Skip to main content
Fig. 3 | BMC Medical Research Methodology

Fig. 3

From: Comparison of different rating scales for the use in Delphi studies: different scales lead to different consensus and show different test-retest reliability

Fig. 3

Test-retest agreement: comparison across rating scales:

Pirateplot of the percentage agreement and the weighted kappa distributions between survey 1 and survey 2. Each data point represents the value for a single treatment goal. The beans represents the smoothed density, the boxes the 95% confidence interval with the middle line as the average. Five-point/nine-point scale*: rating scale mapped onto three categories: - “very important” (five-point scale) and “9, 8, 7” (nine-point scale) as “main goal” - “a little important”, “somewhat important” (five-point scale) and “5, 6, 7” (nine-point scale) as “secondary goal” -“I do not expect this” and “this does not apply to me” (five-point scale) and “3, 2, 1” (nine-point scale) as “no goal

Back to article page