Journal of Child Sexual Abuse

ISSN: 1053-8712 (Print) 1547-0679 (Online) Journal homepage: https://www.tandfonline.com/loi/wcsa20

Child and Interviewer Race in Forensic

Interviewing

Amy K. Fisher, Tomiko D. Mackey, Carol Langendoen & Marie Barnard

To cite this article: Amy K. Fisher, Tomiko D. Mackey, Carol Langendoen & Marie Barnard
(2016) Child and Interviewer Race in Forensic Interviewing, Journal of Child Sexual Abuse, 25:7,
777-792, DOI: 10.1080/10538712.2016.1208705

To link to this article: https://doi.org/10.1080/10538712.2016.1208705

Published online: 12 Aug 2016.

Submit your article to this journal

Article views: 604

View related articles

View Crossmark data

Citing articles: 2 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=wcsa20
JOURNAL OF CHILD SEXUAL ABUSE
2016, VOL. 25, NO. 7, 777–792
http://dx.doi.org/10.1080/10538712.2016.1208705

Child and Interviewer Race in Forensic Interviewing

Amy K. Fishera, Tomiko D. Mackeyb, Carol Langendoenc, and Marie Barnardd
a
Department of Social Work, University of Mississippi, Oxford, Mississippi, USA; bDirector of Clinical
Services, Family Crisis Services of Northwest Mississippi Inc., Oxford, Mississippi, USA; cPrivate Practice,
Oxford, Mississippi, USA; dDepartment of Leadership & Counselor Education, University of Mississippi,
Oxford, Mississippi, USA

ABSTRACT ARTICLE HISTORY

The purpose of this study was to examine the potential effect of Received 16 June 2016
child race and interviewer race on forensic interviewing outcomes. Revised 19 June 2016
The results of the regression analysis indicated that child race and Accepted 19 June 2016
interviewer race had a significant effect on interview outcome KEYWORDS
category (no findings, inconclusive, or findings consistent with Child; child sexual abuse;
sexual abuse). Furthermore, the results indicate that the interaction disclosure; forensic
of child and interviewer race had predictive value for rates of interviewing; interviewer;
findings consistent with sexual abuse but not in the direction interview outcome; race
predicted. Cross-race dyads had significantly higher rates of inter-
view outcomes consistent with sexual abuse. These findings sug-
gest that more research into the effect of race on disclosure of child
sexual abuse is needed.

In 2011, child protective services (CPS) agencies nationwide reported

receiving a staggering 3.4 million allegations of child maltreatment. CPS
investigated and substantiated only a fraction of these allegations.
However, over 60,000 of the cases that were ultimately substantiated
contained allegations of sexual abuse (U.S. Department of Health and
Human Services, 2012).
As part of the investigative process in child sexual abuse cases, CPS workers
commonly use forensic interviewing services offered by child advocacy centers
(CACs). The National Children’s Alliance (NCA) estimates that in 2011, over
185,000 forensic interviews occurred in conjunction with CACs nationwide
(NCA, 2012). The results of these interviews are often used in legal proceedings,
not only to determine custody but also in criminal proceedings against the
accused. The sheer number of interviews, coupled with the enormity of the
stakes involved and the many factors involved in the forensic interview process
itself, have resulted in a field that has been extensively examined.
As part of this examination, researchers have scrutinized many factors
affecting the disclosure of child sexual abuse. However, one area that has
remained largely unexplored is the effect of child and interviewer race on

CONTACT Amy Fisher [email protected] Assistant Professor, University of Mississippi, Social Work, P.O.
Box 1848 University Mississippi 38677, United States.
© 2016 Taylor & Francis
778 A. K. FISHER ET AL.

disclosure (Springman, Wherry, & Notaro, 2006). The present study exam-
ines the effect of child and interviewer race on disclosure of child sexual
abuse within the context of forensic interviews conducted in a CAC setting.
American helping professionals live and practice in a country where race
and racial issues are very sensitive topics to broach (Sue, 2015). In spite of a
national movement to increase knowledge and respect for cultural diversity
and diversity-related issues in the field of child maltreatment (Behl, Crouch,
May, Valente, & Conyngham, 2001), the impact of the history of racism in
our nation is evident still today.
Life experiences largely determine the way in which people interact with those
who are racially and culturally similar or dissimilar to them (Staats, 2014). While
some adopt a color-blind approach to dealing with people of a different cultural
or racial background, the reality is that when a person walks into a room, race
and gender are immediately perceived, perhaps triggering bias. When a child is
interviewed in a child-friendly setting by a trained forensic interviewer, does the
child consider the race of the interviewer, and if so, what does that child do with
the information? Alternatively, does the interviewer consider the child’s race
when conducting a forensic interview, and is the interview conducted differently
for African-American and Caucasian children either consciously or
unconsciously?

