This article was downloaded by: [University of Aberdeen]

On: 27 December 2014, At: 02:33

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number:
1072954 Registered office: Mortimer House, 37-41 Mortimer Street,
London W1T 3JH, UK

Ergonomics
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/terg20

Evaluation of a prototype
multi-posture office chair
a b c
S.J. Legg , H.W. Mackie & W. Milicich
a
Centre for Ergonomics, Occupational Safety
and Health, Department of Human Resource
Management, Massey University, Private Bag
11222, Palmerston North, New Zealand
b
Sports Science Department, Unitec Institute of
Technology, Auckland, New Zealand
c
Hamilton Physiotherapy Clinic, 12 Vialou
Street, Hamilton, New Zealand
Published online: 09 Nov 2010.

To cite this article: S.J. Legg , H.W. Mackie & W. Milicich (2002) Evaluation
of a prototype multi-posture office chair, Ergonomics, 45:2, 153-163, DOI:
10.1080/00140130110120042

To link to this article: http://dx.doi.org/10.1080/00140130110120042

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all
the information (the “Content”) contained in the publications on our
platform. However, Taylor & Francis, our agents, and our licensors
make no representations or warranties whatsoever as to the accuracy,
completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of
the authors, and are not the views of or endorsed by Taylor & Francis.
The accuracy of the Content should not be relied upon and should be
independently verified with primary sources of information. Taylor and
Francis shall not be liable for any losses, actions, claims, proceedings,
demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in
relation to or arising out of the use of the Content.
 This article may be used for research, teaching, and private study
purposes. Any substantial or systematic reproduction, redistribution,
reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access
and use can be found at http://www.tandfonline.com/page/terms-and-
conditions
Downloaded by [University of Aberdeen] at 02:33 27 December 2014
 ERGONOMICS, 2002, VOL. 45, NO. 2, 153 ± 163

Evaluation of a prototype multi-posture o ce chair

S. J. LEGG*{, H. W. MACKIE{ and W. MILICICH}

{Centre for Ergonomics, Occupational Safety and Health, Department of Human
Resource Management, Massey University, Private Bag 11222, Palmerston
North, New Zealand
{Sports Science Department, Unitec Institute of Technology, Auckland, New
Zealand
}Hamilton Physiotherapy Clinic, 12 Vialou Street, Hamilton, New Zealand
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

Keywords: Seating; Chair; O ce; Furniture; Musculoskeletal discomfort;

Ergonomics.

O ce chairs have often been designed to promote a single `correct’ rather rigid
and upright posture, yet it is acknowledged that allowing changes in posture is
good ergonomics practice. The present study investigated o ce worker’s
preferences for a standard shaped typist’s chair (ST) and a prototype multi-
posture (PMP) o ce chair designed to allow its users a variety of sitting positions.
Forty-two (22 male and 20 female) telesales personnel (12), clerical staV (12) and
researchers (18) used ST or PMP in their workplace for the ®rst week of a 2-week
study (with an even number in each work area). The PMP chair was introduced to
participants with a brief lecture on how to use it and with an information booklet.
Following this, each participant completed a chair comfort questionnaire. In the
second week, participants swapped chairs and again completed the chair comfort
questionnaire. At the end of the second week participants were also asked to
complete a separate questionnaire about the usability of the information booklet
that accompanied the PMP chair. Statistically signi®cant diVerences in subject’s
rating of the two chairs were observed in 7 out of 19 questions. On a 100 mm
scale, the ST chair was rated as having a greater mean overall acceptability,
desirability and suitability for body build than the PMP chair. Participants also
claimed to achieve better posture in the ST chair, that they tipped forward less
and were more satis®ed with its width. Although the participants generally
preferred the ST chair, the PMP chair received more favourable ratings among
the researchers who were quite mobile in their work, and in whom there was a
trend for less neck, shoulder and upper back discomfort. More participants
reported an overall preference for the PMP chair. The ®ndings suggest that a
more aesthetically acceptable PMP chair should be developed, peoples’ reasons
for preferring a more traditionally designed chair should be explored, and that the
eVect of postural stability education on personal preconceptions should be
examined to obtain an optimal combination of healthy sitting habits, comfort and
aesthetic qualities in an o ce chair.

