View the most recent version.

Modified on November 10, 2009

Education Reference Guide,
2006 Census

Catalogue no. 97-560-GWE2006003

• Definitions and concepts
• Changes in the 2006 Census questions and concepts
• Change in the mode of collection for 2006
• Questions
• Classification
• Data
• Data quality
• Comparison with other data sources
• Historical comparability

Education information is collected in the census because education plays a major role in Canadian society. Education has an impact on the quality of life of Canadians (in areas such as income, work, health and participation in the community) and the knowledge and skills of the Canadian labour force have an impact on Canada's overall economic performance.

Definitions and concepts

The 2006 Census measured four main concepts through its education questions: completed education credentials: certificates, diplomas and degrees (questions 26 to 29), major field of study (Question 30), location of study (Question 31), and attendance at school (Question 32).

Within these concepts, there are six analytical variables that have been released: Highest certificate, diploma or degree (see section below on Data Quality); Attendance at school (see section below on Data Quality); Major field of study - Classification of Instructional Programs (see section below on Data Quality); Location of study; High school certificate or equivalent (see section below on Data Quality); and Degree in medicine, dentistry, veterinary medicine, or optometry.

Changes in the 2006 Census questions and concepts

The Education questions on the census have undergone significant changes since 2001. These changes were motivated by changes to the educational profile of the Canadian population and to the education system itself, changing demands in the Canadian labour market, data quality concerns from the 2001 Census and availability of other data sources for education.

The variable that measures 'highest certificate, diploma or degree' has been divided from one question into four separate questions (high school, trades/apprenticeship, college, university). More detailed information has been collected on the type of high school certificate (diploma or equivalency), the type of trades certificate or diploma (Registered Apprenticeship or other trades certificate or diploma) and the duration of college programs.

A new question on the 'location of study,' added to the 2006 Census, collects information on the province, territory or country where the highest postsecondary certificate, diploma or degree was completed. Consultations indicated that it was important to understand interprovincial and international flows of the population with education beyond high school. This is particularly true with increasingly highly-qualified Canadian-born and immigrant populations.

The concept of 'years of schooling' is not available from the 2006 Census. While recognizing the importance of the years of schooling variable (as a continuous variable) to advanced statistical analysis, some of the years of schooling questions in the 2001 Census are not as pertinent today as they were when they were first included in the census. In 2001, 75% of the population aged 15 years and over had completed high school. Of those who had not completed high school, about 75% had completed at least grade 10 and the majority (60%) were over the age of 50 years. This made information on the years of schooling at the elementary and secondary levels redundant for the majority of the population and shifted the focus to high school completion.

In addition, qualitative testing of the years of schooling questions indicated that respondents were often confused as to whether to report the number of years it actually took them to complete their program or how long it would normally take to complete the program as a full-time student. This was particularly true at the postsecondary level where students can combine full-time and part-time attendance or have extended periods of non-attendance before a program is completed.  Overall, consultations with stakeholders indicated that the years of schooling questions were not as important as increasing the detail on completed qualifications in view of the limited space available on the census questionnaire.

Consequently, the concept of 'some postsecondary education' is not collected in 2006, with the focus being instead on completed qualifications rather than participation. The removal of the questions on years of schooling also means that the variable 'highest level of schooling,' as defined in 2001, cannot be derived from the 2006 Census. Data for the concept of 'some postsecondary' and 'highest level of schooling' are available from the Labour Force Survey. For more information on the Labour Force Survey, please see the following link.

The focus for the concept 'attendance at school' has changed from full-time/part-time attendance to type of school attended. The variety of attendance patterns possible today made it difficult for respondents to fit themselves into the full-time, part-time response categories, especially given the nine-month reference period. Question 32 on attendance at school now has categories for type of school attended (high school, college, CEGEP, other non-university institution and university).

