












































Forensic Anthropology Vol. 4, No. 4: 309–318
DOI: 10.5744/fa.2021.0017

© 2022 University of Florida Press

“If we wish to get the greatest efficiency in forensic 
identification, we should stop trying to do it in a way in 
which the very real local variation becomes mere static 
to confuse the results, and try to do it in a way in which 
local variation makes a positive contribution to correct 
diagnosis.” (Brues 1992:127)

Introduction

Though it has been stated that it is not possible to sort humans 
into more categories than continental origins (Williams et al. 
2005), forensic anthropologists do this all the time, often 
working within a continent. Forensic anthropologists com-
monly refer to this practice as ancestry estimation. But is 
ancestry what we are really estimating? What we call ances-
try and how we refer to it, harkens back to 19th- century racial 
typology— when skulls were observed and forced into one of 
three to five groups. Then along came the Modern Synthesis 
which shifted the focus from categorization to exploration of 
phenotypic diversity through an evolutionary lens (Huxley 
1942). Inspired by Huxley, Sherwood Washburn proposed the 

New Physical Anthropology (1951) which brought the tenants 
of the Modern Synthesis into biological anthropology. With 
this paradigm shift and to distance ourselves from the typol-
ogy of the past, biological and forensic anthropologists 
changed the terms Caucasoid, Mongoloid, and Negroid, to 
European, Asian, and African. This was a small step to dis-
tance the discipline from the typology, as was the use of the 
term ancestry as a replacement for the word race. However, 
the change in terminology was a substitution of names while, 
at the time, maintaining the status quo with the often used 
three group model. New methods of estimation have been 
developed since the name change, along with data collection 
representing more diverse population groups. However, the 
term ancestry estimation is still used frequently in practice, 
what does it mean and is it what we are really estimating?

Ancestry refers to ancestral origins which could include 
continental origins or ethnic affiliation. Ancestral categories 
such as African, Asian, or European are broad and not infor-
mative for describing global or local population structure or 
for use in forensic case work. Although publications have 
highlighted the uniqueness of United States population 
groups with respect to migration, gene flow, genetic drift, and 
secular change, (Edgar 2009; Manthey et al. 2017; Ross et al. 
2003; Spradley et al. 2008; Tise et al. 2014) the notion of con-
tinental ancestry persists. Ousley et al. (2009) among oth-
ers, demonstrated that population groups can be accurately 
classified, using craniometric data, into social race catego-
ries within the United States (U.S.), by time period, ethnic 
affiliation, and geographic boundaries further demonstrating 
that ancestry (i.e., categorization into African, Asian, and 

RESEARCH ARTICLE

What Are We Really Estimating in Forensic Anthropological 
Practice, Population Affinity or Ancestry?

Kate Spradleya* ● Richard L. Jantzb

ABSTRACT: While American forensic anthropologists often state that they estimate ancestry, is that what they are really estimating? 
Although typological terminology, the oids, was replaced with continental terminology, the change was linguistic rather than substantive. 
The American population is comprised largely of immigrants. Genetic data suggests a high degree of admixture within American population 
groups. Further, data from documented skeletal collections suggest that Americans have undergone secular changes. Our paper addresses 
the uniqueness of the American population as compared to ancestral continental and geographic origin groups to address what it is that 
forensic anthropologists are really estimating, ancestry or something else? We conclude, based on uniqueness of American population 
groups, that what forensic anthropologists are estimating is best described as affinity, a term that indicates similarity and is not exclusively 
attached to definitions of race and ethnicity.

KEYWORDS: forensic anthropology, population affinity, ancestry, craniometric, anthropological genetics, biological distance

aDepartment of Anthropology, Texas State University, San Marcos, 
TX 78666, USA

bDepartment of Anthropology, University of Tennessee, Knoxville, 
TN 37996, USA

*Correspondence to: Kate Spradley, Department of Anthropology,
Texas State University, 601 University Dr., San Marcos, TX 78666

E-mail: mks@txstate.edu

Received 06 April 2021; Revised 16 June 2021; Accepted 02 July 2021



310 What Are We Really Estimating?

European) is not what forensic anthropologists are estimat-
ing. Yet, despite evidence to the contrary, introductory text-
books frequently suggest that forensic anthropologists, in 
practice, are estimating continental ancestry (Bass & Trimble 
2005; Black & Ferguson 2011; Byers 2016; Christensen 
et al. 2019; Gill & Rhine 1990; Klepinger 2006).

The term population affinity is an estimation of group 
membership indicating morphological or genetic similarity to 
a well- defined group and is based on some measure of statis-
tical distance. Recent publications highlight an increase in the 
use of this term (Berg & Ta’ala 2014; Pilloud & Hefner 2016; 
Ross et al. 2004; Ross et al. 2011). Forensic anthropologists 
are good at estimating population affinity within the U.S. 
because assortative mating practices shaped by miscegena-
tion laws, racism, and social class have shaped our current 
population structure (Gross & Edgar 2021). The ability of 
forensic anthropologists to estimate population affinity is an 
application of anthropological genetic studies, however this is 
rarely if ever discussed within forensic anthropology text-
books and neglected in the broad literature with the exception 
of a few (Algee-Hewitt & Goldberg 2016; Relethford 2009).

The purpose of this paper is to explore American popu-
lation structure through comparison with ancestral groups to 
provide further insight to the estimation of population affin-
ity and how it differs from ancestry.

