Publications

    Longitudinal and International Study of Adults Research Paper Series

    Evaluation of the Canadian Household Panel Survey pilot content

    Evaluation of the Canadian Household Panel Survey Pilot Content

    by Andrew Heisz 1

    Abstract

    In the fall of 2008, Statistics Canada, in partnership with Human Resources and Social Development Canada (HRSDC) and the Canadian academic community, put into the field the Canadian Household Panel Survey Pilot (CHPS-Pilot). This paper describes the background of the project, the steps taken in the development of the pilot survey, and the results of a series of explorations of the data collected.

    1. Introduction

    In the fall of 2008, Statistics Canada, in partnership with Human Resources and Social Development Canada (HRSDC) and the Canadian academic community, put into the field the Canadian Household Panel Survey Pilot (CHPS-Pilot). This paper describes the background of the project, the steps taken in the development of the pilot survey, and the results of a series of explorations of the data collected. Separate reports examine the methodological results of the pilot survey (Heisz (2011)) and the use of administrative data in the survey (Heisz and Langevin (forthcoming, 2013)).

    2. Pilot development

    2.1 Background

    In January 2006, Statistics Canada, the Social Science and Humanities Research Council of Canada and the Canadian Institute of Health Research hosted the conference: "Longitudinal Social and Health Surveys in an International Perspective". The conference identified an important data gap for Canada: Canada lacks a "general household panel survey". A general household panel survey is a multitopic longitudinal household survey with a sample representative of the population. Such a survey allows for research that stretches beyond traditional subject matter domains, enabling researchers to see how events in one domain may affect others, perhaps much later in life. In general household panel surveys, all household members are interviewed, which allows for analysis of family dynamics and their interactions with other domains in ways that would be impossible for other surveys2,3. Following the Montreal conference, the Policy Research Data Gaps fund provided three years of funding for the Canadian Household Panel Survey Pilot (CHPS-Pilot).4

    2.2 Governance and content development

    The CHPS-Pilot project was developed under a tripartite governance system with each of Statistics Canada, Human Resources and Social Development Canada (HRSDC) and the academic community represented. A steering committee made up of two Director Generals from Statistics Canada, a Director General from HRSDC and two academics directed the project. The survey was managed at Statistics Canada in the Income Statistics Division.

    Content development took place from February 2007 through March 2008, and was motivated by reports from four academic expert groups, each responsible for one of four major subject matter domains: (1) Labour and Income, (2) Family, (3) Human Capital Development, and (4) Health. These expert groups, comprising about 20 Canadian academics, advised on content needs and priorities, and discussed possible research uses for the new survey. In addition to the activities of the expert groups, other reports were contracted to study more specific areas.5

    Policy research experts from federal departments were invited to comment on the draft survey. Researchers from HRSDC, the Canadian Mortgage and Housing Corporation, and the Bank of Canada gave their feedback, resulting in numerous changes to the draft survey. Qualitative testing of the questionnaire in the form of one-on-one interview testing was undertaken from January through March 2008.

    2.3 Survey content and design

    The pilot survey was intended to develop a wave-1 questionnaire; the intent of a wave-1 questionnaire is to establish a foundation of data upon which the later waves of the survey could build. Accordingly, the pilot survey included some questions of a retrospective nature, as well as questions designed to generate a baseline picture of human capital, labour, health and the family. Additional variables related to demographics, and outcome variables related to family income and well-being were also included.

    The CHPS-Pilot survey was conducted between October 15th and December 31st, 2008. The name of the pilot, for the purposes of field work, was the Living in Canada Survey – Pilot (LCS-Pilot). The sample comprised 1,200 dwellings selected from the Labour Force Survey (LFS) Rotate-out frame and 1,400 dwellings selected from the LFS Area-frame. Roughly equal sized samples were drawn from each of four provinces - Ontario, Quebec, Saskatchewan and New Brunswick. To minimise collection costs, the pilot-survey had a small sample size and used a highly clustered design. As a result, the pilot survey does not yield estimates representative of the population.

    Statistics Canada (2008) contains additional information on the content and design of the CHPS-Pilot.

