Office Technology and People

AN ERGONOMIC EVALUATION OF OFFICE CHAIRS

Martin G. Helander Sara J. Czaja Colin G. Drury James M. Cary George Burri
Article information:
To cite this document:
Martin G. Helander Sara J. Czaja Colin G. Drury James M. Cary George Burri, (1987),"AN ERGONOMIC EVALUATION OF
OFFICE CHAIRS", Office Technology and People, Vol. 3 Iss 3 pp. 247 - 263
Permanent link to this document:
http://dx.doi.org/10.1108/eb022651
Downloaded on: 30 January 2016, At: 21:31 (PT)
References: this document contains references to 0 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 447 times since 2006*
Users who downloaded this article also downloaded:
Ann Brooks, (1998),"Ergonomic approaches to office layout and space planning", Facilities, Vol. 16 Iss 3/4 pp. 73-78 http://
dx.doi.org/10.1108/02632779810205602
Jeffrey E. Fernandez, (1995),"Ergonomics in the workplace", Facilities, Vol. 13 Iss 4 pp. 20-27 http://
Downloaded by Temple University At 21:31 30 January 2016 (PT)

dx.doi.org/10.1108/02632779510083359
Sara J. Czaja, James M. Cary, Colin G. Drury, Barbara G. Cohen, (1987),"AN ERGONOMIC EVALUATION OF
TRADITIONAL AND AUTOMATED OFFICE ENVIRONMENTS", Office Technology and People, Vol. 3 Iss 3 pp. 231-246
http://dx.doi.org/10.1108/eb022650

Access to this document was granted through an Emerald subscription provided by emerald-srm:310011 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please visit
www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication
Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

 Office: Technology and People, 3 (1987) 247-262 247
Elsevier Science Publishers B.V., Amsterdam — Printed in The Netherlands

AN ERGONOMIC EVALUATION OF OFFICE CHAIRS

Martin G. Helander, Sara J. Czaja, Colin G. Drury, James M. Cary

Department of Industrial Engineering, State University of New York at Buffalo, Buffalo, NY 14260 (U.S.A.)

and George Burri

IBM Corporation, Real Estate & Construction Division, Stanford, CT, (U.S.A.)

SUMMARY

Ten ergonomic office chairs, chosen from a sample of eighty-

four, were evaluated in an ergonomic field study. Twenty office
employees used each of the chairs for one day. The chairs were
evaluated using four different subjective methods: body part dis-
Downloaded by Temple University At 21:31 30 January 2016 (PT)

comfort, general comfort rating, a chair feature evaluation check-

list, and a ranking procedure. The chair users generally had
difficulties in perceiving and expressing comfort and discomfort
which was required in the two first methods. The latter methods
produced more informative results including significant differ-
ences between the chairs. The study identified several distinct
features related to chair comfort including the design of the seat
pan, back rest, arm rests, and ease of adjustability.

INTRODUCTION

Ergonomic design of office environments is a concept that

has developed during the last ten years. Interest in office ergon-
omics first started in Sweden and Germany where it was sup-
ported by Federal Standards, which regulated the design of
visual display terminal workplaces [ 1 ] . The ensuing revolution
in design of office furniture has had a major impact on furniture
manufacturers; today ergonomic chairs alone represent a bil-
lion dollar business in the United States. It is not surprising
that many of the manufacturers have had difficulties in quickly
adopting the appropriate design principles. As a result, many
of the chairs advertised as "ergonomic" actually are not well
designed. This paper will first provide a brief overview of im-
portant design principles and research techniques and then re-
view them.
There are two major factors that effect the design of work-
 248

stations and chairs: task characteristics and operator charac-

teristics. The most important operator characteristics are
biomechanical and anthropometric factors, and most of the re-
search on chair design have dealt with these issues. Less is
known about the effect of task characteristics, such as different
preferences by managers and secretaries. Below we will briefly
review the research on biomechanics and anthropometry.

