Measuring test-retest reliability (TRR) of AMSTAR provides moderate to perfect agreement – a contribution to the discussion of the importance of TRR in relation to the psychometric properties of assessment tools

Table 1 Median Test-retest-reliability (Gwet’s AC1) per Amstar item 1–11 reviewer (R1-R7), for n = 16 SRs

AMSTAR-Item	1	2	3	4	5	6	7	8	9	10	11	Median (Range)
R1	1	0.72	0.64	0.75	0.9	1	0.86	0.75	0.64	0.63	0.92	0.75 (0.63–1)
R2	1	0.9	0.92	0.63	0.89	0.75	1	0.68	0.84	0.64	1	0.89 (0.63–1)
R3	1	0.8	0.84	0.34	0.63	0.34	0.86	0.75	0.5	−0.02	0.93	0.75 (0.2–1)
R4	1	0.63	0.82	0.63	0.88	1	0.86	0.4	0.82	0.88	0.72	0.82 (0.4–1)
R5	1	0.8	0.8	0.6	0.53	0.86	0.58	0.69	0.82	0.5	−0.02	0.69 (−0.02–1)
R6	1	0.75	0.77	0.53	0.15	0.4	0.93	0.68	1	0.76	0.72	0.75 (0.4–1)
R7	1	0.88	0.92	0.88	0.51	0.77	1	0.68	0.53	0.6	0.75	0.77 (0.51–1)
Median (Range)	1 (1–1)	0.8 (0.63–1)	0.82 (0.64–1)	0.6 (0.53–1)	0.53 (0.15–1)	0.77 (0.34–1)	0.86 (0.58–1)	0.68 (0.4–1)	0.82 (0.5–1)	0.6 (0.2–1)	0.72 (−0.02–1)

Legend: light gray: moderate agreement, medium dark colored: substantial agreement, dark colored almost perfect agreement

ISSN: 1471-2288