THE RELATKDN OF CAR OWNERS‘ OPINIONS TO
CERTAIN OiARACTERlSTiCS OF THE OWNER
AND HIS CAR.

Thesis for {'50 Degree of M. A.
MICHEGAN STATE UNWERSET

Peter Wing Hemingway
1957

 

 

 

 

WNWMbmmMWMMM.
TOAVOIDFINESMumonorhdondCOM

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Nflnnativo Action/Equal Opportunity Inflation
Wanna-9.1

 

 

 

 

 

 

 

THE W10! 0? CAR OWNERS' OPINIONS T0 CERTAIN

CHARACTERISTICS OF THE OWNJSB AND HIS CAR

By
Peter Wing Buningny

A THESIS

submitted to the College of science and Arts

Ilichigen State University of Agriculture and

Applied Science in partial fulfillment of
the requirelnente for the degree of

MASTER OF ARTS

Department of Psychology

1957

Peter W. Hemingway

ABSTRACT

This investigation was designed to study the interrelations be-
tween car owners' Opinions and certain characteristics of the car owners
and their cars.

Two forms of a questionnaire were mailed to a random sample of
800 Ingham County, Michigan, car owners, 400 receiving each form.

122 respondents returned Form A questionnaires. These were analyzed
in terms of responses to individual features and to groups of features
on the questionnaire.

Opinions were scored on each feature for each respondent. The sum
of these scores for each group of features was obtained for each re-
spondent. These sums were used as indices of the respondent's satis-
faction with his present car, of his desire for different features in
his next car, of approval of trends in automotive design, and of the
extent to which certain features and trends contribute to safe Opera-
tion of a car.

The relationship between certain aspects of response and certain
other variables were tested by appropriate non-parametric statistical
techniques. The results indicated that the opinion of a particular
feature is to some degree a function of the make and model of car
owned, the height, weight and age of the respondent, distance driven
per year, the age of his car and his intention to buy or not buy a new
car. Moreover, the results indicated that satisfaction with the present

car was independent of how car owners felt about trends and safety,

but there was a smll negative relationship between satisfaction with

Peter W. Hemingway

the present car and the desire for different features on the next car.
Satisfaction with.the present car was also related to loyalty; re-
spondents planning to buy the sane make of car attaining significantly
higher satisfaction scores than those planning to buy a different make
of car.

The next car score showed significant relationships with the age
of the respondent and distance driven per year, as well as a high re-
lationship with the scores on trends and safety.

The trends score showed a high relationship with the age of the
respondent and with the model owned, as well as with the safety scores.

The safety score showed a high relationship with the marital status
of the respondent: married persons attaining significantly higher
scores than unmarried persons.

Certain owner characteristics were also related; age showing a
significant relationship with plans to buy a new car and distance driven
being significantly related to the sex and the marital status of the re-
~spendent, his plans to buy a new car, and the age ofihis present car.

All four kinds of scores were also related to the way respondents
expressed themselves with reference to the statements section of the
questionnaire.

However, there was no relationship between various characteristics
of the car owner or his car and the way in which he ranked certain fea-
tures in a paired-comparison section of the questionnaire.

The principle conclusion based on the study is that, on the whole,
opinions, with reference to features and groups of features, are related

to the characteristics of the car owner and his car.

The author of this thesis would like to express his thanks to
Doctor James S. Karslake, his major professor, whose patient and

helpful advice and assistance made this study possible.

LIST OF TABLES . . .
INTRODUCTION . . . .
BACKGROUND . . . . .
HYPOTHESES . . . . .

NETHOD AND PROCEDURE

The Questionnaire

The Sample . .

The.ana1ysis . . .

FINDINGS . . . . . .

RESULTS . . . . . .

CONCLUSIONS AND DISCUSbION

COMMENTS . . . . . . . . .

BIBLIOGRAPHY . . . .

AEPENDIX.. . .

TABLE OF CONTfiNTS

a. Form.a of the questionnaire

B. Characteristics of Respondents

C. scoring Procedure and Results

D. analysis of Score RelationShips

3. Item Reaponse Characteristics

10

10

12

13

18

23

28

34

36

. 38

TABLE I

TABLE II

TABLE III

TABLE IV

TABLE V THIDUGH VIII

TABLE IX THROUGH XV

TABLE XVI THROUQ XXIV

TABLE XXV THROUGH XXIX

LIST OF TABLES

Page

saunas? or RETURNS TO THE MAILING or
Fons A or THE QUESTIONNAIRE ... 19

SUMMARY OF SIGNIFICANT RELATIONSHIPS
OBTAINED BS‘T"EEN SCORES DRIVE FROM

EACH OF FOUR SECTICNS OF THE QUESTION-
NAIRE AND CERTAIN CHARACTERISTICS OF

THE WEB OR HIS CAB. ... 20

saunas! or RELaTIONSHIPS newness

UPPER AND Iowan 27% scone GROUPS AND

ITEM BY Iran Reasonsss TO THE STATE-

HENTS SECTION or was QUESTIONNAIRE ... 21

SWY OF SICE‘IIFICANT RELATIONSHIPS
OBTAIND BETWsEN CERTAIN CHARACTER-

ISTICS AND THE RESPONSE TO CERTAIN

FEaTURrIS, HOLDING CONSTANT ANOTHER
CHARACTERISTIC ... 22

DISTRIBUTIONS or CERTAIN CHARACTERr
ISTICS or THE ownsn.ann HIS CAR ... appendix 3

SCORE DISTRIBUTIONS 0! THE Ra}-

SPOiiDmTS TO THE FEATURES AND

GROWS OF FEaTURES ON THE

QUSST IONNaIRA: . . . appendix C

ANALYSSS 0? RSIATIONSHIPS BETWESN

SCORES DERIVED FROM TBS SEVERAL

SECTIONS OF THE QUESTIONNAIRE AND

CERTAIN CHARaCTSRISTICS OF THE

OWNER AND HIS CAR . .. Appendix D

TWO-WAY NON-PARAMETRIC ANaLYSfi

OF RELATIONSHIPS BETWEEN RESPONSES

TO FEATURES am: CERTAIN CHARACTER-

ISTICS OF THE OWNER AND HIS CAB,

HOLDING CONSTaNT ANOTHER

CHARACTrIRISTIC ... appendix E

INTRODUCTION

The primary purpose of this study was to investigate the relation-
ship between consumers' opinions about a product and certain character-
istics of the consumer or of the specific product that he uses. The
method proposed for obtaining the data required was a mail questionnaire
sent to a sample of cmsumers of a particular product.

It was assumed that the manner in which a person responds to a
question about his satisfaction with a particular feature of a product
is a function both of the person and of the product. Thus the degree
of satisfaction indicated for a specific feature or characteristic of a
product would depend upon the absolute satisfaction value of that feature
and upon the ability of that feature to fulfill the particular demands
of that person. It was further assumed that the degree of satisfaction
‘sith the product is a function both of the degree of satisfaction with
the individual features making up the product and the perceived import-
ance of these features.

The product selected for study was the car. The reasons for util-
ising this product for the study were many and varied. Prhmarily, it
was felt that the car is of major interest and concern to a large seg-
ment of our economy at the present time. Am information about the
consumers' opinions, both of the car they presently have and of possible
changes in the features of cars they may have in the future should be
of interest to both the consumer and the producer of the product.

The automdbile industry in the United States has grown from an
obscure spare-time activity of a few inventors and doodlers to one of the

largest single industries in our country in a period of fifty years.

The dynamic and highly competitive nature of this industry has led to
the rapid change (and sometimes improvement) of,features and character-
istics of the car. Every year each firm brings out new models, with
claims of sweeping improvement in this or that feature, and advertising
and news sources carry the claims and countereclaims to the eyes and
ears of the consumer in a continual stream. The basis for this constant
change and claim aspect of the industry is of course canpetit ion. The
only way to sell is to convince the consumer that what he has is obso-
lete--what he needs is something better--and this or that company is
prepared to sell it to him on time:

What is the effect of this all-out attempt to create a desire for
new cars and dissatisfaction with.what is within a year an obsolete,
out-of-date car? Does the consumer really believe that his car, once it
is at least one year old, is unsatisfactory? Does he believe that the
changes made or proposed are really improvements over what he has? And
what characteristics does the consumer or his present car have which are
related to what he believes about any one or all of the features of his
present car or of trends in never care? These are the questions which
this study will attempt to answer.

In order to determine a person's degree of feeling about any par-
ticular subject, it is possible to merely ask.him.how strongly he feels
about that subject and how much he likes or dislikes, is satisfied or
dissatisfied with, is interested or uninterested in, or values or
doesn't value that particular subject. But this is often a poor measure--
Pfiople don't know, they all say it's "o.k.", there is very little
difference between people. But it can be hypothesized that a person's

overall opinion on a subject is some sort of composite of his opinion

about each specific item making up that subject. This item opinion may
be based upon the importance placed upon that item, how that item com-
pares to similar items previously or subsequently experienced, and the
amount of improvement felt possible in that item.

Thus, a driver's Opinion about h0W‘W611_hIB car rides may be based
upon how important he considers a comfortable ride, how the ease of ride
of his car compares with cars he has ridden in before and other care he
has ridden in since he has had this car, and how well he feels a car
should ride to be ”ideal”. The driver who says that his car rides "like
a wagon on a railroad" may base this opinion solely on a comparison with
an ideal he wishes for.

The sum of these Opinions on the individual items may be considered
to represent the person's Opinion on the subject made up of those items.
That is, the degree of satisfaction with a particular feature combined
with the perceived importance of that feature forms a unit of the overall
satisfaction with the entire subject.

If one considers the features on which a car owner is asked to give
his opinion as a sample of the features making up the characteristics of
his present car, it may be assumed that, by weighing each feature in
terms of both the expressed satisfaction and the expressed importance,
then sunning these weighted values, a score expressing the owner's
overall Opinion of his present car may be obtained. Sheilar scores may
be obtained on the owner's overall desire for different or new features
on his next car, his overall liking of current trends or changes in
general, and his overall evaluation of such trends or changes as con-

tributors to safety.

It should follow’that such scores will show predictable relations

ships with certain characteristics of either the car owner, the car he

owns, or both. This study is designed to examine these relationships.

BACKGROUND

There is available a great body of literature dealing with
questionnaire methods and techniques and Opinion and attitude research
in general. However, only a brief survey of those sources which.were
particularly relevant to this investigation will be given.

The general area Of questionnaire studies includes a great deal on
methods and procedures, especially in governmental and educational areas,
but there is very little specific material to be found on consumer
opinion studies other than general discussions of methods and prOblems.
The only specific reference found on surveys of car owner's opinions
was an article by Weaver (23). He discusses the methods and motives
of the General Motors Consumer survey, but specific results are not
given. It is certainly reasonable to assume that the autmmobile manu—
facturers are regularly conducting consumer research, but, as in other
competitive fields, the results are not made public in order to maintain
trade secrets.

A great deal of literature is available on questionnaire methods.
Blankenship (l) devotes an entire book to the methods of conducting
opinion research by questionnaire techniques, and Kornhauser (ll) gives
explicit directions on the construction of questionnaires, including
the content, placement and writing of questions, the use of check lists
and the advantages and disadvantages of mail and interview methods of
obtaining respondents.

The specific use of mail questionnaires has received a great deal
0f study, primarily of shortcomings in the method, but recent studies

have been more uncritical. The returns to be expected utilizing mail

questionnaires are well discussed by Parten (13), results varying widely
depending upon the population sampled, the subject matter of the
questionnaire, and the mailing method. If the pOpulation is hetero-
geneous and very large, such as all Americans, the returns are generally
quite low, from 5 to 20 percent being the usual result when the other
factors are not controlled. However, the subject matter plays a large
role, with 10 to 20 percent higher returns on subjects of high general
interest, other factors being equal. Stanton (18) has shown that the
length of the questionnaire plays a large part in the number of returns.
A single question post card form.was returned by 50.2 percent of the I
sample, while a 3 page form on the same subject was returned by only
28.3 percent.

The use of follow-up letters and telephone calls can increase re-
turns to well over 90 percent. Stanton (18) found that the use of 3
follow-up letters, plus a personal call to those still not responding
gave a return of 94 percent. Waianen (20) found that a telephone call
before mailing the questionnaire increased returns by 12 percent.

The use of various inducements to increase returns have shown in-
teresting results. A study by Watson (22) showed an.appreciable in-
crease in returns (from 19 percent to 52 percent) when twenty-five cents
was enclosed with the questionnaire, but only a small increase in returns
(from 9.6 percent to 17.6 percent) when the sane amount was to be sent
upon return of the questionnaire. Parten (13) suggests the possibility
of using large inducements, such as one thousand dollars, On a lottery
or prise basis.

The amount and nature of bias in.mail questionnaires has been

widely studied, but the results are somewhat conflicting. Reuse (15)
found considerable bias in intelligence, education, background and
loyalty to the sponsor of the survey between returnees and non-returnees.
shuttleworth (16) reports a similar result, but notes that the bias
shifts with the time waited for returns, earlier returns showing the
greatest bias. a recent study by Wallace (21) indicates that bias is
a function of the sample and is usually negligible for questionnaires
of high interest sent to homogeneous samples. He concludes that bias
is most common on matters causing defensive or antagonistic responses
on the part of the subject.

Clausen and Ford (4) advocate a dual method for reducing bias,
utilizing all methods of increasing returns and making corrections and
allowances for existing bias. Good and scates (7), however, point out
that returns of even 90; may contain significant bias which is not de-
tectible from examination of the returns.

The methods available for analyzing questionnaire data range frmn
the simple frequencies and differences reported in.many polls and sur-
veys to the elaborate scaling techniques used in more theoretical in-
vestigations of attitudes and Opinions. While the method used depends
largely upon the nature and purpose of the study, there is usually a
wide choice of possible alternatives. The more common methods are dis-
cussed by Blankenship (l) and others, while inmormation on scaling
methods as related to questionnaires are covered by Guttman (9) and
Hyman (10). Clark and Kriedt (3) present a study on the advantages of
using Guttman's scaling method for Opinion questionnaires.

The general subject of opinion measurement, especially with ref-

erence to consumer research, is covered in the book.by Churchman and
others (2), which is a report of a conference on research methods.

Day (5) gives a good summary of the various methods used in the more
theoretical research, and the article by McNemar (12) gives a critical
survey of the shortcomings Of the more common methods. The entire sub-
ject is well covered in the books by Remmer (14) and Parten (13). The
paper by White and White (24) presents a good summary on the uses of the
results of Opinion research in the improvement of products.

The specific statistical methods used in this investigation were
based primarily on Guilford's (8) discussion on paired-comparison and
ranking methods, which was also used as a theoretical basis for the for-
mat of the questionnaire. The two-way coefficient of concordance tech-
nique was adapted from a suggestion by Edwards (6) on its applicability
to similar data. The specific computation methods for the other non-

parametric tests were obtained from Siegal (17).

HYPOTHBSES

gypgthgsis 1 - There will be no relationship between the way in which a
car owner responds to a group of features and any of certain character-
istics of the car owner or the car he owns.

That is, the scores derived from each section of the questionnaire
will be independent of: the height, weight, age, sex, marital status,
distance driven per year, loyalty, and plans to buy of the respondent;
the make, model, company and age of the respondent's present car;

scores and responses to other sections of the questionnaire.

mathegis ll - There will be no relationship between the way in which

a car owner responds to any featm-e and any of certain characteristics
of the car owner or the car he owns, even when any other such character-
istic is held constant.