Biomechanics

Static work postures are rarely observed when humans per-

form tasks. Most systems in the body are in a dynamic equilib-
rium with continuous changes around an average value. Seated
operators have a tendency to keep moving around in the chair,
thereby relieving symptoms of postural strain [2]. Such pos-
tural changes may result in a massaging action of the discs in
the spine which increases the flow of nutrients to the disc [3].
The design of a chair should therefore make it easy to change
posture. For example, a flat seat pan is better than "an anatom-
ical" seat pan molded after the contours of the posterior, which
would inhibit movements [ 4 ] . Adjustable or flexible provisions
Downloaded by Temple University At 21:31 30 January 2016 (PT)

at the workstation, such as an adjustable chair or a detachable

keyboard may also encourage movements and are important in
preventing postural strain.
When a person moves from a standing to a sitting position,
the hip angle decreases from 180 degrees to 90 degrees. Anatom-
ically this movement is fairly complicated; about 60 degrees of
the bending takes place in the hip joint and the remaining 30
percent is due to a forward rotation of the pelvis which flattens
the lumbar curve of the spine. Most of the angular changes in
the lumbar takes place in the fourth and fifth lumbar discs [5].
Since these discs are involved in many lower back problems, it
is important to use a chair with a lumbar support which may
restore the lordosis (forward bend) of the spine [6]. The lor-
dosis of the spine is also restored if the chair seat back angle is
increased to 110-120 degrees, in which case the lumbar curve
unbends to its original position. The same effect can be ob-
tained by using a forward-sloping chair seat pan [7]. Chairs
with this design feature have recently appeared on the market.
Most of these chairs use a padded knee support to prevent the
operator from sliding out of the seat. Many users have com-
plained about the difficulty of ingress and egress and uncom-
fortable pressure on the knees [8]. At present, it would therefore
appear that a well-designed traditional chair provides the best
comfort, and accordingly this study concentrates on such chairs.
Most of the sitting pressure is absorbed by the bones at the
 249

base of pelvis (ischial tuberosities). The distribution of pres-

sure between the body and the seat changes constantly due to
the different postures that are assumed, such as crossing ones
legs, supporting an arm on a table or armrest, or leaning for-
ward or backward in the chair. In order to distribute the pres-
sure appropriately, the seat should be firmly upholstered, neither
too soft nor too hard, with a padding thickness of about 1.5
inches [9].
The seat edge of the chair must be well rounded and there
must be adequate clearance between the seat pan and the table.
Otherwise the seat may cut into the tissue of the underpart of
the thighs, which causes reduced blood flow in the leg.

Anthropometry

The body dimensions of the user population are of primary

importance for the design of workstations. In anthropometric
investigations, a person is measured in a standardized erect sit-
ting posture with joint angles at 0,90 or 180 degrees referred to
Downloaded by Temple University At 21:31 30 January 2016 (PT)

as an anatomical position. Unfortunately, this method does not

provide functional dimensions that describe the continuously
changing positions of the human body. Until there are better
translation techniques to convert static anthropometric meas-
ures to functional dimensions, the following guidelines may be
used to design an office workstation [9,10]:
(1) Sitting height and eye height should be reduced by three to
five percent from the erect sitting measures.
(2) For comfortable reach, forward and lateral reach measures
should be decreased by thirty percent from the erect sitting
measures.
(3) Adjustability of seat height should be determined by the
height of the lower leg (popliteal height) to provide a range
of adjustability from the fifth percentile (small) female to
the ninety-fifth percentile (large) male operator. This
translates into an adjustability range of about 16-20 inches.
(4) For the determination of the length of the seat pan, the
fifth percentile measure of the upper leg length (buttock-
popliteal length) should be used. Otherwise the seat pan
length would be too long for the smaller operator who would
not be able to use the lumbar support. This limits the seat
pan length to about seventeen inches.
(5) The minimum width of the seat pan and separation of arm-
rests should be determined by the ninety fifth percentile
female hip breadth (about 18 in.).
 250

Ease of adjustability

Adjustability of seat height is mandatory in most design

guidelines for visual display terminal workplaces [ 1 ] . Since
people differ in size and postural preference, adjustability of the
chair seat height appears necessary, despite the fact that few
experiments have proven how desirable and how necessary it
is. One study investigated preferences for chair height and
backrest height [11]. These heights could be set at three dif-
ferent levels. Overall, the medium setting was perceived as most
comfortable. However, many individuals remain insensitive to
adjustability features. Unless the features are explained and op-
erators trained, there is a great risk that they will never be used.
Chair adjustability may actually cause negative effects, since
maladjusted seating in an adjustable chair may cause more
problems than a non-adjustable regular chair [11]. There are
obviously trade-offs among the benefits of adjustability and its
cost; the actual use of adjustability features depends on the
awareness of the user population, the task performed, and the
ease of adjustment.
Downloaded by Temple University At 21:31 30 January 2016 (PT)