    3. Content evaluations

    In the subsections that follow, we summarize the conclusions of 12 evaluations of the Canadian Household Panel Survey Pilot (CHPS-Pilot) content. In each of these evaluations, researchers analysed the merits of a particular question or set of questions. The objective was to discuss the value of the question(s) in a panel study, and to demonstrate their possible use, even though the data that was available for analysis was from one wave only, and the sample was too small to produce meaningful results in many cases. Overall, the content evaluations uncovered a number of areas for improvement in the question modules, but even so, they provide examples of their potential use in multitopic longitudinal analyses.

    3.1 Human capital and skills measures

    As indicated above, human capital development was highlighted as one of the four core subject matter areas to be covered by the panel survey. For the wave-1 pilot, content focussed on education history, current school status, and a self-assessment of skills used on the job.

    Lance Lochner (Lochner (2010)) investigated the relationship between family background characteristics, family resources and high school enrolment and completion among 15-18 year old respondents. In all, the pilot survey identified 139, 15-18 year olds living at home who were either enrolled in or had graduated from high school. The results showed that family income and mother's education significantly affected enrolment or completion of high school for their children.

    Chris Robinson (Robinson (2010)) examined new self-assessed skills measures developed for the CHPS-Pilot. The skills questions consisted of six two-part questions, the first asking the importance of a particular skill, and the second asking for a self assessment of the level of skill required. To aid in the self assessment, the second question used a scale that was anchored by skill requirement labels. For example, for reading comprehension, the first question asked:

    How important is reading comprehension to the performance of your current job?

    1. Not important
    2. Somewhat important
    3. Important
    4. Very important
    5. Extremely important

    Respondents who answered somewhat important or higher to this question were then asked the second question, with a show card provided to indicate the anchor points on the scale:

    What level of reading comprehension is needed to perform your current job? Please select a number from 1 to 7.

    1.  
    2. Read step-by-step instructions for completing a form
    3.  
    4. Read a memo from management describing new personnel policies
    5.  
    6. Read a scientific journal article describing surgical procedures
    7.  

    The objective of this examination was to see if the new skills measures can help to better understand differences between occupations. To do this, there needs to be sufficient variation within both the importance of the skill indicated, and the level of mastery of the skill required at the job. Moreover, the two parts of the skills questions should not be too highly correlated.

    Table 3.1-1 shows the frequency response for 1522 workers who said that reading comprehension was at least somewhat important to the performance of their current job. The correlation between importance and level of skill was positive (0.57) but there was considerable variation in level within importance category. For instance, among respondents reporting reading comprehension as important, 44 reported a reading level of 2, 69 reported 3, 155 reported 4, and so on.

    Table 3.1-1
    Distribution of level by importance for reading comprehension

    Table 3.4.1-1 Distribution of level by importance for reading comprehension
    Importance categories Level of skills required
    1 2 3 4 5 6 7
    Somewhat important 7 66 43 54 15 8 -
    Important - 44 69 155 62 16 6
    Very important - 24 28 176 241 101 24
    Extremely important - 6 - 48 136 90 94
    (-) indicates fewer than 5 responses
    Source: Robinson (2010), using data from the Canadian Household Panel Survey – Pilot

    Robinson also noted that these new skill questions on the CHPS would allow us to examine skill heterogeneity within occupations. This would allow analysis of occupational mobility that can take into account the skill ranking of movers within occupations. In addition, it will allow for the measure of skill acquisition within occupation codes which would shed new light on how such skills are acquired. Again, what is important is that there is substantial variation in the data acquired within detailed occupational classes, which Robinson confirmed was the case. Robinson also provided some evidence that these measures of skill level are also useful in explaining wage variations within occupations.

    3.2 Family dynamics measures

    Among the family dynamics measures, the CHPS-pilot collected information on pre-survey marriage, common-law and parenting spells. It also included questions on fertility intentions for respondents aged 20 to 49 and parental leave for respondents who had children after 1997. It is important to capture some detail on the life histories of respondents in a panel survey, as the events and trajectories constitute important areas of research, and are likely to influence outcomes later in life.

    In their evaluation of the conjugal and parental histories in the CHPS-Pilot, Dana Hamplova and Céline LeBourdais (Hamplova and Le Bourdais (2010)) examined the probability of entering into a first union through marriage or common-law for five different cohorts of women who were aged 20-69 at the time of the survey, and compared these to the results of the 2001 and 2006 General Social Surveys. The results were consistent across the three surveys, with marriage probability declining and cohabitation probability rising across the three years represented by the surveys.

