Skip to content

Please be aware that some material may contain words, descriptions or illustrations which will not reflect current scientific understanding and may be considered in today's context inaccurate, unethical, offensive or distressing.

The Royal Society: Science in the Making

Referee's report by George Ballard Mathews, on two papers 'On a property which holds good for all groupings of a normal distribution of frequency for two variables, with applications to the study of contingency-tables for the inheritance of unmeasured qualities' and 'On the influence of bias and of personal equation in statistics of ill-defined qualities: An experimental study' by George Udny Yule

Reference number: RR/16/416

Date: 13 December 1905

Description

Sectional Committee: not stated.

Suggests Yule recast his shorter paper adding the essential features of his experiment and withdraw the long paper. Queries what Larmor thinks. Thinks the papers should be very considerably cut down and as Larmor notes, the cost of the printing the tabular results would be very expensive. If he thought fit, Yule could for a small sum have his tables graphed or lithographed and sent round to Pearson and others. Thinks Yule has made very clear that tables recording qualities, not quantities, should be scrutinised very carefully before basing conclusions upon them. Each has to be taken on its own merits and for this reason Mathews does not see the use of making a permanent and elaborate record of Yule's colour experiment.

Briefly discusses the difference between the conclusions of the pairs of tints observers, and the horse-coat colour observers. Does not understand what Pearson means in his postscript. The examination is not confined to 18 series but many more, the result for 100 chestnuts being given. How far the latter support Mendelian theory Mathews must leave a biologist to decide. Explains why he would expect more divergence with the eye-colour results. Does not consider Yule's papers as anything like an attack on Pearson. There is bound to be controversy on these difficult questions of statistics and heredity for a long time to come and he does not see why different views should not appear in the Proceedings as well as elsewhere, provided nothing is printed which is in the least degree personally offensive.

[Both papers published in Proceedings of the Royal Society A, 1906].

Reference number
RR/16/416
Earliest possible date
13 December 1905
Physical description
Letter on paper
Page extent
3 pages
Format
Manuscript

Creator name

George Ballard Mathews

View page for George Ballard Mathews

Use this record

IIIF manifest

View​ IIIF ​ manifest Opens in new tab

Open in Mirador viewer Opens in new tab

Export this record

Download PDF Opens in new tab

Citation

George Ballard Mathews, Referee's report by George Ballard Mathews, on two papers 'On a property which holds good for all groupings of a normal distribution of frequency for two variables, with applications to the study of contingency-tables for the inheritance of unmeasured qualities' and 'On the influence of bias and of personal equation in statistics of ill-defined qualities: An experimental study' by George Udny Yule, 13 December 1905, RR/16/416, The Royal Society Archives, London, https://makingscience.royalsociety.org/items/rr_16_416/referees-report-by-george-ballard-mathews-on-two-papers-on-a-property-which-holds-good-for-all-groupings-of-a-normal-distribution-of-frequency-for-two-variables-with-applications-to-the-study-of-contingency-tables-for-the-inheritance-of-unmeasured-qualities-and-on-the-influence-of-bias-and-of-personal-equation-in-statistics-of-ill-defined-qualities-an-experimental-study-by-george-udny-yule, accessed on 10 February 2025

Link to this record

https://makingscience.royalsociety.org/items/rr_16_416/referees-report-by-george-ballard-mathews-on-two-papers-on-a-property-which-holds-good-for-all-groupings-of-a-normal-distribution-of-frequency-for-two-variables-with-applications-to-the-study-of-contingency-tables-for-the-inheritance-of-unmeasured-qualities-and-on-the-influence-of-bias-and-of-personal-equation-in-statistics-of-ill-defined-qualities-an-experimental-study-by-george-udny-yule

Embed this record

<iframe src="https://makingscience.royalsociety.org/embed/items/rr_16_416/referees-report-by-george-ballard-mathews-on-two-papers-on-a-property-which-holds-good-for-all-groupings-of-a-normal-distribution-of-frequency-for-two-variables-with-applications-to-the-study-of-contingency-tables-for-the-inheritance-of-unmeasured-qualities-and-on-the-influence-of-bias-and-of-personal-equation-in-statistics-of-ill-defined-qualities-an-experimental-study-by-george-udny-yule" title="Referee's report by George Ballard Mathews, on two papers 'On a property which holds good for all groupings of a normal distribution of frequency for two variables, with applications to the study of contingency-tables for the inheritance of unmeasured qualities' and 'On the influence of bias and of personal equation in statistics of ill-defined qualities: An experimental study' by George Udny Yule" allow="fullscreen" frameborder="0" width="100%" height="500px"></iframe>

Related Publications

Related Records

Hierarchy

This item is part of:

Related Fellows

Explore the collection

  • Referee Reports

    Date: 1832-1954

    This collection contains reports on scientific papers submitted for publication to the Royal Society. Started in 1832 when the system was formalised, it is a record of the origins of peer review publishing in practice.

    View collection