This is the first study aiming to evaluate the reliability of assigning a difficulty grade to a climb. Each of the 70 climbers, divided by their abilities into group A (intermediate) and group B (advanced), climbed 5 different routes, and independently suggested a grade for the routes. Although the reliability was generally high (Kendall W = .89 and .86, intraclass correlation coefficient ICC = .82 and .89 in groups A and B, respectively), the grades proposed by a single climber have too large a standard error of measurement to confidently claim that two routes with different grades are truly of different difficulties. The most important factor associated with the accuracy and bias of grading, was the gap between a climber's ability and a route's difficulty (easier routes were graded less accurate, advanced climbers underestimated the true difficulty of the routes). The following factors were also found important for grading; their effect however, was lower, or was only found in some routes or in one group: success (failed attempts were graded higher), tiredness, sex and body height (but only unadjusted for sex). Surprisingly, no significant effects were found for climbing experience, ability level, and style of climbing (onsight, flash).
© Copyright 2021 Kinesiologia Slovenica. Faculty of Sport Universität Ljubljana. Alle Rechte vorbehalten.