Canadian Community Health Survey
Cycle 1.2
Mental Health and Well-being
Public Use Microdata File Documentation
Table of Contents
1. Introduction............................................................................................................................... 1
2. Background ............................................................................................................................... 2
3. Objectives ................................................................................................................................. 4
4. Survey Content.......................................................................................................................... 5
4.1 Consultation Processes ............................................................................................... 5
4.2 Content........................................................................................................................ 6
5. Sample Design ........................................................................................................................ 33
5.1 Target Population...................................................................................................... 33
5.2 Sample Size and Allocation...................................................................................... 33
5.3 Sample Buy-ins......................................................................................................... 34
5.4 Sampling of Households from the Area Frame ........................................................ 35
5.5 Sampling of Respondents ......................................................................................... 37
5.6 Sample Allocation over the Collection Period ......................................................... 38
6. Data Collection ....................................................................................................................... 39
6.1 Questionnaire Design and Data Collection Method ................................................. 39
6.2 Supervision and Control ........................................................................................... 39
6.3 Field Tests................................................................................................................. 39
6.4 Interviewing .............................................................................................................. 40
6.5 Minimising Non-response......................................................................................... 40
6.6 Special Circumstances during Cycle 1.2 Collection Operations .............................. 41
7. Data Processing....................................................................................................................... 42
7.1 Editing....................................................................................................................... 42
7.2 Coding....................................................................................................................... 42
7.3 Creation of Derived and Grouped Variables ............................................................ 42
7.4 Imputation................................................................................................................. 43
7.5 Weighting.................................................................................................................. 43
7.6 Suppression of Confidential Information.................................................................. 44
8. Weighting................................................................................................................................ 45
8.1 Adjustments Applied to the Initial Weight ............................................................... 45
9. Data Quality ............................................................................................................................ 50
9.1 Response Rates ......................................................................................................... 50
9.2 Survey Errors ............................................................................................................ 52
9.2.1 Non-sampling Errors...................................................................................... 52
9.2.2 Sampling Errors ............................................................................................. 53
10. Guidelines for Tabulation, Analysis and Release................................................................. 54
10.1 Rounding Guidelines .............................................................................................. 54
10.2 Sample Weighting Guidelines for Tabulation ........................................................ 55
10.2.1 Definitions: Categorical Estimates, Quantitative Estimates........................ 55
10.2.2 Tabulation of Categorical Estimates............................................................ 56
iii
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
10.2.3 Tabulation of Quantitative Estimates .......................................................... 57
10.3 Guidelines for Statistical Analysis.................................................................. 57
10.4 Release Guidelines.................................................................................................. 59
11 Approximate Sampling Variability Tables ............................................................................ 60
11.1 How to Use the CV Tables for Categorical Estimates............................................ 60
11.1.1 How to Use the disorder CV tables ............................................................. 62
11.1.2 How to Use the general CV tables............................................................... 65
11.2 Examples of Using the CV Tables for Categorical Estimates.............................. 67
11.3 How to Use the CV Tables to Obtain Confidence Limits .................................... 72
11.4 Example of Using the CV Tables to Obtain Confidence Limits .......................... 73
11.5 How to Use the CV Tables to do a Z-test............................................................. 73
11.6 Example of Using the CV Tables to do a Z-test................................................... 74
11.7 Exact Variances/Coefficients of Variation........................................................... 74
11.8 Release Cut-off for the CCHS .............................................................................. 76
12. File Usage ............................................................................................................................. 85
12.1 Use of Weights........................................................................................................ 85
12.2 Variable Naming Convention ................................................................................. 85
12.2.1 Variable Name Component Structure in CCHS .......................................... 85
12.2.2 Positions 1 to 3: Variable / Questionnaire Section Name............................ 86
12.2.3 Position 4: Cycle.......................................................................................... 87
12.2.4 Position 5: Variable Type ............................................................................ 87
12.2.5 Positions 6 to 8: Variable Name .................................................................. 88
12.3 Access to Master File Data ..................................................................................... 88
Appendix A:
Questionnaire
Appendix B:
Record Layout
Appendix C: Derived and Grouped Variables
Appendix D:
CV Tables
Appendix E:
Data Dictionary
Alphabetical Index
Topical Index
iv
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
1. Introduction
The Canadian Community Health Survey (CCHS) is a cross-sectional survey that
collects information related to health status, health care utilization and health
determinants for the Canadian population. The CCHS operates on a two-year
collection cycle. The first year of the survey cycle “.1” is a large sample, general
population health survey, designed to provide reliable estimates at the health region
level. The second year of the survey cycle “.2” is a smaller survey designed to
provide provincial level results on specific focused health topics.
This Microdata File contains data collected in the second year of collection for CCHS
(Cycle 1.2). Information was collected between May 2002 and December 2002, for
the ten provinces. Cycle 1.2 collects responses from persons aged 15 or older, living
in private occupied dwellings. Excluded from the sampling frame are individuals
living on Indian Reserves and on Crown Lands, Health Care institution residents, full-
time members of the Canadian Armed Forces, and residents of certain remote regions.
This document has been produced to facilitate the manipulation of the Cycle 1.2
cross-sectional microdata file, which is described in detail in the following text and
appendices.
Any questions about the data sets or their use should be directed to:
• For technical/general data support:
Electronic Products Help Line: 1-800-949-9491
• For custom tabulations/general data support:
Client Custom Services, Health Statistics Division: 1-613-951-1746
E-mail:
• For remote access support: 1-613-951-1653
E-mail:
Fax: 1-613-951-4198
1
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
2. Background
In 1991, the National Task Force on Health Information cited a number of issues and
problems with the health information system. These problems were that: data was
fragmented; data was incomplete; data could not be easily shared; data was not being
analysed to the fullest extent; and the results of research are not consistently reaching
Canadians. In responding to the needs, the Canadian Institute for Health Information
(CIHI), Statistics Canada and Health Canada have joined forces to create a National
Health Information Roadmap.
The Roadmap is a direct response to the concerns and desires of more than 500
individuals representing a wide range of organizations and interest groups. Early in
1999, the three national organizations listed above conducted a broadly based
national consultation on health information needs. Participants stressed that national
agencies must work together to strengthen Canada’s health information system, and
must build on and contribute to the considerable investments and expertise at local,
regional, and provincial/territorial levels.
The Roadmap represents an important contribution to building a comprehensive
national health information system and infrastructure to provide Canadians with the
information they need to maintain and improve Canada’s health system and the
population’s health. What is needed is a co-ordinated plan of action. No single
government or organization can combat the above-noted problems alone. Co-
operation at all levels – national, provincial, territorial, regional and local health
organizations – is a prerequisite for success.
The plan of action starts by seeking answers to two crucial questions:
1. How healthy is the health care system?
2. How healthy are Canadians?
The first question encompasses the effectiveness, efficiency and responsiveness of the
health care system. Generally, an effective, efficient and responsive health care
system is one that offers the quality of care Canadians expect.
The second question is broader, and addresses the basic objective of the system: Is the
health of Canadians improving? To answer this, a strong health information system is
needed. This information system must embrace six principle characteristics.
2
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
The information system must be:
• Secure and respectful of Canadians’ privacy;
• Consistent;
• Relevant;
• Integrable;
• Flexible;
• User-friendly and accessible.
This health information system needs to be timely, provide person-oriented
information, and have common data standards with other Canadian health surveys,
such as the National Population Health Survey (NPHS). The new system must also
provide: new or expanded data sets; data on health services; data on outcomes, health
status and non-medical determinants of health; data on outcomes of selected health
interventions; implement special studies involving priority issues; data on costs per
service; information exchange protocols; expanded analytical and dissemination
capacity, and public reports on the health care system.
Given this mandate, the Canadian Community Health Survey (CCHS) was conceived.
The format, content and objectives of the CCHS evolved through extensive
consultation with key experts, federal, provincial and community health region
stakeholders to determine their data requirements.
The purpose of this publication, the Public Use Microdata File, is to follow through
on the mandate of collecting reliable, relevant information on health services, health
status, and health issues of importance to Canadians - at the provincial and national
level - and disseminating this information to the public.
3
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
3. Objectives
The Cycle 1.2 mainly measures aspects linked to the mental health of Canadians.
This cycle was then named “Canadian Community Health Survey - Mental Health
and Well-being”. The primary objectives of the CCHS Mental Health and Well-
being are to:
• Provide timely, reliable, cross-sectional estimates of mental health determinants,
mental health status and mental health system utilization across Canada;
• Determine prevalence rates of selected mental disorders to assess the impact of
burden of illness;
• Juxtapose access and utilization of mental health services with respect to
perceived needs; and
• Assess the disabilities associated with mental health problems to individuals and
society.
As a key component of the Population Health Surveys Program of Statistics Canada,
the CCHS helps fulfil broader requirements of health issues in Canada. These are to:
• Aid in the development of public policy;
• Provide data for analytic studies that will assist in understanding the determinants
of health;
• Collect data on the economic, social, demographic, occupational and
environmental correlates of health;
• Increase the understanding of the relationship between health status and health
care utilization.
4
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
4. Survey Content
The first sub-section of this section provides a general discussion of the consultation
process used in survey content development and gives a summary of the final content
selected for inclusion in this study.
4.1 Consultation Processes
One of the main CCHS Mental Health and Well-being objectives is to address
priority mental health determinants, mental health status and mental health system
utilization data gaps at the provincial level. Topic selection for the content of the
Cycle 1.2 was conducted through a process of extensive consultations with
regional, provincial, federal representatives and the research community. Expert
consultation was seen as an integral part of the content development. The
selection of priority areas in terms of mental disorders as well as mental well-
being have been a result of discussions within the Mental Health Expert Group
assembled for the survey, as well as the Population Health Advisory Committee.
Consultations also include contacts with representatives of the World Health
Organization, academia, federal and provincial governments, consumers and
professional associations.
Table 4.1: Questionnaire Modules
• Administration
• Agoraphobia
• Alcohol Dependence
• Alcohol Use
• Chronic Conditions
• Distress
• Eating Troubles Assessment
• General Health
• Height and Weight
• Household Contact and Demographics
• Illicit Drug Use and Dependence
• Income
• Labour Force
• Major Depressive Episode
• Manic Episode (Mania)
• Medication Use
• Mental Health Services
• Panic Disorder
• Pathological Gambling
• Physical Activities
• Psychological Well-being
Manifestation Scale
• Restriction of Activities
• Screening Section
• Socio-demographic Characteristics
• Social Phobia
• Social Support
• Spiritual Values
• Stress
• Two-week Disability
• Work Stress
5
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
4.2 Content
The content for Cycle 1.2 is partly based on a selection of mental disorders from
the WMH-CIDI (World Mental Health – Composite International Diagnostic
Interview Instrument). The WMH-CIDI is a lay-administered psychiatric
interview that generates a profile of those with a disorder according to the
definitions of the Diagnostic and Statistical Manual of Mental Disorders, 4
th
Edition (DSM-IV). The WMH-CIDI questions and algorithms were
operationalised to meet the needs of CCHS 1.2. For the purposes of this survey,
the questions and algorithms are referred to as “CCHS 1.2/WMH-CIDI”. Similar
to the WMH-CIDI, the CCHS 1.2/WMH-CIDI can not be used to measure all
aspects associated with the DSM-IV definitions and classification.
The well-being and determinants of health in Cycle 1.2 are based on sources used
on such surveys as the National Population Health Survey (NPHS), the CCHS
(Cycle 1.1), the Health Promotion Survey (HPS) and other surveys. The following
Table 4.2 provides a detailed breakdown of sources and changes from earlier
usage in CCHS and NPHS.
6
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table 4.2 CONTENT DESCRIPTION – CCHS 1.2
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Introduction Social survey standards. Social survey standards.
1
Household/
Entry
CCHS 1.1 Information collected on each
household member:
Date of birth
Sex
Marital status
Relationship of everyone to everyone
else in the household
Highest grade of elementary or high
school completed
Highest degree, certificate or
diploma
Type of dwelling
Owner/tenant
Number of bedrooms in dwelling
Language of interview
2
GEN
General Health CCHS 1.1 Respondent’s general health status.
General health status compared to
one year ago.
Sleep patterns.
General level of stress in life.
Sense of belonging to one’s
community.
Respondent’s general health status.
General health status compared to
one year ago.
Sleep patterns.
General level of stress in life.
Sense of belonging to one’s
community.
GENB_04 – GENB_06: Replaced
scale with one that is used in Social
support.
New scale :
None of the time
A little of the time
Some of the time
Most of the time
All of the time
Old scale :
Most of the time
Sometimes
Never
Scale was replaced with more
specific/definitive one to give
respondents more options; better
capture what they are feeling.
3
SCR
Screener for
mental disorders
WMH-CIDI In order to reduce response burden,
screening questions are used to
identify and filter respondents who
may experience symptoms that are
associated with specific mental
disorders.
New Module A screener question is found for
generalized anxiety disorder
although CCHS 1.2 did not collect
other information about this specific
disorder. Only the Canadian Forces
2002 Canadian Community Health
7
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
The screening questions are grouped
in one module, which is located near
the beginning of the questionnaire, to
avoid false negatives and learning
pattern where a respondent answers
‘no’ in order to screen out of a
subsequent module.
The screening questions ask if the
respondent has experienced general
symptoms that are usually associated
with different affective and anxiety
disorders. Positive answers are
flagged for further questioning
within disorder-specific modules.
These modules contain more in-
depth questions and relate to more
specific symptoms associated with
the mental disorder. The purpose of
these additional questions is to
evaluate if the respondent has
experienced symptoms that are
identified with the clinical diagnosis
associated with the reviewed mental
disorder.
SCRB_081: Self-perceived rating of
physical health.
SCRB_082: Self-perceived rating of
mental health.
SCRB_20 through SCRB_35:
WMH-CIDI screener questions
(excluding suicide, eating troubles,
gambling, alcohol use and
dependence and illicit drug use and
dependence).
Survey Supplement on Mental
Health collected information on this
disorder.
Screener questions should not be
used as a proxy measure for the
disorder.
8
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
4
CCC
Chronic
conditions
CCHS 1.1
NPHS
Cycle 4
Allergies
Asthma
Fibromyalgia
Arthritis or rheumatism
Back problems excluding
fibromyalgia and arthritis
High blood pressure
Migraine headaches
Chronic bronchitis
Emphysema or COPD
Diabetes
Epilepsy
Heart disease
Cancer
Stomach or intestinal ulcers
Effects of a stroke
Bowel disorder, Crohn’s or Colitis
Alzheimer's disease or other
dementia
Cataracts
Glaucoma
Thyroid condition
Chronic fatigue syndrome
Multiple chemical sensitivities
Following not measured in NPHS or
CCHS 1.1:
Schizophrenia
Other psychosis
Obsessive-compulsive disorder
Dysthymia
Post-traumatic stress disorder
Autism or any other developmental
disorder such as Asperger’s
syndrome or Rett syndrome
Learning disability
Type of learning disability (Attention
Deficit Disorder, no hyperactivity
Allergies
Asthma
Fibromyalgia
Arthritis or rheumatism
Back problems excluding
fibromyalgia and arthritis
High blood pressure
Migraine headaches
Chronic bronchitis
Emphysema or COPD
Diabetes
Epilepsy
Heart disease
Cancer
Stomach or intestinal ulcers
Effects of a stroke
Urinary incontinence (not measured
in CCHS 1.2)
Bowel disorder, Crohn’s or Colitis
Alzheimer's disease or other
dementia
Cataracts
Glaucoma
Thyroid condition
Parkinson’s disease (not measured in
CCHS 1.2)
Multiple sclerosis (not measured in
CCHS 1.2)
Chronic fatigue syndrome
Multiple chemical sensitivities
Other long term condition
Not included:
All questions related to the age when
the diagnosis was made, sub-type of
condition or specific type.
Not included:
CCCA_161: urinary incontinence
CCCA_231: Parkinson’s disease
CCCA_241: Multiple sclerosis
Added new categories on mental
disorders:
CCCB_271: Schizophrenia
CCCB_281: Other psychosis
CCCB_291: Obsessive-compulsive
disorder
CCCB_301: Dysthymia
CCCB_311: Post-traumatic stress
disorder
CCCB_321: Autism or any other
developmental disorder such as
Asperger’s syndrome or Rett
syndrome
CCCB_331: Learning disability
CCCB_33A-D: Type of learning
disability (Attention Deficit
Disorder, no hyperactivity (ADD),
Attention Deficit Hyperactivity
Disorder (ADHD), Dyslexia, Other)
CCCB_341: Eating disorder such as
anorexia or bulimia
Time saving measure.
Low counts and little relevance to
mental health.
Can be picked up by other
categories.
Long term conditions are needed to
better understand the impacts
associated with the experience of
mental health symptoms. There are
links between long term conditions
and mental health problems. These
were added because they were not
measured in the diagnostic modules
and mental health experts had shown
interest in these areas.
