First of all, the article is from 2010. TEP only started the year prior, and the city/state did not have the metrics to publish. Now they do. I think merit pay

First of all, the article is from 2010. TEP only started the year prior, and the city/state did not have the metrics to publish. Now they do. I think merit pay (or just higher pay) and high demands are too different things, and the latter often destroys the benefit of the former I have to respectfully disagree, and I don't think you truly believe that given your 2nd to last paragraph. Following the logic, more money -> more competition for that employer -> employer can have greater selectivity in hiring people who can handle higher demands. Also, TEP is largely if not entirely exempt from lengthy union due process in dismissing underperforming teachers. This was as good a lab experiment as you'll get in educational research, and the empirical data simply was not positive. There is legitimate discussion to be had about potentially improving the model - there are even legitimate arguments NOT to improve the model but run it on its own merits for moral reasons if people like it - but we cannot use a data-driven philosophy and then let our eagerness to disprove the null hypothesis lead us to ignore the data point.