Background

Forensic anthropologists in the U.S. can estimate population 
affinity because of population structure. Population structure 
refers to the genotype and phenotype frequencies that reflect 
mating structures. Population structure has long been stud-
ied by both population geneticists and anthropologists. 
Population geneticists study the genetic composition of 
populations, including changes in genotype and phenotype 
frequency resulting from the evolutionary processes of 
selection, genetic drift, mutation, and gene flow over time. 
Anthropological geneticists focus on biocultural processes 
relating to genetic and environmental interactions at the 
population level, most often studying the normal variation of 
complex traits (Crawford 2007), including craniometric data. 
Further, anthropological genetic analyses include theoretical 
models to test hypotheses focusing on microevolutionary 
processes related to population history (Relethford & Lees 
1982:113). Anthropologists have long been at the forefront of 
studying the cultural and behavioral factors that impact pop-
ulation structure through the study of anthropological 
genetics (Algee- Hewitt & Goldberg 2016; Crawford 2001, 
2007; Livingston 1958; Relethford & Lees 1982). Anthropo-
logical genetics uses model- bound methods to test hypothe-
ses concerning human evolution, selection, and population 
differentiation (Algee- Hewitt & Goldberg 2016:558). 

However, model- free methods explore population structure 
when testing explicit models of gene flow and drift are not 
central questions. Human behavior, influential on human 
history, is not static and as such, neither are populations. 
Population size, migration, and secular change all impact 
population structure.

The U.S. has a generally well- documented population 
history with an ever- changing population structure that has 
been the focus of much research (Algee-Hewitt 2017; Bryc 
et al. 2015; Go et al. 2019; Reich et al. 2012). Archaeological 
and historical documentation provides knowledge of the 
ancestral origins of the current population groups within the 
United States. The narrative of these origins is largely sup-
ported by early studies of classical genetic markers and elab-
orated on through more recent studies focusing on genome 
sequencing. Using SNP data with self- reported U.S. Census 
Bureau categories, Bryc et al. (2015) found that African 
Americans are genetically similar to Africans with variable 
amounts of European admixture and detectable, although low 
proportions of Native American admixture. Previous stud-
ies have specifically linked the African ancestry to West 
Africa (Baharian et al. 2016; Bryc et al. 2010) which is also 
consistent with historical narrative. Further, there is detect-
able structure in the variation of European admixture; Afri-
can Americans in the Southeast have less European 
admixture than in the North and West (Baharian et al. 2016; 
Bryc et al. 2010).

The U.S. Hispanic1 population is the second largest pop-
ulation group with the majority of the population having 
geographic origins in Mexico, Puerto Rico, El Salvador, and 
Cuba (Noe- Bustamante 2019). Further, foreign born individ-
uals comprise a significant portion of these population 
groups, Mexican (31%), Salvadoran (57%), and Cuban (56%) 
(Noe- Bustamante 2019). Genetic studies highlight the com-
plex population structure of Hispanic origin countries which 
influences further structure within the United States. Mex-
ico exhibits clinal variation with higher amounts of European 
admixture in the north and less in the central and peninsular 
region (Ruiz- Linares et al. 2014), which has been documented 
with genetic and quantitative data (Rubi- Castellanos et al. 
2009). Puerto Rico and Cuba also have documented complex 
genetic structures with differing levels of Native American, 
European, and African admixture resulting from unique pop-
ulation histories while more African admixture is found in 
Cuba and Puerto Rico (Cintado et al. 2009; Hanis et al. 1991). 
Thus, the U.S. Hispanic population is not genetically homog-
enous, rather it is complex with detectable structure evident 

1. The term Hispanic is used over Latinx for several reasons.
According to the Pew Research Center Hispanic Trends, approximately 
76% of individuals living in the U.S. with geographic origins/ancestral 
ties to Latin America have not heard of the term Latinx and only 3% use 
it (Noe- Bustamante et al. 2020). Additionally, at this time, it is the most 
common term utilized for reporting missing and unidentified persons.



Spradley and Jantz 311

across the country. European Americans have considerable 
admixture with other European groups and with African and 
Native Americans, although in relatively small amounts 
(Bryc et al. 2015).

In addition to the influences of population history and 
gene flow on population structure, skeletal secular changes 
have been documented in African and European Americans 
(Jantz 2001; Jantz & Meadows Jantz 2016). In particular, an 
increase in cranial base height has provided an increase in 
overall vault height for both groups (Jantz 2001; Wescott & 
Jantz 2005) along with vault width and cranial base length 
(Jantz & Meadows Jantz 2016). Investigating limb bone allom-
etry in European Americans, Jantz and Meadows Jantz (2017) 
found that significant changes occurred in a short period of 
time suggesting that the unique American environment has 
disrupted developmental canalization. Although significant 
secular changes in skeletal morphology have occurred over 
the past century, such changes do not obscure underlying 
genetic structures (Relethford 2004) and support the use of 
skeletal morphology in exploration of population structure.

Given the complex population history, varying degrees 
of genetic admixture among all U.S. groups, and secular 
changes in skeletal phenotype due to a unique American 
environment, are American population groups similar to 
their ancestral populations? In the following analyses we use 
craniometric data and biological distance analysis to explore 
the pattern of quantitative variation among individuals who 

self- identified into three U.S. Census Bureau racial and ethnic 
categories and geographically defined population groups in 
the U.S. (African American, Hispanic American, and Euro-
pean American) as compared to their ancestral geographic 
origin groups. The resulting structure from the biological 
distances will be discussed within the framework of the 
ancestry vs the affinity of local populations.

