Upload
clifford-richard
View
214
Download
0
Tags:
Embed Size (px)
Citation preview
Survey of Labour and Survey of Labour and Income Dynamics (SLID)Income Dynamics (SLID)
All Congress SymposiumAll Congress Symposium“The riches of the Research Data Centres”“The riches of the Research Data Centres”
June 2, 2004June 2, 2004
University of Manitoba, WinnipegUniversity of Manitoba, Winnipeg Heather LatheHeather Lathe
Income Statistics DivisionIncome Statistics Division
2
Points to coverPoints to cover
Survey design highlights, content, Survey design highlights, content, analytical usesanalytical uses
New content as of 2002New content as of 2002
Getting started: SLID retrieval system Getting started: SLID retrieval system (SLIDRET)(SLIDRET)
Also available in photocopies:Also available in photocopies:
Variance calculation – bootstrapVariance calculation – bootstrap
Sources for more informationSources for more information
3
Original objectives of SLIDOriginal objectives of SLID
Labour market and income flowsLabour market and income flows Determinants of changeDeterminants of change Impacts on the familyImpacts on the family
Also:Also: Main source for cross-sectional income Main source for cross-sectional income
datadata
4
Design highlightsDesign highlights
Longitudinal data, among the first at Statistics CanadaLongitudinal data, among the first at Statistics Canada Labour market and income data togetherLabour market and income data together A wide variety of additional “explanatory” variablesA wide variety of additional “explanatory” variables Family make-up and changes are keyFamily make-up and changes are key
TimeframeTimeframe Individuals followed for 6 year periodIndividuals followed for 6 year period Annual contact to update informationAnnual contact to update information
5
Sample designSample design
Up to 2 interviews per yearUp to 2 interviews per year Labour in January and Income in May if no T1 Labour in January and Income in May if no T1
access permissionaccess permission Preliminary information asked during first contact Preliminary information asked during first contact
Reference year
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Panel 1
Panel 2
Panel 3
Panel 4
6
Who is interviewed?Who is interviewed? Longitudinal respondents selected at start of Longitudinal respondents selected at start of
panelpanel
Cohabitants also interviewed (every member of Cohabitants also interviewed (every member of the household)the household)
Movers followed (longitudinal)Movers followed (longitudinal)
Labour, education and income collected for Labour, education and income collected for persons 16+, labour not collected 70+persons 16+, labour not collected 70+
Housing variables collected for every in-scope Housing variables collected for every in-scope household, disability for every personhousehold, disability for every person
10 provinces, non-institutional, off-reserve10 provinces, non-institutional, off-reserve
7
2001 sample sizes2001 sample sizesPersons Persons HouseholdsHouseholds
CANADACANADA 62 50062 500 31 04731 047
NFLDNFLD 2 7652 765 1 2851 285
PEIPEI 1 7461 746 866866
NSNS 4 2544 254 2 1292 129
NBNB 3 8713 871 1 8821 882
QUEQUE 11 92811 928 6 1066 106
ONTONT 18 36618 366 8 8218 821
MANMAN 4 5974 597 2 3382 338
SASKSASK 4 3244 324 2 2372 237
ALBERTAALBERTA 5 2895 289 2 6432 643
BCBC 5 3605 360 2 7402 740
8
Sample size by CMA – 2001 householdsSample size by CMA – 2001 households
TorontoToronto 1 546 1 546
MontrealMontreal 1 326 1 326
WinnipegWinnipeg 1 117 1 117
VancouverVancouver 957 957
Ottawa-GatineauOttawa-Gatineau 850 850
EdmontonEdmonton 737 737
CalgaryCalgary 653 653
Kitchener Kitchener 501 501
HalifaxHalifax 555 555
St.Catharines-NiagaraSt.Catharines-Niagara 455 455
Quebec CityQuebec City 523 523
LondonLondon 418 418
HamiltonHamilton 405 405
Victoria Victoria 321321
9
SLID SLID contentcontent Over 1000 variablesOver 1000 variables Database organized into logically related Database organized into logically related
groups of variablesgroups of variables Database key variables:Database key variables:
person person jobjob job-absencejob-absence economic familyeconomic family household household household relationshipshousehold relationships monthly receipt of EI / WC /SAmonthly receipt of EI / WC /SA jobless spelljobless spell census familycensus family marital statusmarital status education certificateseducation certificates
10
Organization of contentOrganization of contentPERSON
LABOUR INCOME & WEALTH
EDUCATION PERSONAL CHARACTERISTICS
LABOUR MARKET ACTIVITY PATTERNS
WORK EXPERIENCE
JOBLESS PERIODS
JOB INFORMATION
JOB CHARACTERISTICS
ABSENCES FROM WORK
EMPLOYER ATTRIBUTES
INCOME SOURCES
MONTHLY RECEIPT OF UI / WC / SA
ASSETS (not yet available)
EDUCATIONAL ACTIVITY
LEVEL OF SCHOOLING
DEMOGRAPHICS
ETHNO-CULTURAL
DISABILITY
CHILDREN
GEOGRAPHY
HOUSEHOLD & FAMILY INFO
TRAINING
HOUSING
11
Employment and unemployment Employment and unemployment dynamics researchdynamics research
Gross changes in employment, unemployment and Gross changes in employment, unemployment and inactivity between months or yearsinactivity between months or years
Gross flow data of persons or jobs by industry, Gross flow data of persons or jobs by industry, occupation, worker characteristics and job occupation, worker characteristics and job characteristicscharacteristics
Duration of spells (jobs or unemployment)Duration of spells (jobs or unemployment) ExamplesExamples: :
To what extent are long spells of unemployment To what extent are long spells of unemployment experienced by the same individuals?experienced by the same individuals?