That is, the response to any feature on the questionnaire will be
independent of: make when model is held constant; model when make is
held constant; compamv when model is held constant, model when compam'
is held constant; age when yearly distance driven is held constant; dis-
tame driven when age is held constant; height when weight is held con-

Itant; weight when height is held constant; plans to buy when age of

car is held constant.

muonmmocxmms

W

The questionnaire used in this study was designed by the investi-
gators to give a maximum amount of information about car owner opinions
on various features and trends in the development of cars, with a mini-
mum of effort on the part of the respondent. The features inquired
about were drawn from a list obtained by surveying advertisements,
articles and other sources of information on recent and proposed trends
in automotive construction and design. From this list twenty-four
features were drawn which were considered most relevant to cars manu-
factured between 1946 and 1956. These features were listed on the
Present Car section, and the respondent asked to check how satisfied
he was with this feature as it appears on his present car, and how im-
POrtant this feature is to him. It was felt that the opinion of a car
owner about a feature would be a function of the importance of that
feature to him. That is, satisfaction or dissatisfaction with a feature
thought to be important would indicate a stronger Opinion than with a
reams-e thought to be unimportant.

On the Next Car section, twelve of the features in the Present Car
Ioction were listed in terms of changes in that feature, along with
M1VO features or trends mentioned in the sources as current or pro-
P0.“ changes in cars. The respondent was asked to check whether he
“at” each feature on his next car, and how important it was that he
1"" each feature on his next car. The attempt was made to use features

on this and the next two sections which would represent changes from the

features found on the respondent's present car.

11

On the Trends section, the same twelve features were again listed
in terms of possible changes in that feature, along with twelve new
features mentioned in the sources as currently popular trends in auto-
motive construction and design. The respondent was asked to check
whether he liked or disliked each trend, and how important the trend was
to hum.

On the safety section, the twelve present-car features were again
listed in terms of changes from the present, along with twelve addition-
al features or trends from the sources mentioned as contributing to the
.safety aspect of driving. The respondent was asked to check whether he
felt each feature increased or decreased safety, and how important he
felt each feature is in terms of safety.

The features were then divided into two forms,.a and B, each form
containing in each section eight of the twelve features carried through
all of these sections and eight of the twelve features unique to each
section. This allowed an overlap of four of the features in each of the
groups on each section on bOth forms, so that a comparison of the re-
spondents to each fom could be made. Thus, each form contained in each
of these sections four features common to both forms and carried through
all four sections; four features common to both forms but unique to
each section; four features unique to each form but carried through all
four sections; and four features unique to each form and each section.
This made it possible to trace the consistency of responses to the
features appearing on all four sections, and the consistency with.which
the respondents on the two different fonms responded to the common items.

To provide additional checks on the consistency of response, both

by groups and by individuals, two additional sections were prepared.

12

The Features, or paired-comparison section, was composed of two sets of
five features each, taken from the Present Car section. Each feature
was paired with every other feature in its set, and the respondent was
asked to check the member of each pair which he liked the best (Form a)
or which he thought was most important (Fons B). The same pairs appear-
ed on both forms, so that the order of the features in terms of liking
and in terms of importance could be compared between the groups re—
sponding to the two forms.

The Statements section was composed of features drawn from the first
four sections, and put into statements with which the respondents were
asked to indicate how strongly they agreed or djsagreed. Twelve of
these statements were connon to both forms, the remainder unique to
each form.

The Questions section was included to provide additional infor-
mat ion on the features which were most important to the respondents,
and as a source of additional features for revising the questionnaire.
This section was the ease on both forms.

The first page of the questionnaire was composed of questions con-
cerning those characteristics of the respondent and his present car
which were desired for the analysis. although much more information
would have been desirable, it was felt that a minimum of personal
questions should be asked, to avoid unduly antagonizing any of the re-

spondents. These questions were the same on both forms.

W
The pOpulation chosen for this investigation consisted of the reg-

istered car owners of Ingham County, Michigan. The 1956 Motor vehicle

13

Registrations list was obtained and a random sample of 800 private
passenger car owners was drawn from this list, using a table of random
numbers. Two lists of 400 owners each were compiled from this master
list. One form of the questionnaire (either Form a or Form B) was sent
to each one of the 400 owners on each list. While it was not considered
that the pOpulation sampled was necessarily representative of car own-
ers in general, it was considered to be reasonably heterogeneous in re-
spect to the characteristics studied. also, as it was not presumed that
the returns would be of sufficient quantity to allow the claim of random
sampling in tense of the respondent group, it did not matter too much
what differences existed between this population and the total pOpu-
lation of car owners. It was felt that getting sufficient heterogeneity
for a meaningful analysis would provide enough information to indicate

whether or not a more adequate technique of sampling would be worthwhile.

W

The Form a questionnaires which were returned were scanned upon
arrival for canpleteness and the responses tabulated for each feature
and characteristic. a scoring system was set up to give a single value
for the responses to each feature on each section of the questionnaire.
For example, the set Sat isfied-lmportant was Judged to represent the
strongest degree of satisfaction with a present car feature, while the
set Dissatisfied-Important was Judged to represent the strongest feeling
of dissatisfaction. The remainder of the response sets to present car
features were placed between these extremes in the order Judged to be
most indicative of a continuum of feeling strength ranging from most

satisfied to most dissatisfied. The satisfaction scale was used first,

14

ordering the responses Satisfied, ?, Dissatisfied. The response on the
importance scale was then used to order responses within each satis-
faction scale, and numbers assigned to each combination, giving a 9 to
the combination Satisfied-Important, 8 to Satisfied-t, 7 to Satisfied-
Uhimportant, 6 to ?-lmportant, 5 to ?-?. 4 to ?-Unimportant, 3 to Die-
satisfied-Unhlportant, 2 to Dissatisfied-?, and l to Dissatisfied-
Important.

a similar scoring system was used on the three other sections
having two scales, ordering first along the right-hand scale and using
the importance scale to order within each of these categories. Thus,
every respondent had recorded a single number from 1 to 9 for each
feature on these sections which.told what both of his responses were to
the feature and which represented as well a measure of the strength and
direction of his Opinion, 5 being considered the neutral or no opinion
point. These scores were then summed for each of these sections, this
total being considered the best estimate of a respondent's overall
Opinion about his satisfaction with his present car (Present Car score),
his wants or needs on his next car (Next Car score), his approval of
trends (Trends score), and his concern with safety (Safety score).

On the paired-comparison (postures) section, the number of times
each feature was picked over a paired feature was recorded, ordering
for each respondent each of the features within each set. There were
two such sets of five features each.

For the Statements section, each.response was assigned a score
from 1 to 5 in such a way as to make 5 mean a favorable and 1 mean an

unfavorable response.

‘ A'" t‘

rwr.V-.

 

15

The information available allowed the features on all six sections
to be ordered in terms of the Opinion of the respondents. That is, a
feature could be stated to be more or less satisfactory, wanted,
approved or contributing to safety than another feature in the same
section. Also, the features in each set in the paired-comparison
section could be ordered in terms of being more or less liked than an-
other feature in the same set. Each item on the Statements section
could be ordered in a single continuum, and the amount of agreement of
the group‘on am one item could be easily determined.

The total score obtained on each of the first four sections pro-
vided a method for ordering the respondents in terms of their overall
satisfaction with their present car, their degree of wanting changed
features on their next car, their approval of trends in cars and their
feeling of the contribut ion of changed features to safety.

As there was no reason to believe that the arbitrary assigment
of numbers to the set of responses provided an equal-interval scale,
but'was at best a fairly accurate device for ordering (the strength of
the Opinion, it was felt that the total score derived from these num-
bers would provide only a ranking of the respondents. For this reason,
non-parametric tests of the relationship between both total scores and
individual feature scores were used.

In order to examine the relationship between the scores and certain
characteristics of the car owners and their car, a rank-order coefficient
of correlation (r.) was computed. The relationships examined by this
test were those between the total scores on each of the first four

sections (Present Car, Next Car, Trends and Safety) and the age, height

an! weight of the owners, and the yearly distance driven. The inter-

16

relationships between each of these scores and between each of these
characteristics were also examined by this test.

The relationships between these same scores and the sex, marital
status, present car age, and car buying plans of the respondent were
examined by means of the Mann-Whitney U-test.

The relationships between these same scores and car loyalty, make,
model and company of the respondent's present car were examined by the
use of the Kruskal-Wallis H-test. (Loyalty is defined as the expressed
intention of an owner to buy a new car of the same make-“Loyal, or of a
different make--Not Loyal).

The relationships between the way in which owners ranked each of
two sets of features on the paired-comparison section and certain other
characteristics were examined by means of the coefficient of concordance
test (We) or the rank-order correlation (r.) as a measure of the degree
of agreement between or among groups. The characteristics studied were
the age, height, weight, and loyalty of the respondent, yearly distance
driven, and the make, model, company, and age of the respondent's car.

The relationships between the responses to each itan in the State-
ments section and each of the scores derived from the first four sections
of the questionnaire were examined by a Chi Square test. The responses
to each item were combined into agree and disagree categories, with the
7 response being combined with the category representing a negative
attitude. That is, if the stat-nent was positive, such as "my present
car is well designed", ‘2 responses would be combined with D and SD as
indicating dissatisfaction with the present car. After combining the

responses in this manner, two by two tables (Upper and Lower 27% score

17

groups and agree and disagree response groups) were fanned and chi
Square computed, using Yates correction for continuity, except that, in
those tables where there were expected cell frequencies of less than 5,
Fisher's exact test of probability was used.

The relationships between the scores obtained on each feature in
each of the six sections of the questionnaire and the characteristics
of the car owners or their cars were examined by a two way non-para-
.metric analysis of variance technique suggested by Edwards (6 ). This
allowed the computation of the coefficient of concordance (We) as a
measure of the relation between a characteristic and a feature holding
another characteristic constant. The characteristic groups used for
this analysis were make and model, company and model, height and weight,
and age and distance. Car age and plan to buy were also used, but since
plan to buy only gave two categories, yes and no, a rank-order co-
efficient was run between the plan to buy groups matched on car age.

as it was not considered feasible to analyze all of the features,
it was decided that it would be most advantageous to study features felt
to be most likely to show a relationship and which could be compared
with other sections and with the other form of the questionnaire. The
features chosen were all the features on the paired-comparison, as they
were both from other sections and common to both forms, and two features
which were carried through all of the first four sections (horsepower
and ease of ride). Certain other features (two from each section) felt
to be related to certain of the characteristics and which had a reason-
able spresd of responses for the total respondent group, were also

included.

FIEDIN GS

Table 1 presents a summary of the results obtained from the single
mailing of the 400 Form a questionnaires.

Table 11 presents a sumnary of the relationships detemined to be
significant between the scores derived from.the several sections of the
questionnaire and certain characteristics of the car owner or his car.

Table III presents a summary of the relationships between the
scores derived from four of the sections of the questionnaire and the
item by item responses of car owners to the statements section of the
questionnaire.

Table IV presents a summary of the relationships determined to be
significant between the responses to certain features and certain
characteristics of the car owner or his car with another such charac-

teristic held constant .

TABLE!

SUM-{HY OF RETURNS TO MILLING OF
m A OF THE QUESTIONNAIRE

 

 

Porm.a Questionnaires Number Percent
Hailed 460 100%
Returned undelivered 12 3.0
Returned uncompleted 14 3.5
Returned completed 122 30.5

 

Total returned 148 37.0

TABLE II

SUMMARY OF SIGNIFICANT RELATIONSHIPS OBTAINED BETWEEN SCORES
DERIVED IRON EACH OF FOUR SECTIONS OF THE QUESTIONNAIRE AND

CERTAIN CHARACTERISTICS OF THE OWNER OR HIS CAR

 

Relationship between and

 

Score on Score or Characteristic Test Result
Present Car Owner loyalty H 11.65“
Present Car Next car score ra -.21'
lext Car Owner age re -.29*‘
Next Car Owner age H 7.16‘
Next Car Distance driven r8 .20‘
next Car Trends score r‘ .63“
Next Car Safety score rI .50“
Trends Owner age r8 -.30"
Trends Owner age H 6.05‘
Trends ledel owned H 12.08‘
Trends Safety score ra .57"
Safety Marital status U 1.93‘

‘ Significant relationship at 5% level.
" Significant relationship at 1% level.

Note: r. indicates rankrorder correlation test.
B indicates Kruskal-Wallis H-test.
U indicates Kenn-Whitney U-test.

TABLE III

amount or REIATIONSHIPS BETWEEN UPPER an Iowa 27% scone GROUPS

SECTION OF THE QUESTIONNAIRE

AND ITEM BY ITEM RESPONSES TO THE STATEMENTS

 

Item Score Group
Present Car Next Car Trends Safety
x2 p 12 p x2 p x2 p
1 .57 .55 .54 .56
2 2.27 .54 2.17 .55
3 4.25 .05 .36 .08 2.38
4 .36 2.71 .10 ‘ .237 .11 '
5 .00 .00 ‘ .307 ‘ .355
6 .06 1.51 .96 3.90 .05
7 .33 4.87 .05 7.15 .01 7.73 .01
8 12.52 .01 2.38 .00 .00
9 1.53 2.17 6.09 .02 .96
10 .09 6.49 .02 .08 .85
11 .00 .89 10.15 .01 4.67 .05
12 10.15 .01 ‘ .074 ‘ .237 .11
13 .26 .00 3.61 .10 5.39 .05
14 .08 2.74 .10 4.21 .05 5.26 .05
15 12.81 .01 3.61 .10 .28 .07
16 1.04 2.44 .06 5.20 .05
17 .00 ‘ .027 ‘ .028 ‘ .130
18 2.97 .10 1.56 1.01 .00
19 2.17 .00 .24 .24
20 1.62 2.38 1.62 1.31
21 1.84 8.92 .01 5.64 ‘.02 .27
22 .09 1.44 .85 1.62
23 .07 4.57 .05 8.43 .01 6.80 .01

k

’ Indicates use of Pinher's exact test of probability.

TABLE IV

SUINARY OP SIGNIFICANT REIAT IONSHIPS OBTAINED BETWEH CERTAIN
CHARACTERISTICS AND THE RESPONSE TO CERTAIN FEATURES, HOIDING
CONSTANT ANOTHER CHARACTERISTIC

 

 

Relationship between Holding Constant Result
Characteristic feature Characteristic We a
Make Acceleration (r) .nOdel .765*
.Age Leg room.(r) Distance .85"
.Age Luggage space (r) Distance .65‘
.Age .Power brakes (F) Distance .85“
.Age Smoother riding (T) Distance .85“
age smoother riding (S) Distance .65‘
Distance Head rocm (r) Age .65“
Distance .Power steering (F) 48° .65‘
Distance Easier ride (N) Age .65‘
‘histance Higher horsepower (T) Age .85“
‘Distance More horsepower (St) Age .65‘
Comparw Head room (1!) Model .65“
Company Low purchase price (P) Model .85“
Company Trouble-free operation (1) Model .85“
Model Exterior design (1) Company .85“
Model Acceleration (2) Company .65‘
Model Low purchase price (E) Company .65’
Height Head room (F) Weight .65‘
Height Trouble-free operation (F) Weight .65*
Weight Low purchase price (r) Height .65‘
Plans to buy More horsepower (S) Age of car rs=l.00

 

‘ Significant at 10% level.
“ Significant at 575 level.

NOIO:

The letter in parentheses after each feature indicates the section
of the questionnaire listing the feature:

N - Next Car section

T - Trends section

S - Safety section

I - Features section

St - Statements section

 

RESULTS

The analysis of the 1;; Form.a questionnaires shows the following
results in terms of the hypothesis stated.

Hypothesis 1. The hypothesis that there would be no relationship
between the way in which a car owner reSponds to a group of features and
any of certain characteristics of the car ouner or the car he owns, was
found to be untenable for the following relationships.

