Causation vs. Correlation

  • Correlation means two things happen at the same time or change together, but one doesn’t necessarily cause the other.

  • Causation means one thing directly influences or produces the other.

A common mistake is assuming that because two things are related, one must cause the other.

Example 1: Ice Cream Sales & Drowning Incidents

  • Observation: When ice cream sales go up, drowning incidents also increase.

  • Correlation: These two rise together.

  • Possible Flawed Reasoning: Eating ice cream increases risk of drowning. 

  • But not causation: Eating ice cream doesn’t cause drowning. The actual cause is hot weather — more people buy ice cream and go swimming in the summer.

Example 2: Shoe Size & Reading Ability in Children

  • Observation: Kids with bigger shoe sizes often read better than kids with smaller feet.

  • Correlation: Shoe size and reading skill increase together.

  • Possible Flawed Reasoning: Students in Third Grade with Smaller Feet need proactive tutoring for reading. 

  • But not causation: Big feet don’t make someone a better reader. The real cause is age — older children naturally have larger feet and more developed reading skills.

How Do Child Abuse Pediatricians Use This Flawed Reasoning?

Examples

How CAPs might interpret the Correlation as Causation

Correlations

Example: Misunderstanding Correlation vs. Causation

Topic: Twins

Observation:
Studies show that infant twins are reported to have more injuries than singletons.

Child Abuse Pediatrician Reasoning:

  • Parents of twins must be at a higher risk of being abusive.

  • Since twins have more injuries, they must be abused more often.

  • This must have an explanation — perhaps parents of twins are more frustrated and overwhelmed, leading to abuse.

  • Alternatively, because many twins are born through fertility treatments, these parents may have had unrealistic expectations about parenting and became angry when reality was harder than expected.

This reasoning assumes that a higher number of injuries means a higher rate of abuse — without considering other explanations.

Reasonable Interpretation (Understanding Correlation ≠ Causation):
Why might twins have more accidental injuries?

  1. Practical care challenges: Caring for two infants at once is complex. One caregiver managing two babies increases the likelihood of accidental bumps, falls, or mishandling — even with attentive care.

  2. Interaction between twins: Even young infants can move enough to bump into each other or cause minor accidental injuries.

  3. Biological factors: Twins often share nutrients in the womb, which can lead to developmental differences affecting skin, bone density, or overall health — potentially making them more susceptible to accidental marks, bruising or fractures.

Conclusion:
A higher rate of diagnosed abuse in twins may reflect a failure to consider differential diagnoses. Correlation (more injuries in twins) does not equal causation (more abuse).