Martin Beaulieu, Chief - Quality and Data Ethics Secretariat , Stat ist ics Canada Workshop on Ethics in Modern Stat ist ical Organisations March 27, 2024

The Role of Data Ethics to Maintain and Improve Public Trust : The Stat ist ics Canada Experience

Delivering insight through data for a better Canada

Outline

• What is Data Ethics? • Why is it important for National Statistical Organizations?

• Public Trust • Statistics Canada’s Data Ethics Approach

oNecessity and Proportionality Framework oGuiding Principles oConcrete Implementation oChallenges and Lessons Learned

2

Data Ethics • Field of study that addresses questions about the appropriate use of data,

at various stages of the life cycle. • Examples of possible ethical issues related with the use of data: Asking questions that could potentially cause harm to respondents Using biased data to inform decision makers Using data that can reinforce existing inequalities Privacy intrusion that is not proportional to the benefit of a project Producing information that does not correspond to the needs of the

population

3

Importance of Data Ethics • Ever increasing demand for more timely and more granular statistics • May lead to surveys asking more sensitive questions • New data sources available and new tools to process these data combined

with a constant decline in surveys response rates • In this context, with we are innovating and therefore introducing practices that can raise

new ethical challenges. • Less transparent by default

• Impact on Public Trust • Data Users

• Quality • Data Providers (population, enterprises, other organizations)

• Confidentiality/ Security • Responsible acquisition and use throughout the lifecycle

• Gather – Guard – Grow – Give (Rancourt, 2019) 4

Necessity and Proport ionality Framework (N&P) • Objective to optimize privacy protection and the production of

information when designing a data-gathering approach. • Necessity: The description of the needs should include who requires

the information, the reasons why such information is needed and concrete expected benefits.

• Proport ionality: • The balance between privacy intrusion and the quality of the data is coherent

with the expected benefits of a project; • Ensuring we do not acquire or use more information than we need to produce

such benefits. • Concrete application of the N&P Framework:

• Creation of the Data Ethics Secretariat • Foundations of ethical considerations: Six Guiding Principles

Foundations of Ethical Considerations: Six Guiding Principles

Benefits for Canadians Privacy & Security

Transparency & Accountability Trust & Sustainability

Data Quality Fairness & Do no Harm

Guiding Principles Benefits for Canadians • Define and illustrate

the expected benefits with concrete examples

• If not going through with the project has negative side-effects, explaining them and listing the consequences will help justify the project.

Privacy & Security • Measures to avoid

a security breach/ Potential impacts of a breach

• Gathering only the necessary information

Transparency & Accountability • Proactive steps taken to

be transparent • Accountability when

using ML or AI “Human in the loop”

Guiding Principles Data Quality

• Known sources of error and/ or bias and ways to address them

• The expected benefits of the project should be proportional to the quality of the data.

Fairness & Do no Harm • Making sure harmful

profiles that could propagate stereotypes are not created

• Especially important with information on vulnerable or marginalized segments of the population

• Data to be used for the purpose of the project that justified its acquisition

• Plans to reuse data for different purposes

Trust & Sustainability • Relationships with

partners/ data providers • Ensure our own rules

are followed

Examples • Personal questions responded by proxy (Privacy and security) • Difficult survey questions (fairness and do not harm; data quality) • Using administrative data to compensate for refusals

(Transparency and Accountability; Trust and sustainability)

Concrete implementat ion • Reviews are done by Data Ethics Secretariat (DES) on new data

acquisitions, surveys and microdata linkages, working alongside other groups within the agency:

• Legal/ Privacy team • Subject-Matter area • Specialized teams depending on context

• Short questions based on the six guiding principles used by Data Ethics Secretariat to determine whether a request/ project is sensitive.

• Integrated existing processes

Governance Structure • Data Ethics Secretariat • Internal Data Ethics Committee • Principal Data Ethics and Scientific Integrity Officer • Advisory Council on Ethics and Modernization of Microdata

Access

Challenges and Lessons Learned

Challenges • DES role not always well known • Some definitions not well known

(e.g. personal data) • Perception of adding a roadblock • Time management for a small team

Lessons Learned • COMMUNICATION • Support from governance structure • Get involved early • Feasibility/ Exploratory studies • Don’t get rushed by tight timelines

Thank you/ Merci Pour de plus amples renseignements, veuillez contacter:

For more information, please contact:

[email protected]

Ethics of Technology

Workshop on Ethics in Modern Statistical Organisations March 2024

Content • What is Ethics of Technology • Why is it important for NSOs • Examples of ethical dilemmas for technology

Technology – Changing Paradigm… “ ….The merging of geopolitics, technology, and economics is reshaping the rules of the international order, and changing the existing balance of power and influence. Geostrategic power is increasingly determined by the ability to develop, control, and exploit frontier technologies. These technologies that are expected to reshape industries and societies, and potentially provide solutions to some of the major global challenges, range from advanced semiconductors to quantum, from biodigital to hypersonic technologies, and from 5G to artificial intelligence (AI) and Big Data….” “ … Technology is both a source of power and a tool to exert power. Technological capabilities not only determine economic and military success, but also play a decisive role in shaping future prosperity and the ability to tackle systemic challenges such as food security, pandemics, and climate change…The rapid pace of technological advancement in frontier technologies is increasing the urgency and opportunity cost of government inaction….” Geotechnomics, Policy Horizons Canada, May 16, 2023 Geotechnomics – Policy Horizons Canada

