This week’s rant is brought to you by Ryan Kroll, Michaels’ Program Evaluation Manager. Last week Ryan issued a Program Brief discussing 90/10 confidence and precision sampling that is the industry norm for energy efficiency program impact evaluation. The 90/10 simply means the results of the sampled projects have a 90% probability of being within plus or minus 10% of properly representing the entire population – and NOT necessarily the right answer. Sample representation and the right answer are obviously different things. Here is the perversion in impact evaluation that I’ve never seen written anywhere: The less you learn about…
Read More
