CICAREVIE SMOKING, If EALTH AND DISSONANCE:
(PROJEH-CfBRiT III: PSYCHOGRAPHIC DATA
AND FURTHER TOP-LINE ANALYSIS
REPORT NO. RD.1743-C
20.5.1980
ISSUED BY: D.J. Wood
DISTRIBUTION:
Dr. L.C.F. Blackman
Dr. I.W. Hughes
Dr. R.A. Sanford
R.M. Gibb, Esq.
R.S. Wade, Esq.
R.G. Nicholls, Esq.
Herr E. Rittershaus
Dr. F. Seehofer
Dr. C.J.P. de Siqueira
Mr. A.J. Kruszynski
Dr. D.G. Felton
Library
AUTHOR: M. Oldman
Copv No. 1, 2, 3, 4
5
6,7
8
9, 10, 11
12, 13
14
15
16
17
18
19, 20
PROG. REF.: 12.01.000
COPY NO.:
C)
~_n
L-n
BATCo document for Province of BritiSh Columbia 29
October 1999


Group Research & Development Centre,
British-American Tobacco Co. Ltd.,
SOUTHAMPTON.
MO/CAL/46J 20th May 1980
CIGARETTE SMOKING, HEALTH AND DISSONANCE.
(PROJECT LIBRA) III: PSYCHOGRAPHIC DATA AND
FURTHER TOP-LINE ANALYSIS
(Report No. RD.1743-C)
SUM11ARY AND CONCLUSIONS
This report concludes the presentation of the
basic data recorded in
Project LIBR.A. Attention is focussed on the
various psycho.graphic items
presenLed- in the questionnaire and it is shown
that, in respect of several
indices of individual difference, the responses
are consistent with the
behavioural and attitudinal characteristics of the
subject groups defined
previously.
Further top-line categories of respondent are
identified and the
data inspected with regard to the following:
smoking and health concern
amongst smokers, health concern amongst smokers
and never-smokers, never-
smokers and other people smoking, and smokers
under pressure to quit
smoking.
The results confirm the general utility of the
consonance/dissonance
dimension but also indicate that further
differentiation within the
consonant/dissonant categories is likely to yield
a more valuable
segmentation of the smoker population.
c::>
Xt.
BATCo document for Province of BritiSh Columbia 29
October 1999


INTRODUCTION
In an earlier report (1) it was suggested that an
individual's
attitudes and behaviour regarding his own health
could he expected to be
reflected in various indices of individual
difference. It will be recalled
that the Project LIBRA questionnaire incorporated
items for the measurement
of six parameters of individual difference:
locus of control
neuroticism
anxiety
ego strength
social desirability
authoritarianism.
PrL-sentation of these data was excluded from the
report dealing with
the top-line analysis (2) but is undertaken here
together with inspection
of some further pertinent top-line categories.
For smokers in the LIBRA sample it was decided to
distinguish,
amongst those who believe that smoking is harmful
to health, between
those concerned about their own health and those
who are not concerned
about their own health. It is also possible to
compare and contrast the
responses given by "health conqcious" smokers with
those given by "health
conscious" never-smokers, and to distinpuish,
amongst never-smokers,
between those concerned for their own health from
other people smoking
and those not expressing this concern. Those
smokers under pressure to
quit smoking are also identified and their
distinguishing characteristics
described.
C__
\-C
4 z_-
BATCo document for Province of BritiSh Columbia 29
October 1999


This report, therefore, comprises the following
sections:
1. Psychographic data.
2. Smoking and health concern amongst smokers.
3. Health concern amongst smokers and
never-smokers.
4. Never-smokers and other people smoking.
5. Smokers under pressure to quit smoking.
As with the previous report of the top-line
analysis (2), the
opportunity is taken to present the fullest
description of the data
obtained. Integration of these data, together with
a more sophisticated
analysis will be reported on later occasions.
1. PSYCHOGRAPHIC DATA
For the purposes of tabulating the results to be
discussed in this
section, most of the original primary top-line
categories ha~-C- been
retained. No distinction is made at this stage,
however, between recent
and established ex-smokers. The major categories
are sho;,,m in Table 1.
CD
NJ
BATCo document for Province of BritiSh Columbia 29
October 1999


TABLE I
PRIMARY TOP-LINE CATEGORIES-
Description
SMOKERS OF PACKETED CIGARETTES
Degree of Consonance/Dissonance
Highly consonant - have not tried to give up
smoking
cigarettes, nor seriously want to.
Consonant - have tried to give up smoking
cigarettes, but do not seriously
want to.
Dissonant - have not tried to give up sTroking
cigarettes, but seriously want to.
Highly dissonant - have tr1ed to give up smoking
cigarettes, and seriously want to.
EX-SMOKERS OF PACKFTED CIGARETTES
NEVER-S~fOKERS OF PACKETED CIGARETTES
Abbreviation
SM0
HIC
C 0
DIS
MID
EXS
NEV
In the proceeding analysis, the smoker sub-groups,
ex-smoker and
never-smoker samples are differentiated by sex.
Reference to Table 2
will indicate the distributions by sex within the
major categories.
BATCo document for Province of BritiSh Columbia 29
October 1999


-5-
TABLE 2
COMPOSITION OF PRIMARY CATECORIES BY SEX
Male Fema I e
N (7.) 14 (7)
SMOKERS
HIC (154) 69 44.8 85 55.2
CON (213) 117 54.9 96 45.1
DIS (56) 22 39.3 34 60.7
HID (320) 145 45.3 175 54.7
EX-SMOKERS (508) 309 65.0 178 35.0
NEVER-SMOVERS (740) 265 35.8 475 64.2
It is interesting to note that within the smoker
category, only the
consonant group has a preponderance of men. The
distributions by sex for
the ex-s-~noker and the ne~er-smoker categories
were set by quota.
(a) L-ccus of Control
The dimension of locus of control (LoC) recognises
that some people
have an internal locus of control (i.e. they
perceive themselves as having
considEz-able influence In determining the course
of their lives) while
others are more externally (i.e. fatalistically)
controlled. This
psychological concept is related to smoking in
tt~7o principal ways.
Firstly, people with an internal LoC are less
likely to be cigarette
smokers. Secondly, people with an external LoC who
in fact do smoke are
less likely than those people with an internal LoC
to successfully stop
smokinz.
As measured by Rotter's scale (3), LoC has
appeared to be a promising
variable, both in terms of distinguishing between
smokers and non-smoler!~,
r\j
C7 \
BATCo document for Province of BritiSh Columbia 29
October 1999


and in predicted success in efforts to quit
smoking. Investigations by
James et al. (4) and Straits and Sechrest (5)
showed smokers to be
significantly more fatalistic than non-smokers.
James et al. also found
that male ex-smokers were significantly more
internal than other males,
and Hjelle and Clouser (6) found that external
females were more likely
to be smokers. Other studies have related
internality to the successful
reduction of smoking rates (7, 8). However,
several studies have vielded
contrary results (9, 10, 11, 12, 13 and 14).
It has been suggested that the failure to
consistently find a
significant relationship between LoC and smoking
hehaviour is due Co
inadequacies of the LoC scale in measuring
self-control (15). flo-,;cver,
since Rotter's scale has been found to Tneasure
several factors rather
than a unidimensional trait (16, 17 and 18), the
discrepancies a~onp_
previous findings may be due to the existence of a
relationship bet,,een
smoking and only some of the factors of the LoC
scale, with variance on
the other factors occasionally functioning to
obscure the relationship.
It has generally been assumed that smokers score
externally on this
scale because of their presumed lack of
"will-power" or self control
(evidenced by their inability to quit). Indeed, a
s tudy by Mlott and
Mlott (19) showed that individuals who stopped
smoking were more internally
controlled than those who smoked or never smoked.
At a gross level, the data from the present
research would appear to
support this latter conclusion. Although no
estimates of statistical
significance are made at the present time, the
mean LoC scores for smokers
(12.95), ex-smokers (12.86) and never smokers
(12.93) are tending to
accordance with Mott and Mlott's conclusions.
BATCo document for Province of BritiSh Columbia 29
October 1999


