Effects of Variable Names on Comprehension: An Empirical Study —- Eran Avidan

Another paper about naming! (this whole session is)

Eran studies the impact of variable names on comprehension, by doing a controlled experiment with 8 developers and 6 methods in 38 sessions over 22 hours. One group (control) got the code as is, the full method without the method name, in the other group, all variables were replaced with single letters.

The findings were interesting: the one letter group took significantly more time to find the meaning of the methods, but the one letter group was more likely to be correct! Because in some cases, there were wrong or misleading names, that made comprehension harder.

A few wrong/misleading names:

Furthermore the participating developers indicated that parameters matter more to comprehension then do local variables names.

Read the full paper if you want to know more.