Political Science Research Foundations and Methods Midterm Study Guide

Comprehensive vocabulary study cards covering units of analysis, measurement, validity, reliability, causal inference, experimental design, sampling techniques, controlled comparisons, levels of measurement, and descriptive statistics based on the midterm study guide.

Aggregate-level unit of analysis

A unit of analysis where each case being studied is a group or larger unit, such as a state, country, or election.

EX:

Your conclusions are about groups, not individual people. For example, studying voter turnout across U.S. states means each row represents a state.

Codebook

A guide that explains the variables in a dataset, what they measure, and how their values are coded.

EX:

It helps you understand what the data actually mean. Without a codebook, you might misinterpret a variable or its values.

Concept

An idea or phenomenon a researcher wants to study, such as political trust, partisanship, or political knowledge.

EX:

Research begins with concepts, but you need to define and measure them before you can test a hypothesis.

Conceptual definition

An explanation of what a concept means and which characteristics it includes.

Data

The information researchers collect or use to study a question, which can include survey responses, numbers, dates, or observations.

Dataset

An organized collection of data, usually arranged in rows and columns.

Ecological fallacy

Making assumptions about individuals based on information about the groups they belong to.

Individual-level unit of analysis

A unit of analysis where each case being studied is an individual person.

Unit of analysis

The type of case being studied, such as an individual, state, country, or election.

Unit of measurement

The specific unit or scale used to measure a variable, such as years, dollars, percentages, or a $0\text{--}100$ feeling thermometer.

Value labels

Descriptions assigned to numerical codes to explain what those codes represent.

Variable

A measurable characteristic that can take different values across cases.

Cross-sectional dataset

A dataset that contains information about multiple cases at one point in time.

Cross-sectional study

A study that examines multiple cases at one point in time to identify patterns or relationships between variables.

Panel study

A study that collects information from the same cases at multiple points in time.

Time-series dataset

A dataset that tracks one unit over multiple points in time.

Research design

The overall plan for how a researcher will investigate a research question, including what data to collect, which variables to measure, and how to analyze them.

Posttreatment measurement

Measuring an outcome after a participant has received a treatment or been exposed to an experimental condition.

Pretreatment measurement

Measuring an outcome before a participant receives a treatment or is exposed to an experimental condition.

Construct validity

The extent to which a measure behaves as expected based on theories about the concept being measured.

Face validity

The extent to which a measure behaves as expected based on theories about the concept being measured.

Validity

The extent to which a measure accurately measures the concept it is intended to measure.

Reliability

The extent to which a measure produces consistent results when the same thing is measured under similar conditions.

Operational definition

A specific explanation of how a concept will be measured, including the instrument or procedure used.

Random measurement error

Unpredictable errors that cause measurements to vary by chance.

Systematic measurement error

An error that consistently distorts measurements in a particular direction.

Hawthorne effect

When people change their behavior because they know they are being observed or studied.

Feeling thermometer

A measure that asks people to rate how warmly or favorably they feel toward a person, group, or political institution, usually on a $0\text{--}100$ scale.

Likert scale

A scale that measures attitudes by asking respondents how strongly they agree or disagree with a statement.

Test-retest method

A method of testing reliability by giving the same measure to the same participants at two different times.

Split-half method

A method of testing reliability by dividing a measure into two parts and comparing the results from each part.

Deductive reasoning

Starting with a theory and using it to develop a hypothesis that can be tested with data.

Inductive reasoning

Reasoning that starts with specific observations and uses them to develop broader patterns or theories.

Theory

A set of connected statements that explains why something happens and may help predict future outcomes.

Model

A simplified representation of a theory or a relationship between variables.

Hypothesis

A specific, testable prediction about the relationship between variables.

Research hypothesis

A hypothesis stating the expected relationship between variables if the researcher's explanation is supported.

Null hypothesis

A hypothesis that states there is no relationship or difference between the variables being studied.

Independent variable

The variable used to explain or predict changes in another variable; in a causal hypothesis, it is the presumed cause.

Dependent variable

The variable researchers are trying to explain; in a causal hypothesis, it is the presumed effect or outcome.

Intervening variable

A variable that helps explain the process through which an independent variable affects a dependent variable.

Deterministic relationship

A relationship in which a particular value or change in $X$ leads to a specific outcome in $Y$ with certainty.

Probabilistic relationship

A relationship in which $X$ changes the likelihood that $Y$ will occur, but does not guarantee the outcome.

Direct relationship

A relationship in which higher values of $X$ are associated with higher values of $Y$, or lower values of $X$ are associated with lower values of $Y$.

Inverse relationship

A relationship in which higher values of $X$ are associated with lower values of $Y$.

Positive relationship

A relationship in which higher values of one variable tend to be associated with higher values of the other.

Negative relationship

A relationship in which higher values of one variable tend to be associated with lower values of the other.

Linear relationship

A relationship in which changes in $X$ are associated with a relatively constant change in $Y$.

Causal diagram

A visual representation of the proposed causal relationships among variables, usually using arrows.

