Small-Step Live Programming by Example [pdf]
cseweb.ucsd.edu
cseweb.ucsd.edu
"There are also several threats to the validity of our results. Our survey was conducted in the presence of one of the authors, which could lead to a social desirability bias. Additionally, the phrasing of questions was not neutral (e.g. “SNIPPY helped me write my code” instead of “How helpful was SNIPPY in writing your code”). (...)
Finally, the small sample size and short length of tasks could be a threat to the internal validity of the study."