9
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
(ADD), Attention Deficit
Hyperactivity Disorder (ADHD),
Dyslexia, Other)
Eating disorder such as anorexia or
bulimia
Other long term physical or mental
problem
5
HWT
Height and
Weight
CCHS 1.1
NPHS
Cycle 4
Height/weight (self-reported).
Self-perceived body image.
Height/weight (self reported).
Self-perceived body image.
Added:
HWTB_01: question on pregnancy.
This question was included in 1.1 but
in a different location within the
questionnaire.
In order to properly calculate BMI,
need to know if respondent is
pregnant.
Correlate of mental health.
6
PAC
Physical
Activity
CCHS 1.1
NPHS
Cycle 4
Participation in physical activity in
the past 3 months.
Frequency of participation.
STRB_66: respondents are asked if
they jog or do other exercises to deal
with stress.
Participation in physical activity in
the past 3 months.
Frequency of participation.
Added:
PACB_6: added sentence at
beginning of question; “Now I’m
going to read you 4 sentences that
can be used to describe the amount
of physical activity that people
usually do.”
Correlate of mental health.
7
PWB
Psychological
Well being Scale
Massé’s
Well-being
scale
New
Self-esteem
Balance
Social involvement
Sociability
Self-control
Happiness
N/A New module Massé’s scale was used because it
encompassed the aspects of the
Bradburn Scale as well as the
Mastery module.
Correlate of mental health.
8
DIS
Distress CCHS 1.1
Kessler’s scale called the K10
measuring the frequency of feeling:
NPHS
Cycle 4
Sad
Nervous
Restless
Hopeless
Worthless
Everything was an effort
Kessler and Mroczek scale called the
K6 measures frequency of feeling:
Sad
Nervous
Restless or fidgety
Hopeless
Worthless
Everything was an effort
Replaced by Kessler’s 10-items scale
(very similar to the old one):
Frequency of feeling:
Tired for no good reason
Nervous
So nervous that nothing could calm
down
Hopeless
Replaced with new scale to be
consistent with Kessler (K10).
Measures important aspect of mental
health and can be superimposed with
diagnostic modules.
10
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
And:
Tired for no good reason
So nervous that nothing could calm
down
“…or fidgety”
So restless couldn’t sit still
“…or depressed”
Chronic aspects of distress are also
examined (i.e. more often, less often,
or same as usual in the past month).
Chronic aspects of distress are also
examined (i.e. more often, less often,
or same as usual in the past month).
Restless or fidgety
So restless couldn’t sit still
Sad or depressed
So depressed that nothing could
cheer up
Everything was an effort
Worthless
New question order:
DISB_10A (so sad…) became
DISB_10H (so depressed…).
DISB_10C became DISB_10E.
DISB_10E became DISB_10J.
DISB_10F became DISB_10I
Added:
INT: intro text emphasising the 1 -
month reference period
DISB_10A: tired out for no good
reason
DISB_10C: so nervous that nothing
could calm you
DISB_10F: so restless you could not
sit still
DISB_10G: sad or depressed
DISB_10N: text emphasising 1 -
month reference period
Using a booklet to display response
categories.
Clear emphasis on reference periods
is important because there are so
many different reference periods in
the survey.
Will reduce interview time and
response burden.
9
STR
Stress New STRB_1: Rating of the ability to
handle unexpected and difficult
problems (adapted from NPHS 96).
STRB_2: Rating of the ability to
handle the day-to-day demands of
life (adapted from NPHS 96).
New module
Items for ways of coping questions
(STRB_61 to STRB_611) are
derived and modified in wording
from several coping scales. The
majority of questions are selected
from Folkman and Lazarus (1985)
Ways of Coping Revisited (WOC-
11
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
STRB_3: Respondent’s self report of
the source of stress in day-to-day life
contributing most to feelings of
stress:
Time pressures / not enough time
Own physical health problem or
condition
Own emotional or mental health
problem or condition
Financial situation (e.g., not enough
money, debt)
Own work situation (e.g., hours of
work, working conditions)
School
Employment status (e.g.,
unemployment)
Caring for own children
Caring for others
Other personal or family
responsibilities
Personal relationships
Discrimination
Personal and Family’s safety
Frequency use of ways of dealing
with stress:
STRB_61: Try to solve the problem
STRB_62: Talk to others
STRB_63: Avoid being with people
STRB_64: Sleep more than usual
STRB_65A: Try to feel better by
eating more or less than usual
STRB_65B: Try to feel better by
smoking more cigarettes than usual
STRB_65C: Try to feel better by
drinking alcohol
STRB_65D: Try to feel better by
using drugs or medication.
R). Several questions are selected
from Amirkhan (1990) Coping
Strategy Indicator (CSI), and Carver
et al. (1989) COPE scale.
12
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
STRB_66: Jog or do other exercise
STRB_67: Pray or seek spiritual help
STRB_68: Try to relax by doing
something enjoyable
STRB_69: Try to look on the bright
side of things
STRB_610: Blame oneself
STRB_611: Wish the situation would
go away or somehow be finished
10
Each of the
following
“disorders”
(10A Major
Depressive
Episode,
10B Manic
Episode (Mania),
10C Panic
Disorder,
10D Social
Phobia,
10E Agoraphobia)
WMH-CIDI NOTE:
CCHS 1.2 questionnaire is based on
a modified version of the WMH-
CIDI instrument recognized by the
WMH2000 Project.
Diagnostic disorders were selected
on the basis that they were prevalent
enough to provide reliable estimates,
that they were balanced across
disorders, that they were treatable
and program/policy relevant and
were guided by recommendations
from the Survey’s Expert
Committee.
N/A New modules
10A
DEP
Major depressive
episode
(including
suicide thoughts
and attempts)
WMH-CIDI Episodes of:
Being “sad, empty or depressed”
Losing interest in most things
Feeling discouraged about how
things are going in life
Duration:
Episodes of 2 weeks or longer
recurring month after month for a
year or longer
Frequency:
Most of the day, nearly everyday
Used short CIDI scale for major
depressive episode. This is a subscale
of an earlier version of the CIDI
instrument.
New module
Note that the Suicide module was
integrated into major depressive
episode (see Suicide below).
Streamlined skips and wording for
services section of major depressive
episode to take into consideration
services mentioned in suicide
section.
Integrating suicide into major
depressive episode facilitates
programming, ensures consistency in
questions (between those who go
through major depressive episode
and those who don’t), and keeps
sensitive questions on a similar topic
together (instead of asking about
suicide in major depressive episode
and again in the suicide module).
13
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Severity:
Worst and most recent episodes
Specific symptoms:
At least 5 symptoms such as change
in weight or appetite, energy level,
ability to concentrate
Added booklet for suicide questions
and refer to thoughts of suicide,
suicide plan, and attempted suicide
as Experience A, B, and C.
Added word “seriously” to thoughts
of suicide.
Booklet increases confidentiality if
others within earshot of interview,
eases respondent discomfort with
potentially sensitive topic.
Qualitative testing showed its
necessity. Many people said it is
normal to think of suicide as
teenagers but that they did not
“seriously” consider it.
10a
SUI
Suicidal
thoughts and
attempts
CCHS 1.1
WMH-CIDI
Measure of lifetime and past 12
months suicidal thoughts, plan and
attempt.
Age when experiences happened/ last
happened.
Most recent time when experiences
happened (in relation to the last or
worst event).
Medical attention received because
of attempt.
Hospitalization overnight or longer
because of attempt.
Professional resources sought or
talked to after suicide attempt.
Location where contacts took place.
Considered committing suicide or
taking own life in past 12 months.
Attempted to commit suicide or tried
taking own life in past 12 months.
Medical attention received following
suicide attempt.
New module
Questions on suicide are integrated
in the major depressive episode
module (see major depressive
episode above).
A confidential approach was used to
collect this information through the
reference to various experiences
(suicidal thoughts, plan and attempt)
as Experience A, B, C and the use of
a booklet.
10B
MIA
Manic Episode
(mania)
WMH-CIDI
Episodes of:
Being “so happy or excited that it
leads to trouble”
Duration:
Episodes of several days or longer
New module
14
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Severity:
Worst and most recent episodes
Specific symptoms:
At least 3 symptoms such as
restlessness, increased interest in sex,
being more friendly or outgoing than
usual, and other behaviours that
might not be usual for the individual
10C
PAD
Panic Disorder WMH-CIDI
Episodes of:
“Attacks of fear or panic” and
“attacks of suddenly experiencing
symptoms associated with a panic
disorder”
Duration:
Multiple sudden, short attacks
Severity:
Number of symptoms and episodes.
Specific symptoms:
At least 4 worsening symptoms such
as shortness of breath, heart
pounding, chest pain or discomfort,
choking or smothering sensations
New module
10D
SOP
Social Phobia WMH-CIDI
Episodes of:
“Persistent and impairing fears of
social and performance situations”
Severity:
Number of symptoms and episodes
Feeling of disappointment with
oneself
Specific symptoms:
New module
15
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
At least 2 symptoms such as feeling
dizzy, having chest pain or
discomfort, trembling, sweating or
having trouble breathing normally
10E
AGP
Agoraphobia WMH-CIDI
Episodes of:
“Persistent and impairing fears of
leaving the house or being alone
away from home”
Severity:
Number of symptoms and episodes
Feeling of disappointment with
oneself
Specific symptoms:
At least 2 symptoms such as
trembling, sweating, heart
palpitations, jitters, fatigue, tingling
in the hands and feet, nausea, rapid
pulse, sense of impeding doom
New module
11
ALC
Alcohol use CCHS 1.1
NPHS
Cycle 4
Consumption of number of drinks in
past 7 days.
Rate of use in the past 12 months and
lifetime.
Indicator of 12 month period where
once per month respondent drank 5
or more drinks in lifetime.
Reason for reducing or stopping
drinking.
Regular consumption of 5 drinks or
more on one occasion.
Consumption of number of drinks in
past 7 days.
Rate of use in the past 12 months.
Regular consumption of 12 or more
drinks a week in lifetime.
Reason for reducing or stopping
drinking.
New questions added:
ALCB_4: regular consumption of
more than 12 drinks a week during
past 12 months
ALCB_10: Lifetime occurrence of a
period where there was regular
consumption of 5 drinks or more in
one single occasion, more than once
a month for a 12 month period
Modified :
ALCA_6 became ALCB_10 in
CCHS Cycle 1.2.
Questions were added to screen for
high risk consumption patterns in
respondents.
16
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
12
ALD
Alcohol
dependence
CCHS 1.1
WMH-CIDI
Lifetime and past 12 month’s alcohol
dependence.
Questions measure Kessler and
Mroczek CIDI short form for DSM-
IIIR Alcohol Dependence (7
symptoms).
Added 5 symptoms from WMH-
CIDI:
drank more than intended, spent
excessive time recovering from
alcohol effects, reduced activities
due to alcohol, drank despite
exacerbating physical/mental
condition
Past 12 months interference in life
and activities, such as:
Home management activities
Ability to attend school
Ability to work at a job
Ability to form and maintain close
relationships
Social life
Past 12 months alcohol dependence.
Questions measure Kessler and
Mroczek CIDI short form for DSM-
IIIR Alcohol Dependence (7
symptoms).
Use of a booklet to display response
categories.
Added:
ALDB_QINT2: new introduction.
5 symptoms from WMH-CIDI
(ALDB_10 through ALDB_14):
Experience of symptoms or
situations associated with alcohol
dependence.
Examine lifetime occurrence. New
question: “Has that ever happened?”:
ALDB_01A, ALDB_03A,
ALDB_04A, ALDB_05A,
ALDB_06A, ALDB_07A,
ALDB_09A, ALDB_10A,
ALDB_11A, ALDB_12A,
ALDB_13A, ALDB_14A
Past 12 months interference in life
and activities, such as:
Home management activities
Ability to attend school
Ability to work at a job
Ability to form and maintain close
relationships
Social life
Will reduce interview time and
response burden.
Booklet also provides confidentiality
for sensitive modules and is less
intrusive (especially desired by
young respondents during testing).
To provide indication of respondents
experiencing symptoms or situations
associated with alcohol dependence
occurring in their lifetime.
Assess lifetime alcohol dependence.
Add questions to be consistent with
disorder modules.
17
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
13
DRG
Illicit Drugs
CADS
WMH-CIDI
IDGB_01 through IDGB_24:
Lifetime use (or tried) and past 12
months frequency use of:
Marijuana, cannabis or hashish
Cocaine or crack
Speed (amphetamines)
Ecstasy (MDA)
Hallucinogens, PCP or LSD
Sniffing glue, gasoline or other
solvents
Heroin
Steroids, such as testosterone,
Dianabol or growth hormones
WMH-CIDI questions partially
assessing DSM-IV substance
dependence (IDGB_25x): Lifetime
and past 12 month experience of
symptoms or situations associated
with drug dependence.
IDGB_26x: Past 12 months
interference with life and activities,
such as:
Home management activities
Ability to attend school
Ability to work at a job
Ability to form and maintain close
relationships
Social life
New module
Added:
IDGB_5AA – took drugs 1 to 3
times a month or more for a period of
12 months.
Added reference to lifetime use of
drugs (IDGB_5AL – IDGB_5HL).
New question: “Has that ever
happened?”:
IDGB_5A1, IDGB_5B1,
IDGB_5C1, IDGB_5D1,
IDGB_5E1, IDGB_5F1,
IDGB_5G1, IDGB_5H1
Added WMH-CIDI questions on
substance dependence.
Use of a booklet to display response
categories.
Past 12 months interference in life
and activities, such as:
Home management activities
Ability to attend school
Ability to work at a job
Ability to form and maintain close
relationships
Social life
Changes based on Canadian alcohol
and drug survey (CADS).
To provide indication of respondents
experiencing symptoms or situations
associated with drug use in lifetime.
CADS focused on past 12 months
only.
Expand module to examine
substance dependence.
Will reduce interview time and
response burden and is much more
confidential (especially desired by
young respondents during qualitative
testing).
Add questions to be consistent with
disorder modules.
18
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
14
CPG
Pathological
gambling
CPGI
(Canadian
Problem
Gambling
Index)
Participation in different kinds of
gambling (frequency in past 12
months) (modified from CPGI).
Amount of money spent on gambling
in past 12 months.
Past 12 months attitudes and
experiences of gambling which
indicate severity of problem
gambling (modified from CPGI).
Family history of gambling.
Past 12 month alcohol and drug use
while gambling.
N/A New module
Combined similar types of gambling
activity questions (modified from
Canadian problem gambling Index):
Instant win or daily lottery tickets (2
questions in CPGI).
Lottery tickets or raffle / fundraiser
(2 questions in CPGI).
Casino games other than coin
slots/VLT (5 questions in CPGI:
poker, blackjack, roulette, keno, and
craps).
Sports lotteries, pools, sporting
events (2 questions in CPGI)
Internet or arcade gambling (2
questions in CPGI).
Other (not included in CPGI).
Not included from CPGI:
Screener on casino games.
Use of booklet for response
categories.
Changed the names of some of the
examples associated with gambling
activities to suit region (i.e. added
Mise-au-jeu for Quebec).
Gambling activities were conducive
to grouping.
Will reduce interview time and
response burden.
All changes were approved by the
authors of the scale.
Casino games questions were
combined to one question on CCHS,
hence screener was not needed.
Booklet provides confidentiality for
sensitive modules and is less
intrusive (especially desired by
young respondents during testing).
Reduce length and response burden.
To be more tailored to regional
differences.
19
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Past 12 months interference with life
and activities, such as:
Home management activities
Ability to attend school
Ability to work at a job
Ability to form and maintain close
relationships
Social life
CPGI questions on amount of time
(in minutes or hours) spent on
gambling activities were not
included.
Combined amount of money spent to
encompass all types of gambling
activity instead of one variable per
activity and to be total in 12 months
instead of per month/per day and
added response categories instead of
open-ended.
Some questions on correlates (like
depression and suicidal thoughts)
were not included since they are
measured elsewhere in the
questionnaire.
Added the interference scale.
“Do not gamble” is not used in same
way as in CPGI.
Changed skip pattern so that if
respondent indicated that they
gambled 1-5 times per month or less
(or said don’t know or refuse) for
each type of gambling activity from
CPGB_01A to CPGB_01M or
refused the first question
(CPGB_01A), they were skipped out
of the module and not asked the
problem gambling questions. This
skip pattern is not part of the CPGI.
Reduce length and response burden
(CPGI can be scored without their
inclusion).
Reduce length and response burden
(CPGI can be scored with this
modification).
Measured elsewhere in
questionnaire.
To indicate interference caused by
gambling activity and to be
consistent with other modules.