Methodologies for chair evaluation

There are three basic methods for evaluating chairs: the use
of anthropometric data to define chair measures, the use of fit-
ting trials to adjust the chair to the operator and experiments
involving subjection evaluations of chairs, either in a labora-
tory setting or at the workplace. Many research studies have
combined two or more of these techniques [12]. The use of
anthropometric methods for design of chairs has already been
discussed. In this section, we will comment on some of the dif-
ficulties on making subjective evaluations on comfort.
There are four major difficulties in using subjective evalua-
tions of comfort [13]. First, individuals may not be aware of
their feelings of comfort. Although varying degrees of discom-
fort may be perceived, it may be more difficult to sense varying
degrees of comfort. According to Branton, comfort is some-
times defined as the "absence of discomfort", implying that
comfort may be a nominal variable without varying levels. The
second problem is that individuals may have difficulties ver-
balizing feelings of comfort. Branton noted that comfort is a
"very primitive and ingrained feeling and not readily accessible
to introspection and verbalization" [13]. The third difficulty
is that individuals may have problems in attributing which of
the design features of the chair is the source of comfort or dis-
comfort. Finally, it is difficult for individuals to compare the
 251

comfort of different chairs, since it is difficult to memorize the

sensation of comfort sufficiently long for a comparison [13].
Since comfort is such an elusive variable and since individ-
uals have difficulties in producing valid assessments of comfort,
it should be expected that there may be a great amount of var-
iability in data for comfort assessment. In addition, there are
differences among individuals in body shape and task activities,
and it should not be surprising if one chair can be perceived as
both comfortable and uncomfortable by different individuals.
Some of this variability may be reduced by using pre-trained
subjects rather than untrained subjects. In addition to giving
more reliable results, pre-trained testers are more sensitive to
low levels of discomfort [14]. This was the approach taken in
this study.

Assessment of seat comfort

Branton [13] reviewed the literature on subjective assess-

ment of seat comfort and concluded that:
Downloaded by Temple University At 21:31 30 January 2016 (PT)

(1) Simple, relatively unstructured scales should be used.

(2) Subjects should indicate discomfort by body part.
(3) Checklists should be used to have subjects evaluate specific
features of the chair.
In this study, scales were therefore developed to evaluate these
features, see below.

OBJECTIVES

The objectives of this study were to evaluate ten ergonomic

office chairs and give recommendations which could be used as
a guideline for purchase of new chairs. Each chair was evaluated
with and without armrests for a total of twenty chairs.

METHOD

The evaluation was conducted in the offices of IBM Corpo-

ration at the Real Estate and Construction Division in White
Plains, New York. Twenty subjects, ten secretaries and ten pro-
 252
Downloaded by Temple University At 21:31 30 January 2016 (PT)

grammers participated in the study. These job types were se-

lected, since they constitute the target population for whom the
chairs were intended.
The secretarial population consisted of ten females ranging
in age from twenty-one to fifty-five years with an average age
of twenty-eight years. The average amount of experience in the
current job position was three years, and they were sitting for
about six hours per day. The programmer population consisted
of three males and seven females ranging in age from twenty
years to forty-one years, with an average age of thirty-one years
and two years job experience. The programmers spent about
five hours per day sitting. There were ten different makes of
chairs. These chairs were chosen from a sample of eighty-four
different chairs using criteria such as: appearance, availability
of a family of chairs, engineering characteristics, safety and du-
rability, ergonomics features and ease of adjustability. Since
there were two models of each chair, with or without armrests,
a total of twenty chairs was used. Each subject evaluated ten
chairs, five with armrests and five without. The order of pres-
entation of chairs was counterbalanced to control for order ef-
fects. Each day, a subject was given a different chair and asked
to sit in it for an entire workday. The subject was shown how
to use the adjustability controls to adjust the chair to the best
position.
To measure overall comfort, the general comfort rating
 253
Downloaded by Temple University At 21:31 30 January 2016 (PT)

