Ch.8, Surveys Flashcards
(35 cards)
Non-Probability Sampling:
uses non random selection, not everyopmne has a chance of being incllued
Does not always provide a completely accurate representation of the population of interest and may therefore be biased
Only used when constraints to time and cost may make it impossible to attain a probability sample
Probability Sampling:
uses random selection, strong external validity
Random Sampling: e
each individual of population has equal chance to be chosen
Convenience Sampling, Non-Probability:
inexpensive, method based on the convenience of surveying people who are easy to reach (whoever is closest in proximity)
Snowball Sampling:
method used to collect data when it’s difficult to locate respondents, commonly used on social media sites to collect data, usually requested that people tell their friends and therefore creates the snowball
Simple Random Sampling, Probability
Simple Random Sampling: subset of population where each member has an equal chance to be chosen
Stratified Random Sampling, Probability
population is divided into subgroups called strata (stratum: shared characteristic like gender or age that can be used to segment the population), superior to simple random sampling because it reduces the difference of the selected sample from the actual population; it ensures each subgroup within the population receives proper representation within the sample.
Cluster Sampling, Probability. Difference from Stratified sampling?
uses natural but relatively heterogeneous population, randomly selected strata are sampled (women, middle aged), quick, easy, inexpensive
—In Cluster Sampling, the sampling is done on a WIDE population of clusters therefore, cluster/group is considered a sampling unit UNTO ITS OWN. In Stratified Sampling, SPECIFIC elements within each stratum are sampled. In Cluster Sampling, only selected clusters are sampled. In Stratified Sampling, from each INDIVIDUAL stratum, a random sample is selected.
Sampling Distribution:
a probability distribution of a statistic that is obtained through repeated sampling of a specific population. It describes a range of possible outcomes for a statistic, such as the mean or mode of some variable, of a population.
Sampling error
difference between the estimate of the population characteristic and its true, unknown value, occurring when the results from a sample do not
Research-Based Survey: vs Evaluation Survey:
Research-Based Survey: survey used to make decisions (entail collecting, analyzing, and interpreting results to describe the level of an attribute/behavior etc)
Evaluation Survey: survey used to make decisions (drive organizational change, improve effectiveness, etc.)
Mixed-Mode Surveys:
in which multiple methods are used to contact and survey respondents to increase response rate (that way not all participants are limited to using technology)
Group Administration:
respondents are brought together and asked to respond to list of questions
Focus Groups:
: individuals are asked about their attitudes or perceptions toward a specific thing should be 10-12 people
Dichotomous Questions, qualitative
question can only be answered in one of two indicated ways
Forced Choice:, qualitative
makes respondent choose definitive answer
Scale Format Questions, Quantitative
ask respondents to describe a situation along a continuum; Likert Type Items
Interval/Numeric Question:
most precise measure available, used when we want a precise difference assesses along a skil (ADVANTAGE OVER LIKERT)
Semantic Differential Question, quantitative
Collects interval data in which the respondent is asked to assess an attribute on a set of contradicting adjective pairing (light/dark)
Used to measure attitudes about idea, concepts, items, people/events
Cumulative/Guttman Scale
Interval data
Respondent checks all items with which they agree
Filter/Contingency Questions, Quantitative
Determine whether the respondent is qualified to answer the subsequent question
Survey Habituation:
series of questions that all have the same answer choices; respondents may learn to ignore a question because of repeated exposure to the format
iNTRA-Rater Reliability
rater rates the same behavior consistently over time
Internal Consistency:
refers to the degree that survey items are measuring the same construct