Day 257 - 06 Jun 96 - Page 15

     
     1
     2   Q.   But are the 2,000 people the same each quarter?
     3        A.  No, it is 2,000 different people.
     4
     5   Q.   Right.  This survey was done by telephone; is that right?
     6        A.  This data would be based upon the Taylor Nelson
     7        database, which is actually derived from interviews within
     8        home.
     9
    10   Q.   Last time you gave evidence you did say that both the
    11        surveys had been done by telephone?
    12        A.  I may be mistaken in that.  On further checking of the
    13        data I think I can confirm that Taylor Nelson is interview
    14        within home and that fast track is telephone interview, so
    15        AF1 is, in fact, interview within home and AF2 is, in fact,
    16        telephone interview.
    17
    18   Q.   AF2 is fast track, and that is by telephone?
    19        A.  Yes, that is right.
    20
    Q.   Right.
    22
    23   MR. MORRIS:  Was that, the 6,000 people, the telephone, the fast
    24        track telephone?
    25        A.  No, Taylor Nelson is the larger 60,000 sample and that
    26        would be a home interviewing sample.
    27
    28   Q.   Right.  How many people were interviewed by phone in
    29        compiling these figures that you used in your first?
    30        A.  The first document that starts 'Frequency of visit' is
    31        the one that is derived from telephone interview.
    32
    33   Q.   Was it the sample?
    34        A.  The sample would be 5,000 on a yearly basis.
    35
    36   Q.   5,000?
    37        A.  Yes.
    38
    39   MR. JUSTICE BELL:  Can I just make sure I have not made an
    40        obvious blunder?  The figures, on the face of it, it may be
         just the method you used to work them out, appear to be
    42        very different, because on the 'frequency of visit' sheets
    43        you have several times a week 4 per cent?
    44        A.  Yes.
    45
    46   Q.   And if one then moves from that survey, or that analysis,
    47        to the, what I call, document 1, which is the number of
    48        people eating at McDonald's et cetera, 4 per cent of the
    49        21,735,000 eating at McDonald's is nearly 870,000, 869,400,
    50        which is very considerably more than the 385,000 you have 
    51        come out with on this analysis? 
    52        A.  Yes. 
    53
    54   Q.   Is there an explanation for this?
    55        A.  Yes.  Let me clarify the reason for the two sources of
    56        information.  On your previous request to try and quantify
    57        the numbers of people by visit occasion, which we were not
    58        satisfied that the AF2 or the fast track sample gave us a
    59        robust enough sample on about 5,000 people to give an
    60        answer that we would be happy returning to the courts, so