Literature review
Despite the potential relevance or impact of race on an interview outcome, there
is very little direct guidance to be found in either forensic interviewing protocols
or empirical research investigating this issue (Springman et al., 2006). Decisions
pertaining to whether a child is interviewed by someone of a similar racial
background may be based on speculation, personal experiences, or agency
policy/protocols. In many cases the question may be a non-issue since agencies
may lack an ethnically or racially diverse forensic interviewing staff.

Forensic interviewing protocols

Practitioners might look to forensic interviewing protocols for guidance on the
issue of race within the interviewing context. There are over 700 CACs nation-
wide that adhere to national accreditation standards (National Children’s
Alliance, 2012). Services differ, but generally, the forensic interviewing services
offered in conjunction with CACs include investigative techniques supported by
the NCA and the American Professional Society on the Abuse of Children
(APSAC) (Hlavka, Olinger, & Lashley, 2010). Most CACs utilize one of several
nationally recognized interviewing protocols; the CornerHouse Forensic
Interview Protocol (formerly referred to as RATAC) and NICHD are two
examples (Anderson et al., 2010). Race and diversity are not specifically
 JOURNAL OF CHILD SEXUAL ABUSE 779

addressed in the rapport-building sections of these protocols, although the

intensive trainings such as the five-day ChildFirst (formerly known as Finding
Words) training, for example, includes discussion of and literature on cultural
competence issues. Nonetheless, there are no guidelines for child and inter-
viewer race matching or mixing in these protocols.

Related areas
For practitioners of forensic interviewing who attempt to find guidance on
the issues within the professional literature, there are currently three general
areas that are relevant to the establishment of best practices: the effect of race
in the context of professional counseling relationships, the effect of race in
the context of the field of child sexual abuse in general, or the body of
research that attempts to identify predictors of disclosure of child sexual
abuse. While none of these areas directly address the issue of race in the
forensic interviewing context, each lends some guidance that can be valuable.
Literature examining the effect of race in the context of professional counsel-
ing relationships has moved from attempts to quantify the effect of race to a
consensus that cultural competence is the best practice for addressing racial
differences between the counselor and client (Cabral & Smith, 2011; Sullivan &
Cottone, 2010; Swift, Callahan, Tompkins, Connor, & Dunn, 2015). Cultural
competence stresses the importance of understanding both the counselor and
the client’s cultural context, with attention to language differences and other
details that may contain hidden meaning (Chao, 2012). Above all, cultural
competence calls for awareness of cultural differences in attitudes and beliefs,
knowledge of the worldview of culturally diverse clients, and the use of appro-
priate skills (Chao, 2012). In a meta-analysis, matching the race of the counselor
and client, without more, was found to have no positive effect on the outcome of
the counseling experience (Cabral & Smith, 2011).
Within the field of child sexual abuse prevention and response, some
research has examined the issue of race and ethnicity. However, these studies
have focused on racial and ethnic differences in symptom presentation
(Clear, Vincent, & Harris, 2006), differences among ethnicities in the experi-
ence and meaning of child sexual abuse (Ullman & Filipas, 2005) and
characteristics of child sexual abuse within certain ethnicities (Futa, Hsu, &
Hanson, 2001). Although knowledge of this body of research could increase
the cultural competence of a forensic interviewer, it sheds little light on the
effect of the race of the child and interviewer on the forensic interview itself.
In attempting to determine what predicts the disclosure of child sexual
abuse, researchers have examined many factors. These factors include the
child’s developmental level, gender, relationship to the perpetrator, severity
of the abuse, age at the time of the forensic interview, consequences of
disclosure to the child and family, caregiver support, and the child’s
780 A. K. FISHER ET AL.

perception that he or she is responsible for the abuse (Crisma, Bascelli, Paci,
& Romito, 2004; Goodman-Brown, Edelstein, Goodman, Jones, & Gordon,
2003; Lippert, Cross, Jones, & Walsh, 2009; Paine & Hansen, 2002). There is
scant attention paid to the effect of race on the disclosure of child sexual
abuse.

