Quiz 3
Terms
- Survey Research
-
A data collection technique
in which information is collected
by having individuals respond to questions
- 3 Types of survey research
-
Self-administered questionnaires
Interview surveys
Telephone surveys
- Closed ended questions
-
The respondent is asked to select an answer
from among a fixed set of alternatives provided by the researcher
these are mutually exclusive categories
and exhaustive
-
Open-ended questions
-
The respondent is asked to
provide his/her own answer to the question.
These are more exploratory
and used when all response categories are not known.
-
Questions should be
-
Clear and unambiguous.
Avoid reference to things that arenot clearly defined or that respondents would have to interpret.
-
Questions should
-
Express only one idea.
Avoid double barreled questions.
- Question wording
-
Should not be biased
-
Six other things to think about in question design:
-
- Avoid negative items.
- Respondents must be competent to answer
- Respondents must be willing to answer
- Questions should be relevant.
- Short items are best.
- Be sure to pretest your instrument.
-
Compositve measures
-
Composed of several items that have a logical or empirical structure.
-
Answers to multiple questions are calculated together to form a value of construct by
-
- Summing the values of individual items
- Weighting and summing values of items.
-
Most standardized instruments are
-
Composite measures
-
3 formats of composite measures
-
- Likert scales
- Dichotomous questions
- Semantic differential format
- Likert scales
-
A series of statements, each followed by 4 or 5 response alternatives
-
Semantic differential format
-
Respondents choose between two opposite positions on a ranked scale
- 2 ways to avoide response sets
-
- Make sure all items are not scaled in the same direction/reverse score items.
- Use an even number of response choices so that a middle or neutral option is not given.
- 4 methods of administrating social surveys
-
- Computer assisted (i'net, computer/phone, laptop)
- Mail distribution
- Face-to-face interviewing
- Telephone interviewing
- 4 advantages of questionnaires
-
- Quick and inexpensive
- Allows one to collect data from geographically disperse locations
- Eliminates interviewer bias
- Good for sensitive issues
- 4 disadvantages of questionnaires
-
- Requires certain degree of literacy and facility in English
- Questions must be sifficiently simple to comprehend
- No opportunity to probe
- Possible selection bias/low response rates (~20-90%)
- Unstandardized interview guide approach
- Least structured; interviewer has general topic area
- Non-schedule-standardized
-
More structure; more narrow topic & specific questions asked of all respondents
-
Schedule-standardized interview guide approach
-
Interview schedule with specific questions; fixed order and transition phases are used.
-
5 advantages of interviews
-
- may collect more accurate and observational information
- Interviewer can explain questions and clarify answers
- Offers more control and more flexibility
- Higher response rates
- Good for complex issues
-
4 disadvantages of interviews
-
- Potential for interviewer bias/contamination of results by interviewer influence
- Safety concerns
- Increased costs
- Increased time for interviewer training and travel
-
Group depth interview
-
A facilitated interview with a whole group of people at the same time
- 5 adavantages of focus groups
-
- Able to assess people's subjective reactions
- Allows interviers to get at many levels of meaning
- Flexible
- Inexpensive
- Quick Results
- 2 disadvantages of focus groups
-
- Less generalizability
- Difficult to analyze - lots of information
-
4 items about secondary analysis of surveys
-
- Uses previously collected data
- Generally inexpensive and fast
- Questions aren't necessarily asked in the format that would best suit your research purpose
- May be misused (i.e., develop questions or hypotheses to fit the
-
4 reasons to draw a sample
-
- Some times entire population is too large to study
- A sample may be more time-efficient, more cost effective, & more feasible
- Samples can provide a close approximation of the population (limited by type of sampling techniques us
- Element
- That unit about which information is collected and that provides the basis for analysis
- Population
- The theoretically-specified aggregation of study elements; all possible cases of what we are interested in studying and to whom we want to generalize
-
Study population
-
That aggregation of elements from which the sample is actually drawn
- Sampling frame
-
A listing of all the elements in a population
- Sample
-
One or more elements selected from a population
-
Representative sample
-
A sample that accurately reflects distribution of relevant variables in the population; allows generalization to population
-
2 types of sampling methods
-
- Probability sampling
- Non-probability sampling
-
Probability sampling
-
Each element has a chance, often equal, of being included in the sample.
Used for generalization to the population
-
Non-probability sampling
-
The probability of each population element being included in the sample is not known; generally a less accurate representation of the population
Used when probability sampling is not an option or generalization is not needed.
- 4 nonprobability sampling designs
-
- Convenience/availability
- Purposive/judgmental
- Quota sampling
- Snowball sampling
-
Convenience or availability sampling
-
Using elements readily available to the researcher
--cannot develop a complete sampling frame; exploratory
--not very accurate; cannot generalize
- Purposive/judgmental
-
Inverviewer's own judgment and prior knowledge of the population are used to choose persons for sample that best serve the purposes of the study
--provides some control of sample
--exploratory studies
-
Quota sampling
-
Uses a matrix of the target populations characteristics to establish a quota sample specifying the number of elements to be selected from each category
--similar to stratified sampling
--used when sampling frame cannot be estab
-
Snowball sampling
-
Collect data on the few members of target population available, then request them for information leading to more people, who lead to more people, and so on
--good for hard to access populations
--often used in qualititative fi
-
Random selection
- Each element has an equal chance of being selected
- Parameter
-
Aggregate data from a population; the summary description of a given variable in a population
- Statistic
-
Aggregate data from a sample; the summary description of a given variable in a sample; used to estimate population parameters
-
Standard error
- Estimate of the extent to which the values of the sample differ from those of the population.
-
Probability theory
-
Ultimate purpose of sampling is to selecta set of elements from a population so that sample statistics accurately portray total population parameters.
--requires random selection
-
Sampling distribution
-
Theoretically taking repeated random samples of a single population and plotting the summary statistics on a graph
- Probability theory states that if we collected all possible samples of a population, the statistical means would cluster in a formation called
-
a normal curve, such that the distribution of sample statistics cluster around the population parameter in a known way