In the present study golden ratio was applied to classify the human facial form as normal, short, and long based on the their variation in the growth of craniofacial configuration in vertical and horizontal directions. Even though the underlying minor hard tissue abnormalities may be masked by the soft tissue drape, the determination of facial shape can be helpful in earlier identification of progressive developmental abnormalities, which can affect the morphology and functional values of soft tissue in the later stages.[11
] In the present study, we evaluated the mean values of physiognomical facial height, width of the face, and physiognomical facial index according to gender and race. We compared our data with other racial and ethnic groups . Secondarily, we evaluated the golden ratio between the measurements to classify the facial shape of the subjects.
Compilation of facial measurement of various populations
The comparative anthropometric analysis in the present study shows difference and similarities in the facial measurement among the 3 racial groups. The physiognomical facial height of the face was greater in males with significant sexual dimorphism in all the racial groups. The width of the face was greater in males with significant sexual dimorphism only in Malaysian Chinese and Malaysian Indian subjects. Conversely, in Malay subjects, the female value was greater than males with no statistical difference. The vertical height of the face was greater in Malaysian Chinese compared to other racial groups, may be because of higher hairline; however, most of them had short face because of greater width of the face. The measurements and index were greater in Malaysian Chinese and Malaya compared to Malaysian Indians. The measurements showed statistically significant interracial difference in Chinese versus Indians in both the sexes. The width of the face between Indians versus Malays showed no significant difference in male groups. In male group, the physiognomical height of the face showed significant difference between Malaysian Chinese versus Malay. Conversely, in female group, showed no significant difference in width of the face. Physiognomical height of the face showed significant difference between the sexes in Malay's and in male groups between Malay and Malaysian Chinese.
Comparison of the data with other studies revealed variations and similarities in the facial measurement between Malaysian population and other population . In the present study Physiognomical facial height was greater in males than the females in all the racial groups; in contrast, Farkas et al
. reported higher value for females in Indians.[7
] Facial height Malaysian Chinese male group of our study was similar to the male mean value (191.4 mm) in Japanese population.[7
] The Malaysian Indian facial height (male 182.5 mm; female, 172.7 mm) of both the sexes was similar to Angolan population (male 182.5 mm; female, 172.4 mm).[7
] The female value of Malaysian Indians was similar to female value of Croatian (172.6 mm), Thai (172.8 mm), and North American population (172.5 mm).[7
] The female value of Malay population (179.2 mm) was similar to female value of Zulu population (179.1 mm).[7
] In the present study, the face was wider in females than the males in Malay's. This finding was in agreement with the studies conducted by Farkas et al
] on Greeks and Ngeow et al
] on Malays. The mean value of Malaysian Chinese male subjects was similar to the mean value of the corresponding sex in Angolan (139.8 mm) and Turkish (140.1 mm) population.[7
] The female value (135.2 mm) of Malaysian Chinese population was similar as that of polish population (135.5 mm).[7
] The female male mean value of Malaysian Indian was near to Indian (124.9 mm) population mean value.[7
] Similarly, the female Malay populations mean value was close to Turkish population (134.5 mm).[7
Based on paleoanthropology, in the present study, Malaysian Indians facial measurements were different from Malaysian Chinese and Malay, because most of the Malaysian Indians are descendants of South India and they belong to Dravidian race. As excepted, Malays and Chinese showed more similarities in the measurements, since they belong to Mongoloids race.
The golden proportion is considered as a universal standard for aesthetical predictions, biological efficiency, and wellbeing, regardless of sex, age, and ethnicity. Jefferson stated that deviation from the standards can result in the development of facial abnormalities and disorders. In the present study,  20 (M-10, F-10) Indians, 16 (M-5, F-11) Chinese, and 24 (M-15, F-9) Malay subjects had a normal facial shape. The other subjects whose facial proportion deviated from golden ratio were considered as short face (229) and long face (23). The subjects with short and long face may have or develop maxillofacial, temporomandibular joint disorder. The number of subjects with long face were lesser in all the 3 the racial groups; it was interesting to note that none of the Malay female had long face. Saraswathi, in a study on 75 subjects, reported that 14 individuals had normal shape of the face, 61 individuals had long (11) and short face (50).[17
The facial dimensions are studied by direct (spreading or sliding caliper) and indirect anthropometric techniques (camera and computer analysis). In the present study, direct measuring techniques which is considered to be accurate was used in the measurement.[18
] To minimize the errors, the calipers with instrumental error zero were used by one observer for taking the measurement on all the subjects. This is a baseline study done on the Malaysian population; results may not represent the whole Malaysian population due to a relatively small sample size and included students were from various states in Malaysia.