Interviewer and interviewee race

The authors found only three studies specifically focusing on interviewer and
interviewee race and its effects on disclosure of negative events such as child
sexual abuse. First, Dunkerley and Dalenberg’s 1999 analog study involved
artificial nonabuse scenarios and evaluated disclosure rates related to positive
and negative secrets. The study revealed that only 20% of African-American
children disclosed a negative secret to a Caucasian interviewer but almost
67% disclosed to an African-American interviewer. The study further showed
that 60% of Caucasian children disclosed a negative secret to an African
American interviewer and almost 83% disclosed to a Caucasian interviewer
(Dunkerley & Dalenberg, 1999). In sum, although there were differences in
the rate at which African-American and Caucasian children disclosed nega-
tive secrets, the authors posited that race matching of the interviewer and
child strongly predicted disclosure of a negative secret.
In contrast, Dailey and Claus (2001) found that matching adult client and
interviewer race did not increase the disclosure of child sexual abuse. The
authors gathered retrospective data on 8,276 clients and 22 interviewers from
intake interviews conducted with adults seeking publicly funded substance
abuse treatment. The intake interview included use of the Addiction Severity
Index (ASI), a structured interview that includes questions on history of
abuse. Although Caucasian interviewers and female interviewers elicited
more reports of abuse and Caucasian males were more likely to report
abuse, the researchers found that matching interviewer and client race, age,
or gender did not increase nor decrease the likelihood of disclosure of child
sexual abuse (Dailey & Claus, 2001).
Although these articles shed some light on race and disclosure, neither
involved children being interviewed as part of an ongoing investigation of
child sexual abuse. However, in 2006, Springman and colleagues published
research retrospectively examining actual forensic interviews of 220 alleged
child sexual abuse victims using the interview protocol disseminated by the
National Children’s Advocacy Center (NCAC). They found that the child’s
and interviewer’s race “significantly predicted degree of disclosure” (p. 109),
though not in the way Dunkerley and Dalenberg’s research would suggest.
While Dunkerley and Dalenberg (1999) found that same-race dyads strongly
predicted disclosure of a negative secret, Springman and colleagues found
that cross-race dyads resulted in a greater degree of disclosure than same-race
 JOURNAL OF CHILD SEXUAL ABUSE 781

dyads. This finding was the opposite of that hypothesized by the study
authors prior to beginning their research (Springman et al., 2006).

The present study

The current lack of agreement in the limited research examining the role of
race as a factor that may influence disclosure of child sexual abuse indicates
the need for further study of this important issue. Dunkerley and Dalenberg
(1999), using artificial interview scenarios, concluded that race matching
significantly increased disclosure of negative secrets. However, two studies
specifically examining the disclosure of child sexual abuse utilizing actual
client data did not support that conclusion, though the results were subtly
different. Specifically, Dailey and Claus (2001) found that race matching did
not increase disclosure by adults while Springman and colleagues (2006)
found that race mixing increased the likelihood of disclosure by children.
Additional investigation is warranted to further examine the potential impact
of race on disclosure in order to give better guidance to practitioners.
The purpose of the present study is to examine the potential effects of the
race of the child and the interviewer on forensic interview outcomes, speci-
fically concerning child sexual abuse. It was expected that interviews con-
ducted by interviewers of the same race as the child would lead to more
interview findings consistent with sexual abuse than cross-racial dyads.
This study includes real-world data in a retrospective analysis of existing
files of forensic interview outcomes in cases of alleged child sexual abuse
involving African-American and Caucasian children at a CAC in a rural,
southern state. Interviews were conducted by six forensic interviewers, three
African American and three Caucasian, all trained in the CornerHouse
Forensic Interview Protocol: RATAC (Rapport, Anatomy Identification,
Touch Inquiry, Abuse Scenario, and Closure). Although addressing the
implications of racial issues as they relate to forensic interviews encompasses
far more than just African Americans and Caucasians, this study is limited to
these two groups as the population of the state in which the CAC is located is
37% African American and 60% Caucasian (U.S. Census Bureau, 2010).

Methods
Participants
This retrospective analysis included 522 cases of suspected child sexual abuse. The
study included all interviews conducted between 2000 and 2009 in which the
child’s race was identified as African American or Caucasian; there was an inter-
view outcome of findings consistent with child sexual abuse, inconclusive, or no
findings; and the interviewee was aged 2 to 17 years. Furthermore, only interviews
782 A. K. FISHER ET AL.

Table 1. Child and Interviewer Race.