(GCR) developed by Shackel, Chidsey and Shipley was used,

see Fig. 1. This scale has been validated for assessment of chair
comfort [15]. To measure body part discomfort, the scale de-
veloped by Corlett and Bishop was used, see Fig. 2. This scale
is easy to administer, and requires almost no training [16]. Fi-
nally, to identify what chair features were most important and
to obtain an overall assessment of comfort, a chair feature
checklist was used [8].
The general comfort rating and the body part discomfort scale
had to be completed three times per day, in the morning, just
before the lunch break and before the end of the day. The chair
feature checklist was completed only at the end of the day. Two
anthropometric measures were also taken, the popliteal height,
which has implications for seat height, and the buttock-poplit-
eal length which has implications for seat length.
Following the two weeks of evaluation period, the chairs were
lined up in a conference room and subjects were individually
asked to rank the chairs from best to worst (1-10) and assign
a single rank to each chair.
 254

Objective measurements

Following the evaluation, the dimensions of each chair were

measured using the chair measurement procedures outlined in
the proposed ANSI standard for design of VDT workplaces
[17]. These measurements were taken to assess the compli-
ance of the different chairs with the proposed ANSI standard.

RESULTS

Results of the evaluation are presented in the following for-

mat: a comparison of the chairs against the ANSI standard,
frequency of body part discomfort (FBPD), severity of body
part discomfort (SBPD), results of the chair evaluation check-
list (including several chair design features), results of the gen-
eral comfort rating (GCR), ratings of chair comfort, appearance
and ease of adjustability, and overall ranking of the chairs.
Downloaded by Temple University At 21:31 30 January 2016 (PT)

Comparison of chair measures against ANSI draft

standard

For each chair, measurements were obtained of seat height,

seat width, seat length, backrest width, lumbar height, seat pan
slope, backrest angle, and arm rest separation. The measures
were compared to the ANSI measures, see Table 1.
As seen in Table 1 most of the chairs complied with all re-
quired measures, except seat height, which typically does not
adjust low enough.

Frequency and severity of body discomfort

An overall measure of body part discomfort frequency

(FBPD) was computed by summing across the body part rat-
ings, see Fig. 2. In addition, an average severity of body discom-
fort (SBPD) across the body parts was computed. A 2 x 10 x 2
factorial analysis of variance was performed for FBPD and
SBPD. The factors were JOB TYPE with two levels (secre-
taries vs. programmers), ten different makes of CHAIR and
two ARMREST (armrest vs. no armrests). The analyses of
variance (ANOVA) showed no significant differences in body
discomfort for any of the independent variables.
Downloaded by Temple University At 21:31 30 January 2016 (PT)

TABLE 1

Comparison of chair measures of ten chairs to the ANSI draft standard

Chair dimension ANSI standard Average Number of

value or complying
value Mandatory range chairs
Seat height (in.) 16-20.5 yes 16.7-20.5 0
Seat width (in.) Min. 18.2 yes 18.8 10
Seat length (in.) Min. 15 no 15.9-17.0 9
Max. 17
Backrest width (in.) Min. 12 no 16.8 10
Lumbar height (in.) Min. 6 no 6.5-8.0 9
Max. 10
Seat inclination (°) Min. 0 no -0.2-4.8 9
Max. 10
Backrest angle (°) Min. 90 no 90-105.9 10
Max. 105
Armrest Min. 18.2 yes 18.6 9

255
separation (in.)
 256

General comfort rating

An ANOVA was performed for the General Comfort Rating

with the same factors as above. There was no significant differ­
ence in GCR at the end of the day as a function of CHAIR type.
However, there was a significant difference in rating of comfort
between JOB types ( F = 11.4, df= 1/160,p<0.001). On the av­
erage, programmers reported that they experienced more dis­
comfort at the end of the typical workday than did secretaries.
None of the interactions were significant with regard to this
measure, indicating that there were no differences in GCR rat­
ings between secretaries and programmers as a function of chair
type or armrests.

