The aim of this study was to compare the validity of self-reporting and video-recording as methods of measuring the duration of knee-straining work postures, and to evaluate the reproducibility of timings of the video-recordings. Thirty-nine carpenters and 33 floorlayers were video-filmed while working, and were asked immediately afterwards to quantify the amount of time spent in knee-straining positions. The investigators recorded the periods of knee-straining work with a stopwatch during playback of the video-film, and the agreement between the two investigators' measurements was studied in 13 pairs of measurements. The video-observation method was very effective for timing knee-strain work. Furthermore, Spearman's correlation coefficient (0.88) indicated a good association between observed and self-reported knee-straining work. Estimation of self-reported knee-straining work collected by interview showed good correlation with measurements of video-recordings. When judging the repeatability of timings of knee-straining work from the video-recordings, a high level of agreement was seen between the two observers in the 13 dual measurements of time spent in knee-straining work positions.