* Author for correspondence. e-mail: S.J.Legg@massey.ac.nz

Ergonomics ISSN 0014-013 9 print/ISSN 1366-584 7 online # 2002 Taylor & Francis Ltd
http://www.tandf.co.uk/journals
DOI: 10.1080 /0014013011012004 2
 154 S. J. Legg et al.

1. Introduction
O ce chairs have been the focus of much study, largely as a result of an increasing
awareness of the stresses associated with o ce work by ergonomists and o ce
workers themselves. Many studies have addressed factors causing discomfort while
sitting (Marschall et al. 1995, Helander and Zhang 1997) while others have studied
adjustment capabilities of chairs (Kleberg and Ridd 1987, Ridd and Nicholson 1988,
Helander et al. 1995) and the instruction or evaluation process itself (Corlett 1989,
Verbeek 1991). Discomfort, fatigue and overuse syndromes are commonly related to
long periods of repeated o ce work and reduced performance (Liao and Drury
2000) and as a result of this, chairs have been designed to provide support for the
user in such a way that minimizes discomfort while promoting `good’ posture.
However, it is often taught that there is a single `correct’ (i.e. rigid and upright) way
to sit when working (Pustinger et al.1985, Oborne 1987: 216) and consequently,
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

o ce chairs have in the past been designed to promote a single `correct’ posture. This
poses a problem as it is acknowledged that the requirement to allow changes in
posture is good ergonomics practice (Corlett 1989).
Coleman et al. (1998) reported that o ce chairs with traditional padded ®xed
height lumbar supports were unlikely to provide a comfortable or appropriate seat
for the wide range of potential users. Van Dieen et al. (2001) reported that dynamic
o ce chairs oVer a potential advantage over ®xed chairs, at least in terms of changes
in spinal shrinkage measured by stadiometry. Their ®ndings were supported by
Callaghan and McGill (2001), who reported that the constant spinal loading
associated with little dynamic movement characterized by standing and sitting would
provide little rest/change of muscular activation levels or low-back loading.
These studies imply that dynamic chairs, which promote changes in posture in
the user, should reduce perceived discomfort. Evidence in support of this assertion
was reported by van Deurden et al. (2000). They demonstrated, by stadiometry, that
spinal shrinkage was less in a new concept of o ce chairÐa dynamic chair that
imparted passive forced motions to the seated personÐthan when sitting in a `static’
chair. Stranden (2000) also reported that dynamic seatingÐin the form of a free-
¯oating tilt o ce chair that permitted variation in seat angleÐpromoted leg
movements when seated, thereby activating vein pumps to counteract the formation
of local oedema and swelling in the lower legs in seated work postures.
Herbert et al. (2001) found, in an ergonomics intervention study, that
musculoskeletal symptomatolog y was diminished in apparel manufacturing
workers who used adjustable chairs and who were also the recipients of an
ergonomics education programme that focused on training the workers in proper
use of the chair. This ®nding was of particular interest, since it has been widely
reported that many users, particularly those who have not had speci®c training in
the use of dynamic/adjustable chairs, fail to make use of the adjustment features
that are provided (Shute and Starr 1984, Webb et al. 1984, Lueder 1994, Vitalis
et al. 2000).
In the present study, a new type of dynamic o ce chair was evaluated. It was a
prototype developed by a New Zealand manufacturer and was designed in
collaboration with one of the authors (WM) using fundamental kinematic principles
to promote the adoption by the user of a multiplicity of sitting postures. In order to
further promote the adoption of a wide variety of sitting postures when using the
chair, an educational lea¯et about how to use the chair and a short educational
training programme were also developed and used in this study. It was hypothesized
 Evaluation of a multi-posture office chair 155

that o ce workers would experience increased comfort if they used a chair that was
designed to promote the use of a variety of sitting postures and in which they also
received varying levels of training in the use of the chair. The purpose of the present
study was to investigate o ce worker’s preferences after using two types of o ce
chairsÐa standard-shape d typists’ chair and a prototype multi-posture o ce chair
designed to allow its users a variety of sitting positions.