    France-Pascal Ménard, Dana Hamplova and Céline LeBourdais (Ménard, Hamplova and Le Bourdais (2010)) examined the information on parental leaves to find evidence of whether recent increases of the generosity of parental leave policies, which increased in 2000 for all of Canada, and then in Quebec only in 2006, had an effect on the frequency of parental leaves in this data. Ménard, Hamplova and LeBourdais reported several descriptive statistics and regressions suggesting a measurable increase in the take-up of parental leave over the period, after controlling for other influential demographic and background characteristics.

    Martine Lemonde, Solène Lardoux and Céline Le Bourdais (Lemonde, Lardoux and Le Bourdais (2010)) examined determinants of fertility intentions in the CHPS-Pilot. In all, the data provided 1,249 observations of adults aged 20-49 on fertility intentions. Fully 53% of respondents indicated no intention to have more children, 33% indicated an intention to have at least one more, and 3% said they did not know. The results of a multinomial logit regression indicated that being older, male, having more children, being separated, and having high or primary school as your highest level of schooling reduced the likelihood of saying "yes" to the question "Do you intend to have (more) children sometime?". Being married, in a common law partnership, speaking a language other than English or French, and living in Quebec each increased the likelihood of intending to have more children. In a separate regression, Lemonde, Lardoux and Le Bourdais found that most of the same variables also affected the number of children desired.

    3.3 Health measures

    The CHPS-Pilot provided a number of baseline measures of health status. The inclusion of these, along with a variety of other well-being measures, offered an opportunity to evaluate the linkages between health, labour and income, human capital and family.

    One area investigated by Émilie Renahy, Beatriz Alvarado Llano, Sam Harper and Amélie Quesnel-Vallée (Renahy, Alcarado Llano, Harper and Quesnel-Valée (2010)) was the possibility of using the multitopic nature of the CHPS-pilot to develop an index of economic exclusion. They used variables related to material deprivation, food security, financial and housing insecurity, and ability to access savings or credit in an emergency, to construct an index of economic insecurity. These variables were reduced to one index using factor analysis, and then the authors assessed the impact of this variable on self reported health, independent of household income using a nested multilevel logistic regression. The results suggested that both household income and economic exclusion were significant factors related to poor health.

    The CHPS-Pilot took an innovative approach to the measurement of Child Health and its impact on the family. The approach was to use the Children with Special Health Care Needs Screener (CSHCN-Screener)6 to identify families having a child/children with special health care needs (CSHCN), and then follow these with questions on whether these special needs impacted the parent's participation in the labour market (Questions used in the pilot survey are shown in Appendix table 1). Lisa Strohschein (Strohschein (2009)) reported that of the 617 children aged 0-12 identified through the pilot survey, 13.0% were identified by their parent as having a special health care need; an incidence which was quite similar to that seen in other countries for children of this age. As in other countries, boys were significantly more likely to have a special health care need than girls.

    Among households with at least one child aged 0-12, 20.5% were identified as a household with one or more CSHCN. Fully 21.6% of households with a CSHCN reported having financial problems, and 35.1% had a family member who cut back or stopped working, as a result of the extra care required for a CSHCN. These incidences were also similar to those found in US data.

    3.4 Labour, income and well-being measures

    While the previous section highlighted two studies that examined the intersection between economic well-being and health, this section discusses five reports that focused on aspects of labour, income and well-being.

    3.4.1 Material deprivation

    Michael Ornstein (Ornstein (2009)) analysed the measures of material deprivation on the pilot survey. There were two basic approaches to material deprivation measurement followed in the CHPS-Pilot. The first approach was to directly measure individual dimensions of material deprivation that are of interest, such as questions on food security, housing security and financial security. The second approach was through administering a bank of 10, 2-part questions on whether the household consumed specified items deemed to be necessities, and if the item was missing, whether this was because the respondent's household could not afford it. The results of these questions could then be used to derive an index describing the respondent's material deprivation (The ten questions in the material deprivation module are shown in appendix table 2).