… Fundamental shift in the role of technology brings the need to reconsider how technology is embedded in all organizational processes –

including ethics, values and governance …

Why Ethics of Technology? “ … Technologies are often embedded with the biases or values of those who develop them.[38] This may be unintentional, or intentional with the explicit aim of exerting political or economic power and influence. For instance, AI is trained on data selected by the developer and may reflect the biases of the data source.[39] When countries import technologies, they may inadvertently import undesirable values too… “ “ … Policy makers, decision makers, and stakeholders may need to identify biases embedded in technology, and which of these might be most harmful to their values and interests. They may also have to explore the feasibility of developing minimum ethical standards… “ Geotechnomics, Policy Horizons Canada, May 16, 2023 Geotechnomics – Policy Horizons Canada

• Ethical considerations in technology date back to Ancient Greece, however the formal field of technoethics is fairly recent. The term “technoethics” was created by the philosopher Mario Bunge in 1977.

• Technoethics is about ethical aspects of technology systems and practices: – Ethical questions surrounding invention and implementation of new technologies – How standard ethical questions are changed by new technology

Why Is It Important for NSOs? • NSO mission is to provide trustworthy statistics

– Traditionally, ‘trustworthy’ has been defined around quality of statistics and ethical practices for treating data – data ethics

– Both have been cornerstones for maintaining public trust in NSOs • Societal expectations changed – more and more, expectations around environmental

and social responsibilities and governance matter in the public eyes – Impacting public trust and public perception of NSOs – Not only data, but all aspects of organizational performance are being viewed through this

lens

… With the rapid paradigm shift in the role of technology, ethical, environmentally and socially responsible IT could have a significant

impact on public trust in NSOs ...

Technology Ethics ≠ Data Ethics

Data Ethics

• Focus on the information that is created by technologies

• Everything related to data throughout the statistical process

• And beyond - data retention / archival / disposal

• Various practices in place such as • Confidentiality protection • IT security • Ethical gathering and acquisition of data • How data is used • How data is disseminated

Technology Ethics

• Intellectual Property and Copyrights – impacts of new technologies

• Social responsibility in use of technology (e.g., to counteract mis-information)

• Privacy vs security in workplace • Technoethics and cognition • Digital Divide – inequalities due to various

technologies • Ethical impacts of specific technologies –

AI, quantum computing, digital twin, etc. • Environmental footprint of technologies • Etc.

Dilemma: Quantum Computing (QC)

QC is reaching a tipping point to show real impact

Most applications are centered around optimization problems, quantum censing, quantum communication

Experts estimate that within the next 4-5 years, quantum computing will be applied to simulation of physical systems, search, and machine learning

Just like Gen-AI brought a new set of ethical considerations with its applications, QC technology will do the same

Familiar Problem of Data Privacy and Data Protection – Post-QC Cryptography Impacts

Experts estimate that within 5 years, QC will make conventional cryptography unsafe to use

The USA requested all departments and agencies to prepare an

inventory of impacted systems and start planning mitigation

implementation

Data that is protected now could be stolen and stored for a few years and then cracked using

QC? Will QC make it easier to compare different data to

triangulate confidential data?

Dilemma: Environmental Sustainability

IT, and data centers specifically, are amongst the most carbon-intensive aspects of business The International Energy Agency (IEA) predicts that the energy use of global data centers will jump from 1% of worldwide electricity demand to 6.4% in 2030 and to ›20% in a decade AI operations require heavy computing - › Microsoft is considering nuclear power to power AI operations

It is about technologies but also how technologies are used within organization: Desktop: 49% of GHG emissions (GCG estimated at 605 kg of CO2e over 4 year life-cycle) attributed to the use Servers, 88% of GHG emissions (GCG estimated at 8,104 kg of CO2e over 4 year life-cycle) attributed to the use

To advance NSO agenda and face new data product demands, though, programs ask for more computing power, more servers, scalable computing to handle bigger volumes of data and more sophisticated models

Environmental sustainability should be an important consideration for use of technology by NSOs, for example: • Sustainable infrastructure, data

centres and cloud: from always-on to always-available

• Sustainable digital workplace: educate employees; modular architecture

• Sustainable data: data storage • Sustainable code and software:

trade off between accuracy, speed, costs and energy consumption; business technologists and IT professionals

Dilemma: Social Responsibility

Negative side of a new technology – production of disinformation, misinformation, deepfakes, etc.

NSOs have significant expertise in ensuring that statistics and data produced by NSOs are trustworthy, using various ethical frameworks such as Responsible Machine Learning

Is it enough from a perspective of social responsibility?

• Do NSOs now have a social responsibility to share our expertise to help citizens and companies to learn how to use new info technology in a responsible and proper way?