Materials and Methods

Craniometric data from U.S. and worldwide groups (Table 1) 
are used to generate biological distances. U.S. groups repre-
sent self- reported or next- of- kin- reported group labels that 
fit within the Office of Management and Budget’s (OMB) 
Statistical Directive No. 15 that outlines race and ethnic cat-
egories for federal bureaucratic reporting purposes. These 
categories are also used by the National Missing and Uniden-
tified Persons System or NamUs, which is a national miss-
ing and unidentified persons database used in medico- legal 
investigations. The OMB does not capture the range of vari-
ation or local group level for all U.S. citizens. However, it 
does capture the most often used terminology employed by 
forensic anthropologists when providing estimates of group 
membership for the biological profile. The data utilized in the 
subsequent analyses for American groups were based on 
availability and represent the three largest population groups 

TABLE 1—Sample Groups

Group Female Male Time Period Curator /Reference

African
Gold Coast
Ashanti
Cameroon
Calabar
Dogon
Togo
South African

10
13
10
7

52
7

44

10
17
31
16
47
28
57

19th Century
19th Century
19th Century
19th Century
17th– 18th Centuries
19th Century
20th Century

American Museum of Natural History
”
”
”

https:// web . utk . edu / ~auerbach / HOWL . htm
Berlin Society for Anthropology, Ethnology and Prehistory 

20th Century
American

African American
American/Hispanic
European American
Zuni

101
13

169
20

171
59

343
24

20th Century
20th Century
20th Century
16th– 17th Centuries

Forensic Anthropology Data Bank
”
”

Smithsonian Institution
Asian

Japan
 Hainan

55
38

64
45

20th Century
18th– 19th Centuries

Dudzik, 2015  
https:// web . utk . edu / ~auerbach / HOWL . htm

European
Croatian
Czech
German
Italian
Portuguese

0
35
21

114
64

31
44
58

106
68

20th Century
19th Century
19th Century
20th Century
20th Century

Vienna Natural History Museum
Vienna Natural History Museum and Charles University Smithsonian 

Institution and Charite Anatomy
Cattaneo et al. 2018

Weisensee 2008
Latin American

Mexican Migrant
Mexican Mayan
Guatemalan Migrant
Guatemalan Mayan

15
20
15
0

156
42
25
74

20th Century
20th Century
20th Century
20th Century

Mexican, Guatemalan, and Migrant Craniometric Dataset
”
”

Forensic Anthropology Foundation of Guatemala



312 What Are We Really Estimating?

in the U.S., African American, Hispanic American, and 
European American. There is currently a lack of skeletal data 
available for Asian Americans and contemporary Native 
Americans. However, with the development of New Mexico 
Decedent Image Database (https:// nmdid . unm . edu / ) and 
other large datasets derived from CT images (Colman 2019; 
Stull 2020), such data should increase in the future.

Population Groups and Data

Craniometric data (following definitions outlined in Howells 
1973) from African American, European American, and His-
panic Americans come from the Forensic Anthropology 
Data Bank (https:// fac . utk . edu / background / ) and represent 
20th- century birth years. These groups consist of positively 
identified individuals with self- reported or next of kin 
reported affinity. Although self- reported and next of kin 
reported mostly consist of the terms Black or White, in this 
paper we use the terminology African American and Euro-
pean American to convey the combination of ancestry and 
geographic birthplace/nationality. While the African and 
European Americans were born in the U.S., the Hispanic 
American sample consists of individual with ancestral ties 
to Mexico who were either born in the U.S. or born in Mex-
ico and moved to the U.S. shortly after birth.

The West and Central African data were collected from 
the American Museum of Natural History and were part of 
the Felix von Luschan collection within the Royal Museum 
of Ethnology in Vienna, Austria (Smith 2002). The exact 
time period they represent is unknown, although likely 
18th and 19th century (Smith 2002). The Dogon come from 
the (Howells 1973) craniometric data set and are published 
online. Togo are part of the von Luschan collection at the Ber-
lin Society for Anthropology, Ethnology and Prehistory. 
These African data represent primary areas of origin for the 
enslaved Africans forced to immigrate to the New World. 
The South African data is from L’Abbé et al. (2013) and 
comes from the Pretoria Bone Collection.

Latin American data are represented by indigenous Gua-
temalan and Mexican Mayans and migrants from Mexico and 
Guatemala that died in U.S./Mexico border states, predomi-
nately Arizona and Texas and (Spradley 2020). All Latin 
American groups, Mexican and Guatemalan Mayan and 
migrants, have 20th century birth years and are well defined in 
the literature (Hughes et al. 2017; Spradley 2021). A Southwest 
Native American group, Zuni, was also included in the analysis 
in addition to two Asian groups, Hainan (Howells 1989) and 
Japanese (Dudzik 2015). Native Americans, including Indige-
nous groups in Mexico and Central America, have Asian 
ancestry from prehistoric migration (Hughes et al. 2019).

Data from Europe spans the West and East and includes 
a Czech sample that contains known individuals collected 
from the Vienna Natural History Museum in Austria and 

Charles University in the Czech Republic. The Croatian data 
are from the Vienna Museum of Natural History. Italian data 
were obtained from Manthey et al. (2018). The Portuguese 
data come from the New Lisbon Skeletal Collection (Wei-
sensee & Jantz 2011).

Methods

Our research question addresses the pattern of variation 
among U.S. population groups as related to their ancestral, 
geographic origins, and as such the purpose is not classifica-
tion. Rather we explore the pattern of population relation-
ships through biological distance. The U.S. population 
groups are compared to ancestral and local groups based on 
historical narrative and supported by genetic studies. Afri-
can Americans are compared to West and Central Africans, 
along with European Americans. European Americans are 
used in lieu of the European groups as the European Amer-
ican sample more closely approximates the population group 
with contributing gene flow to African Americans. The His-
panic American group is compared to all Latin American and 
European Americans. The Hispanic comparison also includes 
Asian and Native American groups due to shared ancestry. 
European Americans were compared to European groups as 
well as South Africans as the latter have a similar popula-
tion history with European Americans.

Analyses were completed following procedures outlined 
in (Jantz & Spradley 2014) to produce Mahalanobis and Fst 
distances. Sexes were pooled to increase the overall sample 
size and sex differences were removed by centering on sex 
specific means. FORDISC 3.1 was used to generate the cova-
riance matrices and means for each group which were cen-
tered and pooled. The eigenvalues and eigenvectors were 
extracted to obtain the principal component scores to com-
pute Mahalanobis distances (D2) and obtain the canonical 
variates to create two- dimensional plots. The plots are used 
to assess population structure and structure coefficients are 
used to interpret the craniometric variation.