The score derived from the Present Car section was related to; the
loyalty of the Car owner, those who planned to buy a new car of the same
make (loyal) attaining significantly higher satisfaction scores on Eresent
Car than car owners planning to buy a new car of a different make (not
loyal); the Next Car score, car owners who attained high scores on the Ire-
sent Car section attained significantly lower scores on the Next Car section
than car owners who attained low scores on the fresent Car section; state-
ments 3, 8, 12, 15 and 18 in the statements section, car owners who attained
high scores on the Irescnt Car section responded significantly more often

in a manner indicating satisfaction with a present car feature as covered
in Statements 3, 8, 12, 5 and 18 than owners who attained low scores on
the tresent Car section.

The score derive; from the Next Car section of the questionnaire was
ralnated to; the age of the owner, older owners attaining significantly
lower scores than younger owners; the yearly distance driven, Owners driv-
ing there than 15,000 miles yearly attained significantly higher scores
that! owners driving less than 15,000 miles yearly; the score on the Trerfs
seCtion, owners who attained high scores on the Next Car section attained

Significantly higher scores on the Trends section than owners who

attained low scores on the Next Car section; statements

24

4., 7, 10, 12, 14, 15, 17, 21 and 23 in the statements section, owners
who attained high scores on the Next Car section responded significantly
more often in a manner indicating desire for a feature on their next

car as covered in statements 4, 7, 10, 12, 14, 15, 17, 21 and 23 than
owners who attained low scores on the Next Car section.

The score derived from the Trends section of the questionnaire was
related to; the age of the owner, older owners attained significantly
lower scores than younger owners; the model of car owned, owners of
convertible and hardtop models attained significantly higher scores than
owners of other models; the score on the Safety section of the question-
naire, owners who attained high scores on the Trends section attained
significantly higher scores on the Safety section than owners who
. attained low scores on the Trends section; statements 7, 9, ll, 13, 14,

17, 21 and 23 on the Statements section, owners who attained high scores
on the Trends section responded significantly more often in a manner in-
dicating approval of trends as covered in the foregoing statements than
owners who attained low scores on the Trends section.

The score derived from the safety section was related to; the
marital status of the owner, married owners attained significantly
higher scores than unmarried owners; statements 6, 7, ll, 13, 14, 16
and 23 in the statements section, owners who attained high scores on
the safety section responded significantly more often in a manner indi-
cating belief in the safety value of a feature as reflected in each
0f the foregoing statements than owners who attained low scores on the
“‘9‘! section.

No other significant relationships between the sections and
,.

25

characteristics were found. Tables 11 and III present a summary of

the significant findings. Tables XVI through XXII, Appendix D.

present the results of all the analyses of the data.

Hypothesis II. The hypothesis that there would be no relationship

between the way in which a car owner responds to any of certain features
and am of certain characterist ics of the car owner or the car he owns,
even when another such characteristic is held constant, was found to be
untenable for the following relationships.

The response to the feature "acceleration" in the paired-comparison

section was significantly related to the make of car owned, holding model

of car owned constant. The owners of Chevrolets and liercurys indicated

more liking of this feature than owners of Fords and Oldsmobiles.

The responses to the features "leg room", "luggage space" and
"power brakes" in the paired-comparison section, "smoother riding" in
the Trends section, and "smoother riding“ in the Safety section were

significantly related to the age of the owner, holding the yearly dis-

tance driven constant. Older owners indicated less liking of "leg room"

than younger owners. The older owners indicated more liking of ”luggage

Space" and "power brakes" than younger owners. The older owners also

indicated more approval of the trend "smoother riding" as well as judging

the safety value of "smoother riding" higher than the younger owners.
The responses to the features "head room" and "power steering" on

“10 Paired-comparison section, "easier ride" on the Next Car section,

"higher horsepower" on the Trends section, and "more horsepower" in

item 14, on the Statements section were significantly related to the

yearly distance driven by the owner, holding the age of the owner con-

26

stant. Owners driving greater distances indicated more liking of "head
room", more desire for "easier ride" in their next car, less approval of
the trend ”higher horsepower", and less desire for "more horsepower" on
their present car than owners driving a shorter distance. The results
on the liking of "power steering" were somewhat different, owners driv-
ing between 10 and 15 thousand miles yearly indicated more liking of
this feature than owners driving either more or less than this.

The responses to the features "head room", "low purchase price"
and "trouble-free Operation" in the paired-comparison section were sig-
nificantly related to the company manufacturing the present car owned,
holding the model of the present car constant. Owners of Ford products
indicated a greater liking of all these features than owners of either
General Motors or Chrysler rroducts. Owners of General.uotors products
also indicated more liking of "head room" and "low purchase price" than
owners of Chrysler products, but owners of Chrysler products indicated
greater liking of "trouble-free Operation" than owners of General Motors
products.

The responses to the features "exterior design", "acceleration"
and "low purchase price” on the paired-comparison section were signifi-
cantly related to the model of car owned, holding the company manu-
facturing the car constant. Owners of hardtOps indicated greater liking
of all these features than owners of either two-door or four-door sedans.

The responses to the features "head room" and "trouble-free
°P°r8tion" in the paired-comparison section were significantly related
‘0 the height of the mar, holding the weight of the owner constant.

Th9 tWiller owners indicated greater liking of both of these features

27

than shorter owners.

The response to the feature "low purchase price" on the paired-
comparison section was significantly related to the weight of the owner,
holding the height of the owner constant. Owners who weighed the least
indicated a greater liking of this feature than heavier owners.

The response to the feature "more horsepower" in the safety section
was significantly related to the owner's plans to buy a new car, when
matched on the age of the car owned. Owners who planned to buy a new
car indicated a higher safety value for this feature as the age of car
owned increased, while owners who did not plan to buy indicated a lower
safety value for this feature as the age of car owned increased.

There were no other significant relationships obtained between the
features examined and these characteristics. Table 1V presents a
sunnary of the significant findings under this hypothesis, and Tables
m through XXIX, Appendix 1:), present the results of all the analyses

Of th is data.

CONCLUSIOND' 4ND DISCUSSION

The major overall conclusion to be drawn from this study is that
most of the Opinions of consumers on specific and general features of a
product are related to the characteristics of the consumer and to the
characteristics of the product used (owned) by the consumer. While such
a conclusion is by no means original or startling, the further con-
clusion that these relationships can be specified and measured, and the
results used to improve either (or both) the consumer's opinion or the

product itself, should be of more interest to manufacturers and con-
sumers alike.

The specific Conclusions to be made on the basis of the results of
this investigation may be Considered tentative, due to the nature of the
sample and the questionnaire, but they do indicate the possibilities of
further investigation of these and other relationships with more
adequate techniques.

The first conclusion is made on the basis of the responses to the
B130c:1fic features. a visual inspection of the number checking each re-
SPOnse allows the conclusion that these car owners are, in general,
Quite satisfied with practically all features of their present car. The
feature indicated as most unsatisfactory is the purchase price, and even
this is not too great a source of dissatisfaction.

The desire for features on the next car is primarily a desire for
those features increasing the comfort and convenience of the driver,
With little desire for novel changes, or increases in power and speed.
This is also substantiated by the responses to the Trends section.

“3811:: the features most approved are those increasing comfort, while

29

horsepower and speed, and completely new features, receive little en-
couragement.

One other finding is obvious from these responses. Any feature
directly affecting the consumer's pocketbook such as purchase price,
miles per gallon, repair costs, etc., is definitely near the top or
bottom of the list on every section, depending upon whether it repre-
sents an inCrease or decrease in cost.

The results based on the relationships between the scores used as
measures of overall Opinion and other characteristics allow several
conclusions. First, satisfaction with the present car is independent
of all of the characteristics of the owner and his car, and relatively
unrelated to desires for approval of and judged value of change. This
indicates that satisfaction is wholly a function of the ability of the
Present car to fulfill the needs of the owner, and is not based upon
any comparison with possible changes in car features. also, the re-
lation between satisfaction and loyalty indicates that the owners
generalize their feelings from satisfaction or dissatisfaction with
their particular car to anticipation of satisfaction or dissatisfaction
with newer cars of the same make.

The high relationships between the desire for new features and
aPPI'Oval of trends and the age of the car owner indicates that the older
Where are less interested in and desirous of changes than the younger
0"Hull's. This may be due to a generally more conservative outlook on the
Part of older persons, but it may also represent to some extent a more
critical attitude toward the claimed benefits of changes, based on

19:1801' experience with such changes in cars.

The relationship between the Judged safety value of changes and the
marital status of the owner indicates a greater concern with safety on
the part of married persons. This may also be true of older persons,
their desire for safety negating their disapproval of changes in this
section. This may account for the lack of relationship between age and
safety, which should be present if the safety score represents a judge-
ment on the value of changes.

The results obtained on the analysis of individual features further
confirm these conclusions. Older owners tend to like those features
adding to comfort and convenience, while younger owners seem to be more
concerned with performance and styling.

While no specific analysis was made on the consistency and relia-
bility of responses, a feature by feature comparison of the results on
the two forms of the questionnaire indicated that the two respondent
groups were quite similar, both in terms of their and their car's
Characteristics and in terms of the types of responses to the cannon
features. There were no major differences noted in the proportion giv-
ing each response to each common feature and the two sets of features on
“1° Paired-comparison section were ranked in the same order by both
groups, The consistency of individuals was noted by a scan of the
questionnaires when received, with few respondents giving any contra-
di¢t0ry responses to the same feature on different sections. The con-

4:
$1813811cy of the group in this respect was checked by comparing the total
resporises to the features carried through all sections of the question-
hairs. As can be seen on the Form A Questionnaire in Appendix a, a

\ggture (i.e., horsepower) which was highly satisfactory on the present

51

car was least wanted changed on the next car (more horsepower), changes
in it least approved of as trends (higher horsepower), and changes in it
least valued as adding to safety (more horsepower).

The relationships between the scores on the first four sections and
the appropriate statements section items indicate further the consisten-
cy of the respondents, as those indicating higher satisfaction with
their present car also indicated higher satisfaction on items relating
to present car features. Thus, it may be concluded that the responses
were fairly reliable and valid and allowed measures to be obtained which
were also reliable, and to outward appearances reasonably valid, as in-
dicated by their relationships with other measures and characteristics.

rRho implications of these conclusions in reference to the present
structure of the automotive industry are quite complex. The basis of
the constant changes in car features is supposedly the thesis that
People want changes and won't accept a car that doesn't represent a con-
Siderable change from their present car. To make certain that people
follow this thesis, the advertising claims are designed to create
dissatisfaction on the part of the owner with his present car and create
de311‘s for the "new, improved, years ahead", characteristics of the most
re(tent. cars produced. But, the finding that satisfaction with the
preSent car is mainly independent of the desires for and interest in
Changes, seems to indicate that the attempt to create desire by making
the 0anr dissatisfied with what he has doesn't work, or at least any

“issatisfaction created is not matched by a corresponding increase in

“Sire for new features. It may be that, if the advertising claims do

\\ least create some dissatisfaction, this complicates the picture even

()1

to

more. If loyalty to a make or company arises from generalization of the
feelings toward the present car owned, increasing dissatisfaction will
lead to owner disloyalty. Thus, a company may, by creating dissatis-
faction, attract new buyers from the owners of different cars, but at
the same time indirectly forces owners of the cars made by that company
into the hands of competing manufacturers.

The findings on the relation of the desire for new features and
approval of trends with the age of the owner has additional implications.
The sales emphasis up to the present time has been directly upon chang-
ing design and increased performance, features most attractive to
younger consumers and least attractive to the older ones. Thus, the
major appeal has been aimed at a limited segment of the consumer market,
and this segment is not only limited in terms of numbers, it is even
more limited in terms of purchasing power, the younger consumer usually
having both a lower income and a greater variety of demands upon that
income than will be true when he reaches middle age. This factor may
account for the less rapid and less advertised changes in the more ex-
pensive cars, which depend more upon traditional prestige and comfort
factors for their sales appeal.

That the opinions of the consumer do have an effect upon the sales
appeal of certain features is evident by the recent happening in the
automobile industry. One of the findings of this study was that these
owners were, at least in relation to the other features listed, extremely
well satisfied with the horsepower of their present cars, had little
desire for higher speed or more horsepower, and did not approve of the

trends toward or safety Value of higher horsepower and acceleration.

all this after several years of constant increases in these features,
and loud claims as to the advantages of such increases, seems to in-
validate the claim of creating desire. This finding supports a state-
ment by the automobile manufacturers association announced in May of
this year that the manufacturers would no longer use in advertising cam-
paigns direct horsepower and performance claims, or take part in speed
races. although the reason stated for such a decision was that such
comparisons and activities had become too excessive, surely this would
not have mattered unless they were also affecting sales. It may be
presumed that the manufacturers have become avare of a similar lack of
interest in these features on the part of the total consumer market.
rhus, it is clear that such opinion questionnaire procedures can
contribute to the determination of which changes in features will be
acceptable or unacceptable to certain groups of consumers. It would
seem clearly worthwhile to find out in advance at least some indication
of the opinions to be expected on a contemplated change, rather than
going ahead with the change and learning afterward that it is not
acceptable. The use of Opinion questionnaire data should enable the
manufacturer to determine which changes are most desired by the majority
of the consumer market and to adjust his product accordingly, at least
to the extent of avoiding clearly unpopular changes. This would result
in a better market for the manufacturer, and a better product for the

e 0118111381“ 0

come "BS

The findings of this study point to the possibilities of further
studies in this area which would produce more significant and pertinent
data on the relationship of opinions to characteristics. Certainly the
findings of this study Cannot be taken as any final word on such matters.
The limitations of the study were many and important, but both the
method used and the results obtained have significance in further in-
vestigations.

It was unfortunate that the respondent sample was so limited in
size, both from considerations of sampling bias, and the resulting
limitation on the analysis. The two-way non-parametric analysis of
variance is a useful statistical test for such explication, but the
small number of categories which could be formed with the data available
did not allow a fair test of the method. With this limitation, which
meant that nearly perfect agreement by all groups would give a signifi-
cant result, 21 of the 182 tests were significant, which, while possibly
accounted for by random factors, are to some extent validated by the
findings of sterling in using Form B of the questionnaire (19). His
results on a similar analysis of features substantiated 6 of the 19
significant results on common items. Certainly an analysis based on
more extensive data would be desirable in substantiating these findings
and in determining relationships which may have been obscured by the
limitations of this stmiy.

There are, of course, limitations on any study of Opinions or
attitudes, especially when.questionnaire methods are used to obtain the

data. However, these problems and limitations are much better and more

fully discussed by the several authors mentioned in the Background

chapter than is appropriate in this paper. The use of scaling techniques

and analysis, rather shortly used in this investigation, offers another
more widely applicable method for similar investigations on a larger
scale.

While the results of this investigation are considered to be the
major relevant findings obtainable from the data, they by no means rep-

resent an exhaustive analysis. There remain many aspects and problems

to be studied. One of the most promising would be a validation study,

running an identical analysis on the data available on Form B. Certain-

1y if these findings are tenable and the assumption that the features
used are a sample of the population of features is correct, similar re-
sults should be forthcoming. also, a more complete and thorough com-
Parison of the respondents to the two forms should be made, as well as a
study of the cannon features and the features carried through several
sections.

It is further proposed that another study in this same area would be
advisable. With the information available from the two investigations
or the present questionnaire, and the use of the Questions sect ion as a

source of additional features of interest to consumers, it should be

fairly easy to prepare a revised questionnaire. The use of such a

Questionnaire with a more adequate sample, or as a tool to investigate
the Opinions of selected groups, should give additional infomation on

"19 findings of these investigations, and should reveafadditional

fac t ore as well.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

13.

14.

15.

BIBLIOGRAPHY

Blankenship, A. B., Consumer and Qpinion Research, New‘York;
Harper & Brothers, 1943.

Churchman, C. W., Ackoff, H. L., and Wax, u., Heasurement of Con-
sumer Interest, Philadelphia: university of Pennsylvania Press,

Clark, E. B., and Kriedt, P. H., WAn.4pplication of Guttman's
New Scaling Technique to an attitude Questionnaire", Educ.
Psychol. Heas., 1948, 83215-224.