CPG was identified during testing as
a sensitive module and respondents
felt that it was burdensome. In order
to reduce the burden on respondents,
interviewer fatigue and to maintain
high response rates, this skip pattern
was adopted.
20
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Added a category to CPGB_02,
called “I do not gamble” that is not
read by interviewers but if
volunteered by the respondent they
will be skipped out of the module.
This response category and skip
pattern is not part of the CPGI.
Some questions on severity were not
included.
These were non-essential for the
CPGI algorithm.
15
ETA
Eating Troubles EAT-26
(Eating
Attitude
Test)
WMH-CIDI
Non-diagnostic module using a
health population approach to
examine peoples' attitudes and
behaviours in relation to food and
their physical appearance.
New module
Added WMH-CIDI screener
questions to filter people who have
experienced eating problems
(ETAB_01A, ETAB_01B).
Underlined reference periods (last 12
months).
The EAT instrument is usually used
with a target population who have
eating problems. Without any
screener, the questions appear
irrelevant for many respondents.
Will reduce misreporting and
ambiguity; clarify questions.
16
RAC
Restriction of
activities
CCHS 1.1 Difficulty with hearing, seeing,
communicating, walking, climbing
stairs, bending, learning or doing
similar activities.
Question asking respondents if they
have a physical or mental condition
or health problem that reduces the
amount or the kind of activity that
they can do at home, work, school or
other activities (such transportation
or leisure).
The cause of the condition that
reduces the kind or amount of
activity the respondent can do is also
asked.
Difficulty with hearing, seeing,
communicating, walking, climbing
stairs, bending, learning or doing
similar activities.
Question asking respondents if they
have a physical or mental condition
or health problem that reduces the
amount or the kind of activity that
they can do at home, work/school or
other activities (such transportation
or leisure).
The cause of the condition that
reduces the kind or amount of
activity the respondent can do is also
asked.
RACB_6C: dropped the word
“normal” in “doing normal everyday
housework”.
Added or modified:
RACB_2B1 and RACB_2B2: Work
and school environment broken
down into 2 different questions.
RACB_5: New answer categories
listing the main cause of the
condition responsible for the activity
reduction. It now specifically
identifies “emotional or mental
health problem or condition” or “use
of alcohol or drugs”.
Qualitative testing revealed that the
word “normal” was not understood
by everyone the same way
(subjective definition).
To measure each concept separately.
To identify those whose main cause
was due to “emotional or mental
health problem or condition” or “use
of alcohol or drugs”.
21
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Embarrassment and discrimination
experienced because of the
condition.
Questions identifying the type of
activities of daily living which are
limited because of the condition.
Impact of condition (physical or
mental) on everyday activities such
as cooking, shopping, paying bills,
transportation, leisure activities, etc.
Impact of condition or health
problem on everyday activities such
as cooking, shopping, housework,
personal care, moving about inside
the house.
(New) RACB_5A1 and RACB_5A2:
Experience of embarrassment
because of a physical or mental
condition or health problem.
(New)RACB_5B1 and RACB_5B2:
Experience of discrimination or
unfair treatment because of a
physical or mental condition.
Modified:
Type of help needed from another
person because of any physical
condition or mental condition or
health problem.
RACB_6B1: Added “help with
getting to appointments and running
errands such as shopping for
groceries”.
RACB_6D: Added: “such as spring
cleaning” (Deleted “washing walls”).
RACB_6E: help “or taking
medication”.
Added: RACB_6G: help with
looking after your personal finances
such as making bank transactions or
paying bills.
Questions added concerning the
difficulties experienced by the
respondent because of any physical
condition or mental condition or
health problem. (New)
To identify those who needed help
due to “emotional or mental health
problem or condition”.
To clarify question for respondents.
To measure difficulties experienced
by the respondent because of any
physical condition or mental
condition or health problem.
22
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
RACB_7A: difficulty making new
friends or maintaining friendships.
RACB_7B: difficulty dealing with
people you don’t know well.
RACB_7C: difficulty starting and
maintaining a conversation.
RACB_8: Cause of condition
associated with required help or
difficulty experienced by the
respondent:
Physical health
Emotional or mental health
Use of alcohol or drugs
17
TWD
2 week disability CCHS 1.1 Health during the past 14 days:
Stay in bed because of illness or
injury (number of days)
Stay in bed was due to mental health
or use of alcohol or drugs (number of
days)
Days when activities were cut down
due to illness or injury (number of
days)
Days when activities were cut down
due to mental health or use of
alcohol or drugs (number of days)
Report on past 2 week disability.
Number of days spent in bed.
Number of days where respondent
cut down on activities.
Deleted:
TWDA_5: do you have a regular
medical doctor?
Added:
TWDB_2A and TWDB_2B: # of
days the respondent stayed in bed all
day because of his/her emotional or
mental health or use of alcohol or
drugs.
TWDB_4A and TWDB_4B: # of
days the respondent cut down on
things for all or most of the day
because of his/her emotional or
mental health or use of alcohol or
drugs.
TWDB_5A, TWDB_6, TWDB_6A
and TWDB_6B: # of days when it
took extra effort for the respondent
to perform to his/her usual level
23
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
because of his/her emotional or
mental health or use of alcohol or
drugs.
TWDB_1, TWDB_2: Replace “alité”
with “dû garder le lit”.
Moved:
Entire module to the end of
questionnaire so that respondents are
more comfortable about reporting the
cause of the two-week disability.
More familiar term for French
respondents.
Too much taboo/stigma associated
with mental health to be easily
reported as cause of two-week
disability.
18
SER
Mental Health
Services
Utilisation
New SERB_02 through SERB_09:
Number of hospitalisations for
problems related to emotional or
mental health or the use of alcohol or
drugs
Age at the time of first admission
Age at the time of last admission
Number of nights hospitalized during
the past 12 months
SERB_10A- SERB_10I: Contacts
with professionals about emotions,
mental health or use of alcohol or
drugs:
Psychiatrist
Family doctor or general practitioner
Other medical doctor such as a
cardiologist, gynaecologist or
urologist
Psychologist
Nurse
Psychotherapist/Social
worker/counsellor
Religious or spiritual advisor such as
a priest, chaplain or rabbi
Other professional (listed in
SERB_90 as: acupuncturist,
New module New module
24
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
biofeedback teacher, chiropractor,
energy healing specialist, exercise or
movement therapist, herbalist,
homeopath or naturopath, hypnotist,
guided imagery specialist, massage
therapist, psychic, relaxation, yoga,
or meditation expert, and dietician)
SERB_12: Age when first had a 15-
minute session of psychological
counselling or therapy.
SERB_20 through SERB_99A:
Roster of questions including:
Age when first and last talked to the
specific professional about emotions
mental health or use of alcohol or
drugs
Number of times saw or talked to the
professional about emotions, mental
health or use of alcohol or drugs
during the past 12 months
Locations where the professional
contacts took place during the past12
months
Level of satisfaction with treatments
and services received from the
professional during the past 12
months
Perceived level of help received from
the professional with regard to
problems with emotions, mental
health or use of alcohol or drugs
Reason why stopped talking to the
professional about emotions, mental
health or use of alcohol or drugs
Insurance coverage for all or part of
the cost of contacts with the specific
professional
25
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Professional recommending another
professional or other type of clinic or
program specializing in mental
health services
SERB_QA0C: Past 12 month’s use
of an internet support group or chat
room as help for problems with
emotions, mental health or use of
alcohol or drugs.
SERB_A1A-SERB_A1D: Age when
first and last used a support group as
help for problems with emotions,
mental health or use of alcohol or
drugs and past 12 months use of
support group.
SERB_A2A- SERB_A2D: Age when
first and last used a telephone
helpline service for problems with
emotions, mental health or use of
alcohol or drugs and past 12 months
use of helpline.
SERB_A4A-SERB_A4I: Kind of
help needed but not received during
the past 12 months:
Information about mental illness and
its treatments
Information on availability of
services
Medication
Therapy or counselling
Help with financial problems
Help with housing problems
Help with personal relationships
Help with employment status or
work situation
26
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
SERB_A5A- SERB_A5M: Reasons
did not receive help:
Preferred to manage oneself
Didn't think anything more could
help
Didn't know how or where to get
more help
Afraid to ask for more help or of
what other would think
Couldn’t afford to pay
Problems with things like
transportation, childcare or
scheduling
Professional help was not available
in the area
Professional help was not available
at time required (e.g. doctor on
holidays, inconvenient hours)
Waiting time too long
Didn’t get around to it / didn’t bother
Language problems
Personal or family responsibilities
Total amount spent on services and
products used to help with problems
with emotions, mental health or use
of alcohol or drugs during the past 12
months
19
MED
Medication Use NPHS
Cycle 4
CCHS 1.1
Past 12 month use of medication
groups (sleep, diet pills, anxiety,
mood stabilizers, anti-depressants,
psychotic, behaviours, and
stimulants).
Dropped from NPHS/CCHS:
Use of medication during past
month:
Use of the following medication
groups (med groups) during past
month:
Pain relievers
For each medication group, the list of
examples of medications are stated in
brackets which means they are not
mandatory to be read by the
interviewer but are there to clarify
and provide examples of the
medication group in question (i.e. not
read automatically).
27
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Pain relievers
Codeine, Demerol
Allergy medicine
Asthma medications
Cough or cold remedies
Penicillin or other antibiotics
Medicine for the heart
Diuretics or water pills
Steroids
Insulin
Pills to control diabetes
Stomach remedies
Laxatives
Birth control pills
Hormones (type and date of start of
hormone therapy)
Thyroid medication
Use of other health products, past 2
days
Kept from NPHS:
Use of medication past 2 days
Exact name of medication
Was it prescribed by a medical
doctor or dentist
Tranquilizers
Diet pills
Anti-depressants
Codeine, Demerol
Allergy medicine
Asthma medications
Cough or cold remedies
Penicillin or other antibiotics
Medicine for the heart
Diuretics or water pills
Steroids
Insulin
Pills to control diabetes
Sleeping pills
Stomach remedies
Laxatives
Birth control pills
Hormones (type and date of start of
hormone therapy)
Thyroid medication
Other medication
Added:
MEDB_n1- MEDB_n4J: For
each medication group:
Medication taken under the
supervision of a health professional
Type of professional who prescribed
the medication
Past 12 months misuse of medication
(forgot to take or took less than
supposed to)
Reason why took less medication
MEDB_2: Past 2 days medication
use:
Total number of medications taken.
MEDBF3n: Exact name of
prescribed medication (up to a
maximum of 12).
MEDB_3nA: was prescription from
a medical doctor or dentist.
MEDB_4: Insurance coverage for
prescribed medication.
MEDB_5: Past 12 months use of
other health products such as
vitamins and herbs for problems
related to emotions, alcohol or drug
use, energy, concentration, sleep or
ability to deal with stress.
MEDB_6: List of other health
products:
St. John’s Wort, Valerian,
Chamomile, Ginseng, Kava Kava /
Kava Root, Lavender,
Chasteberry/Chaste Tree Berries,
Black Cohosh, Ginkgo Biloba,
28
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
NeuRecover – DA
MEDB_7: Type of professional who
recommended the use of other health
products.
20
SSM
Social support CCHS 1.1
NPHS
Cycle 4
(MOS)
Type of support received (same as
cycle 1.1).
Availability of support.
Frequency of use of different kinds
of support.
Type of support received:
To help you if you were confined to
bed
To listen to you when you need to
talk
To give you advice about a crisis
To take you to the doctor
Shows you love and affection
To have a good time with
Give you information to help you
understand a situation
Confide in or talk to about yourself
or your problems
Who hugs you
To relax with
To prepare your meals
Whose advice you want
Gets your mind off things
Help with daily chores
Share your private worries
Turn to for suggestions
To do enjoyable things with
Understands your problems
Love you and make you feel wanted
Added:
For each domain of support used
(support for activities of daily living,
support through affection, social
interaction support and informational
support) follow-up questions were
asked:
SSMB_21A, SSMB_22A,
SSMB_23A, SSMB_24A: In past 12
months, did you receive this kind of
support?
SSMB_21B, SSMB_22B,
SSMB_23B, SSMB_24B: When you
needed it, how often in the past 12
months did you receive this kind of
support?
The follow-up questions were newly
developed for 1.2. Questions were
grouped to avoid repetition and
response burden following
complaints from interviewers and
respondents.
21
SDC
Demographics CCHS 1.1 Country of Birth
Ethnic or cultural origin
Current languages
Mother tongue
Cultural or racial background
Country of Birth
Ethnic or cultural origin
Current languages
Mother tongue
Cultural or racial background
Updated lists of:
SDEB_1: Country of Birth (Sri
Lanka)
SDEB_4A-SDEB_4S: Ethnic or
cultural origin (Black, Norwegian,
Welsh, Swedish, Aboriginal)
SDEB_5A- SDEB_5S: Current
Updated lists.
29
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
languages (Dutch, Hindi, Russian,
Tamil)
SDEB_6A- SDEB_6S: Mother
tongue (Dutch, Hindi, Russian,
Tamil)
22
SPV
Spiritual Values CCHS 1.1 Questions on the role of spirituality
in the respondent’s life:
Past 12 months religious services
attendance
Role of spirituality in life
Extent to which spirituality gives
meaning to life, strength to face
difficulties and help to understand
difficulties in life
Religious affiliation
Self-perception of religious identity
Questions on the role of spirituality
in the respondent’s life:
Past 12 months religious services
attendance
Role of spirituality in life
Self-perception of religious identity
Not included:
SPRA_3: Self perception of how
religious or spiritual the respondent
is.
Added:
SPVB_2: Extent to which thought
that spirituality helps find a meaning
to life
SPVB_3: Extent to which thought
that spirituality gives strength to face
everyday difficulties
SPVB_4: Extent to which thought
that spirituality helps understand the
difficulties of life
SPVB_5: Religion, if any
SPVB_7: Self-perception of religious
identity
Modified:
SPVB_QINT1, SPVB_1, SPVB_2,
SPVB_3, SPVB_4: changed
“spirituality” for “spiritual values”.
Qualitative testing revealed that
respondents associated very religious
connotation to “spirituality” whereas
“spiritual values” had a more general
meaning. To deal with this,
significant changes were made to the
module.
23
LBF
Labour Force
participation
CCHS 1.1 Current employment status.
Occupation.
Work pattern past 12 months.
Current employment status.
Occupation.
Work pattern past 12 months.
Not included:
LBFA_35: Restrictions on smoking
at workplace
30
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
Health reason – not currently
working; absent from work; not
looking for work.
LBFA_43: Respondent’s preferred
choice of number of hours worked
versus pay
LBFA_74: Weeks without work but
looking all in one period, in separate
periods or in 3 or more
Added:
LBFB_13A: Health reason - not
currently working
LBFB_41A: Health reason - absent
from work last week
LBFB_73A: Health reason - not
looking for work
The additional info asks the
respondent to specify if the reason is
due to:
Physical health
Emotional or mental health
(including stress)
Use of alcohol or drugs
24
WST
Work stress CCHS 1.1 Derived scale which measures
different aspects related to work
stress, including:
Decision latitude
Psychological job demand
Physical workload
Job insecurity
Social support at work
Job satisfaction
Derived scale which measures
different aspects related to work
stress, including:
Decision latitude
Psychological job demand
Physical workload
Job insecurity
Social support at work
Job satisfaction
Added:
Use of a booklet to display response
categories.
Reduce interview time and response
burden.
31
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Order Modules
Source/
past use
Summary Description
(CCHS 1.2)
Summary Description
(NPHS Cycle 4/CCHS 1.1)
Changes from past CCHS and
NPHS cycles
Reason(s) for change /
Additional Comments
25
INC
Income CCHS 1.1
Personal and household income.
NPHS
Cycle 4
Personal and household income. Both cycles are identical.
26
ADM
Administration CCHS 1.1 Permission to link information
collected to provincial health
information.
Permission to share information with
provincial health ministries / Institut
de la Statistique du Québec and
Health Canada.
Permission to link information
collected to provincial health
information.
Permission to share information with
provincial health ministries / Institut
de la Statistique du Québec and
Health Canada.
Modified:
Wording for permission to link/share
questions (ADM_Q01A, Q01B,
Q04A).
ADM_Q4A: shortened question for
data sharing (all provinces except
Québec).
ADM_Q4A (data sharing – Québec):
share with “provincial ministries of
health, Institut de la Statistique du
Québec and Health Canada.
Deleted:
Data sharing question specific to
NWT, Yukon, Nunavut.
Frame evaluation questions (FE_Q1
– FE_Q3A).
The wording for the share and
linking questions are very important
and are guided in part by privacy
guidelines.
Survey is not asked to people living
in the territories.