    Ornstein found that both approaches gave useful information on well being. For example, table 3.4.1-1 shows results from the block of questions on financial security. He found that items which reflect the more acute states of deprivation were also more negatively related to income. For example, the mean log monthly household income of families who used a food bank was 7.45 ($1,720) compared to 8.33 ($4,146) among those that did not (3.4.1-1). Similarly, income was lower among those who reported missing each of the 10 deprivation items. For example, income was 7.52 ($1,844) among households who did not have fresh fruits and vegetables every day, and 8.35 ($4,230) among those that did.

    Table 3.4.1-1
    Financial security measures, by household income

    Table 3.4.1-1 Financial security measures, by household income
    Questions on financial security % Mean log monthly household income
    In the last 12 months, did you ever… Yes Yes No
    miss paying an electricity, gas or utility bill on time because you were short of money? 12.8 8.03 8.34
    miss paying the rent or mortgage on time because you were short of money? 5.6 7.89 8.33
    pawn or sell something because you were short of money? 2.8 7.79 8.32
    ask for financial help from friends or family because you were short of money? 13.2 7.85 8.37
    use a food bank? 13.1 7.45 8.33
    ask for help from welfare or community organizations because you were short of money? 5.6 7.47 8.36
    Source: Ornstein (2009), using data from the Canadian Household Panel Survey-Pilot.

    However, Ornstein questioned the value of including the material deprivation index items in a longitudinal survey, as many of the 10 items listed in the material deprivation index do not have research value outside of their contribution to the index. For example, while the questions on food security could be used in a number of different ways, the question on whether the household could afford to replace worn out furniture (one of the items on the material deprivation index list) would have limited application outside of the construction of an index of material deprivation.

    3.4.2 Expenditures

    Thomas Crossley and Yuqian Lu (Crossley and Lu (2010)) examined the set of expenditures questions on the CHPS-Pilot. While there are a number of good reasons to collect data on household expenditures in a longitudinal survey, there are challenges in measuring these expenditures related to recall error and response burden. In the CHPS-Pilot, two questions on food expenditures were asked, one related to expenditures on food to be consumed at home, and a second on food to be consumed outside of the home, such as in a restaurant. Additionally there were other expenditures questions related to transportation, clothing, housing and a residual "other items" category, allowing for a calculation of total expenditures for the month preceding the survey. Crossley and Yu compared the values collected from the CHPS-Pilot on expenditures to those collected by the expenditures-based Survey of Household Spending (SHS) for 2008. In the aggregate, the CHPS-Pilot expenditures drastically underestimated total expenditures compared to the SHS - mean expenditures reported in the CHPS-Pilot were 63% of SHS expenditures, with expenditures on Clothing, Transportation, and "other expenses" being the most under represented. Estimates of median expenditures on housing and utilities (at 85% the SHS values), and food consumed inside (at 83% SHS values) and outside the home (at 96% of SHS values) were closer to the SHS estimates. However, the underreporting of expenditures on food consumed inside the home was seen as somewhat surprising, and a larger problem than experienced with other surveys.

    3.4.3 Job prospects

    David Green (Green (2010)) examined responses to the CHPS-Pilot questions around expectations for future job changes. Specifically, the CHPS-Pilot asked workers:

    I would like you to think about your employment prospects over the next 12 months:

    Q1. how likely is it that you will leave your job voluntarily, that is, quit or retire?

    Q2. how likely is it that you will lose your job, that is, get laid off or fired or not have your contract renewed?

    Q3. If you were to lose your job, how likely is it that you would eventually find and accept a job that would be at least as good as your current job, in terms of wages and benefits?

    Possible determinants of responses to these questions include the education and experience of the worker, as well as the persons "wage residual". The wage residual is the difference between the actual wage the worker earns, and the expected wage the worker would earn given their age, education, gender and other observable factors. Green suggested that the wage residual might be negatively related to expectations about future jobs; people with disproportionately high wages, given their characteristics, would be less optimistic about future job prospects.