-7-
From other researches mentioned above, It might
also be expected
that male ex-smokers were more internal than other
males. At the present
level of analysis this hypothesis is only
partially supported. From
Table 3, it can be seen that male ex-smokers are
more internal only in
relation to the two consonant groups (HIC, CON).
Those expressing the
wish to quit smoking (the highly dissonant and
dissonant groups) are
relatively more internal than the ex-smoker rales
group.
TABLE 3
MEAN LOCUS OF CONTROL SCORES
SMOKERS
HIC
CON
DIS
HID
EX-SHOKERS
NEVER-SMOKERS
All Male Female
12.95
12.75 12.86
13.051 13.65
12.321 12.59
12.5-5 13.09
12.86 12.69! 13.19
12.93 12.91 1 12.94
It would seem reasonable to assign the greatest
relative externality
to the consonant smoker since he/she has attempted
to quit in the past
but no longer expresses the wish to do so. For
both male and female
smokers the highest mean LoC score (i.e. greatest
externality) is, in
fact, obtained by the consonant category.
The lowest mean scores were recorded in respect of
the dissonant
group. This is somewhat surprising in view of
their earlier
characterisation (2) as indulgent smokers who,
more than any other
CX:)
BATCo document for Province of BritiSh Columbia 29
October 1999


-8-
sub-group, admit to smoking out of habit and
because cigarettes are being
handed around.
After studying smokers' perceptions of the dangers
of smoking, Foss (20)
concluded that many smokers do not believe that
smoking is dangerous
enough to be a cause of death. In other words,
many people may start
smoking, or not try to quit once they have
started, because they believe
they have no reason to avoid the hahit. This
suggests that nianv smokers
have not internalized the belief that smoking is a
life-threatening
behaviour. Of the present qa-ple of 746 smokers,
346 (i.e. 467) expressed
the belief that smoking is harmful to health. Of
these, 66% were very/qxiite
concerned for their own health and the remainder
were not very/not at all
concerned for their own health. For these two
groups the mean LoC scores
were 12.95.and 12.93 re-,portively. Whilst this
differencc is unlikely to
be statistically significant, the direction of the
difference supports
Foss' view that the more fatalistic a smoker is,
despite acknowledging
the health risk attached to smoking, the less
likely he is to express
concern for his own health.
(b) Neuroticism
The items used to measure neuroticism were those
developed by
Eysenck (21). The results obtained in the present
study are sho-..7n in
Table 4.
CZD
\~c
\0
BATCo document for Province of BritiSh Columbia 29
October 1999


-9-
TABLE 4
KEAN NEUROTICISM SCORES
All Male Female
SMOKERS 3.58
HIC 1 3.12 3.31
CON 3.14 3.74
DIS 3.18 3.67
HID 3.48 4.20
EX-SMOURS' 3.14 2.86 3.64
NEVER-SMOKERS 3.42 3.10 3.59
Inspection confirms the established finding (22)
that women score
consistently higher than men on this parameter.
The previous finding by Cherry and Kiernan (23),
that non-smokers
are less neurotic than smokers is supported, for
men, by the present
data. Amongst women, highly consonant smokers
appear to be the most
stable.
Within the smoker category it is quite consistent
with expectation
to find the highly dissonant smoker, of either
sex, revealed as the most
neurotic. It will be recalled that the highly
dissonant smoker is the
one most likely to think about his physical
condition a lot, often feels
tense and is least likely to be a very relaxed
person.
This profile is entirely consistent with the
relatively high mean
neuroticism scores obtained by this group which
would also predict the
relatively higher usage of tranquillizers and
hypnotic drugs previously
recorded for this group.
BATCo document for Province of British Columbia 29
October 1999


(c) Anxiety
The Taylor-Spence Manifest Anxiety Scale (24),
from which the items
used here were taken, is known to correlate highly
with Eysenck's dimension
of neuroticism. Accordingly, it is not surprising
to find some similarities
between the anxiety scores recorded (Table 5) and
the neuroricism scores
above.
TABLE-5
t IE A.NXIETY SCORES
All Male Female
SMOKERS 9.90
HIC 8.86 10.29
CON 9.06 10.68
DIS 10.64 10.36
HID 9.65 10.26
EX-SMOKERS 9.48 9.12 10.15
NEVER-SMOKERS 9.49 8.83 9.85
Whether male or female, never-smokers exhibit the
lowest mean level
of anxiety and, for females, ex-smokers score
lower than smokers. This
pattern is not maintained by male respondents,
however, where the highly
consonant and consonant smokers achieve the lowest
levels aniongst smokers
and ex-smokers.
The high mean score for the dissonant male smoker
group cannot be
readily explained. Closer insoection of the
original data, however,
reveals that in respect of one item ("I have
trouble getting to sleep")
the dissonant group scored disproportionately
higher than respondents
Ur
BATCo document for Province of BritiSh Columbia 29
October 1999


in the other categories. Yet they are the least
likely group to resort
to the use of Mogadon, a common hypnotic used to
induce sleep in insomniacs.
(d) Ego Strength
The construct of self-esteem, or ego strength, has
been used by many
people in diverse ways. Nevertheless, despite its
popularity, no standard
theoretical or operational definition of the
concept exists.
Ego strength is defined liere simply ne, liking
and ru~:pect for or),.F-.elf
which has some realistic basis. The Thomas-Zander
ego strength scale (25),
part of which was employed here, conceptualizes
ego strength in two parts.
First is a person's ability to be self-directing
and to translate intentions
consistently into behaviour. Second is the ability
to control and
discharge tension withour disrupting other
psychological processes.
The results in respect of the ego strength
mcasures taken in the
present study are presented in Table 6.
TA.B LE 6
MEA.N EGO STRENGTH SCORES
All Male Female
iSMOKERS 20.53
HIC 22.17 20.80
CON 21.50 18.97
DIS 21.36 20.49
RID Zl.Zb 19.30
EX-SMOKERS 21.11 21.87 19.72
NEVER-SMOKERS 20.71 21.85 20.07
\1 C
N)
BATCo document for Province of BritiSh Columbia 29
October 1999


-12-
Beyond noting the apparent but slight progression
in mean ego strength
scores from highly consonant to highly dissonant
in male smokers, little
can be inferred from these data. It might have
been expected, on the basis
of the definition of ego strength offered above,
that ex-srokers and
never-smokers would have achieved higher mean
scores than smokers. There
is, however, only the merest support for the
hypothesis that ex-smokers
will have the higher mean ego strength scores when
all ex-smokers are
compared with all smokers and all never-smokers.
(e) Social Desirability
The tendency for some questionnaire respondents to
"fake good" is
well kno,.m. It is particularly important in the
present study to attempt
to measure and control an), such tendency. It was
suggested earlier (1)
that soine check on the truthfullness of responses
would be required,
particularly in relation to the other
psychographic data. Also, the
actual prevalence of "faking good" within and
between the various smoking/
non-smoking groups was thought to be intrinsically
interesting and of
direct relevance to an understanding of the modes
of conflict resolution
being investigated. The items used were taken from
Eysenck and E senck (22).
y
The overall mean levels recorded in respect of all
smokers (2.81),
ex-smokers (2.87) and never-smokers (2.99) are
unremarkable. Within the
smoker category, for both sexes, it is interesting
to note (Table 7) that
the respondents who have never attempted to quit
(RIC, DIS) score higher
than those who have attempted to quit (CON, HID).
BATCo document for Province of British Columbia 29
October 1999


TABLE 7
MEAN SOCIAL DESIRABILITY
All Male Fema I e
SMOKERS 2.81
HIC 2.97 3.10
CON 2.68 2.62
DIS 3.32 3.21
HID 2.79 2.70
EX-SMOURS 2.87 2.94 2.73
NEVER-SMOURS 2.99 2.84 3.07
(f) Authoritarianism
Th& ricasure of authoritarianism adopted in the
present study was the
4-ite= F-scale (26). The higher an individual
scores on this test, the
higher are his inferred attributes of
conservatism, conventionalism,
superstition, intolerance of ambiguity, and
feelings of threat and
insecurity. In the study by Mlott and. Mlott (19)
previously cited, a
fuller but similar index was used to test the
hypothesis that individuals
who have stopped smoking are less dogmatic, less
rigid and less close-
minded than those who have never smoked or who
continue to do so. Their
results indicated that there was a significant
interaction between sex and
smoker in relation to his or her authoritarianism
with all the males
scoring higher than the females. Both the male
smoker and ex-smoker were
equally high on authoritarianism while the
never-smoker group reflected
the lowest score. The females showed an opposite
trend to the males.
BATCO document for Province of British Columbia 29
October 1999


