Transcript of the chat session on Income and Housing data, 2011 National Household Survey, which occurred on Friday, September 13, 2013 from 12:30 to 1:30 p.m. EDT

Note: This was a bilingual chat session, which means that the participants were able to submit their questions in English or French. Statistics Canada respects the Official Languages Act  and is committed to ensuring that information products of equal quality are available in both English and French. For that reason, all the questions and answers have been translated in the other official language.

 Moderator:  Hello everyone. The chat session will begin in just 5 minutes. Our experts on Income and Housing data, Brian Murphy, Eric Olson, Jerry Situ and Sandrine LeVasseur will be answering your questions regarding this third and last release of the 2011 National Household Survey. Stay tuned!
Friday, September 13, 2013 12:25:01

 Moderator:  Welcome everyone. It is 12:30 p.m. EDT and the session is now open. This is a bilingual chat session, which means that you can submit your questions in English or French. The experts will try to get to them in a timely manner and they will respond in the official language in which the question was asked.
Friday, September 13, 2013 12:30:48

 Rose: Thank you for the opportunity to submit questions to Stat Can Analysts. I have two questions that I hope can either be addressed in the chat or can be directed to where I can find the information. 1. In 2006 I was able to find income profiles by Occupation Code broken down by gender. Is that available this time. (Source: Occupation - National Occupational Classification for Statistics 2006 (720C), Sex (3) and Selected Demographic, Cultural, Labour Force, Educational and Income Characteristics (273) for the Population 15 Years and Over of Canada, Provinces, Territories, Census Metropolitan Areas and Census Agglomerations, 2006 Census - 20% Sample Data) 2. I'm interested in understanding trends in seniors income compared to 5 and 10 years ago. Any insights on that?
Friday, September 13, 2013 12:31:11

 Eric Olson: Hi Rose, thank you for your questions. 1. Occupation data broken down by gender is available from the NHS at 99-014-X2011042. We have avoided making inferences about trends from the NHS compared with other surveys. Many of our tables provide information by age group.
Friday, September 13, 2013 12:38:49

 sonya: In one of the NHS data tables and publications, you provide detailed economic and demographic characteristics for the top 1%, top 5% and top 10% of earners. Do you also have this data for top 20%, 30% etc? If so, where are these data located?
Friday, September 13, 2013 12:31:26

 Brian Murphy: Hi sonya. We do not have individual deciles associated with the high income estimates. We do have family income deciles for income composition at Table 32.
Friday, September 13, 2013 12:39:11

 Shawn Menard FCM: The 'cautions' on the data have been fairly consistent about comparisons with the 2006 release. However, in the release it mentions that the number of Canadian families who pay more than 30% of their income is about 1 in 4. Is it likely that this number is actually higher given the new survey methodology and likely lower response rates from lower-income households?
Friday, September 13, 2013 12:32:00

 Jerry Situ: Thank you for your question Shawn. The estimates on housing affordability from the NHS are the best estimates that can be produced from the NHS data. The data was evaluated through a comparison with other data sources. I invite you to take a look at the Data Comparability section of our Housing Reference Guide.
Friday, September 13, 2013 12:56:22

 Sameer: Will any of the data, in particular the income data be available at the DA level?
Friday, September 13, 2013 12:32:03

 Brian Murphy: Hello Sameer, Sorry we don't have data available at the DA level. Tables at other geographical levels such as province, CMA or CA levels are available: 2011 National Household Survey: Data tables.
Friday, September 13, 2013 12:44:07

 rariste: To what extent the income findings from NHS 2011 can be compared to that of the 2006 Census (long form questionnaire)?
Friday, September 13, 2013 12:33:43

 Eric Olson: Hi rariste, We have avoided presenting comparisons with data from the 2006 Census. More information on comparability can be found in the Income Reference Guide, National Household Survey, 2011 and the National Household Survey User Guide. Thank you for your interest in our data.
Friday, September 13, 2013 12:45:38

 Pbov: Hello, I see that Aboriginal population data is available on the public data tables for Income data, however, the Aboriginal population data is not available on the public data tables under the Housing tables (only at the community level). Is there any reason why Statistics Canada has not released Aboriginal population Housing data under these tables? Do you plan on releasing this data at a later date? Thanks, Paul
Friday, September 13, 2013 12:33:58

