High School Economics • Foundations of Economic Thinking

Evaluating Economic Evidence — Evaluate the strength of evidence for an economic argument (intro)

Learn to separate strong economic claims from weak ones by critically analyzing the evidence behind them.

High School Economics emblem with a market chart, dollar symbol, and balance scale
Inside this lesson 10 sections 5 questions 7 figures

SECTION 1

Historical Context & Motivation

Every day you encounter economic claims—on social media, in the news, or even in political debates. Someone might say, "Raising the minimum wage always destroys jobs," or "Tax cuts pay for themselves." But how do you know whether these claims are backed by solid evidence or are just opinions dressed up as facts? The skill of evaluating economic evidence has been developing for centuries, evolving alongside the discipline of economics itself.

In the early days of economic thought, arguments were largely philosophical. Thinkers like Adam Smith and David Ricardo relied on logic, observation, and narrative examples to make their cases. It was not until the twentieth century that economists began to insist on empirical evidence—data collected from the real world—as a standard for evaluating claims. This shift transformed economics from a purely theoretical field into one that uses statistics, experiments, and careful measurement to test ideas.

1776 • Adam Smith's Wealth of Nations

Smith used logical reasoning and real-world observations about factories and trade to argue for free markets, but relied more on anecdote than systematic data.

1930s • The Keynesian Revolution

John Maynard Keynes used economic data from the Great Depression to argue that government spending could boost a struggling economy, pushing economists toward evidence-based policy debates.

1960s–1970s • Rise of Econometrics

Economists developed advanced statistical tools called econometrics to test hypotheses with large data sets, making evidence evaluation more rigorous and systematic.

2000s–Present • Natural Experiments & Randomized Trials

Researchers like Esther Duflo and Joshua Angrist won the Nobel Prize for using experimental and quasi-experimental methods to generate stronger causal evidence in economics.

This history reveals a key question that still matters today: How can we tell the difference between a well-supported economic argument and one that only sounds convincing? Answering that question is the central goal of this lesson.

SECTION 2

Core Principles of Evidence Evaluation

Before you can judge whether economic evidence is strong or weak, you need a framework—a set of guiding ideas that help you ask the right questions. Think of these principles as a toolkit you carry with you whenever you read an article, hear a speech, or discuss policy. Each principle highlights a different dimension of evidence quality.

1

Source Credibility

Who is making the claim? Evidence from peer-reviewed research, government statistical agencies, or established universities tends to be more reliable than evidence from anonymous blogs or organizations with a financial stake in the outcome.
2

Relevance of Data

Does the evidence actually address the specific claim being made? Data about U.S. wages may not be relevant to an argument about wages in Brazil. The evidence must match the argument's scope and context.
3

Sample Size & Representativeness

How much data was collected, and does it represent the broader population? A survey of ten people in one city cannot reliably represent a national trend. Larger, more diverse samples produce stronger evidence.
4

Correlation vs. Causation

Just because two things happen together does not mean one caused the other. Ice cream sales and drowning rates both rise in summer, but ice cream doesn't cause drowning—hot weather drives both. Strong evidence carefully distinguishes correlation from causation.
5

Bias & Objectivity

Is the evidence presented fairly, or has someone cherry-picked data that supports only one side? Look for whether the argument acknowledges limitations or counterevidence. Transparent, balanced analysis signals stronger evidence.
✦

KEY TAKEAWAY

Think of evaluating economic evidence like being a judge at a science fair. You wouldn't award first place just because a project has a flashy poster. You'd check the hypothesis, the data, the experiment's design, and whether the conclusions actually follow from the results. In the same way, evaluating economic evidence means looking past persuasive language to examine the quality of the data, the logic of the argument, and the credibility of the source.

SECTION 3

The Evidence Evaluation Framework — Visual Guide

The diagram below illustrates a step-by-step process for evaluating any economic claim. Start at the top with the claim itself, then move through four evaluation filters. Each filter asks a critical question, and the answers help you decide whether the overall evidence is strong, moderate, or weak.

ECONOMIC CLAIMFILTER 1Is the source credible and unbiased?✓FILTER 2Is the data relevant and representative?✓FILTER 3Does the evidence show causation, not just correlation?✓FILTER 4Are the conclusions logical and free from cherry-picking?✓STRONG EVIDENCE
Start with an economic claim at the top. Each filter asks a critical question about the evidence. If the evidence passes all four filters, it is considered strong. Failing any filter weakens the overall argument.

Notice that the framework is sequential: you work through each filter in order. An argument that fails at Filter 1—for example, an anonymous source with a clear financial interest—should already raise red flags, even if the data itself looks impressive. Conversely, evidence from a credible source that still confuses correlation with causation (Filter 3) has a significant weakness. The more filters the evidence passes, the more confidence you can place in the argument.