-14-
The data obtained here partially support Mlott and
Mlott's original
hypothesis. From Table 8 it can be seen that, for
females, authoritarianism
is lowest amongst ex-smokers when compared with
all smokers and never-
smokers.
TABLE 8
MEAN AUTHORITARIANISM SCORES
All Male Female
SMOKERS 14.34
HIC 14.40 14.35
CON 14.46 14.13
DIS 15.19 14.84
Hit) 14.22 14.24
EX-SMOKERS 14.06 14.25 13.70
NEVER-SMOKERS 13.72 13.23 13.99
For nale respondents, ex-smokers achieved a lower
mean score than
current smokers but not as low as never-smokers.
For both male and female
groups, the highest scores were associated with
the dissonant group.
SU~24ARY
Compared with all ex-smokers and never-smokers,
current _srokers tend
to score
- higher on authoritarianism
- higher on neuroticism
- higher on anxiety
- lower on ego strength
- lower on social desirability
and are the most fatalistic.
N-)
BATCo document for Province of BritiSh Columbia 29
October 1999


-15-
Compared with all smokers and never-smokers,
ex-s.okers tend to score
- lower on neuroticism
- higher on ego strength
and are the least fatalistic.
Compared with all smokers and ex-smokers,
never-smokers tend to score
- lower on authoritarianism
- higher on social desirability.
Within the smoker category, and relative to the
other smoker sitb-
groups, the hizhly-consonant smoker tends to score
lower on neuroticism
lower on anxiety (males)
higher on ego strength.
Relative to the oth(2r smoker suh-proups, the
consonant smoker tends
to score
- lower on authoritarianism (females)
- higher on anxiety (females)
- lower on ego strength (females)
- lower on social desirability
and is the most fatalistic.
Relative to the other smoker sub-groups, the
dissonant st-..ol:er tends
to score
- higher on authoritarianism
- higher on anxiety (males)
- higher on social desirability
and is the least fatalistic.
BATCO document for Province of British Columbia 29
October 1999


-16-
Relative to the other smoker sub-groups, the
highly dissonant smoker
tends to score
- lower on authoritarianism (males)
- higher on neuroticism
- lower on ego strength (males).
DISCUSSION
Although presented here as univariate comparative
data, the responseE:
to the psychographic items given are intended for
more rigorous application
in later stages of the LIBRA analysis.
Nevertheless it is interesting to
note that the direction of the gross differences
observed between the
smoker azid non-smoker cateporic- are In 11iie
with previous research
findings.
The s-moker's greater reliance on "external"
rather than "internal"
controls noted by Smith (27) is supported here.
Similarly, the smoker's
relatively less "strength of character" in
comparison with non-smokers (28)
is reflected in the lower mean ego strength scores
recorded in the present
study in respect of the smoker group. Various
researches have also
reported the higher anxiety level found amongst
smokers (29, 30, 31).
Whilst less comparative data for ex-smokers exists
in the literature, the
finding here thatj of the three primary groups,
they are the least neurotic
and have the highest ego-strength is quite
consistent with the expected
profile of the individual who has stopped smoking.
That they are also
the least fatalistic suggests appropriately that
they perceive themselves
as having greater personal control over their
behaviour.
When the smoker group is differentiated according
to degree of
consonance/dissonance, a number of apparently
valid differences in the
11 C
Qill
BATCo document for Province of BritiSh Columbia 29
October 1999


psychographic data emerge, particularly in respect
of dissimilarities
between the highly consonant and highly dissonant
groups. That members
of the former category exhibit lower levels of
neuroticism and anxiety
reinforces the belief that they relatively lack
the constitutional
disposition to be concerned about their smoking.
Their higher scores on
the ego strength measure suggests that they
possess a firm self-image,
itself consistent with their attitudes to
sTroking. In conLrast the hipher
mean neuroticism scores recorded for the highly
dissonant smokers provide
an appropriate substrate for their wish to quit,
with the lower ego
strength scores suggesting a good reason for their
failure to do so.
The consonant smoker's emergence as the most
fatalistic is consistent
with his current wish not to quit despite having
attempted to do so in
Lite past. - of the smok~~rs, the dissonant group
revealed the most "internal"
score on the locus of control T-easure. Noting
that ex-smokers were, more
than any other category, the least fatalistic
suggests that some members
of the dissonant group are indeed likely to become
successful quitters.
Clearly, inspection of these data at a raw mean
level allows only
the crudest indications of difference to be
inferred. Whilst these
inferences support a view of the major categories
which is apparently
consistent with the various behavio.rs and
attitudes previously noted (2),
more sophisticated analysis will be required
before a comprehensive
understanding of smoker and non-smoker types can
be gained.
2. SMOKING AN"D HEALTH C014CERN AMONGST SMOKERS
All smokers in the sample were asked to indicate
their agreement/
disagreement with the statement "cigarette smoking
is harmful to health"
and to express the level of concern in relation to
smoking and their own
cc
BATCo document for Province of British Columbia 29
October 1999


health ("How concerned are you about the possible
effects YOUR smoking
has on YOUR OWN HEALTH?"). of the total smoker
sample of 746, 212 were
identified who agreed strongly/agreed with the
(first) statement and were
very/quite concerned in response to the (second)
question and 134 were
found who agreed with the statement but who
claimed to be not very/not at
all concerned about the effects of their smoking
on their o-.n health.
For the purposes of this discussion, the first
careVt-ry (28% of all
smokers) will be designated as CONGRUENT smokers
and the second (187 of
all smokers), INCONGRUENT smokers. In this report,
the inz:re overt
differences between these two groups are presented
with t1he intention of
better understanding the apparent mismatch in
attitudes re-Flected by the
"incongruent" group of smokers.
Considering first the psychographic data, it
appears that the
incor.2ruent smoker is slightly more fatalistic
than the congruent smoker
but has considerably lower levels of manifest
anxiety (mean score = 9.42
versus 10.20) and neuroticism (3.35 versus 4.08).
He tends to be more
authoritarian (14.51 versus 13.71) and has a
slightly higher ego strength
(20.48 versus 20.06).
His tendency to "fake good" is also higher (3.04
vers-,:s 2.47). in
many ways this "profile" is similar to that of the
highly consonant
smoker. It should be noted, however, that only
70/. of the highly consonant
group are not very/not at all concerned about the
possible effects their
cigarette smoking has on their health. The
equivalent percentages for
the consonant, dissonant and highly dissonant
groups are 68, 36 and 23
respectively. It would appear, therefore, that the
incongrient smoker group
draws its membership from all consonant/dissonant
categories, albeit with
an emphasis on the consonant and highly consonant
sectors.
BATCO document for Province of BritiSh Columbia 29
October 1999


-19-
The responses to the battery of questions related
to general attitudes
to health are recorded in Table 9. For comparison,
the percentages of
all smokers, highly consonant, consonant,
dissonant and highly dissonant
smokers giving the same response (i.e. agree
strongly/agree) are also shown.
TABLE 9
PERCENTAGE R.ESPONDENTS AGREEI14G STRONGLY /AGREE
ING
Item SMO HIC CON DIS HID Conguent Inconguent
I consider myself to be a very
healthy person 67 75 68 59 63 61 70
I eat more than I should 40 32 43 34 45 48 45
1 am in very good physical condition 53 62 58 42
48 45 58
I exercise regularly 39 43 39 30 41 43 42
I eat more sweets than I should 25 27 23 18 27 30
21
I am careful to eat a balanced diet 40 42 41 52 38
39 39
I
h nk about m ph ysIcal. condition
I y
a lot 47 35 45 48 54 61 37
T here Is too much fuss about what
one should and should not eat 67 68 69 68 66 58 73
I am overweight 35 26 37 39 39 37 35
I get out of breath easily 34 26 34 30 39 40 31
I often feel tense 46 34 45 39 54 54 46
I am very health conscious 31 35 24 31 34 36 29
I am very weight conscious 34 31 27 41 39 45 22
1 am generally a very relaxed person 63 72 71 57
54 50 67
The congruent/ incongruent disti nction ext ends
the discrimination
between smokers beyond the consonanc e/dis sonan
ce dimension in a number of
CZ-
Qj
Ur
11 C
BATCo document for Province of British Columbia 29
October 1999