Results

The CV plot featuring African Americans (Figure 1) shows 
the African American group intermediate to the West and 
Central Africans and European Americans on the first axis. 
The second axis highlights geographical patterns among the 
West and Central African groups with Nigeria (Gold Coast) 
most differentiated from Mali (Dogon) and situates African 
Americans with West Africans. Almost all variables contrib-
ute to CV1 as indicated by the structure coefficients (see 
Table 2). The pattern of loadings presents a clear pattern of 
morphometric variation. High scores on CV1 result from a 
wide cranial base, long cranial base length and vault height, 

F1

T2



Spradley and Jantz 313

FIG. 1—African American CV Plot.

TABLE 2—Structure Coefficients

Variable

African 
American

European 
American Hispanic

Can1 Can2 Can1 Can2 Can1 Can2

AUB 0.947 0.036 – 0.671 0.593 – 0.316 – 0.732
BBH 0.768 0.087 0.783 0.222 0.616 0.729
BNL 0.881 0.423 0.700 0.086 0.849 0.464
BPL – 0.723 0.608 0.300 0.242
DKB – 0.875 0.260 – 0.507 0.672 – 0.685 0.381
EKB – 0.658 0.228 – 0.097 0.160
FOL 0.237 0.266 0.887 0.226 0.744 0.153
FRC 0.839 – 0.202 0.868 – 0.018 0.869 0.284
GOL 0.763 0.556 0.935 – 0.249 0.938 0.104
MDH 0.700 0.468 0.611 0.770 0.765 0.080
NLB – 0.971 – 0.088 – 0.764 0.332
NLH 0.895 0.202 0.500 – 0.451
OBB 0.942 – 0.067 0.389 0.369 0.764 – 0.413
OBH 0.156 0.093 0.229 – 0.765 – 0.385 – 0.572
OCC 0.938 – 0.147 0.952 0.189 0.753 0.335
PAC 0.006 0.910 0.738 – 0.349 0.824 0.349
XCB 0.661 – 0.581 – 0.908 0.357 0.313 – 0.496
ZYB – 0.269 0.001 – 0.723 0.513 – 0.766 0.147

high nasal height and narrow nasal aperture and interorbital 
distance. The reverse is the case for low scores on CV1. 
CV1 reflects a pattern of variation normally associated with 
African- European variation. High scores on CV2 mainly 
reflect longer basion- prosthion and basion- nasion lengths, 
and a longer, narrow cranial vault.

What is noteworthy in Figure 1 is that African Ameri-
cans are about equidistant from West Africans and European 
Americans. This is shown quantitatively in the first column 
of Table 3. African Americans are slightly more similar to 
three African groups (Ashanti, Cameroon, and Togo) than to 
European Americans. But African Americans are slightly 
more similar to European Americans than they are to three 
other African groups (Calabar, Dogon, and Gold Coast). The 
distances show African Americans to occupy the approxi-
mate mid- point between Africans and European Americans. 
The Fst distances in the first row of Table 3 are all around 
0.06, presenting the distances as a proportion of total varia-
tion attributable to between groups.

The CV plot for Hispanics (Figure 2) situates the  U.S.- born 
group intermediate to European Americans and Latin Ameri-
can and Asian groups, with European Americans the most dif-
ferentiated. CV1 mainly serves to separate Euro- Americans 
from the Indigenous Maya groups and Hainan. The second 
axis separates the Asian groups from all other groups. Consid-
ering Hispanic and Latin American groups, the Hispanic 
American group is most similar to the Mexican migrant group, 
followed by the Guatemalan migrants and most differentiated 
from the Mexican and Guatemalan Mayan groups (Table 4). 
Further, the Mexican and Guatemalan migrant groups are 
intermediate to Hispanic American and Guatemalan Mayan 
groups. High scores on CV1 mainly reflect long cranial bases 
(Table 2), generally long crania with narrow nasal apertures 
and narrow bizygomatics. CV2 reflects narrow cranial bases 
and high vaults, small orbits and wide interorbital breadth.

The European American CV plot (Figure 3) highlights 
that European Americans, South Africans, and Portuguese 
are distanced from all other European Groups on the first axis. 
The Croatians and Czechs are more similar to one another as 
are the Germans and Italians. The second axis emphasizes the 
intermediate nature of the European Americans as compared 
to the South African and Portuguese groups (Table 5). The 

TABLE 3—Mahalanobis distances (below diagonal) and Fst distances (above diagonal) for groups used in African analysis.

Af_Am Ash Cal Cam Dog Eu_Am GoC Togo

Af_Am 0.000 0.058 0.078 0.057 0.062 0.059 0.063 0.056
Ash 8.931 0.000 0.025 0.018 0.027 0.157 0.009 0.011
Cal 12.200 3.668 0.000 0.014 0.027 0.173 0.027 0.031
Cam 8.643 2.700 2.082 0.000 0.027 0.159 0.019 0.018
Dog 9.544 3.981 4.059 3.985 0.000 0.151 0.047 0.029
Eu_Am 9.102 26.727 30.075 27.285 25.522 0.000 0.175 0.155
GoC 9.682 1.381 4.019 2.836 7.176 30.635 0.000 0.022
Togo 8.481 1.605 4.572 2.676 4.344 26.325 3.304 0.000



314 What Are We Really Estimating?

structure coefficients show that high scores on CV1 are asso-
ciated with long cranial bases, high vaults, generally longer 
vaults and narrow vaults and faces (Table 2). High scores on 
CV2 reflect greater vault and face breadth, low orbits, and 
greater mastoid height.