2. Methods
2.1. Subjects
Forty-two (22 male and 20 female) telesales personnel (12), clerical staV (12) and
researchers (18) were asked to participate in the study. The participants had a mean
and standard deviation (SD) height and weight of 172.2 (SD = 10.9) cm and 76.6
(SD = 14.1) kg, respectively.
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

At the beginning of the study a questionnaire asking about participants’ past and
recent discomfort in the neck, shoulders, upper back, elbows, low back, wrist/hands,

Figure 1. Standard (ST) o ce chair.

 156 S. J. Legg et al.

hips/thighs, knees and ankles/feet was administered to all participants, who were
required to mark their answer to each question on a 100 mm line.

2.2. Study design

The participants were then given one of two types of o ce chair, a standard typists’
(ST) chair (®gure 1), or the newly-designed prototype multi-posture (PMP) chair
(®gure 2). In each workplace (telesales, clerical and research) one-half of the
participants were given the ST chair and one-half were given the PMP chair.
Participants using the ST chair received no instructions on how to use it, while those
using the PMP chair were given about 10 min of practical `postural stability’
instruction by one of the authors (WM) on how best to use the chair and were given
an instructional booklet containing information on diVerent postures that could be
adopted.
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

The information booklet named the PMP chair as `the Options chair’ and
described nine posture options that had been devised by one of the authors (WM).

Figure 2. Prototype Multi-Posture (PMP) o ce chair.

 Evaluation of a multi-posture office chair 157

The posture options included three optimally balanced sitting positions in which the
body’s centre of gravity was above and between the participants’ feet and the
backrest was not used when working or leaning forward (one of these positions in
which both feet were under the chair brought the person into a similar con®guration
as for a `kneeler chair’, and it is argued that improved comfort would be associated
with the body’s centre of gravity passing over and between a person’s feet, when
seated); two traditional (upright) sitting positions in which the centre of gravity of
the body was well behind the feet, one with the trunk supported by the backrest, the
other unsupported; and two positions using the optimally balanced positions (vide
supra) but with the feet either thrust forward or tucked under the seat. A ®nal
optional sitting position was a reclining posture in which the trunk leaned on the
backrest and the feet were thrust forward.
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

2.3. Questionnaire
After 1 week of using the chairs, participants were asked to complete a questionnaire
regarding the chair that they had used. Questions asked about comfort,
acceptability, how much they would want to use the chair, suitability for body
build and height, how often they changed sitting position, how many positions of
comfort could be found, how much normal discomfort was eased, satisfaction with
posture, stability, mobility, tendency to tip forwards, width of the chair, how easy it
was to get on and oV the chair, how much normal work patterns, amount and
quality of work were aVected, how many times they would change their posture in
1 h and what discomfort they felt (in the same areas asked in the preliminary
questionnaire). In a ®nal question, they were asked to identify the chair that they
preferred overall. Participants were also asked to give overall positive and negative
comments about the chair. Most of the questions required marking their answers on
a 100-mm line and providing comments to support their answer, while other
questions required `yes/no’, multiple choice, or numerical answers.
After the ®rst week, participants in each workplace swapped chairs, and the
procedures for the previous week were repeated. This cross-over design was used in
order to negate order eVects. After the week in which they used the PMP chair,
participants were also asked to complete a questionnaire about the usefulness of the
educational material they received before using the PMP chair.

