Received: 9 December 2017 Revised: 19 September 2018 Accepted: 21 September 2018 DOI: 10.1111/1748-8583.12217 R E V I E W P A P E R When your resume is (not) turning you down: Modelling ethnic bias in resume screening Eva Derous1 | Ann Marie Ryan2 1 Department of Personnel Management, Work and Organizational Psychology, Ghent University, Ghent, Belgium 2 Michigan State University, East Lansing, Michigan, USA Correspondence Eva Derous, Department of Personnel Management, Work and Organizational Psychology, Ghent University, Henri Dunantlaan 2, Ghent 9000, Belgium. Email: [email protected] The review paper is based on a keynote held by th Organizational Psychology, May 18, 2017, Dublin, Hum Resour Manag J. 2018;1–18. wileyo Abstract Resume screening is the first hurdle applicants typically face when they apply for a job. Despite the many empirical studies showing bias at the resume‐screening stage, fairness at this funnelling stage has not been reviewed systematically. In this paper, a three‐stage model of biased resume screening is presented. We first discuss relevant theoretical perspec- tives (e.g., job market signalling and impression formation theories) to explain why resume screening is vulnerable to biased decision‐making and ethnic discrimination in particu- lar. On the basis of the best available evidence, we consider contingencies of ethnic discrimination in the applicant, the decision‐maker, and the broader context (e.g., organisation), as well as the effectiveness of interventions that might counter ethnic bias in resume screening. The paper ends with a critical agenda for further research and practice. KEYWORDS discrimination, diversity, ethnicity, recruitment, resume screening 1 | INTRODUCTION Despite decades of legislation and HR professionals' commitment to equal opportunities, ethnic minority members still suffer a weaker labour market position compared with equally qualified majorities (Shen, Chanda, D'Netto, & Monga, 2009). Human capital factors may explain some of the differences in hiring outcomes, but discrimination has also been put forward as a possible explanation (Derous & Ryan, 2018; Hoque & Noon, 1999). Ethnic minorities, for instance, still need to complete 50% more applications to get invited for a job interview when compared with equally qualified ethnic majorities (Zschirnt & Ruedin, 2016). e first author at the 18th Conference of the European Association for Work and Ireland. © 2018 John Wiley & Sons Ltdnlinelibrary.com/journal/hrmj 1 http://orcid.org/0000-0001-7874-5836 mailto:[email protected] https://doi.org/10.1111/1748-8583.12217 http://wileyonlinelibrary.com/journal/hrmj 2 DEROUS AND RYAN Resume screening, the first hurdle applicants typically face seems particularly vulnerable to hiring discrimination. Fairness in resume screening, however, is less well investigated when compared with the number of studies on the adverse impact of personnel selection tests (Outt.

1. Received: 9 December 2017 Revised: 19 September 2018
Accepted: 21 September 2018
DOI: 10.1111/1748-8583.12217
R E V I E W P A P E R
When your resume is (not) turning you down:
Modelling ethnic bias in resume screening
Eva Derous1 | Ann Marie Ryan2
1 Department of Personnel Management,
Work and Organizational Psychology, Ghent
University, Ghent, Belgium
2 Michigan State University, East Lansing,
Michigan, USA
Correspondence
Eva Derous, Department of Personnel
Management, Work and Organizational
Psychology, Ghent University, Henri
Dunantlaan 2, Ghent 9000, Belgium.
Email: [email protected]
The review paper is based on a keynote held by th

2. Organizational Psychology, May 18, 2017, Dublin,
Hum Resour Manag J. 2018;1–18. wileyo
Abstract
Resume screening is the first hurdle applicants typically face
when they apply for a job. Despite the many empirical
studies showing bias at the resume‐screening stage, fairness
at this funnelling stage has not been reviewed systematically.
In this paper, a three‐stage model of biased resume screening
is presented. We first discuss relevant theoretical perspec-
tives (e.g., job market signalling and impression formation
theories) to explain why resume screening is vulnerable to
biased decision‐making and ethnic discrimination in particu-
lar. On the basis of the best available evidence, we consider
contingencies of ethnic discrimination in the applicant, the
decision‐maker, and the broader context (e.g., organisation),
as well as the effectiveness of interventions that might
counter ethnic bias in resume screening. The paper ends with
a critical agenda for further research and practice.
KEYWORDS

3. discrimination, diversity, ethnicity, recruitment, resume
screening
1 | INTRODUCTION
Despite decades of legislation and HR professionals'
commitment to equal opportunities, ethnic minority members
still suffer a weaker labour market position compared with
equally qualified majorities (Shen, Chanda, D'Netto, &
Monga, 2009). Human capital factors may explain some of the
differences in hiring outcomes, but discrimination
has also been put forward as a possible explanation (Derous &
Ryan, 2018; Hoque & Noon, 1999). Ethnic minorities,
for instance, still need to complete 50% more applications to get
invited for a job interview when compared with
equally qualified ethnic majorities (Zschirnt & Ruedin, 2016).
e first author at the 18th Conference of the European
Association for Work and
Ireland.
© 2018 John Wiley & Sons Ltdnlinelibrary.com/journal/hrmj 1
http://orcid.org/0000-0001-7874-5836
mailto:[email protected]
https://doi.org/10.1111/1748-8583.12217
http://wileyonlinelibrary.com/journal/hrmj
2 DEROUS AND RYAN
Resume screening, the first hurdle applicants typically face

4. seems particularly vulnerable to hiring discrimination.
Fairness in resume screening, however, is less well investigated
when compared with the number of studies on the
adverse impact of personnel selection tests (Outtz, 2010). This
is remarkable for several reasons. First, resumes are
worldwide one of the most frequently used screening tools.
Over 98% of North‐American companies use resume
screening as the first selection hurdle (Piotrowski & Armstrong,
2006). Second, the quantity and quality of the supply
of applicants sets limit on what subsequent HRM practices can
achieve (Newman & Lyon, 2009; Thorndike, 1949).
Finally, the influential resource‐based view of the firm (Barney,
2001) spawned several decades of research to illustrate
that HRM practices are important sources of sustainable
competitive advantage. Although the HRM literature has inves-
tigated recruitment within this framework, the lack of focus on
this critical “entry gate” of resume screening is surprising.
This review focuses on resume screening. Resumes1 are
applicant‐generated, annotated career summaries of job
qualifications. HR professionals are expected to screen resumes
in an objective and fair way based on applicants' job‐
relevant characteristics (like work experiences or educational
credentials). Resume screening, however, might be
unfair when resulting in differential treatment discrimination or

5. differential effect discrimination (National Research
Council, 2004). Differential treatment discrimination arises
when applicants are treated in an unequal way based
on characteristics related to their group membership (like
screening out applicants based on ethnic‐sounding names
as appearing on resumes). Differential effect discrimination
results when applicants are treated in an unequal way
based on inadequately justified, nonjob‐related factors that
covary with minority characteristics.
The central aim of this paper is to review literature on ethnic
bias in the resume‐screening phase so as to inform
HRM practice around this critical point of organisational entry.
Below, we present a three‐stage model that integrates
theoretical perspectives to explain why resume screening is
vulnerable to biased decision‐making and ethnic discrimi-
nation (Section 2). This section is followed by research findings
on contingencies of discriminatory resume screening
and focuses on factors situated at three different levels (i.e.,
applicant information in resumes, the decision‐maker,
and the broader resume‐screening context) that may moderate
biased decision‐making against ethnic minorities
(Section 3). Whereas these first two sections consider
microlevel processes, we follow with a critical reflection upon
several practical HRM interventions to avert ethnic

6. discrimination in resume screening that are also situated at
different
levels of our model (i.e., the screening tool, the decision‐maker,
and the resume‐screening context; Section 4). We
conclude this review with a discussion of limitations, future
research opportunities, and implications for HR practice.
2 | BIASED RESUME SCREENING
Despite the abundant literature on hiring discrimination, little
research has considered why resume screening may be
prone to biased decision‐making. Integrating assumptions from
job market signalling and impression formation theo-
ries, we present a three‐stage model on biased decision‐making
in resume screening (see Figure 1, Part A). This model
states that when nonjob‐related, stigmatising applicant
information is presented in resumes and job‐related,
personalised information is rather limited (Stage 1: Applicant
information in resumes), decision‐makers might engage
in categorisation/Type 1 processing (Stage 2: Impression
formation), which increases the risk of biased applicant
impressions/ratings and discriminatory decision‐making and—
hence—may undermine workforce diversity (Stage 3:
Resume‐screening outcomes). Below, we explain each of the
stages in more detail.
2.1 | Stage 1: Applicant information in resumes
The first stage represents the building block of decision‐making

7. in resume screening, namely, applicant information in
resumes, and is based on job market signalling theory.
According to this theory (Spence, 1974), hiring managers and
job seekers have partly conflicting interests and will
communicate and interpret signals of the other party's unknown
characteristics (like applicants' competencies or the
organisation's culture) to obtain the biggest gains (like getting
the
best employees on board or getting hired). Typically, signalling
theory in selection considers the cues job seekers use
FIGURE 1 Model of biased resume screening [Colour figure can
be viewed at wileyonlinelibrary.com]
DEROUS AND RYAN 3
to make inferences about prospective employers (Carter &
Highhouse, 2014). However HR professionals also look
for signals of applicants' job qualifications, like work
experiences in resumes. Besides job‐related information,
resumes might also signal nonjob‐related information, like
applicants' social group status, through both explicit and
implicit cues. Applicants' skill sets on resumes, for instance, are
explicit/observable signals of applicants' job qualifi-
cations that—at the same time—may also reveal information
about applicants' chronological age in an implicit/subtle

8. way (Abrams, Swift, & Drury, 2016). Similarly, certain
extracurricular activities on resumes can subtly signal nonjob‐
related, stigmatising information like applicants' ethnic
background (Dovidio & Gaertner, 2000), which might affect
recruiters' information processing.
2.2 | Stage 2: Impression formation
The second stage of the model builds on impression formation
theories and represents the way applicant information
is further processed by decision‐makers. Impression formation
theories (like the continuum model; Fiske, Lin, &
Neuberg, 1999) specifically explain how signals of applicants'
group status affect HR professionals' decision‐making.
When only a limited amount of personalised information is
available (like a one‐page resume), individuals will auto-
matically engage in categorisation that in turn may activate
group stereotypes. Whereas these processes are auto-
matic or unconscious (i.e., Type 1 processing; Kahneman,
2003), resume screening can also involve high levels of
conscious involvement, such as assessments of the congruence
of applicants' characteristics with job and organisa-
tion characteristics (i.e., fit; Kristof‐Brown, 2000). When more
personalised information becomes available, recruiters
might engage in attribute‐based processing instead of
category‐based processing of applicant information (i.e., Type

9. 2 processing; Kahneman, 2003). The point, however, is that
Type 1 processing readily occurs during resume
screening because of the limited amount of applicant
information and, hence, may colour resume‐screening outcomes.
2.3 | Stage 3: Resume‐screening outcomes
The third stage of the model focuses on the outcomes that result
from the information processing stage during
resume screening. Specifically, Type 1 processes may affect
decision‐makers' first impressions/ratings and trigger
http://wileyonlinelibrary.com
4 DEROUS AND RYAN
discriminatory decision‐making. Perceptions of similarity, for
instance, may automatically induce interpersonal
attraction (Byrne, 1961) and explain why
recruiters/organisations tend to attract, select, and retain
applicants
that are similar to job incumbents (Schneider, 1987). Such
cognitive processes may lead towards homogeneous
workforces and may undermine organisational diversity. As we
will delineate more in discussing interventions,
when more personalised information becomes available about an
applicant, HR professionals should be better
able to monitor Type 1 processing, which may result in a more

10. fully informed and unbiased decision about the
applicant.
The three‐stage model illustrates why resume screening is
vulnerable to biased decision‐making (see Figure 1,
Part A), but not why ethnic discrimination occurs. Several
social‐economic and psychological theories may further
clarify why HRM systems may lead to such discrimination.
Typically, social‐economic theories stress macrolevel fac-
tors, like resource availability, institutional ideologies, industry
culture, and local labour market practices (Almeida,
Fernando, & Sheridan, 2012; Béret, Mendez, Paraponaris, &
Richez‐Battesti, 2003; Blair, Culkin, & Randle, 2003;
Sidanius & Pratto, 1999). Social‐psychological theories discuss
microlevel factors; among which are demographic dis-
similarity to others (Goldberg, 2005), individuals' need to
protect their in‐group and self‐identity (Tajfel & Turner,
1979), stereotypes and prejudiced attitudes (McConahay,
Hardee, & Batts, 1981), and personality and dispositions
that trigger prejudice (Altemeyer, 1981; Sidanius & Pratto,
1999). Table 1 describes and illustrates some of the most
cited theories on ethnic discrimination in hiring.
3 | CONTINGENCIES OF BIASED RESUME SCREENING
Resume screening may be more prone to ethnic discrimination
when applicants' ethnic minority status is cued in