The 'attendance at school' variable was made available on March 4, 2008, in limited release, due to some concerns with the data. Initial analysis had pointed to important discrepancies between the 2006 data and data from previous censuses, as well as to important variations with the Labour Force Survey. The analysis indicates that the 2006 Census data overestimated attendance at school for the population aged 45 years and over.

As a result, data on attendance at school were not included in any of the various standard or specialized data products, nor in the online analytical document on education released on March 4, 2008. Upon subsequent historical analysis, data for attendance at school (with a caution for those aged 45 and over) were released. Standard tables using an aggregate variable (focusing on total attendance) were released on October 28, 2008.

Data on full- or part-time attendance at school are available from the Labour Force Survey for the population aged 15 to 64. The Labour Force Survey, with a reference period of ‘last week' and calculations of attendance at school across academic months, is a source of information on full-time/part-time attendance at school.

For more information on the Education questions for the 2006 Census, please see the document Education questions on the 2006 Census of Population: New questions for a new century. Also see the document 2006 Census New Education Module: follow-up report, Catalogue No. 92-133-XWE.

Change in the mode of collection for 2006

For the first time, the 2006 Census offered all households in Canada the option of completing their census questionnaire online. This secure and convenient option was available in English and French.

For more information on the census questionnaire online, please see this link.

Questions

The Education questions for private households are found in the 2B Questionnaire of the census.  The 2B questionnaire was used to enumerate a 20% sample of all private households in Canada. To assist people whose first language was neither English nor French, the census questions were translated into 62 other languages, including 18 Aboriginal languages.

Questions 26 to 29 collected information on all certificates, diplomas and/or degrees that the respondent had completed from secondary/high school; trades; college, CEGEP, other non-university institutions; or university. The response categories collected detailed information about the level of each completed certificate, diploma, or degree. Question 30 collected information on the major field of study of the highest certificate, diploma or degree and Question 31 asked about the province, territory or country where the highest certificate, diploma or degree was completed.  Question 32 collected information on whether the person attended school at any time in the nine months previous to Census Day and what type of school was attended.

The Education questions appear on pages 16 to 19 of the 2B (long) questionnaire.

The 2006 Census also collected information from persons living in private households on Indian reserves, Indian settlements and in remote areas and enumerated them with the 2006 Census Form 2D questionnaire. The 2D Education questions were the same as the 2B questions.

For more information on all the questions included in the 2006 Census, please refer to 2006 Census questions and reasons why the questions are asked.

Classification

The field of study of the highest certificate, diploma or degree of the respondent was first classified using the Classification of Instructional Programs, Canada 2000 (CIP [Canada 2000]). The CIP Canada 2000 consists of about 1,400 categories; programs are assigned a six-digit code at the most detailed level. In order to allow comparisons with previous censuses, the variable was also classified a second time using the Major Field of Study (MFS) classification. The MFS classification, created in the mid-1980s, was used in the 2001 Census and in previous years and consists of about 450 categories.

The CIP Canada 2000 was adopted for several reasons: it includes new and emerging fields of study; its hierarchical taxonomy allows for more accurate tracking, assessment and reporting of field of study data; and it allows for enhanced data compatibility between Mexico, the U.S. and Canada.

Both theoretical and empirical concordance tables are available that provide mappings of the relationships between the CIP Canada 2000 and the Major Field of Study classification. The theoretical concordance is definitional in nature and provides users with a means of summarizing the similarities and differences between the two systems. However, users are cautioned that this type of concordance cannot be used to convert counts from one classification system to another. For more information on the CIP Canada 2000 – Major Field of Study theoretical concordance, please see this link (Appendix O in the 2006 Census Dictionary).

The empirical concordance tables provide proportional breakdowns of both the CIP components of individual MFS codes as well as the MFS components of individual CIP codes, based on classification of 2006 Census responses. These tables can be used to produce approximations of CIP fields of study in censuses prior to 2006, or to produce approximations of MFS counts in 2006 and beyond. For more information on the CIP Canada 2000 – Major Field of Study empirical concordance, please see the entry 'Major Field of Study – Classification of Instructional Programs (CIP) and Major Field of Study (MFS) empirical concordance' of the 2006 Census Dictionary.