2.4. Statistical analysis

Means and standard deviations were calculated for all participants’ answers to each
question. Paired t-tests were used to examine the diVerences in subject’s preferences
between the two chairs. For binary category of response (`yes/no’ answers), the total
`yes’ and `no’ answers were compared using non-parametric binomial tests (Siegel
and Castellan 1988) for matched pairs (McNemar’s test, Agresti 1990) and for
independent samples (w2 test, Fleiss 1973). Statistically signi®cant diVerences were
accepted at the 5% level of probability (p50.05).

3. Results
3.1. Quantitative preferences
When participants were asked about their overall chair preference, 23 preferred the
PMP chair and 18 preferred the ST chair. However on the 100-mm line scale used for
responses, statistically signi®cant preferences in favour of the ST chair were observed
in seven out of 19 questions about the chairs (table 1). The ST chair was rated as
 158 S. J. Legg et al.

Table 1. Mean (SD) and statistical signi®cance of participants’ responses to chair comfort
questions.
ST chair MP chair Statistical
(n=42) (n=42) DiVerence signi®cance

Overall comfort
(0=very uncomfortable, 100=very
comfortable) 63.1 (19.5) 52.6 (28.2) 710.5 *
Overall acceptability
(0=very bad, 100=very good) 62.9 (22.2) 50.5 (28.5) 712.4 *
How much would you want to use at work?
(0=never, 100=all of the time) 66.0 (29.2) 45.7 (37.7) 720.3 *
Suitability for body build
(0=very unsuitable, 100=very suitable) 68.5 (22.4) 52.7 (32.0) 715.8 *
Suitability for height
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

(0=very unsuitable, 100=very suitable) 70.6 (21.5) 62.2 (28.7) 77.2 NS

Position changed more than usual? y=8, y=28, y=20,
(Yes/no) n=30 n=12 n=718 ***
Positions of comfort achieved
(number) 2.6 (1.6) 3.1 (1.7) 0.5 NS
DiVerent positions ease usual discomfort? y=13 y=13 y=0
(yes/no) n=18 n=17 n=71 NS
Satis®ed with posture?
(0=very dissatis®ed, 100=very satis®ed) 64.9 (23.0) 52.7 (31.3) 710.8 *
Stability of chair
(0=very unstable, 100=very stable) 66.5 (20.2) 58.3 (25.5) 78.2 NS
Ability to move around
(0=very badly, 100=very well) 69.0 (22.8) 61.7 (28.0) 77.3 NS
Extent tipped forward
(0=not at all, 100=very much) 19.7 (21.6) 33.2 (31.6) 13.5 *
Satisfaction with width
(0=very dissatis®ed, 100=very satis®ed) 74.1 (21.3) 63.2 (27.0) 710.9 *
Ability to get on and oV
(0=very di cult, 100=very easy) 73.1 (18.7) 63.5 (27.8) 79.6 NS
How much it aVects normal work pattern?
(0=not at all, 100=a great deal) 31.1 (27.8) 33.9 (29.8) 2.8 NS
AVect amount of work?
(0=not at all, 100=a great deal) 26.3 (27.1) 28.1 (28.5) 1.8 NS
AVect quality of work?
(0=not at all, 100=a great deal) 24.2 (25.8) 27.7 (27.8) 3.5 NS
Number of times posture would change
in 1 hour (number) 8.0 (10.1) 8.2 (5.6) 0.2 NS
Amount of time would feel comfortable
in chair (minutes) 63 (44) 65 (53) 2.0 NS

*Statistically signi®cant at the 5% level of probability (p<0.05); ***Statistically

signi®cant at the 0.1% level of probability (p<0.001); NS=not statistically signi®cantly
diVerent.

having a mean greater overall comfort (63.1, SD = 19.5) than the PMP chair (52.6,
SD = 28.2, p50.05), mean greater overall acceptability (62.9, SD = 22.2) than the
PMP chair (50.5, SD = 28.9, p50.05), mean greater desirability as a work chair
(66.0, SD = 29.2) than the PMP chair (45.7, SD = 37.7, p50.05) and mean greater
suitability for body build (68.5, SD = 22.4) than the PMP chair (52.7, SD = 32.0,
p50.05). Participants also claimed to achieve better posture in the ST chair (64.0,
 Evaluation of a multi-posture office chair 159