SECTION 4

How Evidence Evaluation Works in Practice

While evaluating economic evidence is not primarily a math exercise, it helps to understand a few basic concepts that economists use to measure the strength of evidence. Even at an introductory level, knowing what these terms mean will sharpen your ability to ask the right questions.

Understanding Sample Size

When economists collect data to support a claim, the sample size (often represented as n) refers to the number of observations or data points they used. A larger sample generally produces more reliable results. If someone claims that a new trade policy boosted exports based on data from only 5 companies, that's a very small sample. If the study examined 5,000 companies, the evidence is far stronger.

CONFIDENCE PRINCIPLE

Largern→MoreconfidenceinconclusionsLarger n → More confidence in conclusions

Where n = the number of observations in the data set. As n increases, random errors tend to cancel out, producing results closer to the true value.

The Correlation vs. Causation Test

One of the most common errors in economic arguments is treating correlation (two things moving together) as if it were causation (one thing directly causing the other). To test for causation, economists look for a plausible mechanism—a logical explanation for why A would cause B—and they try to control for confounding variables, which are outside factors that could explain both A and B. Strong evidence accounts for confounders; weak evidence ignores them.

CAUSATION TEST

Causation=Correlation+Mechanism+ControlledConfoundersCausation = Correlation + Mechanism + Controlled Confounders

A valid causal claim requires not just that two variables move together, but that there is a logical reason one causes the other, and that alternative explanations have been ruled out.

Checking for Cherry-Picking

Cherry-picking occurs when someone selects only the data points that support their argument while ignoring data that contradicts it. For example, if a company reports its stock price on only the three best days of the year, it paints a misleading picture. Evaluating evidence means asking: Is all the relevant data being presented, or just the convenient data? Strong arguments acknowledge contradictory evidence and explain why their conclusion still holds.

SECTION 5

Types of Economic Evidence — A Classification

Not all evidence is created equal. Economists rely on several types of evidence, and each type carries different strengths and weaknesses. Understanding this hierarchy helps you quickly assess how much weight to give to a particular piece of evidence. The diagram below arranges the most common types of economic evidence from strongest to weakest.

EVIDENCE STRENGTH PYRAMIDSTRONGESTWEAKESTRandomized Controlled TrialsResearchers randomly assign groups; isolates causationNatural ExperimentsReal-world events create quasi-random conditionsStatistical Studies (Large Datasets)Correlational analysis with controls for confoundersCase Studies & Historical ExamplesIn-depth look at specific events; limited generalizabilityAnecdotes & Personal OpinionsIndividual stories; high bias risk, low reliability
The Evidence Strength Pyramid ranks evidence types from most reliable (top) to least reliable (bottom). Randomized controlled trials sit at the peak because they best isolate cause and effect, while anecdotes sit at the base because they are easily distorted by personal bias.
Evidence TypeStrengthExample
Randomized Controlled TrialVery High — directly tests causationA government randomly selects towns to receive a job training program and compares employment outcomes to towns without the program.
Natural ExperimentHigh — exploits real-world randomnessComparing minimum wage effects between two neighboring states where one raised its minimum wage and the other did not (the famous Card and Krueger study).
Statistical StudyModerate — shows patterns but correlation risk remainsAnalyzing GDP growth and education spending across 50 countries over 20 years.
Case StudyLow-Moderate — limited generalizabilityExamining how one company's pricing strategy affected its sales in one market.
Anecdote / OpinionLow — high bias, not systematic"My uncle's restaurant went out of business after the minimum wage went up, so minimum wage increases always hurt businesses."

Comparison of common evidence types used in economic arguments

SECTION 6

Worked Example — Evaluating a Real Economic Claim

Let's apply the evidence evaluation framework to a real-world-style economic claim. Suppose you read the following argument in a news article:

ℹ

THE CLAIM

"A recent study of 2,500 small businesses by the National Bureau of Economic Research found that states that reduced corporate taxes by 2% between 2015 and 2020 saw, on average, 8% higher job growth compared to states that kept taxes unchanged. The researchers controlled for differences in population growth, industry mix, and state spending. Therefore, cutting corporate taxes leads to job growth."