Percent N
Child
Race
African American 44% 231
Caucasian 56% 291
Interviewer
Race
African American 50% 3
Caucasian 50% 3

supervised by the clinical director were selected, as a means of assuring consistent

analysis of the interview results.
As this is a retrospective study of deidentified existing charts, the relevant
institutional review board deemed it exempt from the need for IRB approval. All
interviews included were conducted for the purpose of investigation of the
allegations of sexual abuse and ensuring the safety of the child. Child and
interviewer race is summarized in Table 1. African-American children com-
posed 44% of the cases (n = 231) and Caucasian children composed 56% of the
cases (n = 291). There were 96 males (18.5%) and 426 females (81.5%) in the
sample. All children were aged 2 to 17 years with a mean age of 8.43 years. Boys
(M = 6.70 years, SD = 3.16) were significantly younger that girls (M = 8.84,
SD = 4.04, t(520) = −4.86, p < .0001). Caucasian children (M = 7.77, SD = 3.91)
were significantly younger than African-American children (M = 9.29,
SD = 3.92, t(520) = −4.39, p < .0001). Complete data were available on all
participants and thus all analyses utilized the full sample of 522 cases.
Six agency interviewers who worked between March 2000 and July 2009
conducted the interviews included in this study. Three African-American
interviewers conducted 43.7% of the interviews (n = 228), and three
Caucasian interviewers conducted 56.3% of the interviews (n = 294). The
interviewers were female with the exception of one African American male
interviewer, who conducted 4.8% of the interviews. All interviewers were
trained in the use of the CornerHouse RATAC model of forensic interview-
ing and maintained ongoing continuing education. There was no deliberate
racial or gender matching as available interviewers alternated interviews as
referrals were received.

Interview setting and procedures

The forensic interviews were conducted for investigative and safety evalua-
tion purposes. The CAC accepted referrals from law enforcement, the
Department of Human Services Family and Children’s Services, and the
District Attorney’s Office. The CAC follows a multidisciplinary team
approach to serving possible child victims; therefore, interview requests
 JOURNAL OF CHILD SEXUAL ABUSE 783

were screened by a trained case manager or forensic interviewer to avoid

multiple extensive interviews as well as to ensure there was reasonable
suspicion the abuse may have occurred. Reasonable suspicion constituted
either disclosure of abuse by the child, a witness to the abuse, medical
evidence of abuse, or other behaviors indicating abuse (i.e., sexually inap-
propriate behavior). Forensic interviews were conducted only if the request-
ing agency was present to observe the live interview. All interviews were
audio- or videotaped and were conducted in a child-friendly, neutral setting
at the CAC. The clinical director directly witnessed, supervised, and/or
evaluated all interviews included in this study.

Independent variables
The race of the child and the race of the interviewer served as dichotomous
predictor variables. Each of these independent variables had two levels:
African American (coded as 0) and Caucasian (coded as 1).

Outcome measure
There were three potential interview outcomes based on the report provided
by the child during the interview: no findings, inconclusive, or consistent with
sexual abuse. Interview outcomes had been determined following each inter-
view by the forensic interviewer and clinical director working in conjunction
to evaluate several factors provided by the child during the interview. The
factors included, but were not limited to, the content, consistency, and
feasibility of the report; the child’s ability to provide peripheral details and
sensory information; the affect and demeanor of the child during the inter-
view; and the ability of the child to explain the report.

Analysis
Observed frequency data was calculated using all 522 cases. An ordinal
logistic regression analysis was conducted utilizing SPSS 20.0 to assess the
influence of the race of the child and of the interviewer on interview
outcomes (no findings, inconclusive, and consistent with sexual abuse).
The race of the child and the race of the interviewer were both treated as
dichotomous variables (African American or Caucasian). Model fit and
assumptions were tested and odds ratios were calculated for each level of
interview outcome as a function of both predictor variables as well as for
their interaction.
784 A. K. FISHER ET AL.

Results
Interviewer and child race was mixed in 54.2% and matched in 45.8% of
the interviews (see Figure 1). Three interview outcomes were predicted
from the data: no findings, inconclusive, and findings consistent with
sexual abuse. The majority of the interview outcomes fell into the
consistent with findings of sexual abuse interview group (61.7%), fol-

160
140
120
100
80
60
40
20
0
African-American African-American Caucasian Caucasian
Interviewer and Interviewer and Interviewer and Interviewer and
African-American Caucasian African-American Caucasian
Child Child Child Child

Figure 1. Number of cases by interviewer race and child race.

Table 2. Interview Outcomes by Race of Child and Race of Interviewer.

% n
Findings Consistent with Sexual Abuse 61.7 322
Matched Interviewer/Child
African-American Child and Interviewer 10.3 54
Caucasian Child and Interviewer 14.6 76
Mixed Interviewer/Child
African-American Child and Caucasian Interviewer 19.2 100
Caucasian Child and African-American Interviewer 17.6 92
Inconclusive 24.5 128
Matched Interviewer/Child
African-American Child and Interviewer 3.1 16
Caucasian Child and Interviewer 8.6 45
Mixed Interviewer/Child
African-American Child and Caucasian Interviewer 6.3 33
Caucasian Child and African-American Interviewer 6.5 34
No Findings 13.8 72
Matched Interviewer/Child
African-American Child and Interviewer 3.5 18
Caucasian Child and Interviewer 5.7 30
Mixed Interviewer/Child
African-American Child and Caucasian Interviewer 1.9 10
Caucasian Child and African-American Interviewer 2.7 14
Total 100 522
 JOURNAL OF CHILD SEXUAL ABUSE 785

