Trade Practices Commission v Arnotts Ltd
[1989] FCA 645
•26 OCTOBER 1989
Re: TRADE PRACTICES COMMISSION
And: ARNOTTS LIMITED; ARNOTT'S BISCUITS LIMITED; FLEDSPAC PTY LIMITED and
THE DICKENS CORPORATION PTY LIMITED
No. G1316 of 1988
FED No. 645
Evidence
COURT
IN THE FEDERAL COURT OF AUSTRALIA
NEW SOUTH WALES DISTRICT REGISTRY
GENERAL DIVISION
Beaumont J.(1)
CATCHWORDS
Evidence - reference of questions to court expert - whether purported findings constitute a "report" for the purposes of 0.34 r.2(1).
HEARING
SYDNEY
#DATE 26:10:1989
Counsel and Solicitors Mr. B.C. Oslington Q.C.,
for Applicant: Mr. D. Staelhi and
Mr. M.R.J. Ellicott instructed by the Australian Government Solicitor
Counsel and Solicitors for Mr. C.A. Sweeney Q.C. and
the first and second Mr. P. Comans instructed by
Respondents: Clayton Utz
ORDER
Rule that the document MFI "50" be admitted into evidence as exhibit 138.
Note: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.
JUDGE1
On 14 September 1989, an order ("the order") was made, by consent, appointing Eden Brinkley as court expert for the purposes of 0.34 r.2 By r.2(1), where a question for an expert witness arises, the Court may -
"(a) appoint an expert as court expert to inquire into and report upon the question;
(b) authorize the court expert to inquire into and report upon any facts relevant to his inquiry and report on the question;
(c) direct the court expert to make a further or supplemental report or inquiry and report; and
(d) give such instructions as the Court thinks fit relating to any inquiry or report of the court expert."
Under the terms of the order, the question which Mr. Brinkley was to inquire into and report upon was stated in a letter to him written jointly by the solicitors for the Trade Practices Commission and by the solicitors for Arnotts relevantly as follows:
"As you may be aware, the parties to the litigation described above have agreed that a 'court expert' should be appointed, to report to the Federal Court about certain survey material prepared by the Roy Morgan Research Centre Pty. Limited ('Roy Morgan'). You have been nominated by the Australian Bureau of Statistics to undertake the above role and we understand you are willing to undertake that role but we would be grateful if you would confirm that this is the case. It has been agreed that the applicant and the first and second respondents in these proceedings shall be jointly liable for the reasonable costs of the Australians Bureau of Statistics in having you carry out the role of Court expert.
The relevant questions were included in one of Roy Morgan's omnibus surveys and about 1,200 people were interviewed. Roy Morgan carried out a statistical analysis of the results, using a computer, and Mrs. Michele Levine produced a report concerning the results of the survey.
We propose to forward to Roy Morgan for your use a copy of part of MFI 27 (being "ML1") and exhibits 118, 119, 120, 121, 122, 123, 124 and 126 which consist of a 'survey on biscuits', the records made by interviewers, their summary forms, instructions to them and letters returned by interviewees.
The relevant questions are numbered 21a to 25c and are found at the back of the annexure to MFI 27 and you are asked to make your report as soon as practicable in relation to these questions and the responses to them. You are requested to:-
1. Provide a report on the question whether the computer analysis ML1:
(a) provides a sound basis;
(b) is based on proper and professional survey practices;
for drawing conclusions as to the views of the population of Australia on questions 21(a) to 25c set out in the survey questionnaire.
2. Comment on such of the following issues as are relevant to the above questions:
(i) whether the appropriate universe was examined;
(ii) whether a representative sample was drawn from the universe;
(iii) the expertise of those who conducted the survey, including the experience of the interviewers;
(iv) whether the questionnaire, sample design and interviewing were in accordance with generally accepted standards of objective procedure and statistics in the field of such surveys;
(v) the methods by which the universe and sample were selected, and of the techniques employed for selecting and instructing the interviewers.
