Eh, they were just using a hypothetical example to demonstrate how we interpret data based on context in general. Nitpicking how the hypothetical example doesn't fit the validity of the population kind of misses their point since that wasn't what they were trying to explain.