Clausen, J. a. and Ford, H. N., "Controlling Bias 1n.ua11
Questionnaires", J. Amer. Statis. ass., 1947, 42:497-512.

Day, D., "Methods in attitude Research", Amer. Sociol. Rev., 1940,
5:395-410.

Edwards, a. L., statistical Methods for the aehgvioral §ciences,
New York: Rinehart, 1954.

Good, C. V. and Scates, D. B., gathods of Research, new York;
Appleton-Century-Crofts, 1954.

Guilford, J. P., Psychometric gethodg, (second Edition) New‘York;
HcGraw-Hill, 1954.

Guttman, L., "The Cornell Technique for Scale and Intensity
Analysis", Educ. & Psychol. Heas., 1947, 7:247-279.

Hyman, B., "Problems in the Collection of Opinion-Research Tata",
amer. J. sociol., 1950, 55 (No. 4) 362-370.

Kornhauser, 3., "Constructing questionnaires and Interview
schedules”. In Jahoda, Deutsch & Cock, Research.Hethods in Socigl

Belations; Eart 11, aglgcted gechgigggs, New York; Dryden Press,
1951, Chapter 12.

Housmar, Q., "Opinion-attitude Methodology”, Psychol. Bull., 1946,
45: 289-374.

Parten, M. B., fiurveyg, 29115 and fl mles, New York: Harper and
Brothers, 1950.

Hammers, H. H., lntroductio to in o and ttitude Measurement,
New York: Harper and Brothers, 1954.

Reuss, C. F., ”Differences between Persons Responding and not
Responding to a Mailed Questionnaire", Amer. sociol. Rev., 1943,
8: 433-438.

16.

17.

18.

19.

20.

21.

22.

23.

Shuttleworth, F. K., "sampling Errors Involved in Incomplete Re-
turns to Mail Questionnaires", J. Appl. Psychol., 1941,
25: 588-5910

Siegal, 5., Non-Parametric fitgtigtics, New York: KcGraw-Hill, 1956.

Stanton, 3., "Notes on the Validity of Mail Questionnaires", J.
Appl. Psychol., 1959, 23:95-104.

Sterling, R. L., "An Investigation of Certain Preferences of auto-
mobile Owners in Terms of their Similarities and Differences",
Unpublished Master's Thesis, Michigan state University, East
Lansing, 1957.

Waisanen, F. B., RA Note on the Response to a Hailed Questionnaire",
Public Opinion quarterly, 1954, 18:210-213.

Wallace, D., "a Case for and against Mail Questionnaires", Public
Opinion quarterly, 1954, 18:40-57.

Watson, R., "Investigations by Rail", mkt. Research, 1937, 5:11-16.

Weaver, R. G., "Proving Ground on Public Opinion", J. Consult.
Psychol., 1941, 5:149-153.

White, P., and White, M. "Research for Product Development". In

Blankenship, 5. (ed.), How :0 Conduct gogggmer 3nd 0212102 Begearch,
New York: Harper and Brothers, 1946, pp. 86-95.

LPPIIDII

A...’ :43] IL A

CONSUMER ATTITUDE SURVEY

Department of Psycholoy
Michigan State University

General Information
Please answer the following questions to the best of your ability.
Thank you.

1. Age__ Sex Married? Height Weight

2. Approzdmately, how far do you drive your car in a year?
3. Present car: Make , Model , Year

 

it. Last car: Make , Model . Year
(e.g., 2-door sedan, etc.)

5a. Do you plan to purchase a new car within the next 2 or 3 years?

Yes_ No . (If yes. answer 5b)

5b. Probable Make , Model , Year

 

(If you own more than one automobile , consider the one with which you are
most familiar when answering this questionnaire.)

lets: the figures entered on the questionnaire indicate the amber of
respondents selecting each response to each feature.

YOUR PRESENT GAR

Listed below are sixteen features of my; resent car. These are to be rated by
you according to the indicated scales. Rate all of the features on one scale first.
then rate them on the other scale.

 

On this scale indicate the degree On this scale indicate the degree.
to which 193; are satisfied with of importance to m of each one
each one of the features listed of the features listed below.
below. Check the column which Check the column which best in—
best indicates how satisfied 19;; dicates how important you consider
are with each of these features each of these features of your
of your present car. present car.
Satisfied 'l Dissatisfied Important 1 Unimportsnt
2‘ .111... .2... ...}... horsepower It A II A
'- 31, ° 1' windshield desigi st 101 10 11
5. m G u shifting lb 88 1? 17
.' 39L ‘. u head room 8 98 14 1‘
h 3...... .2... .32.... acceleration ‘ A A L
’0 .29.. .9... .19... visibility ‘ .13.... ...L. ...1...
" 22‘... .2... .19.. weight of ...-b ...‘3. 32.. .21...
’° .91.. .9__ _1_I___ seat comfort . 3531 .11.. ___t___
I. 593.. V .9... 1.1..- luggage Space ' .19.. .52... P...
“' ‘23.. .2... .33... ease of parking ' .321. .3... ..L
n. 10...... 3...... ' location of spare tire. i 32“ L
1" .2. 1°... a... purchase price .229. 3.1.. .1.
16. 1.0.2... ' 11 exterior desig 7‘ u 35
u’ B.,... .3... .” interior deaig: ‘7 '7 3‘
u. ...“... ...... 30 heater and defroster Ill ‘ O

 

w- Dentures ...... nap... on both ran a and n.- s e! the questionnaire.

|.' restores which are repeated in modified form on the Present Car, lest Car,
trade and safety .sssticns of the questionnaire.

Listed below are sateen possible features of M.
rated by you according to the indicated scales.

YOUR NEXT CAR

scale first. then rate them on the other scale.

On this scale indicate whether
of not :93 want each feature on

your next car.

Want
1. .fl.
30 .15...
5.4.

FPkkFPPFFFFFttFF~
I=I=|=IskkEEEELMiEFF

These are to be
Rate all the features on one

On this scale indicate how
important it is to 29;; that

193 have each feature on your

next

Don' 15 Want

better craftsmanship
sinner oar
higher speeds
gas turbine engine

power seats I

more miles per gallon I

automatic windows I
manual gearshift I
heavier car b
more visibility h
more acceleration b
more head room I

easier ride at

push-button shifting a

wraparound windshi eldIb

more horsepower 0

care

Important

I‘MHWPFFFFFFI‘FFF
FFFFFFFFFFFFISM‘F
FN‘I‘FI‘PPFFtFPFI‘F

? .

Unimportant

e.- features which appear on both tor-Lena For-I er the questionnaire.

b.- lectures eIieh are repeated in modified ton on the Present Ger. lent on.
Erode. ed Beat: sections of the «meet ionneire.

m

We are interested in your gpigions about current trends in automotive design.
Listed below are sixteen trends which are to be rated by you according to the
Rate all the trends on one scale first. then rate them on the

indicated scales .

other scale.

On this scale indicate how
19}; feel about each trend.

Like

s.”—

I...—

|e_‘..

tel...
5.119...
“.15...
70.8..
...“.
’0.”—
“M.”—

'8

Mkkl’ilfiBFFPFHHF
FH-I‘IBMMWFMFFFF

Dislike

Important

torsion-bar suspension A
higher repair costs A
dual exhaust system A

fancier interior desip A
smoother riding Ii
more accessories I

push—button shifting I)

larger rear fenders I

Hotels}:
FMHH‘FM‘I‘FPFFPH“:
IsleLislala-lslslslslslsHal-Ji-

wraparound windshi elds It

 

lower cars I
higher horsepower I.
longer cars I
hatter cars h
greater visibility t

faster acceleration b

I: it ‘5 I: I: is I:

less head room i

7

On this scale indicate the
importance of each trend to m.

Unimp or tent

‘0' tenures mieh appear on beth Ion A. and torn I of the questionnaire.

h- restores me. are repeated in modified form on the Present car. lent Oar.
We. and Safety sections ef the questionnaire.

A

SM‘E‘I'Y

Safetyin the automobile is important to all of us. and the manufacturers claim

many advances in this area. However. we are interested in obtaining W
as to the mtribution and W to safety for each of the features listed
below. Rate all the features on one scale first. then rate them on the other scale.

On this scale indicate the On this scale indicate the

cgptri'pgtig to safety of W of each feature

each feature. in tense of safety.
Increases ‘3 Decreases Important 2 Unimportant
Safety Safety
1. .2. __§__ 3;, pap-out windshi elds __§9___ AL 33“
£40.. .32. __1_§_, heavier cars h __§_5_ __§_l__ A
£29.. ..3. ...9... turn signals 332.. .3... .3...
‘0 J1, A A faster acceleration‘ A A 31,
'0 A ___‘_'__ A shock~absorbing bumpers __‘_1__ A i
.' __11_I_ ___Z_ ___9_ increased visibility ' ELL _,_!_ ...}.
'0 .JL, 3. ___1__, recessed steering columns __!_°_ it ___‘___
3‘ ___L A *1 less head room ‘ i i .32..
" ___2_ A A smaller wheels . 1 .31. 31,
10' .21. i ___2_ smoother riding a i A i 1L
1:. __‘_'__ _1{_ J: tinted windshields ‘ __‘__‘_’__ it, i
1"}; ___‘_’;__ _}_’__ push—button shifting '9 3L :1 f;
”'12. .33. .3. padded dashboards ' 139.. .32.. ...‘L.
is. .11 it ___E_ wraparound windshi elds “ 1;: 1L ,3};
15‘ 321, ___1__‘_ _____!_, seat belts ' it A i
I" i i __‘__1_ more horsepower I8 3 A it

In Features which appear on both For! A and For: B of the questionnaire.

b.- features which are repeated in modified form on the Present tar. heat Car.
frauds, and Safety sections of the questionnaire.

W

l‘or each of the following pairs of features check the one up, like the best.
Do this for every pair.

HHHHHHHHHH

lots:

leg room
acceleration

exteri or desiai
poser brakes

head room
trouble-ones operation
pushnbutton shifting
low purchase price
exterior desig
power steering

head room

power brakes

leg room

low purchase price
push—button shifting
acceleration
luggage space

power steering
lugage space

trouble—free operation

01'

01'
or

01‘

01'
or
or
01‘
01‘
01‘
01'
01‘
01‘
01'
01'
01'

01'

head room

low purchase price
push-button shifting
power steering
exterior desig:
acceleration

leg room

power brakes

1% IP80.
acceleration
push-button shifting
trouble-free operation
exterior design

power steering
luggage space

power brakes

head room
trouble-free operation
leg room

low purchase price

tabulation of responses to these items are smiled in Table XV.

Append in Us

HHHHHHHHHH

All of these futures are common to Forms A and I of the questionnaire.

SWMS

Indicate how strongly m or W with each of the following state»
ments “by encircling the appropriate symbol in front of each statement.

who symbols

indicate degree of agreement as listed below

>3 bS»B>S>:>:>:>2»8>t>:>.agpg»:egpgsxe. bgsgsges

6:

U§65686a6a6e6868686$686:6:6:636E686: 63636863

sn'

SA - suongly agree

A - agree

i - neither agree nor disagree

D .. disagree

8D ... strongly disagree
Adding more safety features has'not reduced accident injuries.
the advantages of smaller cars outweigh their disadvantages
The interior of my present car is very well desiped.

A car with the engine in the rear would be a great improve-
ment.

Automatic turn signals do not make driving any safer.

The newer the car. the more any needed repairs will cost.
lwould likealargergas tankinmynextcar.

It is an effort to get in or out of nor present car.
Experienced drivers would welcome sports~car handling.

I will demand better craftsmanship in my next car.

'Bhe trend toward longer care has gone too far.

My car is Just too hard to steer.

minted windshields make driving more dangerous.

I wish w car had more horsepower.

‘l'he visibility from my car is excellent.

If it were heavier. my car would be more satisfactory.

I would not pay extra for power seats in my next car.

the heater and defroster in my car could be greatly improved.
My next car has to be easier to park.

fiberglass bodies would not make cars any better.

My next car should make my present car look obsolete.

I will insist on a complete demonstration before choosing a
care

The controls and dials on my dashboard could be made less
complex.

-1..- - I \ A I , A A O l

- .
.
.' ' y _
. e
0
. . I
l l
.
‘.. ‘ .
, .
. Q
,1
. _e
' s
. e
‘ .
3.
9
.
e ‘3. '4 .
' " I
.
‘s ' i .
O
I , ‘ ' I
.
.V 1‘ ' e e
‘ I
.
e' u
.
. I
.
I
. ‘ t
e
.
u
.
C

 

 

a
I
s

4.

i$UESlflKnfl3

In answering the following questions. do not limit yourself to the features
mentioned in the questionnaire. Include anything that :93 feel is relevant to the

ques ti on.

Be as specific as possible. (Use the back of this page if you need

additional space. )

1.

2.

3.

1+.

5.

Ignoring the cost. what features would you like in your next car?

What features of today's cars do you think need the most improvement?

What features do you like most in your present car?

What features do you like least in your present car?

How did you like this questionnaire? Make any comments you wish.

THANK YOU FOR YOUR OSOPMATION

MICHIGAN STATE UNIVERSITY

CWAQUGEHRBANDAHUEJKMNGIONGTUWNNG

 

DEMEDGNTOPHWGKROGY

April 1! 1957s

Dear Sir:

The Industrial Section of the PsycholOgy Department at Michigan
State University is conducting a survey of the Opinions of automobile
owners. We think you will enjoy this Opportunity to indicate how you
feel about some of the features of your automobile.

Please fill out the enclosed questionnaire in accordance with
the directions at the top of each page. Winn completed, please mail
it back to us in the enclosed return envelope. .

Thank you for your oOOperation.

Sincerely yours,

may J. WW

James S. Karslaloe
Associate Professor

JSK:BS
Enc.