• Will use of new technologies contribute to deepening and re- defining digital divide? For example, those who could afford using the latest version of ChatGPT vs. the rest?

• Do NSOs now have a social responsibility of working with big technology giants to build in ethical principles in how new info technology is defined and is evolving?

Questions and Discussion

Stay connected!

StatsCAN app Eh Sayers podcast StatsCAN Plus The Daily

Website Surveys and statistical programs Data service centres My StatCan

Questions? Contact us: [email protected]

Statistics Canada— Your National Statistical Agency

Delivering insight through data for a better Canada

{Classification}

Rates of out-of-home care among children in Canada:

An analysis of national administrative child welfare data

UNECE/UNICEF Expert Meeting on Statistics on Children

March 5, 2024

Nathaniel Pollock,1 Alexandra Ouédraogo,1 Nico Trocmé,2 Wendy Hovdestad,1 Amy Miskie,3

Lindsay Crompton,1 Aimée Campeau,1 Masako Tanaka,1 Cindy Zhang,1 Claudie Laprise,1 and

Lil Tonmyr1

1 Family Violence Epidemiology Section, Public Health Agency of Canada 2 School of Social Work, McGill University 3 Ministry of Children and Family Services, Government of Alberta

{Classification}

How many children are in

out-of-home care

in Canada?

2

{Classification}

Estimates of the number of children in out-of-home care in Canada

15,071 26,680

202220192019 2021

59,283 15,701 26,680 ?

Canadian Incidence

Study of Reported

Child Abuse &

Neglect2

Canadian

Child Welfare

Information System4

Prevalence Incidence Prevalence Prevalence

Analysis of

public data from

government

departments1

National

Population

Census3

(Children aged 0-15; Placement

following a new investigation)

(Age of protection in each

province/territory; Children in care on

March 31; some missing data on kinship)

(Children aged 0-14; foster

children; point-in time)

(Children aged 0-17+; all

placement types; March 31, 2022)

(New placements)(Existing placements) (Existing placements) (Existing placements)

{Classification}

Canadian Child Welfare Information System

4

Out-of-home care = Alternative care

Collaboration with child welfare sector partners

Aggregate and record-level data about children and

youth aged 0-17+ in out-of-home care in Canada

Point-in-time counts on March 31, 2022

{Classification}

Analysis

Percentage (% of all children in out-of-home care)

To describe the characteristics of children in out-of-home care.

Rate (per 1,000 population)

To estimate the prevalence of out-of-home care among children.

Rate Ratio (X times higher)

To compare rates of out-of-home care among different groups.

As available, analyze

data by:

• Sex/gender

• Age group

• Province/territory

• Year

• Placement type

Sensitivity

analysis with

public

ISC data

Notes: ISC, Indigenous Services Canada; ISC data was based on a March 31, 2020 count of children in out-of-home care whose parents or guardians were “ordinarily resident on reserve” and whose placement was under

the authority of a First Nations child and family services agency or a provincial/territorial department in a jurisdiction for which delegated First Nations agencies do not exist, such as Yukon and Newfoundland and Labrador.

+

{Classification}

Results

6

{Classification}

How many children are in out-of-home care in Canada?

61,104 Children in out-of-home care in Canada

on March 31, 2022

Note: Data from Prince Edward Island is based on a total count for the 2020/2021 fiscal year, not a March 31 point-in-time count..

{Classification}

61,104 Children in out-of-home care in Canada

on March 31, 2022

Sex/gender data from 6 provinces/territories*

11,489 Children in out-of-home care in Canada

on March 31, 2022

52.4% 47.6%

Male Female Notes: *Based on data from 6 provinces and territories (Newfoundland and Labrador, Nova Scotia, New Brunswick, Alberta, Northwest Territories and Nunavut),

covering 18.8% of children in out-of-home care on March 31, 2022.

Percentage of children in out-of-home care by sex/gender, 2021/2022

{Classification}

Rates of out-of-home care by age group, 2021/2022

Notes: Based on data from 6 provinces and territories (Newfoundland and Labrador, Nova Scotia, New Brunswick, Ontario, Alberta and Northwest Territories), covering 26.3% of children in out-of-home care

in 2021/2022. Age-stratified data from Ontario included children aged 0-17 years who were in care and in customary care. The rate for 18 to 21 years (25 years in Yukon) was 0.7 per 1,000. Data not shown

because three provinces and territories (Newfoundland and Labrador, Alberta, Ontario) were excluded from the analysis due to missing (i.e. age span did not include 18+ years) or suppressed counts).

Rate per 1,000 population

4.96

3.93

3.27

3.62

4.36

3.46

16 to 17 years

12 to 15 years

8 to 11 years

4 to 7 years

1 to 3 years

<1 year

A g e G

ro u p

{Classification}

0

5

10

15

20

25

30

35

NL PEI NS NB QC ON MB SK AB BC YK NT NU Canada

R a te

p e r

1 ,0

0 0 p

o p u la

ti o n

Province/Territory

Rates of out-of-home care in Canada by province/territory, 2021/2022

Notes: Data for the 2021/2022 fiscal year from Prince Edward Island were not available so count data from the most recent year (2020/2021) were used as proxy. Rates are based on data from 13

provinces/territories; data from Indigenous Services Canada were not included.