 Jerry Situ: Thank you for your question. Analysis of housing data for the Aboriginal population was not the focus of the analysis released on September 11. Aboriginal variables and housing variables were released with the NHS data, and tables can be created through a custom tabulation. There will be additional data tables released this year, and some of them with have Aboriginal variables and housing variables.
Friday, September 13, 2013 12:40:18

 Davidson: How significant is the impact of short form census questionnaire on the results of the NHS?
Friday, September 13, 2013 12:34:33

 Moderator: Davidson, thank you for your question. Unfortunately, our subject matter experts are not able to provide an answer during this session. We will follow up with you by email.
Friday, September 13, 2013 12:39:07

 Maria A: Can we get the GNR rate for the low income question? By municipality?
Friday, September 13, 2013 12:35:53

 Brian Murphy: Hi Maria. The GNR applies to specific geographies and not specific questions. Thank you for your question.
Friday, September 13, 2013 12:55:37

 Angella: "for returned questionnaires, only a certain proportion contained usable information for the income question. This is considered the item response rate." I just want to confirm the meaning of the item response rate for income: is it 60% of those that received the survey, or 60% of those that responded to the survey.
Friday, September 13, 2013 12:36:04

 Eric Olson: Hi Angella, Sorry that this was unclear in the guide. The correct interpretation is "60% of those that were selected in the survey".
Friday, September 13, 2013 13:03:39

 Maria A: Do we know the prevalence of low income families? (not just individuals)
Friday, September 13, 2013 12:36:15

 Brian Murphy: Hello Maria. Low-income rates are available for individuals but neither for families nor households.
Friday, September 13, 2013 12:51:11

 DerrickB: With such differences between NOC 2011 and NOC 2006, are there plans to release the data in the NOC 2006 format as well to better serve the end users?
Friday, September 13, 2013 12:41:33

 Eric Olson: Hi DerrickB. There is currently no plan to publish income data based on the NOC 2006. Custom tables based on the NOC 2006 are available upon request.
Friday, September 13, 2013 13:18:16

 Jay: Do you have monthly Shelter Costs in Vancouver, BC? And Toronto, ON? Where can I find this info? How do I get the specific bar graphs info (above)?
Friday, September 13, 2013 12:42:32

 Sandrine Levasseur: Thank you for your question. The average monthly shelter cost paid by households in Vancouver (CMA) was $1,294 per month,compared with $1,366 for households in Toronto. You can find more information in the 2011 National Household Survey: Data tables. All graphs have accessibility text information. If you click on the description of the figure, you will get all data related to the graph.
Friday, September 13, 2013 12:52:16

 Sara Mayo: Will target group profiles for NHS data be released on the website? If so, is there a planned date you can share? Will there be additional basic-cross tabs (data tables) that you are planning to add to the website, or is that it and everything else has to be a custom request? Can you share a date for the release of the next round of data tables if that is planned?
Friday, September 13, 2013 12:42:33

 Jerry Situ: Hello, thank you for your question. Additional data tables will be released later this year. Furthermore, all variables in the NHS are now available for custom requests.
Friday, September 13, 2013 12:48:09

 JimCuddyBuddy: I'm happy to have this chance to ask you all my question! I am curious about the affordability of detached homes in rural / urban Canada. In your video on YouTube you have info on affordability of various types of homes. I'm wondering if there is a significant gap between the amount of homeowners of detached homes above the affordability threshold in rural Canada vs urban Canada (on average). Many thanks
Friday, September 13, 2013 12:43:48

 Jerry Situ: Thank you for your question. We did not look specifically at the comparison that you are interested in. Generally, we do see that the proportion of households that exceed the affordability threshold is higher in the core of metropolitain areas. I invite you to take a look at some of the housing affordability maps to see how housing affordability varies across the country and in the largest 6 Census Metropolitain Areas: Thematic maps by theme
Friday, September 13, 2013 13:08:42

 ChantelleT: We have been told that data at national and provincial levels from the NHS is useable and useful. We understand the issues about comparing to previous years, but how useful does Statistics Canada think the data (income data) is, at a Municipal and Regional level?
Friday, September 13, 2013 12:43:52