Appendix B
TABLE V

REPORTED PERM CHARACTERISTICS OF THE 122 RESPCUDENTS
TO FOR! A 01' THE QUESTIONNAIRE

 

 

Characteriet is I Charao terist ic
3.1 Height (inches)
Male 107 78 - 79 1
relale 15 76 - 77 l
74 - 75 e
liar-ital Status 72 - 73 13
Single 24 7o - 71 43
Iarried 98 es — e9 28
55 - s7 17
Age (years) 64 - 66 8
so - e4. 1 52 - 65 s
75 - 79 1 so - 51 2
7o - 7e 1
65 - 69 3 Weight (pounds)
60 - 64 5 230 - 239 2
55 - 59 1° 220 - 229 1
60 - 54 17 210 - 219 2
45 - 49 13 200 - 209 4
40 - 44 13 190 - 199 12
35 - 39 2° 180 - 189 1a
30 - 34 1" 17o - 179 22
25 - 29 14 150 - 159 22
20 - 24 7 150 - 159 16
140 - 149 10
Distance Driven per Year 130 - 139 4
(thousands) 120 - 129 5
35 - 39 2 110 - 119 -
so - 34 1 100 - 109 1
25 - 29 7
20 - 24 e
15 - 19 29
10 - 14 , . 49
5 - 9 25
o - 4 s

TABLE VI

REOBTED MAKE, MODEL, AND YEAR OF “WACTURE OP PRESEWT
AND PAST CAB 0WD BY THE 122 RESPONDENTS TO
FORM A 01' THE QUESTIONNAIRE

 

Present Car

 

 

 

 

 

__m L godgl g . Year anufaoture H
Buick 11 2-door Sedan 42 1957 11
Ch0vrolet 50 4-door Sedan 30 1956 24
Chrysler 2 Hardtop 17 1955 32
DeSoto 4 Coupe 4 1954 15
Dodge 1 Station Iagon 7 1955 16
Ford 29 Convert ib 1e 5 19 52 9
Kaiser 1 Hot Given 17 1951 6
Hercury 7 1950 4
Oldsmobile 17 1949 3
Packard 1 1948 -
Plymouth 7 1947 1
Peat iac 10 1946 l
Studebaker 1
Volkswagen 1
Last Car

lake I Hodel ll Year of lanufacture 1!
Buick 8 2-door Sedan 43 1956 3
Chevrolet 23 4-door sedan 41 1955 4
Chrysler 2 HardtOp 7 1954 9
Desoto 3 Coupe 6 1955 18
Dodge 4 Stat ion Wagon 4 1952 18
Ford 26 Convertible 3 ' 1951 19
Hudson 1 Not Given 12 1950 12
Kaiser 2 Ho Car 6 1949 13
Hercury 3 1948 11
Hash 5 1947 3
Oldsmobile 19 1946 1
Packard 2 1942 2
Plymouth 5 1941 1
Pontiac 11 1940 1
Studebaker 3 1936 1
Willys l 1954 1
Ho Car 6 1953 l

 

TABLE VII

BEFORTED MAKE AND HIDE]. 01‘ CAR WHICH THE 122 RESPONDENTS
Pm TO BUY WITHIN THE NEXT 3 YEARS

 

 

 

Hake H Model H
Buick 7 2-door Sedan 13
Chevrolet 18 4-door Sedan 25
DeSoto 2 Hardtop 10
Dodge 4 Coupe I
tore 10 Stat ion Wagon 12
Hercury 3 Convert ible 3
Hash 2 Not sure 23
Oldsmobile 13 Do not plan to buy 55
Packard 1
Plymouth 2
Pontiac 4
"Foreign" 5
not Sure 18
no not plan to buy 55

TABLE VIII

m IDIALT! 01‘ THE RISPONDENTS

 

 

Loyalty Classification H
Loyal 7 plan to buy sane make of car 50
f - do not plan to buy or uncertain

of make 48

Hot loyal - plan to buy a different
make of car 24

 

Appendix C
TABLE IX

SYSTEI UTILIZED FOR ASSIGNING SCORES TO EACH COMBINATION OF RESPONSES
WITH BEFEEEE TO EACH FEaTURE ON THE PRESENT CAR, NEXT CAR, TRENDS,
AND SAFETY SECTICNS OF THE QUESTIONNAIRE

 

Response checked on the
Satisfaction Scale

Present Car Features §core

Response checked on the Importance Scale

 

 

 

 

.naasassai 1. Sudanezisni
Satisfied 9 8 7
7 6 5 4
Dissatisfied 1 2 8
ext estures ‘ ' re
Response checked on the
W W
12232.11 1 3219221831.!
Want 9 8 7
7 6 5 4
Don't Want 1 2 3
new; upturn Sgorg
Response checked on the
M the t 9 81
man: 1 Grant
Like 9 8 7
7 6 5 4
Dislike 1 2 3

 

Response checked on the
M

Safety Fumes ficorg

 

Increase Safety
?
Docreasee Safety

 

8 III
.1
8 7
5 4
2 3

 

NUMBER OF BESPWDENTS ATTAINING EACH 30cm. FOR THE

TABLEX

COMBINATION OF RESPONSES CHECKED FOR EACH FEATURE
ON THE mm 0113 SETIG 0! THE QUESTIWIAIBR

 

 

 

 

Score
Feature ,
—‘1f""5_* ‘7"' 6’ <5 (' ‘3"“2“"I"'
Horsepower 5e 17 45 1 l l 1 O 4
Windshield Design 85 9 10 0 O O 1 1 16
Shifting 76 14 15 1 2 1 1 1 11
Ease of Ride 84 7 5 ll 2 0 0 1 12
Head 300:! 79 12 15 4 O 0 l 2 9
Aocelorat ion 69 13 17 3 4 2 O O 14
Visibi1ity 94 2 2 8 0 O O O 15
mum of Car 50 25 26 4 2 3 2 a 7
Seat comfort 84 9 6 7 1 O o l 14
Luggage Space 64 21 21 3 2 O l 2 8
Ease of Parkim 91 5 5 3 3 O 1 1 13
Location of Spare fire 50 1e 40 2 1 '3 1 4
Purchase Price 48 5 5 16 4 O O 42
Exterior Design 60 18 25 4 O o 2 10
Interior Delia: 61 24 22 0 3 2 4. 0 6
Heater an! Defroster 92 4 0 6 0 O O O 20

 

lots: for emple, line 1 cf the table shows that 54 respondents

checked the responses, satisfied-important on horsepower,

while 4 respondents checked the responses.
important, using the scoring systas presented in Table 111.

Unsst 18! led-

TABLEXI

NUMBER OF HESPONDENTS ATTAINING EACH SCORE FOR THE
COMBINATION OF RESPONSIE CHECKED FOR EACH FEATURE
on THE NEXT CAR SMTIOH OF THE QUESTIONNAIRE

 

 

 

Score
Feature 3

2 :8 L :L a L a 4.4.
Better Craftmanship 79 7 4 3 12 4 9 2 1
Smaller Car 17 6 2 0 1o 8 46 13 21
Higher Speeds 3 O 2 O 5 5 9O 9 8
Gas Turbine Engine 10 4 7 2 41 11 41 5 1
Power Seats 6 7 8 O 4 9 80 4 4
lore Ililes per Gallon 111 3 2 1 4 0 O 0 1
Autonst ic Windows 6 l 10 O 12 8 78 3 4
Manual Gearshift 16 7 2 O 16 8 42 10 21
Heavier Car 22 5 6 1 17 8 44 11 11
Here Visibility 62 l 4 8 18 5 ll 5 8
More Acceleration 28 3 4 13 6 48 9 9
More Hesd Roan 26 1 3 6 14 ‘ 7 44 6 15
Easier Ride 64 3 3 6 16 7 15 3 5
Push-Button Shifting 29 3 3 l 12 17 48 4 5
Irsp'cround Iindshie 1d 66 3 7 O 20 7 10 5 4
lore Horsepower 7 0 2 l 16 7 72 6 11

 

TABLE 111

NUMBER OF RESPONDENTS ATTAIHING-EACH SCORE FOR THE
COHBIHATIOH Ol‘ RESPWSEB CHECKED FOR EACH mm
on THE TRHDS SECTION OF THE QUESTIONNAIRE

 

 

 

 

Score
{gature
2 J 7 i 2 A.____=.5___2___L_
Torsion-bar Suspension 57 12 9 1 20 17 5 l 0
Higher Repair Costs 0 O O 2 2 0 7 7 104
m1 Exhaust System 26 7 5 o 20 22 32 2 a
rancier Interior Design 14 3 l3 3 12 26 45 2 4
Smoother Hiding 95 8 7 1 7 4 O O 0
Here Accessories 4 2 9 1 15 17 61 3 lo
Pushrbutton Shifting 25 5 9 O 15 21 37 4 6
Larger Hear renders 4 2 8 0 10 15 68 2 l3
wraparound‘Iindahields 72 2 6 l 14 9 11 1 6
Lower Cars 38 7 13 O 18 12 23 2 9
Higher Horsepower 13 1 4 3 12 16 55 4 l4
Longer Cars 11 3‘ 6 o 11 11 59 5 16
Heavier Cars 24 8 ll 2 l7 14 35 3 8
Greater Visibility 98 1 4 3 12 1 1 O 2
raster Acceleration 39 3 7 2 19 ll 31 3 7

Less Head Room 1 0 8 1 15 4 27 12 54

—_

TABLE XIII

mm or WTS manna non soon: FOR 232
COIBIIATIOI or RESPONSES cancxxn ran EACH FEATURE
on THE SAEETY snorlal or THE QUESTIOIIAIBE

 

 

 

 

Score
£16.19”: - .

9 8 7 6 5 4 3 2 1
Pop-out Windshields 52 12 4 O 30 9 8 1 6
Heavier Cars 49 lo 10 0 26 13 6 2 6
turn Signals 117 l 2 1 l O 0 0 0
Pastor Acceleration 4O 4 3 2 27 5 25 4 12
Shock-absorbing Dupers 59 10 6 2 29 12 14 O 0
Increased Visibility 110 4 l 1 5 1 O O O
Recessed steering Calms 76 8 5 4 28 O 1 0 0
Less Head Room 2 O 3 O 28 14 23 14 38
Smaller heels 13 4 4 O 56 l9 l5 4 7
Smoother Hiding 55 4 7 2 26 22 2 O 3
Tinted Windshields 56 8 4 1 26 14 8 2 3
Push-button Shifting 14 3 5 1 40 4o 13 2 4
Padded Dashb cards 100 6 5 O 7 3 0 1 O
Wraparound Windshields 75 9 6 0 22 4 3 O 3
Seat Bolts 93 5 3 O 9 5 6 0 1
Here Horsepower l4 4 4 2 25 12 37 5 19

 

TABLE XIV

TOTAL SCORE DISTRIBUTIOI 0’ THE 122 RRSPONDENTS TO PORI.A
Cl FOUR OF THIJIEASURES DERIVED ERG! THE QUESTIONNAIRE

 

 

 

 

[assure
Present lent Trends Safety
'1" '3' ' 'T' '3'

Total Score 9 r 9 r
140 - 144 11
135 - 139 8 1
130 - 134 16 1
125 - 129 15 5
120 - 124 17 l 9
115 - 119 16 1 11
110 r 114 12 2 3 20
106 - 109 10 4 4 23
100 - 104 4 5 5 17
95 - 99 5 8 12 14
90 - 94 2 l5 9 7
85 - 89 1 14 15 10
80 - 84 2 18 16 3
75 - 79 1 19 20

70 - 74 1 14 17

65 - 69 l 10 5

60 - 64 8 3

55 - 59 3 7

5O - 54 l 1

45 - 49 3 1
4O - 44

35 - 39 l

 

TABLE XV

RANK OF FEATURES 0N PAIRED-COMPARISON (FEATURES) SECTICI,
BASED UPC! THE NUMBER 0? TIMES EACH FEATURE WAS CHOSEN
OVER ANY OTHER EEATURE WITH WHICH IT WAS PAIRED

 

 

feature number of First Choices Hank
Set A
Dog roc- 407 1
Head room 282 2
Exterior desig 190 4
Puahrbntton shifting 121 5
Luggage space 218 3
Set I
Acceleration 129 5
Low purchase price 331 2
Power brakes 173 3
Power steering 153 4
Trouble-free operation 431 A 1

 

Hots: All the features in each of the two sets were paired with every
other feature in the acne set.

Appendix D
TABLE XVI

TESTS FOR THE PRESENCE OR ABSENCE OF A RELATICNSHIP

W THE SCORES ATTAINED on THE PRESMT CAR, NEXT

CAR. TREES. AND SAFETY SECTIONS AND THE INDICATE
CHARACTERISTICS OF THE 1133mm

 

goers on Present Car

 

 

Characteristics ngt Result
Age of respondents r.l r. = .114
Height of respondmts - r8 '.019
Height of respondents r. -.048
Distance driven per year r. .035
Sex of respondents U z :: . 15
Marital status of respondents U 1.25
Plan to buy new car U 1.14
Age of present car U 1.10
Loyalty to present make R H a 11.65”
lake of present car H 1.20
lodel of present car H 5.49
Ccnpany of present car H .65
Age of respondents H 2.00
Distance driven per year H .15

re Next
to t c 19!! Result

Age of respondents rs r, = -.290"*
Heiait of respondents r. -.015
Height of respondents r, -.066
Distance driven per year r" .201“
Sex of responduts U z : .30
Harital status of respondents U .56
Plan to buy new car U .52
Age of present car U .24
Loyalty to present car H H = 2.63
lake of present oar H 7.83
Hcdel of present car H 2.25
Company cf present car H 7.10
Age of respondents H 7.16"
Distance driven per year H 4.55

TABLE XVI (Continued)

fig 0:. 0!! grendg

gmggteristics

Age of respondents
Heigit of respondmts
Ieigit of respondents
Distance driven per year
Sex of respondents
larital status of respondents
Plan to buy new car

Age of present car
Loyalty to present make
lake of present car
Model of present car
Company of present car
Age of respondents
Distance driven per year

000 St

W

Age of respondents
Ueigit of respondents
'eigit of respondents
Distance drin per year
So: of respondents
larital status of reaponients
Plan to buy new car

Age of present car
Loyalty to present maloe
Make of present car
Hodel of present car
Company at present car
Age of respondents
Distance driven per year

mmmmmmccczdnu

.032
.093
. 139

.63
.15
1.02
1.28
.97
9.09

12. 08"

2.82
6.03"
2.20

 

' Significant at a; level
” Significant at 1% level

Note: r. indicates rank-order coefficient of correlation test.

U indicates lam-Whitney U-test of the significance of the

difference between the mean ranks of two groups ranked on

a common variable. .

H indicates Kruskal-Wallis H-test for the significance of the

differences among the mean ranks of three or more groups

ranloed on a common variable.

‘8
Set B

Features

TABLE XVII

Set A

on THE rim-cmmlsm (FEATURES) snc'rxon

TEST OF THE DEGREE 03‘ AW AMONG CERTAIN (HiOUPS
IN THE WAY IN WHICH {II-IE! BARK THE TWO SETS 01" FEAT

 

Cotdsdb 0
04.39

0 jfiDOLV/l
$5100in
>800

W OX 4: Q
$01.00

dualnv
099.0530

31
1.2%}
doom nv
006 .0094

9.95“ rm
sup—v34. £33m.

2&600
Louiekx m

.semNWAWQv:

.100! .00..

 

Respondents

who own:

- .860”

5

3
'b 3 e864“

2

1

1
liodel not given 1

1
1

2-door Sedan
4-door Sedan
Hardtop

Test
manufactured in:

Coupe
Station Wagon

Convertible
Respondents
who own cars

1a1a1a1a1a9u9~9~1n159~

354.444.434.51

54333334334.

42222111243

23555555525

44333332541.