    The first step in his analysis was to estimate the wage residual. This was done by differencing the observed wage and the predicted wage. The regression equation used to generate the predicted wage was noted by Green to yield coefficients that were comparable to estimates obtained with other datasets, for example, the union wage and gender wage differentials of about 0.22 log points were within standard ranges. Next Green created dummy variables equalling one of the respondent responded "Likely" or "Very Likely" to the three expectations questions reported above and regressed these on the wage residual as well as education, experience and other controls. The wage residual was a significant negative predictor of whether a person thought he or she would leave his current job and of whether he thought he would be laid off from his current job. It was not significant in the model that estimated whether the person was optimistic in finding a job of equal quality to his or her current job.

    3.4.4 Parental leave

    Chaowen Chan, Dana Hamplova and Céline Le Bourdais (Chan, Hamplova and Le Bourdais (2010)) examined the information collected in the employment histories of respondents. In the CHPS-Pilot, the approach was to ask workers to report up to 6 past non-employment spells lasting 6-months or longer, and then to infer the work spells (other questions allowed researchers to determine whether these work spells were full time or part time). They compared the resulting jobless spells collected with responses to questions on another part of the survey on the use of parental leave from work following the birth of a child. They found that among the 244 observations where parents reported taking a 6-month parental leave, only 128 mentioned a corresponding jobless spell. It may be that many of the remaining respondents did not regard the parental leave spell as a jobless spell. More study is recommended to assess the quality of the jobless spells module.

    3.4.5 Self employment

    In his review of the CHPS-Pilot data Herb Schuetze (Schuetze (2010)) focused on the area of content related to self employment. First, he found that the overall rate of self employment in the survey, at 12.5% was quite similar to that identified in the 2006 census (12.4%). He then modeled the determinants of self-employment in each of the CHPS-Pilot and census data, and raised a concern that in some cases determinants had opposite signs when estimated with the CHPS compared to the Census. Some of these were likely due to small sample size considerations, but others, Schuetze wrote, might have been due to differences in the ways variables were measured, and in particular the highest level of education variable. The highest level of education variable was measured in the survey using information gathered in a detailed education history module, rather than through a "single highest level of education" question as was used in the census.

    Schuetze also highlighted two ways that the innovative questions included in the CHPS-Pilot could contribute to our knowledge of self-employment. First, he added variables denoting the presence of medical insurance, dental insurance and life insurance to the regression model described above. Both medical and dental insurance had a negative effect on the incidence of self-employment. While this result is not surprising, in the context of a panel dataset, these types of information could be crucial to understanding transition in or out of self employment. For example, the presence of medical or dental insurance may inhibit movement into entrepreneurial activity, or alternately, if the presence of such benefits in the spouse's employment plan may help the transition to self-employment. Only through measuring these data longitudinally could we answer these questions.

    Second, Schuetze examined the question on the survey which asks "at what age would you like to retire" to see if the self employed had a higher preference for working later in life. Reporting on the raw data, he found that the self employed wanted to retire at age 70 compared to age 61 for paid workers. He then estimated a model of the determinants of the age at which workers would like to retire. After controlling for a number of other characteristics, the model suggested that a little more than half of the raw differential in preferred retirement age remained unexplained (about 5.5 years), suggesting that it was something left unmeasured about the self employed that made them want to retire later. Panel data could be used to overcome the issue of unobserved heterogeneity in the context by examining transitions into self employment, holding fixed individual characteristics constant.

    4. Conclusion

    Readers are invited to look at Heisz (2011) for a summary of methodological results from the Canadian Household Panel Survey Pilot (CHPS-Pilot) as well as a forthcoming evaluation of administrative data linking (Heisz and Langevin, 2013). The results from these reports and other expertise gained from the CHPS-Pilot data will inform future work on longitudinal household survey development at Statistics Canada.

