Ontario Material Deprivation Survey (OMDS)
Detailed information for 2008
The purpose of the survey is to develop indicators to gauge the ability of families to satisfy basic material needs such as food, clothing, housing as well as social needs of participation and leisure.
Data release - December 2, 2009
The Ontario Material Deprivation Survey (OMDS) was conducted on behalf of the Ontario Government to inform the Government of Ontario's poverty reduction plan. Specifically, the survey was designed to produce estimates of the incidence of missing two or more items (out of ten) for Ontario and various geographic subgroups, at the individual level, as well as for the estimation of indicators necessary for computing a Material Deprivation Index. The survey includes ten items deemed to be necessities, due to lack of resources. The list of ten items comes from a 2008 study by the Daily Bread Food Bank of Toronto. It also includes questions on income. The OMDS was fielded in March and April 2009 as a supplement to the Labour Force Survey (LFS, record number 3701).
Reference period: Calendar year 2008 (income), the LFS reference week in March and April 2009
- Household, family and personal income
- Income, pensions, spending and wealth
- Low income and inequality
Data sources and methodology
The target population is composed of Ontario economic families, excluding persons living on Indian reserves or in military barracks, and persons who had been living in an institution for more than six months. The targeted respondent was a family member 15+ years of age who was the most knowledgeable person in the family to answer the survey. In the event that there were multiple economic families in one household, the OMDS randomly selected one of these.
The majority of the questions asking about material deprivation were tested in a pilot of another Statistics Canada survey, while the income questions are from harmonized content standards for Statistics Canada income questions. The questionnaire was evaluated by the Questionnaire Review Committee (QRC) in Special Surveys Division.
This is a sample survey with a cross-sectional design.
The sample size for the OMDS is about 12,000 households, drawn from five LFS rotation groups.
In the event that there were multiple economic families in one household, the OMDS randomly selected one of these. The respondent for the survey was the member of an economic family who felt comfortable and knowledgeable in answering questions about the family's circumstances. This individual may or may not have been the same person as the LFS respondent.
Data collection for this reference period: 2009-03-22 to 2009-05-04
Responding to this survey is voluntary.
Data are collected directly from survey respondents.
The Ontario Material Deprivation Survey was a voluntary Labour Force Survey (LFS) independent supplement. OMDS was conducted as both a computer assisted telephone interview (CATI) and computer assisted personal interview (CAPI) supplement to the LFS. Since it was a "proxy" survey, any knowledgeable household member could be the respondent for the household.
View the Questionnaire(s) and reporting guide(s) .
Some editing was done at the time of data collection where information was inconsistent with previous entries. During data processing responses to income questions were examined for consistency; for example, reporting wage/salary as a source of family income and reporting of wage/salary values for any member of the family. Also, cases in which families reported no income for 2008 were analyzed using socio-demographic and labour information. Due to a low number of cases with possible discrepancies, the original responses were retained. Very low as well very high incomes were also examined. The impact of these values on key estimates, such as the median household/economic family income, was determined to be minimal, so no corrective action was taken.
Donor imputation was simultaneously conducted for both personal income and family income. Family income was derived by summing the total personal income of all members of a given family. The imputation rate for personal and family income are shown in the following table.
The principles behind the calculation of the weights for the OMDS are identical to those for the LFS. However, further adjustments are made to the LFS sub-weights in order to derive a final weight for the individual records on the OMDS microdata file.
1. An adjustment to account for the use of a 5/6 (five-sixth) sub-sample, instead of the full LFS sample.
2. An adjustment to account for the additional non-response to the supplementary survey; i.e., non-response to the OMDS households that did respond to the LFS. The procedure is similar to the LFS non-response weight adjustment, but groupings are based on different variables.
At this stage the weight is comprised of two components: the inverse of the sampling rate and the non-response adjustment. A third component, the family weighting adjustment described below, was added to improve accuracy of estimates.
Independent estimates are available monthly for various age and sex groups by province. These are population projections based on the most recent census data, records of births and deaths, and estimates of migration. Using a linear regression model, auxiliary information is used to arrive at the final weight. The regression is set up to ensure that the final weights it produces sum to the census projections for the auxiliary variables, namely various age-sex groups, economic regions and census metropolitan areas. This improves the reliability of estimates that can be produced by the OMDS. At the same time as ensuring consistency with external census counts, the family weighting procedure also ensures that every member of the economic family is assigned the same weight.
Estimates of material deprivation were compared with results obtained by other recent surveys. A material deprivation index was calculated and analyzed in the context of several socio-demographic characteristics.
Statistics Canada is prohibited by law from releasing any information it collects which could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.
The master file does not include any personal identifiers.
Revisions and seasonal adjustment
This methodology type does not apply to this survey.
Since the OMDS was a supplement to the LFS, the frame used was the LFS frame. Any non-response to the LFS had an impact on the OMDS frame. The quality of the sampling variables in the frame was very high. The OMDS sample consisted of Ontario economic families in five LFS rotation groups. The criteria used for the OMDS selection (such as a rotation group) were not missing for any LFS records.
Note that the LFS frame excludes about 2% of all households in the 10 provinces of Canada. Therefore, the OMDS frame also excludes a similar proportion of Ontario households. It is unlikely that this exclusion introduces any significant bias into the survey data.
Table 1 (in the "Additional documentation" link below) summarizes the response rates to the Labour Force Survey (LFS) and to the Ontario Material Deprivation Survey (OMDS).
The item non-response to the set of deprivation questions was very low, varying from 0 to 0.8%. Although there were some differences between those families who responded to all the deprivation questions and those who did not answer one or more deprivation questions, given the fact that the item non-response is so low, the impact on the final estimates should be negligible.
The item response rate for the deprivation questions combined with the OMDS response rate was between 88.0% and 88.8 %. The overall item response rate (accounting also for the LFS non-response) ranged from 83.0% and 83.7%.
The OMDS uses a multi-stage survey design and calibration, which means that there is no simple formula that can be used to calculate variance estimates. Therefore, an approximate method was needed. The Rao-Wu bootstrap method was used because the sample design and calibration needed to be taken into account when calculating variance estimates.