Skip to main content

Advertisement

Table 3 Bias [90% prediction intervals] of variance component estimates for percentage time above 90° for a subset of the investigated sampling strategies

From: Accuracy and precision of variance components in occupational posture recordings: a simulation study of different data collection strategies

n s 10 10 10 10 20 20 20 20
ttot 120 120 480 480 120 120 480 480
nd 2 4 2 4 2 4 2 4
3a. Between-subjects variance, σ 2 BS
t b 1 r 0.0 −0.2 −0.1 −0.1 −0.1 −0.1 −0.2 −0.2
[−3.0, 5.3] [−3.0, 4.1] [−3.0, 4.1] [−2.8, 3.0] [−3.0, 3.6] [−2.6, 2.7] [−2.9, 2.7] [−2.0, 1.9]
  1 f 0.0 −0.1 0.0 −0.3 −0.1 −0.2 0.0 −0.3
[−3.0, 5.0] [−3.0, 3.8] [−3.0, 4.2] [−2.9, 2.5] [−3.0, 3.4] [−2.4, 2.5] [−2.7, 2.7] [−2.1, 1.7]
  15 r 0.8 0.4 −0.1 −0.2 0.2 0.0 −0.2 −0.2
[−3.0, 11.1] [−3.0, 9.7] [−3.0, 4.8] [−3.0, 4.1] [−3.0, 7.6] [−3.0, 6.3] [−3.0, 3.1] [−2.6, 2.6]
  15 f 0.5 0.3 −0.5 −0.6 0.1 0.0 −0.6 −0.7
[−3.0, 9.9] [−3.0, 9.4] [−3.0, 3.3] [−3.0, 2.8] [−3.0, 6.6] [−3.0, 6.2] [−3.0, 2.1] [−2.8, 1.6]
  60 r 1.6   0.0 −0.1 0.7   −0.2 −0.2
[−3.0, 17.3]   [−3.0, 5.8] [−3.0, 5.5] [−3.0, 11.5]   [−3.0, 3.7] [−3.0, 3.6]
  60 f    −0.3 0.1    −0.4 −0.1
[−3.0, 5.0] [−3.0, 5.7]    [−3.0, 3.2] [−3.0, 3.6]
3b. Between-days variance, σ 2 BD
t b 1 r −0.2 −0.1 −0.1 0.0 0.0 0.0 0.0 0.0
[−4.0, 6.1] [−4.0, 5.3] [−3.4, 4.8] [−2.5, 3.2] [−3.5, 4.6] [−3.3, 3.8] [−2.6, 3.3] [−1.9, 2.3]
  1 f −0.8 −1.1 −0.3 −0.3 −0.7 −1.1 −0.2 −0.3
[−4.0, 5.3] [−4.0, 3.7] [−3.4, 4.7] [−2.8, 2.9] [−4.0, 3.6] [−4.0, 2.4] [−2.8, 3.4] [−2.2, 1.9]
  15 r 8.2 20.2 0.9 3.5 8.7 20.3 1.0 3.6
[−1.8, 24.5] [4.0, 44.6] [−3.0, 6.7] [−0.7, 9.0] [0.5, 20.5] [7.5, 37.5] [−2.1, 5.1] [0.5, 7.5]
  15 f 7.1 19.2 0.1 2.6 7.3 19.4 0.3 2.7
[−1.9, 20.9] [3.9, 41.9] [−3.3, 5.0] [−1.0, 7.4] [0.1, 16.9] [7.4, 35.1] [−2.6, 4.0] [−0.1, 6.2]
  60 r 17.2   2.0 7.3 17.6   2.1 7.4
[−0.5, 51.0]   [−2.7, 9.0] [0.7, 17.0] [2.7, 42.0]   [−1.6, 7.2] [2.4, 14]
  60 f    1.3 7.7    1.5 7.9
[−2.8, 7.2] [1.0, 17.9]    [−1.8, 6.0] [2.6, 14.7]
3c. Within-day variance, σ 2 WD
t b 1 r 0.1 −0.1 −0.9 0.2 0.5 0.2 −0.3 −0.1
[−50.5, 54.8] [−46.7, 50.8] [−42.9, 42.3] [−37.3, 38.9] [−35.6, 39.1] [−33.1, 34.9] [−29.8, 31.4] [−27.2, 27.1]
  1 f 0.5 −0.2 1.1 0.5 0.6 0.4 1.4 0.3
[−48.9, 54.1] [−44.3, 47.7] [−42.0, 46.5] [−35.9, 38.4] [−34.3, 37.3] [−32.2, 33.4] [−29.9, 33.6] [−26.2, 25.9]
  15 r −8.8 −19.0 0.7 −2.1 −6.9 −18.7 0.6 −1.9
[−72.9, 61.4] [−74.3, 43.3] [−43.7, 47.3] [−44.3, 42.9] [−53.8, 43.5] [−60.0, 26.1] [−31.6, 34.4] [−31.5, 29.5]
  15 f −8.5 −16.7 −2.4 −6.2 −9.2 −17.6 −1.5 −5.9
[−70.0, 60.7] [−73.8, 46.3] [−44.4, 43.5] [−44.6, 35.2] [−55.3, 39.9] [−59.3, 26.5] [−31.0, 29.2] [−33.8, 23.6]
  60 r −13.0   1.2 −3.2 −13.3   1.0 −2.7
[−80.6, 65.7]   [−45.6, 53.2] [−47.5, 45.8] [−61.0, 39.5]   [−34.0, 36.9] [−34.9, 32.5]
  60 f    −1.1 3.1    −0.5 4.1
[−47.2, 47.9] [−42.1, 52.8]    [−32.0, 34.0] [−29.0, 39.7]
  1. a, b, c: between-subjects, between-days, and within-day variance.
  2. n s , number of subjects; t tot , total sampling time per subject (minutes); n d , number of days per subject; t b , size of sampling blocks (minutes); r, random sampling; f, fixed interval sampling.
  3. Bias and prediction intervals are presented relative to the “true” variance components of the parent data set, i.e. σ 2 BS  = 3.0, σ 2 BD  = 4.0 and σ 2 WD  = 164.7 (cf. Table 2). Thus, the result in bold and italics tells that for the sampling strategy (n s , t tot , n d , t b ) = (10, 120, 4, 1) with random distribution of blocks, σ 2 BS was downward biased by 0.2, i.e. its average estimated value was 2.8, and the 90% prediction interval, presented as [−3.0, 4.1], reached from 3.0-3.0 to 3.0 + 4.1, i.e. from 0.0 to 7.1.