Why Measuring Matters

Catégories :

Draft – For discussion only

Monitoring Brief: The callousness of not asking how children are doing
By Dr. Fred Wulczyn

This analysis was produced at the request of IFSD to support ongoing research in First Nations child and family services. IFSD’s work is undertaken through a contract with the Assembly of First Nations (AFN). The views and analysis of the independent authors do not necessarily reflect the views of the AFN or IFSD.

In late May 2021, after hundreds of unmarked graves were uncovered at a former Indian residential school near Kamloops, British Columbia, there was a shock that resounded globally. First Nations children in Canada had suffered unspeakable outcomes. Callous disregard is a hard phrase to use when talking about someone else’s children but there are times when it is an inescapable conclusion. That was one of those times.

The repercussions of that dark chapter in Canadian history raise many, many questions. There is one that stands out at this moment of truth and reconciliation: what next?

The Canadian Human Rights Tribunal (CHRT) ruled that the First Nations child and family services system – linked to the intergenerational trauma and repercussions of residential schools – was discriminatory and underfunded. Subsequently, the CHRT defined two critical elements in its rulings: end discrimination and ensure it does not recur. There have been expenditures (approximately $20B) committed to invest in the past – a righting of wrongs – for children and families that suffered from the child protection system. There must be a connection to how those same wrongs might be avoided going forward. While an additional roughly $20B has been committed to investing in a better system for the future, there must be a commitment to knowing how young people are doing (by and for the First Nations that care for them). Is the government of Canada prepared to invest more wisely in the tools and technology needed to uphold what is a fundamental obligation to ensure discrimination does not recur? It is an uncomfortable and costly question to answer, but as it often does, history tells us that ignoring the question is even more costly than addressing it.

It is, of course, fair to ask about the way forward. History is again instructive. Speaking before the U.S. Congress in 1975, Orville Brim called on the United States to document the well-being of children in the same way we worked to document the health of the economy. His words ring true today. We are obliged to know, with a sense of purpose, how children are doing – the state of the child if you will. “Nothing esoteric is meant by using the term ‘indicators of the state of the child’ or ‘indicators of child development,’ or even ‘childhood indicators.” Children rely on the adults in their community to understand how well they are doing.

Brim also had advice for how we might come to know the state of the child. “To produce these facts clearly requires there be identical measures, repetitively applied over time, to comparable populations of children.” As a collaborator and confidant of Urie Bronfenbrenner, Brim understood the importance of rigor. One does not come to know the state of the child haphazardly. A sense of purpose has to be matched with investment in the collection of data that provides the evidence needed to judge the well- being of children and whether our efforts to improve the state of the child are paying off in the expected ways.

In its relationship with First Nations people, Canada finds itself at a critical moment in time. One path forward is built around a commitment to empowering communities so that they know the state of their child. What of the other path? It is the path with which we are already familiar.

Even if the choice is clear, the way forward requires due diligence. Again, history offers guidance. There are two parts to the answer. First, is what we might call the data capture phase. Brim said it best – to know the state of the child one needs measures repetitively applied over time. The second phase involves interpretation – using the data so that everyone with an interest in the state of the child knows the state of the child.

Best practices suggest the following steps:

  1. It starts with a conceptual model of child development and the factors the influence healthy development: the child, the child’s family, the child’s community, and the investments made to support healthy development.

    The Measuring to Thrive Framework provides the groundwork for the measurement model and the accountability framework.

  2. An investment in data capture systems. Data capture refers to the recording of the relevant data, laid out in the Measuring to Thrive Framework. Data capture requires investment in data management technology (this is how the data is recorded) and the human resources needed to enter data if data entry is not accomplished in some other way. For relatively simple tasks, such as recording the dates of placement or the reason for leaving placement, the types of data have to be defined but no specialized data collection instruments are required. In the case of a child’s well-being, assessment instruments have to be selected along with an operational plan for measurements.

Data capture systems require investment. Given the lack of attention to this form of investment in the past, investment in the context of the overall settlement agreement by the central government is the most sensible. First Nations agencies must have the resources needed to know the state of their children. We see in the past how the First Nations are profoundly disadvantaged when they are not able to carry out this most basic function.

It is important to remember that the form of data capture significantly affects the monitoring function. Child development is a life course phenomenon. Data capture has to support the life course perspective: the same measures are recorded repetitively over time for cohorts of children from birth going forward.

For purposes of data capture, the data capture may involve already existing data from administrative data systems or other sources (e.g., birth records, school records, health records) as available. These other data sources have to be integrated into the conceptual framework.

If needed there are abundant examples of how this is accomplished.

Last, data capture, with support from the government, is a local or community responsibility. The data captured must serve the community first and foremost.

3. To support regular reporting, the data captured have to be used in ways that help communities manage their investments in children and families. This is the evidence-building phase of the monitoring function.

For this purpose, the conceptual model has to be transformed into a data model that renders the data suitable for analysis.