11. resumes and job‐related, personalised information is limited
(Stage 1). Decision‐makers, however, may not be equally
affected by applicant information in resumes (Stage 2). Whether
they engage in categorisation or move to more
individualised information processing and decision‐making
(Stage 3) might depend on several contingencies that acti-
vate or inhibit categorisation (Kulik, Roberson, & Perry, 2007)
and that are situated in the applicant (ethnic markers,
qualifications), the decision‐maker (beliefs, attitudes,
experience), and the broader screening context (screening task,
job, organisation, society). Below, we review contingencies that
are discussed in the literature and that may moderate
impression formation and resume‐screening outcomes (see
Figure 1, Part B).
3.1 | Applicant
3.1.1 | Ethnicity cues
Applicants' ethnic‐sounding names are one of the most
investigated and explicit ethnic markers on resumes. In their
seminal correspondence audit study,2 Bertrand and
Mullainathan (2004) showed that resumes with
African‐American
sounding names received 50% less of a chance of a positive
callback compared with those with White‐sounding
names. Ethnic name discrimination in resume screening has

12. been reported around the world (see Zschirnt & Ruedin,
2016, for a meta‐analysis), and diverse lab studies also showed
applicants with ethnic‐sounding names to be least
liked and hired (e.g., Cotton, O'Neill, & Griffin, 2008). Yet
resumes may even include more explicit ethnic markers, like
cues about one's appearance in pictures attached to resumes.
Research, for instance, shows an overall preference for
light over dark‐skinned applicants, even among darker skinned
recruiters (Harrison & Thomas, 2009). Because visual
cues are immediately available and more rapidly processed than
verbal and behavioural cues, visual markers of one's
ethnic background (like skin tone) might trigger social
categorisation more than ethnic‐sounding names
(Weichselbaumer, 2017). Although it is common to ask
applicants for their picture in some countries (Belgium), in
others, it is not (like the Netherlands). Nevertheless, trends
towards pre‐screening via social media (SHRM, 2016)
make ethnic markers (like skin tone, ethnic attire, and even
speech) more salient in early screening stages and there-
fore potential for bias should be considered.
T
A
B

13. L
E
1
S
o
ci
o
‐e
co
n
o
m
ic
an
d
p
sy
ch
o
lo
g
ic
al
th
e
o
ri
e
s
o
n

14. e
th
n
ic
d
is
cr
im
in
at
io
n
in
h
ir
in
g
T
h
e
o
ri
e
s
(d
o
m
ai
n
/l
e
v

15. e
l/
d
e
sc
ri
p
ti
o
n
)
Il
lu
st
ra
ti
o
n
E
co
n
o
m
y
(m
ac
ro
le
v
e
l)

16. T
as
te
‐b
as
e
d
d
is
cr
im
in
at
io
n
th
e
o
ry
M
aj
o
ri
ty
w
o
rk
e
rs
/e
m
p

17. lo
y
e
rs
av
o
id
in
te
ra
ct
in
g
w
it
h
m
in
o
ri
ti
e
s
b
e
ca
u
se
o
f
ta
st

18. e
‐b
as
e
d
fa
ct
o
rs
in
st
e
ad
o
f
as
cr
ib
e
d
p
ro
d
u
ct
iv
it
y
(p
re
fe
re
n

19. ce
‐b
as
e
d
h
ir
in
g
).
H
ir
in
g
m
an
ag
e
rs
se
t
d
if
fe
re
n
t
g
ro
u
p
‐b
as

20. e
d
th
re
sh
o
ld
s
o
f
jo
b
‐
ir
re
le
v
an
t
ch
ar
ac
te
ri
st
ic
s
(li
k
e
e
th
n

21. ic
ac
ce
n
t)
,l
e
ad
in
g
to
lo
w
e
r
ab
ili
ty
e
st
im
at
io
n
s
o
f
m
in
o
ri

22. ty
m
e
m
b
e
rs
(i
.e
.,
ta
st
e‐
b
a
se
d
d
is
cr
im
in
a
ti
o
n
).
S
ta
ti
st
ic

23. a
l
d
is
cr
im
in
a
ti
o
n
o
cc
u
rs
w
h
e
n
o
b
se
rv
e
d
g
ro
u
p
d
if
fe
re

24. n
ti
al
s
in
p
ro
d
u
ct
iv
it
y
ar
e
m
is
ta
k
e
n
ly
at
tr
ib
u
te
d
to
g
ro
u
p

25. id
e
n
ti
ty
(N
e
ils
o
n
&
Y
in
g
,
2
0
1
6
)
S
ta
ti
st
ic
al
d
is
cr
im
in

26. at
io
n
th
e
o
ry
M
aj
o
ri
ty
w
o
rk
e
rs
/e
m
p
lo
y
e
rs
d
is
av
o
w
m
in

27. o
ri
ty
in
d
iv
id
u
al
s
b
e
ca
u
se
th
e
y
co
n
si
d
e
r
m
in
o
ri
ty
g
ro

28. u
p
s
as
a
w
h
o
le
le
ss
p
ro
d
u
ct
iv
e
fr
o
m
an
e
co
n
o
m
ic
p
o

29. in
t
o
f
v
ie
w
.
S
o
ci
o
lo
g
y
/s
o
ci
al
p
sy
ch
o
lo
g
y
(m
ac
ro
le
v
e

30. l
an
d
m
ic
ro
le
v
e
l)
S
tr
u
ct
u
ra
l
d
is
cr
im
in
at
io
n
th
e
o
ry
S
o

31. ci
e
ti
e
s
sh
ar
e
id
e
o
lo
g
ie
s
th
at
ju
st
if
y
g
ro
u
p
‐b
as
e
d
in
e
q
u
al

32. it
ie
s
an
d
fo
rm
al
ly
/i
n
fo
rm
al
ly
e
m
b
e
d
d
e
d
so
ci
e
ta
l/
o
rg
an

33. is
at
io
n
al
p
ro
ce
ss
e
s
su
p
p
o
rt
su
ch
id
e
o
lo
g
ie
s.
H
ir
in
g
b
as

34. e
d
o
n
e
m
p
lo
y
e
e
re
co
m
m
e
n
d
at
io
n
s
in
st
e
ad
o
f
ad
v
e
rt
is

35. e
m
e
n
ts
le
ad
s
to
d
if
fe
re
n
ti
al
e
ff
e
ct
d
is
cr
im
in
at
io
n
if

36. e
m
p
lo
y
e
e
s
sy
st
e
m
at
ic
al
ly
re
co
m
m
e
n
d
ap
p
lic
an
ts
th
at
ar

37. e
si
m
ila
r
to
th
e
m
se
lv
e
s
(W
al
d
in
g
e
r
&
L
ic
h
te
r,
2
0
0
3
).

38. G
ro
u
p
p
o
si
ti
o
n
th
e
o
ri
e
s,
re
al
is
ti
c
g
ro
u
p
co
n
fl
ic
t,
an
d
co

39. m
p
e
ti
ti
o
n
th
e
o
ri
e
s
S
o
ci
al
g
ro
u
p
s
ar
e
in
co
m
p
e
ti
ti

40. o
n
o
v
e
r
v
al
u
e
d
re
so
u
rc
e
s
an
d
p
e
rc
e
iv
e
d
th
re
at
s
fr
o
m
lo

41. ss
o
f
re
so
u
rc
e
s
re
su
lt
s
in
d
is
cr
im
in
at
io
n
.
S
o
ci
al
‐e
co
n
o

42. m
ic
th
re
at
(li
k
e
re
ce
ss
io
n
)
fo
st
e
re
d
h
ir
in
g
d
is
cr
im
in
at
io
n
ag

43. ai
n
st
e
th
n
ic
fe
m
al
e
ap
p
lic
an
ts
(K
in
g
,
K
n
ig
h
t,
&
H
e
b
l,
2

44. 0
1
0
).
P
sy
ch
o
lo
g
y
(m
ic
ro
le
v
e
l)
R
e
la
ti
o
n
al
d
e
m
o
g
ra

45. p
h
y
th
e
o
ry
M
aj
o
ri
ty
m
e
m
b
e
rs
co
m
p
ar
e
th
e
ir
d
e
m
o
g
ra

46. p
h
ic
ch
ar
ac
te
ri
st
ic
s
to
m
in
o
ri
ty
m
e
m
b
e
rs
.
P
e
rc
e
iv
e
d
d
is

47. si
m
ila
ri
ty
le
ad
s
to
n
e
g
at
iv
e
at
ti
tu
d
e
s
an
d
b
e
h
av
io
u
r
to
w

48. ar
d
s
m
in
o
ri
ty
m
e
m
b
e
rs
.
R
ac
e
si
m
ila
ri
ty
e
ff
e
ct
s
w
e
re
o

49. b
se
rv
e
d
o
n
ca
n
d
id
at
e
s'
o
v
e
ra
ll
in
te
rv
ie
w
as
se
ss
m
e
n
ts
an
d

50. jo
b
o
ff
e
r
d
e
ci
si
o
n
s
b
y
W
h
it
e
re
cr
u
it
e
rs
(G
o
ld
b
e
rg
,
2

51. 0
0
5
)
S
o
ci
al
ca
te
g
o
ri
sa
ti
o
n
an
d
so
ci
al
id
e
n
ti
ty
th
e
o
ry

52. M
aj
o
ri
ti
e
s'
n
e
e
d
fo
r
an
d
p
ro
te
ct
io
n
o
f
a
p
o
si
ti
v
e
id
e
n
ti

53. ty
(s
e
lf
/
g
ro
u
p
)
m
ig
h
t
in
st
ig
at
e
in
‐g
ro
u
p
fa
v
o
u
ri
ti
sm
,
w

54. h
ic
h
m
ay
le
ad
to
d
is
cr
im
in
at
io
n
ag
ai
n
st
m
in
o
ri
ti
e
s
R
e
su
m

55. e
s
o
f
h
ig
h
ly
e
th
n
ic
‐i
d
e
n
ti
fi
e
d
ap
p
lic
an
ts
re
ce
iv
e
d
lo
w
e

56. r
jo
b
su
it
ab
ili
ty
ra
ti
n
g
s
th
an
e
q
u
al
ly
q
u
al
if
ie
d
b
u
t
le
ss

57. e
th
n
ic
al
ly
id
e
n
ti
fi
e
d
co
u
n
te
rp
ar
ts
(D
e
ro
u
s,
N
g
u
y
e
n
,
&

58. R
y
an
,
2
0
0
9
).
S
te
re
o
ty
p
e
co
n
te
n
t
m
o
d
e
ls
S
te
re
o

59. ty
p
e
s
le
ad
to
d
is
cr
im
in
at
io
n
.
S
te
re
o
ty
p
e
s
ar
e
m
aj
o
ri
ti
e
s'

60. in
d
iv
id
u
al
b
e
lie
fs
(c
o
g
n
it
iv
e
sc
h
e
m
a
an
d
e
xp
e
ct
at
io
n
s)

61. ab
o
u
t
g
ro
u
p
ch
ar
ac
te
ri
st
ic
s
th
at
ar
e
ap
p
lie
d
to
m
in
o
ri
ty

62. an
d
th
at
ar
e
ty
p
ic
al
ly
(b
u
t
n
o
t
n
e
ce
ss
ar
ily
)
n
e
g
at
iv
e
in
n

63. at
u
re
.
M
in
o
ri
ti
e
s
m
ay
b
e
aw
ar
e
o
f
m
aj
o
ri
ti
e
s'
st
e
re
o
ty
p

64. ic
al
b
e
lie
fs
ab
o
u
t
th
e
m
(i
.e
.,
m
e
ta
‐s
te
re
o
ty
p
e
s)
an
d
e
v
e

65. n
in
te
g
ra
te
th
e
se
id
e
as
in
to
th
e
ir
se
lf
‐c
o
n
ce
p
t
(i
.e
.,
se
lf
‐s

66. te
re
o
ty
p
in
g
).
D
e
te
rm
in
in
g
w
h
o
g
e
ts
h
ir
e
d
fo
r
w
h
ic
h
jo
b

