ended questions
Pros: easy to implement, do stats on, compare results with other similar surveys, and restricts participant responses; Cons: restricts participant responses, designing questions can be tricky
Rating scale
closed-ended response question that asks participants about degrees of judgement for a dimension e.g. amount of agreement, liking, or confidence
Graphic rating scale
two words appear on either side of a solid line and participants place a mark on the line indicating their relative preference for one or the other word, which is measured in term of distance from one end
Semantic differential scale
two words appear on either side of a series of dashed lines and participants place a mark on the dash indicating their relative preference for one or the other word
3 basic dimensions concepts are rated in using semantic differential scales
evaluation (e.g. good-bad), activity (e.g. slow-fast), potency (e.g. weak-strong)
Non-verbal scale
response options are in the form of images instead of words or numbers
What are some solutions to response set/bias?
use reverse-worded questions, get rid of neutral alternatives
What can you do about question order effects?
insert filler questions, counterbalance the questions
What are the 2 main ways to administer surveys?
questionnaires and interviews
Interviewer bias
intentional or unintentional influence on a respondent by an interviewer to encourage responses consistent with the interviewer’s expectations
Focus group
qualitative method of data collection where 6 to 10 people are interviewed together about a particular topic
Population
group of people of interest to the researchers, from which a sample is typically drawn
Sample
subset of the population
Confidence interval
serves as a range of plausible values that are likely to be observed if the study were to be repeated; a quantification of our uncertainty around our estimates (wider CI = greater uncertainty)
Sampling error
the degree to which the estimate based on a sample deviates from the true population value
External validity
the degree to which study results based on a sample may be generalized to the population from which the sample was drawn or other populations
Sampling frame
the individuals or clusters of individuals in a population who may actually be selected for inclusion in the sample
Response rate
the percentage of people selected for a sample who actually complete a survey
What are the 2 basic techniques for sampling?
probability sampling and non-probability sampling
Probability sampling
participants are randomly drawn and each member of the population has a known and specific probability of being chosen; likely representative of population and high generalizability
Non-probability sampling
participants are not drawn at random and one cannot specify the probability that any member of a population will be included; easy and convenient but low generalizability
What are the types of probability sampling?
simple random sampling, stratified random sampling, and cluster sampling
Simple random sampling
each member of the population has an equal probability of being included in the sample
When is simple random sampling used?
when you have access to your entire population and have more or less equal representation of all groups contained in your population
Stratified random sampling
the population is divided into strata or subgroups followed by random sampling from each stratum to reflect their proportions in the population
When is stratified random sampling used?
when you have access to your entire population and you have unequal representation of all groups contained in your population
Cluster sampling
clusters of individuals are identified, sampled, then all individuals in each cluster are included in the sample
When is cluster sampling used?
when you don’t have access to your entire population
What are the types of non-probability sampling/
convenience sampling, purposive sampling, quota sampling, snowball sampling
Convenience or haphazard sampling
recruiting participants whenever and wherever you can on the basis of availability and without regard for the representativeness of the sample
Purposive sampling
a type of convenience sampling procedure conducted to obtain only a sample of people who meet a predetermined criteria
Purposive sampling
a type of convenience sampling procedure conducted to obtain only a sample of people who meet a predetermined criteria
Quota sampling
the sample is chosen using convenience sampling to reflect the numerical composition of various subgroups in the population
Snowball sampling
asking participants to recommend others to participate and help researchers accumulate participants