8.2

{Classification}

Percentage of children in out-of-home care by placement type in Canada

0%

10%

20%

30%

40%

50%

60%

P e rc

e n

ta g

e (

% )

Foster home (51.2%)

Kinship home (34.3%)

Other (2.4%)

Group care (12.1%)

Notes: Based on point-in-time count data from 9 provinces/territories with available data, stratified by placement type, over the 5-year period: Newfoundland and Labrador, Nova Scotia, New Brunswick, Quebec,

Manitoba, Alberta, Yukon, Northwest Territories and Nunavut. ISC data not included for any year or jurisdiction.

March 31,

2018

March 31,

2022

{Classification}

Rates of children aged 0-17 in alternative care by country

0 2 4 6 8 10 12

Finland

France

Germany

Denmark

Canada

Sweden

Canada

Australia

United Kingdom

United States

Ireland

Notes: Canadian rate estimate of 8.2 per 1,000 based on data for children aged 0-17 (up to age 20, 21, and 25 in some jurisdictions) from all 13 provinces and territories but excluding data from 50

Indigenous child welfare agencies in 4 provinces; Canadian rate estimate of 9.5 based on data for children aged 0-21 from all 13 provinces and territories and Indigenous Services Canada; other rate

estimates from Eurochild (2020/2021),5 U.S. Department of Health and Human Services (2018/2019),6 and the Australian Institute of Health and Welfare (2019/2020).7

8.2

9.5 0-17+ years

Data from 13 provinces/territories + federal government

0-17+ years

Data from 13 provinces/territories

Rate per 1,000 population

{Classification}

Strengths and challenges with national data in Canada

Strengths

• Working conceptual definition

• Near complete national coverage

• Stratified data are feasible

• Placement types are comparable

• March 31 point-in-time count is common

• Multiple types of data can be combined

• Aggregate data mitigates privacy risks

13

Challenges

• Unclear definitions in some regions

• Gaps in coverage by year and local areas

• Limited data on gender and age

• Varied approaches to kinship and youth data

• Underestimate of temporary placements

• Possible double-counting

• No data on Indigeneity, race or ethnicity

{Classification}

Future work

• Refine the definition of out-of-home care

• Increase use of record-level data

• Expand coverage:

+ Geography

+ Years

+ Sex/gender and age

+ Informal, youth, and adoption placements

+ Specific placement types

• Establish a data governance plan

• Formalize data sharing

• Add new indicators:

+ Reason for placement

+ Entry and exit

+ Placement duration

{Classification}

Summary

• An estimated 61,104 children were in out-of-home care on March 31st, 2022.

• The rate of out-of-home care in Canada was 8.2 per 1,000.

• Rates of out-of-home care varied by jurisdiction.

• Rates were highest among males and children aged 1-3 and 16-17 years.

• Most children in out-of-home care (>84%) were in family-based care.

{Classification}

Thank you. Questions?

Contact Us

Nathaniel Pollock

Public Health Agency of Canada

[email protected]

Read the report

Pollock, N. J., Ouédraogo, A. M., Trocmé, N., Hovdestad, W., Miskie, A., Crompton, L., Campeau, A.,

Tanaka, M., Zhang, C., Laprise, C., & Tonmyr, L. (2024). Rates of out-of-home care among children

in Canada: An analysis of national administrative child welfare data. Health Promotion and Chronic

Disease Prevention in Canada: Research, Policy and Practice, 44(4).

https://doi.org/10.24095/hpcdp.44.4.02

{Classification}

References

1. Saint-Girons M, Trocmé N, Esposito T, Fallon B. Children in out-of-home care in Canada in 2019 [Internet]. Montréal (QC): Canadian

Child Welfare Research Portal; 2020 [cited 2023 Jun 22]. Available from: https://cwrp.ca/publications/children-out-home-care-canada-

2019

2. Fallon B, Lefebvre R, Trocmé N, et al. Denouncing the continued overrepresentation of First Nations Children in Canadian Child

Welfare: Findings from the First Nations/Canadian Incidence Study of Reported Child Abuse and Neglect–2019. Montréal (QC):

Canadian Child Welfare Research Portal; 2021 [cited 2023 Jun 22]. Available from: https://cwrp.ca/publications/denouncing-continued-

overrepresentation-first-nations-children-canadian-child-welfare

3. Black T, Trocmé N, Fallon B, Houston E. Children in foster care in Canada in 2016 & 2021. Toronto (ON): Canadian Child Welfare

Research Portal; 2022 [cited 2023 Jun 23]. Available from: https://cwrp.ca/publications/children-foster-care-canada-2016-2021

4. Pollock N, Ouédraogo A, Trocmé N, Hovdestad W, Miskie A, Crompton L, Campeau A, Tanaka M, Zhang C, Laprise C, Tonmyr L.