67. ty
p
e
d
e
p
e
n
d
s
o
n
ap
p
lic
an
ts
'
m
at
ch
to
th
e
jo
b
st
e
re
o
ty
p
e

68. ,
lik
e
ap
p
lic
an
ts
fr
o
m
st
e
re
o
ty
p
ic
al
ly
w
ar
m
g
ro
u
p
s
h
ir
e

69. d
m
o
re
fo
r
so
ci
al
jo
b
s
(C
u
d
d
y
,
G
lic
k
,
&
B
e
n
in
g
e
r,
2
0

70. 1
1
).
(N
e
g
at
iv
e
)
M
e
ta
‐
st
e
re
o
ty
p
e
s
m
ig
h
t
w
it
h
h
o
ld
q

71. u
al
if
ie
d
m
in
o
ri
ti
e
s
fr
o
m
ap
p
ly
in
g
fo
r
ce
rt
ai
n
jo
b
s
(W
ill
e
&

72. D
e
ro
u
s,
2
0
1
7
).
T
h
e
o
ri
e
s
o
n
o
ld
‐ f
as
h
io
n
e
d
ra
ci
sm
,
m

73. o
d
e
rn
ra
ci
sm
,
an
d
m
o
ti
v
at
io
n
to
re
sp
o
n
d
w
it
h
o
u
t
p
re

74. ju
d
ic
e
R
ac
ia
l/
e
th
n
ic
p
re
ju
d
ic
e
ar
e
ra
ci
al
/e
th
n
ic
m
aj
o
ri
ti
e

75. s'
p
re
d
o
m
in
an
tl
y
n
e
g
at
iv
e
at
ti
tu
d
e
s
to
w
ar
d
s
ra
ci
al
/e
th
n
ic

76. m
in
o
ri
ti
e
s
th
at
m
ig
h
t
re
su
lt
in
e
it
h
e
r
b
la
ta
n
t
o
r
m
o
re

77. su
b
tl
e
fo
rm
s
o
f
d
is
cr
im
in
at
io
n
ag
ai
n
st
th
e
se
m
in
o
ri
ti
e
s

78. o
r
su
p
p
re
ss
th
e
m
.
M
o
d
e
rn
ra
ci
sm
af
fe
ct
e
d
h
ir
in
g
d
is
cr

79. im
in
at
io
n
w
h
e
n
a
le
g
it
im
at
e
au
th
o
ri
ty
fi
g
u
re
p
ro
v
id
e
d

80. a
b
u
si
n
e
ss
‐r
e
la
te
d
ju
st
if
ic
at
io
n
fo
r
d
is
cr
im
in
at
io
n
(B
ri
e
f,
D

81. ie
tz
,
C
o
h
e
n
,
P
u
g
h
,
&
V
as
lo
w
,
2
0
0
0
).
T
h
e
o
ri
e
s
o
n

82. p
e
rs
o
n
al
it
y
(a
u
th
o
ri
ta
ri
an
is
m
an
d
so
ci
al
d
o
m
in
an
ce
o

83. ri
e
n
ta
ti
o
n
)
P
e
rs
o
n
al
it
y
u
n
d
e
rl
y
in
g
m
aj
o
ri
ti
e
s'
w
o
rl
d

84. v
ie
w
s
af
fe
ct
in
g
e
th
n
ic
/r
ac
ia
l
p
re
ju
d
ic
e
re
ac
ti
o
n
s
to
w
ar
d
s
m

85. in
o
ri
ti
e
s
ar
e
ri
g
h
t‐
w
in
g
au
th
o
ri
ta
ri
an
is
m
(c
o
n
fo
rm
it
y

86. to
tr
ad
it
io
n
,
su
b
m
is
si
o
n
to
p
o
w
e
rf
u
l
le
ad
e
rs
,
an
d
p

87. o
lit
ic
al
o
ri
e
n
ta
ti
o
n
),
an
d
so
ci
al
d
o
m
in
an
ce
o
ri
e
n
ta
ti
o
n

88. (e
n
d
o
rs
e
m
e
n
t
o
f
g
ro
u
p
h
ie
ra
rc
h
y
)
M
aj
o
ri
ty
o
f
th
e
m

89. e
m
b
e
rs
h
ig
h
in
so
ci
al
d
o
m
in
an
ce
o
ri
e
n
ta
ti
o
n
(S
D
O
)
w

90. e
re
le
ss
lik
e
ly
to
se
le
ct
a
p
o
te
n
ti
al
te
am
m
e
m
b
e
r
fr
o
m

91. lo
w
‐s
ta
tu
s
g
ro
u
p
s
(li
k
e
ra
ci
al
m
in
o
ri
ty
ca
n
d
id
at
e
s)
th
an

92. th
o
se
lo
w
in
S
D
O
(U
m
p
h
re
ss
,
S
im
m
o
n
s,
B
o
sw
e
ll,
&
T
ri

93. an
a,
2
0
0
8
).
DEROUS AND RYAN 5
6 DEROUS AND RYAN
Although studies typically focus on explicit markers on
resumes, resumes also contain more implicit cues to one's
ethnicity, such as applicants' affiliations with socio‐cultural
groups, that may trigger biased information processing.
Cole, Rubin, Feild, and Giles (2007) illustrated that although
HR professionals believed work experience to be the
strongest influence on ratings of applicants' employability,
ratings were mostly affected by affiliations as mentioned
on resumes. Moreover, multiple ethnic cues may also interact
and increase category salience such that resumes of
highly ethnically identified applicants (e.g., with
ethnic‐sounding name and affiliations) might receive lower
employ-
ability ratings due to increased out‐group status (Derous et al.,
2009; Kang, DeCelles, Tilcsik, & Jun, 2016).

94. 3.1.2 | Qualifications
Job‐related cues on resumes typically temper categorisation.
Contemporary models of discrimination (Dovidio &
Gaertner, 2000) suggest less discrimination if applicants either
clearly possess the requested job qualifications or
do not possess them at all. However, when qualifications are
moderate, a more ambiguous situation is created in
which discriminatory hiring decisions could be justified. For
example, Almeida et al. (2012) noted that a lack of
recognition of experience and credentials gained overseas as
well as concerns about language skills affected the
employment outcomes of professional immigrants. Hence,
ethnic discrimination may occur more when decisions
can be rationalised based on some other factors than applicants'
ethnicity (Brief et al., 2000).
3.2 | Decision‐maker
Studies on ethnic bias in resume screening have somewhat
disregarded individual differences in decision‐makers' per-
sonality, beliefs about others (worldviews/stereotypes),
prejudiced attitudes, and recruiting experience. This may be
explained by the long‐standing tradition of audit studies in
which decision‐makers' dispositions are typically not
accessible (for an exception, see Rooth, 2010). Studies that
measure individual differences are predominantly con-

95. ducted in the lab and include beliefs in societal group
hierarchies, like social dominance orientation (Sidanius &
Pratto,
1999), prejudiced attitudes like modern racism (McConahay et
al., 1981), and motivation to respond without preju-
dice (Plant & Devine, 1998, 2009). Although findings are
sometimes mixed, negative beliefs/attitudes towards others
typically result in larger ethnic bias (Derous & Ryan, 2018).
Indeed, although blatant discrimination is still reported,
research also shows that recruiters may not act upon their
prejudice in resume‐screening situations where bias would
be very obvious (e.g., if directly attributable to themselves) or
when there is an external reason not to react in a
biased way (e.g., because of organisational policies; Brief et al.,
2000).
Because individuals have become more sensitive to politically
correct standards to disavow discrimination
(Dovidio & Gaertner, 2000), researchers directed their attention
to implicit prejudice. Rooth (2010), for instance,
showed that Swedish recruiters were less likely to invite
Arab‐Muslim minority applicants when they had more
negative, implicit attitudes towards Arab‐Muslims (as measured
with implicit association tests; Greenwald, Banaji,
& Nosek, 20153). There is, nevertheless, an ongoing debate
about the validity of implicit attitude measures and

96. whether any relation with behaviour can be expected and
established outside the lab (Oswald, Mitchell, Blanton,
Jaccard, & Tetlock, 2015).
Finally, there is also a debate about the role of decision‐makers'
expertise. Predominantly discussed in interviewing
studies, some show more experienced recruiters to be less
biased towards stigmatised applicants, whereas others show
more bias. De Meijer, Born, van Zielst, and van der Molen
(2007), for instance, found that experienced recruiters used
more irrelevant information when judging ethnic minorities,
perhaps because they feel overconfident and hence,
engage in Type 1 processing.
3.3 | Context
Contextual cues, like job and organisation characteristics as
well as the way decision‐makers perform resume‐
screening tasks, also affect impression formation and
resume‐screening outcomes.
DEROUS AND RYAN 7
3.3.1 | Job characteristics
Contextual factors studied most are job stereotypes. Stereotypes
not only exist for people but also for jobs, and
these may orient HR professionals towards viewing applicants

97. as more or less suitable for certain jobs (i.e., cognitive
matching model; Trope & Liberman, 1993). For instance, King,
Mendoza, Madera, Hebl, and Knight (2006) showed
that the effect of applicants' ethnic names on overall resume
evaluation was not significant when applicants'
suitability for high—versus low—status jobs was controlled for,
suggesting job stereotypes affected resume‐screening
outcomes. Audit studies (Carlsson & Rooth, 2008) further
demonstrated lower callback ratios for ethnic minorities
who applied for occupations with higher external client contact
(restaurant workers and shop sales assistants). Yet
mixed findings are also reported (Booth, Leigh, & Varganova,
2012; Derous, Ryan, & Serlie, 2015; Weichselbaumer,
2017), perhaps because matching effects may depend on a
mixture of contextual cues (Goldberg, Finkelstein, Perry,
& Konrad, 2004). For instance, Dietz, Joshi, Esses, Hamilton,
and Gabarrot (2015) showed that bias against qualified
immigrants was mitigated when the fit with the clientele was
emphasised. Derous, Pepermans, and Ryan (2017) fur-
ther showed that discriminatory resume screening of the same
applicant with varying skin tone (dark vs. light)
depended on the particular combination of several job and
industry characteristics.
3.3.2 | Organisation/task characteristics

98. Aside from some studies that considered the demographic
diversity of organisations and their clients (Almeida,
Fernando, Hannif, & Dharmage, 2015), surprisingly little
research considers the role of other organisational character-
istics (like size and policies) and screening task characteristics
(like time and financial pressure) on the way HR profes-
sionals screen resumes (Almeida et al., 2012). However, the
HRM literature has established that line managers do not
fully engage in HRM duties because of time pressures and
prioritising operational over HR tasks (McGovern, Gratton,
Hope‐Hailey, Stiles, & Truss, 1997; Woodrow & Guest, 2014).
Such time and motivational constraints likely
contribute to using categorisation to move through resume
screening more quickly.
In sum, screening out competent people during the
resume‐screening stage because of nonjob‐related applicant
factors, decision‐makers' dispositions, and contextual factors is
worrisome and costly, especially when labour markets
are tight and talented workers are hard to find. Hence, effective
interventions are much needed to avert discrimina-
tory resume screening.
4 | INTERVENTIONS TO AVERT BIASED RESUME
SCREENING
Understanding contingencies may help both researchers and

99. practitioners evaluating selection practices as well as
interventions to mitigate biased decision‐making. Based on the
best available evidence, this section critically dis-
cusses interventions to avert discriminatory resume screening
that are situated at the level of the screening tool,
the decision‐maker, and the resume‐screening context (see
Figure 1, Part C).
4.1 | Screening tool
At the level of the screening tool, three different types of
interventions are discussed, namely, anonymisation,
personalisation, and standardisation.
4.1.1 | Anonymisation
Anonymous application procedures such as blind auditions
(Goldin & Rouse, 2000), blind interviewing (Buijsrogge,
Derous, & Duyck, 2016), anonymous resume screening (Åslund
& Skans, 2012), or “whitened”4 resumes (Kang
et al., 2016) aim to combat illegal discrimination by blotting or
concealing personal identifiers. Although blind
auditions and interviewing have been found to be effective,
studies on anonymous resume screening have shown
positive (Åslund & Skans, 2012; Kang et al., 2016), null, or
even negative effects (Behaghel, Crépon, & Le Barbanchon,