SD = 23.9) compared with the PMP chair (52.7, SD = 31.3, p50.05), that they felt
that they tipped forward more on the PMP chair (35.2, SD = 31.6) than on the ST
chair (19.7, SD = 21.7, p50.05) and were more satis®ed with the width of the ST
chair (74.1, SD = 21.3) than the PMP chair (63.2, SD = 27.0, p50.05). A total of 28
participants reported changing their posture more than usual when they used the
PMP chair in comparison with 8 who reported doing so when using the ST chair.
This diVerence was statistically signi®cant (p50.001 for both the McNemar and
w2tests). There was no signi®cant diVerence between the use of diVerent positions to
ease their usual discomfort on either chair. There were no order eVects.
Some of the reasons underlying the responses given by the participants in their
written comments were as follows. Participants preferred the ST chair as a chair
they would want to use at work because they reported feeling few pressure points
and the seat supported the back well, whereas the PMP chair was reported as
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

having a too small a seat causing too much pressure on subjects’ pelvic ¯oors. The
ST chair was reported as being more suitable for their body build than the PMP
chair because of a greater seat area and better back support. The PMP chair was
reported as requiring a longer seat and a higher backrest. Participants felt that they
tipped forward on the PMP chair more than on the ST chair due to a short seat
length giving them reduced leg support. In this instance, the ST chair was reported
as having a wide base with the edges not falling away, and participants reported
sitting further to the back of the seat. Greater satisfaction with the width of the ST
chair than with the PMP chair was reported as a result of a wide seat giving good
support and stability. Participants were less satis®ed with the apparent width of the
PMP chair as it was perceived to be small and narrow with insu cient upper leg
support, despite both chairs having the same posterior width. In general, the ST
chair was preferred as it had a wide seat with good leg support and good back
support, whereas the PMP chair seemed small and narrow with reduced leg
support and less back support.
Although none of the rest of the answers to the chair comfort questionnaire
were signi®cantly diVerent, the ST chair almost always received a more positive
rating. Participants reported achieving a mean of 3.1, SD = 1.7 positions of
comfort in the PMP chair compared with a mean of 2.6, SD = 1.6 positions of
comfort in the ST chair and the mean reported number of times their posture
changed per hour was 8.2, SD = 5.6 for the PMP chair compared with 8.0,
SD = 10.1 for the ST chair.

3.2. DiVerences between participant groups

When each group of participants’ results were analysed separately, similar
preferences towards the ST chair were observed in the telesales and clerical
participants’ responses. However, in the group of researchers, 14 participants
preferred the PMP chair overall while only 4 preferred the ST chair overall. Also,
participants reported a statistically signi®cant greater overall acceptability in the
PMP chair (64.8, SD = 20.6) (on a 100 mm scale) than the ST chair (57.3,
SD = 22.0, p50.05) and, although not statistically signi®cant, almost all other
questions resulted in answers favouring the PMP chair including their responses to
their perception of personal discomfort for all of their body regions. The research
participant’s main reasons for preferring the PMP chair were that there was no
pressure under the thighs and that it had a softer seat and more comfortable back
support.
 160 S. J. Legg et al.

3.3. DiVerences in body part discomfort

There were no statistically signi®cant diVerences in participants’ reported speci®c
comfort levels for diVerent parts of their body for each chair (table 2). However,
participants reported slightly lower discomfort in the neck, shoulders and upper
back for the PMP chair than for the ST chair, and slightly lower discomfort was
reported in the hips/thighs, knees, ankles/feet for the ST chair than for the PMP
chair.

3.4. EVect of educational material

Seventy-four per cent of participants reported reading the educational material that
accompanied the PMP chair. Participants reported a mean score of 81, SD = 18 (on
the 100 mm scale) for understanding the material provided in the information
booklet. A mean score of 63, SD = 25 was reported when they were asked if they
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

found the information booklet helpful, a mean of 58, SD = 24 was reported when
asked how much they were able to apply the information in the booklet to using the
chairs and a mean of 61, SD = 22 was reported when asked how much they could
transfer what they learnt from reading the information booklet to their o ce.