Evaluating the Claim Step by Step

  1. 1

    Step 1 — Check Source Credibility

    The claim cites the National Bureau of Economic Research (NBER), which is a well-respected, nonpartisan research organization in the United States. NBER working papers are reviewed by other economists. This passes the credibility filter.
    ✓ Source is credible
  2. 2

    Step 2 — Assess Data Relevance & Sample Size

    The study examines 2,500 small businesses across multiple states over a five-year period. This is a reasonably large sample, and the data directly addresses the claim (tax cuts → job growth). The evidence is relevant and the sample is large enough to draw meaningful conclusions.
    ✓ Data is relevant, n = 2,500 is a solid sample
  3. 3

    Step 3 — Test Correlation vs. Causation

    The researchers controlled for population growth, industry mix, and state spending—three important confounding variables. However, we should ask: Were there other differences between the states that could explain the result? For instance, states that cut taxes may also have had more business-friendly regulations or stronger local economies. The study addresses some confounders but may not account for all of them, which is a partial weakness.
    ⚠ Partially passes — some confounders controlled, but not all
  4. 4

    Step 4 — Check for Cherry-Picking & Logical Reasoning

    The article presents an average across many businesses and states, which reduces the risk of cherry-picking individual examples. The conclusion—that tax cuts "lead to" job growth—is somewhat strong given that not all confounders were addressed. A more cautious conclusion would say tax cuts are "associated with" job growth. Still, the overall reasoning is sound.
    ✓ Mostly passes — minor overstatement of causality
  5. 5

    Step 5 — Overall Assessment

    This is a moderately strong piece of evidence. The source is credible, the sample is large, and several confounders were controlled. The main weakness is that the study may not account for all alternative explanations, and the conclusion overstates the causal link slightly. A careful evaluator would say this evidence supports but does not prove the claim that corporate tax cuts lead to job growth.
    Overall: Moderately Strong Evidence

SECTION 7

Common Strengths and Pitfalls in Economic Arguments

As you practice evaluating economic evidence, you will start to notice recurring patterns—some arguments consistently display signs of strong evidence, while others fall into predictable traps. The table below summarizes the most common strengths and pitfalls you should watch for.

Signs of Strong EvidenceCommon Pitfalls (Weak Evidence)
Cites peer-reviewed or government-published data sourcesUses unnamed "experts" or vague attributions ("studies show...")
Uses large, representative sample sizesRelies on one or two examples, anecdotes, or small samples
Controls for confounding variablesTreats correlation as causation without addressing confounders
Acknowledges limitations and counterevidenceCherry-picks data; ignores inconvenient facts
Uses cautious language ("suggests," "is associated with")Overstates conclusions ("proves," "always," "never")
Data is current and relevant to the specific contextUses outdated data or applies findings from one context to a completely different one

Comparing hallmarks of strong vs. weak economic evidence

✦

KEY TAKEAWAY

Think of economic evidence like building materials. A house built with tested steel and inspected lumber (peer-reviewed data, controlled studies) will stand strong in a storm. A house built with scrap wood and guesswork (anecdotes, cherry-picked stats) might look fine on a sunny day, but it will collapse under scrutiny. Your job as an evidence evaluator is to inspect the building materials before you trust the structure.

SECTION 8

Connecting to Advanced Economic Analysis

The evidence evaluation skills you are learning in this introductory lesson lay the groundwork for more sophisticated analysis you might encounter in college economics, AP courses, or real-world policy work. As you advance, the questions become more complex, but the core principles remain the same. The table below shows how introductory concepts connect to their advanced counterparts.

Introductory Concept (This Lesson)Advanced Concept (Future Study)
Checking source credibilityEvaluating research methodology, peer-review processes, and institutional incentives in academic publishing
Sample size mattersStatistical significance testing, confidence intervals, and power analysis in econometrics
Correlation vs. causationInstrumental variables, regression discontinuity, and difference-in-differences analysis for isolating causal effects
Identifying cherry-pickingPublication bias, p-hacking, and meta-analysis (combining results from many studies)
Evidence strength pyramidHierarchy of evidence in policy analysis, cost-benefit analysis frameworks, and systematic reviews

How introductory evidence evaluation skills connect to advanced economic methods

The good news is that by mastering the basics now, you are building a foundation that will serve you whether you pursue a business degree, enter the workforce, or simply want to be a more informed citizen. Every time you read an economic claim in the news, you are practicing the same skills that professional economists and policy analysts use every day—just at a different scale.

💡

LOOKING AHEAD

In future lessons, you will learn how to apply these evaluation skills to specific economic topics like supply and demand, fiscal policy, and international trade. You will also explore how economists design studies to produce the strongest possible evidence—and why even the best studies have limitations.

SECTION 9

Practice Problems

Test your understanding of evidence evaluation with the following five problems. They increase in difficulty from basic recall to critical analysis.

Question 1: CONCEPTUAL

What is the difference between correlation and causation? Give a brief economic example of each.
Show answer
Correlation means two variables move together, while causation means one variable directly causes the other to change. For example, there is a correlation between higher education levels and higher incomes—people with more education tend to earn more. However, causation requires showing that education itself (not family background, talent, or other factors) directly causes the income increase. An economist could test this using a randomized scholarship program, where some students are randomly selected to receive free college tuition and then compared to those who did not. If the scholarship recipients earn more, that is stronger evidence of causation.