A concordance between the International Standard Classification of Education and the Classification of Instructional Programs, Canada 2000 has been developed by Statistics Canada.

Please see Appendix R of the 2006 Census Dictionary for more information on the International Standard Classification of Education (ISCED) – Classification of Instructional Programs Canada, 2000 (CIP Canada 2000) concordance. This information will be updated, as necessary, on Statistics Canada's website.

Data

Education data are reported for all private households, for persons aged 15 years and over, excluding institutional residents.

Census education data can be found on the census website at the following links:

2006 Census

• Highlight tables, 2006 Census
  
• Topic-based tabulations, 2006 Census
• Profiles at each level of geography, 2006 Census
  • Cumulative profile and release components
  • Print profile series ($)
  • 2006 Community Profiles
  • Census Tract (CT) Profiles, 2006 Census
  • Aboriginal Population Profile
  • Federal electoral district (FED) profile
  • Special interest profiles

Census Trends, 2006 Census

2001 Census

1996 Census

Users should be prudent when comparing 2006 Census data with previous census years as many variables are new for 2006 or have changed and some may not have retained comparability over time. Please see the section below on data quality for more information.

Data quality

Imputation for non-response and inconsistent responses

Throughout the process that culminates in the release of census data, considerable effort is made to maintain high standards of data quality. Nevertheless, some degree of inaccuracy is inherent in the data that are produced. Assessment of the education variables consists of the following steps:

• examination of non-response rates
• comparison of the distributions of edited and unedited data to determine whether imputation has introduced bias
• comparison of the distributions by response mode (Internet versus paper)
• chronological comparison with data from previous censuses (this comparison is limited because of the new content in the education module)
• comparison with other data sources, as appropriate.

To assess the usefulness of census data for their intended purposes and to understand the risk involved in drawing conclusions or making decisions based on the data, users need to be aware of the following data quality indicators for the education variables.

Examination of non-response rates, comparison of edited and unedited data, and comparison of data by response mode

One of the important steps in assessing data quality is to look at the data before and after imputation has been applied to ensure that there are no differences revealed which could introduce bias. The examination of the distributions before and after the imputation of non-response showed that there were no differences and that imputation has not introduced bias to the data except in the case of Question 30 on Major field of study. Please see Question 30 below for more information on this.

Question 26: Secondary (high) school diploma or equivalent

• For Canada, the non-response rate for this question was 2.3%, which is similar to the rates observed for the language, mobility and Aboriginal questions. At the provincial level, the rate was higher in the Western provinces (4.0% in Saskatchewan, 3.3% in Manitoba, 3.1% in British Columbia and 2.9% in Alberta). The non-response rate was below the national average in the other provinces. For the territories, it was higher than the national average (5.4% in Nunavut, 4.6% in Yukon Territory and 3.5% in the Northwest Territories).

• For Canada, 19.4% of respondents who used the Internet reported that they did not have a secondary (high) school diploma or equivalent, compared with 27.4% of respondents who completed a paper questionnaire. The non-response rate for this question was higher among those who responded on paper (1.5%) than among those who responded via the Internet (0.6%).

Question 27: Registered Apprenticeship or other trades certificate or diploma

• For Canada, the non-response rate for Question 27, on Registered Apprenticeship or other trades certificate or diploma, was 3.1%, slightly higher than the rate for the question on secondary school diploma or equivalent. As in Question 26, the provincial non-response rates were above the national average in the Western provinces (4.5% in Saskatchewan, 4.1% in British Columbia, 3.9% in Manitoba and 3.5% in Alberta). The rate was below the national average in the other provinces. For the territories, it was higher than the national average (6.5% in Nunavut, 5.3% in Yukon Territory and 4.1% in the Northwest Territories).

