This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: PROBLEM SET 1 ECONOMICS 418 UNIVERSITY OF ARIZONA (1) Explain why the following claims might not imply causal relationships and outline how you would test for a causal effect using an experimental methodology with a treatment and control group (a) People with smaller ears earn more money This relationship is unlikely to be causal. One possibility for the stated correlation could be some anthropological peculiarity of the surveyed individuals, but it is not the size of ears that determines earnings. To test for a causal effect, wed draw a random sample from the population (e.g. all US adults), willfully change the ear size for half of them, and then (after som suitable delay) observe their differences in earnings. (b) Smokers earn less There are likely to be many reasons why smokers earn less that are not directly due to smoking. A simple one is that less educated people are more likely to smoke and also have lower eanings. To test for a causal effect, wed draw a random sample from the population, force half of them to take up smoking, and then (after a delay) observe the differences in earnings. (c) People with American college education earn more than people with European college education Some of the effect is likely to be causal. To test for a causal effect, wed take a random sample from all college students, and give half of them American college education and half of them...
View Full Document