BODY SIZE DATA OF SRI LANKAN WORKERS AND COMPARISON WITH OTHER POPULATIONS IN THE WORLD : ITS IMPACT ON THE USE OF IMPORTED

A national anthropometric survey of Sri tankan workers has provided uptodate information on the body sizes of both males and females. This data (which was lacking in Sri Lanka) can be of great benefit for designers and importers of goods for Sri Lankans. Simple and anthropologically spread techniques have been used to collect the data. Comparisons of body sizes between developing countries (including Sri Lanka) with those of developed counmes revealed wide variations which have been responsible in the incompatibility in the use of imported goods. Anthropometric data of homogenous populations in different countries are not usually available and more so in developing countries. It is emphasized that such data be developed in all countries or in every ethnic group or on a regional basis. The designers, manufacturers and importers of goods should have access to this data. It is suggested that the techniques and methods described in this paper can be used by developing countries lacking similar data.


I. Introduction .
Sri Lanka imports machinery, vehicles and various other items and articles manufactured in foreign countries.These imported items which are manufactured to fit people in those countries, in many instances, cannot be used safely, efficiently and comfortably by the local population.Many problems of inadequacies in design have been experienced by Sri Lankans, either in using imported and even locally produced machines, which have been made disregarding the body sizes of the users.In order to design machines, articles, garments and work places to match a population, an important requirement for the designer is the information regarding body dimensions of the people.There are records of anthropometry of Sri Lankani, but they refer to specific age ggoups or populations viz.University students3 or to school children5 and more recent data collected with the purpose of ascertaining the physical differentiation of the main races in Sri ~anka.'As the required body measurements of the homogenous population of Sri Lanka are not available, a survey was planned and carried out on a national sample of the population of workers, both male and female.The techniques and procedures described in this paper can be of benefit to those planning such surveys.
Sri L d a is a tropical qsian country with a geographical area of approximately 65,000 square kilometers and a population of about 16 million people of whom about 30% consists of the working population.Workers were taken as subjects for the body size survey as it was easy t o reach them, for measuring convenience of random sampling and they represented a homogenous population.Other reasons are that the majority of imported.goods and machines are used by workers.In Industrially Developing Countries QDC) including Sri Lanka, the unavailability of body size data as well as unawareness of the importance of ergonomics principles are the main reasons why machines and goods are not manufactured or adapted to suit the user population.In this paper anthropometric data of Sri Lankas and other IDC are compared with data of Industrialized Countries PC) to show the variability in body sizes that exist.The design incompatibility in'goods seems to be-an important aspect of the probleins involved in the transfer of technology from developed to developing countries.

Methods and Procedures
The national anthropometric survey of workers in Sri Lanka was carried out during a period of 16 months in 1%1/82.The measurements were taken by three $rained investigators.A pilot anthropometric survey lasting for three weeks'was conducted on 100 fibre mill workers which provided the investigators with adequate experience in carrying out body~measurements xcurately and speedily.
A total of 90 body measurements, which were chosen to cover anthropometric informations for designers2$' were taken.The total number of subjects was 724 between the ages of 21 and 51 years, of which 438 (60.5%) were males with mean weight 51 kg, (S.D. 8.3 kg), and 286 (39.5%) were females with a mean weight 43 kg, (S.D. 6.5 kg).This sample included workers from all 24 districts from the 7 provinces of Sii Lanka, (Figure I).
The number of subjects from each district was selected on a proportional basis in accordance with the employment survey statistics of 19747The *five main races which represented the national sample consisted of 581 (8b%) Sinhalese, 78 (11%) Indian Tamils, 53 (7.5%) Ceylon Tamils, 9 (1%) Moors and 3 (0.5%) Burghers.Members of the main relj'gious groups, communities and armed farces and also veddhas (a jungle tribe) have been represented in the sample.The subjects from 80 work establishments and employed in various categories of work were selected at random.Except for the age, sex, race and number of subjects from each district, there was no discrimination of the seleciion of subjects.Sick and disabled persons were excluded.
Simple b u i accurate measuring instruments were used considering the requirements of easy transportation, reliability of results, durability and speed of operation.and use.The main instrument, Harpenden anthropometer was used to take 50% of the measurements (linear measurements).The circumfererices were measured using a canvas flexible measuring tape and for face and head measurements calipers were used.Few measurements were taken with, a steel tape.An adjustable flat seat was used to take seated measurements.
Out of the total of 90 body measurements only 85 were actually taken.The other measurements were derived, (e.g.measurement nasion tumenton was derived by subtracting measurement nasion to, vertex from measurement menton to vertex).The 90 measurements included the weight, 28 standing measurements, 15 sitting, 12 circumference, 25 head, 4 hand and 5 foot.
The measurements were carried out at the respective work places.Male subjects wore briefs, females wore skirt and.blouse of thin material and all were measured without footwear.Two members of the team alternately measured and recorded to reduce boredom and fatigue.The data were recorded on specially designed forms.The recorder also verbally confirmed each measurement before entering it in the form.
The,measurements always were carried out on the left side of the subject.The p ~s t u r e adopted by the subject during measurements and the anatomical points of the measurements were as indicated in Figures 2a to 2h.The hand, feet and head measurements (Figures 2f to 2h) were taken while the subject adopted a standjng posture.
When using the anthropometer and the calipers, care was taken that only light pressure was applied (by the datum probes) at the part of body contact, in order not to indent the flesh.Tape measurements were taken with the tape tension such that the flesh was not significantly indented.data were scnitiniwd for omission and obvious errors.Where errors were apparent, such values were deleted.The checked data were subjected to'statistical analyses.