 Eric Olson: Hi ChantelleT, Thanks for this question. We do think the income data should be considered of good quality at many levels for comparisons and analysis within the NHS. For example, we have written a shortNHS in Brief piece featuring the concentration of low-income by Census Tract. You must still be cautious, when analysing the smallest communities and sub-populations with low counts (though this is true in all surveys).
Friday, September 13, 2013 12:53:17

 sonya: Thanks Bryan. However, the link does not include the requested demographic and economic characteristics of the family. Would a custom data request to Stats Can allow me to capture what I need? Individual or family income is ok.
Friday, September 13, 2013 12:45:06

 Eric Olson: Hi sonya, Yes, a custom data request would certainly be able to answer that. A variable has been created at the individual level that would enable the construction of any groupings down to percentiles. It is possible, as well, to do that at the family level.
Friday, September 13, 2013 13:03:07

 Daphna Mills: Thank you for the opportunity to submit a question. This chat session is a great initiative. For NHS tables that show data for Nunavut, some data is available at the city level for Iqaluit (NHS Profile, Iqaluit, CY, Nunavut, 2011). Is it possible to choose Nunavut excluding Iqaluit (i.e., the rest of Nunavut) as a geographic region, and if so, how can I do this?
Friday, September 13, 2013 12:45:25

 Eric Olson: Hi Daphna, I hope you are well. The geography of Nunavut always includes Iqaluit in our standard products. For counts, you probably could derive the desired counts by subtracting the Iqaluit figure from a Nunavut total. However, your best bet is most likely to request a custom or semi-custom tabluation.
Friday, September 13, 2013 13:13:24

 PlanCannon: Low income prevalence as classified by LICO was instrumental to many programs in the municipality for which I work. How should we proceed, now that the LIM-AT measure is being used instead of LICO? Will low income prevalence as classified by LICO be available at a later date?
Friday, September 13, 2013 12:45:30

 Brian Murphy: Hello PlanCannon. The estimates for LICO prevalence will be ready next week. Please contact STATCAN.infostats-infostats.STATCAN@canada.ca to obtain the data. Thank you for your question.
Friday, September 13, 2013 13:04:00

 Josh Brandon: If 26% did not respond, how do you adjust for the non-responders? 440,000 households in Manitoba, given in the table "2011 National Household Survey: Data tables Tabulation: Shelter Cost Groups" e.g. seems a little low. Are there tens of thousands of households who did not respond, so were left off the table, or did you use some formula to slot in households as though they had responded?
Friday, September 13, 2013 12:48:29

 Jerry Situ: Hello Josh. Thank you for your question. Non-responding households are accounted for through weighting and calibration in the NHS database. So they are included in all tabulations.
Friday, September 13, 2013 13:01:50

 Angella: For income data by socio-economic detail, will we get a data reliability measure, such as the CV?
Friday, September 13, 2013 12:50:32

 Moderator: CVs for selected estimates from the NHS are planned to be released in the next few weeks.
Friday, September 13, 2013 12:51:10

 Sara Mayo: When is the anticipated release date for the technical report on sampling and weighing?
Friday, September 13, 2013 12:52:52

 Moderator: Sara Mayo, thank you for your question. Unfortunately, the subject is not relevant to the topic of this chat. We will follow up with you by email in the following business days.
Friday, September 13, 2013 12:53:39

 Sameer: Brian, you stated that you dont have DA level income data presently. Will this be released in the future? How about CSD level data in the future?
Friday, September 13, 2013 12:57:16

 Eric Olson: Hi Sameer, I do not expect any DA level income to be released in standard products. A researcher could request that data though. Univariate CSD-level data is available in the NHS profile. No cross-tabulations at the CSD level are planned. We could entertain suggestions if you feel there are specific areas that would be valuable to a broad variety of users.
Friday, September 13, 2013 13:06:51

 Annie McEwen: Given in incomparability of the NHS with the Census will CANSIM tables that used Census data be updated with NHS data?
Friday, September 13, 2013 12:58:06