Discussion

Our results provide evidence of geographic structure in that 
American population groups are differentiated from their 

geographic, ancestral origin groups. Thus, these American 
groups represent local population groups and local variation 
found in the U.S.. These results are not surprising given the 
population histories of each group and the documented gene 
flow and secular changes that have taken place in the Amer-
ican population over the past 150+ years (Jantz 2001; Mead-
ows Jantz & Jantz 1999; Spradley 2006; Wescott & Jantz 
2005). It is possible that the dissimilarity of the American 
groups to their ancestral groups is due to a lack of directly 
comparable reference data, especially for the European 
Americans. However, the groups used in the present analysis 

FIG. 2—Hispanic American CV Plot.

TABLE 4—Mahalanobis distances (below diagonal) and Fst distances (above diagonal) for groups used in Hispanic analysis.

Eu_Am Ghisp Hai Jap Hisp Am Mmaya UShisp Zuni Gmaya

Eu_Am 0.000 0.064 0.085 0.058 0.048 0.095 0.031 0.098 0.089
Ghisp 9.801 0.000 0.046 0.032 0.007 0.035 0.020 0.029 0.019
Hai 13.359 7.015 0.000 0.019 0.042 0.069 0.050 0.039 0.045
Jap 8.847 4.713 2.734 0.000 0.026 0.058 0.025 0.031 0.027
Mhisp 7.288 1.012 6.353 3.822 0.000 0.040 0.012 0.031 0.023
Mmaya 15.034 5.210 10.597 8.862 5.964 0.000 0.046 0.061 0.051
UShisp 4.577 2.891 7.584 3.637 1.695 7.022 0.000 0.036 0.025
Zuni 15.585 4.324 5.821 4.623 4.626 9.340 5.443 0.000 0.018
Gmaya 14.080 2.820 6.733 3.974 3.368 7.670 3.753 2.688 0.000

FIG. 3—European American CV Plot.

TABLE 5—Mahalanobis distances (below diagonal) and Fst distances (above diagonal) for groups used in 
European analysis.

Czech Eu_Am Ger Por Saf Ita Cro

Czech 0.000 0.078 0.018 0.071 0.091 0.018 0.023
Eu_Am 9.464 0.000 0.066 0.042 0.040 0.044 0.071
Ger 2.066 7.956 0.000 0.052 0.086 0.015 0.032
Por 8.543 4.954 6.203 0.000 0.067 0.043 0.085
Saf 11.227 4.674 10.501 8.064 0.000 0.074 0.100
Ita 2.047 5.176 1.696 5.002 8.896 0.000 0.024
Cro 2.669 8.543 3.698 10.395 12.502 2.699 0.000



Spradley and Jantz 315

represent groups that are documented or have reliable contex-
tual information along with available craniometric data.

The American population groups used in the current 
analyses have ancestry in Africa, Europe, and Mexico, all of 
which have documented population structure. Structure has 
been documented within Africa using genetic and craniomet-
ric data and is thought to be influenced by ethnicity, lan-
guage, and geography (Keita 1990; Spradley 2006; Stull 
2013; Tishkoff et al. 2009), although genetic distance and 
geography have varied strengths (Tishkoff et al. 2009) likely 
due to the Bantu migration. Europe is also highly structured 
based on geography. Relethford et al. (1980) found variation 
among local populations in Ireland using anthropometric 
data, but also found that as migration increased, among group 
variation decreased. Further, Novembre et al. (2008) demon-
strated that across Europe, an individual’s residence can be 
predicted geographically within a few hundred km based on 
SNP data. Mexico also demonstrates geographic structure on 
a clinal scale, with higher European admixture in the North 
and the least in the South and Peninsular regions with Cen-
tral Mexico presenting intermediate (Algee- Hewitt et al. 
2018; Hughes et al. 2013, 2017). The within group variation 
in Africa, Europe, and Mexico is not found within Ameri-
can population groups.

Using craniometric data and unsupervised clustering, 
Algee- Hewitt (2016) demonstrated that that the American 
population can be resolved into three basic clusters, African 
American, European Americans, and Native American/His-
panic. However, she found relatively little within group 
structure. Algee- Hewitt’s results are also in line with genetic 
studies. Micheletti et al. (2020) found that over 90% of Afri-
can Americans have genes from four major regions of West 
Africa suggesting that any initial structure resulting from the 
African Diaspora has been lost. Although the degree of 
admixture is patterned with higher percentages of African 
ancestry in the southeastern U.S. (Baharian et al. 2016). 
Although European immigrants came from many geographic 
areas within Europe, genome wide SNP data found relatively 
little structure in European Americans, identifying only NW 
Europeans, SE Europeans, and Ashkenazi Jews (Price et al. 
2008). While structure has been found in Hispanic Ameri-
cans, with more Hispanics of Mexican origin in the West and 
Caribbean in the East (Bertoni et al. 2003), Algee- Hewitt 
(2016) did not find structure in her Hispanic sample, most 
likely because her sample was predominantly of Mexican ori-
gin. Our results indicate similarity Hispanic Americans and 

Mexican migrants; however, they are also differentiated. The 
Mexican migrants are less similar to the Mexican and Gua-
temalan Mayans as they are thought to exhibit more admix-
ture (Spradley 2021), yet they are still differentiated from 
Hispanic Americans with Mexican ancestry.

Craniofacial secular change has been documented in 
historic to modern African and European Americans (Jantz 
2001; Jantz & Jantz 2016; Jantz & Meadows Jantz 2000; 
Spradley 2006; Wescott & Jantz 2005) but less so in Hispanic 
Americans (Spradley et al. 2016). Studies focusing on Afri-
can and European Americans find that both groups show 
changes in the same direction with a most notable increase 
in vault height, particularly in the cranial base (Wescott & 
Jantz 2005). Secular change can result from a variety of fac-
tors including gene flow, improvements in overall health con-
ditions, and plasticity. Given that both African and European 
Americans have shared the same environment for over sev-
eral hundred centuries, population structure is still apparent 
among these two groups. However, our analyses suggest that 
African Americans are more similar to European Americans 
than admixture proportions would predict, possibly due to 
the shared American environment. Although secular change 
has influenced the overall craniofacial morphology of Amer-
ican population groups, such changes do not obscure the 
underlying genetic structure (Relethford 2004). Therefore, 
secular change is not the sole driver of dissimilarity between 
American groups and ancestral groups.