Data of Sri Lankan Workers
The 'anthropometric data of Sri Lankan workers have been statistically summarised for each of 90 measurements, with .mean,standard deviation, number of subjects and 5th and 95th percentile values for both males and females, (Figures 2a to 2h).The results of weights are given below.(The results for males appear above and females below).' , m e -h y 1;;i:;;-; ; : : ; -- -;;;+ -; A: -; $ --;;:--;;;--;  To ascertain whether significant differences exist within the same population, the weight and three important dimensions viz.stature, sitting height and head circumference, were selected and their means were compared between the 7 provinces (which covered the whole country) using ANOVA.There were no significant differences in these measurements for both male and female populations, which indicated a homogeniety in body measurements among people living in one country like Sri Lanka.However the differences in means between males and females in the above measurements were highly significant (p < 0.001).It is believed that head and face measurements may not differ significantly between males and females in one e t h i c group or society, unlike other body mea~urements.~In the Sri Lankan survey the percentage differences in mean measurements between males and females were greater ' + measurements taken standing (e.g.stature) (7.09%), sitting (e.g.sitting height) (7.45%) and on hand (7.13%) and on foot (7.28%) than face measurements (5.35%) as shown in Table 1.The percentage difference was least in' the head measurements (3.5% approx.).
Therefore with regard to the use of head gear, special modifications in sizing, seem to be not necessary between males and females.

Populations Compared
Most anthropometric data available today are from the United States and other European countries with emphasis on military purposes.With rkgard to' developing countries, anthropometric data are scarce.National data are almost non-existent, as collection of such data takes ,a long time and effort and is sometimes not useful and feasible, considering the vastness of some countries with regard to their geographical area and clipatic and ethnic differences within the country.Due to the lack of precise data for homogenous populations even in developed countries, in this paper c o m ~o n s were'rnade from the available data which.may not be statistically consistent, but suitable enough to show the variations that exist.

I
The following sources of data were used to make comparisons.(a) Data provided by the national anthropometrie survey of workers in Sri Lanka, (Reported in this paper).
(c) Results of a postal questionnaire conduct.ed in developing countries by the authors.
In oyder to see whether significant variations actually exist between Sri Lankans and people from* an industrially developed country e.g.Sweden, a large number of important dimensions of Sri Lankans obtained fiom the national s m e y were compared with data from sweden.llWherever data from Sweden were not available, data from ~r i t a i n , ~ were substituted, It was observed that significant differences were apparent in h o s t all of the dimensions.

Cdmparison of Stature
The percentile values of stature of-males of seven countries have been compared by ~e n n e d ~. " Values of nine other populations including Sri Lankans.were included and shown in Figure 3. Making use of this graph, the percenfages of accommodation of a design range of 90% of .theBritish with respect to populations of other countries were prepared, (Table 2).It was seen that only 35% of Sri Lankans could be accommodated in such a dedgn.Gross inadequacy of a British design (based on accommodating 90% of the British popdIation) was seen for all countries starting from Tunisia down.Remarkably, except for Japan (an industrialised country) the problem appears to be prevalent only in deyeloping countries.
From the available data, average statures of'males and females are (poupcd into different regiqns and shown in     The conclusion that shorter people live in IDC may not always be true like in the example of the Japanese.Even Taiwanese from the Far East (whose data were not available at the time of writing) may be shorter people than Westerners.But Taiwan may now be classified as an IC.Therefore variability in stature and/or other dimensions can be caused by an ethnjc rather than an industrial factor, although climate may also be a contributory factcr, (3.5).