100. 8 DEROUS AND RYAN
2015; Hiscox et al., 2017; Krause, Rinne, & Zimmermann,
2012). The French government, therefore, decided to aban-
don the idea of making anonymous resume screening mandatory
in the recruitment procedures of their public
employment services (Behaghel et al., 2015).
Why may anonymous resume screening fail? Both signalling
and impression formation theory can help us under-
stand unintended side effects of anonymous resume screening.
Aside from very explicit markers, resumes might also
contain more implicit cues, such as extracurricular activities,
that might signal in a subtle way applicants' ethnic
minority status (Dovidio & Gaertner, 2000). Further, with
anonymous resume screening, resumes are
decontextualised and depersonalised. As a result, HR
professionals have less possibility to understand and attenuate
negative signals (e.g., from gaps in resumes or lower
qualifications; Behaghel et al., 2015) and therefore might—par-
adoxically—engage in categorisation.
4.1.2 | Personalisation
Instead of altering or removing information, applicants could
provide more personalised information, for instance, by
means of video resumes or social network sites. Video resumes
are short videotaped messages of 1–2 min in which

101. an applicant presents himself/herself to potential employers.
Much like paper resumes, video resumes present candi-
date information in an asynchronous way (one can view the
resume information at any time, at any place). However,
they differ from paper resumes in that they provide more and
different cues and allow applicants to show relevant com-
petencies. Interestingly, although ethnic minority applicants
consider video resumes as more fair than paper‐and‐pencil
resumes, HR managers report concerns as more nonjob‐relevant
information (like physical attractiveness) is included
(Hiemstra & Derous, 2015).
Applicants may also provide more personalised information
through social media as individuals increasingly
include links to their social network sites on their resumes
(SHRM, 2016). HR professionals may use “cybervetting”
(i.e., the screening of social media sites like Facebook and
LinkedIn) to extract information from applicants to inform
personnel decisions (Berkelaar & Buzzanell, 2014). About 44%
of HR managers believe candidates' public social
network sites to be good sources for assessing potential (SHRM,
2016). Indeed, social network information might
provide more and different types of personalised information
(like interests, values, and interactions with other users)
that reflect more typical behaviours than resumes do. Therefore,

102. these sources might have incremental validity
beyond traditional screening tools.
However, findings are inconclusive about the validity of social
network information. Kluemper, Rosen, and
Mossholder (2012) found that personality traits could be
reliably assessed via Facebook profiles and were predictive
of future work behaviour beyond applicants' self‐rated
personality and intelligence scores. Yet Van Iddekinge,
Lanivich, Roth, and Junco (2016) showed that across a broad
array of KSAOs, ratings of applicants' Facebook pages
did not predict job performance (i.e., supervisor ratings,
turnover intentions, and actual turnover). Moreover, HR
professionals tended to favour White and female applicants
when they screened applicants' Facebook information,
resulting in adverse impact. Furthermore, the availability of job
irrelevant information may impair the overall validity
of unstandardised social media despite the fact that typical
performance might be reflected in these media.
4.1.3 | Standardisation
Given that standardisation of selection procedures reduces the
chance of judgmental biases in both recruiters and
applicants (Highhouse, Doverspike, & Guion, 2015), structured
application forms might also be considered. This allows
organisations to score applicants' competencies and background

103. information in a more objective way than with
applicant‐generated resumes that lack uniformity. Standardised
application forms may also provide applicants fewer
possibilities to use impression formation tactics and faking than
applicant‐generated resumes (Derous & Ryan,
2018). Equally, more structured, job‐related social network sites
like LinkedIn might make these sources less
vulnerable to biased decision‐making than less structured media
and at the same time increase their validity.
Corroborating this, van de Ven, Bogaert, Serlie, Brandt, and
Denissen (2017) recently showed accurate personality
DEROUS AND RYAN 9
estimates based on LinkedIn profiles. However, the
effectiveness of screening tools will also depend on
decision‐maker's characteristics and the way the decision‐maker
uses the tool.
4.2 | Decision‐maker
This section considers the feasibility and effectiveness of four
different types of interventions focused on the deci-
sion‐maker, whether in HR or line management: selecting out
prejudiced raters, offering training, holding raters
accountable, and replacing human decision‐makers with
algorithms.

104. 4.2.1 | Selection
Selecting out prejudiced raters seems obvious given effects of
raters' particular worldviews (like social dominance ori-
entation) and prejudiced attitudes on judgments (i.e., theories
on modern racism and authoritarian personality; see
Table 1). This intervention, however, might not be feasible as
those chosen to screen resumes might do so because
of their technical expertise or hiring authority (Brief et al.,
2000). Indeed, globally, HRM responsibilities related to
selection are increasing the responsibility of line managers
rather than HR professionals (Brewster, Brookes, & Gollan,
2015). Furthermore, explicit prejudice measures may be
susceptible to socially desirable responding, and their predic-
tive validity in the context of choosing resume screeners still
needs to be demonstrated. The same applies to other
measures of individual predispositions, like social dominance
orientation. Therefore, other interventions like training
are considered.
4.2.2 | Training
Recruiters could be trained to increase awareness about
judgmental biases in resume screening. Dietz et al. (2015),
for instance, demonstrated how developing a common identity
across groups may be a basis for inclusive HRM strat-

105. egies and reduce hiring discrimination against high skilled
immigrants, for example, when a fit with a diverse clientele
is emphasised (i.e., common in‐group identity model; Gaertner
& Dovidio, 2000). Building on social psychological the-
ories on categorisation, stereotyping, and motivation to respond
without prejudice (see Table 1), Devine, Forscher,
Austin, and Cox (2012) further showed evidence for a
multi‐faceted implicit prejudice habit‐breaking intervention
that lasted 8 weeks and included different elements such as
contact, perspective taking, stereotype replacement
(i.e., reconsideration of actions and thoughts to replace biased
response), counter‐stereotypical imaging (i.e., imagin-
ing examples of out‐group members who counter commonly
held stereotypes), and individuating (i.e., considering
out‐group members as individuals instead of stereotyped group
members). However, these interventions are typically
developed for and tested in educational settings and not yet in
corporate contexts, like resume screening.
4.2.3 | Accountability
Holding recruiters accountable for their decisions could also
hold them from acting in prejudiced ways. However,
Self, Mitchell, Mellers, Tetlock, and Hildreth (2015) showed
that type of accountability instruction matters. Holding

106. people accountable for certain outcomes, like an increase in the
representation of minority applicants to face legal
or other pressures (i.e., identity‐conscious accountability),
resulted in more pro‐minority bias and less qualified
applicants than when recruiters were held accountable for
making fair selection decisions based on job‐relevant con-
siderations (i.e., identity‐blind accountability).
Panel recruitment in which a team instead of a single rater
screens resumes may be another avenue to increase fair-
ness. Following predictions from contemporary theories on
prejudice (seeTable 1), the presence of significant others (like
colleagues) might externally motivate recruiters to respond
without prejudice and—hence—to avoid being perceived as
discriminatory and/or to avoid repercussions (Plant & Devine,
1998, 2009). Ethnically mixed screening panels might
even lead to less biased decision‐making. When recruiters work
in ethnically mixed screening panels, they might get
to know each other and, as a consequence, might move from
social categorisation (Type 1 processing) to
individualisation (Type 2 processing) (Fiske et al., 1999;
Kahneman, 2003). Building further on predictions from the
social
10 DEROUS AND RYAN

107. identity theory, ethnic minority and majority recruiters might
even develop a common in‐group identity, which also
reduces the chance on biased decision‐making (i.e., common
in‐group identity model; see Gaertner & Dovidio, 2000).
In general, HRM research has clearly established that HR
departments play a key role in enabling line managers
to successfully implement effective HR practices (Trullen,
Stirpe, Bonache, & Valverde, 2016). Creating accountability
as well as providing recognition for unbiased hiring can be an
important lever in ensuring effective resume screening.
4.2.4 | Algorithms
Instead of screening, training, and making decision‐makers
accountable for fair screening, one could also replace
human decision‐makers by automated resume readers or
algorithms. This idea is not new: In the 1970s, both the
Pentagon and IBM already replaced human decision‐makers by
algorithms to narrow down the large piles of resumes
(O'Neil, 2016). Automated resume readers may boost efficiency
by saving time, money, and energy. The French cos-
metic company L'Oréal, for instance, developed an algorithm to
measure cultural fit based on only three open‐ended
questions candidates answered on their mobile phone, which
released recruiters from the time‐consuming procedure
of screening many resumes.

108. Proponents argue that algorithms may be more accurate and
predictive than human decision‐makers (Danieli,
Hillis, & Luca, 2016). Although professionals still prefer
holistic information processing (Kuncel, Klieger, & Ones,
2014), Kuncel, Klieger, Connelly, and Ones (2013) showed that
mechanical data combination methods resulted in
more than 50% improvement in the prediction of work and
academic criteria when compared with more holistic, intu-
itive methods. Other researchers further showed that algorithms
can rate applicants' accomplishment narratives as
reliably as human raters (Campion, Campion, Campion, &
Reider, 2016), can predict applicants' personality traits
and social/communication skills reasonably well from nonverbal
cues extracted from video resumes (Nguyen &
Gatica‐Perez, 2016) or from Facebook likes (Youyou, Kosinski,
& Stillwell, 2015), and can even predict which candi-
dates would most likely become involved in shooting or be
accused of abuse as police officers (Chalfin et al., 2016).
Still, opponents remain cautious about the overall validity and
fairness of automated resume‐screening tools: If
people have the ability to identify how algorithms work, they
might beat them too through strategic behaviour (like
drafting resumes to fit the system). Although some biases like
friendship bias (Nguyen, 2006) might be countered,

109. automated resume screening might still be vulnerable to
impression management and even faking behaviour (Waung,
McAuslan, DiMambro, & Mięgoć, 2017) as it might be as
difficult for algorithms as human decision‐makers to filter
this out. Moreover, when algorithms are built upon human
decision‐makers' subtly prejudiced rules, they might be
even more precise and persistent in discriminatory
decision‐making than any human decision‐maker. For instance,
Saint George's Hospital Medical School of South‐London was
found guilty of discrimination in its admission policy
because their automated resume reader used nonjob‐related
criteria (like misspellings), which were correlated with
applicants' ethnic group membership (Lowry & MacPherson,
1988).
4.3 | Context
In addition to interventions in the resume‐screening tool and
with decision‐makers, organisations as well as society at
large could develop policies and procedures to record
discriminatory screening practices, to monitor recruitment
messages/sources and to guarantee competence‐based
assessments through discrimination‐free employment
arrangements.
4.3.1 | Recording

110. Organisations could use different techniques, like
correspondence audits (see earlier) and mystery shopping tests,
to
measure and record hiring discrimination at the organisational
and industry level. Mystery shopping involves a con-
federate who makes checks against specified criteria in order to
get insight into system delivery. The self‐regulating
body of recruitment offices in Flanders, for instance, had
fictitious commissioning clients deliberately ask discrimina-
tory questions to recruitment offices in order to uncover
discriminatory intentions (Federgon, 2013). Similar research
DEROUS AND RYAN 11
has asked subsidised cleaning companies to send out only
native, Belgian cleaners to potential employers. Whereas
correspondence audits register actual discrimination, mystery
shopping only capture one's intention to act in a dis-
criminatory way. Hence, one point of debate is whether mystery
shopping might be used in a punitive rather than
a self‐monitoring way. Also, discriminatory intentions might
reflect many different underlying, biasing processes that
are typically not directly measured with these tools (e.g.,
preferences, beliefs about economic productivity and com-
petitiveness, and social dominance; see Table 1). Another point
of discussion is who may administer such tests,

111. whether to encourage HR managers and CEOs to organise audits
and mystery shopping themselves or to consider
using qualified research institutes and/or governmental bodies.
4.3.2 | Targeted recruitment
Organisations may also attract more minority job seekers
through targeted recruitment strategies like diversity state-
ments and the portrayal of minority employees in job
advertisements. These targeted recruitment strategies build on
the social identity theory: Applicants who perceive the best fit
with their social/individual identity may feel most
attracted to the organisation and may apply. Hence, by
increasing the number of ethnic minority applicants that
apply, targeted recruitment strategies may be a way to avert
adverse impact and to increase fairness in assessment.
Though, because effects of such targeted recruitment strategies
on the reduction of adverse impact are rather mixed
(Avery & McKay, 2006), researchers turned their attention
towards qualification‐based targeted recruitment strategies,
aimed to attract highly qualified ethnic minorities. Newman and
Lyon (2009) indeed showed that job postings
designed to attract highly qualified ethnic minorities (e.g.,
requiring applicants high in conscientiousness) resulted
in less adverse impact. However and although promising,
qualification‐based targeted recruitment strategies still

112. tend to disregard stereotypical ideas applicants might have
about job qualifications/requirements. Indeed, applicants
too might have ideas about the stereotypical beliefs out‐group
members hold about in‐group members (i.e., meta‐
stereotypes; Vorauer, Main, & O'Connell, 1998), and they may
even integrate such meta‐stereotypes into their
own self‐concept (self‐stereotyping). Building further on
stereotype content models, Wille and Derous (2017)
showed that organisations should be cautious about sprinkling
job ads with requirements that (minority) candidates
hold negative meta‐stereotypes about, particularly if those
requirements are communicated in dispositional ways
(like “This company is looking for applicants who are high in
integrity”). Such job ads might discourage (highly
qualified) minority candidates to apply instead of attracting
them.
Besides recruitment messages, organisations may also consider
their recruitment sources as some might be less
frequently consulted/used by minority than by majority job
seekers. For instance, video resumes are potentially dis-
criminatory against minority groups who may have less tech
access (i.e., differential effect discrimination; Heathfield,
2016). Remarkably, bias might even be encoded in algorithms
of search engines (Hajian, Bonchi, & Castillo, 2016).

