An affidavit was filed by Mr MJ Collins. He is the managing director of a company which inter alia advises
franchisers and franchisees on all aspects of franchising. He served as Vice Chairman, and later as Chairman, of the
South African Franchising Association from 1983 to 1992. He has addressed numerous meetings, conferences and
seminars on
View Parallel Citation
various aspects of franchising. During such addresses he has on numerous occasions held up the business format
adopted by McDonald's as the model for efficient franchising. During seminars
Page 20 of [1996] 4 All SA 1 (A)
he was questioned about McDonald's and its business system. Moreover, since become Vice Chairman of the South
African Franchise Association, he says, he has received "numerous requests, too numerous even to have counted"
from prospective franchisees and ordinary members of the public for advice as to how to become a McDonald's
franchisee. He also notes that the South African press regards McDonald's as newsworthy. Thus coverage was for
instance given when McDonald's opened its first outlets in Russia and continental China. Objection was taken to the
admissibility of Mr Collins's affidavit, but at least the items of evidence mentioned above must be admissible. I
discuss their weight later.
Finally there was evidence about two market surveys. A great deal of argument was addressed to us about the
admissibility and weight of such evidence. Before going into such matters it is necessary first to set out the nature
and content of the evidence.
Mr CK Corder is the Chairman and Managing Director of Market Research Africa (Pty) Ltd. It is not disputed that
he is an expert on market researching. In his affidavit he sets out first the theoretical principles involved. The basic
theory of market research is that from a given representative sample of the consumer public it is possible to project,
by means of acceptable mathematical methods, results of such sampling to a general population or "universe"
within certain statistical limits. In other words, the researcher first determines the class of persons (or universe)
which is sought to be tested, and then questions individuals from that universe. The confidence that one has in a
projection from the samples to the universe varies according to the number of persons interviewed in the survey,
the sampling technique used and the level of response. Statisticians commonly use the term "inference" to denote
the process of generalising sample evidence to the universe from which the sample is selected. The basis of
inferential statistics is probability theory. A theorem in probability theory, namely the Central Limit Theorem, which
applies in this case, states that if a large number of independent samples are drawn from a universe, the individual
results will be different, but that they will have a "normal" distribution around the real value being measured in the
universe. Based on this theorem, and standard and accepted statistical tables in relation to normal distribution, one
can calculate the probability of the real value in the universe being within a specific interval or variation, which Mr
Corder referred to as the confidence level.
During September 1993 Mr Corder was instructed to conduct a market survey on behalf of McDonald's. He was
informed that the objectives of the study were to establish awareness of the name McDonald's, to measure
recognition of the McDonald's trade marks, to ascertain the association of McDonald's with certain products or types
of business undertakings, and to establish the awareness of McDonald's hamburgers. The method used by him was
the conducting of personal interviews using a structured questionnaire and interviewing aids. The interviewing aids
consisted of two text show cards and one colour picture
View Parallel Citation
show card featuring the main McDonald's trade marks. Copies of the questionnaires and show cards were before
the court.
The universe for the survey was defined as white adult males and females, aged 16 years and over, living in
houses in higher income suburbs of Pretoria, Verwoerdburg, Johannesburg, Bedfordview, Randburg and Sandton. A
sample of 202 persons was taken. Mr Corder gave details about the manner in which the sample was selected. I
need not repeat them there is no suggestion that the sampling was not scientifically correct. The fieldwork was
conducted from 7 December to 24 December 1993 by trained interviewers under the supervision
Page 21 of [1996] 4 All SA 1 (A)
of field supervisors. Twentyone percent of the interviewers were back checked in order to ensure reliability.
Affidavits of supervisors and interviewers were filed to confirm their actions. Mr Corder, who was in overall control,
also confirmed that the survey was properly conducted.
The relevant conclusions were set out as follows:
"A large majority of respondents were aware of the name McDonald's, and/or the McDonald's logos/trademarks (77%). More
than half had heard of both McDonald's, and knew the logos/trademarks too (57%).
...
Most respondents spontaneously associated McDonald's with hamburgers, or knew of 'McDonald's Hamburgers' (80%).
...