-20-
instances. For example, more than any other group
the congruent smoker
admits to
eating more than he should;
eating more sweets than he should;
thinking about his physical condition a lot;
getting out of breath easily;
being very health conscious;
being very weight conscious.
They are also less likely to perceive themselves
as very relaxed and
are less likely to agree that there is too much
fuss about what one should
and should not eat.
In contrast, the inco~lgiuent smoker is, more than
any other category
of smoki-r, likely to agree that there is too much
fuss about what one
should and should not eat. He is also
distinguished as being the type of
smoker least likely to be very weight conscious.
It would appear that in
relation to attitudes to general health, the
differentiation based upon
congruence/ incongruence further separates the
smoker sample in a number
of ways.
Inspection of the data relating to drug, coffee
and tea consumption
revealed no remarkable differences, although it
was evident that the
incongruent smoker admits to a higher average
daily alcohol consumption
than the congruent individual.
Turning to smoking behaviour, the incongruent
smoker has been smoking
for an average of 23.24 years compared with 19.24
years in the case of
the congruent smoker. This latter figure is below
the average for all
smokers (22.86 vears) and is lower than for any of
the consonant /dissonant
(ii
11 C
011
BATCo document for Province of BritiSh Columbia 29
October 1999


groups. Thirty-four percent of congruent smokers
have/had parents who
both smoked cigarettes compared with 25% in the
case of incongruent
smokers. Interestingly, the average daily
consumption for the congruent
group is higher than that of the incongruent group
(19.84 versus 18.68
per day) and only 10% of the former group,
compared with 16% of the
latter, claim to smoke less than ten cigarettes
per day. More congruent
smokers than incongruent smokers currently smoke
low tar brands (23.17,
versus 18.7%). Otherwise there are no major
differentiating features
within the brand usage data. Congruent smokers
were more likely than
incongruent smokers to mention health related
reasons for changing their
current most often brand (44% versus 35%).
The percentages of respondents agreeing
strongly/agreeing with the
smoker motivation items are presented in Table 10.
Again, for comparison,
the consonance/dissonance categories are also
included.
01 \
N-)
BATCo document for Province of BritiSh Columbia 29
October 1999


TABLE 10
PERCENTAGE RESPONDENTS AGREEING STRONGLYAGREEING
Item SMO HIC CON DIS HID Congruent Incongruent
There are t!--es when I light up a
cigarette just out of habit 77 67 71 84 83 85 72
I smoke when relaxing 88 94 88 75 86 87 82
I smoke in order to make a break
in tasks 47 45 47 38 49 49 44
I smoke when feeling bored 67 57 61 61 76 75 60
1 smoke as a reward after a hard
or boring task 57 59 57 47 57 57 55
I smoke in order to relax 65 70 67 50 66 66 66
I need the nicotine in cigarettes 30 24 24 30 37
40 25
I smoke because cigarettes are
being handed around 43 45 42 48 42 48 42
1 smoke In order to break the ice
with other s=okers 21 23 20 18 20 19 23
I smoke when deeply absorbed in
what I'm doimg 32 29 32 28 33 33 29
1 smoke in order to help me
concentrate 33 30 38 29 33 32 31
1 smoke instead of nibbling something 41 36 42 39
43 44 35
1 smoke when working but not deeply
absorbed in the task 47 40 52 43 48 51 44
I smoke in order to feel less tense
or less ir-1---able 56 52 53 50 62 66 48
I smoke whe2 I'm feeling hungry
but I can't or don't want to eat 28 24 25 22 33 32
21
Smoking Is one of my few indulgences 81 78 81 88
82 61 82
I smoke whe= feeling alone 51 51 45 48 55 56 42
Cigarettes are comforting 70 74 72 74 67 66 71
1 smoke when feeling tense or
anxious 73 67 63 67 84 81 71
I like to snake when I have a drink 81 83 81 71 83
85 so
I really emfoy smoking - 69 83 77 63 58 53 71
1 find smoklng very satisfying 71 85 78 66 62 60
70
I smoke whe-a feeling down 70 61 66 61 77 75 65
I smoke after receiving bad news
or a sudden shock 76 77 75 68 78 78 70
Handling a cigarette is part of the
joy of smoking 39 39 40 45 37 43 31
I always hz---:e cigarettes at certain
times or oct-asions during the diy 59 50 58 61 62
64 51
C--
NJ
BATCo document for Province of British Columbia 29
October 1999


As with the attitudes to health battery, it is
possible to identify
a number of items in the motivation set which
extend the differentiation
between smokers obtained when using the
consonance/dissonance criteria.
The congruent smoker is more likely than any other
category to smoke
just out of habit;
instead of nibbling something;
in order to feel less tense or less irritable;
when feeling alone;
when having a drink;
at certain times/occasions during the day.
He is less likely to find cigarettes comforting
and less likelv to
really enjoy smoking or to find it satisfying. He
is more likely to
admit to needing the nicotine in cigarettes.
The incongruent smoker is distinguished by being
the least likely to
smoke
when relaxing;
when feeling bored;
instead of nibbling something;
in order to feel less tense/irrItable;
when feeling hungry but unable or unwilling to
eat;
when feeling alone.
He is also the least likely to perceive handling a
cigarette as part
of the joy of smoking.
Eighty-five percent of congruent smokers have
attempted to quit
smoking, compared with 72% of incongruent smokers.
For the congruent
group the average length of time since the last
attempt to quit was
(J-7
\-O
BATCO document for Province of BritiSh Columbia 29
October 1999


-24-
1.99 years and for the incongruent group, 3.15
years. The superficial
similarity between these values and those recorded
for the highly dissonant
group (=2.18 years) and the consonant group (=3.25
years) is worth noting.
Mentions of general or specific health reasons for
trying to quit were
made by 79Z of the congruent group (cf. 73% of the
highly dissonant group)
but by only 54% of the incongruent group (cf. 53%
of the consonant group).
Only 19% of the incongriient smokers who had
attempted to qijit had made
more than one serious attempt in the past two
years. For the congruent
group the equivalent figure was 32%. Of the total
Incongruent group only
21% would very seriously like to give up smoking
cigarettes compared with
61% of the congruent group.
It is clear that members of the congruent group
are under greater
pressure than the incongruent group to give up
smoking cigarettes.
Forty-five percent of the former compared with
twenty percent of the
latter claim to be under a great deal/a fair
amount of pressure. Similarly,
pressure to modify smoking behaviour is more
accutely felt by the congruent
smoker group (42% cf. 20% are under a great deal/a
fair amount of pressure
to modify).
The responses to the smoking and health items are
recorded in Table N
in similar fashion to the data presented in Tables
9 and 10.
NJ
all
BATCo document for Province of British Columbia 29
October 1999


TABLE I I
PERCENTAGE RESPONDENTS AGREEING ST RONCLY/ACREEING
Item SMO HIC CON DIS HID J Congruent Incongrjent
There Is too much fuss being made
k
about smoking and health ris .s
51
71
62
48
35
25
53
Smoking is not hurting =e as long
as I feel alTir-ht and show no
signs of health problems 65 86 82 54 46 34 76
Smoking might be harmful but .
I can't stop 53 36 45 59 65 68 47
The I pet fr- Inp.
is worth any small risks it might
involve 54 69 74 46 36 30 66
Cigarette smoking simply makes
it more likely you would get an
illness you would get anyway 1 51 48 55 59 49 46
62
Many non-smokers die of illnesses
that are often said to be caused
by smoking cigarettes 84 85 79 84 85 82 87
There Is nothing wrong with smoking
so long, as a person smokes modcr;lLely 75 90 90
71 57 45 85
Mild cigarettes are safer than
strong cigarettes 73 69 79 77 71 73 78
Doctors do not agree amongst
themselves about the possible
dangers of smoking to health 61 68 64 68 55 49 60
You cannot always believe what the
government tells you about smoking
and health 57 67 59 53 49 40 54
You could be run over by a bus
tomorrow so why worry about smoking
cigarettes 72 85 79 62 63 51 77
I've smoked for so long now that
stopping would not make any
difference now 47 67 60 45 30 20 52
Low tar cigarettes are safer than
other cigarettes 71 63 76 70 72 71 75
1 enjoy smoking and don' t see
why I should change my ways 69 91 90 57 47 38 81
I fully intend to give up smoking
some day 36 9 14 52 66 67 26
I could give up smoking today
if I really wanted to 34 38 43 25 29 29 1 21
11 C
011
01.
BATCo document for Province of British Columbia 29
October 1999