    Appendix

    Table 1
    Children with special health care needs screener

    Table 1 Children with special health care needs screener
    Question number Description
    Q05 Does [INSERT CHILDNAME] currently need or use medicine prescribed by a doctor, other than vitamins?
    1 Yes
    2 No (Go to Q20)
    Q10 Is this because of any medical, behavioral or other health condition?
    1 Yes
    2 No (Go to Q20)
    Q15 Is this a condition that has lasted or is expected to last for at least 12 months?
    1 Yes
    2 No
    Q20 Does [INSERT CHILDNAME] need or use more medical care, mental health or educational services than is usual for most children of the same age?
    1 Yes
    2 No (Go to Q35)
    Q25 Is this because of any medical, behavioral or other health condition?
    1 Yes
    2 No (Go to Q35)
    Q30 Is this a condition that has lasted or is expected to last for at least 12 months?
    1 Yes
    2 No
    Q35 Is [INSERT CHILDNAME] limited or prevented in any way in ^DT_HISHER ability to do the things most children of the same age can do?
    1 Yes
    2 No (Go to Q50)
    Q40 Is this because of any medical, behavioral or other health condition?
    1 Yes
    2 No (Go to Q50)
    Q45 Is this a condition that has lasted or is expected to last for at least 12 months?
    1 Yes
    2 No
    Q50 Does [INSERT CHILDNAME] need or use special therapy, such as physical, occupational or speech therapy?
    1 Yes
    2 No (Go to Q65)
    Q55 Is this because of any medical, behavioral or other health condition?
    1 Yes
    2 No (Go to Q65)
    Q60 Is this a condition that has lasted or is expected to last for at least 12 months?
    1 Yes
    2 No
    Q65 Does [INSERT CHILDNAME] have any kind of emotional, developmental or behavioral problem for which DT_HESHE needs or gets treatment or counselling?
    1 Yes
    2 No (Go to Q75)
    Q70 Has this problem lasted or is expected to last for at least 12 months?
    1 Yes
    2 No
    If any the above questions indicate a special health care need that is expected to last at least 12 months, then continue on to Q75
    Q75 Has [INSERT CHILDNAME]'s health condition caused financial problems for your family?
    1 Yes
    2 No
    Q80 Have you or other family members cut down on the hours you work in order to care for [INSERT CHILDNAME]?
    1 Yes
    2 No
    Q85 Have you or other family members stopped working because of
    [INSERT CHILDNAME]'s health conditions?
    1 Yes
    2 No
    Source: Q05 to Q70 comprise the CSHCN screener, and is (c) CAHMI – The Child and Adolescent Health Measurement Initiative

    Table 2
    Material deprivation module

    Table 2 Material deprivation module
    Question number Description
    Q05 Do you and your family save regularly at least about $20 per month for rainy days or for retirement?
    1 Yes (Go to Q15)
    2 No
    Q10 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q15 Do you and your family have fresh fruits and vegetables every day?
    1 Yes (Go to Q25)
    2 No
    Q20 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q25 Do you and your family have a small amount of money to spend each week on yourself?
    1 Yes (Go to Q35)
    2 No
    Q30 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q35 Do you and your family have meat, fish or vegetarian equivalent every other day?
    1 Yes (Go to Q45)
    2 No
    Q40 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q45 Are you and your family able to replace worn out furniture?
    1 Yes (Go to Q55)
    2 No
    Q50 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q55 Do you and your family have appropriate clothes for job interviews?
    1 Yes (Go to Q65)
    2 No
    Q60 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q65 Are you and your family able to get around either by having a car or by using a monthly bus, subway, or commuter train pass (or equivalent)?
    1 Yes (Go to Q75)
    2 No
    Q70 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q75 Are you and your family able to have friends or family over for a meal at least once per month?
    1 Yes (Go to Q85)
    2 No
    Q80 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q85 Do you and your family have at least two pairs of shoes, including one to wear outside in the winter?
    1 Yes (Go to Q95)
    2 No
    Q90 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason
    Q95 Are you and your family able to buy modest presents for family or friends at least once per year?
    1 Yes (Go to End)
    2 No
    Q100 Is this because you cannot afford it or for some other reason?
    1 Cannot afford it
    2 Some other reason

    References

    Bernard, P., Beach, C., Bélanger, P., Charbonneau, J., LeBourdais, C., Livingstone, D., McDaniel, S., McDonald, L., Phipps, S., Potvin, L., Quesnel-Vallée, A. and Douglas Willms J. (2005), Capturing the Lifecourse; The contribution of a Panel Study of Lifecourse Dynamics (PSLD) to public policy analysis in Canada, http://www.ciqss.umontreal.ca/Longit/session7_paper3.html

