Sociology Research Methods: Sampling, Surveys, and Experimental Design

Comprehensive vocabulary flashcards covering fundamental concepts in sociological research methods: sampling terminology, probability vs. non-probability techniques, survey administration, and experimental designs.

Element (Unit of Analysis / Case)

The unit from which information is collected and which provides the basis for analysis (e.g., people, organizations, athletic teams, families).

Universe

The theoretical and hypothetical aggregation of all elements (collection of people) for a given survey.

Population

A theoretically specified aggregation of survey elements that places specific limitations or definitions on the universe (e.g., Ohioans who have lived in the state one year and are 18 years of age or older).

Survey Population

The aggregation or list of elements from which the survey sample is actually selected.

Sampling Unit

An element or set of elements considered for selection in some stage of sampling.

Primary Sampling Unit (PSU)

The unit selected during the first stage of a multi-stage sampling design (e.g., a nursing home or household).

Secondary Sampling Unit

The unit selected in the second stage of sampling (e.g., head of household or city) that can be utilized more than once in the process.

Final Sampling Unit (FSU)

The actual individual person or element selected from the list during the final stage of sampling.

Sampling Frame

The actual list of sampling units from which the sample or some stage of the sample is drawn.

Observation Unit

The element or aggregation of elements from which information is directly collected.

Variable

A characteristic of interest being studied that must vary across cases rather than remaining constant.

Parameter

A summary description of a given variable across the entire population (e.g., average age from the U.S. Census).

Statistic

A summary description of information calculated from a survey sample (e.g., mean age of sample respondents).

Sampling Error

The true difference between the population parameter and the sample summary statistic, which is difficult to calculate directly because population-wide data is rarely available.

Standard Error (SE)

A statistical estimate of the deviation of a summarizing sample measure from the true population parameter, calculated as $SE = \frac{SD}{\sqrt{n}}$.

Standard Deviation (SD)

A measure of the amount of variation or dispersion of a set of values, where a low value indicates scores cluster near the mean and a high value indicates scores spread across a wider range.

Probability Sampling

A sampling method where each case in the population has a known, non-zero probability of being selected into the sample.

Simple Random Sampling (SRS)

A probability sampling design in which each element in the population has an equal and independent chance of being selected.

Systematic Sampling

A sampling method where units are chosen from a list at regular selection intervals ($I = \frac{N}{n}$) following a randomly determined start.

Stratified Random Sampling

A sampling method in which the population is divided into non-overlapping groups (strata) based on specific variables, followed by simple random sampling within each group.

Proportionate Sampling

A type of stratified sampling where the proportion of each stratum in the sample matches its exact proportion in the population.

Disproportionate Sampling

A type of stratified sampling used when a target subgroup is small, drawing a disproportionately larger sample from that subgroup to ensure adequate representation.

Cluster Sampling

A sampling method in which elements are selected from the population in groups or clusters (e.g., city blocks, dorms, classes) rather than individually.

Multi-Stage Cluster Sampling

The procedure of selecting a sample through two or more successive, contingent sampling stages.

Non-Probability Sampling

Sampling techniques in which cases are selected by means other than chance, leaving the selection probability of each unit unknown.

Accidental Sampling (Convenience Sampling)

A non-probability sampling technique where available individuals who agree to participate are enrolled until the targeted sample quota is reached.

Quota Sampling

A non-probability method that sets target sample quotas based on manifest demographic traits of the population and fills them via accidental sampling.

Purposive Sampling

A non-probability sampling method where respondents are hand-picked from a specific subgroup based on researcher judgment to test a hypothesis.

Expert Sampling (Judgement Sampling)

A subtype of purposive sampling in which cases are selected on the basis of informed professional opinion regarding who is representative of the target group.

Snowball Sampling

A non-probability sampling method where existing study participants refer or provide contact information for other individuals possessing the target characteristic.

Survey Research

A method of gathering data about a human population through direct contact with respondents using systematic instruments like questionnaires or interviews.

Sample Survey

A survey design where information is gathered from a representative fraction of the target population rather than an entire census.

Survey Response Rate

The percentage calculated by dividing the number of completed surveys by the total number of individuals sent the survey, multiplied by $100$.

Survey Completion Rate

The percentage calculated by dividing the number of fully completed surveys by the number of surveys initiated by respondents.

Random Digit Dialing (RDD)

A telephone survey technique that dials randomly generated phone numbers, allowing unlisted telephone numbers to be included in the sampling frame.

Experiment

A research design featuring controlled manipulation of independent variables by the investigator, paired with precise observation and measurement to evaluate causality.

Stimulus

The manipulated independent variable ($X$) administered to participants in the experimental group.

Control Group

The group in an experiment that does not receive the manipulated stimulus (often receiving a placebo) and serves as an external baseline for comparison.

Double-Blind Experiment

An experimental protocol where neither the participants nor the administering researchers know who is assigned to the experimental group versus the control group.

Randomization (Random Assignment)

The process of assigning experimental participants to either the control or experimental condition purely by chance, making both groups statistically comparable.

Classic Experimental Design

An experimental structure that utilizes pre-test and post-test observations across both an experimental group and a control group.

Solomon 4-Group Design

An advanced experimental design using four groups (two pre-tested and two un-pretested) across two time points to isolate the pre-test interaction effect on the post-test.

Internal Validity

The degree to which the experimental procedures ensure that observed changes in the dependent variable are strictly caused by the independent variable without procedural distortion.

External Validity

The extent to which the experimental findings can be generalized beyond the laboratory setting to broader, non-test populations.

Hawthorne Effect

An experimental threat to external validity where subjects modify or improve their behavior simply because they know they are being observed.

Experimental Mortality

A threat to internal validity in longitudinal research caused by participants dropping out, moving away, or dying prior to follow-up testing.

Statistical Regression

A threat to internal validity where individuals selected on the basis of extreme initial scores naturally regress back toward the average score upon subsequent testing.