Evaluation of comfort measures

An ANOVA of the overall comfort rating of the chair indi­

cated that there was a significant difference among the CHAIRS
(F=2.15, df= 9/160, p<0.05). A Newman-Keuls test on the
means indicated that two of the chairs were significantly more
Downloaded by Temple University At 21:31 30 January 2016 (PT)

uncomfortable than the others, and one chair was rated to be

the most comfortable. Among the remaining seven chairs there
were no significant differences. The JOB typeXCHAIR type
interaction was insignificant, indicating that there was no dif­
ference in the pattern of secretaries' and programmers' ratings
of chair comfort. Likewise, the JOB X ARMREST and JOB X
ARMREST X CHAIR interactions were also insignificant.
These findings indicate that there was no difference between
how programmers and secretaries rate chairs with and without
armrests, and this finding was consistent across the different
chairs.
There was a significant difference among the CHAIRS with
respect to ease of adjustability (F=2.09, df= 9/160, p<0.05).
Two of the chairs (chairs Η and A) were easier to adjust than
the other chairs and one of the chairs (chair J ) was signifi­
cantly more difficult to adjust. There were no differences be­
tween secretaries and programmers in their ratings of ease of
adjustability as a function of chair type.
With respect to appearance, the results also indicated a sig­
nificant difference in rating according to CHAIR type (F= 2.43,
df= 19/160, p < 0.001). The Newman-Keuls test on the means
indicated that one of the chairs (chair J) was significantly more
displeasing than the other chairs. There was no significant dif­
ference between secretaries' and programmers' rating of chair
appearance.
 257

Relationships among t h e evaluation measures

In order to determine which chair dimensions are important,

Pearson correlation coefficients were computed among the
evaluation measures, see Table 2. All the chair dimensions
except seat height and seat slope were significantly related to
chair comfort. The most important chair dimensions were seat
shape, shape of backrests, and ease of adjustability. The same
measures were the most important for overall comfort.
To determine if body size had any impact on the rating of
difference chair features, correlation coefficients were com­
puted between the anthropometric measures and the ratings of
chair dimensions. One significant correlation was found be­
tween buttock-popliteal length and seat length (r=0.19,
p < 0.01). As expected, the subjects preferred a seat pan length
that was compatible with the buttock-popliteal length.

TABLE 2

Pearson correlation coefficients for chair measures

Downloaded by Temple University At 21:31 30 January 2016 (PT)

Chair measure Overall chair Overall

comfort appearance
Seat height r = 0.01 r = 0.17a
Seat length r= 0.25a r = 0.19a
Seat width r = -0.20 a r = -0.03
Seat slope r = 0.05 r = 0.11
Seat shape r = -0.68a r = 0.47a
Position of backrest r= 0.21a r = 0.23a
Shape of backrest r = -0.74 a r = -0.51 a
Seat cover friction r = -0.43 a r = -0.18 a
Ease of adjustability r = 0.48a r = 0.31a
Overall appearance r = 0.56a —
Armrest comfort r = 0.33a r = 0.23a
a
Indicates a significant relationship between the measures at the 0.01 level.

Overall ranking of chairs

After the conclusion of the evaluation, all chairs were dis­

played in a conference room, and the participants in the study
were invited one at a time to provide an overall ranking from
best to worst of the chairs. They were encouraged to sit in the
chairs, compare them to each other, and then fill out an evalu­
ation sheet. Sixteen of the subjects participated in this evalua­
tion, and the results are shown in Table 3. The distribution of
the rank orders in Table 3 is significant with Friedman's
χ2
= 33.9, p < 0.001.
 258
Downloaded by Temple University At 21:31 30 January 2016 (PT)

TABLE 3

Rank order for overall ratings of chairs (N = 16)

Chair Individual rankinga Average

rankb
A 5 1 5 1 5 1 1 1 1 5 4 2 1 3 7 9 3.2
Β 4 2 2 3 4 5 8 7.5 5 8 2 1 2 4 3 5 4.1
C 2 4 4 5 7 7 7 5.5 2 2 5 5 3 1 1 7 4.2
D 1 8 8 8 1 3 4 5 4 3 6 7 6 8 9 4 5.2
Ε 6 6 3 4 2 2 2 10 8 9 9 9 9 5 2 1 5.4
F 3 5 1 2 3 10 10 4 9 1 1 10 8 8 5 10 5.6
G 7 3 7 7 6 6 6 7.5 6 7 7 3 5 2 8 3 5.7
Η 9 7 6 6 8 8 5 5.5 3 4 8 8 4 8 6 2 6.4
I 8 9 9 9 9 9 9 9 7 6 3 4 7 8 4 6 7.3
J 10 10 10 10 10 4 3 2.5 10 10 10 6 10 8 10 8 8.2
a
Best chair = 1 and worst chair = 10.
b
The ranking is statistically significant with Friedman's χ2 = 33.9, p < 0.001.
 259

Summary of results

The results of the study are summarized in Table 4 which

gives average values of ease of adjustability, appearance and
comfort as derived from the chair evaluation form. In addition,
the average overall ranking from the comparison of the chairs
in the conference room is given.

