Words to know
observational study
A study that observes groups as they are, without assigning who gets what; it can show links but struggles to prove causes.
randomized controlled trial
An experiment that randomly assigns participants to groups, so other differences tend to even out.
confounding variable
A third factor that is linked to both the supposed cause and the effect, creating a misleading connection.
relative risk
How many times larger one group's risk is than another's (for example 1.5 times, or 50% higher).
absolute risk
The actual chance of the outcome in a group (for example 3 in 100).
representative sample
A sample whose members resemble the larger population the claim is about.
SIFT
A four-move method for checking information: Stop; Investigate the source; Find better coverage; Trace claims to the original.
Hook: 50% sounds terrifying
A headline says a habit makes something 50% more likely. That sounds alarming. But 50% more of what? If the risk goes from 2 in 100 to 3 in 100, that is indeed 50% higher, and it is also one extra person in a hundred. Both statements are true. Which one a writer chooses tells you something about the writer.
Evaluating an informational text means judging how good its evidence is for the claim it makes. This lesson gives you the questions researchers, journalists and university instructors ask.
Remember from Grade 11
In ENG3U you evaluated texts with criteria, fact-checked an online biography with lateral reading, and questioned data graphics for misleading axes, small samples and correlation presented as causation. This lesson goes further into research evidence: study design, relative versus absolute numbers, confounders and conflicts of interest.
A text to practise on
Winter Cyclists Face 50% Higher Injury Risk, Study Finds
(An invented news article written for this lesson; the study and numbers are not real.)
(1) Cyclists who keep riding through the winter are 50% more likely to be injured in a fall than those who stop in November, according to a new study.
(2) Researchers surveyed 600 members of three cycling clubs in two Ontario cities. Over one season, 3 in every 100 winter riders reported a fall injury, compared with 2 in every 100 riders who stored their bikes for the winter and walked or took transit instead.
(3) "These results are a wake-up call," said a spokesperson for a company that sells car-share memberships, which helped fund the survey.
(4) Local cyclist Dana Ferreira, who has ridden year-round for twelve years, disagreed. "I've never been hurt once," she said.
(5) The researchers noted that winter riders in the study were, on average, younger and rode longer distances than the comparison group.
Checking a source: SIFT
The researcher Mike Caulfield's SIFT method gives four quick moves for checking a claim online:
- Stop: notice your reaction. A claim that makes you angry or delighted deserves a pause.
- Investigate the source: leave the page and find out who is behind it (this is the lateral reading you practised in Grade 11).
- Find better coverage: look for other reliable reports of the same claim.
- Trace claims, quotes and media to the original context: find the actual study, the full quotation, the original photo.
Expertise must be relevant. A cardiologist is an expert on hearts, not on traffic engineering. Ask whether a source's expertise matches this claim.
Quiz
1. A risk rises from 2 in 100 to 3 in 100. The absolute increase is...
2. Why can't an observational survey easily prove that winter riding causes injuries?
3. A confounding variable is...
4. The car-share spokesperson's quotation should be read with caution because...
6. A single person's experience is strongest as evidence for...