27 Exit CCHS 1.1 Standard module. No change.
32
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
5. Sample Design
5.1 Target Population
CCHS Cycle 1.2 targets persons aged 15 years or older who are living in private
dwellings in the 10 provinces. Residents of the three territories, persons living
on Indian Reserves or Crown lands, clientele of institutions, full-time members
of the Canadian Armed Forces and residents of certain remote regions are
excluded from this survey. The Cycle 1.2 covered approximately 98% of the
population aged
15 or older in the 10 provinces.
5.2 Sample Size and Allocation
To provide reliable estimates at the provincial level, and given the budget
allocated to the Cycle 1.2 component, a sample of 30,000 respondents was
desired. Because provinces vary greatly in population size and reliable estimates
are required both at national and provincial levels, the sample was allocated
among provinces proportionally to the square root of the estimated population in
each province. Table 5.1 gives the targeted provincial sample sizes.
Table 5.1: Targeted sample sizes by province
Province Total sample size
(targeted)
Newfoundland/Labrador 1,525
Prince Edward Island
*
1,000
Nova Scotia 1,960
New Brunswick 1,765
Quebec 5,485
Ontario 6,720
Manitoba 2,165
Saskatchewan 2,045
Alberta 3,370
British Columbia 3,965
Canada 30,000
* The minimum sample size for a province was set to 1,000.
33
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Moreover and in order to have a good urban and rural representation in each
province, the sample was subsequently allocated to two strata: urban and rural.
The provincial sample was proportionally allocated to the urban and rural strata
to the number of dwelling in each stratum. Then sample sizes were enlarged
before data collection to take into account out-of-scope and vacant dwellings
and anticipated non-response.
5.3 Sample Buy-ins
Prior to the start of the data collection, the provinces of Ontario and Nova Scotia
had provided extra funds so that a larger sample of dwellings could be selected.
The purpose of those buy-ins were to get sufficient sample size in order to
provide reliable estimates for sub-provincial areas. Ontario added 7,702 sample
units while Nova Scotia added 790 units.
The province of Ontario was divided in seven regions. The allocation of the
14,422 sample units among the sub-provincial areas was performed using the
root-N approach while respecting a minimum sample size of 1,842 for each
region. This well-known allocation scheme balances the reliability requirements
at provincial and sub-provincial levels.
The province of Nova Scotia was divided into six regions (in fact, they were the
same as for the CCHS Cycle 1.1). The provincial sample of 2,750 units was
equally allocated to the regions in order to have similar levels of reliability in the
estimates for every region. Thus and except for allocating 525 sample units to
the region “Zone 6”, each region was allocated 445 sample units. Table 5.2 gives
the sample allocation by sub-provincial area for the provinces of Ontario and
Nova Scotia.
34
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table 5.2 Allocation by region for Ontario and Nova Scotia with buy-in
samples
Sub-provincial Area Sample Size
Ontario
14,422
Southwest 1,895
Central South 1,842
Central West 2,211
Central East 2,105
Toronto 2,632
East 1,895
North 1,842
Nova Scotia
2,750
Zone 1 445
Zone 2 445
Zone 3 445
Zone 4 445
Zone 5 445
Zone 6 525
5.4 Sampling of Households from the Area Frame
The Cycle 1.2 used the area frame designed for the Canadian Labour Force
Survey (LFS) as its frame. The sampling plan of the LFS is a multistage
stratified cluster design in which the dwelling is the final sampling unit.
1
In the
first stage, homogeneous strata were formed and independent samples of clusters
were drawn from each stratum. In the second stage dwelling lists were prepared
for each cluster and dwellings, or households, were selected from the lists.
For the purpose of the plan, each province is divided into three types of regions:
major urban centres, cities and rural regions. Geographic or socio-economic
strata are created within each major urban centre. Within the strata, between 150
and 250 dwellings are regrouped to create clusters. Some urban centres have
separate strata for apartments or for census enumeration areas (EA) in which the
average household income is high. In each stratum, six clusters or residential
buildings (sometimes 12 or 18 apartments) are chosen by a random sampling
method with a probability proportional to size (PPS), the size of which
corresponds to the number of households. The number six was used throughout
1
Statistics Canada (1998), Methodology of the Canadian Labour Force Survey, Statistics Canada,
Catalogue No. 71-526-XPB.
35
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
the sample design to allow a one-sixth rotation of the sample every month for
the LFS.
The other cities and rural regions of each province are stratified first on a
geographical basis, then according to socio-economic characteristics. In the
majority of strata, six clusters (usually census EAs) are selected using the PPS
method. Where there is low population density, a three-step plan is used
whereby two or three primary sampling units (PSU), which normally correspond
to groups of EAs, are selected and dividing each PSU into clusters, six of which
are sampled. The selection is made at each step using the PPS method.
Once the new clusters are listed, the sample is obtained using a systematic
sampling of dwellings. Table 5.3 gives an overview of the types of PSUs used
for the entire LFS sample. The yield is the number of households selected
within the framework of the LFS for a given month. As the sampling rates are
determined in advance, there is frequently a difference between the expected
sample size and the numbers that are obtained. The yield of the sample, for
example, is sometimes excessive. This especially happens in sectors where there
is an increase in the number of dwellings due to new construction, for example.
To reduce the cost of collection, an excessive output is corrected by eliminating,
from the beginning, a part of the units selected and by modifying the weight of
the sample design. Such an operation, usually conducted at an aggregate level, is
called sample stabilization. Moreover, one increases the required size of the
sample by households to account for dwellings, experience having shown that
12% of all dwellings are not occupied by households that are part of the field of
observation (certain dwellings are vacant or occupied seasonally, others are
occupied by households that are not targeted by the survey).
Table 5.3 Major first-stage units, sizes and yields
Area
Primary
Sampling Unit
(PSU)
Size
(households
per PSU)
Yields
(sampled
households)
Toronto, Montréal, Vancouver
Cluster 200-250 6
Other cities
Cluster 150-200 8
Apartment frame
Apartment Varies 5
Most rural areas / small urban
centres
Enumeration
area
300 10
36
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Requirements specific to Cycle 1.2 led to some modifications to this sampling
strategy. To get a base sample of 38,492 respondents for Cycle 1.2, 54,000
dwellings must be selected from the area frame (to account for vacant dwellings
and non-responding households). On an on-going monthly basis the LFS design
provides approximately 68,000 dwellings distributed across the various
economic regions in the 10 provinces whereas the Cycle 1.2 required a total of
54,000 dwellings distributed in the urban and rural stratum of all provinces or
regions (for Ontario and Nova Scotia), which have different geographic
boundaries from those of the LFS economic regions. Overall, the Cycle 1.2
required a lower number of dwellings than those generated by the LFS selection
mechanism, or an adjustment factor of 0.8 (54,000/68,000). At the urban/rural
level in provinces or regions, however, the adjustment factors varied from 0.5 to
3.0, which required certain adjustments.
The changes made to the selection mechanism in regions varied depending on
the size of the adjustment factors. For regions that had a factor smaller than or
equal to 1, the number of PSUs selected was reduced if necessary. For example,
if the factor was 0.5 then only 3 PSUs were selected in each stratum instead of
the usual number of 6 PSUs. For those with a factor greater than 1, the sampling
process of dwellings within a PSU was repeated for a subset of the selected
PSUs that were part of the same region. For example, if the factor was 1.6 then
the selection of dwellings within a PSU was repeated for 4 of the 6 PSUs in all
strata in that region. When a repeated selection of dwelling within a PSU was
necessary and no more dwellings were available in that PSU, then another PSU
was selected. Where the chosen approach created an unnecessary surplus of
dwellings, stabilisation was performed.
5.5 Sampling of Respondents
Selection of individual respondents was designed to ensure over-representation
of young persons (aged 15 to 24) and seniors (65 or older). The selection
strategy was designed to consider user needs, cost, design efficiency, response
burden and operational constraints
2
. One person aged 15 or older was randomly
selected from the sampled households. The probability of selection for each
person in a household was defined as a function of the household composition.
The Table 5.4 describes the rule for selecting a person within sampled
households.
2
Béland, Y., Dufour, J. and Gravel, R. (2001), “Sample Design of the Canadian Mental Health Survey”,
2001 Proceedings of the Survey Methods Section, Vancouver: Statistical Society of Canada, 93-98.
37
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table 5.4: Selection Strategy based on Household Composition
Number of persons aged 25 or over (m)
Number of
15-24 year
olds
0 1 2 3 4 5+
0 - A A A A A
1 A B B B B A
2 A B B A A A
3+ A A A A A A
A: Selection of one person with equal probability.
B: Selection of one person where those in the 15-to-24 age group would have
a probability of 2.6/(2.6*n+m) of being selected and the others would have
a probability of 1/(2.6*n+m), where n is the number of persons between
15 and 24 and m is the number of persons aged 25 or over.
5.6 Sample Allocation over the Collection Period
In order to balance interviewer workload, the initial sample of dwellings was
equally allocated at random, within each region, over the three collection periods
covering seven months (May to November 2002). The first collection period
covered three months (Q1: May to July 2002) while the other two collection
periods covered two months each (Q2: August and September 2002, and Q3:
October and November 2002). More time was allowed for the first collection
period compared to the other two periods in order to give a chance to the
interviewers to familiarise themselves with the survey. For operational
constraints, all dwellings of a PSU were assigned to the same collection period.
It is also important to mention that data collection continued until the end of
December in order to improve response rates.
38
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
6. Data Collection
6.1 Questionnaire Design and Data Collection Method
The Cycle 1.2 questionnaire was administered using computer-assisted
interviewing (CAI). Sample units selected from the area frame were interviewed
using the Computer-Assisted Personal Interviewing (CAPI) method.
CAI offers a number of data quality advantages over other collection methods.
First, question text, including reference periods and pronouns, is customised
automatically based on factors such as the age and sex of the respondent, the
date of the interview and answers to previous questions.
Second, edits to check for inconsistent answers or out-of-range responses are
applied automatically and on-screen prompts are shown when an invalid entry is
recorded. Immediate feedback is given to the respondent and the interviewer is
able to correct any inconsistencies.
Third, questions that are not applicable to the respondent are skipped
automatically.
6.2 Supervision and Control
CAPI interviewers worked independently from their homes using laptop
computers and were supervised from a distance by senior interviewers.
Completed interviews were transmitted daily to Statistics Canada’s head office
using a secure telephone transmission directly from the interviewer’s home.
6.3 Field Tests
A CAPI field test was conducted in February 2002. The test was conducted in
Alberta and Quebec using a sample of 600 units.
The main objectives of the CAPI test were to evaluate respondent reaction to the
questions and to obtain estimates of completion times for the various sections of
the questionnaire. Field operations procedures, interviewer training and the
computer-assisted interviewing application were also tested.
39
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
6.4 Interviewing
In all selected dwellings, a knowledgeable household member was asked to
supply basic demographic information on all residents of the dwelling.
Depending on the composition of the household, one member aged 15 or over
was then selected for a more in-depth interview.
CAPI interviewers were trained to make an initial personal contact with each
sampled dwelling. In cases where this initial visit resulted in non-response,
telephone follow-ups were permitted.
Every effort was made to conduct the interviews face to face. Collection by
telephone was authorized only when travel was prohibitive or the respondent
refused to conduct the interview in person. As well, household contact and
selection of a respondent was allowed by telephone, after an initial contact was
attempted in person.
In the end, 14% of cases nationally were completed by telephone (slightly higher
in Ontario). No proxy interviews were permitted for this survey.
6.5 Minimising Non-response
Prior to the first contact by an interviewer, an introductory letter and brochure
were delivered to each selected dwelling for which we had a valid mailing
address. These explained the importance of the survey and provided examples of
how Cycle 1.2 data would be used.
Interviewers were instructed to make all reasonable attempts to obtain Cycle 1.2
interviews. When the timing of the interviewer's visit was inconvenient, an
appointment was made to call back at a more convenient time. If no one was
home, numerous call-backs were made. For individuals who at first refused to
participate in the survey, a letter was sent from the Regional Office to the
respondent, stressing the importance of the survey and the household's
collaboration. This was followed by a second call (or visit) from a senior
interviewer, a project supervisor or another interviewer to try to convince
respondents of the importance of participating in the survey. During the final
months of data collection, non-response cases and selected persons who had
previously refused were again approached and encouraged to participate in the
survey. This diligence in contact may have resulted in stronger survey results
by maximising the response rate. To help minimize non-response, some non-
response cases were followed-up one more time in November. Refusals were
excluded from these.
40
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
To reduce the impact of language as a barrier to conducting interviews, each of
the Statistics Canada Regional Offices has recruited interviewers with a wide
range of language competencies. To help these interviewers, an ‘official’
translation of key terms was created in Chinese and Punjabi, the two most
prevalent non-official languages from CCHS Cycle 1.1. Interviewers able to
speak those languages used the guide to translate questions and complete the
interview. Interviewers were restricted from conducting interviews in any other
language because of the complexity of the question concepts.
6.6 Special Circumstances during Cycle 1.2 Collection Operations
Data collection took place between May 2002 and December 2002, a period of
seven months. This plan was carefully designed to ensure that the survey’s
quality objectives would be met. The final month of collection was planned to
provide interviewers with an opportunity for a final attempt to convert non-
responding cases.
For most of Statistics Canada’s household surveys, collection operations
proceed smoothly and within the established parameters. For Cycle 1.2, the total
workload imposed by the lengthy interview, complex content and difficult
respondent burden in some cases, proved to be a challenge for the data collection
infrastructure in place. To ensure the success of collection operations, a number
of strategies were put into place. Among these were, specialized training on
mental illness and how to conduct difficult interviews for interviewers, careful
planning of collection periods and interviewer assignment sizes, addition of staff
in key areas, and the decision to allow limited data collection by telephone.
To ensure that data quality was maintained during collection, a monitoring
system was put in place. Various aspects related to the interview process were
monitored at the interviewer level such as average interview time and item non-
response. Regular weekly feedback from Head Office to the Regional Offices
helped maintain and correct problems as they occurred. A validation process
was also put in place in the field to monitor the quality of the work performed by
the interviewers. At the end of data collection, a national response rate of 77%
was achieved. The reader will find complete details regarding the response rates
in Section 9.
41
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
7. Data Processing
7.1 Editing
Most editing of the data was performed at the time of the interview by the
computer-assisted interviewing (CAI) application. It was not possible for
interviewers to enter out-of-range values and flow errors were controlled
through programmed skip patterns. For example, CAI ensured that questions
that did not apply to the respondent were not asked.
In response to some types of inconsistent or unusual reporting, warning
messages were invoked but no corrective action was taken at the time of the
interview. Where appropriate, edits were instead developed to be performed
after data collection at Head Office. Inconsistencies were usually corrected by
setting one or both of the variables in question to "not stated".
7.2 Coding
Pre-coded answer categories were supplied for all suitable variables. Several
questions in the Cycle 1.2 questionnaire allow write-in responses. For some of
these questions, write-in responses were either coded into one of the existing
listed categories if the write-in information duplicated a listed category or into
new unique categories. Medication questions were coded to the Anatomical
Therapeutic Classification (ATC).
7.3 Creation of Derived and Grouped Variables
To facilitate data analysis, a number of variables on the file have been derived
using items found on the Cycle 1.2 questionnaire. Derived variables generally
have a "D" or "G" in the fifth character of the variable name. In some cases, the
derived variables are straightforward, involving collapsing of response
categories. In other cases, several variables have been combined to create a new
variable. Appendix C provides details on how these more complex variables
were derived.
42
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
7.4 Imputation
Due to some technical problems in certain skip patterns of the suicide module,
some respondents were not asked the questions required for the calculation of
the derived variables ‘12-month suicidal thought’ and ‘12-month suicidal
attempt’. Consequently, important information was missing for those individuals
(this represented 4.83% of all respondents for the ‘12-month suicidal thought’
and 1.03% of all respondents for the ‘12-month suicidal attempt’). Moreover
and because of their profiles, those individuals are more likely to have had a 12-
month suicidal thought and/or a 12-month suicidal attempt which would have
resulted in an underestimation of the prevalence. To fill in these missing
responses, values were imputed using the approach described below.
Two methods of imputation were used, a deterministic method and one based on
a logistic regression model. As it was possible to derive directly the missing
value based on other responses for some respondents, a deterministic imputation
method was first used. This was the case for all missing values for the 12-month
suicidal attempt and for about one fourth of the missing values for the 12-month
suicidal thought. For the remaining missing values of the 12-month suicidal
thought, a logistic regression imputation method was used. The method
consisted in fitting a logistic regression model between the variable to impute
(the 12–month suicidal thought) and correlated characteristics using respondents
without missing values who were similar to those to impute. Using the fitted
model, a probability of response (yes or no) was calculated for each respondent
who needed imputation; a response was then imputed based on that probability.
7.5 Weighting
The principle behind estimation in a probability sample such as Cycle 1.2 is that
each person in the sample "represents", besides himself or herself, several other
persons not in the sample. For example, in a simple random 2% sample of the
population, each person in the sample represents 50 persons in the population. In
the terminology used here, it can be said that each person has a weight of 50.
