Structures of Oppression in the U.S. Child Welfare System: Reflections on Administrative Barriers to Equity

Round 1

Reviewer 1 Report

Thank you for the opportunity to review your paper. This is quite an important topic. I am recommending that the paper be revised to address a variety of concerns. 

The paper reads rather like a polemic which means it appears to be balanced more towards advocacy than towards an academic piece that serves the purpose, as stated in the beginning, to invite other academics into analyzing various ways in which bias exists in child protection. 

1. Lines 31-33 - Wexler's work is very much advocacy as opposed to research or academic framing. The way his work is framed might serve the paper better if it is to show how advocacy organizations are framing the argument.
2. line 60 - knock up against is a rather parochial term - it is important to recall the readership is from many countries and terms such as that may not be well understood in other cultural contexts.
3. The Garcia-Smith case - is this hypothetical? I am presuming so. If it is not, then there needs to be evidence of consent. The idea of using a case is excellent - the case proceeds as though certain decisions are made thus framing the pathways. It might serve the paper better to show what the decision points look like and what possible pathways could be explored - and why certain ones are more likely.
4. line 89 - although you do cite the original Kempe article a bit further into this paragraph, it should be cited as soon as mentioned rather than starting with other pieces that cite Kempe.
5. line 107 - It will help readers to know that the USA legal system allows individual states to lay out these requirements as opposed to national legal framing. Many readers will be in jurisdictions where such standards are set out nationally.
6. Line 120 - while legislation does indicate that a person can be prosecuted for not reporting, the reader will benefit from knowing how often that happens and what penalties occur. Some work has suggested that these prosecutions are quite rare.
7. line 126 - people of color - this term will mean different things in different countries - who is included here? Canada will use BIPOC or racialized populations with a specific intent of including Indiegnous peoples (which the USA refers to as Native American); the UK will use BAME (Black and Minority Ethnic)
8. Line 141 - It seems this argument is incomplete - for example, the impact of the social determinants of Health is missing. The linkage to intersectionality is also missing suggesting that the bias issues are unique to and held tightly within CPS as opposed to the linkages to multiple systems that compound upon each other. The view taken here is too narrow.
9. Lines 146-148 - this presumes that the mandatory reporter is in a position to assess and act which may not be true - many intersecting systems are limited by their own structural biases that act as barriers to intervention.
10. Line 162 - here again comes the question of social determinants of health - child protection is typically not in a position to correct lack of access or underfunding of the needed services. This represents a very difficult decision point for workers - please explore how such dilemmas might be addressed.
11. line 185/186 - first sentence of the paragraph - please expand - this is an interesting contradiction.
12. the paragraph starting at line 188 is an example of where quantitative data is needed. Such statements need this in order for the claim to hold value beyond argument. 
13. Line 224/225 - here the article might benefit from giving consideration to the place of inter generational patterns and traumas that can arise from systems of racism and bias. An example can be the Indian Boarding Schools and the impact they still have.
14. The material around line 269 is another place where quantitative data is needed
15. line 311 - there is good quantitive data to be had - please offer the reader some of that to strengthen the points being made here.
16. Line 334/335 - same point as 15.
17. Paragraph ending at line 356 - there is very good research showing how various assessment tools and approaches being used with child protection populations have not been validated and/or normed on these populations.  This work should be cited here. An example is  Choate, P. W., & McKenzie, A. (2015). Psychometrics in parenting capacity assessments–A problem for First Nations parents. First Peoples Child & Family Review, 10(2), 31-43.  and  

Mann-Johnson, J. (2016). Decolonizing Home Assessment Practice at the Kitchen Table: A Thematic Analysis Identifying the Crucial Elements in the Assessment of Kinship Caregivers (Master's thesis, Graduate Studies).    and   Whitcombe-Dobbs, S. (2020). Through a glass darkly: the assessment of parenting capacity in the context of child protection. 19. Line 367 -  Here you might think of the WEIRD work - Henrich, J., Heine, S. J., & Norenzayan, A. (2010). Most people are not WEIRD. Nature, 466(7302), 29-29.   20. Line 376/377 - please show the data. 21. Line 383/385 - While this is an argument that appears often tells us why this is valid. As framed it is too much correlation v causation. 22. Line 481 - there is a need to bring oversight into the conversation. Various places use, for example, child and youth advocates to routinely review cases and publish reports (very common in Canada) and other independent review processes (New Zealand and Australia). There are also the large independent reviews (UK). Do these processes bring needed change as they have access to material that is otherwise confidential such as the soon to commence hearing into the death of Arthur Labinjo-Hughes in the UK (think also of Baby Peter/ Maria Colwell/ Victoria Climbie in the UK as other examples. What role does the media play in this discussion as it tends towards moral outrage which also pushes CPS workers to less risk tolerance. Recall that the public (and even politicians) have little interest in CPS unless tragedy is leading the public's awareness of CPS. 23. There are quite a number of problems with the reference list which needs a thorough review

• many doi numbers are missing
• several links I tried were broken
• primary sources should be used v secondary such as with American Academy of Paediatrics - the actual research article is available - Danielson, M. L., Bitsko, R. H., Ghandour, R. M., Holbrook, J. R., Kogan, M. D., & Blumberg, S. J. (2018). Prevalence of parent-reported ADHD diagnosis and associated treatment among US children and adolescents, 2016. Journal of Clinical Child & Adolescent Psychology, 47(2), 199-212.
• When materials are drawn from the internet please always offer the link - e.g.  https://www.apa.org/workforce/publications/16-demographics , 

I am hoping these observations are helpful. 