Given the similarities between these two groups and 
their intermediate or dissimilar positions relative to their 
ancestral groups, an additional analysis, to highlight and 
contextualize our results within a forensic anthropological 
framework, was performed. A discriminant functional anal-
ysis (DFA) was performed to assess how well these Ameri-
can groups classify when compared to their ancestral groups. 
A cross- validated classification matrix is presented in Table 6. 
Only 29 African American misclassify into the African 
group, only 7 more than classify into the European Ameri-
can group. These results agree with the intermediate position 
of African American presented in our CV plot. Further, West 
Africans only misclassify into the African American group. 
A total of 61 European Americans classify into the European 
group and 20 as African American. Interestingly, Europeans 
misclassify into all other groups. Given that 75% of African 
Americans and 84% of European Americans are correctly 
classified into their own group, we suggest that the ability to 
predict group membership can be referred to as affinity.

TABLE 6—Classification Matrix

N Afr_Am Eur_Am European W. African % correct

Afr_Am 270 202 22 17 29 74.8
Eur_Am 509 20 427 61 1 83.8
European 406 56 61 279 10 68.7
W. Africa 242 27 0 0 215 88.8



316 What Are We Really Estimating?

Conclusions

Forensic anthropology has traditionally been a subdiscipline 
of skeletal biology. However, it should also be considered a 
subdiscipline of anthropological genetics. Brues (1992) under-
stood population structure (and simulated it) before forensic 
anthropology crystallized as a discipline. Analyses of popula-
tion structure through biodistance, including estimations 
made by forensic anthropologists, employ fundamentals of 
anthropological genetics and skeletal biology, and provide 
insights regarding mating structures, geography, and environ-
ment, and their impact on quantitative trait variation and local 
population structures. We suggest that the ability to predict 
group membership at the level of local variation rather than 
global variation can be referred to as affinity. All of which is to 
say that it is vitally important that forensic anthropologists 
understand the populations from which their cases are drawn.

Our three analyses suggest that when forensic anthropol-
ogists provide an estimation of group membership within the 
U.S., they are providing an estimation of population affinity
rather than ancestry. The estimation of affinity is more specific
than ancestry and considers local variation (e.g., Hispanic
American, African American, European American). If we
compare craniometric data from an unknown individual to
African Americans, Hispanic Americans, European Ameri-
cans, or other contemporary American groups, then we are
estimating which group the unknown is most similar to, and
therefore the estimation would be considered affinity. Is the
suggested difference between ancestry and affinity trivial?
On the surface, perhaps. However, Maier et al. (2021) recently
pointed out confusion and inconsistencies with terminology
in forensic literature concerning the terms ancestry and race
and how they are perceived as interchangeable. They call for
discussions of standardization to improve practice and public
engagement. Affinity provides more flexible terminology that
is dependent on the researcher to fully contextualize and
describe the groups used in the analyses. Further, it is more
flexible and can be used in areas where population structure
exists but is influenced by geographic boundaries, isolation
by distance, or cultural assortative mating practices rather
than prescribed racial or ethnic variation.

References

Algee- Hewitt BF, Hughes CE, Anderson BE. Temporal, geographic 
and identification trends in craniometric estimates of ancestry 
for persons of Latin American origin. Forensic Anthropology 
2018;1(1):4– 17.

Algee- Hewitt BFB. Population inference from contemporary 
American craniometrics. American Journal of Physical 
Anthropology 2016;160(4):604– 624.

Algee- Hewitt BFB, Goldberg A. Better together: Thinking 
anthropologically about genetics. American Journal of Physi-
cal Anthropology 2016;160(4):557– 560.

Algee-Hewitt BF. Geographic substructure in craniometric estimates 
of admixture for contemporary American populations. Amer-
ican Journal of Physical Anthropology 2017;164(2): 260– 280.

Baharian S, Barakatt M, Gignoux CR, Shringarpure S, Errington 
J, Blot WJ, et al. The great migration and African- American 
genomic diversity. PLoS Genetics 2016;12(5):e1006059.

Bass WM, Trimble MK. Human Osteology: A Laboratory and 
Field Manual. Columbia: Missouri Archaeological Society; 
2005.

Berg GE, Ta’ala SC. Biological Affinity in Forensic Identification 
of Human Skeletal Remains: Beyond Black and White. Boca 
Raton, FL: CRC Press; 2014.

Bertoni B, Budowle B, Sans M, Barton SA, Chakraborty R. Admix-
ture in Hispanics: Distribution of ancestral population con-
tributions in the Continental United States. Human Biology 
2003;75(1):1– 11.

Black S, Ferguson E. Forensic Anthropology: 2000 to 2010. Boca 
Raton, FL: CRC Press; 2011.

Brues AM. Forensic diagnosis of race— general race vs specific 
populations. Social Science & Medicine 1992;34(2):125– 128.

Bryc K, Auton A, Nelson MR, Oksenberg JR, Hauser SL, Willia 
S, et al. Genome- wide patterns of population structure and 
admixture in West Africans and African Americans. Proceed-
ings of the National Academy of Sciences 2010;107(2): 786– 791.

Bryc K, Durand EY, Macpherson JM, Reich D, Mountain JL. The 
genetic ancestry of african americans, latinos, and european 
Americans across the United States. The American Journal of 
Human Genetics 2015;96(1):37– 53.

Byers SN. Introduction to Forensic Anthropology. London: Taylor & 
Francis; 2016.

Christensen AM, Passalacqua NV, Bartelink EJ. Forensic Anthro-
pology: Current Methods and Practice. London: Academic 
Press; 2019.