(2024). Rates of out-of-home care among children in Canada: An analysis of national administrative child welfare data. Health

Promotion and Chronic Disease Prevention in Canada: Research, Policy and Practice, 44(4). https://doi.org/10.24095/hpcdp.44.4.02

5. Eurochild. Better data for better child protection systems in Europe: mapping how data on children in alternative care are collected,

analysed and published across 28 European countries [Internet]. Geneva (CH): UNICEF Europe and Central Asia Regional Office;

2021 [2023 Sep 24]. Available from: https://eurochild.org/uploads/2022/02/UNICEF-DataCare-Technical-Report-Final-1.pdf

6. Children’s Bureau. Child welfare outcomes report data [Internet]. Washington (DC): U.S. Department of Health and Human Services;

2023 [cited 2023 Jun 23]. Available from: https://cwoutcomes.acf.hhs.gov/cwodatasite/

7. Australian Institute of Health and Welfare. Child protection Australia 2019–20 [Internet]. Canberra (AU): Australian Institute of Health

and Welfare; 2021 [cited 2021 Sep 19]. Available from: https://www.aihw.gov.au/getmedia/c3b0e267-bd63-4b91-9ea6-

9fa4d14c688c/aihw-cws-78.pdf.aspx?inline=true 17

{Classification}

Additional

Slides

18

{Classification}

Jurisdiction or department Source Type Most recent

year used

# of

years Date of count

Age span of

coverage

Sex/Gender + Age

Group

Placement

Type

Newfoundland and Labrador Custom 2021/2022 9 Point in time (March 31) 0–21 Yes Yes

Prince Edward Island Public 2020/2021 4 Total fiscal year 0–17 No No

Nova Scotia Custom 2021/2022 9 Point in time (March 31) 0–20 Yes Yes

New Brunswick Custom 2021/2022 9 Point in time (March 31) 0–18 Yes Yes

Quebec Public 2021/2022 5 Point in time (March 31) 0–17 No Yes

Ontario Custom 2021/2022 3 Point in time (March 31) 0–17 Partial Yes

Manitoba Public 2021/2022 9 Point in time (March 31) 0–17 No Yes

Saskatchewan Public 2021/2022 9 Point in time (March 31) 0–21 No No

Alberta Custom 2021/2022 9 Monthly point in time average 0–17 Yes Yes

British Columbia Public and custom 2021/2022 9 Point in time (March 31) 0–18 No No

Yukon Public 2021/2022 5 Point in time (September 30) 0–25 No Yes

Northwest Territories Record-level 2021/2022 5 Point in time (March 31) 0–18 Yes Yes

Nunavut Public 2021/2022 9 Point in time (March 31) 0–18 Partial Yes

Indigenous Services Canada Public 2019/2020 1 Point in time (March 31) 0–17 No No

Data coverage for the Canadian Child Welfare Information System

Placement type Definition Examples of placement types included Special considerations

Kinship home A kinship home is a type of family-based care with a caregiver who has a family relationship

or other close tie or attachment to the child, their family, or the child’s cultural community.

Informal kinship placements variously include children whose legal status has not changed

(i.e. parents/guardians maintain legal custody), but the child is placed with an extended family

member or a trusted community member (as in customary care) on an emergency or

temporary basis under voluntary conditions or by court order. Formal kinship placements

typically involve extended family homes and caregivers who have gone through a formal

review, training and approval process that is similar to the process foster homes undergo.

Both informal and formal kinship and extended family placements are classified in CCWIS

data as kinship homes.

Person of sufficient interest

Kinship out-of-care by court order or agreement

Customary care

Extended family care

Kinship service or placement

Provisional home

Relative foster home

Place of safety

Some jurisdictions, e.g. Saskatchewan, do not use the term kinship in any form

(kinship care, kinship home, kinship service, etc.) when referring to any placement

type.

In some jurisdictions, kinship placements are formal placements that involve a

change in the child’s legal custody status, whereas placement with extended

family does not.

Customary care is a placement type that is specific to First Nations, Inuit and

Métis communities. It typically involves a voluntary placement in or close to a

home community with extended family or other community member. The purpose

of customary care is to support a child’s connection to their culture and language.

Foster home A foster home is another type of family-based care. This type of care typically involves one or

two primary caregivers who are not related to the child (i.e. non-family members). Except in

some specific arrangements with agency-based, contracted or treatment foster homes,

caregivers and children live in a private home. Foster homes are a formal placement and

prospective foster parents/caregivers undergo a screening, training and approval or licensing

process. Caregivers are not typically paid a salary but receive financial support to cover the

living costs for each child placed in their home.

Foster home

Treatment foster home

Parent-model, agency-based, or contracted

foster home

Specialized foster home

Some jurisdictions, e.g. Yukon, no longer use the terms foster home or foster

care, but are using community caregiver home instead.

Group care Group care comprises two main subtypes: group home and treatment facility. A group home

is often a large house with multiple children, where the caregivers are paid staff, e.g. child

and youth workers. Group homes may be operated by the child welfare authority; a

contracted resource such as a not-for-profit, charitable or religious organization; or by a for-

profit business.

A treatment facility refers to any placement in a specialized, often secure, institutional or

congregate setting, e.g. a campus-based treatment centre or hospital, that provides access to

therapeutic supports and interventions for behavioural, social, developmental, mental health,

substance use or physical health conditions or issues.