Why do people withdraw from the labour marketWhy do people withdraw from the labour market What precedes a transition into self-employment?What precedes a transition into self-employment?
12
Life-cycle labour market Life-cycle labour market transitionstransitions
Labour market transitions associated with Labour market transitions associated with particular stages of life cycleparticular stages of life cycle
ExamplesExamples:: Transitions from school to workTransitions from school to work Transitions from work to retirementTransitions from work to retirement Work absences taken to raise childrenWork absences taken to raise children What are typical life-cycle patterns in Canada What are typical life-cycle patterns in Canada
today?today? What are the subsequent activities of high school What are the subsequent activities of high school
drop-outs, and what precedes a return to school?drop-outs, and what precedes a return to school?
13
Job qualityJob quality
Examples:Examples: Wage differences between men and womenWage differences between men and women Job benefitsJob benefits UnderemploymentUnderemployment Career changeCareer change Career advancement and earningsCareer advancement and earnings Job polarization, e.g., in terms of wages and Job polarization, e.g., in terms of wages and
hours workedhours worked
14
Family economic mobilityFamily economic mobility
ExamplesExamples:: How stable is family income?How stable is family income? What proportion of families experience a What proportion of families experience a
significant improvement or deterioration in significant improvement or deterioration in income between two points in time?income between two points in time?
What are the determinants of these changes?What are the determinants of these changes? How important are changes in family How important are changes in family
composition (divorce, remarriage) in explaining composition (divorce, remarriage) in explaining a change in financial well-being?a change in financial well-being?
15
Dynamics of low incomeDynamics of low income
Incidence and duration of spells of low incomeIncidence and duration of spells of low income Factors associated with moves into or out of Factors associated with moves into or out of
low incomelow income ExamplesExamples::
Identifying the existence or otherwise of a Identifying the existence or otherwise of a “persistently poor” sub-population, and “persistently poor” sub-population, and associated characteristicsassociated characteristics
Studying flows into and out of low income in Studying flows into and out of low income in relation to government transfer programs and relation to government transfer programs and taxationtaxation
16
Life events and family Life events and family changeschanges
Identifying the various types of families Identifying the various types of families and how families change, including step-and how families change, including step-families and multi-generational familiesfamilies and multi-generational families
What are the determinants or impacts of What are the determinants or impacts of “life events”?“life events”?
ExampleExample:: - What are the family’s economic circumstances - What are the family’s economic circumstances
preceding a marriage break-up, and what are they preceding a marriage break-up, and what are they for each spouse and any children following a for each spouse and any children following a separation?separation?
17
Educational advancement and Educational advancement and combining school and workcombining school and work
Possible to view educational activity in the context Possible to view educational activity in the context of an individual’s other activities or family of an individual’s other activities or family circumstancescircumstances
ExamplesExamples:: Financial well-being, or family responsibilities, of Financial well-being, or family responsibilities, of
people pursuing post-secondary educationpeople pursuing post-secondary education To what degree do high school or post-secondary To what degree do high school or post-secondary
students combine work and schoolstudents combine work and school What is the labour market involvement of high What is the labour market involvement of high
school drop-outs and what precedes a return to school drop-outs and what precedes a return to schoolschool
18
New contentNew content
Starting with 2002:Starting with 2002: HousingHousing Education/trainingEducation/training Geography – new baseGeography – new base
19
Housing contentHousing contentFrom 1994 From 1994
dwelling tenure (rented/owned) and typedwelling tenure (rented/owned) and type
From 1999 From 1999 with/without mortgage; number of bedroomswith/without mortgage; number of bedrooms
From 2002 From 2002 the need for repairsthe need for repairs farm or home business operated from the propertyfarm or home business operated from the property Homeowners – shelter costs:Homeowners – shelter costs:
mortgage payments, utility costs mortgage payments, utility costs property taxes, condominium feesproperty taxes, condominium fees
Renters – shelter costs:Renters – shelter costs: monthly rent, utility costsmonthly rent, utility costs amenities included in the rent (parking, appliances, etc.)amenities included in the rent (parking, appliances, etc.) whether the rent is subsidised by government or an whether the rent is subsidised by government or an
employeremployer
20
New housing research –New housing research –examplesexamples
Do changes in employment trigger a housing move?Do changes in employment trigger a housing move?
Is a new job or promotion associated with a move to Is a new job or promotion associated with a move to more suitable, affordable or adequate housing?more suitable, affordable or adequate housing?
What proportion of families experience a significant What proportion of families experience a significant improvement or deterioration in housing between two improvement or deterioration in housing between two points in time?points in time?
How important are changes in family composition How important are changes in family composition (divorce, remarriage) in explaining a change in (divorce, remarriage) in explaining a change in suitability, affordability or adequacy of housingsuitability, affordability or adequacy of housing??
21
Education coverageEducation coverage
More complete in 2002More complete in 2002 secondary schoolsecondary school community collegecommunity college business schoolbusiness school vocational schoolvocational school universityuniversity courses, workshops, seminars courses, workshops, seminars training related to employmenttraining related to employment
22
Education – Education – collected characteristicscollected characteristics
Main subject / subjects of coursesMain subject / subjects of courses Reason for choosing area of study Reason for choosing area of study Goals related to current or future employmentGoals related to current or future employment Employer supportEmployer support Number of monthsNumber of months Number of weeksNumber of weeks Number of hours per weekNumber of hours per week Total number of hoursTotal number of hours Full time / part time Full time / part time Area of studies - code Area of studies - code Receipt of certificate or diplomaReceipt of certificate or diploma
23
New education-related researchNew education-related research
Impact of programs offered by elementary and high Impact of programs offered by elementary and high schools, trade schools, colleges, CEGEPs and schools, trade schools, colleges, CEGEPs and universitiesuniversities
gender, level of education, age, job tenure, labour market gender, level of education, age, job tenure, labour market experience and occupation. experience and occupation.
Life-long learningLife-long learning
Does training have impact on career progress or Does training have impact on career progress or income?income?
24
Geography Geography
Census 2001 geography introducedCensus 2001 geography introduced 1992-1998 keeps 1991 geography1992-1998 keeps 1991 geography From 1999 new geographyFrom 1999 new geography Some variables needs to be recoded back to Some variables needs to be recoded back to
19921992 Old 1991 geography Old 1991 geography
Old information for recoded variables and Old information for recoded variables and years 1999-2001 stored in new “old years 1999-2001 stored in new “old geography” variablesgeography” variables
25
Access to confidential Access to confidential microdatamicrodata
Custom data retrievalsCustom data retrievals
On-premises access On-premises access Research Data CentersResearch Data Centers
Remote access Remote access
SLIDRET – in RDC’sSLIDRET – in RDC’s
26
SLID database retrieval SLID database retrieval software (software (SLIDRETSLIDRET))
Produces a flat rectangular file in Text (ASCII) Produces a flat rectangular file in Text (ASCII) format to eliminate the requirement to format to eliminate the requirement to understand the database structureunderstand the database structure
By producing this study file, you get started on By producing this study file, you get started on some of the main choices for your study…some of the main choices for your study…
27
SLIDRET basic inputs SLIDRET basic inputs
Type of analysis: longitudinal or cross-Type of analysis: longitudinal or cross-sectional, and years of study.sectional, and years of study.
Unit of analysis: Person, Person-job, Unit of analysis: Person, Person-job, Family, Various spells , etc. (11 Family, Various spells , etc. (11 choices in all).choices in all).
Browse the variable descriptionsBrowse the variable descriptions
Loads the appropriate weight field Loads the appropriate weight field according to other specifications.according to other specifications.
28
SLIDRET variable choiceSLIDRET variable choice
Pick variables of interest.Pick variables of interest.
Can add weights (on top of default one)Can add weights (on top of default one)
If desired, possible to restrict the If desired, possible to restrict the population and sort the file.population and sort the file.
Note: SLIDRET prevents the user from Note: SLIDRET prevents the user from making inconsistent choices.making inconsistent choices.