Cintado A, Companioni O, Nazabal M, Camacho H, Ferrer A, Fer-
nandez De Cossio ME, et al. Admixture estimates for the pop-
ulation of Havana City. Annals of Human Biology 2009;36(3): 
350– 360.

Colman K. Towards Virtual Forensic Anthropology [PhD disser-
tation]. University of Amsterdam; 2019.

Crawford MH. The Origins of Native Americans: Evidence from 
Anthropological Genetics. Cambridge: Cambridge University 
Press; 2001.

Crawford MH. Anthropological Genetics: Theory, Methods and 
Applications. Cambridge: Cambridge University Press; 2007.

Dudzik B. Investigating Cranial Variation in Japanese Populations 
Using Geometric Morphometrics [PhD dissertation]. Knox-
ville: University of Tennessee; 2015.

Edgar HJH. Biohistorical approaches to “race” in the United States: 
Biological distances among African Americans, European 
Americans, and their ancestors. American Journal of Physi-
cal Anthropology 2009;139(1):58– 67.

Gill G, Rhine S, eds. Skeletal Attribution of Race. Albuquerque: 
University of New Mexico Maxwell Museum of Anthropol-
ogy; 1990.

Go MC, Jones AR, Algee- Hewitt BF, Dudzik B, Hughes CE. Clas-
sification trends among contemporary Filipino crania using 
Fordisc 3.1. Forensic Anthropology 2019;2(4):293– 303.

Gross JM, Edgar HJ. Geographic and temporal diversity in dental 
morphology reflects a history of admixture, isolation, and drift 
in African Americans. American Journal of Physical Anthro-
pology 2021;175(2):497– 505.

Hanis CL, Hewett- Emmett D, Bertin TK, Schull WJ. Origins of US 
Hispanics: Implications for diabetes. Diabetes Care 1991;14(7): 
618– 627.

Howells W. Data from: Howells Craniometric Data Set. https:// web 
. utk . edu / ~auerbach / HOWL . htm. 1973.



Spradley and Jantz 317

Howells W. Skull shapes and the map. Papers of the Peabody 
Museum of Archaeology and Ethnology Vol. 79. Harvard Uni-
versity, Cambridge; 1989.

Hughes CE, Tise ML, Trammell LH, Anderson BE. Cranial mor-
phological variation among contemporary Mexicans: Regional 
trends, ancestral affinities, and genetic comparisons. Ameri-
can Journal Physical Anthropology 2013;151(4):506– 517.

Hughes CE, Algee-Hewitt BF, Reineke R, Clausing E, Anderson 
BE. Temporal patterns of Mexican migrant genetic ancestry: 
Implications for identification. American Anthropologist 
2017;119(2):193– 208.

Hughes CE, Dudzik B, Algee-Hewitt BF, Jones A, Anderson BE. 
Understanding (mis)classification trends of Latin Americans 
in Fordisc 3.1: Incorporating cranial morphology, microgeo-
graphic origin, and admixture proportions for interpretation. 
Journal of Forensic Sciences 2019;64(2):353– 366.

Huxley J. Evolution:. The Modern Synthesis. Cambridge: The MIT 
Press; 1942.

Jantz R, Meadows Jantz L. Limb bone allometry in modern 
Euro-Americans. American Journal of Physical Anthropol-
ogy 2017;163(2):252– 263.

Jantz RL, Meadows Jantz L. Secular change in craniofacial mor-
phology. American Journal of Human Biology 2000;12(3): 
327– 338.

Jantz RL. Cranial change in Americans: 1850– 1975. Journal of 
Forensic Sciences 2001;46(4):784– 787.

Jantz RL, Spradley MK. Cranial Morphometric Evidence for Early 
Holocene Relationships and Population Structure. In: Owsley 
DW, Jantz RL, eds. Kennewick Man: The Scientific Investiga-
tion of an Ancient American Skeleton. College Station: Texas 
A&M University Press; 2014:472– 488.

Jantz RL, Meadows Jantz L. The remarkable change in Euro- 
American cranial shape and size. Human Biology 2016;88(1): 
56– 64.

Keita SOY. Studies of ancient crania from northern Africa. Amer-
ican Journal of Physical Anthropology 1990;83(1):35– 48.

Klepinger LL. Fundamentals of forensic anthropology. vol 1. John 
Wiley & Sons; 2006.

L’Abbé EN, Kenyhercz M, Stull KE, Keough N, Nawrocki S. Appli-
cation of Fordisc 3.0 to explore differences among crania of 
North American and South African Blacks and Whites. Jour-
nal of Forensic Sciences 2013;58(6):1579– 1583.

Livingston F. Anthropological implications of sickle cell gene dis-
tribution in West Africa. American Anthropologist 1958;60(3): 
533– 562.

Maier C, Craig A, Adams DM. Language use in ancestry research 
and estimation. Journal of Forensic Sciences 2021;66(1): 
11– 24.

Manthey L, Jantz RL, Bohnert M, Jellinghaus KJ. Secular change 
of sexually dimorphic cranial variables in Euro- Americans 
and Germans. International Journal of Legal Medicine 
2017;131(4):1113– 1118.

Manthey L, Jantz RL, Vitale A, Cattaneo C. Population specific 
data improves Fordisc®’s performance in Italians. Forensic 
Science International 2018;292:263.e1– 263.e7.

Meadows Jantz L, Jantz RL. Secular change in long bone length 
and proportion in the United States,1800– 1970. American 
Journal of Physical Anthropology 1999;110:57– 67.

Micheletti SJ, Bryc K, Esselmann SGA, Freyman WA, Moreno 
ME, Poznik GD, et al. Genetic consequences of the transat-
lantic slave trade in the Americas. American Journal of Human 
Genetics 2020;107(2):265– 277.