TABLE 4

Average values of ease of adjustability, appearance, and comfort compared to aver­

age overall ranking a

Chair Ease of Appearance Comfort Average

adjust­ overall
ability ranking

A 2.9 b 2.8 3.6b 3.2

Β 4.2 3.2 6.2 4.1
C 4.5 2.9 6.9 4.2
D 4.3 2.5 6.3 5.2
Ε 4.9 3.2 7.6c 5.4
F 3.2 4.3 6.0 5.6
Downloaded by Temple University At 21:31 30 January 2016 (PT)

G 3.4 3.0 6.5 5.7

Η 2.3 b 3.3 6.8 6.1
I 3.6 2.7 6.9 7.3
J 5.0c 5.3c 8.4c 8.2
a
The first three measures were obtained from the chair evaluation form, and the
average overall ranking from Table 3. A numerically small value indicates better
performance.
b
This value is significantly smaller than the others.
This value is significantly greater than the others.

DISCUSSION

From the comparison of chair measures with the ANSI stan­

dard it can be seen that most of the chairs complied with all
required measures except for seat height adjustability. The seat
height, typically, did not adjust low enough. The lower limit of
16 inches has been an issue of some controversy. Chair manu­
facturers claim that it is difficult to manufacture height-ad­
justable chairs that go lower, since the adjustability mechanism
and the casters occupy too much space. Human factors experts,
on the other hand, claim that the lower limit should be deter­
mined by the 5th percentile female popliteal (lowerleg) height,
which is approximately 15 inches (including 1.5 in shoe height).
It should be noted, however, that the subjects did not complain
about the chairs not adjusting low enough. Although there was
a significant correlation (r = 0.19) between popliteal height and
preferred chair height, the value was low, which indicates that
 260

the choice of chair height depends on many other factors as

well. (In practice, it may of course be difficult to adjust the
chair to a low position because the table height is fixed.) There
were more complaints about chairs not adjusting high enough.
A common reason was that a higher eye position is preferred,
since it provides a greater display viewing angle, which makes
it easier to use with bifocal eye glasses.
Three different evaluation methodologies were used in this
study: the general comfort rating, the frequency and severity of
body part discomfort, and the chair feature checklist. Measure-
ment of frequency and severity of body discomfort, see Fig. 2,
was the least informative of these evaluations. Many individu-
als ignored this part of the questionnaire, obviously because
they did not experience body discomfort. We have previously
commented on the difficulties in sensing and verbalizing feel-
ings of (dis) comfort. Thus the results are not particularly
surprising.
The results from the general comfort rating (GCR) were only
slightly more informative. In this case there were significant
differences between job types since programmers reported more
discomfort than did secretaries, but there were no differences
Downloaded by Temple University At 21:31 30 January 2016 (PT)

between the chairs. Although this is the only scale that has been
scientifically validated for measurement of seat comfort [ 15 ],
our subjective impression is that the scale needs further im-
provement. Presently, the GCR incorporates several different
variables including comfort, restlessness, stiffness, soreness, and
pain. The present authors believe that GCR contains too many
dimensions which makes it confusing to use. A less compli-
cated, unidimensional scale would probably be preferable.
The chair feature checklist assessed the appropriateness of
several chair design features. In addition there were overall as-
sessments of global variables including appearance and comfort
of the chair. These measurements provided significant differ-
ences between chair types. Admittedly, there were not very many
fine distinctions, since only the chairs with the very highest and
lowest ratings were significantly different.
The overall rank order of the chairs supplied good informa-
tion. In this case all the chairs were brought to a conference
room for simultaneous evaluation and comparison. This fol-
lowed two weeks of evaluation time, and subjects could there-
fore rely on their previous experience, and had the opportunity
to directly compare the chairs without having to rely on mem-
ory. In terms of expressing preferences, we believe that this part
of the evaluation was the most important. The average overall
ranking of three chairs, A, Band C, were higher than the re-
maining chairs. Their ease of adjustability, aesthetic appear-
ance and comfort were also satisfactory. Although the results
 261

