State limitations
Name what could weaken a result: small sample size, measurement error, and confounders: and what the data still can't show.
- Sample vs. whole group: A limitation often comes from testing only a few cases, so you must see how a sample relates to the whole group.
- What a measurement supports: Stating limits means knowing which claims the data can and cannot back up.
Prerequisites are inferred: pending teacher review.
Re-learn the skill with worked practice and clear examples.
State limitations by naming small sample size, measurement error, and confounders: and by saying what the data still can't show.
A team measured lead (ppb) in four water samples from one home, shown below, and concluded the city's water is unsafe. Which is the strongest limitation of that conclusion?
Reviewed| Sample | Lead concentration (ppb) |
|---|---|
| 1 | 6 |
| 2 | 9 |
| 3 | 7 |
| 4 | 10 |
- A.The samples were measured in the wrong units
- B.Four samples from one home can't represent a whole city's water
- C.Lead is not harmful at any level
- D.A data table cannot hold lead measurements
Show the worked solution ▾
Answer: B. Four samples from one home can't represent a whole city's water
- Step 1: Compare sample to claim: Only four samples, all from one home, were taken. The conclusion is about the entire city.
- Step 2: Name the limitation: Such a small, single-location sample can't be generalized to every home in the city: that is the strongest limitation.
Why it's right: Four samples from one home is a small, single-site sample, so generalizing to the whole city is the strongest limitation here.
- A: The units (ppb) are appropriate for lead in water.
- C: Whether lead is harmful is not the flaw in this study's reasoning about sampling.
- D: A data table is a fine place to record lead measurements.
Aligned to HBS investigation: limitations · reading level ~grade 9
- A 'Limitations' section lists small n, measurement error, and possible confounders so readers weight the conclusion fairly.
Fill these in as you work through the lesson.
- Limitation (a reason the result might not be trustworthy or general):
- Sample size (how many cases were tested):
- Measurement error (how far a reading can be off from the true value):
- Confounder (a second factor that could explain the result):
- Generalize (to extend a finding beyond the cases tested):
A small makes it risky to the result to everyone. Measurement means a single reading might not be exact, and a hidden could be the real cause.
- Why is a result from only three water samples hard to trust for a whole city?
- Name one source of measurement error when reading a lead test.
- Give one thing a single lead reading at one tap still can't tell you.
A team measured lead in three tap-water samples from one building and concluded the whole city's water is unsafe. List three limitations of that conclusion.