Comparison of Sitting Height
Figure 4 shows a graph with percentile values of sitting heights prepared and completed in the similar lines as for stature.A feature that was shown in this graph is that Japanese have higher sitting heights than the Swedish, Italians, Iranians, Tunisians and South Africans who were taller than the Japanese in stature.A pro-portionally longer trunk in Japanese than ~u r o ~e a n s , ~~ may be the reason why +e Japanese occupy a higher sitting height.Another conspicuous finding is that the 50th percentile sitting height was more or less similar for European continent, Scandinavia and Japanese populations.This is an advantageous facility when designing machines to be operated seated' including motor vehicles for the above populations.Sri Lanka occupied the last position indicating a warning sign2 for importers of motor vehicles, whether from the West or from Far East.It is noted that the imported motor vehicles (cars and coaches) that are mostly used in Sri Lanka do not have the vertical adjustment of the driver's seat.
3.5 'Comparison of Body Proportions.. Body proportions are also important to consider when designing for different populations.Tanner and others14 studied body proportions in Japanese children and adults and comparisons were made with British and with Japanese Americans.They found that Japanese now have trunk/leg proportions closer to those of North Europeans than was the case 20years ago.Concerning relative lengths of the limbs and trunks, pheasant12 had made generalisations such as, African negroes have proportionally greater limb lengths (and a proportionqlly lesser sitting height) than Eurvpeans and vice versa in Far Eastern samples.
Taking the stature, sitting height and body proportions together, similarities have been observed in groups of populations.Except the Koreans where average stature of 168.7 cm appeared a little out of place, all other dimensions more or less fitted in well into the groups.Therefore taking these basic dimensions, the world populations can be conveniently dividedinto six main ethnic groups, viz.Caucasians from the West, Negroids from Africa, Aryans from the Middle East, Mongoloids from the Far East Asia, Latin .hencansfrom South America and Dravidians/Aryans (include Sri Lankans) from South Asia, (Table 4).
-(Number of countries appears in payenthesis)

Comparison of Hand Dimensions
In the design of machine controls, hand tools and articles where manual dexterity becomes important, the hand dimensions are crucial.Data obtained from studies conductkd in the past, the hand dimensions of females from different ethnic groups and countries were compared, (Table 5).The hands of West Indiaii females were found significantly larger than West Europeans and Indian hands,61n another study, the hands of British females were found to be significantly broader &an Hong Rong Chinese and Japanese hands.The Hong Kong Chinese hands were broader thanJapanese hands. 4From the other available data the hand dimensions of African, Sri Lankan, Swedish and Egyptian females were compared.Swedish hand length was significantly greater than that of Sri Lankans and Egyptians.But African (Sudanese females) hand lengths were significantly greater than Swedish hands and Sri Lankan women's hand width significantly greater than Egyptian women.
The hand dimensions do not seem to follow.the.same pattern as for stature and sitting heights.Women in industriahed countries had generally larger hands.The negroid women from developing coundes also had larger hands, even larger than European or Caucasian women.Women from Far East Asia, Middle East and South Asia seemed to have smaller hands.

Figure 4 .
Figure 4. Percentile values of sitting heights of males from 15 countries (Sitting heights measured from sitting surface)

*
The length and breadth on gloves were measured similar t o measurements on human hands.

Table 3 .
It can be seen from this table that North American, Euiopean and Scmdinavian people.are tall, African and people in the Midde East' are of medium height and Latin

Table 2 .
Percentages of accommodation of 5th t o 95th percentile of a British design for other populations.
* An Industrialized Country American, South East Asian (including Sri Lanka) and Far East Asian people are short.It was noted that except for Japan, people who live in IC are taller than those who live in IDC.

Table 3 .
Human body size variability of males from the different parts of the world

Table 6 .
Results of measurements carried out on head harness of imported industrial safety helmets, goggles and hand gloves in Sri Lanka and the relevant mean anthropomemc measurements of Sri Lankan and British populations (All measurements in centimeters)