113. Sweeney (2013) showed that algorithms for public record
websites were more likely to imply criminal activities (like
arrest records) with searches for Black‐sounding names than
White‐sounding names. Finally, labour market interme-
diaries (temporary work agencies and public employment
services) can play a role in assuming some level of recruit-
ment and selection functions for hiring organisations (Bonet,
Cappelli, & Hamori, 2013). However, Ingold and
Valizade (2017) demonstrated that although intermediaries may
increase likelihood of hiring from disadvantaged
groups, employer selective hiring criteria still led to lower
employability of marginalised groups.
4.3.3 | Employment (economic/societal)
Finally, more radical interventions consider the rethinking of
employment relations at the economic/societal level to
reduce hiring discrimination by promoting open, accessible
labour markets. One way to realise this is through new
types of employment arrangements. eLancing5 (Aguinis &
Lawal, 2013) might address this call: Employers' evaluation
of eLancers based on their past assignments resembles work
sample tests that are known to be valid predictors of
future work performances. Furthermore, hiring for “eLancing”
assignments may be blind, so that freelancers' ethnicity
does not affect decision‐making.

114. 12 DEROUS AND RYAN
Open Badges ecosystems are another way to create more
accessible, discrimination‐free labour markets. The open
badges ecosystem (https://openbadges.org), originally launched
by Mozilla, encompasses a method for packaging
information about one's individual accomplishments, skills,
qualities, or interests in portable image files as a digital
badge that subsequently can be displayed via job seekers' social
media platforms and consulted by potential
employers. The system's infrastructure ensures that badges are
reliably issued by institutions and endorsed within
the open badges ecosystem (e.g., as approved by the Department
of Education or other reliable institutions). Through
open badges backpacks, applicants might provide potential
employers with very personalised, timely, job relevant,
and certified/objective information about their competencies
during the initial screening stage, which in turn might
help countering social categorisation and hiring discrimination.
Indeed, according to impression formation theories
(e.g., Fiske et al., 1999), the more personalised information a
recruiter/HR professional receives about an applicant
in the early screening stage (e.g., through information in open
badges), the more she/he might engage in Type 2 pro-

115. cessing (individuating) and move away from Type 1 processing
(social categorisation).
Technological developments (like Open Badges) not only offer
alternatives for discriminatory resume screening
but also redesign HR practices fundamentally. Whereas
traditionally, companies attract, screen, and select applicants
by presenting job requirements/offers, through technological
developments like open badges, the power nexus shifts
to the applicant, who will attract, screen, and even select
companies/jobs by showing their competencies (i.e., com-
panies bidding for applicants).
5 | DISCUSSION
Diversity in organisations can be effectively managed through
HRM practices. Remarkably, despite societal debates
about fair hiring (Feintzeig, 2016), fairness of HR tools like
resume screening has received less research attention,
especially when compared with the extensive literature on other
selection tools, like the job interview. Resume
screening, however, is worldwide one of the most frequently
used screening tools that determines the quantity, qual-
ity, and diversity of applicant pools. We aimed to address this
gap by formulating a model of biased decision‐making
in resume screening (Figure 1) that includes contingencies of
resume screening as well as interventions to avert dis-

116. criminatory screening, all related to relevant theories and
empirical findings on ethnic discrimination. It further allows
to identify mixed findings and literature gaps. Hence, the model
might steer further research on discriminatory
resume screening as well as interventions to avert this. Below,
we summarise the most important opportunities for
further research, followed by implications for practitioners.
5.1 | Research opportunities
5.1.1 | Applicant information and screening tool
At the heart of the model (Figure 1, Part A) is a cognitive
mechanism of biased decision‐making that we based on
assumptions from job market signalling and impression
formation theory. Research could further investigate
microlevel processes of impression formation in a bottom‐up
way, for example, by tracing decision‐makers' attention
to both nonjob‐related/job‐related and implicit/explicit
information, by investigating whether attention paid to
different resume cues differently affects
categorisation/individualisation, and by investigating their
effect on
resume‐screening outcomes (see for a similar approach on
interview bias: Buijsrogge et al., 2016). These findings
might also provide useful information regarding the
effectiveness (validity) of anonymisation versus personalisation

117. of resumes. Related, more empirical studies are needed on the
effectiveness of structuring applicant information
to avert Type 1 processing in decision‐makers. Further and as
already mentioned, we considered ethnic discrimina-
tion. Systematic reviews of effects of other stigmatising cues in
resumes (and combined effects) are also needed,
given the paucity of reviews on judgmental biases in resume
screening and the necessity to generalise findings to dif-
ferent stigmatised groups.
https://openbadges.org
DEROUS AND RYAN 13
5.1.2 | Decision‐maker
Surprisingly little research considers individual difference
variables that make decision‐makers vulnerable tobiased
decision‐
making in the resume‐screening stage. Hence, more research is
needed on reliable/valid methods of choosing resume
screeners for unbiased decision‐making (like measures of
prejudiced attitudes), on training programs that might change
recruiter bias in resume‐screening contexts, and on the
usefulness of ethnically mixed screening panels. Also,
researchers
have only started evaluating the way algorithms are developed

118. and validated. As human decision‐makers are already being
replaced by algorithms in organisations, algorithms should be
compared with humans regarding reliability/validity of
decisions, levels of adverse impact, vulnerability to impression
management (including faking), and perceived fairness.
5.1.3 | Context
Not only individual difference variables but also contextual
variables may increase the likelihood of Type 1 processing
and trigger biased decision‐making in resume screening. Future
research could consider moderating effect of
microlevel factors like job and resume‐screening task
characteristics (like available time and other task‐related pres-
sures). For instance, some recruitment sources do not reach
potentially qualified candidates from ethnic minority
communities or might discourage qualified applicants to apply.
However, also mesolevel factors (like organisational
diversity policies) and macrolevel factors (like labour market
situation, work arrangements, politics, and cultural
habits) might affect recruitment practices and should be further
considered. For instance, affirmative action plans
cannot be realised with anonymous resume screening as one
needs to be aware of social category membership. As
regards macrolevel factors, one could further investigate
whether effectiveness of Open Badges depends on the kind

119. of information included in badges (like cognitive test
performance).
5.2 | Practical implications
Organisations may keep track on how decision‐makers evaluate
applicants and could set‐up specific training pro-
grams in which recruiters are informed about judgmental
mechanisms and biases (e.g., induced by cultural differ-
ences) as well as potential effective interventions to avert
biases, such as the use of qualification‐based targeted
recruitment and competence‐based screening tools like
structured application forms. However, targeted recruitment
initiatives as well as more technology‐driven applications (like
automated resume‐screening tools) should always be
critically evaluated to assure they are valid and free from bias,
regardless of whether they are developed outside or
inside one's organisation. Relatedly, organisations should keep
up to date about, set‐up, and communicate their pol-
icies on cybervetting so that both recruiters and applicants are
fully aware of the kind of information that might be
evaluated online. Finally, more attention should be paid to
recruiters/decision‐makers' working conditions. Stress
levels (due to time or any other task‐related pressure) should be
reduced as these might increase the risk of Type
1 processes and biased decision‐making in resume screening.

120. Ethnic minority applicants and career counsellors can benefit
too from literature insights. For instance, applicants may
be informed about explicit/implicit cues to both job‐related and
nonjob‐related information on their resumes as well as
about organisational context factors (like client preferences).
Career counsellors might also help applicants to properly
interpret job requirements and check critically whether
applicants' qualifications are not too ambiguously presented but
clearly match the job requirements to minimise risks on
discrimination. Finally, social media profiles may be kept up to
date
and best include professional information only. In general,
applicants might consider recruitment devices that allow for
more competence‐based, individualised screening (like badges,
structured competency lists, or perhaps video resumes).
6 | CONCLUSION
In conclusion, despite the widespread use of resume screening
as well as the plethora of studies on ethnic discrim-
ination in hiring, a model on biased decision‐making in resume
screening that integrates findings was still lacking.
14 DEROUS AND RYAN
One of the strengths of this paper is that we addressed this
literature gap by highlighting an underlying mechanism of

121. biased decision‐making, contingencies that might moderate bias,
and interventions that might avert judgmental bias
in resume screening. This review not only revealed several
interesting insights but also showed that there is still much
to be discovered. Specifically, we discussed biased resume
screening in the context of ethnic discrimination without
considering other stigmatising factors than ethnic makers (or
intersectional effects). Further, we considered biased
resume screening from the HR professional/organisation
perspective rather than the applicant/job seeker perspec-
tive. Applicants' perceived discrimination, however, may be as
important as actual discrimination. Third, this review
focused on operational HR processes to manage organisational
diversity rather than tactical and strategical HR pro-
cesses that also play a role (Shen et al., 2009). As we are among
the first to summarise and integrate literature on
biased decision‐making in resume screening, still more aspects
can be looked at to build an even more comprehensive
model. Finally, latest technology‐driven tools/systems (like
algorithms and Open Badges) reflect not only the
changing nature of our labour market and talent
acquisition/management in HRM but also the potential to
counter
bias in early screening stages, if carefully thought through,
developed, and implemented by HR professionals.

122. CONFLICT OF INTEREST
The authors declare that they have no conflict of interest.
ENDNOTES
1 Some disciplines (medicine, education, and academia) expect
extensive curriculum vitae (CVs) that offer a complete career
history with detailed information on professional activities.
This review focuses on resumes, which are a more abridged
career summary; however, much of the research reviewed may
be applicable to CV screening.
2 Employment audit studies investigate labour market outcomes
of applicants who are equally qualified for a job but differ in
nonjob‐related characteristics, like ethnic background. In
correspondence audit studies, pairs of matched resumes are sent
to the same employer and the type and number of call‐backs are
registered.
3 Implicit association tests are reaction time measures in which
respondents are asked to match concepts (Arab‐sounding
names) to attributes (good/bad). The speed with which
respondents do so is considered to reflect implicit attitudes
towards certain minorities.
4 Whitened resumes are ones where identifying information is
concealed or blotted, for example, by using one's middle
name instead of first name if the former is more race‐neutral or
by removing words referring to racial group membership
(like [Black] students' association).
5 eLancing websites are crowdsourcing internet marketplaces
where employers place assignments (e.g., software develop-
ment and translations) that freelancers can bid for. Work is
completed on an as‐needed basis and freelancers are

123. evaluated on the quality of their previous assignments.
ORCID
Eva Derous http://orcid.org/0000-0001-7874-5836
REFERENCES
Abrams, D., Swift, H. J., & Drury, L. (2016). Old and
unemployable? How age‐based stereotypes affect willingness to
hire job
candidates. Journal of Social Issues, 72(1), 105–121.
https://doi.org/10.1111/josi.12158
Aguinis, H., & Lawal, S. O. (2013). eLancing: A review and
research agenda for bridging the science–practice gap. Human
Resource Management Review, 23(1), 6–17.
https://doi.org/10.1016/j.hrmr.2012.06.003
Almeida, S., Fernando, M., Hannif, Z., & Dharmage, S. C.
(2015). Fitting the mould: The role of employer perceptions in
immi-
grant recruitment decision‐making. International Journal of
Human Resource Management, 26(22), 2811–2832. https://doi.
org/10.1080/09585192.2014.1003087
Almeida, S., Fernando, M., & Sheridan, A. (2012). Revealing
the screening: Organisational factors influencing the
recruitment
of immigrant professionals. International Journal of Human
Resource Management, 23(9), 1950–1965. https://doi.org/
10.1080/09585192.2011.616527
Altemeyer, B. (1981). Enemies of freedom: Understanding right
wing authoritarianism. Winnipeg, CA: Jossey‐Bass.