55555455325

23444544443

3222222324‘

1.1.1A1a1a15151o1a159u

1957
1956
1955
1954
1953
1952
1951
1950
1949
1947
1946

'0 : e568“

WC 3 e551.‘

Test

SOt B

Features

Set A

TABLE XVII (Continued)

 

S. 538

 

Juvezu
..0énaanrmk. 1.1.1.1.11aiaioaalna.l.i.i.l 1.1aaiql
dcwsudknv u
tfiVQJAnV RoxaneAs9nz.4:5.4.n.4.o.e.n mm .4.e.4.o
nunJ’AYLAw .
.vnvaao .4_4.4.a.4.qno.sauaiuu:s:... .. «vague.»
.nuua.:w _m
anAYJJY>fiKw
«eat; “cannanc”onana.in2.ona.2ocna 9.9.9~9~
cJPnngznflflrQ nonono.a.5.5.a.a.n.4.5.4_a.a .o.o.n.4
.uuaanv
“vuAvuoa14 oaququauqu.s.9.aauauauau..¢u .o.e.2.4
42.520
c 3 .
fibk na.Jofimw .a.o.2.4.4.o.o.a.o.o.z.o.o.o w. .o.o.4.o
c9000 fl
Locolodakiu 44555333445432 e 4353
"

(coaryiflddék Hana.4.i.2ncnan4n2n2.eoana.e “success
.r29nr; meuJ 111111qu1a1a1a1a1a11111asa.i : .i.i.i.i
. v.

m. . m
I. r
.t .d 0 rem.
u: . r u an «no
a a .t 11 e m.ms n “I m.
o r .l .n .fl 0 I r
.1 o .o.d.t C .d u .u e
m roam u Oman» m. on»
n. on oflds m Wtdk ”d .m
u. 1m swuium«1wmu um mmmt
.R.m .m.C.CLm.D.E.u“I.U.r.P.P.S.v u» nunufiu

WC .2 e944“
- .858"

WC

WC 3' e802. ‘
- .967“

WC

Test
Test

60” - 67'

68" r 70”
71" ' 78'

Respondents whose

height is:

TABLE XVII (Continued)

 

 

Features
Set A set B
C .g d .e c
§ § cg Le? ., g g . ‘2’ . .2
o 2.9.,21—3'fi a fi’aisl'
0’s» ‘ \c d d 3 :; «u.L
4’ P “ e‘l‘ "' 3‘38 3G, H»
{D {3 q’ "" 9
Respondents 0, J. .J. 0, Q, l’ 0
whose weight
is:
100 - 155 lbs. 1 2 4 5 3 5 2 3 4 l
156 - 184 lbs. 1 2 4 5 3 3 2 4 5 1
185 - 239 lbs. 1 2 3 5 4 4 2 3 5 l
I'lt WC 3 e924'. WC .1 .858"
Respondents
whose age is:
20 - 34 yrs. 1 2 3 5 4 5 2 3 4 l
35 - 49 yrs. 1 2 4 5 3 3 2 4 5 1
50 - 89 yrs. 1 2 4 5 3 4 2 3 5 1
Test Wc = .924“ we a .828“
Respondents
whose yearly
distance
driven is:
0 - 9,990 miles 1 2 3 5 4 4 2 3 5 1
10,000-14,999 " 1 2 4 5 3 4 2 3 5 l
15,000-39,999 " l 2 4 5 3 5 2 3 4 1
Test we : .924“ we = .924“
Respondents whose
feeling towards
their present
car is:
Loyal l 2 4 5 3 5 2 3 4 l
? 1 2 4 5 3 4 2 3 5 1
Hot Loyal l 2 4 5 3 4 2 3 5 1
Test we a .967“ we I .924“

mm xvu (Continued)

 

 

Features
Set A Set 13
(i ' 0 ° ‘ - 5:. ~ -
Respondents j i \I V‘. “3 2 Q: Q \n 0‘
who are: 0' 4 .1 0/ Q, ,
Harried l 2 4 5 3 . 5 2 3 4 1
Single 1 2 4 5 3 4 2 3 5 1
Test ra : 1.00“ r8 : 90“
Respondents who:
Plan to buy 1 2 4 5 3 5 2 4 1
Do not plan
to buy 1 2 4 5 3 4 2 3 1
Test 1‘. = 1e00‘. 1'8 2 e70
Respondents
who are:
liale 1 2 4 5 3 5 2 3 1
Female 1 2 3 5 4 5 2 3 1
Test r.I : .90’ r" - 1.00"
Respondents who
attained scores
in thO:
Upper 27%-
Present Car 1 2 4 5 3 5 2 3 1
Lower 2731-
Present Car 1 2 4 5 3 5 2 3 l
c “ - 5.
Test r. - 1.00 ra .. 1.00
Upper 27%-
Next Car 1 2 4 5 3 5 2 3 1
Lower 27%-
Hext Car 1 2 4 5 3 5 2 3 1
Test r. : " ra - 1.00"
Upper 27%-
Trends 1 3 2 5 4 5 2 3 1
Lower 27%-
Trends 1 2 4 5 3 5 2 4 1
T..t I" 3 e70 1'3 3 090‘

TABLE XVII (Continued)

 

 

Features
Set A Set B
‘9. d. 4 v; «3 5 “9 ‘ v; 6

Respondents who .4 d: in a; j 4 0:3. :2, Q; g;
attained scores

in the:

Upper 27%-

Safety 1 2 3 5 4 5 2 3 4 1
Lower 27%r

Safety 1 2 4 5 3 5 2 3 4 1

Test r. I .90‘ r8 = 1.00"

 

' Significant at the 5% level.
“ Significant at the 1% level.

Note: We - indicates the coefficient of concordance; a test of the

degree of agreemnnt among the rankings of a set of
features, by three or mmre groups of Judges.

r. - indicates the rankrorder coefficient between the ranks

assigned a set of features by a group of Judges, and the

ranks assigned the same set by another group of Judges.

TABLEIXVIII

'Dlsrnnsncss IN RESPONSE 10 ITEMS In THE STATEMENT SECTION
Bsmwssu BESPOKDENTS IN THE UPPER.AND LOWER 27% on THE
PRESENT CAB.DISTRIBUTION or SCORES

 

 

Number of Upper 27% Lower 27%
statement Agree Disagree Agree - Disagree X2 p
1 22 11 19 14 .57
2 16 17 - 10 23 2.27
3 29 4 22 11 4.25 .05
4 6 27 8 25 .36
5 5 28 5 28 ' -
6 18 15 17 16 .06
7 9 24 7 26 .33
8 1 32 12 21 12.52 .01
9 12 21 17 16 1.53
10 26 7 27 6 .09
11 25 8 25 8 - -
12 1 32 11 22 10.15 .01
13 11 22 13 20 .26
14 9 24 8 25 .08
15 28 5 14 19 12.81 .01
16 10 23 14 19 1.04
17 3O 3 _ 30 3 , -
’13 14 19 21 12 2.97 .10
19 13 20 19 14 2.17
20 29 4 25 8 1.64
21 7 26 12 21 1.84
22 26 7 27 6 .09
23 9 24 10 23 .07

 

TABLE XIX

DIFFERENCES IN RESPONSE TO ITEMS IN THE STATMT SECTION
3mm RESPONDmTS IN THE 1113mm END LOWER 27% ON THE
NEXT CAB DISTRIBUTION or 8003118

 

 

Number of Upper 27% Lower 27%
statement Agree Disagree Agree Disagree x? p

l 17 16 20 13 .55

2 14 19 17 16 .54

3 27 6 25 8 .36

4 8 25 3 30 2.71 .10
5 2 31 2 31 -

6 13 20 18 15 1.51

7 13 20 5 28 4.87 .05
8 9 24 4 29 2.38

9 19 14 13 20 2.17

10 31 2 23 10 6.49 .02
11 25 8 28 5 .89

12 7 26 2 31 ‘ .074
13 9 24 9 24 -
14 12 21 6 27 2.74 .10
15 20 13 27 6 3.61 .10
16 14 19 8 25 ' 2.44

17 28 5 33 0 * .027
18 22 11 17 16 1.56

19 18 15 18 15 -
20 24 9 29 4 2.38
21 15 18 4 29 8.92 .01
22 28 5 24 9 1.44
23 14 19 6 27 4.57 .05

 

 

‘ Items with an expected cell frequency of less than five; tested by
Fisher's exact test of probability.

 

TABLEH

DIFFERENCES IN RESPONSE TO ITEMS IN THE STATEMENT SECTION
BETWEEN RESPONDENTS IN THE UPPER END LOWER 273; ON THE
TRENDS DISTRIBUTION OE SCORES

 

 

Number of Upper 27% Lower 27%
statement .Agree Disagree Agree Disagree X2 p
l 18 l5 15 18 .54
2 13 20 19 1.4 2. 17
3 26 7 25 8 .08
4 6 27 3 30 "‘ .237
5 1 32 3 3O ‘ .307
6 14 19 18 15 .96
7 15 18 5 28 7.15 .01
8 8 25 8 25 -
9 20 13 10 23 6.09 .02
10 25 8 24 9 .08
11 22 11 32 1 10.15 .01
12 6 27 3 30 ' .237
13 27 . 13 20 3.61 .10
14 11 22 4 29 4.21 .05
~15 22 11 24 9 ~ .28
16 12 21 11 22 .06
17 26 7 32 1 ‘ .028
18 22 11 18 15 1.01
19 15 18 17 16 .24
20 25 8 29 4 1.62
21 15 18 6 27 5.64 .02
22 25 8 28 5 .85
23 16 17 5 28 8.43 .01

 

"' Items with an expected cell frequency of less than five; tested
by Fisher's exact test of probability.

TABLE XXI

DIEEERENOES IN RESPONSE TO ITEMS IN THE STATEMENT SECTION
BETWEEN RESPONDENTS IN THE CIDER ..ND LOWER 27% ON THE
SAFETY DISTRIBUTION OE SCORES

 

 

 

 

 

Number of Upper 27% Lower 27%
statement Agree Disagree Agree Disagree X2 p

1 19 15 21 12 .55

2 15 20 15 17 .55

5 29 4 24 9 2.59

4 5 29 5 27 .11

5 5 50 5 29 7 .555 3 E
5 14 19 22 11 5.90 .05 .,
7 11 22 2 51 7.75 .01
9 5 29 5 29 -

9 19 15 14 19 .95

10 29 5 25 9 .95

11 25 10 5O 5 4.57 .05
12 5 27 5 29 .11 '
15 7 25 15 17 . 5.59 .05
14 12 21 4 29 5.25 .05
15 25 10 22 11 .07

15 17 15 9 25 5.20 .05
17 27 5 51 2 * .150
19 20 15 20 15 -

19 15 17 19 15 .24
20 25 10 27 5 1.51
21 12 21 10 25 .27
22 25 9 29 4 1.52
25 15 17 5 27 5.90

 

‘ Items with an expected cell frequency of less than five; tested by
Fisher's exact test of probability.

 

TABLE XXII

RANK-ORDER CORRELATIONS BETWEEN EACH 0? THE DISTRIBUTIONS OF

SCORES ON THE PRESENT 01R, NERT CAR, TREIDS

AND SAFETY SECTIONS OF THE QUESTIONNAIRE

 

 

 

 

 

Score Present Next
Car Car . Trends Safety
Score
Presmt Car 1000 'e206. -0016 e110
Next Car 1.00 .627“ .504“
Trends 1.00 .568“
Safety 1.00
‘ Significant at the 5% level.
" Significant at the 1% level.
TABLE XXIII
RANK-ORDER CORRELATIONS BETWEEN A NUMBER
OF CHERECTERISTICS OF THE RESPONDENTS
Characteristic Age Height Weight Distance
Characteristic ‘
.Age of respondent 1.00 -.229‘ .082 -.153
.Height of respondent 1.00 .6151” .185‘
Weight of respondent 1.00 .029
1.00

Distance driven per year

 

‘ Significant at the 5% level.
" Significant at the 1% level.

J" I"

 

TABLE XXIV

Z'VALUES FROM THE MENN‘WHITNEY U'TEST OF THE SIGNIFICANCE
OF THE DIFFERENCES BETWEEN THE MEAN BANKS OF TWO GROUPS
OF RESPONDENTS RANKED ON THE INDICATED VARIABLES

 

 

Groups Hale Married Plan to buy Owners of new cars

vs vs vs . vs

Female Single Don't plan to buy Owners of old cars
Variables :55
13. 1.27 .99 2.49‘ 1.20 ;
Distance 2.45‘ 2.09‘ 2.51* 5.22** ‘

 

‘ Significant at the 5% level.
“ Significant at the 1% level.

Note; For example, the z-score of 1.27 in the first column, first
line, indicates that males do not differ significantly from
fanales in respect to age; while the z-score of 2.43 in the
first colmnn, second line, indicates that males do differ
significantly from females in respect to distance driven.

Appendix E

TABLE XXV

MEAN SCORES ON THE INDICATED ITEMS, RANKED BY COLUMN AND ROW,
FOR A TWO-WAY NON-PMTRIC ANALYSIS OF VARIANCE

 

Paired Comparison (Features) section
Set A "leg room"

 

 

 

 

 

Model Z-Door 4-Door ' Sum of the
Sedan Sedan Hardtop Row Ranks
Make
3 3 4 10
“"1““ 3.33 3.00 2.29
2 4 3 9
Ford 3.40 2. 33 2.50
l 2 l 4
Mercury 4. 00 3. 50 3. 51
4 l 2 ‘7
Old smob ile 3. 17 3.60 3. 50
sum of the
001mm Ranks
We . e15 Wc - e426
0
Set A "head room"
2 2 4 8
Chevro let 2. 33 3. 29 l. 29
3 3 l 7
Ford 2.07 2.33 4.00
4 l 3 8
Mercury 1.00 3.50 2.50
l 4 2 7
Oldsmobile 3.00 2.00 3.50
sum of the
Column Ranks
We a .15 Wc :- .000
c

 

Note: Tables where a 0 appears in the lower right-hand corner cover
features comnon to Forms A and B of the questionnaire.

TABLE XXV (Cont inued)

Set A "exterior desim"

 

 

 

 

 

 

Model Z-Door 4-Door Sum of the
Make Sedan Sedan Hardtop Bow Ranks
4 2 l 7
Chevrolet 1.08 1. 71 2 3.29 l
2 1 2 5
Ford 1.4? 2.33 2 2.50 l
1 4 3 8
Mercury 2.50 . 50 3 2. 00 2
3 3 4 10
Oldsmobile 1.33 1.20 3 1.50 1
Sum of the
Column Ranks 10 5
We = .38 We : .255
c
Set A "push-button shifting"
3 3 l 7
Chevrolet .83 .86 2 2. 14 l
1 2 3 6
Ford 1.08 1.00 3 1.01 2
2 4 2 8
Mercury 1.00 .50 3 1.50 l
4 1 4 9
Oldsmobile .67 1.40 l .50 3
Sum of the
Column Banks 9 7
We a .03 We a .085 0
Set a "luggage space”
1 4 1 6
Chevrolet 2. 33 1. l4 3 l. 29 2
3 1 2 6
Ford 1.80 2.01 1 1.00 3
4 2 4 10
Mercury 1.50 2.00 l 1.50 3
2 3 3 8
Oldsmobile 1.83 1.80 2 .99 3
Smn of the
Column Banks 7 11
We 8 038 W0 ‘3 e213

TABLE XXV (Continued)

Set B "acceleration"

 

 

 

 

 

 

Model 2—Door 4€Door Sum of the
Sedan Sedan Hardtop Row Ranks
Make
2 1 l 4
Chevrolet 1.00 3 1.14 2 2.14 l
3 3 4 10
Ford .87 2 1.00 l .00 3
1 2 2 5
Mercury 2.00 2 1.01 3 2.01 l
4 4 3 ll
Oldsmobile .50 3 .99 1 .51 2
Sum.of the
column ranks 10 7 7
We = .15 we : .765' C
Set B "low purchase price"
1 3 4 8
ChOWOIOt 3e 17 2 3e 29 1 2e 71 3
4 2 2 8
Ford 2.73 3 3.33 2 3.50 l
2 1 1 4
Mercury 8.00 3 3.50 2 4.00 1
3 4 3 10
Oldsmobile 2.83 2 2.40 3 3.49 1
sum of the
column ranks- 10 8 6
W0 e21 'c = .383
Set B "power brakes"
4 2 4 10
Chevrolet 1.25 2 1.14 l .86 3
3 1 1 5
Ford 1.47 2 1.33 3 2.50 l
1 4 3 8
Mercury 3.00 1 . 50 3 1 . 50 2
2 3 2 7
Oldsmobile 1.50 2 .80 3 2.00 l
SHEIOf the
column ranks 7 10 7
We : .15 We = .255

7

TABLE XXV (Continued)

set B "power steering"

 

 

Model Z-Door 4-Door Sum of the
Sedan Sedan Hardtop Row Ranks

m

3 2 4 9
ChMOIQt e92 2 1001 1 e43 3

2 3 ' 1 6
Ford 1.40 2 1.00 3 1.50 l

4 4 3 11
Mercury .00 3 .99 l .50 2

1 l 2 4
Oldsmobile 1.50 2 2.20 l .51 3
Sum of the
column ranks 9 6 9

We :2 .15 WC 3 .596 C

 

Set B "trouble-free operation”

 

 

 

1 3 1 5
Chevrolet 3.67 2 3.43 3 3.71 1

3 4 3 10
Ford 3.20 2 3.33 l 2.50 3

4 2 4 10
Mercury 1.50 3 3.50 1 2.00 2

2 ‘ l 2 5
OldmmObile 3.60 2 3.61 1 3.50 3
Sum of the
column rank: 9 6 . 9

we 3 .15 we : .510 0
Present Car Section
Feature 1 "Horsepower"