lowed by inconclusive interview findings (24.5%) and no findings

(13.8%) (see Table 2).
The full model, including child race, interviewer race, and an interaction term
of the two predictors, was tested for the 522 cases. Because the majority of the
cases fell in the highest category of the dependent variable, findings consistent
with sexual abuse, a complementary log-log link function was utilized. Goodness
of model fit was examined utilizing Pearson chi-square and deviance measures.
A good fit was found for the model (X2(3) = 7.37, p = .06; D(3) = 7.44, p = .06). The
difference between the two log-likelihoods was examined to determine if the null
hypothesis that the model without the predictors is as good as the model with the
predictors could be rejected. The difference (X2(3) = 14.6, p < .002) indicated that
this null hypothesis could be rejected and that the model including child and
interviewer race and an interaction term with these two predictors is a better
predictor of interview outcome.
A test of parallelism indicated that there is a difference in the regression
coefficients for the three categories of case outcome (X2(3) = 7.44, p = .06).
Although the proportion of variance accounted for according to Nagelkerke’s R2
is low (R2 = 033), this should be interpreted with caution as the variance in a 3
level ordinal variable, particularly when the majority of responses are in one of
the categories (findings consistent with sexual abuse in the current study), is
restrained. Overall, the model had statistically significant ability discriminate
between findings consistent with sexual abuse from inconclusive or no findings
results (c = .589, p = .001). The parameter estimates indicated that the race of the
child (β = .678, p < .001), the race of the interviewer (β = .511, p < .006), and the
interaction of their race (β = –.883, p < .003) were all significant predictors of the
interview outcome (see Table 3).
Analysis of interview outcome rates by the race of the child was conducted
to examine potential differences. While the results were not statistically
significant, African-American children were more likely to have an outcome
consistent with sexual abuse compared to Caucasian children and less likely
to have an inconclusive finding or no finding compared to Caucasian
children (Table 4). Table 2 indicates the percentage of children by race in
each interview outcome category.
It was expected that interviews conducted by interviewers of the same race as
the child would lead to more findings consistent with sexual abuse than cross-
racial dyads. Wald chi-square tests demonstrated support for a significant
interaction between interviewer race and child race (see Table 3); however, the

Table 3. Ordinal Regression Analysis Predicting Interview Outcome.

Variable Β SE Wald Statistic p value
Child’s Race .678 .193 12.37 .0001
Interviewer’s Race .511 .186 7.53 .006
Child*Interviewer Race −.883 .295 8.96 .003
786 A. K. FISHER ET AL.

Table 4. Odds for African-American Child’s Interview Outcome.

Outcome Odds Ratio Confidence Intervals
Disclosure 1.46 0.96–2.22
Inconsistent 0.72 0.48–1.09
No Disclosure 0.78 0.28–2.15
Note: Caucasian child’s interview outcome was utilized as the reference group.

Table 5. Odds of Disclosure Compared to Non-disclosure or Inconsistent Interview Outcome.

Race Dyad Odds Ratio Confidence Intervals
Caucasian Child and Caucasian Interviewer 1.0
African-American Child and Caucasian Interviewer 2.31 1.43–3.72
Caucasian Child and African-American Interviewer 1.91 1.18–3.05
African-American Child and African American Interviewer 1.57 0.92–2.68

45%

40% African-American
Child/African-
35%
American
30% Interviewer
Caucasian
25% Child/Caucasian
Interviewer
20%
African-American
15% Child/Caucasian
Interviewer
10%

5% Caucasian
Child/African-
0% American Interviewer
No Findings Inconsistent Consistent
Findings Findings

Figure 2. Interview outcome by race of child and interviewer in percentages for each interview
outcome category.

interaction was not in the direction expected. Examination of the odds ratios
indicated that findings consistent with sexual abuse were more likely in cross-
racial dyads than in same-racial dyads (see Tables 2 and 5). African-American
children interviewed by a Caucasian interviewer had 2.31 increased odds of
having a finding consistent with sexual abuse compared to a Caucasian child
interviewed by a Caucasian interviewer. Caucasian children had 1.90 increased
odds of findings consistent with sexual abuse when interviewed by an African-
American interviewer compared to a Caucasian interviewer. This interaction is
depicted in Figure 2.
 JOURNAL OF CHILD SEXUAL ABUSE 787

Discussion
The current investigation further explored some of the conflicting findings
from the previous literature by examining the effect of the race of the
interviewer and the race of the child on forensic interview outcomes in a
real-world setting in the rural South. The results of this study challenged the
main predicted outcome. As anticipated, the race of the child, the race of the
interviewer, and the interaction of these variables did significantly influence
interview outcome. However, these data indicated that findings consistent
with sexual abuse were more likely in cross-racial dyads compared to same-
racial dyads. It is important to note that while these results are statistically
significant, the effect size is not large.