The weighting phase is a step that calculates, for each person, his or her
associated sampling weight. This weight appears on the microdata file, and
must be used to derive meaningful estimates from the survey. For example, if
the number of individuals who had a major depressive episode is to be
estimated, it is done by selecting the records referring to those individuals in the
sample having that characteristic and summing the weights entered on those
records. Details of the method used to calculate sampling weights are presented
in Section 8.
43
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
7.6 Suppression of Confidential Information
It should be noted that the 'Public Use' microdata file described above differs in
a number of important respects from the survey 'master' file held by Statistics
Canada. These differences are the result of actions taken to protect the
anonymity of individual survey respondents. Protection of respondents is
assured through suppression of individual values, variable grouping, and
variable capping. Users requiring access to information excluded from the
microdata files have three options: to purchase custom tabulations, use one of
the Research Data Centres
3
, or use the remote access option. (See sub-section
12.3)
3
The most current information about the Research Data Centres can be found at www.statcan.ca
44
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
8. Weighting
In order for the estimates produced from survey data to be representative of the
covered population and not just the sample itself, a user must incorporate the survey
weights into their calculations. A survey weight is given to each person included in
the final sample, that is, the sample of persons having answered the survey. This
weight corresponds to the number of persons represented by the respondent for the
entire population.
As described in Section 5, Cycle 1.2 had recourse to the area frame of the LFS. For
this reason, the weighting strategy of the CCHS is very similar to the LFS one. Table
8.1 presents an overview of the different adjustments, part of the weighting strategy,
in the order in which they are applied.
Table 8.1: List of adjustments in the weighting
0 Initial weight
1 Sample increase or decrease
2 Stabilization
3 Removal of out-of-scope units
4 Household non-response
5 Creation of person level weight
6 Person non-response
7 Poststratification
These adjustments will be explained in the following section.
8.1 Adjustments Applied to the Initial Weight
Adjustment 0 – Initial weight
Since the mechanism established for the LFS was used to select the area frame
sample, the initial weights had to be computed with respect to that mechanism.
First, within each stratum defined by the LFS, clusters (primary units) are selected
with probabilities proportional to population sizes (based on 1991 Census counts).
Next, dwellings are sampled within each selected cluster using systematic
sampling. The product of the probabilities for each of these selections represents
the overall probability of selection, and the inverse of that probability is used as
the initial weight of the unit. For more details about the selection mechanism, as
45
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
well as a more complete definition of strata and clusters, refer to Statistics Canada
(1998).
4
Adjustment 1 – Sample increase or decrease
Some modifications were made to the default LFS mechanism at the time of
sample selection for Cycle 1.2. The LFS design provides approximately 68,000
dwellings nationally, which is very close to the needs of Cycle 1.2. However, at
the provincial level or regional level (in Ontario and Nova-Scotia), the number of
dwellings provided by the sample design of the LFS can either exceed the needs
of Cycle 1.2 or be insufficient. Modifications made in order to obtain the needed
sample within a province or a Health Region (HR) consisted, in summary, of
repeating the sampling process of dwellings within a few clusters in the provinces
or HRs where Cycle 1.2 needed more dwellings, or not selecting any dwelling in a
few clusters when it needed less dwellings. This modification had the effect of
increasing or decreasing the sample and had to be accounted for in the weighting
to correctly represent the probability of selection. An adjustment factor
representing the sample increase or decrease rate at the provincial level or
regional level (in Ontario and Nova-Scotia) was calculated. The initial weight was
multiplied by this adjustment factor, which results in weight 1.
Adjustment 2 – Stabilization
In the HRs or provinces where Cycle 1.2 needed more dwellings, increasing the
sample as described in the previous paragraph resulted in a significantly larger
sample than necessary. Stabilization was therefore instituted to bring the sample
size back down to the desired level. The stabilization process consisted of
randomly subsampling dwellings at the provincial level or at the HR level in
Ontario and Nova-Scotia. An adjustment factor representing the effect of this
stabilization was therefore calculated to adjust the probability of selection
appropriately. This factor, multiplied by the weight 1, produces weight 2.
Adjustment 3 – Removal of out-of-scope units
Among all dwellings sampled, a certain proportion are identified during collection
as being out-of-scope. Dwellings that are demolished or in construction, vacant,
seasonal or secondary dwellings, and institutions are examples of out-of-scope
cases for Cycle 1.2. Records for these dwellings were simply removed from the
sample, leaving only in-scope dwellings. They kept the same weight as in the
previous step, which is now called weight 3.
4
Statistics Canada (1998), Methodology of the Canadian Labour Force Survey, Statistics Canada, Cat. No.
71-526-XPB.
46
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Adjustment 4 – Household non-response
During collection, a certain proportion of interviewed households inevitably
resulted in non-response. This usually occurs when a household refuses to
participate in the survey, provides unusable data, or cannot be reached for an
interview. Weights of non-responding households were distributed to
respondents using response propensity classes. The CHAID (Chi-Square
Automatic Interaction Detector) algorithm available in Knowledge Seeker,
5
was
used to identify which characteristics best split the sample into groups that were
dissimilar with respect to response/non-response. Note that groups were formed
independently within each province or HR in the case of Ontario and Nova-
Scotia. Since the information available for non-respondents is limited, only
characteristics such as, collection period (with 3 periods; May 2002 to July 2002,
August 2002 to September 2002 and October 2002 to December 2002) and a
rural/urban indicator could be used in the creation of the classes. In a few
provinces or HRs, both variables were significant in the creation of classes. In a
few others, only one variable was significant, and in the remaining, none was
significant. An adjustment factor was therefore calculated within each class as
follows:
householdsrespondingallforweightofSum
householdsallforweightofSum
3
3
Weight 3 for responding households was multiplied by this factor to produce
weight 4. Non-responding households were dropped out of the process at this
point.
Adjustment 5 – Creation of person level weight
Since the ultimate sampling unit for the CCHS is a person, the household level
weights computed up to this point need to be converted down to the person level.
This weight is obtained by multiplying the weight 4 by the inverse of the
probability of selection of the person selected in the household. This gives the
weight 5. As mentioned before, in the households with a certain number of
persons in the age groups 15 to 24 and 25 and over, this probability of selection is
larger for the persons in the age group 15 to 24 (see Section 5.4 for more details
on the algorithm of the selection of the person). For the other households, this
probability is equal to the inverse of the number of persons aged 15 and over in
the household, no matter which person is selected.
5
ANGOSS Software (1995), Knowledge Seeker IV for Windows - User's Guide, ANGOSS Software
International Limited.
47
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Adjustment 6 – Person non-response
A Cycle 1.2 interview can be seen as a two-part process. First the interviewer
gets the complete roster of the people living within the responding household.
Second, (s)he interviews the selected person within the household. In some cases,
interviewers can only get through the first part, either because they cannot get in
touch with the selected person, or because that selected person refuses to be
interviewed. Such cases are defined as person non-response and an adjustment
factor must be applied to the weights of respondents to overcome this non-
response. As for the treatment of household non-response, the adjustment was
applied within classes based on characteristics available for both respondents and
non-respondents. All characteristics collected when rostering all household
members were in fact available for the creation of the classes. The CHAID
algorithm was once again used to define the classes and the final result presented
definitions that varied from one province to another (or HR in the case of Ontario
and Nova-Scotia). Depending on the province or HR, the following
characteristics were used to form the adjustment classes: sex, age group,
urban/rural indicator, education, marital status and the size of the household. As a
result, an adjustment factor is calculated as follows:
personsselectedrespondingallfor5weightofSum
personsselectedallfor5weightofSum
Weight 5 of responding persons was therefore multiplied by this adjustment factor
to produce weight 6. Non-responding persons are dropped out of the weighting
process from this point onward.
Adjustment 7 – Poststratification
The final step necessary to obtain the final Cycle 1.2 weight was the
poststratification. Poststratification is done to ensure that the sum of the final
weights corresponds to the population estimates defined at the provincial level
and regional level in Ontario and Nova-Scotia, for all eight age-sex groups of
interest, that is, the four age groups 15 to 24, 25 to 44, 45 to 64, 65 and over, for
both males and females. The population estimates for 2002 were based on the
1996 Census counts and estimates of birth, death, immigration and emigration
counts. The average of these monthly estimates for each of the province (or HR)-
age-sex poststrata was used to poststratify. The weight 6 was therefore adjusted
to obtain the final weight 7 with the help of the adjustment factor defined as
follows:
48
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
respondent the of group sex-age-HR) (or province the for 6 weightsof Sum
respondent the ofgroupsex-age-HR) (or province the for estimate Population
Consequently, the weight 7 corresponds to the final Cycle 1.2 weight that can be
found on the data file with the variable name WTSB_M.
49
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
9. Data Quality
9.1 Response Rates
In total and after removing the out-of-scope units, 48,047 households were
selected to participate in Cycle 1.2. Out of these selected households a response
was obtained for 41,560 which results in an overall household-level response rate
of 86.5%. Among these responding households 41,559 individuals (one per
household) were selected to participate in Cycle 1.2 out of which a response was
obtained for 36,984 which results in an overall person-level response rate of
89.0%. At the Canada level, this would yield a combined response rate of 77.0%
for Cycle 1.2. Table 9.1 gives combined response rates as well as relevant
information for calculation of them by province and for seven Ontario regions.
Next we describe how the various components of the equation should be handled
to correctly compute combined response rates.
Household-level response rate
HHRR =
number of responding households
all in-scope households
Person-level response rate
PPRR =
number of responding persons
all selected persons
Combined response rate = HHRR x PPRR
Next is an example on how to calculate the combined response rate for Canada
using the information found in Table 9.1.
HHRR =
41,559 = 0.865
48,047
PPRR =
36,984 = 0.890
41,559
Combined response rate = 0.865 x 0.890
= 0.770
= 77.0%
50
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table 9.1 Combined Response Rates by Province
Household Level Person Level Combined
Province
Health
Region
# In
Scope
Hh
# Resp
Hh
Hh
Resp.
Rates
# Pers.
Select. # Resp.
Pers.
Resp.
Rates
Resp.
Rates
Canada Total 48,046 41,559 86.5 41,559 36,984 89.0 77.0
Newfoundland Total 1,898 1,750 92.2 1,750 1,562 89.3 82.3
Prince Edward Island Total 1,227 1,098 89.5 1,098 1,002 91.3 81.7
Nova Scotia Total 3,476 3,131 90.1 3,131 2,785 88.9 80.1
New Brunswick Total 2,185 1,942 88.9 1,942 1,706 87.8 78.1
Quebec Total 6,829 5,961 87.3 5,961 5,332 89.4 78.1
Ontario Total 17,956 15,038 83.7 15,038 13,184 87.7 73.4
South West 2,307 1,977 85.7 1,977 1,782 90.1 77.2
Central South 2,302 1,940 84.3 1,940 1,675 86.3 72.8
Central West 2,747 2,304 83.9 2,304 2,035 88.3 74.1
Central East 2,622 2,258 86.1 2,258 1,947 86.2 74.3
Toronto 3,378 2,582 76.4 2,582 2,142 83.0 63.4
East 2,298 1,963 85.4 1,963 1,783 90.8 77.6
North 2,302 2,014 87.5 2,014 1,820 90.4 79.1
Manitoba Total 2,705 2,403 88.8 2,403 2,230 92.8 82.4
Saskatchewan Total 2,553 2,245 87.9 2,245 2,045 91.1 80.1
Alberta Total 4,198 3,629 86.4 3,629 3,236 89.2 77.1
British Columbia Total 5,019 4,362 86.9 4,362 3,902 89.5 77.7
51
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
9.2 Survey Errors
The estimates derived from this survey are based on a sample of individuals.
Somewhat different figures might have been obtained if a complete census had
been taken using the same questionnaire, interviewers, supervisors, processing
methods, etc. as those actually used. The difference between the estimates
obtained from the sample and the results from a complete count under similar
conditions is called the
sampling error of the estimate.
Errors which are not related to sampling may occur at almost every phase of a
survey operation. Interviewers may misunderstand instructions, respondents may
make errors in answering questions, the answers may be incorrectly entered on
the computer and errors may be introduced in the processing and tabulation of the
data. These are all examples of
non-sampling errors.
9.2.1 Non-sampling Errors
Over a large number of observations, randomly occurring errors will have
little effect on estimates derived from the survey. However, errors occurring
systematically will contribute to biases in the survey estimates. Considerable
time and effort was made to reduce non-sampling errors in the survey.
Quality assurance measures were implemented at each step of data
collection and processing to monitor the quality of the data. These measures
included the use of highly skilled interviewers, extensive training with
respect to the survey procedures and questionnaire, and the observation of
interviewers to detect problems. Testing of the Computer Assisted
Interview application and field tests were also essential procedures to ensure
that data collection errors were minimized.
A major source of non-sampling errors in surveys is the effect of non-
response on the survey results. The extent of non-response varies from
partial non-response (failure to answer just one or some questions) to total
non-response. Partial non-response to Cycle 1.2 was minimal; once the
questionnaire was started, it tended to be completed with very little non-
response. Total non-response occurred either because a respondent refused
to participate in the survey, or because the interviewer was unable to contact
the selected respondent. Total non-response was handled by adjusting the
weight of persons who responded to the survey to compensate for those who
did not respond. See section 8 for details of the weight adjustment for non-
response.
52
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
9.2.2 Sampling Errors
Since it is an unavoidable fact that estimates from a sample survey are
subject to sampling error, sound statistical practice calls for researchers to
provide users with some indication of the magnitude of this sampling error.
The basis for measuring the potential size of sampling errors is the standard
deviation of the estimates derived from survey results. However, because of
the large variety of estimates that can be produced from a survey, the
standard deviation of an estimate is usually expressed relative to the
estimate to which it pertains. This resulting measure, known as the
coefficient of variation (CV) of an estimate, is obtained by dividing the
standard deviation of the estimate by the estimate itself and is expressed as a
percentage of the estimate.
For example, suppose hypothetically that one estimates that 12% of
Canadians aged 15 and over have had at least one major depressive episode
during their life and that this estimate is found to have a standard deviation
of .007. Then the CV of the estimate is calculated as:
(0.007/0.12) x 100% = 5.83%.
Statistics Canada commonly uses CV results when analyzing data, and urges
users producing estimates from Cycle 1.2 data files to also do so. For details
on how to determine CVs, see Section 11. For guidelines on how to interpret
CV results, see the table at the end of sub-section 10.4.
53
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
10. Guidelines for Tabulation, Analysis and Release
This section of the documentation outlines the guidelines to be adhered to by users
tabulating, analyzing, publishing or otherwise releasing any data derived from the
survey microdata file. With the aid of these guidelines, users of microdata should be
able to produce figures that are in close agreement with those produced by Statistics
Canada and, at the same time, will be able to develop currently unpublished figures
in a manner consistent with these established guidelines.
10.1 Rounding Guidelines
In order that estimates for publication or other release derived from this
microdata file correspond to those produced by Statistics Canada, users are
urged to adhere to the following guidelines regarding the rounding of such
estimates:
a) Estimates in the main body of a statistical table are to be rounded to the
nearest hundred units using the normal rounding technique. In normal
rounding, if the first or only digit to be dropped is 0 to 4, the last digit to be
retained is not changed. If the first or only digit to be dropped is 5 to 9, the
last digit to be retained is raised by one. For example, in normal rounding to
the nearest 100, if the last two digits are between 00 and 49, they are changed
to 00 and the preceding digit (the hundreds digit) is left unchanged. If the
last digits are between 50 and 99 they are changed to 00 and the proceeding
digit is incremented by 1;
b) Marginal sub-totals and totals in statistical tables are to be derived from their
corresponding unrounded components and then are to be rounded themselves
to the nearest 100 units using normal rounding;
c) Averages, proportions, rates and percentages are to be computed from
unrounded components (i.e., numerators and/or denominators) and then are to
be rounded themselves to one decimal using normal rounding. In normal
rounding to a single digit, if the final or only digit to be dropped is 0 to 4, the
last digit to be retained is not changed. If the first or only digit to be dropped
is 5 to 9, the last digit to be retained is increased by 1;
d) Sums and differences of aggregates (or ratios) are to be derived from their
corresponding unrounded components and then are to be rounded themselves
to the nearest 100 units (or the nearest one decimal) using normal rounding;
e) In instances where, due to technical or other limitations, a rounding technique
other than normal rounding is used resulting in estimates to be published or
54
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
otherwise released that differ from corresponding estimates published by
Statistics Canada, users are urged to note the reason for such differences in
the publication or release document(s);
f) Under no circumstances are unrounded estimates to be published or otherwise
released by users. Unrounded estimates imply greater precision than actually
exists.
