I skimmed the Fietelson's paper you linked and this jumped out at me: "Students should generally not be used in studies that depend on specific expertise which

I skimmed the Fietelson's paper you linked and this jumped out at me: "Students should generally not be used in studies that depend on specific expertise which requires significant experience and a long learning curve to achieve, or in studies of professional practices. Such studies are best performed by observing and interviewing professionals, not by controlled experiments." This seems directly relevant to this TDD study. Every one of these contraindications are true in this case.