Contents lists available at ScienceDirect Child Abuse & Neglect journal homepage: www.elsevier.com/locate/chiabuneg Research article Disadvantaged neighborhoods and the spatial overlap of substantiated and unsubstantiated child maltreatment referrals Miriam Marco a, *, Kathryn Maguire-Jack b , Enrique Gracia c , Antonio López-Quílez d a Department of Social Psychology and Methodology, Autonomous University of Madrid, C/Ivan Pavlov, 6, 28049, Madrid, Spain b School of Social Work, University of Michigan, 1080 S. University Avenue, Ann Arbor, MI 48109, USA c Department of Social Psychology, University of Valencia, Av. Blasco Ibáñez, 21, 46010, Valencia, Spain d Department of Statistics and Operations Research, University of Valencia, C/Doctor Moliner, 50, 46100, Burjassot, Valencia, Spain ARTICLEINFO Keywords: Child maltreatment referrals Neighborhood risk factors Victims of maltreatment ABSTRACT Background: Considerable debate exists on whether the substantiation decision is a reliable measure for rates of maltreatment. Studies have shown that risks among children victims of maltreatment versus children investigated but unsubstantiated are similar. Objective: This paper aims to respond to two research questions: (1) Do most child maltreatment referrals, substantiated and unsubstantiated, come from the same neighborhoods? (2) Do sub- stantiated and unsubstantiated referrals share the same neighborhood risk factors? Participants and settings: We used geocoded data from substantiated (n = 1799) and un- substantiated (n = 1638) child maltreatment referrals in Valencia, Spain (2004–2015). As the neighborhood proxy, we used 552 Census block groups. Neighborhood characteristics analyzed were: socioeconomic status, immigration concentration, residential instability, and public dis- order and crime. Methods: To study the geographical overlap of child maltreatment referrals, a Bayesian joint modeling approach was used. To analyze the influence of neighborhood-level characteristics on risk, we used a Bayesian random-effects modeling approach. Results: For substantiated child maltreatment referrals, 90 % of the total between-area variation in risk is captured by the shared component, while for unsubstantiated child maltreatment re- ferrals, the shared component was 88 %. The correlation between substantiated and un- substantiated risks of child maltreatment referrals was .80. These risks were higher in neigh- borhoods with low levels of socioeconomic status, higher immigrant concentration, public disorder and crime. Conclusions: Child maltreatment referrals, regardless of whether substantiated or un- substantiated, overlap in the same disadvantaged neighborhoods. This suggests that in these neighborhoods, families are at a higher risk of being investigated by child protective services suggesting a potential reporting bias. 1. Introduction Child maltreatment is a public health issue with detrimental consequences for victims. Children who are victims of maltreatment https://doi.org/10.1016/j.chiabu.2020.104477 Received 14 October 2019; Received in revised form 7 February 2020; Accepted 18 March 2020 ⁎ Corresponding author. E-mail addresses: miriam.marco@uam.es (M. Marco), kmjack@umich.edu (K. Maguire-Jack), Enrique.Gracia@uv.es (E. Gracia), Antonio.Lopez@uv.es (A. López-Quílez). Child Abuse & Neglect 104 (2020) 104477 0145-2134/ © 2020 Elsevier Ltd. All rights reserved. T