The purpose of this study was to evaluate the validity and responsiveness of the new criterion-based test instrument test for athletes with knee-injuries (TAK) which has been evaluated for reliability in an earlier study. Thirty-five subjects between 18 and 50 years were included in the study. They were all anterior cruciate ligament (ACL)-reconstructed and operated with hamstrings graft. The test-occasions were at 4 and 8 months after operation. The content validity of the TAK was evaluated by determining floor and ceiling effects 4 and 8 months after ACL-operation. Floor or ceiling effects <30% were considered acceptable. The criterion validity was evaluated by implementing correlations between the TAK and the two common used validated and reliable scores, the International Knee Documentation Committee Subjective Knee Form (IKDC) and the Short Form-36 (SF-36). To evaluate the construct validity the performance of the eight tests of the TAK was compared to the performance of the healthy leg that represented normal ability. The hypothesis was that the patient and the physiotherapist who assessed the knee function decreased compared to the healthy leg would have a lower score of the TAK. The responsiveness of the TAK was evaluated by comparing the effect size of the test-instruments between 4 and 8 months after the operation. At 4 months after operation there were no floor or ceiling effects in any of the eight tests in TAK, except in "Test IV-patients' assessment" and in "Test I-physiotherapist's assessment" that both demonstrated a ceiling effects of 31%. At 8 months there were ceiling effects in five of the eight tests in "TAK-patients' assessment" and in all eight tests assessed by the physiotherapist. IKDC/sport activities had ceiling effects in five of the nine activities at 4 months and in all of them at 8 months. SF-36/scale physical functioning had no floor or ceiling effect neither 4 nor 8 months after operation. At 4 months postoperatively the correlation between the TAK/patients' assessment and IKDC/sport activities was moderate (0.59), while the correlation between TAK/physiotherapist's assessment and IKDC/sport activities was low (0.47). At the same test-occasion there were moderate correlations between SF-36/scale physical functioning and TAK/patients' assessment and TAK/physiotherapist's assessment (0.61 and 0.57), respectively. At eight months the correlations were moderate within all areas except between TAK/patients' assessment and IKDC/sport activities where the correlation was high (0.70). Construct validity for TAK was good seeing that all of the tests showed significant differences between the operated and the healthy non-operated leg (P <or= 0.05). The responsiveness of the TAK measured as the effect size was large in five tests and moderate in three tests. The total points for TAK/patients' assessment and TAK/physiotherapist's assessment showed large effect size (1.08 and 1.03), respectively. IKDC/total score and SF-36 scale physical functioning showed moderate effect size (0.79 and 0.41), respectively. The result of this study showed that the TAK is a reliable instrument with a large effect size for changes over time, indicating that the TAK is a very sensitive functional test instrument assessing the ACL-reconstructed patients during the rehabilitation period.