• The proportion of people with a Registered Apprenticeship or other trades certificate or diploma was slightly higher among those who completed a paper questionnaire (19.1%) than among those who responded via the Internet (17.3%). The non-response rate for this question was higher among those who responded on paper (2.6%) than among those who used the Internet (0.6%).

Question 28: College, CEGEP or other non-university certificate or diploma

• For Canada, the non-response rate for Question 28, on college, CEGEP or other non-university certificate or diploma, was 3.1%, the same as the rate for Question 27 and slightly higher than the rate for the question on secondary school diploma or equivalent. As in the case of Questions 26 and 27, the provincial non-response rates were above the national average in the Western provinces (4.5% in Saskatchewan, 4.2% in British Columbia, 3.9% in Manitoba and 3.6% in Alberta). The rate was below the national average in the other provinces. For the territories, it was higher than the national average (6.1% in Nunavut, 5.3% in Yukon Territory and 3.9% in the Northwest Territories).

• The proportion of people with a college, CEGEP or other non-university certificate or diploma was higher among those who used the Internet (31.5%) than among those who responded on paper (24.9%). As in the previous questions, the non-response rate was higher among those who completed a paper questionnaire (2.7%) than among those who responded via the Internet (0.6%).

Question 29: University certificate, degree or diploma

• For Canada, the non-response rate for Question 29, on university certificate, degree or diploma, was 2.5%, very close to the rate for Question 26 and slightly lower than the rates for Questions 27 and 28. As in the previous questions, the provincial non-response rates were above the national average in the Western provinces and the territories (3.6% in Saskatchewan, 3.4% in British Columbia, 3.1% in Manitoba, 2.9% in Alberta, 4.5% in Yukon Territory, 4.2% in Nunavut and 3.0% in the Northwest Territories). The rate was below the national average in the other provinces.

• The proportion of people with a university certificate, degree or diploma was higher among those who responded via the Internet (30.4%) than among those who used a paper questionnaire (21.0%). As in the previous questions, the non-response rate was higher among those who responded on paper (2.0%) than among those who used the Internet (0.5%).

Question 30: Major field of study of the highest degree, certificate or diploma

• For Canada, the non-response rate for Question 30, on major field of study, was 5.5%, higher than the rates for the questions on educational qualifications. In general, it is not surprising for open questions to have higher non-response rates than multiple-choice questions (which can be seen with other questions such as Labour). In contrast to the questions on educational qualifications, the highest provincial non-response rates were not all in Western Canada. Quebec had the highest rate (6.4%), followed closely by British Columbia (6.3%). In the territories and other provinces, the rates hovered around 5.0%.

• Since the question on major field of study was also asked in 2001, a historical comparison of response rates is possible. The non-response rate for this question was 2.5% in 2001. However, the sequence of questions was changed completely, which makes the comparison more difficult. In 2001, the question on major field of study came immediately after – and on the same page as – the question on educational qualifications, which was not the case in 2006.

• The question on major field of study is an open question. The responses were coded first with a new classification (Classification of Instructional Programs (CIP), Canada 2000). Of the 2,530,686 write-in responses to this question, 72.4% were coded automatically from a reference file that contained the most common responses in previous censuses. The remaining write-ins (27.6%) had to be processed by subject-matter coders. The responses were then coded a second time using the Major Field of Study (MFS) classification. The MFS reference file used the same list of common responses from previous censuses as the CIP reference file, resulting in an identical automatic coding rate for MFS as for the CIP Canada 2000 classification. However, in the case of MFS, the remaining write-ins were coded to the MFS classification on the basis of an ID number assigned during manual CIP coding that associated each write-in with its nearest match from the reference file.

• As noted above, the reference file of common major-field-of-study responses from previous censuses contains both MFS and CIP codes for each individual write-in. However, there are a number cases in the reference file in which a single write-in has MFS and CIP codes that are conceptually different. Two main reasons for this are: (1) differing strategies between classifications for the coding of joint major or multiple fields of study and (2) differing coding interpretations under the two classification systems of insufficiently specific write-ins. In order to preserve the historical comparability of the MFS coded variable, MFS codes of individual write-ins on the reference file were not revised to reflect the new CIP interpretation. For more discussion of MFS-CIP comparability, please see Major Field of Study (MFS) in the 2006 Census Dictionary.

• A study was carried out to measure coding quality. A sample of some 3,000 entries was recoded using the CIP Canada 2000 classification, and the overall accuracy rate was found to be 96.0%.

• Examination of the distributions before and after non-response imputation revealed a few differences. It appears that fields of study requiring a Registered Apprenticeship or other trades certificate or diploma, or a college, CEGEP or other non-university certificate or diploma were slightly overrepresented in the non-responses, and as a result, those fields of study make up a slightly larger percentage after imputation. This situation is not unusual and was also noted in the 2001 Census.
• The proportion of people with a postsecondary education was higher among those who responded via the Internet than among those who used a paper questionnaire. An examination of the data on major field of study by broad group shows no substantial difference by response mode. The non-response rate was slightly higher among those who responded on paper (6.1%) than among those who used the Internet (4.4%).

Question 31: Location of study

• For Canada, the non-response (or invalid response) rate for Question 31, on location of study, was 9.3%, appreciably higher than the rates for other questions in the module. It is important to note that this question follows the question on major field of study (Question 30), which instructs respondents who do not have a postsecondary certificate or diploma to skip Question 31 and go directly to Question 32. It appears that many respondents who were in-scope for Question 31 did not answer it. The provincial non-response rates were above the national average in British Columbia (10.7%), Manitoba (10.7%), Saskatchewan (10.6%), and Quebec (9.7%). For the territories, only Nunavut had a non-response rate (14.1%) higher than the national average.

• The question on location of study is an open question. Of the 2,772,064 write-in responses to this question, 95.6% were coded automatically from a reference file. The remaining write-ins (4.4%) had to be processed by subject-matter coders.

• The proportion of people who studied outside Canada was higher among those who responded via the Internet (18.1%) than among those who used a paper questionnaire (13.9%). The non-response rate was slightly higher among those who responded on paper (5.0%) than among those who used the Internet (3.8%).

Question 32: Attendance at school

• For Canada, the non-response rate for Question 32, on attendance at school, was 4.0%, slightly higher than the rates for the questions on educational qualifications. The provincial non-response rates were above the national average in Quebec (6.4%) and the Western provinces (5.8% in Saskatchewan, 5.0% in Manitoba, 4.9% in British Columbia and 4.6% in Alberta). The non-response rate was below the national average in the other provinces. For the territories, it was higher than the national average (7.7% in Nunavut, 5.7% in Yukon Territory and 5.3% in the Northwest Territories).

• The proportion of people who reported attending school in the previous nine months was higher among those who responded via the Internet (21.2%) than among those who used a paper questionnaire (16.3%). As in the questions on educational qualifications, the non-response rate for this question was higher among those who responded on paper (4.0%) than among those who used the Internet (0.9%).

Comparison with other data sources

As with every census, the quality of the 2006 Census education information released was evaluated internally prior to publication. The data were compared, as much as possible, with alternative data sources. The main source of comparison for the education module was the education estimates from the Labour Force Survey for the various variables.

2006 Census counts for the education variables are generally at expected levels, based on 2001 Census results and information obtained from the Labour Force Survey. However, data users should be aware of the following issues (presented by concept rather than by question).

University certificate below bachelor level

For the first time, in the 2006 Census, there was an explicit question on completions of university certificates, diplomas or degrees.  In 2001, Canadians were asked 'What certificates, diplomas or degrees has this person ever obtained?' and one of the answer categories was 'University certificate or diploma below bachelor level.' In 2006, the question was changed to 'Has this person completed a university degree, certificate or diploma?' and one of the answer categories was 'Yes, certificate or diploma below bachelor level.'