-26-
Again, the congruent smoker group is distinguished
by being more
likely than any other smoker sub-group to agree
with an item on two
occasions - "Smoking might be harmful but I can't
stop" and "I fully
intend to give up smoking some day" - and is less
likely to agree with
ten items. The incongruent smoker classification
extends the range of
recorded responses on three items. The
significance of these response
patterns will be discussed later.
Table 12 represents the data in respect of the
"social acceptability"
of sTnoking items.
TABLE 12
PERCENTACE RESPONDENTS ACREEING STRONGLY ACRE E I
NC
Itcm SM 0 HIC CON DIS HID Congruent Incongruent
Cigarette smoking is harmful
to the health of non-smokers 36 22 27 39 50 53 44
Cigarette smoking is a dirty
habit 60 38 49 67 78 84 68
Nowadays it is much less
socially acceptable to smoke 58 47 51 68 65 65 57
Nowadays there are times when
1 feel embarrassed about smoking 28 12 15 34 44 52
20
Just because other people don't
like smoking I don't see why I
should not smoke in their
presence 44 49 49 45 38 33 41
Again the implications of these data will be
discusged later.
It is not surprising, in the light of the data
presented hitherto,
to find that, more than any other smoker
sub-group, the congruent smoker
has a greater awareness of the substances in
cigarettes. Of those
congruent smokers able to mention more than one
substanec In c1garettes,
74% nominated "tar" as that substance most harmful
to health.
BATCo document for Province of British Columbia 29
October 1999


-27-
Amongst incongruent smokers the equivalent figure
was 52%. Of all
congruent smokers 54% believe that nicotine is
definitely harmful (cf. 34%
incongruent smokers) and 697. believe tar to be
definitely harmful (cf.
46% incongruent smokers). The tar band level of
their most often brand
is very/fairly important to 69% of congruent
smokers but to only 39% of
incongruent smokers.
DISCUSSION
Concern for the consequences on one's own health
are but one reason
I-or wishing to quit smoking c"Igarettes. It is
becoming clear that It Is
erroneous to infer that the highly dissonant
smoker group are defined
thus solely because of their health concern.
Indeed, only 53 percent of
HIT) snokers are very/quite concerned about the
possible effects of their
cigarette --miuking on their own health.
The core of the health concerned smoker population
is probably best
defined by those considered here as congruent
smokers. In most ways, the
congruent group presents a more extreme profile of
responses than the
highly dissonant group. Of particular interest,
however, are the group
designated as incongruent smokers for whom, on a
purely rational basis,
the apparent mismatch between views about smoking
and health when applied
generally and when applied specifically to
themselves is difficult to
comprehend. Three alternative ways to explain this
apparL!nL nnoivinly are
logically possible:
(a) Despite their responses to the item, they do
not really believe
that cigarette smoking is harmful to health;
(b) Despite their responses to the item, they are
more concerned
about the consequences of their smoking on their
health;
c0
BATCo document for Province of BritiSh Columbia 29
October 1999


-28-
(c) They are truly incongruent and are able to
entertain concurrently
two apparently contradictory dispositions.
Any combination of the above may be possible, of
course, for any
individual. Also the amalgamation of two response
categories in the case
of both defining items may distort the data as
presented here. Closer
inspection of the data suggests that this is at
least partly the case.
Of the congruent grotip, 49Z st o p.:LN~ arr c
thit cigarette ~mokinp. Is
harmful to health (cf. 16% incongruent group). Of
the incongruent group,
33% were not at all concerned about the possible
effects their cIrqrette
smoking has on their own health (677 were not very
concerned).
On this basis a maximum of only 16% of the
incongruent group are
potentially truly incongruent; that is the 21
respondents who strongly
agree that smoking is harmful to health. If these
same Individuals are
totally accounted for amongst the 90 who were not
very concerned, rather
than not at all concerned about the effects
cigarette smoking has on
their own health, then it is possible to account
for the incongruity
simply on the basis of response selection. Further
analysis will be
required to identify any individuals who are not
at all concerned and
strongly agree that smoking is harmful to health.
For the time being it
should be noted that the "apparently" incongruent
smokers account for
less than 3 percent of the total smoker
population.
Just over 12 percent of all smokers are very
concerned about the
effects of smoking on their health and a further
19 percent are quite
concerned. overall, 21% of smokers strongly agree
that cigarette snoking
is harmful to health and a further 41% agree.
Conversely, only 10%
disagree and 3% disagree strongly. The congruent
smoker category appears
No
011
110
BATCo document for Province of British Columbia 29
October 1999


-29-
to identify the health concerned element within
the two dissonant
groups. In respect of most psychographic indices
and in response to the
attitudinal items, the emergent profile of the
congruent smoker is often
one of greater extreme than the highly dissonant
individual. It would
seem reasonable to concliide that the congrtient
smoker, a-; defined here,
is largely representative of the health concerned
sector of the dissonant
groups.
3. HEALTH CONSCIOUSNESS AMONGST SMOITRS A14D
NEVER-SMOKERS
In this section, attention Is focussed on those
smokers and never-
smokers who agreed strongly/agreed with the
statement "I am very health
conscious". Amongst stnokers, 31% qx~allfled, and
amongst never-smukers
37% qualified. In the following discussion, the
responses of these two
groups are compared and coiitraqtcd, with
occasional reference to their
.1 parent" categories when there is a major
divergence from other smokers
or never-smokers.
Both health conscious smokers (HCS) and health
conscious never-
smokers (HC14) were more likely than other smokers
and other never-smokers
to have visited the doctor at least once in the
previous 12 months (71%
HCS versus 62% other smokers and 71% versus 63%
other never-smokers).
Similarly, HCSs and HCNs had, more than other
smokers or other never-
smokers, been in hospital during the past five
years (37% versus 22% and
28% versus 21% respectively).
The differences in response to the attitudes to
general health items
can be seen in Table 13.
C-D
Qlrl
Un
BATCo document for Province of British Columbia 29
October 1999


-30-
TABLE 13
PERCENTAGE RESPONDENTS AGREEING STRONGLY AGREEING
Item HCS HCN
I consider myself to be a very healthy person 66
74
I eat more than I should 41 37
1 am in very good physical condition 55 66
1 exercise regularly 51 56
1 eat more sweets than I shoiild 24 26
1 am careful to eat a balanced diet 48 59
I think about my physical condition a lot 78 71
There is too much fuss about what one should
and should not eat 59 56
I am overweight 33 35
1 get out of breath easily 34 21
1 often feel tense 50 42
(I am very health conscious) (100) (100)
I am very weight conscious 55 60
1 am generally a very relaxed person 67 59
Whilst both the HCS and HCN group, by definition,
believe themselves
to be very health conscious, the never-smoker
group provides a somewhat
more integrated response pattern on the other
items to support their
health consciousness. For example, relative to the
HCS they are more
likely to perceive themselves as very healthy, to
be in very good physical
condition, to be careful to eat a balanced diet
and to be very weight
conscious. They are less likely to eat more than
they should and to think
CZ-)
BATCo document for Province of BritiSh Columbia 29
October 1999