Group home

Residential care

Treatment centre

Secure treatment

Hospital

n/a

Other Children may be placed in other settings, usually on a temporary or transitional basis. This

small subset of placements are most often used to address extenuating circumstances such

as limited local access to specialized services or limitations in the availability of approved out-

of-home care settings.

Out-of-province/territory

Semi-independent living

Hotel/motel

Shelter

Adoption probation

In some jurisdictions, data on placement types such as independent and semi-

independent living, out-of-province/territory placements and adoption probation

were not included or were not disaggregated.

In the Northwest Territories, out-of-territory placements were not distinguished

from in-territory placements, and so were only included in “other” if the placed

child was not in a form of family or group care.

Table 2. Placement type definitions in CCWIS data

{Classification}

Methods

• Aggregate data (99.6%) and de-identified record-level data (0.4%)

➢ Missing data from ~50 First Nations agencies in four provinces (NB, ON, SK, and QC)

• Sensitivity analysis with public data from Indigenous Services Canada

• Sex/gender data covers 6 provinces/territories

• Age group data for 6 provinces/territories

• Placement type data for 10 provinces/territories

{Classification}

Rates of out-of-home care by year and province/territory, 2013/14 – 2021/22

Notes: Numerator in rate calculations is based on a 31 March point-in-time count. Prince Edward Island data for 2021/2022 were missing; 2020/2021 data were used for the

rate calculation. Nunavut kinship data for 2017/2018 were used for the rate calculation in 2018/2019. ISC data not included for any year or jurisdiction.

0

5

10

15

20

25

30

35

40

2013/2014 2014/2015 2015/2016 2016/2017 2017/2018 2018/2019 2019/2020 2020/2021 2021/2022

R a te

p e r

1 ,0

0 0 p

o p u la

ti o n

Fiscal year (April 1 – March 31)

Manitoba

Nunavut

Northwest Territories

Saskatchewan

Yukon

Newfoundland and Labrador

Prince Edward Island

British Columbia

Quebec

Alberta

New Brunswick

Nova Scotia

Ontario

{Classification}

Canada

(High estimate)

9.5

14% rate increase 8.2

Sensitivity analysis with data from Indigenous Services Canada (ISC)

Data from Provinces and

Territories

Data from Indigenous

Services Canada** (n=9,330)

70,434 Children in out-of-home care

on March 31, 2022

(Data from 13 Provinces/Territories +

Indigenous Services Canada)

61,104 Children in out-of-home care

on March 31, 2022

(Data from 13 Provinces/Territories)

Interpretation Tip:

This sensitivity analysis estimates a

maximum national rate by including ISC data

to account for data gaps in four provinces.

Canada

(Low estimate)

Data from Provinces and

Territories

Missing data*

Rate per 1,000 population

Notes: *Missing data from First Nations agencies in

New Brunswick, Ontario, Saskatchewan, and Quebec.

Notes: **ISC data was derived from a publicly reported count of

children in out-of-home care on March 31, 2020. Some children

counted in ISC data were also likely counted in data from

provinces and territories.

{Classification}

Rate ratio of children in out-of-home care, by province/territory, Canada, 2021/2022

24

0 1 2 3 4 5

Ontario

Nova Scotia

New Brunswick

Alberta

Quebec

British Columbia

Prince Edward Island

Newfoundland and Labrador

Yukon

Saskatchewan

Northwest Territories

Nunavut

Manitoba

Reference group: l National rate (data from 13 provinces/territories and Indigenous Services Canada) l National rate (data from 13 provinces/territories)

P ro

v in

c e /T

e rr

it o

ry

Interpretation Tip

Dots to the right of the red line (greater than 1)

represent a rate that is X times higher than the

national rate of out-of-home care.

Dots to the left (less than 1) represent rates that

are lower than national rate.

Dots that touch or are very close to the red line are

similar to the national rate.

Rate ratio (95% Confidence Interval)

{Classification}

Next steps

25

• Support partnerships with Provincial,

Territorial, Indigenous, and delegated

child welfare departments and agencies

• Collaborate with federal departments

• Design models for data governance

• Harmonize definitions and indicators

• Develop a minimum dataset

• Validate data and seek feedback on

interpretation

• Answer priority research questions

COLLABORATE SHARE

Peer-reviewed

articles

Data Access

for Research

Annual

Reports

Media

Requests

Data

Dashboard Indicator

Framework

ANALYZE

Epidemiology

• Track indicators across time, place, and populations

Research

• Investigate upstream risk & protective factors

• Assess the impact of policy changes

The Canadian Child Welfare Information System can be an evidence hub.

