(1) Recap

• Evaluating Descriptive Claims
• Variables vs. Measures
• Levels of Measurement
• Validity

(2) Measurement Error

• Bias
• Random
• Sources and Solutions

Question/Claim

\(\not\xrightarrow{}\) Concepts not scientific/irrelevant

Concept (and dimensions)

\(\not\xrightarrow{}\) Variable does not map onto concept (lack of validity)

Variables

\(\not\xrightarrow{}\) Procedure does not return the true value (measurement error)

Measure(s)

\(\not\xrightarrow{}\)

“Answer”

Variables for “democracy”: What are their levels of measurement?

1: In each year, a country receives a “Polity Score” describing the level of democracy ranging from -10 to +10, with -10 to -6 corresponding to autocracies, -5 to 5 corresponding to anocracies, and 6 to 10 to democracies.

2. The year in which a country first became a democracy

3. Fraction of adults who can are allowed to vote under suffrage rules

4

POLL

Variables for “democracy”:

1. Ordinal

2. Interval

3. Ratio

4. Nominal

Question: “Was there was extensive election fraud in the 2020 US Presidential Election?”

An Answer:

see here

Concept: election fraud: manipulations of votes or vote counting by changing the candidate or counting ineligible ballots

Variable: Abnormally high voter turnout in constituencies with fraud allegations (would result from higher rates of filling out absentee ballots for people who hadn’t voted, dead people voting, ineligible people voting, or even payments to legally registered people for their votes)

Measure: Statistically significant difference in turnout (Number of votes cast divided by registered voters according to official election results in US 2020 Presidential election) between counties with fraud accusations vs. others. see here

In small groups:

1. What problems, if any, are there with the variable?
2. What problems, if any, are there with the measure?

Eggers et al