-31-
about their physical condition a lot. The HCS is
more likely to get out
of breath easily, to often feel tense and
(paradoxically, perhaps) to
clai:3 to be generally a very relaxed person.
It is interesting to note that whereas 36% of the
HCS group
alwa7s/usually take some medicine or a tablet if
they have n headache
(cf. 33% of all other smokers), only 30% of the
HCN group respond similarly
(cf. 35% of all other never-smokers). That is, the
health conscious
smoker has an inclination greater than other
smokers to take medication
whereas the health conscious never-smoker is less
likely than other
never-smokers to do so. This is particularly
reflected in relation to a
head cold where 40% 11CS compared with 30%. other
smokers always/usually
use 7edication. For never-smokers the difference
is less marked (33% HCN
cf. -11% other never-smokers). Members of the HCS
are more likely to have
beem prescribed Valium (26% cf. 21%) and are more
likely to he currently
taking them (81' cf. 5Z). In all these instances,
the quoted levels are
great.er than for the other smoker or other
never-smoker groups.
Psychographically, the HCS and HCN groups show
higher levels of
anx~sty, authoritarianism and social desirability
than their "parent"
groups of all smokers and all never-smokers. In
each of these cases,
lower mean scores are recorded for the RCN group
than for the BCS group.
With regard to the smoking behaviour data, it is
interesting to note
that, more than any other category of smoker, the
HCS is most likely to
smoke in order to make a break in tasks, in order
to relax, to break the
ice -aith other smokers, when feeling alone, after
receiving bad news or a
sudden shock. He is also most likely to find
cigarettes comforting and
to azree that handling a cigarette is part of the
joy of smoking.
\0
_4 ~.
N-)
__J
N-1
BATCo document for Province of BritiSh Columbia 29
October 1999


-32-
The ECS is less likely to have tried to give up
smoking cigarettes
altogethe7 (69% compared with 72% of all smokers)
and, if he has tried,
has been able to abstain for a lesser period than
the average for all
smokers ('11.9 weeks compared with 13.6 weeks).
Amorlst HCNs, 41% are very/quite concerned about
the dangers to
their health from other people smoking cigarettes
(compared with 35% of
all never-smokers). The equivalent values for the
HCSc; and all smokers
are 10% Ezzd 6% respectively.
DISCUSSICX
Therz- is a general consistency amongst those
subjects claiming to be
very beal:h conscious when their responses to
other health related items
are inspe:ted. Particularly, in the case of the
health conscious never-
smoker, -,)ere is a general awareness of own well
being extending into
appropriE:e eating habits, weight consciousness,
etc.
It %-,uld appear that the health conscious smoker
is less able to
support 11--s health consciousness by reference to
his general heal th/behaviour
patterns. One might venture to suggest that the
he,13th conscffous -;moker
claims menbership of the group more through an
awareness of his failings
(getting out of breath, often feeling tense, etc.)
whereas the corresponding
never-smc;cer presents a life-style which actively
supports his claim to
be health conscious. For both groups, their
attitudes appear to he
related to the elevated levels of anxiety recorded
which were greater
than those recorded for their "parent" groups.
4. NEV--K-SMOKERS AND OTHER PEOPLE SMOKING
A f,=ther sub-set of the total sample questioned
which is worthy of
particula7 attention is that comprising those
persons who ha-.-e never
N.)
BATCo document for Province of BritiSh Columbia 29
October 1999


-33-
smoked but who are very/quite concerned for the
consequences on own health
from other people smoking. Of the total of 747
never-smokers interviewed
in the survey, 261 (35%) qualified for Inclusion
in this category which,
for the purposes of this report we shall designate
as "sensitive" never-
smokers. Members of this group are more strongly
represented in the
higher socio-economic categories.
PsYchographically compared with all
never-smokers, this group produced higher mean
score on all the parnmeters
investigated.
In the following d1scussion the datA for the
sensitive iiever-F~-nkvrs
are compared with those of the other never-smokers
but only where t!~=_re
is a marked deviation will attention be drawn.
The sensitive never-smoker is more likely to have
visited his Eoctor
with an ailment (average number of visits 4.14 cf.
3.70 for all never-
smokers) with 71% of sensitive smokers having made
at least one visit
in the past 12 months compared with 63% of other
never-smokers.
The responses given to the general health items
are shown in Table 14
where the percentages agreeing strongly/agreeing
with each item are
recorded for all never-smokers, sensitive
never-smokers and other never-
smokers.
U1
_rzzl
BATCo document for Province of British Columbia 29
October 1999


-34-
TABLE 14
PERCENTAGE RESPONDENTS AGREEING STRONGLY/AGREEING
Item NEV Sensitive Others
I consider myself to be a very heAlthy person 76
69 79
I eat more than I should 45 49 42
I am in very good physical condition 59 56 60
I exercise regularly 48 48 48
I eat more sweets than I should 29 29 29
I am careful to eat a balanced diet 52 54 50
I think about my physical. condition a lot 45 53
42
Th ere is too much fuss about w)iat one should
and should not eat 61 58 63
1 am overweight 38 38 38
I get out of breath easily 20 21 19
I often feel tense 37 39 35
1 am very health conscious 37 43 33
I am very weight conscious 41 46 38
I am generally a very relaxed person 62 60 64
In comparison with other never-smokers, the
sensitive never-smoker
is more likely to
eat more than he should;
he careful to eat a balanced diet;
think about his physical condition a lot;
often feel tense;
be very health conscious;
be very weight conscious.
CD
BATCO document for Province of BritiSh Columbia 29
October 1999


-35-
He is less likely to
consider himself to be a very healthy person;
be in very good physical condition;
agree that there is too much fuss made about what
one should
and should not eat;
be a generally very relaxed person.
The pattern of responses observed here is entirely
consistent .:ith
the health data reported earlier and the general
psychographic profile of
the sensitive never-smoker. There are some
resemblences also with tlv~
congruent (health conscious) smoker previously
described.
Sensitive never-smokers are more likely than
otliers to have been
prescribed Valium (21% cf. 167) and Mogadon (11%
cf. 7%). Current usage
does not differentiate, liowever. Sensitive
never-smokers liave a lower
average daily consumption of tea, but not of
coffee or alcohol.
Compared with other never-smokers, sensitive
never-smokers are more
aware of the substances, other than tobacco and
paper, that are in
cigarettes. Amongst all never-smokers only 22%
were able to cite more
than one substance in cigarettes or cigarette
smoke which they believed
to be harmful. Nicotine was cited by 45% of
sensitive smokers as the
one substance most harmful to health (cf. 33%
other never-smokers). The
indictment of tar was more prevalent amongst other
never-smokers than
sensitive never-smokers (57% cf. 40%). Of the
sensitive never-smokers,
70% believed nicotine to be definitely harmful
(cf. 49% other never-
smokers) and 73% believed tar to be definitely
harmful (cf. 55% other
never-smokers).
BATCo document for Province of British Columbia 29
October 1999


-36-
When asked, "How far do you personally agree or
disagree with the
statement: cigarette smoking is harMful to
health?" 767. of the sensitive
never-smoker group expressed strong agreement (cf.
47% other never-
smokers). In passing, it is also interesting to
note that two sensitive
never-smokers disagreed strongly with the
statement!
Sensitive never-smokers appear to be more active
in attempting to
perstiade someone to give up smoking cigarettes,
sixLy-nine percent claiming
such an attempt (cf . 52% other never-smokers),
and claim greater vigour
in their efforts (77% of sensitive never-smokers
who had attempted to
persuade someone to quit smoking claimed to have
tried very/quite hard,
cf. 69% other never-smokers).
DISCUSSION
The sensitive never-sinc4-.ers' relatively high
levels of neuroticism
and anxiety, coupled with an exaggerated concern
with their physical well-
being suggests that ;moklng Is just one of the
number of targets upon
which their disapproving sights might be set.
Whilst they are conscious
about their own health and weight they
simultaneously perceive themselves
as less healthy and admit to over-eating.
It would appear that good health is an important
aspect in their
lives, perhaps because they perceive it not to
have visited them in
sufficient manner. This is supported by their
relatively higher incidence
of visits to doctors and hospitals. We would
suggest, at this stage,
however, that it is the characteristically high
levels of anxiety and
neuroticism which is driving both the apparently
hypochondriacal behaviour
and the attitudes to other people smoking.
U_.
~_r
1110
4 ~t_.
N-)
BATCo document for Province of British Columbia 29
October 1999