of the rankings were statistically significant, there were many

individual differences, see Table 1. Chair A, overall the best,
was ranked number 9 by one subject and number 7 by another
subject, and chair F was ranked 1 by three subjects and 10 by
four subjects. Some of the variability is probably due to differ-
ences in body size and some to different aesthetic evaluations.
However, much of the variability is due to difficulty in choosing
between the chairs. This was not surprising since the chairs had
been picked from a sample of 86 chairs; they were all comfort-
able and well designed.
The significant correlations between different chair meas-
ures and chair comfort and appearance indicated that several
important components of a chair are taken into consideration
when judging overall comfort. Seat shape, shape of the back rest
and ease of adjustability seem to be particularly important.
These issues need to be studied in greater detail to understand
the implications of different chair design parameters on overall
chair comfort.
There was a great deal of variability in the data because of
subjective evaluations. It would be preferred if there were phys-
ical measures that could predict comfort, such as measures of
Downloaded by Temple University At 21:31 30 January 2016 (PT)

the seat pan and the seat back. Although some of these data
exist, there is no complete methodology that can be used for
objective assessment of chair comfort. Before such a method-
ology can be developed there are basic questions on sitting com-
fort that must be addressed in research, including the effects of
body size, sensitivity of the body to changes in adjustability pa-
rameters and capability to sense and express comfort and dis-
comfort. Until these issues are resolved, we will keep relying on
subjective evaluations. Evaluation of seating is a multi-dimen-
sional problem, and much research is still needed in order to
develop more reliable methods that can identify the most im-
portant components in seating comfort.

REFERENCES

1 M.G. Helander and B.A. Rupp, An overview of standards and guidelines

for visual display terminals, Appl. Ergon., 15 (1984) 185-195.
2 P. Branton and G. Greyson, An evaluation of train seats by observation
of sitting behavior, Ergonomics, 12 (1967) 316-327.
3 J. Kramer, Biomechanische Veranderungen im Lumbalen Bewegungsseg-
ment, Hippokrates Verlag, Stuttgart, F.R.G., 1973.
 262

4 Η. Vernon (1924), quoted by A.J.S. Lundervold, Electromyographic in­

vestigations of position and manner of working in typewriting, Acta Phys­
iol. Scand., 24 (Supplement 84) (1951) 1-171.
5 H. Schobert, Sitzhaltung, Sitzschaden, Sitzmöbel, Springer, Berlin, 1962.
6 G.J. Andersson and R. Ortengren, Lumbar disc pressure and myoelectric
back muscle activity during sitting, Scand. J. Rehabil., 3 (1974) 115-121.
7 A.C. Mandal, The seated man (Homo sedans), Appl. Ergon., 12 (1981)
19-26.
8 C.G. Drury and M. Francher, Evaluation of a forward-sloping chair, Appl.
Ergon., 16 (1985) 41-47.
9 AT&T Bell Laboratories, Video Display Terminals, Preliminary Guide-
lines for Selection, Installation and Use, Bell Telephone Laboratories, Short
Hills, NJ, 1984.
10 M.G. Helander, Anthropometry and biomechanics in VDT applications,
in: Video Displays, Work, and Vision, National Academy Press, Washing-
ton, DC, 1983.
11 W.P. McLeod, D.R. Mandal and P. Malven, The effects of seating on hu-
man tasks and perceptions, in: Human Factors and Industrial Design in
Consumer Products, Tufts University, Medford, MA, 1980, pp. 117-126.
12 R.K. Leuder, Seat comfort: a review of the construct in the office environ-
ment, Hum. Factors, 25 (1983) 701-712.
13 P. Branton, Behaviour, body mechanics, and discomfort, in: E. Grandjean
(Ed.), Sitting Posture, Taylor and Francis, London, 1969.
14 J.C. Jones, Methods and results of seating research, in: E. Grandjean (Ed.),
Sitting Posture, Taylor and Francis, London, 1969.
Downloaded by Temple University At 21:31 30 January 2016 (PT)

