APEX AP Stats 4.2 Overview Methods of Data Collection- Surveys

Bias

Systematic favoritism toward certain outcomes; it means that the sample may not accurately represent the population in survey sampling.

Cluster Sample

A sample that randomly selects naturally occurring groups (or clusters) from the population, and then samples all or part of each selected cluster.

Convenience Sample

A haphazard sample, or one that is nonrandomly selected on the basis of convenience; doesn't typically yield useful data for research because it doesn't represent the population.

Inference

The probability-based process of drawing conclusions about a population from a sample.

Interviewer-Induced Bias

Response bias that occurs when a respondent's answers are influenced by the personal characteristics or habits of the interviewer.

Nonresponse Bias

The response bias that occurs when some people don't return the survey. This is biasing because we can probably assume that those who don't return the survey are different in some important way from those who do.

Parameter

A numerical fact about a population; typically unknown.

Probability Sample

A sample in which the probability of any one case being selected is known; a type of random sample.

Random Sample

A sample that uses randomness (as opposed to the convenience of the researcher) to assign members to it to control bias.

Representative Sample

A sample that is like the population from which it is drawn with respect to characteristics that are important to the study.

Sampling Frame

A list of population members from which one draws a random sample.

Response Bias

Results when respondents answer survey questions in the manner they think the researcher wants them to answer. This bias is most often a result of the wording of the questions.

Self-Selected Sample

A type of convenience sample where subjects volunteer themselves. It usually does not yield useful data for research.

Simple Random Sample (SRS)

A sample drawn so that every member of the population has an equal probability of being selected into the sample and every population subset of size n has an equal probability of being selected in the sample.

Statistic

A known numerical fact about a sample, used as an estimate of the parameter of the population.

Statistical Significance

A situation in which the probability of an outcome occurring by chance alone is so small that we must conclude there's something other than chance operating to produce that outcome.

Strata

A population subgroup that has a characteristic important to the study, such as gender, age, or race.

Stratified Random Sample

A sample that includes population members who are randomly selected from within strata identified in advance; the strata is chosen so that the individuals within the same stratum are as similar as possible.

Systematic Random Sample

A sample drawn by selecting every nth case from a randomly ordered list, after the first case is randomly selected within the first interval.

Under-Coverage

A form of bias that occurs when the sampling strategy fails to identify a part of the population to be surveyed. Often the poor or very mobile are under-represented because they're more difficult to locate.

Untruthful Answers

A form of response bias introduced when respondents give a socially acceptable answer or an answer they believe the interviewer wants to hear.

Voluntary Response Bias

Self-selected sample; a type of nonresponse bias that is biased because of missing groups in the sample (not knowing what those missing individuals would have responded with or if they would have responded the same way as those who did respond).