A recent study of 10,000 people who were involved in automobile accidents found that a low percentage of those driving large automobiles at the time of their accidents were injured, but a high percentage of those who were driving small automobiles at the time of their accidents were injured.
Conclusion
One is less likely to be injured in a car accident if driving a larger car.
Evidence
A study of 10,000 people who were in an accident found that a much lower percentage of large-car drivers were injured than small-car drivers.
Evaluation
Does this statistic show that drivers of larger cars are less likely to be injured in a car accident? Not quite. As always, with statistics, we have to be cautious about what it slice of the story it's giving us.
To support the claim the conclusion is making, we would want to look at 10,000 drivers and then find out what percentage of large-car drivers have ever been injured in an accident, and what percentage of small-car drivers have ever been injured in an accident.
This stat, meanwhile, looks at 10,000 people who were in an accident, and then finds out what percentage of large/small was injured.
To put it another way, the conclusion is about all drivers of large/small cars, regardless of whether they've been in an accident. The evidence is only about drivers who have been in an accident. The correct conclusion we could draw would be,
Goal
Find an answer pointing out the statistical flaw that we're only looking at a subset of drivers (those who have been in an accident) but concluding something about all drivers. Or find an objection that would help us argue the anti-conclusion, that small car drivers are less likely to be injured in an accident.