1. Getting Started
This guide is primarily designed to help users unfamiliar with CANSIM database find and download data.
a. What is CANSIM?
CANSIM is Statistics Canada’s main socioeconomic time series database. It contains most of the aggregate data collected by Statistics Canada on a regular basis such as data from the Consumer Price Index Survey, the Labour Force Survey, or the National Income and Expenditure accounts.
b. How to access CANSIM?
There are two ways for University of Toronto faculty, staff, and students to download CANSIM series for free. The first one is through CHASS:
- Go to mdl.library.utoronto.ca;
- Under the Statistics menu, select Finding Canadian statistics;
- Under the Searching for Canadian statistics menu, select CANSIM.
- Choose CANSIM multidimensional @CHASS.
- Choose the language and select CANSIM Multidimensional @ CHASS Main Menu.
Note that to access CHASS, you need a University of Toronto IP address.
The other option is to go directly on Statistics Canada’s website www5.statcan.gc.ca/cansim. Since February 1st, 2012, CANSIM data are freely available. The examples below are based on a search on Statistics Canada’s website.
2. Searching for Data
Data are classified by tables and series. A table is the largest unit and contains many series. For instance, there are over 125 tables from the Labour Force Survey. One of these is called “Labour force survey estimates, by National Occupation Classification for Statistics and sex (annual)” (Table 282-0010) and contains over 6700 series. A series is, for instance, the full-time employment of male teachers and professors in Canada (series v2369249).
b. Searching by survey, table, or series number
If you have the name or number of a survey, table, or series, you can easily retrieve the series by typing the information in the search menu.
c. Searching by subject or keywords
Both CHASS and Statistics Canada’s interface allow the user to search for series by subject or keywords. However, unless you know the exact name of the series you are looking for, it is usually better to search by subject in the first stage because the concept we have in mind may not be in the title of the right table or series. For instance, to find the
Suppose we want to find the quarterly Canadian CPI-based inflation rate for the period 1995-2010.
- From CANSIM main page, select Browse CANSIM by Subject (default).
- From the list, choose Prices and price indexes. There are 247 tables on prices and price indexes. The link redirects us to a more precise list of topics.
- Since we are interested in the CPI (consumer price index), select Consumer price indexes.
- The Consumer Price Index statistics is the first table. To access the series, press on the table number (176-0003) to open the CPI table.
Once you have found the relevant table, you are ready to download a series.
3. Playing with Data
We opened Table 17-0003, “Consumer Price Index (CPI) statistics, alternative measures, unadjusted and seasonally adjusted” and are interested in finding the quarterly inflation rate for Canada from 1995 to 2010.
By default, all the active series in the table are selected. To choose which series to download, we go under the tab Add/Remove data.
- Step 1 – Select: Geography. The series are only available for all Canada. We have nothing to do.
- Step 2 – Select: Alternative measures. Different measures of the CPI are available and some series are not collected anymore (indicated by (Terminated)). Since we are interested in the simple inflation rate, select the series Consumer Price Index (CPI) excluding the effect of indirect taxes (index, 2002=100).
- Step 3 – Select the time frame. We want data from Jan 1995 to Dec 2010.
- Step 4 - Select the Screen output format. To easily visualize the data, choose HTML tables, time as rows from the drop-down menu.
- Step 5. Click on Apply to visualize the series
We have the monthly CPI from January 1995 to December 2010, but we are interested in the quarterly inflation rate. To manipulate the data, go under the Manipulate tab.
- Select Quarterly (average) from the first drop-down menu (“Frequency of output data will be”), and Use calendar year from the second drop-down menu (“When converting frequency by summing or averaging”).
- To transform the price index into an inflation rate, select Percent changes, year-to-year from the third drop-out menu (“Data table will contain”).
- Click on Apply to vizualize the series.
To download the data, go under the Download tab.
- Use “Option 1 – Download data as displayed in the Data table tab”.
- Select the language, the output format type (time as rows), the file format, and click on Download data. The output will be a database that can be opened in Excel. To download series in another format (e.g. SPSS, SAS) use CHASS.
4. A more complicated example
Suppose we want the growth rate of real GDP in addition to the inflation rate. These series come from different tables, so we have to use a different approach.
- From CANSIM main page, select Combine series (under the “Search CANSIM” tab.)
- Select Browse by subject.
- As we did before, choose Prices and price indexes, Consumer price indexes, and then select the first table (176-0003).
- The Geography, Alternative measures, and Reference period are the same as in the previous example. Select Canada. Consumer Price Index (CPI) excluding the effect of indirect taxes (index, 2002=100), and From: Jan 1995 To: Dec 2010.
- Click on Retrieve as individual time series.
- To add the GDP series, click on Add more series.
- Click on Browse by subject, and select Economic accounts.
- From the “Economic accounts” menu, choose, Gross domestic product.
- We will use the table titled “Gross domestic product (GDP) expenditure-based, quarterly (Dollars)”. Click on the table number (380-0002).
Select the options:
- Geography: Canada
- Prices: Chained (2002) dollars
- Seasonal adjustment: Seasonally adjusted at annual rates
- Reference period: From Quarter 1 1995 To Quarter 4 2010.
- Click on Retrieve as individual time series
- We have the two series we are interested in, so click on Continue.
- Choose the output format you want from the list. To download a file, choose CSV file, time as rows under DOWNLOADABLE FILE formats.
- Click on Manipulate Data to convert the monthly price index into a quarterly series, and to express both variables in year-to-year percent changes.
Select the options:
- Frequency of output data will be: Quarterly (average)
- When converting frequency by summing or averaging: Use calendar year
- Output will contain: Percent changes, year-to-year
- Click on Apply to download the series.