A large increase in the category ‘university certificate below bachelor level’ was noted in 2006 compared to 2001. The number of adults aged 15 and over who reported having this as their highest credential increased 89% between 2001 and 2006, while the corresponding number in the Labour Force Survey increased by only 5%. In 2001, 2.5% of respondents aged 15 and over declared this as their highest credential, compared to 4.4% in 2006. The Labour Force Survey reported 2.8% in 2001 and 2.7% in 2006. It is suspected that the new design of the 2006 Census questionnaire contributed to significant overreporting of this category.

The data for 'university certificate or diploma below a bachelor level' were published on March 4, 2008, but for the reasons outlined above, the data were not discussed in the analytical report. It is recommended that users not compare this category in 2006 with previous censuses.

Attendance at school

Census data indicate that between 1996 and 2001, there was a marked decrease in school attendance among persons aged 15 to 19 (full-time attendance decreased from 78.5% to 73.4% in this period).  No satisfactory explanation for this decrease was found. As a result, changes were implemented for the 2006 Census and a new version of the school attendance question focusing on the type of educational institution attended was developed.

Further analysis of the 2006 attendance at school variable suggests that the new question was successful in addressing the data quality issues in attendance data for this age group.

The proportion of those aged 15 to 19 reporting having attended school at any time over the nine months previous to Census Day 2006, was 80.6%, an improvement from 2001.

As well, the trends in school attendance for this age group between 1971 and 2006 observed in the Labour Force Survey and the census are generally comparable.

However, it appears that the 2006 Census could have overestimated the proportion of those attending school for the population aged 45 years and over.

For this age group, school attendance was 5.0% in the 2006 Census compared to 2.7% in the 2001 Census, for a difference of 2.3 percentage points. This represents a significant increase of the attendance at school rate for this age group which was not observed in other data sources.

It is recommended that users of the attendance at school variable interpret the 2006 Census results for the population aged 45 years and over with caution and not compare the 2006 Census data for this age group with previous censuses.

Historical comparability

As previously mentioned, the education module in the 2006 Census questionnaire underwent significant changes, including new questions and conceptual changes to existing questions. As a result, there is an impact on historical comparability for some questions. This section outlines the impacts of which users should be aware.

Highest Certificate, Diploma or Degree (presented by category)

No certificate, diploma or degree (less than high school)

Historically, the proportion of Canadians with no certificate, diploma or degree (less than high school) has been higher in the Census than in the Labour Force Survey. In 2006, changes were made to the Census questionnaire to address suspected underreporting of high school completions.

Based on Census results, the proportion of Canadians aged 15 and over with less than high school showed a sizable decrease from 33.2% in 2001 to 23.8% in 2006. This decrease was more pronounced than in the Labour Force Survey where the rate went from 28.8% to 24.9% over the same period.

The convergence of results between the Census and the Labour Force Survey in 2006 is attributed in large part to changes made to the Census questionnaire in 2006. When making historical comparisons, users should be aware that underreporting of high school completions contributed to elevated results for this category in censuses prior to 2006.

High school diploma or equivalency certificate

Historically, the proportion of Canadians with a high school certificate or equivalency as their highest level of educational attainment has been lower in the Census than in the Labour Force Survey. In 2006, changes were made to the Census questionnaire to address suspected underreporting of high school completions.

Based on Census results, the proportion of Canadians 15 years and over with high school as their highest level of education increased from 23.0% in 2001 to 25.5% in 2006. By comparison, the similar rate in the Labour Force Survey went from 26.9% to 26.8% over the same period.

The diminishing gap in results between the Census and the Labour Force Survey is attributed in part to changes made to the Census questionnaire in 2006. When making historical comparisons, users should be aware that underreporting of high school completions has affected the data for this category in censuses prior to 2006.

Apprenticeship or trades certificate or diploma

For the first time in 2006, there was an explicit question pertaining to completions of trade certificates and diplomas.