29
SLIDRET outputsSLIDRET outputs Output file in text formatOutput file in text format
Output file in dbf formatOutput file in dbf format
Record layoutRecord layout
Data dictionaryData dictionary
Variable labelsVariable labels
Code setsCode sets
Univariate statisticsUnivariate statistics
Query - saved for future use or Query - saved for future use or modificationsmodifications
30
Variance estimationVariance estimation
SLID uses bootstrap variance estimationSLID uses bootstrap variance estimation
Bootstrap - replication method (process done Bootstrap - replication method (process done 1000 times1000 times):): n is the number of PSUs in a stratumn is the number of PSUs in a stratum (n – 1) PSUs selected by Simple Random (n – 1) PSUs selected by Simple Random
Sampling with replacement in each stratum to Sampling with replacement in each stratum to obtain a bootstrap sub-sampleobtain a bootstrap sub-sample
Set of bootstrap weights calculatedSet of bootstrap weights calculated
31
Bootstrap weightsBootstrap weights
Bootstrap weights are available for SLID Bootstrap weights are available for SLID reference years 1996-2001reference years 1996-2001
For all types of weightsFor all types of weights Cross-sectional weight (ICSWT26)Cross-sectional weight (ICSWT26) Labour weight (ILBWT26)Labour weight (ILBWT26) Longitudinal weight for each panel (ILGWT26)Longitudinal weight for each panel (ILGWT26) Combined panels longitudinal weight (ILWCP26)Combined panels longitudinal weight (ILWCP26)
1000 sets of weights1000 sets of weights
32
When to use bootstrapWhen to use bootstrap Bootstrap variance estimation can be Bootstrap variance estimation can be
done for all direct estimates:done for all direct estimates: totals, means, proportions, medians, totals, means, proportions, medians,
percentiles, parameters from a regression, percentiles, parameters from a regression, etc etc
Works for estimation of fixed effects Works for estimation of fixed effects does not for random effectsdoes not for random effects technical reason - few PSUs in first stage of technical reason - few PSUs in first stage of
samplingsampling E.g.,E.g.,
multilevel modelingmultilevel modeling random effects models random effects models
33
Why to use bootstrapWhy to use bootstrap
Bootstrap variance estimation accounts Bootstrap variance estimation accounts for complex survey design, specifically:for complex survey design, specifically: cluster samplecluster sample non-uniform response mechanismnon-uniform response mechanism
Use it for population modelsUse it for population models to account for sample designto account for sample design
34
How to estimate the varianceHow to estimate the variance
1.1. Calculate regular estimate Calculate regular estimate θθ
2.2. Execute the estimation program using each set of Execute the estimation program using each set of
bootstrap weights (1000 sets) and keep each bootstrap weights (1000 sets) and keep each
resulting estimate resulting estimate θθii, i=1, 2, …, 1000., i=1, 2, …, 1000.
3.3. The following formula estimates varianceThe following formula estimates variance
1000
1i
2i )(
999
1Variance θθ
35
Which bootstrap weights to useWhich bootstrap weights to use
Always use the bootstrap weights that Always use the bootstrap weights that correspond to the weight used to produce the correspond to the weight used to produce the estimateestimate if an estimate is produced using the cross-sectional if an estimate is produced using the cross-sectional
weight (ICSWT26) for 2001, then the bootstrap weight (ICSWT26) for 2001, then the bootstrap cross-sectional weights for 2001 should be used cross-sectional weights for 2001 should be used for variance estimationfor variance estimation
In order to maintain the maximum precision, In order to maintain the maximum precision, all 1000 weights should be used, especially all 1000 weights should be used, especially for estimations on small domainsfor estimations on small domains
36
More info availableMore info available
A document on how to calculate variance A document on how to calculate variance estimation is available in RDCsestimation is available in RDCs
Some SAS macros and related Some SAS macros and related documentation, as well as examples of documentation, as well as examples of programs doing bootstrap estimation with programs doing bootstrap estimation with SLID data, will be available in a near futureSLID data, will be available in a near future
More information from Longitudinal Data More information from Longitudinal Data Analysis Group at StatscanAnalysis Group at Statscan
37
Survey overviewSurvey overview
Like internet “home page” just for SLIDLike internet “home page” just for SLID www.statcan.cawww.statcan.ca: “: “Our products & services”; Our products & services”;
“Free”; “Personal finance & household finance”“Free”; “Personal finance & household finance” OverviewOverview Products & servicesProducts & services Notes, definitions, methodologyNotes, definitions, methodology QuestionnairesQuestionnaires Research papersResearch papers Data dictionaryData dictionary
38
For more information on For more information on SLIDSLID
Contact the Client Services Section atContact the Client Services Section at
[email protected]@statcan.ca
or toll-free at or toll-free at
1-888-297-73551-888-297-7355oror
951-7355951-7355