← Back to Insights

Module: Reading Research Studies

By SAUFEX Consortium 23 January 2026

[screen 1]

“Scientists discover chocolate helps weight loss!” The headline claims a study proves it. You read the actual study - it was 12 people for two weeks with no control group.

Headlines routinely overstate research findings. Understanding how to read studies helps you evaluate scientific claims without needing a PhD.

[screen 2]

Why This Matters

Research is increasingly used to support claims in public discourse. But most people encounter research through media headlines, not original studies.

This creates opportunities for misrepresentation:

  • Headlines exaggerate findings
  • Important caveats are omitted
  • Single studies are treated as definitive
  • Preliminary research is presented as proven
  • Corporate-funded studies downplay conflicts of interest

Basic research literacy is self-defense.

[screen 3]

Anatomy of a Research Paper

Most research papers follow standard structure:

Abstract: Brief summary of entire study

Introduction: Background and research question

Methods: How the study was conducted

Results: What the data showed

Discussion: Interpretation and limitations

Conclusion: What researchers conclude

You don’t need to read everything - abstract, methods, and discussion/limitations are most important for evaluation.

[screen 4]

Peer Review Explained

Peer review means other experts in the field evaluate research before publication:

  • Reviewers check methodology and reasoning
  • They can request changes or reject publication
  • Process is usually anonymous
  • Not perfect - bad studies get through, politics can interfere

Peer-reviewed publication in reputable journals is a quality signal, but not a guarantee of truth. Non-peer-reviewed work (preprints, press releases) should be treated more skeptically.

[screen 5]

Types of Studies

Different study types provide different levels of evidence:

Randomized Controlled Trial (RCT): Gold standard - random assignment to treatment or control group

Observational: Watching what happens naturally without intervention (can show correlation, not causation)

Meta-analysis: Combining results from multiple studies (strong evidence when done well)

Case study: Detailed examination of individual cases (useful for rare phenomena, not generalizable)

Survey: Asking people questions (subject to response bias)

Understanding study type helps evaluate strength of conclusions.

[screen 6]

Sample Size and Selection

Two critical questions:

How many participants?

  • Small samples (< 30): Very preliminary
  • Medium (30-100): Suggestive
  • Large (1000+): More reliable
  • Very large (10,000+): Can detect small effects

How were they selected?

  • Random sampling from target population: Best
  • Convenience samples (students, volunteers): Limited generalizability
  • Self-selected: Highly biased

A study of 20 college students doesn’t tell you much about humans in general.

[screen 7]

Control Groups and Randomization

Quality experiments include:

Control group: Comparison group receiving placebo or standard treatment (without this, you can’t know if your intervention caused the effect)

Randomization: Participants randomly assigned to groups (prevents selection bias)

Blinding: Participants and/or researchers don’t know who’s in which group (prevents expectation effects)

Studies without these features are much weaker evidence.

[screen 8]

Reading the Limitations Section

The most honest part of a research paper is often the “limitations” section, where researchers acknowledge weaknesses:

  • Small sample size
  • Short duration
  • Limited generalizability
  • Potential confounding factors
  • Measurement challenges

Headlines never mention limitations. Reading them yourself provides essential context for evaluating claims.

[screen 9]

Correlation vs. Causation Revisited

Observational studies can show correlation but not causation. Just because A and B occur together doesn’t mean A causes B:

  • B might cause A (reverse causation)
  • C might cause both A and B (confounding variable)
  • It might be coincidence

“Ice cream sales correlate with drowning deaths” - doesn’t mean ice cream causes drowning. Both increase in summer (confounding variable: warm weather).

Only controlled experiments can establish causation.

[screen 10]

Conflicts of Interest

Who funded the research matters:

  • Tobacco companies funded studies downplaying smoking risks
  • Pharmaceutical companies fund drug trials (positive publication bias)
  • Food industry funds nutrition studies favoring their products
  • Political advocacy groups fund research supporting their positions

This doesn’t automatically invalidate research, but it’s a red flag requiring extra scrutiny. Look for independent replication.

[screen 11]

How Media Distorts Research

Common patterns when research becomes news:

  • Causation claimed from correlation: “Coffee causes cancer” from observational study
  • Exaggerated effect sizes: “50% reduction!” sounds dramatic but might be meaningless if baseline risk is tiny
  • Ignoring limitations: Small sample, short duration, or conflicts of interest omitted
  • Extrapolating beyond data: Animal studies presented as if proven in humans
  • Single study treated as definitive: Science requires replication

Always try to find the original study when evaluating research claims.

[screen 12]

Evaluating Research Claims

Before accepting research-based claims:

  1. Find the original study (not just media coverage)

  2. Check if it’s peer-reviewed

  3. Look at sample size and selection

  4. Check study type (observational or experimental?)

  5. Read the limitations section

  6. Identify who funded it

  7. Check for conflicts of interest

  8. See if it’s been replicated

  9. Look for systematic reviews or meta-analyses

  10. Consider if media coverage matches actual findings

[screen 13]

Red Flags in Research

Be skeptical when you see:

  • Press release but no published paper
  • Extraordinary claims with weak methodology
  • Conflicts of interest not disclosed
  • Refusing to share data or methods
  • Results never replicated
  • Published in low-quality or predatory journals
  • “Breakthrough” claims based on preliminary research
  • Researchers making claims beyond their data

Quality research acknowledges uncertainty and limitations.

[screen 14]

Building Research Literacy

You don’t need to become a scientist to evaluate research claims:

  • Learn to read abstracts and limitations
  • Understand basic study types
  • Know that single studies prove little
  • Check funding and conflicts
  • Find the original source before accepting headlines
  • Look for systematic reviews over single studies

Science is a cumulative process. Individual studies are pieces of larger puzzles, not definitive answers. Media literacy requires understanding this distinction.