Author Response

thank you for your thorough review. Please see the attached document for our response.

Author Response File: Author Response.pdf

Reviewer 2 Report

The article presents a trenchant critique of the child protection system (CPS) in the USA, describes various processes which contstitute the oppressive nature of the system and calls for transformation of the CPS through transparent dialogue.  In doing so the article illustrates the oppressive nature of the system using a case vignette.  Overall the article is well written and obviously addresses an important topic which is the subject of intense academic scrutiny.  However, as the authors acknowledge, this type of criticism of the CPS has been made for over 3 decades now. Indeed this is now the mainstream view of the CPS in academic writing, and it is very rare to find an article which supports the CPS.  It is therefore not very clear what this article adds to the hundreds of other articles which point out the oppressive nature of child protection and the need for fundamental reform of the system.  Thus my first recommendation to the authors is to clearly state in the introduction what the specific contribution of this article will be, and to provide clear objectives for the article. With regards to the case vignette, this does add value to the article in personalising some of the system deficits, but the problem is that the vignette is rather arbitrary and this should be acknowledged.  Sure there are thousands of cases where intervention is inappropriate and families are damaged, but (to play devil's advocate) there are also cases where children who have been severely maltreated are removed from parents and placed in stable, loving foster care.  So the authors need to provide more justification for this particular vignette and how it represents the system.  Finally the conclusion is rather generic, calling for the system to be " revealed and transformed, oppressive structure by oppressive structure".  A bit more specificity about how this might be achieved would greatly strengthen the argument presented here.  Once these issues are addressed the article will be much stronger and will make a contribution to the literature in this area.

Author Response

Thank you for your review. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Thank you for the attention you have given to revisions. There are a few issues following your response to my review

1. The Mann-Johnson article is not in the reference list
2. Line 376/377 - there is a good body of research that shows the degree to which the CPS system is racialized. The child welfare system continues to disproportionately intervene in the lives of  children of color for reasons including poverty, racial bias and discrimination by profes- sionals, geographic location, and inherent structural racism (Boyd, 2015), with Black and  American Indian/Alaskan Native children being overrepresented in foster care --- Thus it is quite possible to show quantitative data to support these statements. In an article such as this, quantitative data bolsters the argument ensuring the reader that this is not just an opinion lacking backing. An example of an article using hard data to support the argument is Cénat, J. M., McIntee, S. E., Mukunzi, J. N., & Noorishad, P. G. (2021). Overrepresentation of Black children in the child welfare system: A systematic review to understand and better act. Children and Youth Services Review, 120, 105714.
3. Line 383/385 - the psychiatric data describes the situation for children in foster care but it does not prove that CPS intervention is causative. It only shows that there is a co-relation between two events - being involved with CPS and having a psychiatric diagnosis with medication. There may be other explanations thus the statement made in the paper is accurate but is written in a way that says to the reader A+B=C when it should be A and B co-exist but A does not necessarily cause B. Being very clear about this matters to the credibility of the article.
4. In terms of 224/225, I agree that addressing my suggestion may be problematic in that part of the argument.
5. In the paragraph starting at line 149 - some mention of the inherent biases imbedded in AI applications in CPS would show how these biases are getting even more imbedded - Lee, N. T. (2018). Detecting racial bias in algorithms and machine learning. Journal of Information, Communication and Ethics in Society.  Chouldechova, A., Benavides-Prado, D., Fialko, O., & Vaithianathan, R. (2018, January). A case study of algorithm-assisted decision making in child maltreatment hotline screening decisions. In Conference on Fairness, Accountability and Transparency (pp. 134-148). PMLR. Valentine, S. (2019). Impoverished algorithms: misguided governments, flawed technologies, and social control. Fordham Urb. LJ, 46, 364. Keddell, E. (2019). Algorithmic justice in child protection: Statistical fairness, social justice and the implications for practice. Social Sciences, 8(10), 281.

Author Response

Thank you again!

Author Response File: Author Response.pdf

Reviewer 2 Report

Thank you to the authors for addressing my comments on this interesting article.  The article is now much stronger and the arguments are well made.  My remaining comment relates to the conclusion.  In my first review I suggested a more comprehensive discussion about the implications of this article.  What would a child protection system look like that did not have mandatory reporting, substantiation, registries, therapeutic support for children and families etc?  As I indicated earlier, pointing out the problems of the CPS is the easy part and has been done extensively.  However replacing the current system with something better is the hard part.  I am not suggesting the authors invent a new system, but a more nuanced discussion on the implications of the reforms suggested, other than that they may reduce disproportionality, would greatly strengthen the article.

Author Response

Thank you; we made significant modifications.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Thank you for your work on the revisions