1 3 2 6
Chevrolet 8.25 2 6.29 3 8.57 l

3 1 4 8
Ford 7.53 2 8.33 1 6.50 3

4 4 l 9
Mercury 4.00 3 4.01 2 9.00 1

2 2 3 7
Oldsmobile 7.67 3 8.20 1 8.00 2
sum of the
column ranks 10 7 7

W0 = 015 We 3 e085

 

Model

TABLE XIV (Cont inued)

Next Car Section
Feature 16 "More horsepower"

 

 

 

 

 

 

2-Door 4-Door Sum of the
Make Sedan Sedan Hardtop Row Ranks
3 l 1 5
Chevrolet 3.50 3 4.00 '2 - 5.86 l
4 4 2 10
Ford 3 .40 2 3. 33 3 3. 50 l
l 3 4 8
Mercury 6.00 1 3.50 2 3.00 3
2 2 3 7
Oldsmobile 3.67 2 3.80 l 3.01 3
sum of the
column ranks 8 8 8
WC : eoo WC 3 e255
0
Trends Section
Feature 11 "Higher horsepower”
2 2 1 5
Chevrolet 4.08 3 4.43 2 5.29 l
4 4 4 12
Ford 3.47 l 3.00 2 1.50 3
l 3 3 7
Mercury 6.00 l 3.01 3 3.50 2
3 1 2 6
Oldsmob ile 3. 67 3 5. 20 l 4. 50 2
Sum of the
column ranks 8 8 8
WC 3 e00 '0 . e596 c
safety section
Feature 16 "More horsepower"
1 3 2 6
Chevrolet 4.08 2 3.57 3 5.86 l
2 2 4 8
Ford 3.53 3 4e33 2 4e 50 1
4 4 l 9
Mercury 2.00 3 3.00 2 7.00 l
3 l 3 7
sum of the
column ranks ll 8 5
We a .50 We - .255

C

TABLE m (Cont inued)

statuents Section
Statement 14 "on horsepower"

 

 

 

 

Model 2-door 4HDoor Sum.of the
Sedan Sedan. Hardtop How Hanks

Make

2 2 . 2 6
Chevrolet 2.42 3 2.43 2 2.44

3 3 1 7
Ford 2.00 3 2e33 2 3.00

l 1 3 5
Mercury 3.00 2 3.50 1 2.00

4 4 4 12
Oldsmobile 1.33 3 1.80 l 1.50
sum of
column.ranks 11 6

W0 3 e38 WC = e595
Next Car Section
Feature 2 "smaller car"

3 4 3 10
Chevrolet 3.42 2 2.57 3 3.43

2 2 2 6
Ford 4.00 3 5.00 2 5.01

4 l 4 9
Mercury 2.50 3 6.00 l 3.00
I l 3 1 5
glidsmo‘oile 5.17 2 4.80 3 6.00
Sum of
column ranks 10 9

‘ It : .38 we = .340

‘ Significant at the ;;% level.

** Significant at the

level.

mam xxvi

MEAN SCORES ON THE INDICATED ITEMS, RANKED BY COMM AND RON,
FOR A TWO-WAY NON-PARAMETRIC ANALYSIS OF VARIANCE

 

Paired-Comparison (Features) Section

Set A "leg roomP

 

 

 

 

 

 

Company General Sum of
Motors Ford Chrysler new Ranks
Medel
2 1 3 6
2‘D00r 3““ 3038 3e47 2e5o 5
l 3 2 6
4HDoor Sedan 3.41 2.80 3.80 2
3 2 l 6
Hardtop 2.56 3.00 4.00 1
8mm of
column ranks 6
WC I 000 NC 2: e00
Set A "head room”
2 3 2 7
2€Door Sedan 2.54 1.94 1.50 3
1 2 l 4
4HDoor Sedan 2.74 2.80 2.12 3
3 1 3 7
Hardtop 1.78 3.25 1.00 3
Sum of
column ranks 9
WC = .65" We I .25 C
Set A "exterior design"
3 3 3 9
2"!)001' 8““ e96 1.59 1.00 2
2 2 2 6
4"D001' Sedan 1.53 1060 1e01 3
1 1 1 3
than of
column.ranks 8
WC 2 e35 WC ' e85’3

 

.‘Nate: Tables where a C appears in the lower right-hand corner cover
features common.to Forms.a and B of the questionnaire.

TABLE. XXVI (Continued)

 

Set A "pushrbutton shifting"

 

 

 

 

 

 

‘WC 3 e05

Company General Sum of
Motors Ford Chrysler Bow Ranks
Model
2 2 1 5
2€Door Sedan .92 3 1.00 2 3.00
3 3 ’ 2 8
4€Doer Sedan .76 3 .80 2 1.88
1 l 3 5
Hardtop 1.78 1 1.25 2 1.00
Sum of
column ranks 7 6
We = .05 We : .25
Set A "luggage space”
1 2 3 6
2‘Door Sedan 2.17 l 1.76 3 2.00
2 l 2 5
4+Door Sedan 1.76 3 2.00 2 2.12
3 3 1 7
Hardtop 1.22 2 .75 3 3.00
Sum of
column ranks 6 8
'c = .35 W0 = .05
C
Set B "acceleration
3 3 2 8
2€Door .96 3 1.00 2 1.01
2 2 3 7
4"D001' 1e18 1 1.01 2 052
1 l l 3
Hardtop 1.78 l 1.02 3 1.50
Sum of
column ranks 5 7

WC I e65t

 

TABLE XXVI (Continued)

 

Set B "law purchase price"

 

 

 

 

 

 

 

 

 

Company General Sum of
Motors Ford Chrysler Row Ranks
3 3 3 9
2€Door Sedan 2.71 2 2.76 1 2.50 3
1 2 ‘ 2 5
4#Door Sedan 2.94 2 3.40 l 2.51 3
2 1 1 4
.3322122—9 §a§2____2_____§a75 1 §a§§___§a
Sum of
column rank: 6 3 9
'0 a .85“ We II .65" C
Set B "power brakes"
l 2 2 5
2€Door Sedan 1.29 3 1.65 1 1.50 2
3 3 1 7
4€Door Sedan .94 3 1.00 2 1.88 1
2 1 3 6
Hardtop, 1.11 2 2.00 l 1.00 3
Sum.of
column ranks 8 4 6
'c = 035 WC 8 065‘ c
Set B "power steering"
2 l 2 5
Z-Door Sedan 1.25 l 1.24 2 1.00 3
1 2 1 4
4€Door Sedan 1.29 2 1.00 3 1.75 l
3 3 3 9
Hardtop_ .45 3 .99 l .98 2
Sum of
column ranks 6 6 6
'0 = .00 W0 . e05 0
Set B "trouble-free operation"
1 2 1 4
2€Door Sedan 3.71 2 3.00 3 4.00 1
3 l 3 7
4€boor Sedan 3.65 2 3.40 3 3.88 1
2 3 2 7
Hardtop 3.67 2 2.75 3 3.98 1
Sum of
column ranks 6' 9 3

We 8 .85” ‘We : .55

TABLE m1 (Continued)

 

 

m

Feature 4 "ease of ride"

 

 

 

 

 

 

WC 3 055

Company General Sum of
Motors Ford Chrysler Roe Ranks
,Model
3 2 . 2 7
B'Door Sod-n 7.54 2 6.47 3 8.90 1
1 3 3 7
4'13001‘ Sedan 8.29 l 5.00 3 8.00 2
2 1 i 4
Hardtop 7.78 3 8.00 2 9.00 1
Sum of
column ranks 6 8 4
'0 3 e35 '0 3 e25
Next Car Section
Feature 13 "easier ride"
1 2 1 5
2-Door sedan 6.46 2 7.47 1 6.00 3
3 1 l 5
4-Door Sedan 5.47 3 9.00 1 7. 75 2
2 3 3 8
H‘rdtog 5e 78 2 7e25 1 5.00 3
sum of
column ranks 7 3 8
WC 3 e55 WC - 025
Trade Section
Feature 5 "smoother ridirg”
1 2 l 4
2-1)oor Sedan 8. 71 2 8.53 3 9.00 l
2 l 3 6
4'3001' 3.0811 7088 3 8.60 2 8.88 1
3 3 2 8
Hardtop 7.67 3 8.25 2 8.98 1
Sum of
column ranks 8 7 3

 

TABLE m1 (Continued)

 

Safety Section
Feature 10 "smoother riding"

 

 

 

 

 

 

 

General Sum of
Company Motors Ford Chrysler Row Banks
11.9491
1 3 . 3 7
Z-DOOr SOdfln 5.83 1 6e29 3 6e50 2
3 1 1 5
4-Door Sedan 5. 53 3 7. 80 2 8.38 l
2 2 l 6
Hardtop 6.00 3 6.75 l 6.51 2
sun of
column ranks ‘ 7 6 5
'c I .05 We a .35 Q
lent Car Section
Feature 9 "heavier car"
2 1 3 6
Z-DOOF 5018!! 4‘58 2 5e65 1 2.50 3
3 2 2 7
4-Docr Sedan 4.47 l 3.00 3 3.38 2
1 3 1 5
m 1.89 2 2,59 3 §.00 _;
Sum of
column ranks 5 7 6
'0 8 .05 WC - e05
statements Section
Statement 11 "on longer cars"
1 3 2 6
2-Door Sedan 4.04 2 3.35 3 4.50 1
2 l 3 6
4-Door Sedan 3.53 3 4.80 l 3.88 2
3 2 1 6
Hardtop 3.44 3 4.00 2 5.00 1
Sum of
column ranks 8 6 4
We 3 e35 WC 3 e00

 

" Significant at the 10% level.
" Significant at the 5% level.

TABLE XXVII

m SCORES ON THE INDICATE ITEMS, RANKD BY COLWN AND ROW,
FOR A I'm-FAY NON-PARABTRIC ANALYSIS OF VARIANCE

 

Paired-Compar ison (Features) Section
Set A "leg room”

 

 

 

 

 

 

 

 

 

 

 

Height ' Sum of
60"-67" 68”-70" 71"-78" Row Hanks
was:
185-239 2 2 2 6
3.25 3 3.56 1 3.53 2
l 3 1 5
3 l 3 7
1529355 3W 0 . 'l l 345:0 2
sum of
column ranks 8 5 5
'0 . e25 '0 - .05 c
Set A "head ram"
1 3 2 6
185-239 2.50 2 2.11 3 2.65 l
3 1 3 7
156-184 2.00 3 2.15 2 2.40 1
2 2 1 5
1292-155— WS 1
Sum of
column ranks 7 8 3
‘ we I .65. WC : e05 c
Set a "exterior design"
3 3 l 7
185-239 1 .00 3 1. 22 2 1. 71 1
1 2 2 5
156-184 1.86 1 1.59 2 1. 55 3
2 1 3 6
M A2144 2 13.9.1___l__..§9 3
sum of
column ranks 6 5 7
We .7: .05 We = .05

 

late: Tables where a 0 appears in the lower right-hand corner cover
features coaunan to Forms A and B of the questionnaire.

mus mu (Continued)

Set A ”push-button shifting"

 

 

 

 

 

 

 

 

 

 

 

 

Sum of
Height 60"-67" 68"-70" 7l"-78" How Ranks
ML
2 1 3 6
185-239 1.00 2 1.44 .1 .41 3
3 2 2 7
156-184 .86 2 1.22 1 .75 3
l 3 1 5
199:199: .1 2 .0 3 00 1
Sum of
column ranks . 6 5 7
'0 . 005 We .- .05 9
Set A "luggage space"
2 2 1 5
185-239 2.00 l 1.67 3 1.82 2
3 1 2 6
l 3 3 7
199-199 2.19 1 1,31 3 1,50 2
Sum of
column.ranks 4 7 7
we = .25 We a .05 C
Set B "acceleration!
2 3 1~ 6
185-239 1.00 2 .33 3 1.35 1
1 l 2 4
156-184 1.71 1 1.44 2 1.00 3
3 2 3 8
M 095 W 099 L
sum of
column.ranks 5 8 5
'0 - 025 WC 3 e35 9
Set B "low purchase price”
2 2 3 7
185-239 2.50 2 3.11 1 2.29 3
3 3 2 8
1 l 1 3
100-155 2. 80 3 3. 19 1 3.00 2
sum of
column ranks 8 4 6
'0 3 .35 '3 = 055.

TABLE XXVII (Continued)

Set B "power brakes"

 

 

 

 

 

 

 

 

 

 

 

 

 

Sum of
Height 60"-67" 68"-70" 7l"-78" Row Hanks
was.
3 1 2 6
185-239 1.25 3 2.00 1 1.35 2
l 2 3 6
156-184 1.86 1 1.41 2 1.15 3
2 3 1 6
100-1§§ 1. 2 3 1,§0 ,1
Sum.of
column ranks 6 6 6
Fe = .05 We = .05 9
Set B "power steering"
1 3 l 5
185-239 2.00 l 1.11 3 1.24 2
2 2 2 6
156-184 1.29 2 1.33 1 1.20 3
3 l 3 7
199;1§§ 1.00 2 1.50 1 .150 3
sum.of
column ranks 5 5 8
'0 = .25 WC 3 e 5 Q
Set B "trouble-free operation"
3 2 3 8
185-239 3.00 3 3.44 2 3.71 l
2 1 2 5
l 3 1 5
m 2 oo 1
sum of
column ranks 8 7 3
'0 II .65" We :1 .ZSL
.Present Car Section
Feature 4 "case of ride"
1 3 1 5
185-239 8.00 2 7.56 3 8.65 l
2 2 3 7
156-184 7.86 l 7.74 2 7.20 3
3 1 2 6
100-155 7.09:: 3 1,75 1 z,§9 2
sum of
column.ranks 6 6 6

“2005

WC '3 005 c

 

TABLE XXVII (Continued)

Next Car Section
Feature 13 "easier ride"

 

 

 

 

 

 

 

 

 

 

 

 

Sum of
Height 60"-67" 68"-70" 71"-78" How Ranks
19.15111:
1 3 3 7
185-239 7. 50 1 5. 78 3 6. 76 2
3 2 2 7
2 l l 4
100-199 7. 1Q 3 z, 12 $2, 9. OO 1
Sum of
column ranks 6 8 4
WC 3 e35 WC — e25 c
Trends Section
Feature 5 "smoother riding"
3 1 3 7
185-239 7 . 75 3 8 . 89 1 7. 76 2
1 3 2 6
156-184 9.00 1 8.11 3 8.35 2
2 2 1 5
loo-155 2.65; 2 9.19 9 9.00 L
Sum cf
column ranks 6 7 5
We a .05 We : .05 0
Safety Sect ion
Feature 10 "smoother riding"
1 ‘ 2 1 4
185-239 8.00 l 6.55 3 6.59 2
' 2 3 2 7
156-184 7. 57 1 6.41 3 6. 5o 2
3 1 3 7
199-159 9,89 ,2____ 9,99 1 9.50 9
Sum of
column ranks 4 7 7
WC 3 e25 WC .- 025 S:
Present Car Section
Feature 5 "head room"
1 1 3 5
185-239 9 . 00 l 8 . 44 2 7. 47 3
2 3 l 6
156-184 8.99 1 7. 74 2 7.70 3
3 2 2 7
100-155 7.40 3 7.88 l 7.50 2
Sum of
column ranks 5 5 8
WC I .25 '0 8 .05 c

 

TABLE XXVll (Continued)

Present Car section
Feature 7 "visibility"

 

 

 

 

Sum.of
Height 60"-67" 68"-71" 727-78" Row Ranks
991ght
3 l 3 7
190-240 7.00 2 9.00 'l 6.47 3
2 3 2 7
150-189 7.43 3 7.44 2 7.80 l
l 2 1 4
loo-149 8.30 2 8.00 3 9.00 1
sum of
column ranks 7 6 5
We 8 e05 WC : e25
statements section
Statement 8 ”on case of entry"
3 3 2 8
190-240 2.25 1 1.56 3 2.24 2
1 2 1 4
150-189 . 2.57 1 2.19 3 2.55 2
2 l 3 6
loo-149 2.30 2 2.31 l 2.00 3
Sum of
column ranks 4 7 7
We .1 .25 WC : .55

 

‘ Significant at the 10% level.