Child race
The Dunkerley and Dalenberg (1999) analog study strongly suggested that
African-American children would be less likely to disclose sexual abuse than
Caucasian children. However, Springman and colleagues (2006) failed to find
evidence to support this prediction, finding instead that the African-American
children in their study were slightly more likely than Caucasian children to
provide a detailed account of sexual abuse. The results of the present study
continue in the direction of the Springman results, with African-American
children being significantly more likely to have the interview outcome of find-
ings consistent with sexual abuse than Caucasian children.
Dunkerley and Dalenberg cited studies finding that African Americans distrust
mental health services in general, and services provided by Caucasians in parti-
cular, as a possible explanation of their findings. However, the Dunkerley study
does not indicate any special efforts toward rapport building or cultural compe-
tence by the interviewers. Both of the interviewing protocols and training pro-
grams involved in Springman and colleagues (2006), and the current study give
attention to rapport building and cultural competence. It is possible that attention
to these factors ameliorated the effects of this distrust. This explanation is con-
sistent with findings from the counseling literature indicating that cultural com-
petence (which includes attention to rapport building) is the best practice for
addressing racial differences (Cabral & Smith, 2011; Sullivan & Cottone, 2010;
Swift et al., 2015).

Interaction of child and interviewer race

Dunkerley and Dalenberg’s (1999) analog study also suggested that inter-
views conducted by interviewers of the same race as the child might predict
more findings consistent with sexual abuse than cross-racial dyads. However,
Dailey and Claus (2001) found that race matching did not increase
788 A. K. FISHER ET AL.

retrospective disclosures by adults. Next, Springman and colleagues’ 2006

study provided some evidence that race mixing increased disclosures by
children, although the effect size was “quite modest” (p. 109). The findings
of the present study are consistent with the Springman findings on the
interaction of child and interviewer race. In the present study, the probability
of having a finding consistent with sexual abuse is statistically significantly
greater for African-American children interviewed by a Caucasian inter-
viewer compared to Caucasian children interviewed by a Caucasian inter-
viewer. Likewise, Caucasian children were more likely to have a finding
consistent with sexual abuse when interviewed by an African American
interviewer compared to a Caucasian interviewer. The Springman study
and the present study have strikingly similar findings (2.76 and 2.31
increased odds respectively for African American children and 1.91 and
1.90 increased odds for Caucasian children with mixed interviewer race).
While the findings of the present study appear to support the findings of
the Springman and colleagues (2006) study, there are differences that warrant
consideration. The present study involved a much larger sample size (n = 522
versus n = 220), more interviewers (n = 6 versus n = 3), and slightly larger
effect size. In addition, the outcome categories in each study were different.
In the present study, the outcome categories were dictated by agency policy
(no findings, inconclusive, and findings consistent with child sexual abuse).
In the Springman and colleagues (2006) study, the outcome categories were
developed by the authors and included no disclosure, tentative disclosure,
and disclosure with detailed account of activities. What remains unknown is
the effective difference between the present study’s category of “inconclusive”
and the Springman study’s category of “tentative disclosure.”

Implications
The process of a child disclosing maltreatment can be burdened with diffi-
culties, such as being interviewed by several professionals from different
disciplines, not being believed, feeling ashamed, and not feeling supported.
The goal of a forensic interview is to provide a safe, supportive, and respect-
ful space that removes impediments to disclosure without crossing the line
into excessive suggestiveness (Gries, Goh, & Cavanaugh, 1996). To conclude
from this study that cross-racial dyads are the best way to promote disclosure
of child sexual abuse would be incorrect. At most, this study may be said to
show some support for the Springman findings that could lead to the
conclusion that race-mixing is not a barrier to disclosure. However, as
Springman and colleagues wrote, “These results may not demonstrate clinical
utility for drawing conclusions regarding the likelihood that children will
differentially disclose sexual abuse as a function of their own race and the
 JOURNAL OF CHILD SEXUAL ABUSE 789

race of the forensic interviewer” (2006, p. 109). Further research is needed to

draw firm conclusions about the effect of race on forensic interviewing.
The interviewers in both Springman and the present study used a semi-
structured forensic interviewing protocol for the interviews, allowing the
interviewer to emphasize (or deemphasize) portions of the protocol in
response to the specific circumstances of a particular child. However, both
protocols emphasize the importance of rapport building. As previously dis-
cussed, one potential implication of this study is that the culturally compe-
tent interviewers followed the protocol guidelines and paid particular
attention to building rapport.

