Medians and percentage distribution within the response categories were used for providing graphically documented results to the participants of the second Delphi round (Figure 3). In some cases, two boxes were marked for one rating scale. Instead of excluding these answers, it was assumed that ...