TABLE MIII

MEAN SCORES ON THE INDICATED ITEMS, BANKED BY COLUMN AND ROW,
rm A TWO-WAY NON-PARAMETRIC ANALYSIS OF VARIANCE

 

Paired-Comparison (Features) Section

Set A "leg room"

 

 

 

 

 

 

 

 

Sum of
Age 20-34 35-49 50-84 How Hanks
M
l 2 2 5
15 ,000-39 ,999 3. 65 1 3. 41 2 3. 23 3
2 3 3 8
10 ,OOO-l4,999 3. 64 1 3. l7 2 2. 77 3
3 1 1 5
0-9 ,999 3. 62 l 3. 50 2 3. 25 3
sum of
column ranks 3 6 9
WC 3 085.‘ We 3 .25 c
Set A "head room"
1 1 l 3
15 ,000-39 ,999 2. 35 3 2. 71 2 2. 85 1
3 2 2 7
2 3 3 8
0-9 .999 2. 12 i1 JLOO .3. 1.75 :3
sum of
column ranks 7 6 5
V10 3 005 WC 3 065‘C
Set A ”exterior design"
2 3 3 8
15,000-39 ,999 1.94 l l . 29 2 .92 3
3 l l 5
10,000-14,999 1.55 3 1.67- 2 1.77 1
1 2 2 5
o-9,999 2.00 1 1.50 3 1.51 2
sum of
column ranks 5 7 6
WC I e05 VIC = e25 c
Note: Tables where a 0 appears in the lower right-hand corner cover

features cannon to Fame A and B of the questionnaire.

TABLE XXVIII (Continued)

Feature 13 "easier ride”

Next Car Section

 

 

 

 

 

 

 

 

sum of
Age 20-34 35-49 50-84 How'Hanks
pistance
2 l 2 5
1 2 1 4
10,000-14,999 8.18 l 6.92 3 7.46 2
3 3 3 9
O-9,999 6.12 1 5.67 2 5.17 3
sum.of
column ranks 4 6 8
We = e35 we I .65. 0
Trends Section
Feature 5 "smoother riding"
1 2 3 6
15,000-39,999 8.12 3 8.29 2 8.54 l
2 1 2 5
10,000-14,999 8.09 3 8.42 2 8.92 1
3 3 l 7
Sum of
column ranks 6 3
we I .85.”.I we I .05 c
Safety Section
Feature 10 "smoother riding" ,
, 1 1 3 5
15,000-39,999 6.35 3 7.00 1 6.71 2
2 2 l 5
10,000-14,999 6.27 3 6.58 2 7.23 1
3 3 2 8
O-9,999 5.75 3 6.50 2 7.08 l
sum.of
column ranks . 9 5
We - .65 We a .25L
Present Car section
Feature 1 "horsepower"
l 2 3 6
15,000-39,999 8.06 1 7.59 3 7.69 2
2 3 ' 1 6
l0,000-14,999 7.73 2 7.58 3 8.31 l
3 l 2 6
sum of
column ranks 6 8 4
WC = 035 We - .00 c

 

TABLE XXVIII (Continued)

Set B "power brakes”

 

 

 

 

 

 

 

 

sum of
Age 20-34 35-49 50-84 Row Ranks
mm
3 2 3 8
15,000-39,999 1.35 2 1.24 3 1.46 1
2 1 - 2 5
10,000-14,999 1.45 2 1.25 3 1.54 1
1 3 1 5
0-9!999 1050 2 lei-7 3 2.00 1
sum.of
column ranks 6 9 3
We - .85" We 3 e25 C
Set 8 "power steering"
2 2 3 7
15,000-39,999 1.24 1 1.18 2 1.08 3
1 1 1 3
10,000-14,999 1.55 2 1.25 3 1.92 1
3 3 2 8
0-9,999 .62 3 .67 2 1.17 1
sum.o£
column ranks 6 7 5
We 3 .05 wc = .65"c
set B "trouble-free operation"
1 3 2 6
3 1 1 5
10,000-14,999 3.45 3 3.67 1 3.62 ' 2
2 2 3 7
0-9,999 3.62 2 3.66 1 3.00 3
sum of .
column rank: 6 4 8
We 3 e35 We - e05 C
Present Car Section
Feature 4 "ease of ride"
3 2 1 6
15,000-39,999 6.65 3 8.00 2 8.85 1
1 1 3 5
10,000-14,999 7.55 2 8.08 1 6.85 3
2 3 2 7
0-9,999 6.88 3 7.33 2 8.08 1
Sum of
column ranks 8 5 5
We

3 e25 We 5' e050

 

TABLE XXYIII (Continued)

Set A."pueh-button shitting”

 

 

 

 

 

 

 

Sum of
Age 20-34 35-49 50-84 How Ranks
nigtggge
3 2 2 7
15,000-39,999 .35 3 1.06 2 1.15 1
1 3 - 3 7
10,000-14,999 1.18 1 1.00 2 .98 3
2 1 1 4
0-9.999 e38 5 1e17 2 1e42 1
Sum.ot
column ranks 7 6 5
W0 3 e05 WC 3 e25 L
Set A "luggage space"
3 2 7 3 8
15,000-39,999 1.71 2 1.47 3 1.77 l
2 1 2 5
10,000-14,999 1.82 3 1.83 2 2.08 1
1 3 1 5
0-9,999 1.88 2 1.33 3 2.17 l
sum.of
column ranke 7 8 3
W0 3: e65. WC - e25
Set B "acceleration"
1 3 2 6
15,000-39,999 1.18 2 1.12 3 1.23 1
7 3 1 '3 7
10,000-14,999 .45 3 1.33 l .54 2
2 2 1 5
0-9,999 1.00 3 1e17 2 1.33 1
sum of
column ranks 8 6 4
W0 3 035 We 3 e05 c
set B "low purchase price"
3 2 1 6
15,000-38,999 2.47 3 2.76 2 2.77 1
2 3 2 7
10,000-14,999 3.09 1 2.54 3 2.62 2
1 1 3 5
0-9,999 3.12 2 3.33 1 2.42 3
sum,or
column rank: 6 6 6

“38.00 Wc-.05c

 

TABLE XXVIII (Continued)

Safety Section
Feature 4 "faster acceleration"

 

 

 

 

sum of
‘139 20-34 35-49 50-84 How Ranks
Dietance
1 2 , 2 5
15,000-39,999 6.06 1 5.88 2 5.15 3
3 1 3 7
2 3 1 6
0-9,999 5.00 2 4.50 3 5.67 1
Sum of _
column.ranke 5 6 7
WC 3 005 We = e05
Safety Section
Feature 11 ”tinted Windshields"
2 2 " 1 5
15,000-39,999 7.00 3 7.06 2 7.15 1
1 3 2 6
10,000-14,999 7.64 1 5.46 3 7.08 2
3 1 3 7
0-9,999 5.75 3 8.00 1 6.75 2
Sum of
column ranks 7 6 5
Wt : .05 we : .05c

 

‘ Significant at the 10% level.
“ Significant at the 5% level.

TABLE XXVIII (Continued)

Next Car Section
Feature 16 "more horsepower"

 

 

 

 

 

 

 

 

Sum of
Age 20—34 35-49 50-84 Row Ranks
Distance
2 1 l 4
15,000-39,999 3.65 3 4.12 ,1 3.77 2
3 2 3 8
10,000-14,999 3.45 2 3.58 1 2.92 3
1 3 2 6
0-9,999 4.25 1 2.33 3 3.25 2
sum of
column ranks 6 5 7
We = .05 Wc - .35 C
Trends Section
Feature 11 "higher horsepower"
2 2 2 6
15,000-39,999 4.18 l 4.06 2 3.85 3
3 3 3 9
10,000-14,999 4.09 1 3.92 2 2.77 3
1 1 1 3
0-91999 4.25 3 4.50 2 4.58 1
sum of
column ranks 5 6 7
WC 3 e05 We - e851
Safety Section
Feature 16 "more horsepower"
1 1 3 5
15,000-39,999 4.58 2 4.65 l 3.62 3
3 2 2 7
10,000-14,999 3.09 3 4.62 l 3.69 2
2 3 1 6
0-9,999 4.25 1 4.00 3 4.17 2
sum of
column ranks 6 5 7
Wc : .05 We = .05 C
Statements Section
statement 14 "on horsepower"
1 1 l 3
15,000-39,999 2.53 2 2.18 3 2.54 1
3 2 3 8
10,000-14,999 1.91 1 1.88 2 1.77 3
2 3 2 7
0-9,999 2.00 2 1.83 3 2.08 1
sum of
column ranks 5 8 5
We 3 e25 WC - 065‘

 

TABLE XXVIII (Continued)

Next Car Section
Feature 16 "more horsepower"

 

 

 

 

 

 

 

 

Sum of
Age 20-34 35-49 50-84 How Ranks
Distance
2 1 1 4
15,000-39,999 3.65 3 4.12 - 1 3.77 2
3 2 3 8
1 3 2 6
0-9 ,999 4e25 1 2e 33 3 3.25 2
Sum.of
column ranks 6 5 7
We = .05 We = .35 c
Trends Section
Feature 11 "higher horsepower"
2 2 2 6
15,000-39,999 4.18 1 4.06 2 3.85 3
3 3 3 9
10,000-14,999 4.09 1 3.92 2 2.77 3
1 1 1 3
Sum of
column ranks 5 6 7
WC . e05 WC - e851
Safety section
Feature 16 "more horsepower"
l 1 . 3 5
3 2 2 7
10,000-14,999 3.09 3 4.62 1 3.69 2
2 3 1 6
0-91999 4.25 1 4.00 3 4.17 2
Sum of
column ranks 6 5 7
WC 3 005 WC 3 e05 c
Statements Section
statement 14 "on horsepower"
1 1 1 3
15,000-39,999 2.53 2 2.18 3 2.54 1
3 2 3 8
10,000-14,999 1.91 1 1.88 2 1.77 3
2 3 2 7
0-9,999 2.00 2 1.83 3 2.08 1
Sum.of
column ranks 5 8 5
We a .25 we = .65‘

 

TABLE XXIX

CORRELATION Bmwm THE RANKINGS FOR EACH 0’ A SET OF FEATURES
BY CAR OWNERS WHO PLAN TO BUY VERSUS THOSE WHO DO NOT PM
TO BUY, HSTCHSD ON THE AGE 0? TBS CAR.NOW OWNED

 

Paired Comparison (Features) Section
Set a "leg room"

 

 

 

 

 

 

Year of
Manufacture 1952 or befor. 1953'1954 1955 1956-1957
M
YOU 3e25 3 3e45 1 3e19 4 3e32 2
No 3.38 2 3.64 1 3.33 3 3.30 4
rs = .40 C
set A "head room"
Yes 2.25 3 2.45 1 2.12 4 2.36 2
No 2.62 1 2.00 4 2.50 2 2.40 3
1" 8 -e80 c
Set A "exterior design”
Yes 1.31 3 1.10 4 1.69 2 2.16 1
No 1.12 4 1.64 1 1.50 2 1.30 3
Set a “puah-button shifting"
Yes 1.00 2 1.05 1 .96 3 .56 4
No 1.25 3 1.09. 4 1.33 1 1.30 2
1.8 I ".80 ' C
Set A ”1uggege space”
YGB 2.31 1 1085 3 1e96 2 1e60 4
No 1.50 3 1.73 1 1.33 4 1.60 2
r 3 -e60 C
Set B "acceleration"
Yes .44 4 .80 3 1.04 2 1.88 1
No 1.00 3 .73 4 1.01 2 1.10 1
r3 - e80 C

 

Note: Tables where a C appears in the lower right-hand corner cover
features common to terms a and B of the questionnaire.

main: mix (Continued)

Set B "low purchase price"

 

 

 

 

 

 

 

Year of 1952 or before 1953-1954 1955 1956-1957
Manufacture
.Plans to buy
Yes 3.38 1 2.75 .2 2.46 3 2.16 4
No 2.75 4 3.09 2 3.33 1 2.80 3
rs 3 .e40 C
Set B "power brakes"
Yes 1.62 1 1.15 3 1.54 2 1.12 4
No 1.75 2 2.00 1 1.17 4 1.30 3
r3 . e00 C
Set B "power steering"
Yes 1.81 1 1.25 4 1.35 2 1.28 3
No .75 4 1.00 2 .83 3 1.01 1
1'8 :3 “.80 c
Set B "trouble-free operation"
Yes 3.06 4 3.85 1 3.62 2 3.44 3
No 3.75 2 3.18 4 3.67 3 3.80 1
r3 3 “.80 C
.Present Car Section
Feature 4 "ease of ride"
Yes 6.44 3 8.15 2 6.92 3 8.20 1
No 8.50 2 8.55 1 6.83 4 8.10 3
1', = em C
Next Car Section
Feature 13 "easier ride”
Yes 7.25 1 6.85 3 7.15 2 5.84 4
No 7.38 1 6.18 4 7.00 2 6.20 3
1‘. I e80 0
Trends Section
Feature 5 ”smoother riding"
Yea 8.69 1 8.55 3 8.68 2 7.84 4
No '8.00 3 8.55 2 7.33 4 8.90 1
1" 8 -080 C

 

TABLE XXIX (Continued)

Safety Section
Feature 10 ”smoother riding"

 

 

 

 

 

 

Y°8P Of 1952 or before 1953-1954 1955 1956-1957
Hanufaeture
Plans to buy
Yes 6.00 4 6.20 ' 3 7.04 6.52 2
No 7.25 2 7.09 3 6.83 7.30 1
r8 I -e4o C
Present Car Section
Feature 1 ”horsepower”
YGS 7e12 3 6e90 4 Bea? 7e96 2
No 7.75 3 8.27 l 7.83 7.90 2
1‘3 3 -080 C
Next Car section
Feature 16 ”more horsepower"
Yes 3.62 3 3.70 2 3.27 4.20 1
No 3.88 1 2.82 4 3.00 3.10 2
1‘8 3 e00 C
Trends Section
Feature 11 "higher horsepower"
Yes 3e56 4 4.00 2 3.58 4e84 1
No 5.00 1 3.45 4 3.50 3.60 2
rs : -040 C
Safety Section
Feature 16 "more horsepower"
Yes 2.31 4 4.10 2 4.08 5.08 1
No 5.25 1 3.91 3 4.00 3.70 4
1‘3 . -1000. C
statements Section
statement 14 "on horsepower"
Yes 2.19 3 2.25 1 2.08 2.24 2
No 2.38 1 1.64 3 2.00 1.70 4

r3 . -e50

 

TLBLS XXIX (Continued)

Trends Section
Feature 3 "dual exhaust system"

 

 

magma 1952 or before 1953-1954 1955 1956-1957
Plans to buy
Yes 5.3]. 3 30 75 I 4 5054 5e56
No 6.00 1 4.27 4 5.33 5.10
r8 2 e20
Trends Section
Feature 4 "fancier interior design"
Yes 4.38 3 3.95 4 4.85 5.28
No 4.38 3 5.09 1 4.50 4.20

 

‘ SignifiCant at the 5% level.

‘-)n x

HICHIGQN STQTE UNIV. LIBRRRIES

213

 

“II III” llllll Ill llllllll
9 1 9

312 30107