Limitations and suggestions for further research

A limitation of this retrospective study is the lack of random assignment of
interviewees to selected interviewers. Although there is great value in the
examination of “real world” cases, there is a real need for a well-designed
randomized study on this issue in which the potential influence of the race of
the child and interview can be more carefully examined. This study did
include six interviewers, but two interviewers conducted approximately
60% of the interviews. In addition, there was only one male interviewer
involved. An important control was that one clinical director supervised all
cases and worked with the interviewers to determine how to categorize the
results. All interviewers received the same training in the RATAC protocol as
provided by CornerHouse.
Moreover, due to the demographics of the state in which the interviews
were conducted, the present study involved only African Americans and
Caucasians. The effects and implications of other races, such as Native
American or Hispanic, should also be studied. Furthermore, other factors
that influence culture, such as religion, region, sexual orientation, and age,
should be studied.
The implications of the use of the RATAC protocol and the NCAC
protocol need to be explored. While this study (using RATAC) replicated
the major findings of the Springman and colleagues (2006) study (using the
NCAC protocol), more information regarding the impact of using the two
protocols would be valuable. Further investigation could help in crafting
amendments to existing protocols that provide guidance to clinicians regard-
ing the impact of child and interviewer race. In addition, further investiga-
tion could result in helpful additions to agency policy.
The present study has potential interest for professionals beyond forensic
interviewers. Health care providers, victim advocates, police officers, prose-
cutors, child protection professionals, and others involved in child welfare
could benefit from these findings. Further research into the impact of race
for other professionals working in the field of child maltreatment is needed.
790 A. K. FISHER ET AL.

In addition, more is becoming known about the long-term health impact of

adverse childhood events (ACEs) such as child abuse, sexual trauma, and
witnessing domestic violence (Felitti & Anda, 2010). As health professionals
move toward screening patients for these events, a deeper understanding of
the effect of race on disclosure could prove valuable. Although race may be a
sensitive topic in this country, it is a topic that must be acknowledged and
explored to determine its effect on helping relationships.

Compliance with ethical standards

All procedures followed were in accordance with the ethical standards of the responsible
committee on human experimentation (institutional and national) and with the Helsinki
Declaration of 1975, as revised in 2000. Informed consent was obtained from all patients for
being included in the study.

Acknowledgments
The authors would like to acknowledge the help of Billy Barrios, Joanna Savarese, Sarah
Clavell Storer, and Olga Berkhout. Each made a meaningful contribution to this work.

Notes on contributors
Amy K. Fisher, JD, LCSW, Assistant Professor of Social Work, University of Mississippi,
University, Mississippi.
Tomiko D. Mackey, LCSW, Director of Clinical Services, Family Crisis Services of Northwest
Mississippi, Inc., Oxford, Mississippi.
Carol Langendoen, LCSW, private practice, Oxford, Mississippi.
Marie Barnard, PhD, Research Assistant Professor, Department of Leadership and
Counselor Education, University of Mississippi, University, Mississippi.

References
Anderson, J., Ellefson, J., Lashley, J., Lukas Miller, A., Olinger, S., Russell, A. . . . Weigman, J.
(2010). The cornerhouse forensic interview protocol: RATAC. Thomas M. Cooley Journal
of Practical and Clinical Law, 12, 193–331.
Behl, L. E., Crouch, J. L., May, P. F., Valente, A. L., & Conyngham, H. A. (2001). Ethnicity in
child maltreatment research: A content analysis. Child Maltreatment, 6(2), 143–147.
10.1177/1077559501006002006
Cabral, R. R., & Smith, T. B. (2011). Racial/Ethnic matching of clients and therapists in
mental health services: A meta-analytic review of preferences, perceptions, and outcomes.
Journal of Counseling Psychology, 58(4), 537–554. 10.1037/a0025266
Chao, R. C. (2012). Racial/Ethnic identity, gender-role attitudes, and multicultural counseling
competence: The role of multicultural counseling training. Journal of Counseling &
Development, 90, 35–44. 10.1111/j.1556-6676.2012.00006.x
 JOURNAL OF CHILD SEXUAL ABUSE 791