10.2 Sample Weighting Guidelines for Tabulation
The sample design used for Cycle 1.2 was not self-weighting. That is to say, the
sampling weights are not identical for all individuals in the sample. When
producing simple estimates, including the production of ordinary statistical
tables, users must apply the proper sampling weight.
If proper weights are not used, the estimates derived from the microdata files
cannot be considered to be representative of the survey population, and will not
correspond to those produced by Statistics Canada.
Users should also note that some software packages might not allow the
generation of estimates that exactly match those available from Statistics
Canada, because of their treatment of the weight field.
10.2.1 Definitions: Categorical Estimates, Quantitative Estimates
Before discussing how Cycle 1.2 data can be tabulated and analyzed, it is
useful to describe the two main types of point estimates of population
characteristics that can be generated from the microdata file.
Categorical Estimates:
Categorical estimates are estimates of the number or percentage of the
surveyed population possessing certain characteristics or falling into some
defined category. The number of individuals who smoke daily is an
example of such an estimate. An estimate of the number of persons
possessing a certain characteristic may also be referred to as an estimate of
an aggregate.
55
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Example of Categorical Question:
SMKA_202: At the present do/does ... smoke cigarettes daily,
occasionally or not at all?
__ Daily
__ Occasionally
__ Not at all
Quantitative Estimates:
Quantitative estimates are estimates of totals or of means, medians and
other measures of central tendency of quantities based upon some or all of
the members of the surveyed population.
An example of a quantitative estimate is the average number of cigarettes
smoked per day by individuals who smoke daily. The numerator is an
estimate of the total number of cigarettes smoked per day by individuals
who smoke daily, and its denominator is an estimate of the number of
individuals who smoke daily.
Example of Quantitative Question:
SMKA_204: How many cigarettes do/does you/he/she smoke each day
now?
|_|_| Number of cigarettes
10.2.2 Tabulation of Categorical Estimates
Estimates of the number of people with a certain characteristic can be
obtained from the microdata files by summing the final weights of all
records possessing the characteristic of interest.
Proportions and ratios of the form
are obtained by:
Y / X
ˆˆ
a) summing the final weights of records having the characteristic of
interest for the numerator (
X
ˆ
);
b) summing the final weights of records having the characteristic of
interest for the denominator (
Y
ˆ
); then
c) dividing the numerator estimate by the denominator estimate.
56
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
10.2.3 Tabulation of Quantitative Estimates
Estimates of quantities can be obtained from the microdata files by:
a) multiplying the value of the variable of interest by the final weight
and summing this quantity over all records of interest to obtain the
numerator (
X
ˆ
);
b) summing the final weights of records having the characteristic of
interest for the denominator (
Y
ˆ
); then
c) dividing the numerator estimate by the denominator estimate.
For example, to obtain an estimate of the average number of cigarettes
smoked each day by individuals who smoke daily, multiply the value of
variable SMKA_204
6
by the weight, WTSB_M, then sum this value over
those records with a value of "daily" to the variable SMKA_202 to obtain
the numerator (
X
ˆ
). Sum the final weight of those records with a value of
"daily" to the variable SMKA_202 to obtain the denominator (
Y
ˆ
). Divide
(
X
ˆ
) by (
Y
ˆ
) to obtain the average number of cigarettes smoked each day
by daily smokers.
10.3 Guidelines for Statistical Analysis
Cycle 1.2 is based upon a complex design, with stratification and multiple stages
of selection, and unequal probabilities of selection of respondents. Using data
from such complex surveys presents problems to analysts because the survey
design and the selection probabilities affect the estimation and variance
calculation procedures that should be used.
While many analysis procedures found in statistical packages allow weights to
be used, the meaning or definition of the weight in these procedures can differ
from what is appropriate in a sample survey framework, with the result that
while in many cases the estimates produced by the packages are correct, the
variances that are calculated are almost meaningless.
6
See Section 12.2 for variable naming convention
57
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
For many analysis techniques (for example linear regression, logistic regression,
analysis of variance), a method exists that can make the application of standard
packages more meaningful. If the weights on the records are rescaled so that the
average weight is one (1), then the results produced by the standard packages
will be more reasonable; they still will not take into account the stratification
and clustering of the sample's design, but they will take into account the unequal
probabilities of selection. The rescaling can be accomplished by using in the
analysis a weight equal to the original weight divided by the average of the
original weights for the sampled units (people) contributing to the estimator in
question.
In order to provide a means of assessing the quality of tabulated estimates,
Statistics Canada has produced a set of Approximate Coefficients of Variations
Tables (commonly referred to as "CV Tables") for the CCHS. These tables can
be used to obtain approximate coefficients of variation for categorical-type
estimates and proportions. See Section 11 for more details.
58
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
10.4 Release Guidelines
Before releasing and/or publishing any estimate from the microdata file, users
should first determine the number of sampled respondents who contribute to the
calculation of the estimate. If this number is less than 30, the weighted estimate
should not be released regardless of the value of the coefficient of variation for
this estimate. For weighted estimates based on sample sizes of 30 or more, users
should determine the coefficient of variation of the
rounded estimate and follow
the guidelines below.
Table 10.1: Sampling Variability Guidelines
Type of Estimate CV (in %)
Guidelines
1. Acceptable 0.0 - 16.5 Estimates can be considered for general unrestricted release.
Requires no special notation.
2. Marginal 16.6 - 33.3 Estimates can be considered for general unrestricted release
but should be accompanied by a warning cautioning
subsequent users of the high sampling variability associated
with the estimates. Such estimates should be identified by the
letter E (or in some other similar fashion).
3. Unacceptable Greater than
33.3
Statistics Canada recommends not to release estimates of
unacceptable quality. However, if the user chooses to do so
then estimates should be flagged with the letter F (or in some
other fashion) and the following warning should accompany
the estimates:
“The user is advised that . . .(specify the data) . . . do not meet
Statistics Canada’s quality standards for this statistical
program. Conclusions based on these data will be unreliable
and most likely invalid. These data and any consequent
findings should not be published. If the user chooses to publish
these data or findings, then this disclaimer must be published
with the data.”
59
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
11. Approximate Sampling Variability Tables
In order to supply coefficients of variation that would be applicable to a wide variety
of categorical estimates produced from this microdata file and that could be readily
accessed by the user, a set of Approximate Sampling Variability Tables has been
produced. These "look-up" tables allow the user to obtain an approximate
coefficient of variation based on the size of the estimate calculated from the survey
data.
The coefficients of variation (CV) are derived using the variance formula for simple
random sampling and incorporating a factor which reflects the multi-stage, clustered
nature of the sample design. This factor, known as the design effect, was determined
by first calculating design effects for a wide range of characteristics and then
choosing, for each table produced, a conservative value among all design effects
relative to that table. The value chosen was then used to generate a table that applies
to the entire set of characteristics.
The design effects, sample sizes and population counts used to produce the
Approximate Sampling Variability Tables are presented in Appendix D. All
coefficients of variation in the Approximate Sampling Variability Tables are
approximate and, therefore, unofficial. Options concerning the computation of exact
coefficients of variation are discussed in sub-section 11.7.
Remember: As indicated in Sampling Variability Guidelines in Section 10.4, if the
number of observations on which an estimate is based is less than 30, the weighted
estimate should not be released regardless of the value of the coefficient of variation.
Coefficients of variation based on small sample sizes are too unpredictable to be
adequately represented in the tables.
11.1 How to Use the CV Tables for Categorical Estimates
The following rules should enable the user to determine the approximate
coefficients of variation from the Sampling Variability Tables for estimates of
the number, proportion or percentage of the surveyed population possessing a
certain characteristic and for ratios and differences between such estimates.
One should note that there are three types of CV tables and that the kind of table
to use depends on the population considered and on the form of the estimate
(number or proportion).
The tables of the first type basically have the same format as the CV tables that
are habitually published (for example, all tables from Cycle 1.1 of CCHS have
this form). These are the general tables. The appendix contains 22 CV tables of
60
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
this type: one for Canada, one for each province, one for each age group at the
Canada level, and one for each health region in Ontario.
The other tables must be used to calculate CVs for nine types of sub-populations
corresponding to mental disorders. Those tables are grouped by pairs. Each of
these pairs is composed of a table for the estimates that take the form of an
aggregate and a table for the estimates having the form of a proportion. They
will be called the disorders tables. There is one of these pairs for each general
CV table. Like the general tables, the disorders tables allow the user to
determine the CV of an estimate calculated at the Canada level, for a province,
for an age group at the Canada level, or for a health region.
The following is a list of the nine disorder domains and the way to identify the
individuals of each of these domains on the data file:
DISORDER
TIME
PERIOD
IDENTIFIER SAMPLE SIZE
Depression
(DEP)
Last 12
months
DEPBDDY=1 1,944
Depression
(DEP)
Lifetime DEPBDDPS=1 4,713
Mood
(MOOD)
Last 12
months
MPHBFYM=1 2,122
Mood
(MOOD)
Lifetime MPHBFLM=1 5,112
Anxiety
(ANX)
Last 12
months
MPHBFYA=1 1,803
Anxiety
(ANX)
Lifetime MPHBFLA=1 4,268
Substances
dependence
(SUBS)
Last 12
months
MPHBFYSA=1 1,215
At least one
disorder
(ALL)
Last 12
months
MPHBFY=1 4,134
At least one
disorder
(ALL)
Lifetime MPHBFL=1 7,585
Since there are three types of tables, certain rules will differ with the format.
Rules 1.1 to 1.5 apply to disorders tables and rules 2.1 to 2.5 must be followed
for the reading of general tables.
61
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
11.1.1 How to Use the disorder CV tables
The CV disorders tables were made in order to determine the CV of an
estimate calculated for one of the following populations:
All the people aged 15 and over, in Canada, and in one of the nine
disorder domains presented in the table of the previous page
All the people in Canada in one of the nine disorder domains presented
in the table of the previous page and belonging to one of the following
age groups: 15-24, 25-44, 45-64, 65+
All the people aged 15 and over, in a province or a health region and in
one of the nine disorder domains presented in the table of the previous
page
For example, suppose we are interested in determining the CV of the
estimate of the proportion of occasional drinkers among
all individuals aged
25 to 44 who have had a depression episode in the last 12 months. Since the
population of interest is one of the populations for which the disorder tables
were made, these latter tables have to be used to determine the appropriate
approximate CV.
On the other hand, if we are interested in determining the CV of the
estimate of the proportion of occasional drinkers among
all females aged 25
to 44 who have had a depression episode in the last 12 months, we have to
use the general tables, since the
females aged 25 to 44 who have had a
depression episode in the last 12 months is not one of the domains of
disorder previously described (all the individuals aged 25 to 44 who have
had a depression episode in the last 12 months is a disorder domain, but not
all the females aged 25 to 44).
62
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Rule 1.1: Estimates of Numbers Possessing a Characteristic
(Aggregates) in one of the 9 disorder domains
On the appropriate Sampling Variability Disorder Table for totals,
locate the estimated number in the left-most column of the table
(headed "Total"). Since not all the possible values for the estimate
are available, the smallest value which is the closest must be taken
(as an example, if the estimate is equal to 1,700 and the two closest
available values are 1,000 and 2,000, the first has to be chosen).
The CV is at the junction of the line where the estimate is and the
column that corresponds to the disorder domain of interest. See
Example 2.1 in Section 11.2.
Rule 1.2: Estimates of Proportions or Percentages Possessing a
Characteristic in one of the 9 disorder domains
On the appropriate Sampling Variability Disorder Table for totals,
locate the estimated number in the left-most column of the table
(headed "Proportion"). Since all the possible values for the
proportion are not available, the smallest value which is the closest
must be taken. The CV is at the junction of the line where this value
is and the column that corresponds to the disorder domain of
interest.
Rule 1.3: Estimates of Differences Between Aggregates or Percentages
The standard error of a difference between two estimates is
approximately equal to the square root of the sum of squares of each
standard error considered separately. That is, the standard error of a
difference (
) is:
12
ˆˆ
ˆ
XXd −=
)
X
(
+
)
X
(
=
2
2
2
1
1
2
d
αα
σ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
where is estimate 1, is estimate 2, and α
1
ˆ
X
2
ˆ
X
1
and the
coefficients of variation of
and respectively.
1
ˆ
X
2
ˆ
X
The coefficient of variation of
d
is given by . This formula
is accurate for the difference between independent populations or
subgroups, but is only approximate otherwise.
ˆ
d /
d
ˆ
ˆ
ˆ
σ
63
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Rule 1.4: Estimates of Ratios
In the case where the numerator is a subset of the denominator, the
ratio should be converted to a percentage and Rule 2 applied. This
would apply, for example, to the case where the denominator is the
number of drinkers among all individuals in Canada who had a
depression episode in the last 12 months and the numerator is the
number of regular drinkers among all individuals in Canada who
have had a depression episode in the last 12 months.
Consider the case where the numerator is not a subset of the
denominator, as for example, the ratio of the number of individuals
who are occasional or regular drinkers compared to the number of
individuals who do not drink at all. The standard deviation of the
ratio of the estimates is approximately equal to the square root of
the sum of squares of each coefficient of variation considered
separately multiplied by
, where is the ratio of the estimates
(
). That is, the standard error of a ratio is:
R
ˆ
R
ˆ
X
/
X
= R
21
ˆˆ
ˆ
αα
σ
ˆˆ
ˆ
ˆ
ˆ
2
2
1
2
R
+ R =
where
1
ˆ
α
and
2
ˆ
α
are the coefficients of variation
of
and respectively.
X
1
ˆ
X
2
ˆ
The coefficient of variation of
R
ˆ
is given by
αα
σ
ˆˆ
ˆ
ˆ
ˆ
2
2
1
2
R
+ = R /
.
The formula will tend to overstate the error, if
and are
positively correlated and understate the error if
and are
negatively correlated.
X
1
ˆ
X
2
ˆ
X
1
ˆ
X
2
ˆ
Rule 1.5: Estimates of Differences of Ratios
In this case, Rules 1.3 and 1.4 are combined. The CV's for the two
ratios are first determined using Rule 1.4, and then the CV of their
difference is found using Rule 1.3.
64
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
11.1.2 How to Use the general CV tables
The general tables are made to determine a CV when the population of
interest is not one of the disorder domains:
All the individuals aged 15 or more in Canada
All the individuals in Canada in one of the following age groups: 15-24,
25-44, 45-64, 65+
All the individuals aged 15 or more in a province or a health region
Any other population that is not one of the disorder domains described
in Section 11.1.1
As an example, suppose we are interested in determining the CV of the
estimate of the proportion of regular drinkers in Canada, in a province, in a
health region, or in one of the following age groups, in Canada: 15-24, 25-
44, 45-64, 65+. Since this population is not one of the disorder domains, the
general tables have to be used.
Rule 2.1: Estimates of Numbers Possessing a Characteristic (Aggregates)
with the general tables
The coefficient of variation depends only on the size of the estimate
itself. On the appropriate general Sampling Variability Table,
locate the estimated number in the left-most column of the table
(headed "Numerator of Percentage") and follow the asterisks (if
any) across to the first figure encountered since not all the possible
values for the estimate are available, the smallest value which is the
closest must be taken). This figure is the approximate coefficient of
variation.
Rule 2.2: Estimates of Proportions or Percentages Possessing a
Characteristic with the general tables
The coefficient of variation of an estimated proportion (or
percentage) depends on both the size of the proportion and the size
of the numerator upon which the proportion is based. Estimated
proportions are relatively more reliable than the corresponding
estimates of the numerator of the proportion when the proportion is
based upon a sub-group of the population. This is due to the fact
that the coefficients of variation of the latter type of estimates are
65
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
based on the largest entry in a row of a particular table, whereas
the coefficients of variation of the former type of estimators are
based on some entry (not necessarily the largest) in that same row.
(Note that in the tables the CV's decline in value reading across a
row from left to right). As an example, the estimated
proportion of
regular drinkers is more reliable than the estimated
number of
regular drinkers.
When the proportion (or percentage) is based upon the total
population covered by each specific table, the CV of the proportion
is the same as the CV of the numerator of the proportion. In this
case, this is equivalent to applying Rule 1.
When the proportion (or percentage) is based upon a subset of the
total population (e.g., those who never drink), reference should be
made to the proportion (across the top of the table) and to the
numerator of the proportion (down the left side of the table). Since
not all the possible values for the proportion are available, the
smallest value which is the closest must be taken (for example, if
the proportion is 23% and the two closest values available in the
column are 20% and 25%, 20% must be chosen) The intersection of
the appropriate row and column gives the coefficient of variation.
Rule 2.3: Estimates of Differences between Aggregates or Percentages
Same as Rule 1.3.
Rule 2.4: Estimates of Ratios
Same as Rule 1.4.
Rule 2.5: Estimates of Differences of Ratios
Same as Rule 1.5.