124. http://orcid.org/0000-0001-7874-5836
https://doi.org/10.1111/josi.12158
https://doi.org/10.1016/j.hrmr.2012.06.003
https://doi.org/10.1080/09585192.2014.1003087
https://doi.org/10.1080/09585192.2014.1003087
https://doi.org/10.1080/09585192.2011.616527
https://doi.org/10.1080/09585192.2011.616527
DEROUS AND RYAN 15
Åslund, O., & Skans, O. N. (2012). Do anonymous job
application procedures level the playing field? ILR Review,
65(1),
82–107. https://doi.org/10.1177/001979391206500105
Avery, D. R., & McKay, P. F. (2006). Target practice: An
organizational impression management approach to attracting
minor-
ity and female job applicants. Personnel Psychology, 59(1),
157–187. https://doi.org/10.1111/j.1744‐6570.2006.00807.x
Barney, J. (2001). Resource‐based theories of competitive
advantage: A ten‐year retrospective on the resource‐based view.
Journal of Management, 27(6), 643–650.
https://doi.org/10.1016/S0149‐2063(01)00115‐5
Behaghel, L., Crépon, B., & Le Barbanchon, T. (2015).
Unintended effects of anonymous résumés. American Economic
Journal:
Applied Economics, 7(3), 1–27.
https://doi.org/10.1257/app.20140185
Béret, P., Mendez, A., Paraponaris, C., & Richez‐Battesti, N.
(2003). R&D personnel and human resource management in
mul-
tinational companies: Between homogenization and

125. differentiation. International Journal of Human Resource
Management,
14(3), 449–468. https://doi.org/10.1080/0958519022000031843
Berkelaar, B. L., & Buzzanell, P. M. (2014). Cybervetting,
person–environment fit, and personnel selection. Journal of
Applied
Communication Research, 42(4), 456–476.
https://doi.org/10.1080/00909882.2014.954595
Bertrand, M., & Mullainathan, S. (2004). Are Emily and Greg
more employable than Lakisha and Jamal? A field experiment
on
labor market discrimination. American Economic Review,
94(4), 991–1013. https://doi.org/10.1257/0002828042002561
Blair, H., Culkin, N., & Randle, K. (2003). From London to Los
Angeles: A comparison of local labour market processes in the
US and UK film industries. International Journal of Human
Resource Management, 14(4), 619–633. https://doi.org/
10.1080/0958519032000057619
Bonet, R., Cappelli, P., & Hamori, M. (2013). Labor market
intermediaries and the new paradigm for human resources.
Acad-
emy of Management Annals, 7(1), 341–392.
https://doi.org/10.5465/19416520.2013.774213
Booth, A. L., Leigh, A., & Varganova, E. (2012). Does ethnic
discrimination vary across minority groups? Oxford Bulletin of
Eco-
nomics and Statistics, 74(4), 547–573.
https://doi.org/10.1111/j.1468‐0084.2011.00664.x
Brewster, C., Brookes, M., & Gollan, P. J. (2015). The
institutional antecedents of the assignment of HRM

126. responsibilities to
line managers. Human Resource Management, 54(4), 577–597.
https://doi.org/10.1002/hrm.21632
Brief, A., Dietz, J., Cohen, R., Pugh, S., & Vaslow, J. (2000).
Just doing business: Modern racism and obedience to authority
as
explanations for employment discrimination. Organizational
Behavior and Human Decision Processes, 81(1), 72–97.
https://doi.org/10.1006/obhd.1999.2867
Buijsrogge, A., Derous, E., & Duyck, W. (2016). Often biased
but rarely in doubt: How initial reactions to stigmatized
applicants affect interviewer confidence. Human Performance,
29(4), 275–290. https://doi.org/10.1080/
08959285.2016.1165225
Byrne, D. (1961). Interpersonal attraction and attitude
similarity. Journal of Abnormal and Social Psychology, 62(3),
713–715.
https://doi.org/10.1037/h0044721
Campion, M. C., Campion, M. A., Campion, E. D., & Reider, M.
H. (2016). Initial investigation into computer scoring of
candidate essays for personnel selection. Journal of Applied
Psychology, 101(7), 958–975. https://doi.org/10.1037/
apl0000108
Carlsson, M., & Rooth, D.‐O. (2008). Is it your foreign name or
foreign qualifications? IZA Discussion Paper No. 3810.
Retrieved from http://ftp.iza.org/dp3810.pdf
Carter, N. T., & Highhouse, S. (2014). You will be known by
the company you keep: Understanding the social identity
concerns of job seekers. In K. Y. T. Yu, & D. M. Cable (Eds.),
The Oxford handbook of recruitment (pp. 454–462). New York:

127. Oxford University Press.
Chalfin, A., Danieli, O., Hillis, A., Jelveh, Z., Luca, M.,
Ludwig, J., & Mullainathan, S. (2016). Productivity and
selection of
human capital with machine learning. American Economic
Review, 106(5), 124–127. https://doi.org/10.1257/aer.
p20161029
Cole, M., Rubin, R., Feild, H., & Giles, W. (2007). Recruiters'
perceptions and use of applicant resume information. Applied
Psychology: An International Review, 56, 319–343.
Cotton, J. L., O'Neill, B. S., & Griffin, A. (2008). The “name
game”: Affective and hiring reactions to first names. Journal of
Managerial Psychology, 23, 18–39.
https://doi.org/10.1108/02683940810849648
Cuddy, A. J. C., Glick, P., & Beninger, A. (2011). The
dynamics of warmth and competence judgments, and their
outcomes in
organizations. Research in Organizational Behavior, 31, 73–98.
https://doi.org/10.1016/j.riob.2011.10.004
Danieli, O., Hillis, A., & Luca, M. (2016). How to hire with
algorithms. Harvard Business Review, October. Retrieved from
https://hbr.org/2016/10/how-to-hire-with-algorithms
De Meijer, L. A. L., Born, PH. M., van Zielst, J., & van der
Molen, H. T. (2007). Analyzing judgments of ethnically diverse
appli-
cants during personnel selection. International Journal of
Selection and Assessment, 15(2), 139–152. https://doi.org/
10.1111/j.1468‐2389.2007.00376.x
https://doi.org/10.1177/001979391206500105

128. https://doi.org/10.1111/j.1744-6570.2006.00807.x
https://doi.org/10.1016/S0149-2063(01)00115-5
https://doi.org/10.1257/app.20140185
https://doi.org/10.1080/0958519022000031843
https://doi.org/10.1080/00909882.2014.954595
https://doi.org/10.1257/0002828042002561
https://doi.org/10.1080/0958519032000057619
https://doi.org/10.1080/0958519032000057619
https://doi.org/10.5465/19416520.2013.774213
https://doi.org/10.1111/j.1468-0084.2011.00664.x
https://doi.org/10.1002/hrm.21632
https://doi.org/10.1006/obhd.1999.2867
https://doi.org/10.1080/08959285.2016.1165225
https://doi.org/10.1080/08959285.2016.1165225
https://doi.org/10.1037/h0044721
https://doi.org/10.1037/apl0000108
https://doi.org/10.1037/apl0000108
http://ftp.iza.org/dp3810.pdf
https://doi.org/10.1257/aer.p20161029
https://doi.org/10.1257/aer.p20161029
https://doi.org/10.1108/02683940810849648
https://doi.org/10.1016/j.riob.2011.10.004
https://hbr.org/2016/10/how-to-hire-with-algorithms
https://doi.org/10.1111/j.1468-2389.2007.00376.x
https://doi.org/10.1111/j.1468-2389.2007.00376.x
16 DEROUS AND RYAN
Derous, E., Nguyen, H.‐H., & Ryan, A. M. (2009). Hiring
discrimination against Arab minorities. Human Performance,
22(4),
297–320. https://doi.org/10.1080/08959280903120261
Derous, E., Pepermans, R., & Ryan, A. M. (2017). Ethnic
discrimination during résumé screening: Interactive effects of
appli-

129. cants' ethnic salience with job context. Human Relations, 70(7),
860–882. https://doi.org/10.1177/0018726716676537
Derous, E., & Ryan, A. M. (2018). By any other name. In U.
Klehe, & E. A. J. van Hooft (Eds.), The Oxford handbook of
job loss and job search (pp. 501–522). Oxford: Oxford
University Press. https://doi.org/10.1093/oxfordhb/
9780199764921.013.017
Derous, E., Ryan, A. M., & Serlie, A. W. (2015). Double
jeopardy upon resumé screening. Personnel Psychology, 68(3),
659–696. https://doi.org/10.1111/peps.12078
Devine, P. G., Forscher, P. S., Austin, A. J., & Cox, W. T. L.
(2012). Long‐term reduction in implicit race bias: A prejudice
habit‐
breaking intervention. Journal of Experimental Social
Psychology, 48(6), 1267–1278. https://doi.org/10.1016/j.
jesp.2012.06.003
Dietz, J., Joshi, C., Esses, V. M., Hamilton, L. K., & Gabarrot,
F. (2015). The skill paradox: Explaining and reducing
employment
discrimination against skilled immigrants. International Journal
of Human Resource Management, 26(10), 1318–1334.
https://doi.org/10.1080/09585192.2014.990398
Dovidio, J. F., & Gaertner, S. L. (2000). Aversive racism and
selection decisions: 1989 and 1999. Psychological Science,
11(4),
315–319. https://doi.org/10.1111/1467‐9280.00262
Federgon. (2013). Jaarverslag 2012. Brussels, Belgium:
Federgon.
Feintzeig, R. (2016). The boss doesn't want your resume. The

130. Wall Street Journal, 5 January. Retrieved from https://www.wsj.
com/articles/the‐boss‐doesnt‐want‐your‐resume‐1452025908
Fiske, S. T., Lin, M., & Neuberg, S. (1999). The continuum
model, ten years later. In S. Chaiken, & Y. Trope (Eds.), Dual
process
theories in social psychology (pp. 231–254). New York:
Guilford Press.
Gaertner, S. L., & Dovidio, J. F. (2000). Reducing intergroup
bias: The common ingroup identity model. Philadelphia, PA:
Psychology Press.
Goldberg, C. B. (2005). Relational demography and
similarity‐attraction in interview assessments and subsequent
offer deci-
sions. Group & Organization Management, 30(6), 597–624.
https://doi.org/10.1177/1059601104267661
Goldberg, C. B., Finkelstein, L. M., Perry, E. L., & Konrad, A.
M. (2004). Job and industry fit: The effects of age and gender
matches on career progress outcomes. Journal of Organizational
Behavior, 25(7), 807–829. https://doi.org/10.1002/
job.269
Goldin, C., & Rouse, C. (2000). Orchestrating impartiality: The
impact of “blind” auditions on female musicians. American
Eco-
nomic Review, 90, 715–741. Retrieved from
www.jstor.org/stable/117305
Greenwald, A. G., Banaji, M. R., & Nosek, B. A. (2015).
Statistically small effects of the Implicit AssociationTest can
have soci-
etally large effects. Journal of Personality and Social
Psychology, 108(4), 553–561.