Child Functioning and Social Participation in Canada using the Washington Group/United Nations Children’s Fund Child Functioning Module

Stacie Kerr

Dafna Kohen, Leanne Findlay, and Rubab Arim

Statistics Canada

Child disability data in Canada

• National reporting is largely based on survey data:

National disability surveys Child health surveys

Participation and Activity Limitation Survey 2001, 2006

National Longitudinal Survey of Children and Youth 1994-2009, biennially

Canadian Survey on Disability 2012, 2017, 2022

Canadian Health Survey on Children and Youth 2016, 2019, 2023

Study objectives

1. Describe the sociodemographics of Canadian children with and without functional difficulties based on the Washington Group/United Nations Children’s Fund Child Functioning Module (CFM)

2. Examine associations with social participation in contexts including child care, school, and organized sports and clubs

3. Examine the use of the CFM to disaggregate Sustainable Development Goal (SDG) indicators

Data source

• Canadian Health Survey on Children and Youth, 2019

• National, cross-sectional survey that collected data on issues impacting the physical and mental health of Canadian children

• N=43,000 children aged 2-17

• Difficulties in various functional domains were assessed using the CFM

Identifying children with functional difficulties in Canada using the CFM

• An estimated 700,000 Canadian children aged 2-17 were identified with functional difficulties (12% of the population)

• The most identified functional difficulties were:

– communication among children aged 2-4 (3%)

– accepting change among children aged 5 to 11 (5%)

– anxiety among children aged 12-17 (6%)

Sociodemographic characteristics of Canadian children aged 2-17 years with and without functional difficulties based on the CFM

0

10

20

30

40

50

60

Without FD With FD

2-17 years

%

Female

Aged 2-4

Lone parent household

Unemployed parents

Parent has highschool education or less

*

*

*

*

*

* Statistically significant difference between children with and without functional difficulties (p < 0.05)

Using the CFM to examine social participation

Disability can be conceptualized as a social disadvantage imposed by an unsupportive environment in addition to an individual’s functional difficulties

Source: UNICEF, 2021

Child care use among children aged 2-11 with and without functional difficulties

25

30

35

40

45

50

55

60

65

Without FD With FD Without FD With FD

2-4 years 5-11 years

%

*

* Statistically significant difference between children with and without functional difficulties (p < 0.05)

Attended school 5 days or more in the past week among children aged 5-17 with and without functional difficulties

* Statistically significant difference between children with and without functional difficulties (p < 0.05)

60

62

64

66

68

70

72

74

76

Without FD With FD Without FD With FD

5-11 years 12-17 years

%

*

*

Participation in organized sports and clubs in the past 12 months among children aged 5-17 with and without functional difficulties

0

10

20

30

40

50

60

70

80

90

Without FD With FD Without FD With FD

5-11 years 12-17 years

%

Physical activities with coach or instructor

Music, drama, art clubs or lessons

Other organized clubs or community groups

*

*

*

*

* Statistically significant difference between children with and without functional difficulties (p < 0.05)

The CFM as a means to disaggregate SDG indicators

Canada adopted the 2030 Agenda for Sustainable Development in 2015

• Residence in low-income households

• Household food insecurity

• Lower general health

• Lower mental health

Low-income, food insecurity, and lower general and mental health among children aged 2-17 with and without functional difficulties

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

Without FD With FD Without FD With FD

2-11 years 12-17 years

%

Low-income

Food insecure

Lower general health

Lower mental health

*

**

*

* *

*

*

* Statistically significant difference between children with and without functional difficulties (p < 0.05)

Limitations and future directions

• Little is known about measurement invariance of the CFM across diverse groups in the population

• Include the CFM and GALI together on a future Canadian child survey to enable a head-to-head comparison and investigate the GALI as a proxy

CFM GALI (Global Activity Limitation Indicator)

Age group covered Children aged 2-17 Children of all ages

Number of items 16 items (ages 2-4) 24 items (ages 5-17)

2 items

Conclusions

• Findings demonstrate the utility of the CFM to report on child functional difficulties in Canada:

– Sociodemographic characteristics were associated with child functioning in the expected directions

– Lower rates of social participation were observed among children with functional difficulties than children without functional difficulties

– Children with functional difficulties had poorer results for four SDG-related indicators than children without functional difficulties

Stacie Kerr

Health Analysis Division, Statistics Canada

[email protected]

Statistics Canada’s Physical Flow Account for Plastic Material

9th Seminar on the Implementation of the System of Environmental-Economic Accounting (SEEA)

March 19, 2024

Outline

Part I – Physical Flow Account for Plastic Material (PFAPM)

Part II – Surveys related to the Circular Economy

Part I – Physical Flow Account for Plastic Material (PFAPM)

Background Canada’s Zero Plastic Waste Agenda

• Ocean Plastics Charter (2018) • Canada-wide Strategy on Zero Plastic Waste (2018) • Canada-wide Action Plan on Zero Plastic Waste:

• Phase 1 of the Action Plan (2019) • Phase 2 of the Action Plan (2020)

4

Background – PFAPM

• How much plastic is produced • What is the fate of that plastic:

• What amount is discarded? • What amount is leaked into the

environment? • How much is collected? • How much is diverted from the waste

stream? • How much of that is baled/recycled? • How much goes to the landfill or Energy

from Waste (EFW)? • All reported in physical units (tonnes)

5

Background – PFAPM

• The PFAPM includes: • 17 variables (from production to end of life) • Two tables:

• 14 product types • 19 resin types

• National and provincial/territorial data • 2012-2020

7

PFAPM Variables

PFAPM – Product Categories (provincial-level data) • Construction materials • Electrical and electronic

equipment • Electronics • Major appliances • Other electrical and

electronic equipment • Packaging

• Bottles • Film • Non-bottle rigid • Other packaging

• Agricultural film • Textiles • Vehicles • Other products • Total

PFAPM – Resin Types (national-level data) • Total, all resins • Polyethylene terephthalate (PET) resins • Other thermoplastic polyester resins • Low-density polyethylene resins • Linear low-density polyethylene resins • High-density polyethylene resins • Other polyethylene resins • Polystyrene resins • Acrylonitrile-butadiene-styrene resins • Polyvinyl chloride resins • Polypropylene resins • Thermoplastic polyurethane resins

• Polyamide (nylon) resins • All other thermoplastic resins • Phenolic resins • Urea formaldehyde resins • All other formaldehyde based resins • Unsaturated polyester (thermosetting)

resins • Thermosetting polyurethane resins • Other thermosetting resins

PFAPM – Resin Types (national-level data)

• The industry-technology construct, derived from the Supply and Use Tables (SUT), is a matrix showing the distribution of plastic resin (input) across products (output)

• The SUT aggregates all plastic resins into one commodity. To disaggregate the plastic resin into the 19 resin types, a custom resin distribution construct was developed to link the Annual Survey of Manufacturing and Logging Industries (ASMLI) data on industries’ resin use with the industry-technology construct showing inputs of the resin commodity by product

• This method also tries to capture the complexity of industries producing multiple products and multiple industries producing the same product

PFAPM Methodology

• The account uses multiple data sources, such as: • Supply and Use Tables • StatCan survey data (e.g., Waste Management Survey,

Manufacturing Survey) • Administrative data

12

Methodology part I: Plastic production

Methodology – Production side

Plastic in domestically produced products

Plastic content in net trade (imports less exports) of newly produced products

Plastic in products produced for Canadian consumption

Methodology – Production side

• Main data sources and their uses • Supply and Use Tables (basis of the model) • Canadian International Merchandise Trade Database (convert $ to

tonnes) • Annual Survey of Manufacturing and Logging Industries (Resin mix

and Prices) • Capital and Repair Expenditure (product life-time) • Reports used for assumptions to fill any remaining gaps

15

Methodology, part II: Tracking the fate of plastic waste

• 3 main types of data sources are used • StatCan surveys (i.e. Waste Management

Survey) • Alternative data sources (e.g. Waste

characterization reports, bale characterization reports, annual reports from industry stewardship programs, provincial/territorial reports)

• Academic or peer-reviewed reports • Production-based estimates and/or information

from reports where data is not available

Methodology – Fate of plastic waste

Plastics Material Flow Account – dissemination products

1 7

Part II – Surveys related to the Circular Economy

Surveys related to the CE - WMS The Waste Management Survey (WMS) • Collects data about waste disposal and diversion • 3 types of respondents:

• Businesses that provide waste management services • Government entities • Retailers and other businesses that carry out diversion activities such as take-it-back programs (new since 2020)

• Biennial survey since 1996 • In 2023, WMS becomes annual • For odd year cycles, WMS will collect data on composting and recycling • Disposal estimates for these cycles will be modelled

• New survey content added to better understand how diverted plastics are managed: • The percentage of material in recycling facilities disposed as processing residue • Whether recycling facilities bale their plastics • If they do, what type of plastics and what quantities are baled (PET, HDPE, Mixed rigid, etc.)

Surveys related to the CE - ASMLI

Annual Survey of Manufacturing & Logging Industries

• Measures the contribution of manufacturing to economic activity in Canada (domestic production)

• Covers all establishments primarily engaged in manufacturing

• Collects financial and operational statistics and commodity information

• Results published by sub-manufacturing industries and by provinces and territories

Surveys related to the CE – ASMLI & PFAPM Introduction of the plastic module for RY2020 (updated for RY2021):

• Questions added related to raw material purchases (inputs) from the commodity section

• Focus on use of virgin and post-consumer/post-industrial plastics, by resin type, in the manufacturing process

• Targets companies that report specific resins as an input, no matter which the industry they are part of

• Increase of the sample by about 600 units from plastics and chemical manufacturing companies

• Main variables are values (cost of purchases) and quantity

Surveys related to the CE - HES Households and the Environment Survey (HES)

• The HES is a biennial household survey that produces estimates at the national, provincial and, for most variables, the municipal levels of geography

• The purpose of the survey is to measure the environmental practices and behaviours of Canadian households

• The 2021 HES included content related to single-use plastics (e.g., plastic drinking straws, plastic bags) and some plastics avoidance activities (e.g., refillable water bottles, and reusable cups and mugs)

• The 2023 HES is expanding content to cover more single-use products and content related to the concept of repair-or-replace

Questions For questions, contact Michelle Hughes [email protected]

OR Michele McMillan [email protected]

List of tables

Table1

Table2

Table 2a

Table 2b

Table 3

Table 4

Table 5

Table 5a

Table 6

Table 6a

Table 6b