A basic data model has been proposed. In the example below, the model is focused on child placement. The data model provides answers to two broad, mission-critical categories of questions:

  • At the community level, how often does placement away from the family take place, and for what reasons?
  • Given an initial contact with a formal agency (i.e., a child protection agency), how likely is child placement?

The first question asks and answers this question: does placement happen frequently despite investments intended to reduce the need to place children away from home? The second question is concerned with the outcome of contact with the organizations in the community working to reduce the risk of placement. These questions are closely related but distinct.The data model has to support both. The proposed data model accomplishes this.

The data model requires a small number of recorded data elements:

CIN – unique child identifier CID – family or case identifier

Geographic information – this refers to a neighborhood, community, First Nation’s community, and/or Province. Granular data (neighborhood) can be aggregated to larger spatially relevant areas for comparative purposes.

Gender

DOB

Race/ethnicity

Event sequence – a counter that keeps track of the event number so that the temporal order of events can be reconstructed.

Event date – the date of the event (best if this is in Y/M/D format).

Event type – the types of events include: an investigation for maltreatment, placement in out-of-home care, the departure from home event, the onset of services, an assessment, or a change in caseworker.

This data model has a long track record of being used in reporting and monitoring contexts around the world.

Wulczyn, F., Harden, A., & Goerge, R. (1997). Foster Care Dynamics 1983-1994: An Update from the Multistate Foster Care Data Archive. (pp. 1–73).

This report illustrates how a simple data model is used to monitor the placement experiences of children placed away from home.

Wulczyn, F., & Huhr, S. (2019). Human Capital Formation During Childhood: Foundations of the Pathways of Care Longitudinal Study. Pathways of Care Longitudinal Study: Outcomes of Children and Young People in Out-of-Home Care (No. 13; pp. 1–31). https://www.facs.nsw.gov.au/download?file=656572

This paper shows how the simple data model is expanded to track the well-being of children.

Wulczyn, F. (2020). Foster Care in a Life Course Perspective. The ANNALS of the American Academy of Political and Social Science, 692(1), 227–252. https://doi.org/10.1177/0002716220976535

This paper provides another example of the data being used to understand the experiences of children within a life course perspective – i.e., longitudinally over place and time.

Testa, M. F. , & Wulczyn, F. (1980). State of the Child: Illinois. Children’s Policy Research Project, University of Chicago.

This report serves as a template for building a state of the child report.

4.  The data captured are used by the local community to manage their investments. Data sharing enriches our understanding of how well our investments in children pay off by way of comparison. To support data comparison, there are several options.

The Data Hub: With this strategy, the data custodian at each First Nation agency ships in a secure manner the data file to a central entity. The data shipped will take one of two forms:

a) Raw data – Subject to privacy rules that prevent the disclosure of personal identifying information, the raw data will come from the back end of the data management system with as little alteration as possible. Before sending the raw data, a list of data elements (rows and columns) is identified for extraction. A data audit of the local information system is important for this purpose. The data request has to be informed by what is known about the data captured in the system and its structure. After the audit, a copy of those rows and columns is extracted in their original, raw form for transmission.

The data are then curated (transformed) in a manner consistent with the objectives of the monitoring system and the data model. Data curation takes place at the central hub.

Once curated, the data goes back to the agency that supplied the data for both verification and use as a decision support tool and for reporting to the local community.

The exchange takes place in both directions. Quality control is an advantage of this approach.

b) Curated data – Subject to how the curated data is assembled (i.e., as per the data model), the data is curated at the local level.

The curated data is then shared in its curated form with the data for reporting out the evidence from multiple communities.

The curated data can be in the form of a common reporting architecture or as a file of child-level records.

Although there are examples of both approaches, the data hub with centralized data curation is generally more economical. Locally curated data means that each agency is spending the resources to curate the data. There is no efficiency that comes with scale. In a central location, tools developed for one application can be shared across applications. Code to generate reports etc. is more easily and easily and economically generated. And, quality  assurance is easier to maintain and more fairly distributed.

Brim, O. G. (1975a). Childhood social indicators: Monitoring the ecology of development. Proceeding of the American Philosophical Society, 19(6), 413–418.

Brim, O. G. (1975b). Macro-Structural Influences on Child Development and the Need for Childhood Social Indicators. Congressional Record, Volume 121-Part 19, 24630–24632.

Crown-Indigenous Relations and Northern Affairs Canada. (2021, August 10). Government of Canada enhances support to Indigenous communities to respond to and heal from the ongoing impacts of residential schools. https://www.canada.ca/en/crown-indigenous-relations-northern-affairs/news/2021/08/government-of-canada-enhances-support-to-indigenous-communities-to-respond-to-and-heal-from-the-ongoing-impacts-of-residential-schools.html

Knitzer, J., & Allen, M. (1978). Children without homes: An examination of public responsibility to children in out-of-home care (pp. 1–292). Children’s Defense Fund.