4. Discussion
4.1. Overall preferences
The fact that more subjects had an overall preference for the PMP chair when most
of the answers to the chair comfort questionnaire clearly indicate that they preferred
the ST chair, can be explained by the large number of research participants
preferring the PMP chair overall while the telesales and clerical participants (who
predominantl y preferred the ST chair) gave the ST chair consistently high ratings in
their answers to the questionnaire. The telesales and clerical participants’ preference
for the more traditional type of chair (ST chair) would suggest that either they
preferred sitting in a more traditional fashion, or that in their opinion the PMP chair
was not designed as well as the ST chair so that even though they changed positions
more often, they could never achieve comfort due to the chair’s shape. Participants’
comments regarding the ST and PMP chairs reinforce the former possibility in that
they preferred good upper leg support and a supportive backrest (qualities of a more

Table 2. Mean (SD) and statistical signifcance of participants’ responses to personal

discomfort questions.
Statistical
ST chair MP chair DiVerence signi®cance
Neck 19.8 (23.0) 17.0 (21.5) 72.8 NS
Shoulders 18.9 (23.0) 14.4 (20.0) 74.5 NS
Upper back 12.0 (17.7) 11.6 (17.6) 70.4 NS
Elbows 9.3 (15.0) 9.3 (15.7) 0 NS
Low back 18.8 (23.4) 18.9 (22.0) 0.1 NS
Wrist/hands 12.3 (25.4) 10.8 (14.2) 71.5 NS
Hips/thighs 11.8 (19.4) 16.9 (24.4) 5.1 NS
Knees 9.7 (19.2) 12.7 (21.1) 3 NS
Ankles/feet 10.5 (19.1) 12.4 (20.5) 1.9 NS
NS=No statistically signi®cant diVerence.
 Evaluation of a multi-posture office chair 161

traditional seating position). The negative comments towards the PMP chair (poor
upper leg support, seat too short and narrow) may have been a result of the
participants sitting in a more traditional manner in a chair that is not designed to be
used in this way. This is further reinforced by their reporting achieving a mean of
only 3.1, SD = 1.7 positions of comfort while using the PMP chair when there were 7
diVerent sitting possibilities in the educational material. If all the participants had
used the PMP chair as was intended in the design of the chair, then the results
concerning the questionnaire about the PMP chair may have been more favourable.

4.2. EVect of educational material on chair preference

Although 74% of participants read the information booklet accompanying the PMP
chair, only 62% of telesales and clerical subjects and 77% of research subjects read
it. The greater preference for the PMP chair amongst the research participants
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

supports the view that information on how to use the PMP chair is an important
element in the formation of their opinion about the chair. The overall mean score of
those who read the booklet, only 58, SD = 24 on the 100 mm scale reported by
participants asked how much they could apply the information in the booklet to the
PMP chair. This may have been improved if there had been more user education
explaining the bene®ts of the PMP chair.
The very diVerent reactions between the telesales and clerical and the research
participants towards the two chairs is interesting. The research participant s may
have been more willing or have had more opportunity, as a result of the nature of
their job, to objectively try new sitting positions in an attempt to gain maximum
comfort. They were also more mobile than the telesales and clerical participants as
they reported standing, reaching, or sitting more often. This observation suggests
that more mobile people may have a greater preference towards the PMP chair due
to the inherent emphasis on mobility as part of the design of the PMP chair. In
addition, telesales and clerical participants may have in the past been recipients of
more advice on sitting (which is likely to have been on traditional sitting postures), as
this would most likely have been targeted at workers `at risk of injury’ (commonly
considered to be clerical participants rather than researchers), whereas the research
participants would have been less likely to have received such advice. These
entrenched ideas would to some degree oppose the concepts that the PMP chair
promotes.