131. https://doi.org/10.1037/pspa0000016
Hajian, S., Bonchi, F., & Castillo, C. (2016). Algorithmic bias.
In Proceedings of the 22nd ACM SIGKDD International Confer-
ence on Knowledge Discovery and Data Mining ‐ KDD'16 (pp.
2125–2126). New York: ACM Press.
Harrison, M. S., & Thomas, K. M. (2009). The hidden prejudice
in selection. Journal of Applied Social Psychology, 39(1),
134–168. https://doi.org/10.1111/j.1559‐1816.2008.00433.x
Heathfield, S. M. (2016). Do employers want video resumes?
Retrieved from https://www.thebalance.com/video‐resumes‐
1918979
Hiemstra, A. M. F., & Derous, E. (2015). Video résumés
portrayed: Findings and challenges. In I. Nikolaou, & J. K.
Oostrom
(Eds.), Employee recruitment, selection, and assessment:
Contemporary issues for theory and practice (pp. 44–60). Hove,
UK: Psychological Press.
Highhouse, S., Doverspike, D., & Guion, R. (2015). Essentials
of personnel assessment and selection. New York: Routledge.
Hiscox, M. J., Oliver, T., Ridgway, M., Arcos‐Holzinger, L.,
Warren, A., & Willis, A. (2017). Going blind to see more
clearly:
Unconscious bias in Australian Public Service shortlisting
processes. Retrieved from https://pmc.gov.au/sites/default/
files/publications/BETA‐report‐going‐blind‐to‐see‐more
clearly.pdf
Hoque, K., & Noon, M. (1999). Racial discrimination in
speculative applications. Human Resource Management Journal,
9(3),

132. 71–82. https://doi.org/10.1111/j.1748‐8583.1999.tb00204.x
Ingold, J., & Valizade, D. (2017). Employers' recruitment of
disadvantaged groups. Human Resource Management Journal,
27(4), 530–547. https://doi.org/10.1111/1748‐8583.12154
Kahneman, D. (2003). A perspective on judgment and choice:
Mapping bounded rationality. American Psychologist, 58(9),
697–720. https://doi.org/10.1037/0003‐066X.58.9.697
https://doi.org/10.1080/08959280903120261
https://doi.org/10.1177/0018726716676537
https://doi.org/10.1093/oxfordhb/9780199764921.013.017
https://doi.org/10.1093/oxfordhb/9780199764921.013.017
https://doi.org/10.1111/peps.12078
https://doi.org/10.1016/j.jesp.2012.06.003
https://doi.org/10.1016/j.jesp.2012.06.003
https://doi.org/10.1080/09585192.2014.990398
https://doi.org/10.1111/1467-9280.00262
https://www.wsj.com/articles/the-boss-doesnt-want-your-
resume-1452025908
https://www.wsj.com/articles/the-boss-doesnt-want-your-
resume-1452025908
https://doi.org/10.1177/1059601104267661
https://doi.org/10.1002/job.269
https://doi.org/10.1002/job.269
http://www.jstor.org/stable/117305
https://doi.org/10.1037/pspa0000016
https://doi.org/10.1111/j.1559-1816.2008.00433.x
https://www.thebalance.com/video-resumes-1918979
https://www.thebalance.com/video-resumes-1918979
https://doi.org/10.1111/j.1748-8583.1999.tb00204.x
https://doi.org/10.1111/1748-8583.12154
https://doi.org/10.1037/0003-066X.58.9.697

133. DEROUS AND RYAN 17
Kang, S. K., DeCelles, K. A., Tilcsik, A., & Jun, S. (2016).
Whitened résumés. Administrative Science Quarterly, 61(3),
469–502.
https://doi.org/10.1177/0001839216639577
King, E. B., Knight, J. L., & Hebl, M. R. (2010). The influence
of economic conditions on aspects of stigmatization. Journal of
Social Issues, 66(3), 446–460. https://doi.org/10.1111/j.1540-
560.2010.01655.x
King, E. B., Mendoza, S. A., Madera, J. M., Hebl, M. R., &
Knight, J. L. (2006). What's in a name? A multiracial
investigation of
the role of occupational stereotypes in selection decisions.
Journal of Applied Social Psychology, 36(5), 1145–1159.
https://doi.org/10.1111/j.0021‐9029.2006.00035.x
Kluemper, D. H., Rosen, P. A., & Mossholder, K. W. (2012).
Social networking websites, personality ratings, and the
organizational context. Journal of Applied Social Psychology,
42(5), 1143–1172. https://doi.org/10.1111/j.1559‐
1816.2011.00881.x
Krause, A., Rinne, U., & Zimmermann, K. F. (2012).
Anonymous job applications of fresh Ph.D. economists.
Economics Letters,
117(2), 441–444. https://doi.org/10.1016/j.econlet.2012.06.029
Kristof‐Brown, A. L. (2000). Perceived applicant fit:
Distinguishing between recruiters' perceptions of person‐job
and person‐
organization fit. Personnel Psychology, 53(3), 643–671.
https://doi.org/10.1111/j.1744‐6570.2000.tb00217.x
Kulik, C. T., Roberson, L., & Perry, E. L. (2007). The

134. multiple‐category problem: Category activation and inhibition
in the hiring
process. Academy of Management Review, 32(2), 529–548.
https://doi.org/10.5465/AMR.2007.24351855
Kuncel, N. R., Klieger, D. M., Connelly, B. S., & Ones, D. S.
(2013). Mechanical versus clinical data combination in
selection
and admissions decisions: A meta‐analysis. Journal of Applied
Psychology, 98(6), 1060–1072. https://doi.org/10.1037/
a0034156
Kuncel, N. R., Klieger, D. M., & Ones, D. S. (2014). In hiring,
algorithms beat instinct. Harvard Business Review, May.
https://
hbr.org/2014/05/in-hiring-algorithms-beat-instinct
Lowry, S., & MacPherson, G. (1988). A blot on the profession.
British Medical Journal, 296(6623), 657–658.
McConahay, J. B., Hardee, B. B., & Batts, V. (1981). Has
racism declined in America? Journal of Conflict Resolution,
25(4),
563–579. https://doi.org/10.1177/002200278102500401
McGovern, P., Gratton, L., Hope‐Hailey, V., Stiles, P., & Truss,
C. (1997). Human resource management on the line? Human
Resource Management Journal, 7(4), 12–29.
https://doi.org/10.1111/j.1748‐8583.1997.tb00286.x
National Research Council (2004). Measuring racial
discrimination. Washington, DC: The National Academies
Press.
Neilson, W., & Ying, S. (2016). From taste‐based to statistical
discrimination. Journal of Economic Behavior and Organization,

135. 129, 116–128. https://doi.org/10.1016/j.jebo.2016.06.001
Newman, D. A., & Lyon, J. S. (2009). Recruitment efforts to
reduce adverse impact: Targeted recruiting for personality, cog-
nitive ability, and diversity. Journal of Applied Psychology,
94(2), 298–317. https://doi.org/10.1037/a0013472
Nguyen, L. S., & Gatica‐Perez, D. (2016). Hirability in the
wild: Analysis of online conversational video resumes. IEEE
Trans-
actions on Multimedia, 18(7), 1422–1437.
https://doi.org/10.1109/TMM.2016.2557058
Nguyen, N. T. (2006). Recruiters' use of friendship in resume
assessment: When qualification does not always make the cut.
Applied H.R.M Research, 11(1), 1–14.
O'Neil, C. (2016). Weapons of math destruction: How big data
increases inequality and threatens democracy. New York:
Crown.
Oswald, F. L., Mitchell, G., Blanton, H., Jaccard, J., & Tetlock,
P. E. (2015). Using the IAT to predict ethnic and racial discrim-
ination: Small effect sizes of unknown societal significance.
Journal of Personality and Social Psychology, 108(4), 562–571.
https://doi.org/10.1037/pspa0000023
Outtz, J. L. (2010). Adverse impact: Implications for
organizational staffing and high stakes selection. New York:
Routledge.
Piotrowski, C., & Armstrong, T. (2006). Current recruitment
and selection practices: A national survey of Fortune 1000
firms.
North American Journal of Psychology, 8(3), 489–496.
Retrieved from http://org8220renner.alliant.wikispaces.net/file/

136. view/Selection_recrutiment_fortune+1000.pdf/33002629/Selecti
on_recrutiment_fortune+1000.pdf
Plant, E. A., & Devine, P. G. (1998). Internal and external
motivation to respond without prejudice. Journal of Personality
and
Social Psychology, 75(3), 811–832.
https://doi.org/10.1037/0022‐3514.75.3.811
Plant, E. A., & Devine, P. G. (2009). The active role of
prejudice: Unpacking the intentions guiding control efforts.
Journal of
Personality and Social Psychology, 96(3), 640–652.
https://doi.org/10.1037/a0012960
Rooth, D. O. (2010). Automatic associations and discrimination
in hiring: Real world evidence. Labour Economics, 17,
523–534. https://doi.org/10.1016/j.labeco.2009.04.005
Schneider, B. (1987). The people make the place. Personnel
Psychology, 40(3), 437–453. https://doi.org/10.1111/j.1744‐
6570.1987.tb00609.x
https://doi.org/10.1177/0001839216639577
https://doi.org/10.1111/j.1540-560.2010.01655.x
https://doi.org/10.1111/j.0021-9029.2006.00035.x
https://doi.org/10.1111/j.1559-1816.2011.00881.x
https://doi.org/10.1111/j.1559-1816.2011.00881.x
https://doi.org/10.1016/j.econlet.2012.06.029
https://doi.org/10.1111/j.1744-6570.2000.tb00217.x
https://doi.org/10.5465/AMR.2007.24351855
https://doi.org/10.1037/a0034156
https://doi.org/10.1037/a0034156
https://hbr.org/2014/05/in-hiring-algorithms-beat-instinct
https://hbr.org/2014/05/in-hiring-algorithms-beat-instinct
https://doi.org/10.1177/002200278102500401

137. https://doi.org/10.1111/j.1748-8583.1997.tb00286.x
https://doi.org/10.1016/j.jebo.2016.06.001
https://doi.org/10.1037/a0013472
https://doi.org/10.1109/TMM.2016.2557058
https://doi.org/10.1037/pspa0000023
http://org8220renner.alliant.wikispaces.net/file/view/Selection_
recrutiment_fortune+1000.pdf/33002629/Selection_recrutiment_
fortune+1000.pdf
http://org8220renner.alliant.wikispaces.net/file/view/Selection_
recrutiment_fortune+1000.pdf/33002629/Selection_recrutiment_
fortune+1000.pdf
https://doi.org/10.1037/0022-3514.75.3.811
https://doi.org/10.1037/a0012960
https://doi.org/10.1016/j.labeco.2009.04.005
https://doi.org/10.1111/j.1744-6570.1987.tb00609.x
https://doi.org/10.1111/j.1744-6570.1987.tb00609.x
18 DEROUS AND RYAN
Self, W. T., Mitchell, G., Mellers, B. A., Tetlock, P. E., &
Hildreth, J. A. D. (2015). Balancing fairness and efficiency:
The impact
of identity‐blind and identity‐conscious accountability on
applicant screening. PLOS One, 10(12), e0145208. https://doi.
org/10.1371/journal.pone.0145208
Shen, J., Chanda, A., D'Netto, B., & Monga, M. (2009).
Managing diversity through human resource management: An
inter-
national perspective and conceptual framework. International
Journal of Human Resource Management, 20(2), 235–251.
https://doi.org/10.1080/09585190802670516
SHRM. (2016). Using social media for talent acquisition:
Recruitment and screening. Alexandria, VA: Society for Human
Resource Management.

138. Sidanius, J., & Pratto, F. (1999). Social dominance: An
intergroup theory of social hierarchy and oppression. New
York:
Cambridge University Press.
Spence, A. M. (1974). Market signaling: Informational transfer
in hiring and related screening processes. Cambridge, MA:
Harvard
University Press.
Sweeney, L. (2013). Discrimination in online ad delivery.
Queue, 11(3), 10. https://doi.org/10.1145/2460276.2460278
Tajfel, H., & Turner, J. (1979). An integrative theory of
intergroup conflict. In W. Austin, & S. Worchel (Eds.), The
social psy-
chology of intergroup relations (pp. 33–47). Monterey, CA:
Brooks‐Cole.
Thorndike, R. L. (1949). Personnel selection: Test and
measurement techniques. New York: John Wiley & Sons.
Trope, Y., & Liberman, A. (1993). The use of trait conceptions
to identify other people's behavior and to draw inferences
about their personalities. Personality & Social Psychology
Bulletin, 19(5), 553–562. https://doi.org/10.1177/
0146167293195007
Trullen, J., Stirpe, L., Bonache, J., & Valverde, M. (2016). The
HR department's contribution to line managers' effective imple-
mentation of HR practices. Human Resource Management
Journal, 26(4), 449–470. https://doi.org/10.1111/1748‐
8583.12116
Umphress, E. E., Simmons, A. L., Boswell, W. R., & Triana, M.

