PBSI 245 Chapter 4: Sample Surveys in the Real World

Flashcards covering key vocabulary related to sample surveys, including sampling and nonsampling errors, different sampling methods, and factors influencing data quality.

Sampling Errors

Errors caused by the act of taking a sample, which can lead to biased estimations of the population parameter.

Random Sampling Error

The difference between a statistic and a parameter caused by chance in selecting a random sample, which is accounted for by margins of error and confidence intervals.

Biased Sampling Methods

Methods like convenience samples or voluntary response samples that inherently introduce bias and are not random.

Sampling Frame

A list of individuals from which a sample is drawn.

Frame Errors

Errors that occur when the sampling frame is not representative of the population.

Undercoverage

A type of frame error where the sampling frame is incomplete, meaning some groups of individuals are left out.

Erroneous Inclusions

A type of frame error where the sampling frame includes individuals who are not part of the target population.

Multiple Inclusions

A type of frame error where some individuals appear more than once in the sampling frame, giving them a higher chance of being selected.

Nonsampling Errors

Errors not caused by the act of taking a sample, which can occur even in a census and include processing errors, response errors, and nonresponse.

Processing Errors

Researcher's mistakes in mechanical tasks such as arithmetic operations or data entry errors.

Response Errors

Errors that occur when an individual provides incorrect, untrue, or misleading data, including social desirability bias, careless responding, and response distortion.

Social Desirability Bias

A type of response error where individuals provide answers that they believe will be viewed favorably by others.

Nonresponse

The failure to obtain data from an individual selected into a sample, which can be due to refusal to participate or unanswered questions.

Careless Responding

A type of response error characterized by inattentiveness, arbitrary response patterns (random responding, straightlining), or unwillingness to comply with testing demands.

Response Distortion

A form of dishonest responding, which includes faking good (e.g., social desirability responding) or faking bad (e.g., malingering).

Wording of Questions

A type of nonsampling error where the phrasing of survey questions can significantly influence individuals' answers, often due to ambiguity or a 'loaded' nature.

Loaded Questions

Survey questions that contain a desirable end goal, subtly influencing respondents toward a particular answer.

Weighting Responses

A method to address nonresponse errors by giving more weight to individuals from groups that tend to have lower response rates, making their data influence the sample statistic more.

Stratified Random Sampling

A sampling method where the sampling frame is divided into distinct groups (strata), and then a separate simple random sample is taken from each stratum.

Strata

Distinct groups of individuals within a sampling frame, chosen because of interest in those groups or similarities among individuals within them during stratified random sampling.

Probability Sample

Any sample chosen by chance, where the possible samples and the chance for each individual/sample to be selected are known, allowing for confidence statements.

Nonprobability Samples

Samples, such as voluntary response samples, that do not meet the requirements of probability samples and therefore cannot support confidence statements.