Table 1. Summary of results from this and previous studies.

The study is identified in the “Reference” column; “.” indicates this study (values in red). Numbers of children for our study refer to testability and are smaller for other columns; see relevant sections for details. Values in italics were inferred from the data provided in the paper, e.g. by reading off values from figures and/or using the formulae in Table 2. Note that Yang et al[15] claim to provide sensitivity and specificity for the Randot Preschool, but this appears to be incorrect. First, their sample is described as “100 normal children without ocular disease” and specifically excluded children with strabismus or amblyopia, so it is not clear who the “true positives” would be. Second, the denominators used in the calculation of sensitivity and specificity are the number of children in each age-group, not the number of children passing/failing the test. Third, the sensitivity values are over 90% in all age-groups, far higher than any other report. Thus, we have excluded their data.

Age in years	Reference	Number of children	Testability (% who understand & cooperate)	Normative data (i.e. in children with apparently normal vision)						Reliability			Validity
				% of testable scoring nil	Stereo thresholds in arcsec			Stereo log-thresholds in log10 arcsec		95% coefficient of repeatability (1.96 SD of differences[20] in log10arcsec)		Pearson correlation	Sensitivity \ True Positive %	Specificitiy \ True Negative %
					Mean	Median	SD	Mean	SD	Estimate	95% CI
2	.	49	65	14	382c	400	287	2.44	0.38
	[14]		33
	[15]	19	47		332
	[16]	411	31
	[17]	130	32
	[13]	400h		2	216			2.17	0.37
3	.	150	92	7	214c	100	234	2.13	0.40	0.71	0.42–1.0	0.72	14j	91 j
	[14]		73
	[15]	34	85		135
	[16]	366	67
	[17]	287	75
	[12]	138			100a		150	1.74	0.47
	[13]	1606h		3	114			1.92	0.35				27h	99h
4	.	161	99	2	134c	100	148	1.97	0.34	0.82	0.60–1.03	0.05
	[14]		96
	[15]	25	96		71
	[16]	365	88
	[17]	297	96
	[12]	217			100a		50	1.95	0.21
	[13]	400h		1	94			1.80	0.38
	[18]	100e								0.64e		0.97e
5	.	101	99	5	87c	60	76	1.84	0.27	0.64	0.42–0.87	0.59	60 j	98 j
	[14]		98
	[15]	22	95		51
	[16]	373	95
	[17]	300	98
	[12]	104			60a		70	1.59	0.40
	[13]	400h		0	71			1.69	0.38
6–7	.	256	100	1	91c	40	120	1.80	0.31	0.56	0.42–0.70	0.72	31 j	99 j
	[12]	46			60ab		20b	1.76b	0.14b
10–11	.	195	100	1	63c	40	93	1.69	0.23	0.28	0.20–0.36	0.86	26 j	99 j
	[12]	56i			40ai	40	10i						24g
	[8]	19								0.60d	0.35–0.85
	[19]	47f								0.23f	0.12–0.35

a: Study [12] states means are “rounded to the next larger disparity level available in the Randot Preschool”, but this is not consistent with the fact that in some age-groups the mean is given as 40, the best available score, and yet the SD is non-zero, meaning that some children must have scored worse than 40. In this case the pre-rounding mean must have been >40 and so “rounding to the next larger disparity level” would have given a rounded mean of 60. Means may have been rounded to the closest available level.

b. Data from 46 children aged 6 years.

c: We calculated means for non-stereoblind children only, which would reduce our estimates compared to those including all children.

d. For 19 children aged 7–18.

e. For 100 children aged 2–12, but mainly aged around 4. This cohort included 75 clinical patients and 31 out of the 100 were stereoblind, thus ensuring perfect agreement (fail both times).

f. For 47 participants with microtropia, heterophoria or orthophoria, aged 3–80 years, around half children.

g: For 242 patients with amblyogenic conditions, aged 3–18 years, counting 800 arcsec as a fail (57/242 failed).

h: For 1606 children aged 2–5 years, screening for strabismus, counting 800 arcsec as a fail. Sensitivity was 24% screening for amblyopia, and 9% screening for anisometropia; specificity was similarly high for all conditions.

i: Data from 56 children aged 9 and 10 years.

j: Our sensitivity/specificity analysis used larger age-groups, see relevant section.