139. (2008). Managing discrimination in selection: The influence of
directives from an authority and social dominance orientation.
Journal of Applied Psychology, 93(5), 982–993. https://doi.
org/10.1037/0021‐9010.93.5.982
van de Ven, N., Bogaert, A., Serlie, A., Brandt, M. J., &
Denissen, J. J. A. (2017). Personality perception based on
LinkedIn
profiles. Journal of Managerial Psychology, 32(6), 418–429.
https://doi.org/10.1108/JMP‐07‐2016‐0220
Van Iddekinge, C. H., Lanivich, S. E., Roth, P. L., & Junco, E.
(2016). Social media for selection? Validity and adverse impact
potential of a Facebook‐based assessment. Journal of
Management, 42(7), 1811–1835. https://doi.org/10.1177/
0149206313515524
Vorauer, J. D., Main, K. J., & O'Connell, G. B. (1998). How do
individuals expect to be viewed by members of lower status
groups? Content and implications of meta‐stereotypes. Journal
of Personality and Social Psychology, 75, 917–937.
Waldinger, R., & Lichter, M. I. (2003). How the other half
works: Immigrants and the social organization of labor.
Berkeley, CA:
University of California Press.
Waung, M., McAuslan, P., DiMambro, J. M., & Mięgoć, N.
(2017). Impression management use in resumes and cover
letters.
Journal of Business and Psychology, 32(6), 727–746.
https://doi.org/10.1007/s10869‐016‐9470‐9
Weichselbaumer, D. (2017). Discrimination against migrant job
applicants in Austria: An experimental study. German
Economic Review, 18(2), 237–265.

140. https://doi.org/10.1111/geer.12104
Wille, L., & Derous, E. (2017). Getting the words right: When
wording of job ads affects ethnic minorities' application deci-
sions. Management Communication Quarterly, 31(4), 533–558.
https://doi.org/10.1177/0893318917699885
Woodrow, C., & Guest, D. E. (2014). When good HR gets bad
results: Exploring the challenge of HR implementation in the
case of workplace bullying. Human Resource Management
Journal, 24(1), 38–56. https://doi.org/10.1111/1748‐
8583.12021
Youyou, W., Kosinski, M., & Stillwell, D. (2015).
Computer‐based personality judgments are more accurate than
those made
by humans. Proceedings of the National Academy of Sciences,
112(4), 1036–1040. https://doi.org/10.1073/
pnas.1418680112
Zschirnt, E., & Ruedin, D. (2016). Ethnic discrimination in
hiring decisions: A meta‐analysis of correspondence tests 1990–
2015. Journal of Ethnic and Migration Studies, 42(7), 1115–
1134. https://doi.org/10.10800/
133693183X.20105.11303279
How to cite this article: Derous E, Ryan AM. When your resume
is (not) turning you down: Modelling ethnic
bias in resume screening. Hum Resour Manag J. 2018;1–18.
https://doi.org/10.1111/1748‐8583.12217
https://doi.org/10.1371/journal.pone.0145208
https://doi.org/10.1371/journal.pone.0145208
https://doi.org/10.1080/09585190802670516
https://doi.org/10.1145/2460276.2460278

141. https://doi.org/10.1177/0146167293195007
https://doi.org/10.1177/0146167293195007
https://doi.org/10.1111/1748-8583.12116
https://doi.org/10.1111/1748-8583.12116
https://doi.org/10.1037/0021-9010.93.5.982
https://doi.org/10.1037/0021-9010.93.5.982
https://doi.org/10.1108/JMP-07-2016-0220
https://doi.org/10.1177/0149206313515524
https://doi.org/10.1177/0149206313515524
https://doi.org/10.1007/s10869-016-9470-9
https://doi.org/10.1111/geer.12104
https://doi.org/10.1177/0893318917699885
https://doi.org/10.1111/1748-8583.12021
https://doi.org/10.1111/1748-8583.12021
https://doi.org/10.1073/pnas.1418680112
https://doi.org/10.1073/pnas.1418680112
https://doi.org/10.10800/133693183X.20105.11303279
https://doi.org/10.10800/133693183X.20105.11303279
https://doi.org/10.1111/1748-8583.12217
October 2014, ScientificAmerican.com 33
The firsT Thing to acknowledge about
diversity is that it can be difficult. In the U.S.,
where the dialogue of inclusion is relatively
advanced, even the mention of the word “diver­
sity” can lead to anxiety and conflict. Supreme
Court justices disagree on the virtues of diver-
sity and the means for achieving it. Corpora-
tions spend billions of dollars to attract and
manage diversity both internally and external-
ly, yet they still face discrimination lawsuits,
and the leadership ranks of the business world remain
predominantly white and male.

142. It is reasonable to ask what good diversity does us. Diversity of
expertise confers bene-
fits that are obvious—you would not think of building a new car
without engineers, de -
signers and quality-control experts—but what about social
diversity? What good comes
from diversity of race, ethnicity, gender and sexual orientation?
Research has shown that
social diversity in a group can cause discomfort, rougher
interactions, a lack of trust, great-
er perceived interpersonal conflict, lower communication, less
cohesion, more concern
i n b r i e f
Decades of research by organizational scientists, psy-
chologists, sociologists, economists and demographers
show that socially diverse groups (that is, those with a
diversity of race, ethnicity, gender and sexual orienta-
tion) are more innovative than homogeneous groups.
It seems obvious that a group of people with diverse
individual expertise would be better than a homoge-
neous group at solving complex, nonroutine problems.
It is less obvious that social diversity should work in the
same way—yet the science shows that it does.
This is not only because people with different back-
grounds bring new information. Simply interacting with
individuals who are different forces group members to
prepare better, to anticipate alternative viewpoints and
to expect that reaching consensus will take effort.
Katherine W. Phillips is Paul
Calello Professor of Leadership

143. and Ethics and Senior Vice Dean
at Columbia Business School.
Being around people
who are different from
us makes us more
creative, more diligent
and harder-working
Katherine W. Phillips
How
Diversity
works
the state of the world’s science 2014
34 Scientific American, October 2014
about disrespect, and other problems. So what is the upside?
The fact is that if you want to build teams or organizations
capable of innovating, you need diversity. Diversity enhances
creativity. It encourages the search for novel information and
perspectives, leading to better decision making and problem
solving. Diversity can im prove the bottom line of companies
and
lead to unfettered discoveries and breakthrough innovations.
Even simply being ex posed to diversity can change the way
you
think. This is not just wishful thinking: it is the conclusion I

144. draw from decades of research from organizational scientists,
psychologists, sociologists, economists and demographers.
InformatIon and InnovatIon
The key To undersTanding the positive influence of diversity is
the
concept of informational diversity. When people are brought
together to solve problems in groups, they bring different infor-
mation, opinions and perspectives. This makes obvious sense
when we talk about diversity of disciplinary backgrounds—
think
again of the interdisciplinary team building a car. The same
logic
applies to social diversity. People who are different from one
another in race, gender and other dimensions bring unique
infor-
mation and experiences to bear on the task at hand. A male and
a
female en gineer might have perspectives as different from one
another as an engineer and a physicist—and that is a good thing.
Research on large, innovative organizations has shown re -
peatedly that this is the case. For example, business professors
Cristian Deszö of the University of Maryland and David Ross of
Columbia University studied the effect of gender diversity on
the top firms in Standard & Poor’s Composite 1500 list, a group
designed to reflect the overall U.S. equity market. First, they
examined the size and gender composition of firms’ top
manage-
ment teams from 1992 through 2006. Then they looked at the
financial performance of the firms. In their words, they found
that, on average, “female representation in top management
leads to an increase of $42 million in firm value.” They also
measured the firms’ “innovation intensity” through the ratio of
research and development expenses to assets. They found that

145. companies that prioritized innovation saw greater financial
gains when women were part of the top leadership ranks.
Racial diversity can deliver the same kinds of benefits. In a
study conducted in 2003, Orlando Richard, a professor of man-
agement at the University of Texas at Dallas, and his colleagues
surveyed executives at 177 national banks in the U.S., then put
together a database comparing financial performance, racial
diversity and the emphasis the bank presidents put on innova-
tion. For innovation-focused banks, increases in racial diversity
were clearly related to enhanced financial performance.
Evidence for the benefits of diversity can be found well be -
yond the U.S. In August 2012 a team of researchers at the
Credit
Suisse Research Institute issued a report in which they exam-
ined 2,360 companies globally from 2005 to 2011, looking for a
relation between gender diversity on corporate management
boards and financial performance. Sure enough, the re
searchers
found that companies with one or more women on the board
delivered higher average returns on equity, lower gearing (that
is, net debt to equity) and better average growth.
Productivity and equity are probably the most
often cited reasons to at tend to diversity in sci-
ence. Gender and culture also affect the sci-
ence itself, however. They influence what we
choose to study, our perspectives when we
approach scientific phenomena and our strate-
gies for studying them. when we enter the
world of science, we do not shed our cultural
practices at the door.
evolutionary biology is one example.
Despite popular images of Jane Goodall

146. observing chimpanzees, almost all early stud-
ies of primate behavior were conducted by
men. Male primatologists generally adopted
Charles Darwin’s view of evolutionary biolo­
gy and focused on competition among males
for access to females. in this view, female pri-
mates are passive, and either the winning
male has access to all the females or females
simply choose the most powerful male.
the idea that females may play a more
active role and might even have sex with many
males did not receive attention until female
biologists began to do field observations. Why
did they see what men missed? “when, say, a
female lemur or bonobo dominated a male, or
a female langur left her group to solicit strange
males, a woman fieldworker might be more
likely to follow, watch, and wonder than to dis-
miss such behavior as a fluke,” wrote anthro­
pologist sarah Hrdy. Her interest in maternal
reproductive strategies grew from her empa-
thy with her study subjects.
Culture also made a difference in ap -
proach. in the 1930s and 1940s U.s. primatol-
ogists, adopting the stance of being “mini­
mally intrusive,” tended to focus on male
dominance and the associated mating ac -
cess and paid little attention to individuals
except to trace dominance hierarchies; rarely
were individuals or groups tracked for many
years. Japanese researchers, in contrast,
gave much more attention to status and
social relationships, values that hold a higher

147. relative importance in Japanese society.
This difference in orientation led to striking
differences in insight. Japanese primatologists
discovered that male rank was only one factor
determining social relationships and group
composition. they found that females had a
rank order, too, and that the stable core of the
group was made up of lineages of related
females, not males. the longer-term studies of
Japanese researchers also allowed them to
notice that maintaining one’s rank as the alpha
male was not solely dependent on strength.
Diversity has had an effect on studies of
education and social science. Lawrence
Kohlberg’s highly influential work on stages
of moral development in children in the early
1970s was later called into question by psy-
chologist Carol Gilligan on the grounds that it
ignored the perspective of women, who tend-
ed to emphasize the ethic of caring. Nor did
kohlberg’s model account for moral principles
associated with eastern religious traditions, in
part because his scheme did not include prin-
ciples of cooperation and nonviolence.
validity in the sciences involves much more
than attending to canons about the need for
proper controls, replicability, and the like. It
involves choices about what problems and
populations to study and what procedures and
measures to use. Diverse perspectives and val-
Particular Points of view

148. By Doug Medin, Carol D. Lee and Megan Bang
October 2014, ScientificAmerican.com 35
How dIversIty
Provokes tHougHt
Large data-set studies have an obvious
limitation: they only show that diver-
sity is correlated with better perfor-
mance, not that it causes better per-
formance. Research on racial di versity
in small groups, however, makes it
possible to draw some causal conclu-
sions. Again, the findings are clear: for
groups that value innovation and new
ideas, diversity helps.
In 2006 Margaret Neale of Stanford
University, Gregory Northcraft of the
University of Illinois at Urbana-Cham-
paign and I set out to examine the
impact of racial diversity on small deci-
sion-making groups in an experiment
where sharing information was a re -
quirement for success. Our subjects
were undergraduate students taking business courses at the Uni-
versity of Illinois. We put together three-person groups—some
consisting of all white members, others with two whites and one
nonwhite member—and had them perform a murder mystery
exercise. We made sure that all group members shared a
common
set of information, but we also gave each member important
clues

149. that only he or she knew. To find out who committed the
murder,
the group members would have to share all the information they
collectively possessed during dis-
cussion. The groups with racial diver-
sity significantly outperformed the
groups with no racial diversity. Being
with similar others leads us to think
we all hold the same information and
share the same perspective. This per-
spective, which is what stopped the
all-white groups from effectively pro-
cessing the in form ation, is what hin-
ders creativity and innovation. Other
re searchers have found similar re -
sults. In 2004 Anthony Lising Anto-
nio, a professor at the Stanford Grad-
uate School of Education, collaborat-
ed with five colleagues from Stanford
and other institutions to examine the
influence of racial and opinion com-
position in small group discussions.
More than 350 students from three
universities participated in the study. Group members were
asked
to discuss a prevailing social issue (either child labor practices
or
the death penalty) for 15 minutes. The researchers wrote
dissent-
ing opinions and had both black and white members deliver
them
to their groups. When a black person presented a dissenting per-
spective to a group of whites, the perspective was perceived as
more novel and led to broader thinking and consideration of