15 B. Shackel, K.D. Chidsey and P. Shipley, The assessment of chair com-

fort, Ergonomics, 12 (1969) 269-306.
16 E.N. Corlett and R.P. Bishop, A technique for assessing postural discom-
fort, Ergonomics, 19 (1976) 175-182.
17 Human Factors Society, American National Standard for Human Factors
Engineering of Visual Display Terminal Workstations, Revised Draft, Hu-
man Factors Society, Santa Monica, CA, July 1986.
18 C.G. Drury and B.G. Conry, A methodology for chair evaluation, Appl.
Ergon., 13 (1982) 195-202.
 Office: Technology and People, 3 (1987) 263 263
Elsevier Science Publishers B.V., Amsterdam — Printed in The Netherlands

MEETING ANNOUNCEMENT

3rd International IFIP Conference on W o m e n , W o r k and Computerization,

Amsterdam, The Netherlands, April 2 7 - 2 9 , 1988

The purpose of this conference, organized under the auspices

of the International Federation for Information Processing, is
to:
— develop strategies enabling women to exercise greater con-
trol over the application of new technologies in their work;
— put forward initiatives which will help to improve the posi-
tion of women on the labour market (or have already done
so) within the framework of the introduction of new
technologies;
— explore future developments in the field of work and com-
puters and the conditions under which they can open up
prospects for women, including those in the Third World.
Downloaded by Temple University At 21:31 30 January 2016 (PT)

The conference will centre upon the following topics:

(a) The possibilities for influencing changes in women's work-
ing situations (computer and job content; participation in com-
puterization projects; training programmes and education;
conditions of employment).
(b) The position of women on internal and external labour
markets as a result of the introduction of new technologies (jobs
in general; jobs in the computerization field; flexibility; women
and power; labour market developments in the Third World).

For more information contact: IFIP Conference "Women,

Work and Computerization", SIC, Paulus Potterstraat 40,
1071 .DB Amsterdam, The Netherlands; telephone: + 31-20-
620681; telex: 10761 nl omega.
 This article has been cited by:

1. S.O. Ismaila, O.G. Akanbi, C.N. Ngassa. 2014. Models for estimating the anthropometric dimensions using standing height
for furniture design. Journal of Engineering, Design and Technology 12:3, 336-347. [Abstract] [Full Text] [PDF]
2. George F. Beard, Michael J. Griffin. 2013. Discomfort during lateral acceleration: Influence of seat cushion and backrest.
Applied Ergonomics 44, 588-594. [CrossRef]
3. Vincenzo Cascioli, Zhuofu Liu, Andrew I. Heusch, Peter W. McCarthy. 2011. Settling down time following initial sitting
and its relationship with comfort and discomfort. Journal of Tissue Viability 20, 121-129. [CrossRef]
4. Vincenzo Cascioli, Andrew I. Heusch, Peter W. McCarthy. 2011. Does prolonged sitting with limited legroom affect the
flexibility of a healthy subject and their perception of discomfort?. International Journal of Industrial Ergonomics 41, 471-480.
[CrossRef]
5. Lia Buarque de Macedo GuimarãesParticipatory Method for Evaluating Office Chairs 365-391. [CrossRef]
6. M.G. Mohamed Thariq, H.P. Munasinghe, J.D. Abeysekara. 2010. Designing chairs with mounted desktop for university
students: Ergonomics and comfort. International Journal of Industrial Ergonomics 40, 8-18. [CrossRef]
7. Gyouhyung Kyung, Maury A. Nussbaum. 2009. Specifying comfortable driving postures for ergonomic design and evaluation
of the driver workspace using digital human models. Ergonomics 52, 939-953. [CrossRef]
8. Gyouhyung Kyung, Maury A. Nussbaum. 2008. Driver sitting comfort and discomfort (part II): Relationships with and
prediction from interface pressure. International Journal of Industrial Ergonomics 38, 526-538. [CrossRef]
9. Gyouhyung Kyung, Maury A. Nussbaum, Kari Babski-Reeves. 2008. Driver sitting comfort and discomfort (part I): Use of
subjective ratings in discriminating car seats and correspondence among ratings. International Journal of Industrial Ergonomics
Downloaded by Temple University At 21:31 30 January 2016 (PT)

38, 516-525. [CrossRef]

10. C. Gregory Shaw. 1993. Seat Cushion Comparison for Nursing Home Wheelchair Users. Assistive Technology 5, 92-105.
[CrossRef]