You may not control the home life of your engineers, but you presumably have at least some control over who you hire and fire. Even if you don't have the author

You may not control the home life of your engineers, but you presumably have at least some control over who you hire and fire. Even if you don't have the authority to fire someone directly you can still give your subordinates poor evaluations. Imagine being a situation where not only did you not have any control over which engineers worked on your team - or even whether they were engineers at all, or even basically literate. Now imagine that your whole industry pretty much worked liked that and that changing from one company to another (which most likely worked the same way) always meant moving cities. Incidentally, what they did in this study is to measure the value-addition of each teacher, not their raw test scores. Presumably you're not evaluated based on things you have absolutely no control over, as you said - you get judged based on the slice of the pie that you do control.