-27-
5. SMOKERS UNDER PRESSURE TO QUIT SMOKING
Amongst the smokers participating in this project,
238 (32%) were
identified as being under a great deal/fair amount
of pressure to quit
smoking/modify their smoking behaviour. In this
section those responses
which distinguish the "pressured" smoker from
other smokers are described.
More than any other smoker category, they are most
likely to be
married and aged hetween 25 and 34 years. Sixty
percent linve at Ivast
one child (cf. 48% of all otber smokers) and 12%
are to be found in social
classes A or B (cf. 4% of all otber smokers). They
are sliglitly more
likely to be members of households in which no one
else smokes (52Z cf.
49% all otber smn~-crs). Compared with all other
smokers, the pressured
smoker is more likely to parLicipate regularly in
sport (22% cf. 16%).
Psychographically, the pressured smoker group
produced lower mean scores
than all other smokers in respect of locus of
control (i.e. they are less
fatalistic), authoritarianism, ego strength,
social desirability, but
higher scores on anxiety and neuroticism.
Amongst smokers they are the most likely to have
visited the doctor
at least once with an ailment during the previous
12 months (72% cf. 62%
all other smokers). Their responses to the general
health items are
presented in Table 15 where the percentages
agreeing strongly/agreeing
with each item are recorded for all smokers,
pressured smokers, and other
smokers.
LY1
(-J1
cc
BATCo document for Province of BritiSh Columbia 29
October 1999


TABLE 15
PERCENTAGE RESPONDENTS AGREEING STRONGLY /AGREE
ING
Item SMO Pressured Other5
I consider myself to be a very healthy person 67
58 71
I eat more than I should 40 45 39
1
am in very good physical condition 53 43
58
I exercise regularly 39 41 39
I eat more sweets than I should 25 27 1 24
I am careful to eat a balanced diet 40 41 41
1 think about my physical condition a lot 47 51 45
There is too much fuss about what one should
and should not eat 67 62 70
1 am overizeight. 35 37 34
T get out of breath easily 34 38 32
1 often feel tensc 46 47 45
1 am very health conscious 31 31 31
I am very weight conscious 34 40 31
.1 am generally a very relaxed person 63 57 66
The profile of responses given by the pressured
smoker, compared
with the others, is consistent with their higher
mean score.~ on the
anxiety and neuroticism scales. More than other
smokers they tend to
eat more than they should;
eat more sweets than they should;
think about their physical condition a lot;
admit to being overweight;
get out of breath easily;
be very weight conscious.
U7
11 C
_r~t-
BATCo document for Province of British Columbia 29
October 1999


_39-
Relative to other smokers, they are less likely to
consider themselves healthy persons;
believe they are in very good physical condition;
believe that there is too much fuss about what one
should
and should not eat;
be generally very relaxed persons.
In many ways this profile resembles that observed
for the sensitive
never-smoker and is consistent with their
qualification of perceiving
themselves to be under pressure to quit and/or
modify their smoking
behaviour.
Amongst the smoker group, the pressured smoker is
most likely to be
currently smoking low tar (24%) and low to middle
tar (8%) cigarettes.
1 qe pressured smoker Is likely to have ch;;nged
from his prevlotir to
current brand for reasons of health (43%). Their
responses to the smoking
cotivation items were broadly similar to those
obtained for the highly
dissonant and congruent smoker groups. Eighty-two
percent of pressured
smokers had made at least one atLempt to give up
smoking cigarettes
altogether and 30% had made more than one serious
attempt in the last two
years. Their average attempt rate was higher than
any other smoking
group and they also achieved the longest average
period for giving up
(21.45 weeks). Seventy-three percent would
very/fairly seriously like to
give up smoking cigarettes altogether, and of
these 61% mentioned general
health reasons for quitting. Pressure to quit came
largely from the
respondent's husband/wife (47%) as did pressure to
modify their smoking
behaviour (39%).
The responses to the smoking and health Items are
shown is Table 16
where the data for the pressured group is compared
with those from all
s7 :okers (SMO), highly dissonant smokers (HID)
and congruent smokers.
-cz:.
N-)
CC)
BATCo document for Province of BritiSh Columbia 29
October 1999


-40-
TABLE 16
PERCENTAGE RESPONDENTS AGREEING STRONGLY AGREEING
Item SHO Pressured HID Congruen'
There is too much fuss being made about smoking
and health risks 51 41 35 25
Smoking is not hurting me as long as I feel
alright and show no signs of health problems 65 47
46 34
Smoking might be harmful but I can't stop 53 61 65
68
The enjoyment I get from smoking is worth any
small risks it might involve 54
Cigarette s:::oking simply makes it more likely
that you would get an illness you would get
anyway 51
Many non-s--ol"ers die of ilincsses that are often
said to be caused Ly sinoking cigarettcs 84
There is nothing wrong with smoking as long as
a person s--okes moderately 75
Mild cigarettes are safer than strong cigarettes
73
Doctors do not agree amongst themselves about
the possible dangers of smoking to health 61
You cannot always believe what the government
tells you about smoking and health 57
You could be run over by a bus tomorrow so why
worry about smoking cigarettes 72
I've smoked for so long that stopping wotild not
make any difference now 47
Low tar cigarettes are safer than other cigarettes
71
I enjoy smoking and don't see why I should change
my ways 69
I fully intend to give up smoking some day 38
I could give up smoking today if I really wanted
to 34
43 36 30
51 49 46
80 85 82
58 57 45
77 71 73
58 55 49
45 49 40
58 63 51
29 30 20
75 72 71
53 47 38
56 66 67
27 29 9
%10
0 C
BATCo document for Province of BritiSh Columbia 29
October 1999


-41-
ComT-=-red with the highly dissonant and congruent
smokers, the
pressured smoker tends to reflect a more
"moderate" disposition in response
to these ftems. In the majority of cases, the
group response magnitude
is positioned between that of all smokers and the
highly dissonant and
congruent groups. The relatively greater resort to
fatalistic heliefs by
the high17 dissonant smokers ("You could get run
over by a bus...")
together with their greater distrust of government
pron(,iinrv,-:r-nts nbout
smoking and health are notable. It is not
surprising to find that the
pressureE smoker expresses a greater faith than
average in mild arid low
tar cigar-2ttes.
The 7esponses to the social acceptability of
smoking items are given
in Table 17.
TABLE 17
PERCENTAGE RESPONDENTS AGREEING STRONGLY /AGREE
ING
Item
Cigarette smokirZ is harmful to the health of
non-smokers
Cigarette smoki-.-z is a dirty habit
Nowadays it is Trach less socially acceptable
to smoke
Nowadays there are times when I feel embarrassed
about smoking
Just because other people don't like smoking I
don't see why I should not smoke in their presence
SMO ; Pressured BID Congruent
36 43 50 53
60 72 78 84
58 64 65 65
28 44 44 52
44 36 1 38 33
The response pattern for the pressured smoker is
generally similar
to that cbserved for the highly dissonant and
congruent smokers except
that he 4--4z rather less likely to accord with
the vie%., that cicarette
smoking is harmful to the health of non-smokers.
Un
L-n
`10
co
N)
BATCo document for Province of British Columbia 29
October 1999


-42-
Eighty percent of pressured smokers agreed
strongly/agreed that
cigarette smoking is harmful to health (cf. 82%
HID) but only 50% were
very/quite concerned about the possible effects
their cigarette s-oking
has on their own health (cf. 53% HID). Nine
percent of pressured smokers
were very/quite concerned about the dangers to
their licalth from other
people smoking (cf. 11% HID, 17% congruent).
DISCUSSION
Pressure to quit/modify smoking behaviour appears
to be largely
demographically determined. reneral awareness of
the potentlil hazar'4s
to health is greatest amongst the young,
middle-aged, upper socioeconcmic
groups. Not surprisingly, therefore, it is within
these same categories
that the greatest incidence of perceived pressure
is observed. In
addition, -their incidence of illness, as
indicated by the freqilency of
visits to a doctor, and their responses to the
general health items
indicate an awareness of their well-being (or
relative lack of It) against
which reasons to quit/modify smoking given by
influential others are
likely to be perceived as pressure.
It is the pressured smoker who is most likely to
seek a modification
of smoking behaviour through mild or low tar
products. Twenty-four
percent of this group currently smoke 1o%.; tar
products, a greater prosortion
than for any other category defined in this study.
It would appear that
the group of pressured smokers draws its
membership largely from the two
dissonant groups although it is worth noting that
13% of consonant sn-okers
and 9% of highly consonant smokers believe
themselves to be under a great
deal/a fair amount of pressure to give up smoking
altogether. For
modifying smoking behaviour, the corresponding
figures are 16% and 7*,
QF1
(Jrl
co
BATCo document for Province of BritiSh Columbia 29
October 1999