 Moderator: Annie McEwan, thank you for your question. Unfortunately, the subject is not relevant to the topic of this chat. We will follow up with you by email in the following business days.
Friday, September 13, 2013 12:57:34

 Rose: Is data available at the FSA level?
Friday, September 13, 2013 13:05:05

 Moderator: Rose, thank you for your question. Unfortunately, the subject is not relevant to the topic of this chat. We will follow up with you by email in the following business days.
Friday, September 13, 2013 13:05:46

 Richard Stadlwieser: I am signed in but not receiving any video or audio It is welcoming me. Do you have some suggestion
Friday, September 13, 2013 13:06:46

 Moderator: This is a written chat session, not a video or audio chat. Once you are signed in, please scroll down the page to read user questions and our experts answers. Should you wish to ask questions, please input text into the "submit a question" box and select "submit" button.
Friday, September 13, 2013 13:08:49

 Angella: Eric, I'm sorry, I'm still not clear on the interpretation of item non-response for income. Is it 60% of the entire sample, or 60% of those that responded.
Friday, September 13, 2013 13:09:04

 Eric Olson: Of the entire sample. We could discuss this offline if it remains unclear.
Friday, September 13, 2013 13:15:47

 Redgoldgreen: Hi - thanks for doing this. Is the education level of the top 1% increasing? Is the same true for the lowest 10%? I believe it is the case for the middle range of income distribution.
Friday, September 13, 2013 13:10:58

 Brian Murphy: Hi Redgoldgreen. We did not look at the change in the level of education among the top 1% or among other groups. Thanks for your question.
Friday, September 13, 2013 13:15:35

 Annie McEwen: Subsequent to Josh Brandon's question - does this mean there was no imputation, item non-response led to case-exclusion on income tabulations, and adjustment was made by weighting responding cases? Is there a document detailing how the weights were derived (and from what data)?
Friday, September 13, 2013 13:11:04

 Eric Olson: Hello Annie McEwen. Imputation for item non-response did take place for all subject-matter areas. There is a document on weighting methodology planned for early 2014, but some information about imputation and weighting is already available in chapter 4 of the NHS User Guide.
Friday, September 13, 2013 13:24:30

 AlexandraT:
Hi, what is the point of a census?
Friday, September 13, 2013 13:15:23

 Modérateur:  AlexandraT, Thank you for your question. Unfortunately, this subject falls outside of the scope of this session. We will follow up with you by email in the next business days.
Friday, September 13, 2013 13:16:46

 Sara Mayo: The confirmation that no cross-tabulations for CDS will be prepared is devastating for municipalities. In 2006, there were 85 cross-tabs on the website for CSDs, so this is an enormous loss of data for the large number of community data users. I realise you can't comment on the impact, but I have two questions I hope you can answer: 1) Are you planning to release any CD level cross tabs on the website? 2) Do you consider CDSs, i.e. municipalities to be "small areas". I always thought CTs and DAs where small areas, but CSDs were not.
Friday, September 13, 2013 13:21:31

 Eric Olson: To clarify: in my original response, I was considering only income-specific tables. 1) In addition to the CSD table already on the website, there will likely be some additional CSD-level data tables released before the end of the year. 2) Some CSDs are small areas, others are larger. My comments on small areas were based on their population, not their geographic name.
Friday, September 13, 2013 13:30:15

 tjcraig: Good afternoon, I am doing some research on housing costs and subpopulations. Is there a means in which to correlate data sets with the census (i.e those who spend more than 30% on shelter costs and recent immigrants)?
Friday, September 13, 2013 13:27:30

 Jerry Situ: Thank you for your question. The NHS has numerous variables on the characteristics of the population, including immigrant status. These variables can be cross-tabulated in the NHS without relating it to the Census database. I invite you to take a look at our data tables to see what cross-tabulations are available in our standard products: 2011 National Household Survey: Data tables
Friday, September 13, 2013 13:33:18

 Moderator: Thank you for all your questions and comments! It is 1:33p.m. EDT, which means that the chat session is now over. If the experts didn't have a chance to respond to your question, they will follow-up by email in the next few days. The full transcript of this chat session will be made available on the website shortly. Have a great day!
Friday, September 13, 2013 13:33:22

Date modified: