PART I
Recommended Guidelines of Smoking and Health Research
Introduction
In some markets there is much sensitivity to Smoking
and Health, in others comparatively little and in the latter
group it is sometimes assumed that smokers will continue to
demand cigarettes of comparatively high tar and nicotine
values and that such response to 'milds I as exists is to
'psychological' mi lds rather than to the reality of low
delivery, should it be offered.
However, in those markets of low sensitivity and not-
withstanding the current state of consumer opinion and need,
this assumption is unlikely to remain valid over time, partly
because of the likely influence on this subject from other
parts of the world (notably the U.S.A.), partly because
governments may be persuaded to follow the precedents of
other countries in imposing limitations on advertising (which
in itself tends to broaden the licence for unfavourable
press comment on smoking) and, not least, because the
exploitation of the issue can offer competitive advantages
to the Competition especially in markets where they, are
making little head-way. The launch of GALAXY in Brazil was
just such a stratagem.
in this case the Competitor, included league-table type
information in his publicity in the absence of there being
any official or unofficial demand for it. This underlines
the fact that it is not only open to Government to order the
publication of such data, or for manufacturers to volunteer
it for competitive advantage, but other sources may supply
it too. Consumer Associations and similar bodies may do so,
for instance. Influence on local 'sensitivity may also be
exercised through the publication of league-table material
in neighbouring countries or in distant countries of
importance, such as the U.S.A.
Thus, in terms of being prepared for these eventuali-
ties and to know in advance the sort of moves we should be
planning, we require to monitor current and developing local
sensitivity so as to have in readiness brand and publicity
propositions as will meet evolving needs. Indeed this
applies not only to markets of low sensitivity but also to
those in 'which a more substantial degree exists since
further changes may take place in :these too. NO
CN
1_-O
U-4
C_n
BATCo document for Province of British Columbia 16 April 1999

2
It will be evident therefore that we need to assess:
(i) Smoker sensitivity in the whole area of smoking and
health
(ii) Attitudes of non-smokers and ex-smokers, including
in their case:
attitudes to passive smoking - i.e. the
sidbstream effects of a smoker's product
which non-smokers around him may believe
is subjecting them to an involuntary
smoking experience.
(iii) Assess the perceived benefits of smoking as such, (as
opposed to the smoking of particular brands). Such
benefits might include, for instance:
- smoking as a regulator of mental and
physical equilibrium
- as a social lubricant
- as complementary to eating and drinking
- etc.
Thus a somewhat different set of reasons might be
advanced for the habit of smoking than would for
smoking particular brands which customarily list
such benefits as:
- satisfaction
- smoothnes's
- social prestige
- group emulation
etc., etc.
Accordingly, we set out below guidelines detailing the
principles of research which should be followed and which
will apply to the various subjects which may in some way be
connected with health sensitivity. In this present note we
deal with the general Monitoring Function that is proposed and
the arrangement of these notes is first to describe the
research objectives in general terms and then to cover in the
Appendixes detailed notes on specific aspects.
C=)
M.)
0%%
BATCo document for Province of British Columbia 16 April 1999

3 -
MONITORING FUNCTION
Two stages (in parallel, where appropriate) are recommended:
1. Qualitative Phase
There should be set in motion as soon as possible a
semi-continuous scheme of exploratory work whose purpose
will be to:
W Take the current temperature of the market with
respect to smoking and health concern. (This
would include some ex-smokers and non-smokers
as well as smokers)
(ii) Spot any indications of a growth in sensitivity
(iii) Identify the appearance of any new dimensions of
the problem.
'Semi Continuouslvl means with a frequency appropriate
to the market and to the research capabilities of the company
concerned. Thus, this might be-once a year in the case of a
Company with a limited res@earch capability operating in an
uncomplicated market whereas with a company with better faci-
lities in a more complex market a half-yearly or quarterly
monitoring might be appropriate. Thus the frequency will be
a matter for local management discretion.
In either event the person employed to perform it should
be skilled in qualitative work and have a capability for
judgement such as enjoys Company trust. He would also need
to be a person who would conduct such a study with sufficient
circumspection so as not to add to the sensitivity he is
investigating. Th is less of a problem in large markets
but needs particular care in small ones. The data acquired
by these studies -would be used to:
alert management to changing attitudes
and the nature and dimensions of these
atzi-tudes
provide a guide as to what questions should
be included in, or added to, a quantified
study so that numerical evidence can be
obtained for any hypotheses arising from
the exploratory stages (see Section 2)
to provide signpost information such as
wtil spark embryo ideas for brand or
advermising propositions or related strategy
NJ
(:N
%@-O
BATCo document for Province of British Columbia 16 April 1999

4
to relate the data to media coverage of
the subject (see Section 3 - Press
Clippings)
Note See Appendix 1. for principles to be followed,
topic-list, and points to consider in analysis.
Quantitative Research
One of the beat approaches to monitoring attitudes and
behaviour in a Smoking and Health context is to append a special
smoking and health section to a general consumer survey-. We cer-
tainly recommend that You obtain quantified data on the issue of
smoking and health on at least an annual basis. Wherever possible
you should try to avoid changing questions that you have used in
past years because retaining the same question allows you to compare
results over time i.e. to look at trends in your data to see whe-
ther there has been any change in either attitude or behaviour.
I-a order to help you in setting up quantitative research on
smoking and health and in order to enable us in Millbank to compare
results from different countries on this subject, we have prepared
a standard basic questionnaire for use on smoking and health
research. This questionnaire contains a basic core of questions
that need to be considered in all markets. We do not claim that
the questionnaire is complete and comprehensive in the sense that
there may be issues that we have not covered that you wish to con-
sider in your own market. One such area would be sources Of
smoking and health knowledge; another would be detailed infor=a-
tion about awareness and comprehension of league tables where they
have appeared; yet another would be characteristics of a brand
that make respondents feel that it is 'milder' or 'safer'. If you
are already engaged in quantitative research on this subject we
would ask you to consider, in spite of our earlier warning, the
possibility of modifying your questions so that they are the same
as ours or including our own questions along with your own im the
future. It is our intention to send out a questionnaire once a
year to each company so that we can have some feed back about the
state of smoking and health in your market. This request, as far
as consumer information is concerned, will be based upon this
international smoking and health questionnaire. A copy is appe.nded
to this document. Should you have any problems whatsoever in
applying it please contact your regional market research adviser.
In preparing this questionnaire we have made the assumption
that You will be looking at a sample of smokers, ex-smokers azid
non-smokers. Obviously in some markets the expense of such an
exercise would be prohibitive and in such cases you will need to
co
BATCo document for Province of British Columbia 16 April 1999

5
modify the questionnaire so that it only includes questions that
apply to smokers. Again the question of the sample size that
would be most appropriate must be determined locally.according to
your available resources.
Nocessarily, because this questionnaire has been prepared
as a single master questionnaire, you will need to do a fair amount
of thinking about how to apply it to your own local market. Such
simple things as question numbering must be considered carefully,
and this will have consequences for the various I skip I instructions
distributed throughout the questionnaire. Also the local language
problem will require particular care. For example, in Span.i.sh
speaking countries the English word 'tar' has a variety of possible
translations of which the two most common are lalquitran' and
I brea I. In such circumstances you must use your own judgement as
to which words should be used. In the case quoted of Spanish
speaking markets our recommendation would be that both words be
included along with any others of which there may be some Consumer
awareness. The best guide in such situations is your qualitative
research on smoking and health. It is most important when you can
analyse such work that you consider very carefully @.he language
used by respondents. This is probably your beat guide when trans-
lazing this questionnaire into local languages.
You are the first companies to be shown this questionnaire
and we are most anxious to ensure that it is a workable i=strument.
If you do have any problems whatsoever in implementing it please
lot us know immediately. We would also ask you to help us in
extending the questionnaire to new question areas. If you feel
that there are important issues about which we should collect inf or -
mation, then once again please tell us this as soon as possible
and we will make every attempt to incorporate these suSgestions
into future versions of the questionnaire.
Role of Desk Research
Sales and related data is of course already regularly
analysed for a variety of marketing purposes. What we
now recommend is that it should also be analysed from a
specific Smoking and Health vieurpoint: this to be perfor=ed
on a regular basis and a management report written on
each occasion highlighting the perceived trends or movements
which suggest tentative hypotheses of a health-related
kind. For instance:
(a) P. D. L. Data
Although these reports are studied (or at any rate
kept by Production people) it is still rare for them
to be studied on a continuous basis by marketing
people. We recommend that they shoixld be 30 studied
GIN
N.-C
BATCo document for Province of British Columbia 16 April 1999
Mr,11ONW1

6
regularly ocith a view to seeing whether:
Brands of given TPM I s, Total Nicotine
and Extractable nicotine characteristics
are progressing or regressing as classes.
There is any geographical pattern
associated with the particular ups and
downs of brands.
There is any other pattern deducible from
the figures in terms of the association
of particular levels of delivery with
particular types of consumers. E.g.
smoker-type, age, sex, occupation, class
etc.
There is an 'oddball' brand or brands,
whose PDL characteristics and associated
smoker profile and market progress places
it as variance in some way with the rest
o-- the market?
Whether there is a trend over time of
competition brands in PDL terms. Can one
infer on the part of a particular manu-
facturer a desire to move his brands
downwards for example?
Note The relationship between TPM, Total Nicotine
and Extractable figures can also imply an
increasing/decreasing interest in taste as
such. One might thus look for indications
of progress for brands which offer high
taste but relatively low delivery, low
taste/high delivery and other permutations
of these characteristics.
,low
Obviously some of the hypotheses that arise will
occasionally require field research to supply missing
bits of information or validation.
(b) Product
The performance of certain physical characteristics
can also be looked at from a specific health
orientation by grouping these characteristics into
classes (irrespective of the brands which may
comprise them). E.g.,
Those cigarettes having a particular tobacco
colour over a range dark to blonde. It may
be that blonde e.g. may signify a health/
purity association.
BATCo document for Province of British Columbia 16 April 1999

7
White tipping versus brown tipping
cigarettes
Pressure drops
Filter length in ratio to overall
cigarette length
Filter brands with rings (which some
may think indicate when to throw awav)
ve@sus cigarettes without them
Recessed versus normal filter (if
appropriate)
Carbon versus non-carbon
Any other 'special' filter classes
Heavily aromatic versus light aromatic
cigarettes
Slimmer than average versus average
circumference products
High versus medium versus low porosity
papers.
We know of course that performance under any of these
headings can be ascribable to factors other than a
health interest but the purpose of this form of analysis
is to ignore this and to look at the data from this
special health orientation to see if it is suggesting
any po'ssible trend or hypothesis which could have a
bearing on this subject.
Note Menthol versus non-menthol would be another obvious
division we should look at from this. perspective.
In this case we should be looking flor any indication
that menthols are progressive because of:
an interest in the taste
an interest in the health associations
of menthol
an interest in menthol purely because it
ameliorates what would otherwise be an
unpleasant smoke.
E.g. Up-market menthol progress would be likely to
indicate a health interest. Down-market menthol
progress might indicate an interest in ameliorants, N-)
and not health as such. Mi market menthol c7N
progress could be either. 1%n
BATCo document for Province of British Columbia 16 April 1999

(c) Packinzs
Groupings that suggest themselves here are dark
versus light Coloured packings. Again, is
there a geographical or other pattern evident
in the progress of packings of particular colours?
E.g. dark red packs might be progressive down.-
market where interest in health is often least.
Light blue packings might be progressive in up-
market locations where interest in health is likely
to be higher.
Are there any other design elements in the market
which could profitably be grouped in this way.
Bars, chevrons, symbols, etc? Again, it is reallsed
that market structuring will trammel the picture
but the data may still yield health associated
hypotheses if looked at in this special way.
(d) Advertisial and Promotion
Brands can be grouped according to how they are
advertised. E.g. (wherever appropriate):
- Those which the advertising explicitly
states to be mild.
- Those which the advertising implies to
be mild (through mood e.g.)
- Those which are unspecific on the matter
of mildness/strength
- Those which explicitly or through mood
indicate a strong or manly product.
(e) Press Clippings and Other Media Record
Clippi.n&s should be kept of all reports and editorial
treatments.of any subject bearing on the smoking and
health issue and a record kept of similar publicity
enamating from other media.
About once a quarter it can be briefly compressed
into a digest which should indicate not only the
range and substance of the material but also
whether the volume or intensity of it has increased
or decreased over previous quarters.
0N
%@o
BATCo document for Province of British Columbia 16 April 1999

9
Apart from its obvious inherent usefulness this
material mav also be correlated W,ith data real Is ed by
the exploratory research described in Section 1.
Thus a new belief, rumour, or heightening of concern
might be correlated with a particular burst of media
treatment of the subject.
Obviously, any indication of a health-associated
movement arising from one area of data would need to
be looked at with a view to possible correlation
with indications arising from other areas of data.
Further in these notes will be sections on how we see
these various aspects of the subject (i.e. product,
packaging and advertising) being dealt with in the consumer
research part of the programme.
General
In summary, it may well be that on reviewing the data
from all these sources there emerges one or two possible
situations one of which, at least, might riot be so evident
without doing this research.
1. There may be signs of a gradual evolution of health
concern such as is beginning to have its effect in all
manner of subtle ways on brand franchises in the perception
of advertising and in other reactions.
2. There may on the other hand be signs of a more
excitable situation in which because of some own or competi-
tive move or because of more vigorous anti-smoking publicity
consumer behaviour is showing some dramatic or erratic shift.
This may be of short-term effect or have longer term
consequences. A similar situation might arise, particularly
at lower levels of the market, if a rumour arose about a
particular brand whether founded or- unfounded.
These possibilizies are not exclusive and whatever
the nature of the dynamic that is operating it will normally
only become apparent if we relate together as many areas of
data as possible so as to come to reasonable hypotheses.
Thus, the purpose of the 'Monitoring Function' described
in the preceding sections is two fold:
to identify opportunities related to smoking
and health which the crompany may be able to
exploit
to identify threats related to smoking and
health which the company should take action NJ
to guard against. ON
BATCo document for Province of British Columbia 16 April 1999

10
Resulting from either or both of these there may well
be a need for new brands, or for the modification of some
of the company's existing brands.
It is essential that the brand objectives are clearly
spelled out in each case, and that the required posi- I
tionIng of the brand is specified both in product terms
(smoke deliveries etc.) and in consumer terms.
Of particular importance is the successful identifica-
tion of what is the optimum 'distance' of a new or modified
brand from other major brands or clusters of brands in the
market, in order to fulfill its proper longer-term function.
For example a new 'health reassurance' brand may, be
introduced which is located relatively close to the major
existing brands. This may very likely obtain more volume
in the short term, but if smoker concern with health matters
increases it may be that the brand is then too close to the
existing major brands and the consumer requirement is then
for a brand which is seen as more distant.
The second matter which is of particular importance
in the monitoring function is the identification of the
sorts of people who are most likely to be affected by the
various smoking and health pressures.
It is these people who should be regarded as the
primary targets, and on whom the next stage of research (on
brand and product development) should be concentrated.
If research on brands and products with a future
smoking and health role is carried out on samples of smokers
which include people who are less affected by such matters
the results can only be misleading.
The further sections of these guidelines continue
with the-detailed aspects of brand and product development as
related to smoking and health.
NJ
BATCo document for Province of British Columbia 16 April 1999

Apoendix 1.
Situations in which League Tables have not yet been published
A. Exploratory Research Notes
Attitude trends bearing on smoking and health
sensitivity may quite commonly begin at two levels:
(a) The sophisticated level - i.e. in places and among
people where there is high media exposure as well as an
awareness of what is happening in other countries, either
eml*_ through the readership of international as -well as informed
local publications or through travel, or both. Thus
there is at this level a greater tendency to be concerned
about environmental and social issues (air pollution,
traffic pollution, additives in food, and other health and
nuisance consequences of modern living) than at any other.
They are usually moderately affluent whereas poorer people
are too concerned -with the busiftess of making ends meet
to bother over much about these things.
(b) At the unsophisticated level - i.e. in places and
among people who are prone to believe in rumour unsubstan-
tiated by evidence. Once rumour begins it tends to spread
rapidly and there is no body of informed or educated
opinion at this level to correct it. Sometimes it takes
the form of believing that some products are made of 'immature'
tobacco. There are many other examples. It is important
when investigating this level of rumour not to add to it and
indeed once the main facts have been discovered from
respondents, to dispel it in their presence.
In the middle of the market where smokers are usually
concerned with getting the best quality they can at a
reasonable price they tend to receive their opinions one
remove behind those in the upper segment. Thus, in terms
of a structuring the research three groups from each
segment would be a reasonable number to interview comprising
no less than eight people in each. An alternative would
be to use a mini-group method of interviewing, in -which case
a slightly lar_zer number from each segment would be
recommended: say, 30 from each. Two-thirds in each case
should be- drawn from the younger. age groups ( -i8- 30) - In the
upper and middle segments it would be useful to have a small
number of females (say, five per segment).
Groups of similar composition should be interviewed rIQ
every three months.
BATCo document for Province of British Columbia 16 April 1999

12 -
As an extra source of information it would be valuable
to cover (say with a frequency of once or twice a year)
some groups of non-smokers (i.e. those 'who have zie@er . smoked)
and also of ex-smokers (those who once smoked but have
abandoned the habit within the last 12 months). One group
of the former and two of the latter should be adequate for
this purpose, their composition arranged as far as possible
along the lines suggested for the smoker groups.
In the case of both of these segments one would be
especially interested in their attitude to any feeling they
might have of being exposed to the effects of smokers
when in their presence (sides-cream effects) and whether
they exert pressure on smokers of any sort because of any
belief that they are being forced to become 'passive'
smokers themselves.
By the same token the views of smokers would be useful
on whether they are aware of this kind of effect on non-
smokers and whether they feel they are subject to any pressure
from them as a result of it.
A. Smokers
The following are not questions one would put to
respondents as such but questions on which the group leader
would like enlightenment. How the topic is actually
introduced to the groups depends on the atmosphere, what
may or may not have been spontaneously introduced by
respondents themselves, an appropriate sense of timing, etc.
To begin with, however, one would establish regular brand
usage, average smoked per day, knowledge of other brands on
the market, how own brand is seen to differ from others
in image and product terms, brands which might be considered
substitutes for regular brand, special occasion brands and
so on.
At this point one might lead in to the real purpose
of the study which is to assess smoking and health
sensitivity by the use of an oblique approach. In some
markets e.g. 'mildness' might be a euphemism for 'healthier'
so in these cases a statement like 11 should like to go into
this business of mildness and satisfaction a little more deeply'
and then going on to put the undermentioned topics to
respondents for discussion might be sufficient to induce them
to air any attitudes to smoking and health as they may have
together with any knowledge they have of the real or
supposed constituents of cigarettes, such as tar, nicotine,
saltpetre, stem, additives, flavourings, etc. and their
imagined effects.
ON
%lo
CrN
BATCo document for Province of British Columbia 16 April 1999

13 -
In the event that this 'mildness' approach does not
reveal any health sensitivity then one might directly ask
informants what constituents they believe go into the
making of cigarettes and what their attitude is to each
of these; what effects they imagine them to produce and
whether these are desirable or not. The aim is to see
whether there is any desire for mild products for their
own sake or because there is a belief that milder 'is
synonymous with' healthier for you. Obviously to get at
the truth without putting ideas into respondents' minds
which were not already there requires some subtlety in
the choice and wording of questions.
(a) Topics: (oblique approach, using 'mildness' as a lead)
What is actually understood by Imildness".
Is it likely to be thought healthier?
What products are considered mild?
Is a mild product likely to be thought
smoother?
Is it likely to involve a sacrifice in
taste?
(b) Topics: (more direct approach, exploring constituents,
mildness and health reassurance)
What constituents do respondents believe to
be present in cigarettes?
Any awareness of tar, nicotine, saltpetre,
stem, flavourings, additives, etc?
What effects do they perceive these to have
on the product; on themselves?
What difference do they detect in the various
brands in these terms?
Ar some more health-reassuring for the
presence or absence of these constituents.
Which brands? Which constituents?
Has anyone ever consciously switched to a
brand of more perceived mildness or health
reassurance than the one they customarily
smoked?
To what extent was advertising influential N.)
(details)? C71%
BATCo document for Province of British Columbia 16 April 1999
NNW

14 -
To what extent was the opinion of
friends influential (details)?
To what extent was a concern for health
influential?
To what extent was editorial coverage of
smoking in national or international
journals of influence (details)?
Is a perceived mild or health reassurance
prod-act less able to assuage smoking
need than a stronger product?
Is the sacrifice in satisfaction or in
other departments of smoking a worth-
while exchange for the mildness/health
reassurance perceived?
Whether interested or not in mild or in
health reassurance products, what view
do respondents take of people who smoke
cigarettes of obvious m:Lldnes,s or health
reassurance. What sort of people are
they who smoke these?
What is their feeling about filters.
What do filters principally do, in
their view?
Are there any brands of especially efficient
filtration. What are they? Any of low
efficiency. What are they?
How do they detect efficiency in these
cases. (Stains on the filter-end may
be volunteered).
Is filter length important? Ideas about
this.
What proportion of a total cigarettes'
length should be filter?
How do white versus brown tipping compare in
terms of suggesting mildness, strength,
purity, cleanliness, menthol. What sort
of people go for each?
Does tobacco colour have any influence in
this direction?
M.)
co
BATCo document for Province of British Columbia 16 April 1999

5
In general what bearing does length have on
how a cigarette is perceived in
mildness or in health reassurance terms.
E.g., regular, King Size, Superkings,
120 mm etc.
Is price associated with quality and thus
with health?
How about packaging. Can particular
pack colours suggest type of contents.
How do they group co I Ours and their
product associations?
How about names like suave, extra suave,
super suave - (in Latin America)
What sort of products do they imagine
these names apply to; how do they'cLiffer?
What sort of people go for them in their
vi ew?
What sort of products are the young seeking
these days - m:L I ds , machismo brands,
braggart brands, health reassurance brands?
What sort of products should they be smoking
in the view of the groups?
Is it good for them to smoke at all?
Why not, if not?
Do they believe smoking to be bad for one,
or a particular level of smoking bad
for one. Which level?
low- What ensues if the level is exceeded in
their view?
How-far has their reading, TV watching, radio
listening influenced them in these views?
How aware are they of smoking and health
publicity in neighbouring countries or in
more distant countries of importance, such
as the U.S.A. Any awareness and understanding
of League Tables. Is there any knowledge
of league-table publicity in other countries C:>
What effect does it have on local sensitivity, r-1)
BATCo document for Province of British Columbia 16 April 1999

i6 -
Has there been an increase in concern over
these matters?
Given that concern (if there is one) what
kind of cigarette products do they think
would now be best for their needs?
What are the beneficial effects of smoking
For example:
- social lubricant - i.e. offering and
being offered cigarettes
- helps to promote physical and mental
wellbeing
- helps one to concentrate
- provides a 'reward' for tasts performed
- completes the pleasure of eating and
drinking
- etc.
What are the harmful effects, if any. How
'do they identify these. What parts of
the product do they think produce these
effects. How could they be mimi-mised ideally?
Do they identify any particular brands
possessing these negatives?
How far does their own brand possess them?
How much pressure do they receive from
friends or family to give up or reduce
smoking?
Has it increased over time?
Is any of this pressure concerned with the
fact that they are causing their friends
to inadvertently smoke (through being
exposed to sidestream effect)?
How does it affect their own current view of
the matter?
If a cigarette high on taste but low on
nicotine were to be produced would this be
of interest?
Do they identify any such products at the
moment? cr%
BATCo document for Province of British Columbia 16 April 1999

17 -
Wh a are the benefits/negatives of menthols?
What sort of people smoke them?
Are respondents ever inclined to smoke them.
When and why?
How about American versus English versus
local cigarettes. How do these differ
in mildness taste, satisfaction and
health connotations?
Obviously there are many other related topics, many
of which may be spontaneously offered. There will also
be purely local market aspects worthy of introduction.
However there should be sufficient triggers among the above
topics to cover sensitivity generally.
Interpretation
The ad hoc nature of topic introduction does not permit
of a formal scheme of analysis. What is requixed is to look
for signposts denoting sensitivity from one area of discussion
and see whether these are confirmed from those arising
from other discussion areas.
One is looking for such things as:
Any informed knowledge of the matter such as may
influence behaviour and attitudes? E.g. knowlddge
of such things &3 tar and nicotine characteristics,
and how these apply to market brands.
Among those professing indifference to the subject,
any sign of sensitivity lurking behind the bland
reply?
Whether interest in milds and health reassurance
is more marked among particular social, geogra-
phical or cultural segments. Whether sensitivity
in general is more marked among any of these
segments.
Whether it is believed that the young are showing more
interest in 'safe' brands than older people.
Whether there is any concern about the cumulative
effects of smoking more satisfying brands, even
though continuing to express a preference for
them.
Interest in any 'oddball? brand - e.g. one wildly
different from normal in PDT" terms.
L-n
BATCo document for Province of British Columbia 16 April 1999

i8 -
Signs of occasional interest in menthols as
an alleviant to the effects of regular brand
smoking.
Usage of perceived milds and health reassurance
brands on an occasional basis.
Whether behaviour or attitudes show any sign
of being affected by what respondents have
read or heard that touches on smoking and health.
Any belief that certain pack colours denote
particular levels of delivery - e.g. blues may
denote mild cigarettes Reds may denote strong
brands. Greens will probably denote menthols.
Is there any sign that smokers are throwing away
longer butt, than before.
Any talk of some brands having I natux-al I tobacco
or 'purer' tobacco than others.
Is it common for smokers to look at the filters
to see to what degree they are stained.
Is there a daily consumption beyond which it
is thought dangerous to smoke.
Any belief that what happens in the U.S.A.
happens eventually in other countries.
Any interest in 'psychological' mi lds and reassurance
brands - i.e. those which have the appearance
without the reality.
Among which segments is interest in true milds
and health reassurance brands greatest?
How do smokers themselves see the future trend
irrespective of their own loyalties). Do they
envisage a growing interest in mild.s. Among 'whom
particularly?
E.g. A person might make various statements at
different times during the interview which would
enable one to come io a reasonable view as to
his real as opposed to his professed state of
concern. He might for example say:
1. He's happy with his current brand (a strong one)
and wouldn't change it.
Nj
2. Milder brands are tasteless for cissies .... C7N
for women.
BATCo document for Province of British Columbia 16 April 1999

9
Young men are the ones who are beginning
to smoke mi Ids; they are not real smokers.
No, he wouldn't smoke menthols. They're
for queers. Well, he might smoke one
occasionally but only when he gets a cough
from smoking too much of hi s regular brand.
Perhaps once or twice a month.
He doesn't read international magazines.
He vaguely recalls reading adverse comment
on smoking in local newspapers; not often
and it doesn't bother him. Yes, some people
are bothered by it. That iswhy they switch
to milder cigarettes.
6. He doesn't recall any advertising and claims
not to be influenced by it.
7. He throws away fairly long butts. He used
to smoke them further down than that.
8. The government, he thinks would like to ban
smoking, but they make too much money out
of it.
9. He wouldn't encourage his children to smoke.
10. He has tried twice to give up the habit,
for reasons of economy.
II. He wouldn't smoke a cigarette without a
filter. The filter has a 'purifying' effect.
12. He doesn't believe the acquafilter claimed
for the Extra-suave brand recently on the
market is any better than the once on his own
brand.
13. He wouldn't like his own brand to have a
white filter because it would show the
stains, which look unpleasant.
etc.
Obviously statements 4,7,9,10,11 and 13 belie his
professed unconcern about smoking. Statement 12 contradicts
statement 6 and 3,5,8 and 9 indicate awareness of the
social momentum affecting his habit.
He is probably susceptible to a 'psychological' offer.
In time might be susceptible to one which was truly CrN
milder than his current regular.
BATCo document for Province of British Columbia 16 April 1999

20
Topic Guide
B. Non-smokers and Ex-smokers
These respondents will include those who have never
smoked and those who once did but have since given it up.
-Many of the topics outlined in the interview-guide for
smokers will be appropriate for these respondents too and
those which would be irrelevant will be evident from
common-sense inspection. However there are some special
topics appropriate only to ex and non-smokers and these are
listed below.
Ex-Smokers
When was the habit abandoned altogether?
- What brand(s) were being smoked mainly before
abandor=ent
- What benefits were then derived from the habit?
- When was the habit first adopted. How long was
it maintained?
- What were the reasons for giving up smoking:
were these to do with media influence, social
pressure, family or religious pressure, own
convictions?
Were the reasons to do mainly -with smoking as
such or to do with the effects of particular
brands?
If brands, would respondents have continued
smoking had there then been a brand on the
market more suited to their taste - or would
they recommence if there were one now?
What is their attitude to smokers now. Is it
tolerant cr hostile?
/Oft-
Do they ever exert pressure an smokers to abandon
the habit?
Do they mind if people smoke in their presence?
Does this make them feel they are being forced
to smoke themselves - i.e. by exposure to the
sidestream of a smoker's product.
Any other negatives, e.g. of a social kind?
N. B. (i) An attempt should be made to obtain spontaneous
coff=ent rather than answers to direct questions. C:)
However some leading may be necessary in order to get NJ
a topic covered. Nevertheless one should try and ON
BATCo document for Province of British Columbia 16 April 1999

avoid putting into respondents' minds such
negatives as they might not have thought of
themselves.
(ii) It is often the case that ex-smokers
are much less tolerant of smokers than are people
who have never smoked and criticise with a zeal
similar to that of religious converts. This
should be borne in mind when receiving comment.
Non-Smokers
Any special reasons why they have never adopted
the habit?
Do any of their family smoke?
-hat benefits do they imagine they derive from
smoking?
What is their current attitude to smoking and to
smokers?
Has this attitude changed at all over time?
In what way and why. What influences caused the
change. E.g. media, social, personal convictions?
Do they exert pressure on smokers to abandon the
habit or not to smoke in their presence?
When people do smoke in their presence do they
feel any invasion of their rights - e.g. being
made unwillingly to smoke themselves by virtue
of the sidestream effect?
Any other objectionable aspeczs; e.g. hygienic,
dirty ashtrays, etc?
Cautionary statement (i) above, applies also to
non-smokers.
3. Situations in which I L@arue Table' Data has been Published
The previous notes apply particularly to those
markets where there has been no publication of league-table
type information either by, official or other bodies.
However, a situation may change and one may suddenly be
confronted with a publication of this kind in which case
one must be prepared to research its effects.
The guidelines already set out continue to apply but
with the addition of those extra topics requiring invest!-
Sation.
BATCo document for Province of British Columbia 16 April 1999

22
E. G.
Awareness of,league table publicity
Media source to 'which attributed
Understanding of the table itself -
this will be influenced by the 'way
it is presented; see note below
Which brands are no w seen to have
improved their images in smoking and
health terms and which have not?
Are some brands obtaining an unjustified
benefit or damage as a result of
positioning misinterpretation (perhaps
due to alphabetical arrangement; see
note below)?
Has the consumer changed brands as a
result of the league table publication
or is he intending to and why?
Is he now making greater attempts to
reduce his smoking or give it up
altogether?
Note
Consumer perception can be influenced by the way
tables are presented. Several forms of presentation are
possible. For instance:
Tables may be published which show brands in
descending or ascending order of tar readings.
or of nicotine readings. A frequent effect of
this is not necessarily for the brand with the
lowest readings to benefit (especially if it
is too much at variance with market taste) but
for other brands ranked close to it in position
order only. Thus, the brand placed mext might
be quite tonsiderably higher in delivery but
because it is ranked close to the bottom of
the scale obtains a halo effect from positioning.
Another method of presentation is to take a number
of constituents such as tar, nicotine, carbon
monoxide, etc. , and index these according to
some formula or other. Except for those who
understand formulae indexing can have the effect
of obscuring the importance of particular
elements (e.g. nicotine). Thus it might be found C:)
that a brand enjoying increased popularity NJ
UN
%IC
X-
---J
Un
ON
BATCo document for Province of British Columbia 16 April 1999

2 3
because it has a low index number m7 Zht still
be delivering a fair amount of satisfaction.
In this case the effect of the other elements
might be responsible for it having a low
number.
The grouping of brands into high, medium and
low groups according to their tar/rLicatine
readings is also popular. Where there are a
lot of brands and where within one 'band' there
are a number with similar readings the compiler
of the table may group alphabetically. Thus
the lowest in any 'band' may simply be the one
with the lowest alphabet letter and not
necessarily with the lowest readings. Consumers
often miss this and simply take positioning
itself as being the most important.
Some tables are presented by giving low delivery
brands one dot, medium delivery two dots and
high delivery brands three dots. Again the
consequences of alphabetical arrangement within
these bands should be looked for.
Thus one should keep in mind the possible influence of
positioning on consumer perception when undertaking research
into a market where league tables are published.
N.)
Cr%
BATCo document for Province of British Columbia 16 April 1999

24 Appendix 2.
OTM;MTICNAL S%40KZNZ AND HEALTH QUESTICNNAIn
coal,
Prepared and distributed by:
Market Research Ceoartnent,
rIJ
B.A.T. Co. Ltd., @kstminster Hcuse,
7 Millbank, Eondon Swl
April 1979 Co
BATCo document for Province of British Columbia 16 April 1999

25
Pace
SECTION 1. 1
QUkLIFICATICN
SECTION 2. 2
CURRENT MAWFAC-MM CIGAREMTE
SMCKERS
SECTION 3. 4
RY-0 CMRENT SMOKEFS
SECTION 4. 4
EX-5%,1.CKERS OF MANUFAL'MMED
C I G AP-;'=
SECTIM 5. 5
SMOK'6-@ BLNEFI7S
SECTION 6. 7
S&MOKZ.\C' AND hTALTH
SECTICN 7. 10
ATTI'ITMES 10 DE 9A.CKING FLASIT
C71%
LP
11@0
BATCo document for Province of British Columbia 16 April 1999

SECTICN 1. CUALIFICATICN
1(a) "Do ycu wmke any of these tobacco products nowadays?
(b)Have you ever smoked any (other) of these tobacco product regularly?
Q.la - s-,mked Q.lb - ever
nowadays wwked
manufactured cigarettes 1 1
RYO cigarettes 2 2
cigars 3 3
(Add any other tobacco products
of ir=rtance in your market).
A.CURRENT SiOKERS CP MANUFACTURED CIGARETTES AT Q.la
SECTICN 2.
B.CMMENT 9MMS OF RYO CIGARETTES AT Q.la
SECTION 3.
C.IF RESPONDENT SACKES WM MkWA=,,YRM = RYO AT Q.la,
cCmP= BM SECTION 2. AND SECTION 3.
D.EX--SmOKEpS OF LMANUFACTURED CIGARETT CES ( i.e. MANUFACTUMED
CIGARETTES AT Q.lb BUT NOT 0.1a)
SECTION 4
E. ALL CIMER FESPCM&'NTS SECTION S.
C=)
r-j
(ON
01%
BATCo document for Province of British Columbia 16 April 1999

2.
S=CN 2. CLTRW KANUFACr=
CIGARETIIS SNOZERS AT Q. la CNLY
(a) How long ago did you first start smoking manufactured cigarettes
Less than one month
one month but less than three months
Three months but less that six months
S ix months Wt less than cne year
one year but less than eighteen months
Eighteen months but less than two years
Two years but less than three years
Three years but less than five years
Five years but less than ten years
2 (b) Sow many manufactured cigarettes do you sake an average each day?
Less than 1
1
2
3
4
5
6-7
8-9
10
11-12
13-15
16-19
20
21-25
26-30
31-40
41-50
51-60
'w
more than 60
(c) Which (one) brand of manufactured cigarettes do you smoke most often?
R=RD EXACT BRAND AND METHER FILTER TIPPED CR PLAIN EZM.
(BRAND LIST)
BATCo document for Province of British Columbia 16 April 1999

(d) How long has (BRAND MENTICNED AT Q. 2c)
been your regular brand?
Less than one month
One month but less that three months
Three months bit less than six months
Six months but less than one year
One year but less than eighteen months
Eighteen months but less than two years
Two years but less than three years
Three years but less than five years
Five years but less than ten years
Ten years or more
2 (e) What did you smoke instead of (BRAND ZMENTICNED AT Q. 2c) before it
became your regular brand?
1) Did not smoke at all SKIP TO Q. 2q
/01*1 2) Smoked another brand of
manufactured cigarettes
(ESMMISH EXACT BRAND MC
AND MiETHER FILTER TIP CR
PLA.LN M)
Write in:
Filter
Plain
3) Smoked RYO cigarettes
4) Smoked a pipe
5) Smoked c4gars
6) Any other answer
Write in:
...........................
...........................
(f) Why did you &,ange from (P-rCMCT) (BRAND M=ICNED AT Q. 2e) to
(BRAND 'A AT Q. 2c) ?
................................................
................
PROBE: were there any other reasons why you changed
to (BRAND MENTICNED AT Q. 2c) ?
................................ I
................................................
BATCo document for Province of British Columbia 16 April 1999

SZ;OW CARD X
Have you changed your smcking habits in any of these @eys in the last
year?
... SKIP M NE.Vr SECTI01 1
CARD X Yes: Changed to filter tipped cigarette2
Smoked less cigarettes 3
Tried to give up smoking cigarettes....4
Changed to a milder brand 5
Changed to a brand with less tar6
Changed to a brand with less nicotine..7
Inhaled less 8
Left a longer butt at end of cigarettes
(h) Why have you changed your smking habits in the last year?
Are there any other reasons?
SECTION 3. RYO CURRENT SMOKERS
AT 0. la ONLY
(Section contains 0.3a - 3h exactly as Q.2a - h except Eor appropriate
wording changes).
IF RESz-r.N'D&'qT USE 10 SAM MAINUFACTUMM CIGAP0ITES
AT Q. lb SECTION 4. OTHERWISE SECTICN S.
SECTION 4. EX-SMOKERS OF
IMANUFAC'711@ CIGAIIL=,.-S ONLY
(Establish from Q-lb)
4(a) How long ago did @vu stoo Banking manufactured cigarettes
(LIST AS AT Q.2d)
(b) How many manufactured cigarettes did you sake on average per day just
before you stopped srmking them? (LIST AS AT Q.2b) ON
(c) %hic-h brand of manufactured cigarettes did you smoke most often just -C@-
before you stopped making t!,em?
c7N
RECMD ECCr SPAND AND FILTER TIP CR Pr-A-r.-I em.
BATCo document for Province of British Columbia 16 April 1999

4(d) W-hy did you stop smoking manufactured cigarettes?
..............................................
.................................................
PROBE: Were there any other reasons?
............................................
..........
SE=ICN 5 - ASK EVEMB=
5(a) 'Mhat benef its or enjoyments; do you think smokers get from cigarette
smoking?
RECORD ALL H=CNED LN FIRST COUM OF
(b) SHOW CAM x
"Are there any other benef its or enjoyments on this card that you think
smokers get from cigarettes?
RECORD ALL M=ICNED -rlq SECOM COLUM OF
NCN-SMOKERS OF MANLTACTURED OR FM AT Q.1a GO TO
SECTION 6.
SMOFMRS OF &V.NUFAZX= OR FM AT O- la ASK O. Sc
AND Sa.
(c) '"Ahat benef its or enjoyme-nts do you personally derive L-= smoking
cigarettes?
RECORD ALL MENTIONED IN 7AMD COLUM OF GUD
(d) "Which one benefit or enjoyment is most bruportant to ycu personally as
a cigarette smker?
RECORD CNE ANS'vqER IN FWRTH COLQ'M OF GR.:D
Cr1%
BATCo document for Province of BritiSh Columbia 16 April 1999

3 1
Sa Q. so Q. 5c Q. 5d
Spontaneous Prompted Personal Most in=
Benefits Benefit
Relieves strain or tension/
calms nerves/helps 4,.en I have 1 1 1 1
a problem.
Selps when relaxing 2 2 2 2
Relieves boredom /it's company 3 3 3 3
Nice after food/meal 4 4 4 4
Nice with a drink 5 5 5 5
Gives me en4oyment 6 6 6 6
J
It's satisfying 7 7 7 7
It's comforting 8 8 8 8
Helps me concentrate 9 9 9 9
it stimulates me/wakes me up 0 0 0 0
Like taste/taste is enjoyable V V V V
Good with =manv-/makes me
feel more sociable X X X X
Something to do with my. hands I 1 1 1
Stops ea oeight down 2 2 2 2
Looks good/makes me look more
C
.Jroun up 3 3 3 3
Habit 4 4 4 4
Ncne/No benefits 5 5 5 5
Any others (write in):
.................................
/001*1
N-0
BATCo document for Province of British Columbia 16 April 1999

3 2.
SEMCN 6. ASK EVEIMMMY
6 (a) Soffe people say that --,=king is bad for health ard others say it is
,'I C e e
not. How far do you personally agree with the statement t2at -igar tt
s.,mking is harmful to @ealth?
Agree Strongly 4
Agree 3
SKIP M - --(Disagree 2
SECTION 7 Disagree strongly 1
,:@Neither agree nor disagre-e/don't
know/no opinion 0
(b) In wbat @ays do you think cigarette zwking is harmful to health?
...................................................
...................................................
PROBE: Are there any particular illnesses or disease that
you associate with cigarette smoking'?
...................................................
PROBE: Any others?
....................................................
....................................................
(c)(1) Are there any particular substances in cigarettes that you think of as
harmful to he-Alth?
PROBE: Any others?
RECORD ALL MUMCNED M F= COLLM OF TABLE BEIICW.
SHCW CARD X
Which of the substance Mentioned on this card do you think are
contained in cigarettes?
RECORD ALL MDMIONED LN SECOND COLUM OF MU BELGW.
CrN
N-0
CN
BATCo, document for Province of British Columbia 16 April 1999

3 3
(111) %hich one substance of those that ycu have mentioned as being in
cigarettes do -you think is most harmful to health?
RECCRD AINSIER IN VIRD C0t0IN CF MBLE BELJOK.
Q.6cl Q.6clI Q.6cllI
CARD X Spontaneous Aided Most harmful
Tobacco
Flavourinc
Tar
Nicotine
opium/Chandu
Added Chemicals
Carbon Monoxide
Licorice
Fluoride
Hydrogen Sulphide
Igasp
Icloves'
Icinnamcn'
'sugar'
Artificial Tcbacco Material
Goudron
Kondensat
Teer
Isaltpetre'
Ibrea'
lalcuitran'
valcuir-raol
residuos'
the paper %hen it burns'
'Yin Yao'
one of them
All of them
Don't @c-.ow
Any others (write in)
......................................................
............................... 6......................
......................................................
NB: (items marked * should be included in all Questionnaires Others
should be included according to local requirements).
NZICN-TMOKERS SKIP M SE-=ICN 7.
%10
BATCo document for Province of BritiSh Columbia 16 April 1999

3@L -
(d) How concerned are ycu about the rossible effects that cigarette smoking
may have on ycur own hea.1 th?
READ CUr SLUDUNTS AND ASK RESPCNDENT M CECOSE CNE.
(OR CSE CARD IF REMMENT CAN READ)
,Verry concerned
Quite concerned
Slightly concerned
Hardly concerned at all
Not at all concerned
C71%
co
BATCo document for Province of British Columbia 16 April 1999

- 35 -
SEMM 7. ASK EVERYBODY
7 (a) I am go ing to read cut some statements to vou. AS I read cut each, one
will you tell me how much you personally agree or disagree with it.
You may say:
I agree strongly
I agree
I disagree
I disagree strongly
(USE CARD IF RESPCNDENT CAN READ)
READ CUT EACH MTEMENT IN WM.
Don't Know
Agree Agree, Disagree Disagree Don't Care/
Strongly Strongly No answer
Most of my friends smoke
Non-smkers make too much
fuss about sirr-
The decision as wtether to
take up smoking or not should
be a matter of choice for
every adult.
Smoking should be forbidden
in -more public places.
I think that there are
probably more non-smokers than
smokers among adults in
(YCUR ccu-my).
Smoking. is a dirty and
unpleasant habit.
Cigarette sake can be harmful
to the health of non-smkers
in the same room as the
smoker.
NCN-SMKERS OD INTEERVIMI
SMOKERS (MANUFACMMM AND RYO) CCNrINEE
C=;@
r-0
0-
BATCo document for Province of British Columbia 16 April 1999

36 -
7 (b) ASK SMOKERS (MANUFACTLMM AND RYO) CNLY
Don't Khow/
Agree Agree Disagree Disagree Don't Care/
Strongly Strongly No answer
I sometimes refrain from 4 3 2 1 0
smking because I don't
want to offend nonsmokers.
During the last year people 4 3 2 1 0
have been trying to persuade
me to give up waking More
than thev have ever done
,e@ before.
END LVr-vZVIEW
BATCo document for Province of British Columbia 16 April 1999

37 -
;zLr..OLkRY
There is a need for companies to monitor smoking and health
sensitivity in their markets especially in those with low current
concern but where the situation may change over time. Companies
may thus be forewaz-ned of any changes in attitude and be better able
to develop correct brand and publicity propositions. These notes
are guidelines on the research principles to be followed and the
principal information to be obtained covers:
Smoker attitudes to smoking as such, the perceived
benefits of smoking and the degree/extent of smoking
and health concern,
the incidence of gains and losses to the market or
to our brands as a result of smoking and health matters,
attitudes of non-smokers and ex-smokers to the smoking
habit generally and to passive smoking.
Three sorts of research are called for:
Qualitative Research
A quarterly investigation into the attitudes of smokers, non-
smokers and ex-smokers to take the temperature of the market and
identify any new awareness or new dimension of the smoking and
health issue. This would be small scale work employing group dis-
cussions or mini-groups.
Quantitative Research
All companies should conduct quantitative research at least
once a year to monitor the state of smoking and health concern in
their local markets. Wherever possible this should be done by
using the appended international smoking and health questionnaire
modified to meet your local circumstances and extend it in ways
that meet your lacal needs. Translation of this questionnaire into
your local language should be done by referring to the language
used by respondents in qualitative studies. If you have any prob-
lems in applying this questionnaire to your local market refer to
your regional market research adviser in Xi Ilb ank.
Desk-Research
The re-organi3atiOn of data already available to companie 21
and the scrutiny of this from a strictly, smoking and health view-
point. This would embrace PDL, product, packing, advertising and
press-clipping information so as to see if there are any hidden 1,0
relationships.
BATCo document for Province of British Columbia 16 April 1999

38 -
A regular digest of this qualitative, quantitative and
desk-research data is recommended together with any theories
arising therefrom. By this means we can provide a barometer of
opinion on the subject over time and infer the best positions to
adopt.
Appendix 1.
This gives the researcher technical information on how to
carry out the qualitative phase, suggests what he should look for
and gives a list of topics which should be discussed with smokers,
non-smokers and ex-smokers, and an interpretation a-aide.
too*, Avoendix 2.
This is a copy of the inter-national smoking and health ques-
tionnaire recommended by Marketing -5erv-ices Department of Mil lbank.
BATCo document for Province of British Columbia 16 April 1999