Statistically comparing the power usage of a 500 watt refrigerator to a 100 watt lightbulb.?
Question by Daniel Aureli: Statistically comparing the power usage of a 500 watt refrigerator to a 100 watt lightbulb.?
A company claims that its 500 watt refrigerator uses less power than a 100 watt light bulb when they are both constantly running. What is the best statistical method to use to compare data gathered on the power usage. I was thinking Hypothesis testing. I would really appreciate your input on what evaluation method would be most accurate. Thank You.
Best answer:
Answer by cidyah
Yes, hypothesis testing is OK. You need a random sample of refrigerators (some with power greater than 100 watts light bulbs and some with less than 100 watts light bulbs).
The parameter of interest is the amount of time when they are both constantly running.
H0: No difference in time regardless of what light bulb is used
HA : Bulbs with power less than 100 watts runs the refrigerator longer.
Your sample mean should reflect what you hypothesize. If the mean time is not more with less than 100 light bulbs, you should change your alternative to a two-tailed hypothesis.
Know better? Leave your own answer in the comments!