Question 2: BASIC CALCULATION

A blog post argues that a new trade deal increased U.S. exports by citing data from 8 companies over 6 months. Using the evidence evaluation framework, identify at least two weaknesses in this evidence.
Show answer
First, the sample size is very small (n = 8). Eight companies cannot reliably represent the thousands of businesses affected by a national trade deal. Second, six months is a short time frame—trade effects often take years to materialize, so the data may be capturing short-term fluctuations rather than a genuine trend. Third, a blog post is generally not a peer-reviewed source, which raises concerns about credibility and potential bias. Finally, without knowing whether the analysis controlled for other factors (such as global demand changes or currency fluctuations), we cannot rule out confounding variables.

Question 3: INTERMEDIATE

Consider this claim: "Countries with more cell phones have higher GDPs, so giving people cell phones will increase GDP." Evaluate this argument using the correlation vs. causation framework. What confounding variables might explain this relationship?
Show answer
This argument confuses correlation with causation. While it may be true that cell phone ownership and GDP are correlated, the relationship is likely driven by confounding variables rather than a direct causal link. Wealthier countries have more money to spend on technology, including cell phones, so higher GDP may cause more cell phone ownership—not the other way around. Additional confounders include infrastructure development, education levels, and urbanization, all of which increase both GDP and cell phone access. Simply giving people cell phones would not address these underlying factors. To establish causation, a researcher would need a controlled experiment or natural experiment that isolates the effect of cell phone access from these other variables.

Question 4: APPLIED

You are on the student government and must decide whether to extend school lunch hours based on a claim that "longer lunch periods improve student academic performance." A parent presents the following evidence: their child's grades improved after transferring to a school with longer lunches. A teacher presents a university study of 1,200 students across 30 schools that found a small positive correlation between lunch length and test scores after controlling for socioeconomic status. Which evidence is stronger and why? What additional information would you want before making a decision?
Show answer
The university study is significantly stronger evidence than the parent's anecdote. The parent's story is a single case (n = 1) with no controls—the child's improved grades could be due to better teachers, a different curriculum, or personal maturation. The university study uses a large sample (n = 1,200), covers 30 schools (increasing representativeness), and controls for socioeconomic status, which is a key confounder. However, even the university study has limitations: correlation does not prove causation, and "a small positive correlation" suggests the effect is modest. Before making a decision, you would want to know whether the study was peer-reviewed, whether other confounders (such as school quality or student motivation) were controlled, the actual size of the academic improvement, and whether similar findings have been replicated in other studies.

Question 5: CRITICAL THINKING

A prominent economist publishes a study in a respected journal showing that increasing the minimum wage by $1 in a particular state did not reduce employment. A business lobbying group publishes its own report, using data from the same state and time period, showing that small business employment dropped 3% after the minimum wage increase. Both studies use real data. How can two studies using the same setting reach opposite conclusions? What questions would you ask to determine which study provides stronger evidence?
Show answer
Two studies can reach opposite conclusions even with overlapping data for several reasons. First, they may define the outcome differently: the economist's study may measure total employment (including large businesses), while the lobbying group focuses only on small businesses, which may be more sensitive to wage increases. Second, they may use different time frames, comparison groups, or statistical methods. Third, the lobbying group may have cherry-picked its sample or chosen a metric that makes the effect look worse. To determine which is stronger, you should ask: (1) Was the study peer-reviewed? The journal article likely underwent external review, while the lobbying report may not have. (2) Does the lobbying group have a financial interest in the outcome? Industry groups may have an incentive to find negative effects. (3) What variables did each study control for? (4) Does either study acknowledge limitations or alternative explanations? The study that is more transparent, uses a broader and more representative sample, and controls for more confounders provides stronger evidence—even if its conclusion is less dramatic.

SUMMARY

Lesson Summary

Summary

Evaluating economic evidence is a critical thinking skill that helps you distinguish strong arguments from weak ones. You learned that every claim should be tested against four key filters: source credibility, data relevance and representativeness, the correlation vs. causation distinction, and logical reasoning free from cherry-picking. Evidence types range from randomized controlled trials at the top of the strength pyramid down to anecdotes and personal opinions at the bottom.

Strong economic evidence comes from credible, unbiased sources that use large, representative samples and control for confounding variables. Weak evidence often relies on small samples, confuses correlation with causation, or cherry-picks data. By applying these evaluation principles consistently, you become a more informed consumer of economic information—a skill that is valuable in business, in civic life, and in your everyday decision-making.

Varsity Tutors • High School Economics • Evaluating Economic Evidence