Untitled Quiz

A measurable characteristic of something (person, group, event) that can change.

A way to determine how much of a concept applies to a specific example based on observations.

The different parts or aspects of a larger concept that stand alone.

Categorizing objects or observations into groups, without any ranking.

Categorizing objects and ranking them – but the distances between categories aren't meaningful.

Numerical measurements with meaningful intervals between values, but no true zero point.

Abstract descriptions of characteristics, phenomena, or groups

Nominal, Ordinal, Interval, and Ratio

A numerical value where zero means absence.

The variable believed to cause a change in another variable.

The variable that is expected to change due to the independent variable.

How well a measurement truly captures the concept it aims to measure.

Consistency of a measurement. Same result when repeated.

Inaccurate measurement, stemming from either validity or reliability issues.

Systematic error in measurement, always too high or too low.

Measurement consistently overestimates the true value.

Measurement consistently underestimates the true value.

Inconsistent measurement due to random factors.

The entire group being studied.

A smaller part of the population studied.

The process of selecting a sample from a population.

Each member of the population has an equal chance of selection.

Error in results from the sample not representing the population exactly.

A method to control for confounding variables by measuring and adjusting them. Similar cases: unchanging variables that differ between cases and changing variables that differ between cases. Same case over time: confounding variables that change over time. Difference in difference: confounding variables that change over time but differ across cases.

Controlling for confounding variables by comparing cases with similar characteristics. Assumes no constant/time-invariant differences affect the outcome and no changing/time-variant differences affect the outcome.

Controlling for confounding variables by comparing the same case over time. Assumes no confounding variables change over time for the case and no confounding variables change over time for other cases.

Controlling for confounding variables by comparing the difference in change between cases with and without the cause. Assumes no confounding variables change over time differently across cases and trends are parallel across cases without the cause.

A situation where the cause is randomly assigned, allowing researchers to isolate the effect of the cause. Requires minimal assumptions about confounding variables.

The degree to which a study can confidently claim that the cause led to the effect, controlling for confounding variables.

The degree to which the findings of a study can be generalized to other populations, settings, and times.

Different methods to control for confounding variables have a tradeoff between internal and external validity. Adjustment: high external validity, but at the cost of internal validity. Similar cases: higher internal validity, but lower external validity. Same case over time: high internal validity, but low external validity. Difference in difference: high internal validity, but low external validity. Natural experiment: high internal validity, but low external validity.

A statement about what would have happened if something had been different. It's used to test causal claims.

A method that compares two cases that are similar in all relevant respects except for one factor, to see if that factor causes the observed difference.

A statistical relationship between two variables, meaning they change together. It doesn't necessarily mean one causes the other.

When two variables move in the same direction. As one increases, the other also increases.

When two variables move in opposite directions. As one increases, the other decreases.

A correlation between two variables that appears to be causal, but is actually caused by a third, hidden variable.

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

A variable that acts as a mediator between two other variables. It explains how one variable causes another.

A variable that affects one variable, which then affects another variable. It can lead to spurious correlation.

We can never directly observe what would have happened to a case if it had been exposed (or not exposed) to a particular cause. We can only observe its actual outcome.

The hypothetical outcomes that a case would have had if it had been exposed (or not exposed) to a particular cause.

Because we are comparing hypothetical outcomes for the same individuals. We are asking what would have happened to individual A if they had been exposed to the cause, compared to what actually happened to individual A.

They're connected. A study with poor internal validity may have trouble achieving good external validity because we may be generalizing from an inaccurate effect.

The difference between a case's potential outcome if exposed to the cause (Y₁ᵢ) and its potential outcome if not exposed (Y₀ᵢ).