4.3. EVect on comfort

The PMP chair served its purpose in that more participants reported changing
position more often. However, in this case changing position more often didn’t
correspond with greater seating comfort, which suggests that participants were
changing their seating position as a result of not being comfortable rather than as a
result of consciously changing position to reduce long-term fatigue. However, it is
not known to what degree discomfort is a result of incompatibility with the PMP
chair or, alternatively, a general distrust of new sitting positions.
Reported discomfort levels after using both the ST and PMP chairs re¯ect the
participants’ comments about each chair. Slightly lower discomfort reported in the
hips/thighs, knees, ankles/feet for the ST chair than for the PMP chair further
suggests that participants found greater lower body comfort when sitting in the more
traditionally shaped (ST) chair. Slightly less discomfort reported in the neck,
shoulders and upper back for the PMP chair is positive, and perhaps is a result of
 162 S. J. Legg et al.

some of the more upright, although less rigid, sitting positions that the PMP chair
promotes.

4.4. EVect of prior back pain

The eVects of a biased sample resulting from some participant s having a
predisposition towards pain due to past discomfort or injury should be minimal in
the present study. For example, 26% of the participants reported a reasonable level
of past neck pain (greater than 20 on the 100 mm scale) and 17% reported a
reasonable level of recent neck pain. This is comparable to the ®ndings of Buckle
(1988) whose combined male and female reported incidence of neck pain was
approximately 22%. Also Buckle’s combined male and female reported incidence of
back pain was approximately 19%, which is comparable to the combined reported
upper and lower back pain of 25% in the present study.
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

4.5. Psychosocial in¯uences

The shape of the PMP chair was described by some participants as being like a
`bicycle seat’. Even if the participants were comfortable working in the various
sitting positions outlined in the educational material, their preconception of a chair
that looked diVerent may have been negative solely on that account. One subject
mentioned that the shape of the seat pan of the PMP chair made it di cult to use
eVectively if the user were wearing a skirt or dress. However, it is considered to be
unlikely that female o ce workers would sacri®ce wearing skirts or dresses for the
sake of comfort. In this case, the challenge would be to educate people about the
bene®ts of the PMP chair in order to gain greater overall appeal or to design a type
of chair that promotes many sitting positions while maintaining aesthetic appeal to
the majority of people who use it.
Future research should include the development of a multi-posture chair that is
more aesthetically acceptable to o ce workers. The reasons underlying peoples’
apparent ambiguity towards the prototype multi-posture chair should be further
investigated. A more thorough examination of the eVect postural stability education
may have on people’s preconceptions about seating should be undertaken so that an
optimal combination of healthy sitting habits, comfort and chair aesthetic qualities
can be reached.

Acknowledgements
This study was conducted in collaboration and with the support of John Wells,
Managing Director, Formway Furniture Limited, Upper Hutt, Wellington, New
Zealand.

References
AGRESTI, A. 1990, Categorical Data Analysis (Chichester: Wiley).
BUCKLE, P. W. 1988, Work-related musculoskeletal disorders, in A. S. Nicholson and J. E.
RIDD (eds), Health and Safety Ergonomics (London: Butterworths), 113±125.
CALLA GHAN, J. P. and MCGILL, S. M. 2001, Low back joint loading and kinematics during
standing and unsupported sitting, Ergonomics, 44, 280±294.
COLEMAN, N., HULL, B. P. and ELLITT, G. 1998, An empirical study of preferred settings for
lumbar support on adjustable o ce chairs, Ergonomics, 41, 401±419.
CORLETT, E. N. 1989, Aspects of the evaluation of industrial seating, Ergonomics, 32, 257±269.
FLEISS, J. L. 1973, Statistical Methods for Rates and Proportions (Chichester: Wiley).
 Evaluation of a multi-posture office chair 163

