Book contents
- Frontmatter
- Contents
- Preface
- List of contributors
- 1 Statistical consultancy
- 2 Consultants' cameos: a chapter of encounters
- 3 Straight consulting
- 4 A two-period crossover trial
- 5 Consultancy in a medical school, illustrated by a clinical trial for treatment of primary biliary cirrhosis
- 6 The analysis of response latencies
- 7 Acid rain and tree roots: an analysis of an experiment
- 8 On identifying yeasts and related problems
- 9 Uneven sex ratios in the light-brown apple moth: a problem in outlier allocation
- 10 Collaboration between university and industry
- 11 Inspection for faulty components before or after assembly of manufactured items
- 12 Statistical modelling of the EEC Labour Force Survey: a project history
- Bibliography on statistical consulting
- Name index
- Subject index
10 - Collaboration between university and industry
Published online by Cambridge University Press: 05 June 2012
- Frontmatter
- Contents
- Preface
- List of contributors
- 1 Statistical consultancy
- 2 Consultants' cameos: a chapter of encounters
- 3 Straight consulting
- 4 A two-period crossover trial
- 5 Consultancy in a medical school, illustrated by a clinical trial for treatment of primary biliary cirrhosis
- 6 The analysis of response latencies
- 7 Acid rain and tree roots: an analysis of an experiment
- 8 On identifying yeasts and related problems
- 9 Uneven sex ratios in the light-brown apple moth: a problem in outlier allocation
- 10 Collaboration between university and industry
- 11 Inspection for faulty components before or after assembly of manufactured items
- 12 Statistical modelling of the EEC Labour Force Survey: a project history
- Bibliography on statistical consulting
- Name index
- Subject index
Summary
Introduction
In this paper we describe two of the several different ways in which statisticians in a university can act as consultants for industry. In both cases the consulting is effectively carried out by more than one statistician, but there the similarity ends. We hope that these examples will provide a flavour of the activities of an applied statistics department and an applied statistics research unit working together within a university. A number of problems are considered, including: the analysis of dominant lethal assay data; the analysis of quantal assay data incorporating time to response; the analysis of pain data relating to episiotomy; the analysis of aggression in mentally handicapped patients.
The analysis of dominant lethal assay data
Tables 1 and 2 present two sets of data from the paper by Haseman and Soares (1976). In each case, for over 500 litters of mice, the number of dead fetuses was recorded. Tables 1 and 2 are control groups from dominant lethal assays (taken from Haseman and Soares, 1976). In this experiment a drug's ability to cause damage to reproductive genetic material, sufficient to kill the fertilised egg or developing embryo, is tested by dosing a male mouse (typically) and mating it to one or more females. A significant increase in fetal deaths is indicative of a mutagenic effect.
As in many areas of statistics, typically two questions arise:
Can we describe such sets of data in a relatively simple manner?
How might we make comparisons between such data sets?
- Type
- Chapter
- Information
- The Statistical Consultant in Action , pp. 134 - 152Publisher: Cambridge University PressPrint publication year: 1987