PS 193: Quantitative Methods in the Social Sciences - Modules 1 and 2

Comprehensive vocabulary flashcards covering foundational quantitative methods concepts, research paradigms, causal hurdles, types of data sets, and statistical biases from Modules 1 and 2.

Survey Research

Quantitative social research in which a researcher systematically asks many people the same questions, then records and analyzes their answers.

Measurement Validity

A property of an empirical measure indicating how well it captures the underlying theoretical concept it is intended to evaluate, including face, content, and construct validity.

Internal and External Validity (Research Design)

Criteria evaluated in a causal study; internal validity refers to whether the observed covariation is genuinely a causal effect of $X$ on $Y$, while external validity refers to whether findings generalize to other contexts, populations, or settings.

Causal Hurdle 1 (Causal Mechanism)

The first step in evaluating a causal claim, assessing whether there is a credible and plausible causal mechanism connecting the independent variable ($X$) to the dependent variable ($Y$).

Causal Hurdle 2 (Reverse Causality)

The step in evaluating a causal relationship that tests whether researchers can eliminate the possibility that the outcome variable ($Y$) causes the explanatory variable ($X$).

Causal Hurdle 3 (Covariation)

The step in evaluating a causal claim that determines whether empirical covariation or statistical association exists between the independent variable ($X$) and dependent variable ($Y$).

Causal Hurdle 4 (Confounding)

The step in evaluating a causal claim that asks whether all confounding variables ($Z$) have been controlled for to ensure the observed association between $X$ and $Y$ is not spurious.

Confounding Variable ($Z$)

An extraneous variable that is causally related to both the independent variable ($X$) and dependent variable ($Y$), which can produce a spurious empirical correlation if not controlled.

Survivorship Bias

A logical error and form of selection bias where conclusions are drawn solely from entities that passed an observational filter (such as returning aircraft surviving bullet damage) while ignoring those that failed to survive.

Simpson's Paradox

A statistical phenomenon in which a trend or relationship appears across several individual sub-groups but disappears or reverses when the groups are aggregated together.

Spurious Correlation

A mathematical relationship in which two variables show high statistical correlation (such as autism diagnosis rates and organic food sales) without having a direct causal relationship, typically caused by coincidence or common lurking variables.

Cross-Sectional Data

A dataset consisting of multiple individual observational units (such as individuals, firms, or regions) collected at a single fixed slice in time.

Time-Series Data

A dataset consisting of observations on one unit or entity tracked across multiple regular intervals of time.

Pooled Cross-Sectional Data

A dataset formed by combining cross-sectional samples gathered across different time periods, where different random individual units are observed in each time period.

Panel Data (Longitudinal Data)

A dataset following the exact same cross-sectional units (such as specific cities, countries, or individuals) across multiple points in time.

Correlation Coefficient ($r$)

A standardized statistical metric ranging from $-1.0$ to $+1.0$ that measures the strength and direction of a linear relationship between two quantitative variables.

Modernization Theory

A comparative political theory asserting that rising economic development and per capita income lead to greater democratization and democratic stability in nations.

Spatial Voting Model (Proximity Voting)

A voting behavior model positing that an elector's probability of voting for a candidate decreases as the ideological distance between the candidate's position and the voter's position increases.

Four Critical Inquiries for Reading a Poll

The four methodological criteria researchers use to critically evaluate public opinion surveys: (1) Who was asked, and how selected?, (2) What was the exact wording?, (3) Sample size and margin of error?, and (4) Who paid, and could that shape it?

Misleading Cumulative Chart

A graph plotting running totals that continuously trends upwards and masks periods of stagnation or decline occurring in discrete periodic measurements.

Research Paradigm Framework

The layered philosophical structure of scientific inquiry consisting of ontology (nature of reality), epistemology (nature of knowledge), methodology, axiology, and practical methods.