3. Record the reasons, and a statement of the facts, upon which your report is based. You are free to approach officers of Roy Morgan and other employees of Roy Morgan to inform yourself of the facts necessary to make your report."
Pursuant to 0.34 r.3(1), under cover of a letter dated 19 October 1989, Mr. Brinkley has forwarded to the Registrar, and to the parties, a copy of his report ("the report"). Arnotts now seek to tender the report. The Commission objects to the tender.
Arnott's tender is made pursuant to 0.34 r.3(3) which is as follows:
"(3) The report shall, unless the Court otherwise orders, be admissible in evidence on the question on which it is made, but shall not be binding on any party except to the extent to which that party agrees to be bound by it."
It is not suggested that any party has agreed to be bound by any part of the report. The present question is one of its admissibility only.
It is common ground that the questions referred to Mr. Brinkley were matters that could properly be put to an expert (see Minnesota Mining and Manufacturing Company v. Beiersdorf (Australia) Limited (1980) 144 CLR 253 at pp 268-270). The Commission's objection is that the document tendered is not a "report" within the meaning of the rules because it does not deal with the questions referred to Mr. Brinkley but, rather, is a report dealing with different matters. Thus, it is said, the document proposed does not satisfy the description of a "report" found in 0.34 r.2(1) because the document addressed the wrong questions.
The report is entitled "Evaluation of Computer Analysis ML1". It consists of 47 pages of text and seven appendices. The "Table of Contents" indicates the scope of the documents as follows:
"1 INTRODUCTION
1.1 Background
1.2 Aims of this Report
2 AIMS AND OBJECTIVES
2.1 Defining Aims and Information Requirements 2.2 Aims and Information Requirements for the RMRC Survey
3 DATA COLLECTION METHODS
3.1 Collection Methods
3.2 Collection Method for the RMRC Survey 4 QUESTIONNAIRE DESIGN
4.1 Introduction
4.2 An Introduction to Questionnaire Design 4.3 Questionnaire Design for the RMRC Survey 5 SAMPLE DESIGN
5.1 Introduction
5.2 An Introduction to Sampling Theory 5.3 Sample Designs in Social Research 5.4 Sample Design for the RMRC COT Survey 6 ERRORS IN STATISTICAL DATA
6.1 Sampling Error
6.2 Non-Sampling Error
6.3 Errors in the RMRC COT Survey 7 SURVEY MANAGEMENT
7.1 Introduction
7.2 Recruitment and Training
7.3 Survey Management for the RMRC COT Survey 8 DATA PROCESSING
8.1 Introduction
8.2 Stages in Data Processing
8.3 Data Processing for the RMRC COT Survey 9 PRESENTATION AND ANALYSIS OF RESULTS 9.1 Presentation of Results
9.2 Tabulation and Analysis of Results 9.3 Results for the RMRC COT Survey 10 CONCLUSION
APPENDICES
1. Questionnaire for COT 984
2. Method for Selecting Startpoints
3. Interviewer Instructions
4. Tables of Sampling Variation for ML1 Analysis
5. Revised Table A for ML1 Analysis
6. Revised Tables B and C for ML1 Analysis
7. RMRC Research Consultants"
In his introductory remarks, Mr. Brinkley states the aims of his report as follows:
"1.2 Aims of this Report
The report aims to address the questions raised in the facsimile, forwarded to me on 14 September 1989. In particular, to provide sufficient information to answer the question 'whether the computer analysis ML1: a. provides a sound basis;
b. is based on proper and professional survey practices; for drawing conclusions as to the views of the population of Australia on questions 21a to 25c as set out in the survey questionnaire.' Before such a question can be answered, it is necessary to put the survey conducted by RMRC in context. In the sections which follow, each of the main areas of survey design are introduced, and the terms, methods, and principles relevant to each area discussed. It is hoped that by presenting the report in this way, the reader can obtain an overall picture of the range of survey methods available to the researcher and that this can form a basis for assessing the RMRC survey. My findings in regard to the RMRC survey are provided following the general introduction to each section." (Emphasis added)
Having dealt with items 2 to 9 inclusive in the Table of Contents, Mr. Brinkley states his conclusion in these terms:
"10 CONCLUSION
