1. J Forensic Sci. 2008 Nov;53(6):1257-72. doi: 10.1111/j.1556-4029.2008.00852.x.
 Epub 2008 Sep 9.

Facial soft tissue depths in craniofacial identification (part I): An analytical 
review of the published adult data.

Stephan CN(1), Simpson EK.

Author information:
(1)Anatomy and Developmental Biology, School of Biomedical Sciences, The 
University of Queensland, Brisbane, Qld 4072, Australia. 
carl.stephan.AU@jpac.pacom.mil

With the ever increasing production of average soft tissue depth studies, data 
are becoming increasingly complex, less standardized, and more unwieldy. So far, 
no overarching review has been attempted to determine: the validity of continued 
data collection; the usefulness of the existing data subcategorizations; or if a 
synthesis is possible to produce a manageable soft tissue depth library. While a 
principal components analysis would provide the best foundation for such an 
assessment, this type of investigation is not currently possible because of a 
lack of easily accessible raw data (first, many studies are narrow; second, raw 
data are infrequently published and/or stored and are not always shared by some 
authors). This paper provides an alternate means of investigation using an 
hierarchical approach to review and compare the effects of single variables on 
published mean values for adults whilst acknowledging measurement errors and 
within-group variation. The results revealed: (i) no clear secular trends at 
frequently investigated landmarks; (ii) wide variation in soft tissue depth 
measures between different measurement techniques irrespective of whether living 
persons or cadavers were considered; (iii) no clear clustering of non-Caucasoid 
data far from the Caucasoid means; and (iv) minor differences between males and 
females. Consequently, the data were pooled across studies using weighted means 
and standard deviations to cancel out random and opposing study-specific errors, 
and to produce a single soft tissue depth table with increased sample sizes 
(e.g., 6786 individuals at pogonion).

DOI: 10.1111/j.1556-4029.2008.00852.x
PMID: 18783476 [Indexed for MEDLINE]


2. J Forensic Sci. 2008 Nov;53(6):1273-9. doi: 10.1111/j.1556-4029.2008.00853.x. 
Epub 2008 Sep 9.

Facial soft tissue depths in craniofacial identification (part II): An 
analytical review of the published sub-adult data.

Stephan CN(1), Simpson EK.

Author information:
(1)Anatomy and Developmental Biology, School of Biomedical Sciences, The 
University of Queensland, Brisbane, Qld. 4072, Australia. 
carl.stephan.AU@jpac.pacom.mil

Prior research indicates that while statistically significant differences exist 
between subcategories of the adult soft tissue depth data, magnitudes of 
difference are small and possess little practical meaning when measurement 
errors and variations between measurement methods are considered. These findings 
raise questions as to what variables may or may not hold meaning for the 
sub-adult data. Of primary interest is the effect of age, as these differences 
have the potential to surpass the magnitude of measurement error. Data from the 
five studies in the literature on sub-adults which describe values for single 
integer age groups were pooled and differences across the ages examined. From 1 
to 18 years, most soft tissue depth measurements increased by less than 3 mm. 
These results suggest that dividing the data for children into more than two age 
groups is unlikely to hold many advantages. Data were therefore split into two 
groups with the division point corresponding to the mid-point of the observed 
trends and main data density (0-11 and 12-18 years; division point = 11.5 
years). Published sub-adult data for seven further studies which reported 
broader age groups were pooled with the data above to produce the final tallied 
soft tissue depth tables. These tables hold the advantages of increased sample 
sizes (pogonion has greater than 1770 individuals for either age group) and 
increased levels of certainty (as random and opposing systematic errors specific 
to each independent study should average out when the data are combined).

DOI: 10.1111/j.1556-4029.2008.00853.x
PMID: 18783475 [Indexed for MEDLINE]