66
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
11.2 Examples of Using the CV Tables for Categorical Estimates
The following "real life" examples are included to assist users in applying the
foregoing rules. Examples 1.1 and 1.2 are related to the general CV tables.
Examples 1.3 and 1.4 show how to use these coefficients to evaluate the
variability of estimates that are differences of totals, differences of proportions
or ratios. Finally, Examples 2.1 and 2.2 show how to use the disorder CV tables.
Example 1.1: Estimates of Numbers Possessing a Characteristic
(Aggregates)
with the general tables
Suppose that a user estimates that 3,037,049 individuals in Canada have had at
least one depression episode in their life. How does the user determine the
coefficient of variation of this estimate?
1) Refer to the CANADA level general CV table since the target population
is all the Canadians (except the out-of-scopes)
2) The estimated aggregate (3,037,049) does not appear in the left-hand
column (the "Numerator of Percentage" column), so it is necessary to use
the smallest figure closest to it, namely 3,000,000.
3) The coefficient of variation for an estimated aggregate (expressed as a
percentage) is found by referring to the first non-asterisk entry on that
row, namely, 2.1%.
4) So the approximate coefficient of variation of the estimate is 2.1%.
According to the Sampling Variability Guidelines presented in Section
10.4, the finding that there were 3,037,049
individuals have had at least
one depression episode in their life is publishable with no qualifications.
Example 1.2: Estimates of Proportions or Percentages Possessing a
Characteristic with the general tables
Suppose that the user estimates that 3,037,049/24,996,593 = 12.1% of all the
individuals in Canada have had at least one depression episode in their life. How
does the user determine the coefficient of variation of this estimate?
1) Refer to the CANADA level general CV table since the estimate is
a proportion based on a subpopulation of the total population that is not a
disorder domain.
67
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
2) It is necessary to use both the percentage (12.1%) and the numerator
portion of the percentage (24,996,593) in determining the coefficient of
variation.
3) The numerator (3,037,049) does not appear in the left-hand column (the
"Numerator of Percentage" column) so it is necessary to use the smallest
figure closest to it, namely 3,000,000. Similarly, the percentage estimate
does not appear as any of the column headings, so it is necessary to use
the smallest figure closest to it (on the right), namely 10.0%.
4) The figure at the right of the intersection of the row and column used
(since there are asterisks at this intersection, the first number available at
the right must be taken), namely 2.1% is the coefficient of variation
(expressed as a percentage) to be used.
5) So the approximate coefficient of variation of the estimate is 2.1%.
According to the Sampling Variability Guidelines presented in Section
10.4, the finding that 12.1% of all the individuals in Canada have had at
least one depression episode in their life can be published with no
qualifications.
Example 1.3: Estimates of Differences Between Aggregates or Percentages
Suppose that a user estimates that, among men, 1,125,806/12,286,111 = 9.1%
have had a depression episode in their life (estimate 1), while for women, this
percentage is estimated at 1,911,243/12,710,482= 15.0 % (estimate 2). How
does the user determine the coefficient of variation of the difference between
these two estimates?
1) The Canada level CV general table have to be used to obtain the CV’s of
the two estimates since the target populations are men for the first estimate
and women for the second. Using this table in the same manner as
described in example 1.2 gives the CV for estimate 1 as 3.8% (expressed
as a percentage), and the CV for estimate 2 as 3.0% (expressed as a
percentage).
68
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
2) Using rule 1.3, the standard error of a difference ( d = - ) is :
ˆ
X
2
ˆ
X
1
ˆ
)
X
(
+
)
X
(
=
2
2
2
1
1
2
d
αα
σ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
where
X
1
is estimate 1, is estimate 2, and
ˆ
X
2
ˆ
1
ˆ
α
and
2
ˆ
α
are the
coefficients of variation of
and respectively.
X
1
ˆ
X
2
ˆ
The standard error of the difference
d
= (.091 - .150) = -.059 is :
ˆ
006.
030.150.038(.091.
ˆ
ˆ
=
)])([(
+
)])[(
=
22
d
σ
3) The coefficient of variation of
d
is given by
=
ˆ
d /
d
ˆ
ˆ
ˆ
σ
.102.0|059.|006. =−/
4) So the approximate coefficient of variation of the difference between the
estimates is 10.2% (expressed as a percentage). According to the
Sampling Variability Guidelines presented in Section 10.4, this estimate can
be published with no qualifications.
Example 1.4: Estimates of Ratios
Suppose that the user estimates that 14,498,721 individuals are regular drinkers,
while 4,760,435 individuals are occasional drinkers. The user is interested in
comparing the estimate of daily to occasional smokers in the form of a ratio.
How does the user determine the coefficient of variation of this estimate?
1) First of all, this estimate is a ratio estimate, where the numerator of the
estimate (=
X
) is the number of regular drinkers. The denominator of the
estimate (=
X
) is the number of occasional drinkers.
1
ˆ
2
ˆ
2) Refer to the CANADA level general CV table.
3) The numerator of this ratio estimate is 14,498,721. The smallest figure
closest to it in the left column of the CV table is 12,500,000. The
coefficient of variation for this estimate (expressed as a percentage) is
found by referring to the first non-asterisk entry on that row, namely,
0.6%.
69
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
4) The denominator of this ratio estimate is 4,760,435. The smallest figure
closest to it is 4,000,000. The coefficient of variation for this estimate
(expressed as a percentage) is found by referring to the first non-asterisk
entry on that row, namely, 1.8%.
5) So the approximate coefficient of variation of the ratio estimate is given
by rule 1.4, which is,
αα
α
ˆˆ
ˆ
ˆ
2
2
1
2
R
+ =
,
That is,
9
86
ˆ
ˆ
0.01 =
)
(.01 +
)
(.00 =
22
R
α
where α
1
and α
2
are the coefficients of variation of and
respectively. The obtained ratio of regular drinkers to occasional drinkers
is 14,498,721/4,760,435 which is 3.05:1.
X
1
ˆ
X
2
ˆ
The coefficient of variation of this estimate is 1.9% (expressed as a
percentage), which is releasable with no qualifications, according to the
Sampling Variability Guidelines presented in Section 10.4.
Example 2.1 : Estimates of Numbers Possessing a Characteristic
(Aggregates) with the disorder tables
Suppose that a user estimates that in Canada, 664,642 individuals among those
who have had at least one depressive episode in their life are regular drinkers.
How does the user determine the coefficient of variation of this estimate?
1) Refer to the CANADA level disorder CV table for an estimate expressed
as a total since the target population is all Canadians who have had at least
one depressive period in their life.
2) The estimated aggregate (664,642) does not appear in the left-hand
column (the "Total " column), so it is necessary to use the smallest figure
closest to it, namely 500,000.
70
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
3) The coefficient of variation for an estimated aggregate (expressed as a
percentage) is found at the intersection of the column « DEP – Lifetime »
and the line found in 2), namely, 5.4%.
4) So the approximate coefficient of variation of the estimate is 5.4%. The
finding that there were 664,642 individuals among those who have had at
least one depressive episode in their life are regular drinkers is publishable
with no qualifications.
Example 2.2 : Estimates of Proportions or Percentages Possessing a
Characteristic with the disorder tables
Suppose that the user estimates that 1,195,955 / 3,037,049
= 39.4% of all
Canadians, among those who have had at least one depressive episode in their
life, have had one of these episodes in the last 12 months. How does the user
determine the coefficient of variation of this estimate?
1) Refer to the CANADA level disorder CV table for a proportion since the
target population is all Canadians who have had at least one depressive
episode in their life.
2) The percentage (39.4%) does not appear in the left-hand column (the
"Percentage" column) so it is necessary to use the smallest figure closest
to it, namely 35.0 %.
3) The figure at the intersection of the column « DEP – Lifetime » and the
line found in 2), namely 3.1%, is the coefficient of variation (expressed as
a percentage) to be used.
4) So the approximate coefficient of variation of the estimate is 3.1%.
According to the Sampling Variability Guidelines presented in Section
10.4, the finding that 39.4% of all Canadians, among those who have had
at least one depression episode in their life, have had one in the last 12
months, can be published with no qualifications.
71
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
11.3 How to Use the CV Tables to Obtain Confidence Limits
Although coefficients of variation are widely used, a more intuitively meaningful
measure of sampling error is the confidence interval of an estimate. A confidence
interval constitutes a statement on the level of confidence that the true value for
the population lies within a specified range of values. For example a 95%
confidence interval can be described as follows: if sampling of the population is
repeated indefinitely, each sample leading to a new confidence interval for an
estimate, then in 95% of the samples the interval will cover the true population
value.
Using the standard error of an estimate, confidence intervals for estimates may be
obtained under the assumption that, under repeated sampling of the population,
the various estimates obtained for a population characteristic are normally
distributed about the true population value. Under this assumption, the chances
are about 68 out of 100 that the difference between a sample estimate and the true
population value would be less than one standard error, about 95 out of 100 that
the difference would be less than two standard errors, and about 99 out 100 that
the differences would be less than three standard errors. These different degrees
of confidence are referred to as the confidence levels.
Confidence intervals for an estimate
X
ˆ
, are generally expressed as two numbers,
one below the estimate and one above the estimate, as (
X
ˆ
-k,
X
ˆ
+k), where k is
determined depending upon the level of confidence desired and the sampling error
of the estimate.
Confidence intervals for an estimate can be calculated directly from the
Approximate Sampling Variability Tables by first determining from the
appropriate table the coefficient of variation of the estimate
X
ˆ
, and then using the
following formula to convert to a confidence interval CI:
] Xz + X , Xz - X [ =
CI
XXX
αα
ˆ
ˆˆ
ˆ
ˆˆ
ˆˆ
where
α
ˆ
ˆ
X
is determined coefficient of variation for
X
ˆ
, and
z = 1 if a 68% confidence interval is desired
z = 1.6 if a 90% confidence interval is desired
z = 2 if a 95% confidence interval is desired
z = 3 if a 99% confidence interval is desired.
72
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Note: Release guidelines which apply to the estimate also apply to the
confidence interval. For example, if the estimate is not releasable, then the
confidence interval is not releasable either.
11.4 Example of Using the CV Tables to Obtain Confidence Limits
A 95% confidence interval for the estimated proportion of individuals who have
had a depressive episode in the last 12 months (from example 2.2, sub-section
11.2) would be calculated as follows:
X
ˆ
= 0.394
z = 2
α
ˆ
ˆ
X
= 0.031 is the coefficient of variation of this estimate as determined
from the tables.
CI
X
= {0.394 - (2) (0.394) (0.031), 0.394 + (2) (0.394) (0.031)}
CI
X
= {0.370 , 0.418}
11.5 How to Use the CV Tables to do a Z-test
Standard errors may also be used to perform hypothesis testing, a procedure for
distinguishing between population parameters using sample estimates. The
sample estimates can be numbers, averages, percentages, ratios, etc. Tests may
be performed at various levels of significance, where a level of significance is
the probability of concluding that the characteristics are different when, in fact,
they are identical.
Let
and be sample estimates for 2 characteristics of interest. Let the
standard error on the difference
- be
X
1
ˆ
X
2
ˆ
X
1
ˆ
X
2
ˆ
σ
ˆ
ˆ
d
. If the ratio of - over
X
1
ˆ
X
2
ˆ
σ
ˆ
ˆ
d
is between -2 and 2, then no conclusion about the difference between the
characteristics is justified at the 5% level of significance. If however, this ratio
is smaller than -2 or larger than +2, the observed difference is significant at the
0.05 level.
73
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
11.6 Example of Using the CV Tables to do a Z-test
Let us suppose we wish to test, at 5% level of significance, the hypothesis that
there is no difference between the proportion of men who have had a depressive
episode AND the proportion of women who have had one. From example 1.3,
sub-section 11.2, the standard error of the difference between these two
estimates was found to be = .006. Hence,
8.9
6
59
6
15091
ˆˆ
ˆ
−
−
=
.00
.0
=
.00
. - .
=
X
-
X
=z
d
21
σ
Since z = -9.8 is smaller than -2, it must be concluded that there is a significant
difference between the two estimates at the 0.05 level of significance. Note that
the two sub-groups compared are considered as being independent, so the test is
correct.
11.7 Exact Variances/Coefficients of Variation
All coefficients of variation in the Approximate Sampling Variability Tables
(CV Tables) are indeed approximate and, therefore, unofficial.
The computation of exact coefficients of variation is not a straightforward task
since there is no simple mathematical formula that would account for all CCHS
sampling frame and weighting aspects. Therefore, other methods such as
resampling methods must be used in order to estimate measures of precision.
Among these methods, the bootstrap method is the one recommended for
analysis of CCHS data.
The computation of coefficients of variation (or any other measure of precision)
with the use of the bootstrap method requires access to information that is
considered confidential and not available on the public use microdata file. This
computation must be done via other avenues, such as remote access. Remote
access, as well as other alternatives to obtain exact coefficients of variation, is
discussed in sub-section 12.3.
The remote access service allows users to gain access to the bootstrap method
for the computation of coefficients of variation. A macro program, called
“Bootvar”, was developed in order to give users easy access to the bootstrap
method. The Bootvar program is available in SAS and SPSS formats, and is
made up of macros that calculate the variances of totals, ratios, differences
between ratios, and linear and logistic regressions.
There are a number of reasons why a user may require an exact variance. A few
are given below.
74
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Firstly, if a user desires estimates at a geographic level other than those available
in the tables (for example, at the rural/urban level), then the CV tables provided
are not adequate. Coefficients of variation of these estimates may be obtained
using "domain" estimation techniques through the exact variance program.
Secondly, should a user require more sophisticated analyses such as estimates of
parameters from linear or logistic regression models, the CV tables will not
provide correct associated coefficients of variation. Although some standard
statistical packages allow sampling weights to be incorporated in the analyses,
the variances that are produced often do not take into account the stratified and
clustered nature of the design properly, whereas the exact variance program
would do so.
Thirdly, for estimates of quantitative variables, separate tables are required to
determine their sampling error. Since most of the variables for the CCHS (Cycle
1.2) are primarily categorical in nature, this has not been done. Thus, users
wishing to obtain coefficients of variation for quantitative variables can do so
through the exact variance program. As a general rule, however, the coefficient
of variation of a quantitative total will be larger than the coefficient of variation
of the corresponding category estimate (i.e., the estimate of the number of
persons contributing to the quantitative estimate). If the corresponding category
estimate is not releasable, the quantitative estimate will not be either. For
example, the coefficient of variation of the estimate of the total number of
cigarettes smoked each day by individuals who smoke daily would be greater
than the coefficient of variation of the corresponding estimate of the number of
individuals who smoke daily. Hence if the coefficient of variation of the latter is
not releasable, then the coefficient of variation of the corresponding quantitative
estimate will also not be releasable.
Lastly, should users find themselves in a position where they can use the CV
tables, but this renders a coefficient of variation in the "marginal" range (16.6%
- 33.3%), the user should release the associated estimate with a warning
cautioning users of the high sampling variability associated with the estimate.
This would be a good opportunity to recalculate the coefficient of variation
through the exact variance program to find out if it is releasable without a
qualifying note. The reason for this is that the coefficients of variation produced
by the tables are based on a wide range of variables and are therefore considered
crude, whereas the exact variance program would give an exact coefficient of
variation associated with the variable in question.
75
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
11.8 Release Cut-off for the CCHS
The following tables present the release cut-offs for estimates of totals at the
health region, provincial and Canada levels, for some age groups (at the Canada
level only), as well as for the nine disorder domains. Estimates smaller than the
value given in the "Marginal" column may not be released under any
circumstances.