In preparing this report, I have attempted to adopt a broad perspective when answering the questions given in the brief. In the absence of details on how the data is to be used and how it is being analysed, I have found it difficult to summarize my findings and draw a clear conclusion as to whether the ML1 analysis: 'provides a sound basis, and is based on proper and professional survey practices'. However, I am able to draw the following somewhat qualified conclusion. In the preceding sections I have raised a number of concerns in regard to questions 21a to 25c of COT survey 984. While these concerns give me reservations about the conclusions which can be drawn from the results, it will ultimately be necessary for the parties in this case to assess whether the questions are acceptable given the hypothesis under test. It is hoped that the comments I have made in relation to the questionnaire form a sufficient basis for the decision. Should the parties decide that the questions are acceptable, then the following conclusions apply: i) If finely disaggregated data is required, or if statistical tests are being applied to the results in the ML1 analysis, then I have reservations about the reliability of the sampling procedures used and about the manner in which the data has been aggregated to form population estimates.
ii) If a broad qualitative perspective is required to assess an overall trend in consumption, then I expect the analysis does provide a valid indication of the behaviour of the Australian population. However, I doubt whether the accuracy of this trend could be assessed in a statistically defensible manner. I hope this report has been able to clarify the matters of concern to the court. If the soundness of the COT survey is still unclear, a number of options for further work exist. Some form of sensitivity analysis applied to the results may help. With a sensitivity analysis, the results could be adjusted by an arbitrary 10% or 20%, or perhaps by the level of the 'sampling' error. The statistical anmalysis could then be repeated to see if the adjusted data gives the same conclusions. While such an option can lend support to a particular argument, there must be general agreement on the assumptions that are used. For instance, the parties must agree on a likely upper bound for the error in the survey results, an upper bound which is at best subjective.
An expensive alternative might be to repeat the survey using a questionnaire and sampling procedures acceptable to both parties. Obviously this alternative is expensive, and it may delay the findings of this case by many months."
It is true, as Mr. Brinkley says, that he can offer a "somewhat qualified conclusion" only. But, in my opinion, it does not follow that Mr. Brinkley's findings, albeit rather tentative, should not be characterised, for the purposes of O.34 r.2(1), as a report upon the questions referred to him.
The present question arises at the adjectival level only. That is to say, at the level of the admissibility of the material, the substance or merits of the findings do not arise. Those matters, that is, the merits, indeed the whole of the expert's reasoning process, are available for scrutiny at a later stage of the proceedings, with the benefit of any cross-examination of the expert which may take place pursuant to r.4 of O.34.
Put differently, in my view, it is sufficient for the purposes of O.34 r.2(1) that the findings made are a genuine attempt to address the questions referred. Mr. Brinkley purports to respond to the questions asked. It is a genuine attempt to resolve the issues raised. This is enough to satisfy the description of a "report" within the meaning of the rules. The accuracy, or otherwise, of the response is not a relevant consideration for present purposes. It is sufficient that the findings attempt, or seek to deal with the matters raised in the terms of reference. As Mr. Brinkley said (para. 1.2, cited), his report "aims to address the questions raised . . ." It follows, in my view, that the document tendered qualifies as a "report" within the meaning of O.34 r.3(1).
An alternative submission was put on behalf of the Commission that, if the document were to be admitted, there should be excluded from evidence the last three paragraphs of para.10. I reject this submission also. The paragraphs in question are part of the reasoning process used by Mr. Brinkley in producing his report. As such, they are at least adjectivally relevant.
I rule that the document MFI "50" be admitted into evidence as exhibit 138.
55
1
0