HELANDER, M. G. and ZHANG, L. 1997, Field studies of comfort and discomfort in sitting,
Ergonomics, 40, 895±915.
HELANDER, M. G., ZHANG, L. and MICHEL, D. 1995, Ergonomics of ergonomic chairs: a study
of adjustability features, Ergonomics, 38, 2007±2029.
HERBER T, R., DRO PKIN, J., WARR EN, N., SIVIN, D., DOUCETTE, J., KELLO GG, L., BARDIN, J., KA SS,
D. and ZOLOTH, S. 2001, Impact of a joint labor-management ergonomics program on
upper extremity musculoskeletal symptoms among garment workers, Applied Ergo-
nomics, 32, 453±460.
KLEBERG, I. G. and RIDD, J. E. 1987, An evaluation of o ce seating, in E. D. Megaw (ed.),
Contemporary Ergonomics 1987 (London: Taylor & Francis), 203±208.
LIAO, M.-H. and DRURY, C. G. 2000, Posture, discomfort and performance in a VDT task,
Ergonomics, 43, 345±359.
LUEDER, R. 1994, Adjustability in context, in R. Lueder and K. Noro (eds), Hard Facts About
Soft Machines (London: Taylor & Francis), 25±35.
MARSCHALL, M., HARRINGTON, A. C. and STEELE, J. R. 1995, EVect of work station design on
Downloaded by [University of Aberdeen] at 02:33 27 December 2014

sitting posture in young children, Ergonomics, 38, 1932±1940.

OBOR NE, D. J. 1987, Ergonomics at Work, 2nd edn (Chichester: Wiley), 216.
PUSTINGER, C., DAINOFF, M. J. and SMITH, M. 1985, VDT workstation adjustability: eVects on
worker posture, productivity, and health complaints, in R. E. Eberts and C. G. Eberts
(eds), Trends in Ergonomics/Human Factors II (Amsterdam: Elsevier), 445±451.
RIDD, J. E. and NICHOLSO N, A. S. 1988, The modern o ce, in A. S. Nicholson and J. E. Ridd,
Health and Safety Ergonomics (London: Butterworths), 126±142.
SHUTE, S. J. and STARR, S. J. 1984, EVects of adjustable furniture on VDT users, Human
Factors, 26, 157±170.
SIEGEL, S. and CASTELLAN, N. J. 1988, Nonparametric Statistics for the Behavioral Sciences, 2nd
edn (New York: McGraw-Hill).
STRANDEN, E. 2000, Dynamic leg volume changes when sitting in a locked and free ¯oating tilt
o ce chair, Ergonomics, 43, 421±433.
VAN DEURDEN, D. L., GOOSSENS, R. H. M., EVERS, J. J. M., VA N DER HELM , F. C. T. and VAN
DEURSEN, L. L. J. M. 2000, Length of the spine while sitting on a new concept for an
o ce chair, Applied Ergonomics, 31, 95±98.
VAN DIEEN, J. H., DE LOOZE, M. P. and HERMANS, V. 2001, EVects of dynamic o ce chairs on
trunk kinematics, trunk extensor EMG and spinal shrinkage, Ergonomics, 44, 739±750.
VERBEEK, J. 1991, The use of adjustable furniture: evaluation of an instruction programme for
o ce workers, Applied Ergonomics, 32, 179±184.
VITALIS, A., MA RMA RRAS, N., POULAKIS, G. and LEGG , S. J. 2000, Please be seated. Ergonomics
for the New Millennium. Proceedings of the XIVth Triennial Congress of the International
Ergonomics Association and the 44th Annual Meeting of the Human Factors and
Ergonomics Society, 29 July ± 4 Aug, San Diego, California, Published by the Human
Factors and Ergonomics Society, PO Box 1369, Santa Monica, California, USA, VI,
43±46.
WEBB, R. D. G., TACK, D. and MCILROY, W. E. 1984, Assessment of musculoskeletal
discomfort in a large scale clerical o ce: a case study, Proceedings of the 1984
International Conference on Occupational Ergonomics (Downsville, Ont: Human Factors
Association of Canada), 392±396.