-43-
respectively. It would appear therefore that a few
highly consonant and
consonant Smokers maintain their behaviour despite
pressure from others
to quit.
CONCLUSION'S
This report terminates the presentation of the
top-line data from
Project L13RA. The results obtained from this
inspection have confirmed
the utili~:, of the consonance/dissonance
dimension in differentiating thf-
smoker po-,ulation. The further categories defined
for the present report
additionally suggest that the simple distinctions
hased upon r--sponses to
items relEcing to attempts to quit and current
intention to do so support
the vie. :hat further differentiation within the
consonant/dissonant
categorie-z is likely to yield a more valuable
segmentation of the smoker
population.
For :he time being we would offer the following
interim conclusions.
1. The :sychographic indices of anxiety,
neuroticism, ego strength, and
locus of control appear to offer differentiation
between the smoker,
ex-smoker and never-smoker categories of
respondent which have a
high face validity in respect of the attitudes and
behaviours measured
elsewhere in the survey. Similarly, within the
smoker group, those
differentiated according to degree of
consonance/dissonance present
gross psychographic profiles in accordance with
expectation.
2. The '-dghly dissonant smoker category includes
persons seeking to
quit smoking for reasons other than health
concern. The congruent/
incoagruent distinction between smokers may prove
more useful in
isolating those individuals particularly sensitive
to the issues
rela:ing to smoking and own health. Again, the
value of certain
NJ
CO
BATCo document for Province of BritiSh Columbia 29
October 1999


-44-
psychographic parameters in describing smokers
differentiated
according to congruence is noted.
3. Whether or not an individual, smoker or
never-smoker, believes
himself to be very health conscious was found to
be related to the
incidence of hospitilization and number of visits
made to the doctor.
General health consciousness appears to be related
to higher than
average levels of anxiety among, respondent-.
4. Those never-smokers expressing concern for
their own health from
other people smoking were found to be relatively
more T)(-urotic and
anxious than those not so concerned.
5. Smokers under pressure to quit smoking are
particularly identified
by socioeconomic and marital status factors.
Again, compared with
those not tinder pressure, hirher inean scores on
anxiety and neurot!cisln
are obtained by this group.
In further reports attention will be focussed on
the further
differentiation of the smoker category in an
attempt to better describe
the current motivational, behavioural and
attitudinal substrates underl-.-Ing
the U.K. market. It is intended, by multivariate
techniques, to atte---.:
a contemporary segmentation based on
psychographic, smoking behavlour,
and health response data, for both current and
ex-smokers in order to
elucidate the concerns of smokers and the ways
chosen to ameliorate the
effects of the conflict between smoking and health
perceived by many
smokers today.
The next report in this series will deal with the
cigarette consumPtion
and tar band switching data in an attempt to
determine the relationshi-~s
between cigarette consumption patterns, switching
behaviour and the
consonance/dissonance parameter.
N-)
CC)
BATCo document for Province of BritiSh Columbia 29
October 1999


_45-
REFERENCES
1. BAT Report No. RD.1670 Restricted, 23.4.79.
2. BAT Report No. RI).1697-C, 18.10.79.
3. Rotter, J.B. Generalized expectancies for
internal versus external
control of reinforcement. Psychol. Monogr. 1966,
80, 1. (Whole No. 629).
4. James, W.H., Woodruff, A.B. and Werner, W.
Effect of internal and
external control upon changes in smoking
behaviour. J. Consult.
Psychol. 1965, 29, 184-186.
5. Straits, B.C. and Sechrest, L. Further support
of soine findings abol;:
the characteristics of smokers and non-smokers. J.
Consult. Psychol.
1963, 27, 282.
6. Hjelle, L.A. and Clouser, R. Internal-external
locus of control of
reinforcement in smoking behaviour. Psychol. Rep.,
1970, 26, 562.
7. Marston, A.R. and Mcrall, R.M. Comparison of
behaviottral mod1ficatl'..-i
approaches to smoking -redtiction. J. Consult. &
Clin. Psychol., 1971,
36, 153-162.
8. Platt., F.S., Krassen, K. and 11ausner, B.
Individual variation in
behavioural change follo.ing role playing.
Psychol. Rep., 1969,
24, 155-170.
9. Berman, A.L. Smoking behaviour: How is it
related to locus of
control, death anxiety, and belief in afterlife?
Omega, 1973,
4, 149-155.
10. Keutzer, C.S. Behaviour modification of
smoking: The experimental
investigation of diverse techniques. Behav. Res. &
Therapy, 1968,
6, 137-157.
11. Lefcourt, H.M. Personality Correlates of
CiFarette Smoklng,
Department of National Health and Welfare, Ottowa,
1965.
12. 'Lichtenstein, E. and Yeutzer, C.S. Further
formative nnd correIntional
data on the internal-external (1-E) control of
reinforcement "-cale.
Psychol. Rep., 1967, 21. 1014-1016.
13. Bernstein, D.A. Modification of smoking
behaviour: An evaluative
review. Psychol. Bull., 1969, 71, 418-440.
14. Lichtenstein, E. and Keutzer, C.S.
Experimental investigation of
diverse techniques to modify smoking: A follow-up
report. Behav. RES.
& Therapy, 1969, 7, 139-140.
15. Best, J.A. and Steffy, R.A. Smoking
modification procedures tailor&,-:
to subject characteristics. Behav. Therapy, 1971,
2, 177-191. CD
c7 N
BATCo document for Province of BritiSh Columbia 29
October 1999


-46-
16. Collins, B.E. Four components of the Rotter
internal-external
scale: Belief in a difficult world, a just world,
a predictable
world and a politically responsive world. J.
Personality and Soc.
Psychol., 1974, 29, 381-391.
17. Mirels, H.L. Dimensions of internal versus
external control.
J. Consult. & Clin. Psychol., 1970, 34, 226-228.
18. Raemmerer, W.F. and Schwebel, A.I. Factors of
the Rotter internal-
external scale. Psychol. Rep., 1976, 39, 107-114.
19. Mlott, S.R. and Mlott, Y.D. Dogmatism and
locus of control in
Ind1vtduals who smoked, stopped vmr)king and never
qmok-1. .1. Comm.
Psychol., 1975, 3, 53-57.
20. Foss, R. Personality, social influence, and
cigarette smoking.
J. Health & Soc. Behav., 1973, 14, 279-286.
21. Eysenck, B.J. A short questionnaire for the
measurement of two
dimensions of personality. J. Appl. Psychol.,
1958, 14, 14-17.
22. Eysench, H.J. and Eysenck, S.B.G. The Eysenck
Personality Inventory,
London: University of London Press Ltd., 1963.
23. Clierry, N. and Vieriian. K. Perst-Alty ~corcs
and smoking behavlo'lr:
A longitudinal study. Brit. J. Prev. Soc. Med.,
1976, 30, 123-131.
24. Taylor, J. A personality scale of manifest
anxiety. J. Abnorm. Soc.
Psychol., 1953, 48, 285-290.
25. Zander, A. and Thomas, E. The validity of a
measure of ego strength.
Unpub. paper, The University of Michigan,
Institute of Social Research,
1960.
26. Lane, R. Four item F-scale. In: Political
personality and electoral
choice, Am. Political Science Rev., 1955, 49,
173-190.
27. Smith, G.M. Personality and smoking: A review
of the experimental
literature. In W.A. Hunt (Ed.), Learning
mechanisms and smoking.
Chicago: Aldine, 1970.
28. Smith, G.M. Relations between personality and
smoking behaviour
in pre-adult subjects. J. Consult. Clin. Psychol.
1969, 33, (6),
710-715.
29. Walker, R.E., Nicolay, R.C., Klucsny, R. and
Riedal, R.G. Psychological
correlates of smoking. J. Clin. Psychol., 1969,
25, (1), 42-44.
30. Srole, L. Social and psychological factors in
smoking behaviour. The
Midtown Manhatten study. Bull. N.Y. Acad. Med.
1968, 44, (12), 1502-1513.
31. Thomas, C.B. On cigarette smokinp, coronary
heart diseanc 'ind tlic.
genetic hypothesis. John Hopkins Med. J., 1968,
122, 69-76.
NJ
CO
BATCo document for Province of British Columbia 29
October 1999