Causal mechanism

The process that explains how or why one variable causes a change in another variable.

Confounder

A third variable, often called $Z$, that is related to both the independent variable ($X$) and dependent variable ($Y$), offering an alternative explanation for their relationship.

Fundamental problem of causal inference

The dilemma that researchers cannot directly observe both possible outcomes for the same case at the same time: what happens when it receives a treatment and what would have happened to that exact case without the treatment.

Spurious relationship

An apparent relationship between $X$ and $Y$ that disappears or substantially weakens after controlling for a third variable, $Z$.

Experimental design

A research design in which researchers introduce a treatment or condition and examine its effect on an outcome.

Field experiment

An experiment conducted in a real-world setting rather than an artificial laboratory setting.

Laboratory experiment

An experiment conducted in a controlled setting where researchers can carefully manage the conditions.

Treatment group

The group of participants who receive the treatment or condition that the researcher wants to test.

Control group

The group that does not receive the treatment being tested, or receives a baseline or placebo condition for comparison.

Random assignment

A process in which participants are randomly placed into the treatment group or control group.

Internal validity

The extent to which a study supports the conclusion that the treatment caused the observed outcome.

Cluster sample

A sampling method where the population is divided into groups, called clusters, and researchers randomly select some of those groups to study.

Convenience sample

A sample made up of people who are easiest for the researcher to reach.

Purposive sample

A sample in which researchers deliberately choose people because they have specific characteristics or experiences relevant to the research question.

Snowball sample

A sampling method where current participants help researchers recruit additional participants by referring people they know.

Random sample

A sample selected through a random process, giving members of the population a chance of being selected.

Random selection

The process of using chance to choose cases from a population or sampling frame.

Random number generator

A tool that produces random numbers to help researchers select cases by chance.

Simple random sample

A probability sample in which every possible sample of a particular size has an equal chance of being selected, giving each population member an equal chance of selection.

Stratified random sample

A sampling method where researchers divide the population into subgroups, called strata, and randomly select participants from each subgroup.

Systematic sample

A sampling method where researchers select cases at regular intervals from a list after choosing a starting point randomly.

Sampling frame

The list or source that identifies the cases from which a sample is selected.

Sampling weights

Numerical adjustments applied to observations to account for differences in how people were selected or represented in a sample.

Selection bias

A systematic problem that occurs when the process of selecting cases creates a sample that differs from the population in ways relevant to the study.

Response bias

A systematic distortion in the answers people provide, causing their responses to differ from their actual beliefs, behaviors, or experiences.

Controlled comparison

A comparison that examines the relationship between an independent variable ($X$) and a dependent variable ($Y$) while holding a third variable ($Z$) constant.

Controlled comparison design

A research design that examines the relationship between $X$ and $Y$ across different categories or values of a third variable, $Z$.

Controlled comparison table

A table that displays the relationship between $X$ and $Y$ separately for each category of $Z$.

Controlled effect

The relationship between $X$ and $Y$ within one specific category or value of $Z$.

Partial effect

A summary of the relationship between $X$ and $Y$ after accounting for $Z$.

Partial relationship

The relationship between $X$ and $Y$ after taking one or more additional variables into account.

Additive relationship

A relationship in which $X$ continues to be related to $Y$ after controlling for $Z$, and the relationship between $X$ and $Y$ is roughly similar across the different values of $Z$.

Interaction relationship

A relationship in which the relationship between $X$ and $Y$ changes depending on the value of $Z$.

Mean comparison table

A table that compares the average value of a continuous dependent variable across categories of an independent variable, often separately for each category of a control variable.

Nominal-level variable

A variable that divides observations into categories that have no meaningful order or ranking.

Ordinal-level variable

A variable with categories that can be ranked in a meaningful order, but the distance between categories is not necessarily equal.

Interval-level variable

A numerical variable with ordered values and equal distances between them.

Discrete values

Values that are separate and countable, usually whole numbers.

Continuous values

Values that can take any numerical value within a range, including decimals.

Per-capita statistics

A statistic calculated by dividing a total amount by the population size to find the amount per person ($\text{Per capita} = \frac{\text{Total amount}}{\text{Population}}$).

Summary statistics

Numerical values used to summarize the main characteristics of a dataset.

Central tendency

A measure that describes the center or typical value of a distribution.

Mean

The average value, calculated by adding all observations and dividing by the number of observations.

Median

The middle value when observations are arranged from lowest to highest.

Mode

The value or category that appears most frequently in a dataset.

Resistant measure of central tendency

A measure of central tendency that is not strongly affected by extreme values or outliers.

Dispersion

The amount of variation or spread in a dataset.

Range

The difference between the highest and lowest values in a dataset.

Interquartile range

The difference between the third quartile ($Q_3$) and first quartile ($Q_1$), measuring the spread of the middle $50\%$ of observations ($Q_3 - Q_1$).

Variance

A measure of dispersion that describes how far observations are spread from the mean using squared differences.

Standard deviation

A measure of how much observations typically vary from the mean, calculated as the square root of the variance ($\sqrt{\text{Variance}}$).