Table of Release Cut-offs for totals
Canada and Provinces – Total population
CV
CV BETWEEN
0% AND 16.5%
CV BETWEEN
16.5% AND
33.3%
PROVINCES
ACCEPTABLE
MARGINAL
Newfoundland-and-Labrador 14,500 3,500
Prince Edward Island 6,000 1,500
Nova Scotia 18,500 4,500
New Brunswick 18,500 4,500
Quebec 96,000 24,000
Ontario 52,500 13,000
Manitoba 25,500 6,500
Saskatchewan 20,500 5,000
Alberta 44,500 11,000
British Columbia 56,500 14,000
CANADA 57,000 14,000
76
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table of Release Cut-offs for totals
Age groups, Canada – Total population
CV
CV BETWEEN
0% AND 16.5%
CV
BETWEEN
16.5% AND
33.3%
Age Group
ACCEPTABLE MARGINAL
15-24 52,500 13,000
25-44 61,500 15,000
45-64 64,500 16,000
65+ 41,500 10,500
Table of Release Cut-offs for totals
Ontario Health Regions – Total population
CV
CV BETWEEN
0% AND 16.5%
CV BETWEEN
16.5% AND
33.3%
HEALTH REGION
ACCEPTABLE
MARGINAL
South West Ontario 41,000 10,500
Central South Ontario 33,000 8,500
Central West Ontario 72,500 18,500
Central East Ontario 68,500 17,500
Toronto 56,000 14,000
East Ontario 50,000 12,500
North Ontario 50,000 13,000
77
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table of Release Cut-offs for totals
Canada and Provinces – Disorder domains
CV
CV BETWEEN
0% AND 16.5%
CV BETWEEN
16.5% AND
33.3%
PROVINCE DOMAIN
ACCEPTABLE MARGINAL
DEP – 12 months 11,500 3,000
DEP – lifetime 11,500 3,000
MOOD -12 months 11,500 3,000
MOOD – lifetime 11,000 3,000
ANX – 12 months 20,500 5,000
ANX –lifetime 17,000 4,500
SUBS – 12 months 23,500 6,000
ALL – 12 months 17,500 4,500
Newfoundland-and-
Labrador
ALL – lifetime 14,500 3,500
DEP – 12 months 3,500 1,000
DEP – lifetime 4,500 1,000
MOOD -12 months 3,500 1,000
MOOD – lifetime 5,000 1,500
ANX – 12 months 3,000 500
ANX –lifetime 6,500 1,500
SUBS – 12 months 5,500 1,500
ALL – 12 months 4,500 1,000
Prince Edward Island
ALL – lifetime 5,500 1,500
DEP – 12 months 15,000 4,000
DEP – lifetime 20,500 5,000
MOOD -12 months 14,500 3,500
MOOD – lifetime 23,500 6,000
ANX – 12 months 23,000 6,000
ANX –lifetime 20,500 5,000
SUBS – 12 months 31,000 8,000
ALL – 12 months 20,500 5,000
Nova Scotia
ALL – lifetime 22,500 5,500
78
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
DEP – 12 months 12,000 3,000
DEP – lifetime 13,500 3,500
MOOD -12 months 12,000 3,000
MOOD – lifetime 13,500 3,500
ANX – 12 months 19,000 4,500
ANX –lifetime 16,500 4,000
SUBS – 12 months 15,500 4,000
ALL – 12 months 16,500 4,000
New Brunswick
ALL – lifetime 14,500 3,500
DEP – 12 months 64,500 16,000
DEP – lifetime 84,000 21,000
MOOD -12 months 64,000 16,000
MOOD – lifetime 87,500 21,500
ANX – 12 months 74,000 18,500
ANX –lifetime 87,500 21,500
SUBS – 12 months 74,000 18,500
ALL – 12 months 71,500 17,500
Quebec
ALL – lifetime 90,000 22,500
DEP – 12 months 54,500 13,500
DEP – lifetime 51,000 12,500
MOOD -12 months 53,000 13,000
MOOD – lifetime 51,000 12,500
ANX – 12 months 53,500 13,000
ANX –lifetime 50,500 12,500
SUBS – 12 months 40,500 10,000
ALL – 12 months 48,500 12,000
Ontario
ALL – lifetime 51,000 12,500
DEP – 12 months 28,500 7,000
DEP – lifetime 23,500 6,000
MOOD -12 months 28,000 7,000
MOOD – lifetime 23,000 5,500
ANX – 12 months 26,000 6,500
ANX –lifetime 26,500 6,500
SUBS – 12 months 20,500 5,000
ALL – 12 months 24,000 6,000
Manitoba
ALL – lifetime 25,000 6,500
79
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
DEP – 12 months 18,000 4,500
DEP – lifetime 18,500 4,500
MOOD -12 months 17,500 4,500
MOOD – lifetime 19,500 5,000
ANX – 12 months 22,000 5,500
ANX –lifetime 23,500 6,000
SUBS – 12 months 21,500 5,500
ALL – 12 months 22,000 5,500
Saskatchewan
ALL – lifetime 22,000 5,500
DEP – 12 months 45,500 11,500
DEP – lifetime 44,500 11,000
MOOD -12 months 44,000 11,000
MOOD – lifetime 42,000 10,500
ANX – 12 months 43,000 10,500
ANX –lifetime 40,500 10,000
SUBS – 12 months 31,500 8,000
ALL – 12 months 42,000 10,500
Alberta
ALL – lifetime 41,500 10,500
DEP – 12 months 37,000 9,000
DEP – lifetime 41,500 10,500
MOOD -12 months 39,500 10,000
MOOD – lifetime 44,000 11,000
ANX – 12 months 48,000 12,000
ANX –lifetime 56,000 14,000
SUBS – 12 months 54,000 13,500
ALL – 12 months 52,000 13,000
British-Columbia
ALL – lifetime 52,500 13,000
DEP – 12 months 50,500 12,500
DEP – lifetime 57,500 14,000
MOOD -12 months 50,000 12,500
MOOD – lifetime 59,500 14,500
ANX – 12 months 52,000 13,000
ANX –lifetime 55,500 13,500
SUBS – 12 months 43,500 10,500
ALL – 12 months 51,000 12,500
CANADA
ALL – lifetime 58,500 14,500
80
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table of Release Cut-offs for totals
Ontario Health Regions – Disorder domains
CV
CV BETWEEN
0% AND 16.5%
CV BETWEEN
16.5% AND
33.3%
HEALTH REGION DOMAIN
ACCEPTABLE MARGINAL
DEP – 12 months 19,500 6,500
DEP – lifetime 24,000 7,000
MOOD -12 months 19,000 6,500
MOOD – lifetime 24,500 7,000
ANX – 12 months 19,500 7,000
ANX –lifetime 36,000 11,000
SUBS – 12 months 15,500 6,500
ALL – 12 months 26,500 8,000
South West Ontario
ALL – lifetime 35,000 9,500
DEP – 12 months 15,000 5,000
DEP – lifetime 23,000 6,500
MOOD -12 months 17,500 6,000
MOOD – lifetime 25,000 7,500
ANX – 12 months 18,000 6,000
ANX –lifetime 28,500 8,500
SUBS – 12 months 16,500 6,500
ALL – 12 months 22,000 6,500
Central South Ontario
ALL – lifetime 33,000 9,500
DEP – 12 months 64,500 28,000
DEP – lifetime 70,000 23,000
MOOD -12 months 64,000 26,500
MOOD – lifetime 69,500 22,000
ANX – 12 months 51,000 20,500
ANX –lifetime 52,000 16,000
SUBS – 12 months 20,000 7,500
ALL – 12 months 57,000 17,500
Central West Ontario
ALL – lifetime 60,500 17,000
81
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
DEP – 12 months 21,000 7,000
DEP – lifetime 33,500 10,000
MOOD -12 months 20,500 6,500
MOOD – lifetime 35,500 10,500
ANX – 12 months 39,500 15,000
ANX –lifetime 42,500 12,500
SUBS – 12 months 24,500 10,000
ALL – 12 months 41,000 12,500
Central East Ontario
ALL – lifetime 46,000 13,000
DEP – 12 months 25,000 8,000
DEP – lifetime 34,500 10,000
MOOD -12 months 25,000 8,000
MOOD – lifetime 34,000 9,500
ANX – 12 months 29,500 9,500
ANX –lifetime 33,500 9,500
SUBS – 12 months 13,500 5,500
ALL – 12 months 29,000 8,000
Toronto
ALL – lifetime 39,000 10,500
DEP – 12 months 25,000 8,500
DEP – lifetime 34,000 10,000
MOOD -12 months 26,500 9,000
MOOD – lifetime 36,500 10,500
ANX – 12 months 29,000 11,000
ANX –lifetime 40,000 12,000
SUBS – 12 months 20,000 9,000
ALL – 12 months 40,000 12,500
East Ontario
ALL – lifetime 42,000 11,500
DEP – 12 months 19,000 7,000
DEP – lifetime 22,000 6,500
MOOD -12 months 19,000 7,000
MOOD – lifetime 21,500 6,500
ANX – 12 months 12,000 4,000
ANX –lifetime 18,000 5,500
SUBS – 12 months 14,500 6,500
ALL – 12 months 22,500 7,000
North Ontario
ALL – lifetime 24,000 7,000
82
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
Table of Release Cut-offs for totals
Age groups, Canada – Disorder domains
CV
CV BETWEEN
0% AND 16.5%
CV BETWEEN
16.5% AND
33.3%
AGE GROUP DOMAIN
ACCEPTABLE
MARGINAL
15-24 DEP – 12 mois 53,000 13,000
DEP – à vie 48,000 12,000
MOOD – 12 mois 50,500 12,500
MOOD – à vie 50,000 12,500
ANX – 12 mois 46,000 11,500
ANX –à vie 57,500 14,500
SUBS – 12 mois 42,500 10,500
ALL – 12 mois 50,000 12,500
ALL – à vie 54,000 13,500
25-44 DEP – 12 mois 42,000 10,500
DEP – à vie 59,000 14,500
MOOD – 12 mois 41,500 10,000
MOOD – à vie 61,500 15,000
ANX – 12 mois 52,000 13,000
ANX –à vie 50,000 12,500
SUBS – 12 mois 32,500 8,000
ALL – 12 mois 45,500 11,000
ALL – à vie 54,500 13,500
45-64 DEP – 12 mois 59,000 14,500
DEP – à vie 61,500 15,000
MOOD – 12 mois 58,000 14,500
MOOD – à vie 62,000 15,500
ANX – 12 mois 60,000 15,000
ANX –à vie 67,500 16,500
SUBS – 12 mois 68,000 17,000
ALL – 12 mois 57,500 14,000
ALL – à vie 65,000 16,000
83
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
65+ DEP – 12 mois 46,500 11,500
DEP – à vie 40,500 10,000
MOOD – 12 mois 46,500 11,500
MOOD – à vie 40,500 10,000
ANX – 12 mois 45,000 11,000
ANX –à vie 39,000 9,500
SUBS – 12 mois N/A N/A
ALL – 12 mois 38,500 9,500
ALL – à vie 39,000 9,500
84
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
12. File Usage
This section starts by describing the
weight variable of the master file and explains
how it should be used when doing tabulations on the public use microdata file. This is
followed by an explanation of the variable naming convention that is employed for
the CCHS. The last part of the section discusses alternate approaches to data access
available to analysts.
12.1 Use of Weights
Only one weight, WTSB_M, appears on the file. This weight is applicable to all
age groups, provinces and territories. ALL VARIABLES ON THE FILE
SHOULD BE ANALYZED USING THIS WEIGHT.
(For a more detailed explanation on the creation of this weight, see Section 8 of
the documentation on weighting.)
12.2 Variable Naming Convention
The CCHS adopted a variable naming convention that allows data users to easily
use and identify the data based on module and cycle. The variable naming
convention includes the following mandatory requirements: restrict variable
names to a maximum of 8 characters for ease of use by analytical software
products; identify the survey cycle (Cycle 1.1, 1.2 …) in the name; and allow
conceptually identical variables to be easily identifiable over survey cycles. The
variable names for these identical modules and questions should only differ in
the cycle position identifying the particular survey occasion in which they were
collected.
12.2.1 Variable Name Component Structure in CCHS
Each of the eight characters in a variable name contains information about
the type of data contained in the variable.
Positions 1 to 3: Module/Questionnaire section name
Position 4: Survey cycle
Position 5: Variable type
Positions 6 to 8: Question number
85
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
For example: The variable from question 38B, Major Depressive Episode
Module, Cycle 1.2 (DEPB_38B):
Positions 1-3: DEP major depressive episode module
Position 4: B Cycle 1.2
Position 5: _ ( _ = collected data)
Positions 6-8: 38B question number and answer option
12.2.2 Positions 1 to 3: Variable / Questionnaire Section Name
The following values are used for the section name component of the
variable name:
ADM Administration MED Medication Use
AGP Agoraphobia MIA Manic Episode (Mania)
ALC Alcohol Use PAC Physical Activities
ALD Alcohol Dependence PAD Panic Disorder
CCC Chronic Conditions PWB
Psychological Well-being
Manifestation Scale
CPG Pathological Gambling RAC Restriction of Activities
DEP Major Depressive Episode SAM Sample Identifiers
DHH
Household Contact and
Demographics
SCR Screening Section
DIS Distress SDC
Socio-demographic
Characteristics
MHP Mental Health Profile SER Mental Health Services
EDU Education SOP Social Phobia
ETA Eating Troubles Assessment SPV Spiritual Values
GEN General Health SSM Social Support
GEO
Geographic Identifiers
(Methodology)
STR Stress
HWT Height and Weight TWD Two-week Disability
IDG Illicit Drug Use and Dependence WST Work Stress
INC Income WTS Sample Weights
LBF Labour Force
86
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
12.2.3 Position 4: Cycle
Cycle Description
A
Cycle 1.1: Canadian Community Health Survey
: Regional level survey, stratified by health region;
: Common content and optional content selected by health region;
: Estimates for health regions, provinces, territories and Canada.
B Cycle 1.2: Canadian Community Health Survey, Mental Health and
Well-Being
: Provincial level survey;
: Focus content with additional, general content;
: Estimates for the provinces and Canada.
12.2.4 Position 5: Variable Type
_
Collected variable A variable that appeared directly on the questionnaire
C
Coded variable A variable coded from one or more collected variables
(e.g., Standard Industrial Classification code (SIC))
D
Cross-sectional
derived variable
A variable calculated from one or more collected or
coded variables, usually calculated during head office
processing (e.g., Health Utility Index)
F
Flag variable A variable calculated from one or more collected
variables (like a derived variable), but usually
calculated by the data collection computer application
for later use during the interview (e.g., work flag)
G
Grouped variable Collected, coded, suppressed or derived variables
collapsed into groups (e.g., age groups)
I
Imputation flag
variable
A flag variable associated with another variable in the
data file and that indicates whether the latter was
imputed or not. Refer to the data dictionary to
identify which variable the flag variable is associated
with.
87
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
12.2.5 Positions 6 to 8: Variable Name
In general, the last three positions follow the variable numbering used on
the questionnaire. The letter "Q" used to represent the word "question" is
removed, and all question numbers are presented in a two- digit format.
For example, question Q01A in the questionnaire becomes simply 01A,
and question Q15 becomes simply 15.
For questions which have more than one response option, the final position
in the variable naming sequence is represented by a letter. For this type of
question, new variables were created to differentiate between a “yes” or
“no” answer for each response option. For example, if Q2 had 4 response
options, the new questions would be named Q2A for option 1, Q2B for
option 2, Q2C for option 3, etc. If only options 2 and 3 were selected,
then Q2A = No, Q2B = Yes, Q2C = Yes and Q2D = No.
12.3 Access to Master File Data
In order to protect the confidentiality of respondents participating in the survey,
microdata files must meet stringent security and confidentiality standards
required by the
Statistics Act before they are released for public access. To
ensure that these standards have been achieved, each microdata file goes through
a formal review process to ensure that an individual cannot be identified. Rare
values in variables that may lead to identification of an individual are suppressed
on the file or are collapsed to broader categories so that individual disclosure is
minimized. Frequently, these are the variables that are most critical for doing a
complete and comprehensive analysis of the survey data. Since a significant
amount of resources is spent on collecting these data, ensuring that the
microdata files reach their full analytical potential is important for a complete
return on the statistical investment.
Remote access to the survey master file is one way to have access to these data.
Each user of the microdata product can be supplied with a ‘dummy’ test master
file and a corresponding record layout. With this, the user can spend time
developing a set of analytical computer programs using the test file to confirm
that the program commands are functioning correctly. At that point, the code for
the custom tabulations is then sent via e-mail. The code will then be transferred
into Statistics Canada’s internal secured network and processed using the
appropriate master file of CCHS Cycle 1.2 data. Remote access allows computer
programs to be submitted by users for processing at Statistics Canada. Estimates
generated will be released to the user, subject to meeting the guidelines for
analysis and release outlined in Section 10 of this document. Results are
88
CCHS Cycle 1.2, Mental Health and Well-being Public Use Microdata File Documentation
screened for confidentiality and reliability concerns and, once these have been
addressed, the output is returned to the client. There is no charge for this service.
For more information, contact the Data Access Unit at the following address:
A second approach for any client is the production of custom tabulations done
by the Client Custom Services staff in Health Statistics Division. This service
allows users who do not possess knowledge of tabulation software products to
get custom results. As with remote access, the results are screened for
confidentiality and reliability concerns before release. Unlike remote access,
there is a charge for this service. You can have access to this service by writing
to
Finally, the Research Data Centre’s Program allows researchers to submit to
Statistics Canada, a research project that uses data from the Master File. These
projects are accepted based on a set of specific rules. When the project is
accepted, the researcher is designated as a "deemed employee" of Statistics
Canada for the duration of the research, and given access to the Master File data
from designated Statistics Canada sites. For more information on this program,
please consult the Statistics Canada website at the following address:
http://www.statcan.ca/english/rdc/index.htm.
89