Clear, P. J., Vincent, J. P., & Harris, G. E., (2006). Ethnic differences in symptom presentation
of sexually abused girls. Journal of Child Sexual Abuse 15(3), 79–98. 10.1300/
J070v15n03_05
Crisma, M., Bascelli, E., Paci, D., & Romito, P. (2004). Adolescents who experienced sexual
abuse: Fears, needs and impediments to disclosure. Child Abuse & Neglect, 28, 1035–1048.
10.1016/j.chiabu.2004.03.015
Dailey, R. M., & Claus, R. E. (2001). The relationship between interviewer characteristics and
physical and sexual abuse disclosures among substance users: A multilevel analysis. Journal
of Drug Issues, 31(4), 867–888. 10.1177/002204260103100404
Dunkerley, G. K., & Dalenberg, C. J. (1999). Secret-keeping behaviors in black and white
children as a function of interviewer race, racial identity, and risk for abuse. Journal of
Aggression, Maltreatment & Trauma, 2(2), 13–35. 10.1300/J146v02n02_02
Felitti, V. J., & Anda, R. F. (2010). The relationship of adverse childhood experiences to adult
medical disease, psychiatric disorders and sexual behavior: Implications for healthcare. In
R. A. Lanius, E. Vermetten, & C. Pain (Eds.), The impact of early life trauma on health and
disease: The hidden epidemic. New York, NY: Cambridge University Press.
Futa, K. T., Hsu, E., & Hanson, D. J. (2001). Child sexual abuse in Asian American Families:
An examination of cultural factors that influence prevalence, identification, and treatment.
Clinical Psychology: Science and Practice, 8(2), 189–209.
Goodman-Brown, T. B., Edelstein, R. S., Goodman, G. S., Jones, D. P. H., & Gordon, D. S.
(2003). Why children tell: A model of children’s disclosure of sexual abuse. Child Abuse &
Neglect, 27, 525–540. 10.1016/S0145-2134(03)00037-1
Gries, L. T., Goh, D. S., & Cavanaugh, J. (1996). Factors associated with disclosure during
child sexual abuse assessment. Journal of Child Sexual Abuse, 5(3), 1–19. 10.1300/
J070v05n03_01
Hlavka, H. R., Olinger, S. D., & Lashley, J. L. (2010). Forensic evaluation techniques and
issues: The use of anatomical dolls as a demonstration aid in child sexual abuse interviews:
A study of forensic interviewers’ perceptions. Journal of Child Sexual Abuse, 19, 519–553.
doi:10.1080/10538712.2010.511988
Lippert, T., Cross, T. P., Jones, L., & Walsh, W. (2009). Telling interviewers about sexual
abuse: Predictors of child disclosure at forensic interviews. Child Maltreatment, 14(1), 100–
113. 10.1177/1077559508318398
National Children’s Alliance. (2012). Children’s Advocacy Centers. Retrieved from http://
www.nationalchildrensalliance.org/NIAAG
Paine, M. L., & Hansen, D. J., (2002). Factors influencing children to self-disclose sexual
abuse. Clinical Psychology Review, 22, 271–295. 10.1016/S0272-7358(01)00091-5
Springman, R. E., Wherry, J. N., & Notaro, P. C. (2006). The effects of interviewer race and
child race on sexual abuse disclosures in forensic interviews. Journal of Child Sexual Abuse,
15(3), 99–116. 10.1300/J070v15n03_06
Staats, C. (2014). State of the science: Implicit bias review 2014. Retrieved from he Kirwan
Institute for the Study of Race and Ethnicity, http://kirwaninstitute.osu.edu/research/under
standing-implicit-bias/
Sue, D. W. (2015). Race talk and the conspiracy of Silence. Hoboken, NJ: Wiley.
Sullivan, C., & Cottone, R. R. (2010). Emergent characteristics of effective cross-cultural
research: A review of the literature. Journal of Counseling & Development, Summer (3).
88357–362 10.1002/j.1556-6678.2010.tb00033.x
Swift, J. K., Callahan, J. L., Tompkins, K. A., Connor, D. R., & Dunn, R. (2015). A delay-
discounting measure of preference for racial/ethnic matching in psychotherapy.
Psychotherapy, 52(3), 315–320. 10.1037/pst0000019
792 A. K. FISHER ET AL.

U.S. Census Bureau. (2010). Estimates of the resident population by race and Hispanic origin
for the United States and states: July 1, 2009 (SC-EST2009-04). Retrieved July 29, 2010,
from http://www.census.gov/popest/states/asrh/
U.S. Department of Health and Human Services. (2012). Child maltreatment 2011. Retrieved
from https://www.acf.hhs.gov/sites/default/files/cb/cm11.pdf#page=57
Ullman, S. E., & Filipas, H. H. (2005). Ethnicity and child sexual abuse experiences of female
college students. Journal of Child Sexual Abuse, 14(3), 67–89. 10.1300/J070v14n03_04