1. Epidemiology and Global Health
Download icon

Science Forum: Unit of analysis issues in laboratory-based research

  1. Nick R Parsons  Is a corresponding author
  2. M Dawn Teare
  3. Alice J Sitch
  1. University of Warwick, United Kingdom
  2. University of Sheffield, United Kingdom
  3. University of Birmingham, United Kingdom
Feature Article
Cite as: eLife 2018;7:e32486 doi: 10.7554/eLife.32486
4 figures and 3 tables

Figures

A strip plot showing observed lymph node size data by subject (1-12) and sample, after none and a short course of radiotherapy (Short RT).
https://doi.org/10.7554/eLife.32486.004
Boxplots of residuals (observed values - fitted values) for each subject; symbols () are medians, boxes are interquartile ranges (IQR), whiskers extend to 1.5×IQR and symbols () outside these are suspected outliers (a).

Quantile-quantile (Q–Q) plot of the model residuals () on the horizontal axis against theoretical residuals from a Normal distribution on the vertical axis (b).

https://doi.org/10.7554/eLife.32486.005
Appendix 1—figure 1
Naive use of a conventional t-test on correlated (grouped by subject) data, ρ = 0 (black circle ), ρ = 0.2 (red circle) ρ = 0.5 (blue circle) and ρ = 0.8 (green circle), inflates the type I error rate (set at 5%).

(a). The type I error rate can be controlled to the required level by randomly selecting a single measurement for each subject, ρ = 0 (black circle), ρ = 0.2 (red circle), ρ = 0.5 (blue circle) and ρ = 0.8 (green circle), or made conservative (5%) by taking the mean of the measurements for each subject, ρ = 0 (black open circle), ρ = 0.2 (red open circle), ρ = 0.5 (blue open circle) and ρ = 0.8 (green open circle) (b). The relative efficiency of treatment effect estimates declines as the number of clusters become smaller and is always higher for the mean than the randomly selected single measurement strategy (c). The scenarios (i) – (vi) are as described in the text.

https://doi.org/10.7554/eLife.32486.008
Appendix 2—figure 1
Design options for a putative laboratory study testing n samples of experimental material.
https://doi.org/10.7554/eLife.32486.010

Tables

Table 1
Variance inflation factors for cluster sizes (m) 2, 5, 10 and 20, and intraclass correlation coefficients (ICC) 0.01, 0.05, 0.1 and 0.5.
https://doi.org/10.7554/eLife.32486.002
mICC
0.010.050.10.5
21.011.051.101.50
51.041.201.403.00
101.091.451.905.50
201.191.952.9010.50
Table 2
Lymph node sizes (mm), by sample slice and subject, by radiotherapy (RT) group, subjects 1 to 6 no RT and subjects 7 to 12 short RT; highlighted cells are those removed to unbalance the design.
https://doi.org/10.7554/eLife.32486.003
NoneShort RT
SubjectSampleSliceSubjectSampleSlice
123123
111.711.981.88712.372.362.20
21.721.981.8522.362.622.60
212.512.552.65811.331.351.15
22.983.202.8021.901.871.85
311.691.721.80911.701.781.78
21.821.971.7322.071.761.85
411.721.782.041012.232.142.21
22.502.652.7722.502.332.16
513.323.273.071112.101.891.75
23.113.033.1122.112.162.12
612.332.482.531212.582.542.59
22.862.872.5222.772.652.60
Table 3
Number of five selected lymph nodes with maximum diameters 2mm, for up to five tissue samples per subject (1-12), after either none or a short course of radiotherapy (Short RT).
https://doi.org/10.7554/eLife.32486.006
NoneShort RT
SubjectSampleSubjectSample
1234512345
144---710000
23452-812---
32332-910102
4241211021402
5344351142433
62553312343--

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)