Noe- Bustamante L. Key facts about US Hispanics and their diverse 
heritage. Pew Research Center. https:// www . pewresearch . org 
/ fact - tank / 2019 / 09 / 16 / key - facts - about - u - s - hispanics / . Created 
September 16, 2019. Accessed March 28, 2021.

Novembre J, Johnson T, Bryc K, Kutalik Z, Boyko AR, Auton A, 
et al. Genes mirror geography within Europe. Nature 2008; 
456(7218):98– 101.

Ousley S, Jantz R, Freid D. Understanding race and human varia-
tion: Why forensic anthropologists are good at identifying 
race. American Journal of Physical Anthropology 2009;139(1): 
68– 76.

Pilloud MA, Hefner JT. Biological Distance Analysis: Forensic and 
Bio archae ol o gi cal Perspectives. London: Academic Press; 
2016.

Price AL, Butler J, Patterson N, Capelli C, Pascali VL, Scarnicci 
F, et al. Discerning the ancestry of European Americans in 
genetic associations studies. PLoS Genetics 2008;4(1):e26.

Reich D, Patterson N, Campbell D, Tandon A, Mazieres S, Ray, N, 
et al. Reconstructing native American population history. 
Nature 2012;488(7411):370– 374.

Relethford J, Lees FC, Crawford MH. Population structure and 
anthropometric variation in rural western Ireland: Migration 
and biological differentiation. Annals of Human Biology 
1980;7(5):411– 428.

Relethford J. Boas and beyond: Migration and craniometric 
 variation. American Journal of Human Biology 2004;16(4): 
379– 386.

Relethford JH, Lees FC. The use of quantitative traits in the study 
of human population structure. American Journal of Physical 
Anthropology 1982;25(S3):113– 132.

Relethford JH. Race and global patterns of phenotypic variation. 
American Journal of Physical Anthropology 2009;139(1): 
16– 22.

Ross A, Slice D, Ubelaker D, Falsetti A. Population affinities of 
19th century Cuban crania: implications for identification cri-
teria in south Floridian Cuban Americans. Journal of Foren-
sic Sciences 2004;49(1):1– 6.

Ross AH, Ubelaker DH, Kimmerle EH. Implications of dimor-
phism, population variation, and secular change in estimat-
ing population affinity in the Iberian Peninsula. Forensic 
Science International 2011;206(1– 3):214.e1– 214.e5.

Rubi- Castellanos R, Martínez- Cortés G, Francisco Muñoz- Valle J, 
González- Martín A, Cerda- Flores RM, Anaya- Palafox M, 
et al. Pre- Hispanic Mesoamerican demography approximates 
the present- day ancestry of Mestizos throughout the terri-
tory of Mexico. American Journal of Physical Anthropology 
2009;139(3):284– 294.

Ruiz- Linares A, Adhikari K, Acuña- Alonzo V, Quinto- Sanchez M, 
Jaramillo C, Arias W, et al. Admixture in Latin America: geo-
graphic structure, phenotypic diversity and self- perception 
of ancestry based on 7,342 individuals. PLoS Genetics 
2014;10(9):e1004572.

Smith J. W.E.B. Dubois, Felix von Luschan, and racial reform at 
the Fin de Siecle. Amerikastudien 2002;47:23– 38.

Smith JD. WEB Du Bois, Felix von Luschan, and racial reform at 
the fin de siècle. Amerikastudien/American Studies 2002: 
23– 38.

Spradley K, Stull KE, Hefner JT. Craniofacial secular change in 
recent Mexican migrants. Human Biology 2016;88(1):15– 29.

Spradley M. Biological Anthropological Aspects of the African 
Diaspora: Geographic Origins, Secular Trends, and Plastic 
Versus Genetic Influences Utilizing Craniometric Data [PhD 
dissertation. Knoxville: The University of Tennessee; 2006.

Spradley MK, Jantz RL, Robinson A, Peccerelli F. Demographic 
change and forensic identification: Problems in metric identi-
fication of Hispanic skeletons. Journal of Forensic Sciences 
2008;53(1):21– 28.

Spradley MK. Data from: Mexican, Guatemalan, and Migrant 
Craniometric Dataset. http:// doi . org / 10 . 5281 / zenodo . 4011199. 
Created September 1, 2020.



318 What Are We Really Estimating?

Spradley MK. Use of craniometric data to facilitate migrant iden-
tifications at the United States/Mexico border. American Jour-
nal of Physical Anthropology 2021;175(2):486– 496.

Stull KE. South African Long Bone Database. https:// zenodo . org 
/ record / 3950301 #  . YbO91mDMLIW. Created July 17, 2020.

Stull KE. An Osteometric Evaluation of Age and Sex Differences in 
the Long Bones of South African Children from the Western 
Cape [PhD dissertation]. University of Pretoria; 2013.

Tise ML, Kimmerle EH, Spradley MK. Craniometric variation of 
diverse populations in Florida: Identification challenges within 
a border state. Annals of Anthropological Practice 2014;38(1): 
111– 123.

Tishkoff SA, Reed FA, Friedlaender FR, Ehret C, Ranciaro A, Fro-
ment A, et al. The genetic structure and history of Africans 
and African Americans. Science 2009;324(5930):1035– 1044.

Washburn SL. Section of anthropology: The new physical anthro-
pology. Transactions of the New York Academy of Sciences 
1951;13(7 Series II):298– 304.

Weisensee KE, Jantz RL. Secular changes in craniofacial mor-
phology of the Portuguese using geometric morphometrics. 
American Journal of Physical Anthropology 2011;145(4): 
548– 559.

Wescott DJ, Jantz RL. Assessing craniofacial secular change in 
American blacks and whites using geometric morphometry. 
Modern Morphometrics in Physical Anthropology. Springer; 
2005:231– 245.

Williams FL, Belcher RL, Armelagos GJ. Forensic misclassifica-
tion of ancient Nubian crania: Implications for assumptions 
about human variation. Current Anthropology 2005;46(2): 
340– 346.


