When studying discomfort glare, researchers tend to rely on a single questionnaire item to obtain user evaluations. It is unclear whether the choice of questionnaire item affects the distribution of user responses and leads to inconsistencies between studies. This study aims to investigate if different glare questionnaire items yield similar distributions of user discomfort in daylit environments. We conducted a comparative study of selected questionnaire items from previous glare experiments, testing them in three independent user studies with different lighting conditions and glare stimuli. We compared the resulting outputs across questionnaire items with 540 data points from 149 participants. Results indicated that ordinal questionnaire outputs show strong correlations (0.68 < ρ _< 0.85), high internal reliability (α _= 0.93), and captured the same latent construct. Binary questionnaire items reflected different glare thresholds but still correlated well with ordinal items. The construct validity of tested questionnaire items was confirmed through responses to an open-ended question. These findings suggest that the tested questionnaire items may be used for category rating-type discomfort glare evaluations and consistently capture the same construct.