    Chaowen, C., Hamplová, D. and Le Bourdais, C. (2010), Evaluating the Information Collected in the Employment Histories and Comparing Work Interruption and Parental Leave Data in the Pilot Living in Canada Survey (LCS), Statistics Canada, Mimeo

    Crossley, T. F. and Yuqian, L. (2010), An Evaluation of the Expenditure Questions in the Living In Canada Survey – Pilot, Statistics Canada, Mimeo

    Green, David (2010), Brief Investigation of LICS, Labour Component, Statistics Canada, Mimeo

    Hamplová, D. and Le Bourdais, C. (2010), Evaluation of the Conjugal and Parental Histories in the Living in Canada Survey (LCS) Pilot Data, Statistics Canada, Mimeo

    Heisz, Andrew (2011), Results from the Canadian Household Panel Survey Pilot, Statistics Canada, catalogue no. 89-648-XIE no. 001

    Heisz, A. and Langevin, M. (2013), Historical Data Linkage of tax records on earnings: the case of the Canadian Household Panel Survey Pilot, Statistics Canada, 2013, forthcoming.

    Lemonde, M., Lardoux, S. and Le Bourdais, C. (2010), Determinants of Fertility Intentions in the Canadian Household Panel Survey, Statistics Canada, Mimeo

    Lochner, Lance (2010), Family Background, Family Resources, and Schooling in CHPS Pilot Survey, Statistics Canada, Mimeo

    Ménard, F-P, Hamplová, D. and Le Bourdais, C. (2010), Evaluation of the Measures of Parental Leave in the Pilot Data of the Living in Canada Survey (LCS), Statistics Canada, Mimeo

    Ornstein, Michael (2009), Measuring Food, Housing and Financial Security and Material Deprivation in the Living in Canada Survey, Statistics Canada, Mimeo

    Picot, G., Berthelot, J-M and Webber, M. (2006), Possible Future Directions for Longitudinal Surveys at Statistics Canada, paper presented at the Longitudinal Social and Health Surveys in an International Perspective Conference, Montreal, Canada.

    Renahy, É., Alvarado Llano, B., Harper, S. and Quesnel Vallée, A. (2010), Data Evaluation and International Comparability of the CHPS Pilot Survey, Statistics Canada, Mimeo

    Renahy, É., Alvarado Llano, B., Harper, S. and Quesnel Vallée, A. (2010), Health Inequalities over the Life Course: Social Exclusion, Statistics Canada, Mimeo

    Robinson, Chris (2010), Analysis of the Skills Measures in the Canadian Household Panel Survey Pilot, Statistics Canada, Mimeo

    Schuetze, Herbert (2010), Evaluation of LCS Pilot Self-Employment Content, Statistics Canada, Mimeo

    Statistics Canada (2008), Microdata User Guide: Living in Canada Survey – Pilot, Statistics Canada

    Strohschein, Lisa (2009), Data Evaluation of Living in Canada - Pilot: CSHCN Screener, Statistics Canada, Mimeo

    Notes

    1. Andrew Heisz, Income Statistics Division, Statistics Canada, 5th Floor, Jean Talon Building, 170 Tunney's Pasture Driveway, Ottawa, Canada, K1A 0T6, andrew.heisz@statcan.gc.ca
    2. The survey design and content would be motivated by the life course perspective. The life course perspective states that individual well-being reflects three complementary ideas: That lives are multifaceted, they are linked and they develop over time. Please see Bernard et al (2005) for more detail on the life course approach to analysis.
    3. Other general household panel surveys include the German Socio Economic Panel (GSOEP), the British Household Panel Survey (BHPS) and the Household Income and Labour Dynamics in Australia survey (HILDA).
    4. Please see Berthelot, Picot and Webber (2006) for more discussion of longitudinal surveys at Statistics Canada.
    5. Content recommendations are found in Alvarado, Quesnel-Vallée, Harper and Lynch (2007); Belleau, Charbonneau, Hamplova, Lapierre-Adamcyk, Laplante, Lardoux, Le Bourdais, Ménard and St-Amour (2007); Bowlus, Lochner, Robinson and Stinebrickner (2007); Fortin, Green, Lemieux and Milligan (2007); Strohschein (2009); and Sweetman and Warman (undated).
    6. More information on the CSHCN screener can be found at www.cahmi.org
    Date modified: