To the author’s knowledge, no research has earlier been done on the development of a sport specific strength measurement system for para kayak classification.Aim: In order to work in line with the International Paralympic Committee, to develop more evidence-based methods for classification, the overall purpose of this study was to design and develop a sport specific test battery to measure force in isometric contractions. It was also: (1) to test-retest the battery on able-bodied participants, to ensure the reliability of the method for further research and (2) to evaluate whether performing two or three trials is sufficient when testing the test battery for reliability. Method: Ten able-bodied participants volunteered to this study. All participants performed a test battery of four isometric strength tests. The participants completed three voluntary maximal isometric contractions for each exercise with a duration of 5 seconds, and a 30 second rest separating each trial. Data used in the analysis, was the mean value of two respectively three trials. The Shapiro-Wilk test of normality was used to assess the normality of the data. The level of significance when testing normality was set to p < 0.05. A dependent t-test and an intraclass correlations coefficient (ICC) with a 2-way mixed effects- model, absolute agreement, method 3.1 were used to assess the test-retest reliability for both trial 1-2 and trial 1-3. Results: All tests showed high reliability and no systematic error were found. The data were normally distributed. When analysing data from three trials, one test showed good reliability; knee extension left (ICC=0,77). All other data showed excellent reliability. When analysing data from two trials, two tests showed good reliability; leg press left (ICC = 0,88) and knee extension left (ICC = 0,78). All other data showed excellent reliability (> 90). Conclusion: The results of this study indicate that all tests can reliably be applied in a classification system for para-kayak. Both of the methods, using three data points and using two data points, can be used when analysing the data.