HLSA 721 Midterm Review Spring 2025

Vocabulary flashcards based on key terms and concepts from the HLSA 721 Midterm Review lecture notes.

The Federal Statistical System

A decentralized and loosely organized set of federal offices and agencies responsible for producing, disseminating, and analyzing data.

The Office of Management and Budget

The federal office that loosely oversees the federal statistical system.

Administrative Data

Data generated from program administration or market transitions, not collected for research purposes but often used in research.

It’s designated for specific purposes, and it’s not always timely or available.

Public Opinion Surveys (Polls)

Surveys that collect information on policy and candidate preferences.

Population Surveys

Surveys that collect information on characteristics and behaviors from well-defined populations.

Economic Data

Data that refers to employees and/or firms, collected by the US Census Bureau.

Demographic Data

Data that refers to people and households, collected by the US Census Bureau.

Population Surveillance

Data collection activities meant to monitor trends and characteristics of a population.

Forecasting and Predicting

Policy evaluation activities that allow decision-makers to predict the outcome of a proposed policy.

Evaluation

Policy evaluation activities meant to measure the impact of an enacted policy.

Operational Uses of Federal Surveys

Usage of federal survey data in public policy implementation and resource allocation.

The Decennial Census

A constitutionally mandated survey conducted every 10 years to enumerate the entire population.

National Health Interview Survey

The official source of information on the health of the US population, a repeated cross-sectional survey.

Current Population Survey

A federal labor market survey that is the official source of employment and poverty information.

American Community Survey

A survey of housing, social, economic, and demographic characteristics conducted by the Census Bureau.

IPUMS (Integrated Public Use Microdata Series)

A data access system that facilitates access to US and international surveys by governmental entities.

Independent Variable

A variable that is thought to cause an outcome of interest.

Dependent Variable

An outcome that is affected by the independent variable.

Mediator

The variable that links an independent variable and an outcome, the mechanism of action.

Moderator

A variable that changes the causal effect of the independent variable on the dependent variable.

Confounder

A variable that causes both the independent variable and the dependent variable, creating a spurious correlation.

Target Population

The population of interest for the survey.

Survey Error

The divergence between a survey statistic and the true population parameter.

Survey Methodology

The study area concerned with reducing survey error in the presence of financial and normative constraints.

Variable Error

Errors that differ over repeated administration of the same survey.

Bias

Errors that are consistent over repeated administrations of the same survey.

Measurement Errors

Survey errors that occur due to the process of measuring the variable of interest.

Errors of Non-Observation

Survey errors caused when the survey is administered to a subset of the full population of interest.

Total Survey Error

The full amount of survey error in a survey statistic, including measurement errors and errors of non-observation.

Interviewer Error

A measurement error arising from the use of interviewers during data collection.

Response Formation

The cognitive process a survey respondent goes through to answer a survey question.

Social Desirability Bias

A measurement error that occurs when respondents provide answers that are not consistent with their own judgment to maintain social reputation.

Satisficing

A measurement error occurring when respondents provide a 'good enough' answer rather than a high-quality one.

Recall Error

Measurement errors that occur due to respondents' imperfect memory.

Telescoping

Measurement error that occurs when respondents misplace an event in time relative to the reference period of a survey question.

Record-check Studies

Studies linking survey responses to a gold standard to compare survey results with a truth source.

Cognitive Interviews

Studies wherein respondents recount their survey-taking experience to gauge measurement error.

Split Ballot Test

A study where different groups receive different surveys to compare response variation.

Instrument Errors

Survey errors arising from the interaction between respondents and the survey instrument.

Question Stem

The actual survey question in a survey instrument.

Response Options

Pre-specified answers from which respondents must choose.

Write-in Responses

A generic term for response options allowing respondents to provide any answer they choose.

Skips

Conditional subsequent questions based on answers to initial survey questions.

Mode of Interview

Different forms in which surveys can be conducted, such as in-person, telephone, or written formats.

Sampling

Methods used to select a subset of the population for data collection.

Non-Probability Samples

Sample designs based primarily on convenience to the researcher.

Probability Samples

Sampling designs in which every unit of the population has a known non-zero probability of selection.

Sampling Frame

The operationalization of the population, often as an actual list of units in the population.

Simple Random Sample

A sampling design where every population element has the same probability of selection.

Sampling Error

Survey errors resulting from using a probability sample design.

Hypothetical Sampling Distribution

The distribution of results from repeatedly drawn probability samples from the same population.

Standard Error

A measure of sampling error, equivalent to the standard deviation of the hypothetical sampling distribution.

Complex Sample Designs

Probability samples where elements have known non-zero probabilities of selection but may not be independent or equally probable.

Stratified Sampling

A sampling design that divides the population into groups and draws samples from each group.

Oversampling

A sampling strategy that increases the selection probability of certain groups to ensure sufficient sample size.

Cluster Sampling

A sampling design where the population is divided into groups, and only some groups are sampled.

Multi-Stage Sampling

A sampling design involving multiple phases, like drawing cluster samples and then individual units from those clusters.

Weights

Survey weights that represent how many units in the population each sampled unit represents.

Post-Stratification Weighting

Methods used to adjust sample weights to match known population values.

Raking

Adjustment methods to sample weights ensuring weighted estimates match external population values.

Coverage Error

Survey error caused by a sampling frame that does not perfectly match the population.

Non-Response Error

Survey error arising from not obtaining information from sampled elements.

Unit Non-Response

Non-response occurring when no information is collected from a designated sampled unit.

Item Non-Response

Non-response occurring when not all requested information is obtained from respondents.

Response Rate

A measure of the fraction of sampled individuals that participated in the survey.

Response or Non-Response Bias

The amount of non-response error in a survey statistic, dependent on non-response error and response rate.

Imputation

Methods used to correct for item non-response by filling in missing information.

List-wise Deletion

Exclusion of cases with one or more missing items from analysis.