Comm 304 (8-10)
Terms
undefined, object
copy deck
- estimation
- Through the use of inferential statistics, we can estimate the characteristics of a population based upon our sample
- survey
- we need to carefully explicate our concepts and generate valid, easily understood measures of each concept; hoosing an appropriate sample.
- Response rate
- reflects the percentage of those individuals selected for the sample who actually participate in the survey
- Cohort studies
- a particular age group is tracked over time
- Panel Studies
- contact the same individuals at each wave of the survey
- Sampling error
- deviation between the sample statistic and the population parameter
- Confidence interval
- represents the interval within which we are reasonably confident the parameter falls
- Stratified sampling
- researchers first divide the overall population into different subgroups, or strata; A random sample is then drawn from each group;
- cross-sectional
- Surveys that involve a single, one-time measure
- Trend studies
- ask the same questions multiple times, looking for patterns or trends in data; a new sample is selected for each wave of the survey
- Requirements for random selection
- all individuals in the population must have an equal chance of being selected; the selection of each individual must be independent
- Statistic/Parameter
- the equivalent numerical summary for the larger population from which the sample is drawn
- interviews
- involve having a trained individual ask questions and then record the responses
- questionnaires
- printed sets of questions that are filled out by the participants, can be completed quickly and by many individuals at once, without requiring a research team to pay a large number of interviewers
- Longitudinal design
- measures are made at multiple points in time
- Cluster sampling
- we first randomly select a sample from some broader variable, such as county or census district; we would then randomly select a narrower variable, such as street or community, and eventually we would randomly select individuals from within those narrow categories
- sampling frame
- a list of every member of the population
- sample
- a subset of some larger population
- population
- group of individuals about whom we wish to make conclusions
- Deliberate sampling
- researchers choose to focus on locations or contact methods likely to attract particular types of participants
- external validity
- he results from our particular sample will not effectively represent the larger population
- Confidence level
- When we decide just how sure we want to be that our estimate includes the population parameter
- Margin of error
- plus-minus value; will only apply to statistics describing the entire sample