The overall rates of completion in the trades in the Census have remained steady over the last twenty-five years. While rates of trades certification (as the highest credential attained) declined slightly in all provinces from 2001 to 2006, the proportion aged 15 and over in Quebec increased from 10.8% to 15.3%. By comparison, the similar rate in Quebec in the Labour Force Survey went from 11.5% to 13.6%.

A change was made in the 2006 Census questionnaire to specifically include certification by a type of educational institution found only in Quebec, the ‘centres de formation professionelle’, and this may have influenced the data on trade certifications in Quebec.

The components of this category that specify the type of trades certificate obtained were newly introduced in the 2006 Census and therefore cannot be historically compared.

College, CEGEP or other non-university certificate or diploma

The total category for college, CEGEP or other non-university certificate or diploma is comparable over time.

The components of this category that specify the duration of the program of study were newly introduced in the 2006 Census and therefore cannot be historically compared.

University certificate below bachelor level

A large increase in the category ‘university certificate below a bachelor level’ was noted in 2006 compared to 2001. It is recommended that users not compare this category in 2006 with previous censuses because of the issues described in the section above on “Comparison to other data sources”.

University certificate or degree (Bachelor’s degree, Certificate or diploma above bachelor’s level, Degree in medicine, dentistry, veterinary medicine or optometry, Master’s degree, Earned Doctorate)

Questions pertaining to university degrees attained in 2006 (for example bachelor's degrees or master's degrees) were similar to those asked in 2001.

Data for the university categories (Bachelor’s degree through to Earned doctorates) are comparable over time.

Secondary (High) School Certificate or Equivalent

Historical comparisons using the secondary (high) school graduation variable are complicated by changes to the Census questionnaire and the consequent impacts of these changes on respondent behaviour.

As a result of changes to the questionnaire, the secondary (high) school graduation variable consists of different categories in 2006 than in 2001, including categories that appear similar between years based on their labels, but are in fact different based on their underlying definition and derivation. Specifically, the variable in censuses prior to 2006 accounts for whether an individual has some postsecondary education ('further training – no certificate') with or without high school certificate or equivalency, while the variable in 2006 considers only whether the individual has completed a postsecondary certificate, diploma or degree – with or without a high school certificate or equivalency.

Furthermore, for the first time in 2006, the Census included an explicit question asking Canadians aged 15 and over to report whether they had completed high school, with a separate category for completion of high school equivalencies. This change was made to address suspected underreporting of high school completions. In previous censuses, respondents were asked to select all levels of education they had completed from a single list, with the result that many respondents reported only their highest credential rather than all credentials earned.

As evidence of this tendency, the proportion of those aged 15 and over who had a trades certificate but no high school diploma decreased 45% from 2001 to 2006. Conversely, the rate of those having a trades certificate and a high school certificate or equivalent increased 51% in the same time period. The increase in those with a high school certificate and a trade indicates that respondents are more likely to report the presence of their high school diploma when it is asked in a separate question (as in 2006). 

Another indicator of this tendency is seen when comparing the data for the total population 15 and over with a high school certificate (regardless of whether or not they had completed further studies) in the Census and the Labour Force Survey. The proportion of those with a high school certificate (regardless of further studies) has historically been much lower in the Census than in the Labour Force Survey. However, in 2006 when the explicit question for high school completion was included in the Census (as it is in the Labour Force Survey), the results of the two sources converge. According to the Census, 58% of Canadians aged 15 and over had completed a high school certificate in 2001, increasing to 73% in 2006. The corresponding proportions in the Labour Force Survey are 68% in 2001 and 72% in 2006.

Due to changes to the questionnaire and the derivation of the variable as described above, only six of the categories in 'high school certificate or equivalency' may be considered for comparison with previous censuses:

• 'Without high school certificate or equivalent – without further schooling' (2006) is conceptually comparable to the sum of 'Without high school graduation certificate – without further training' and 'Without high school graduation certificate – with further training: no certificate, diploma or degree' (2001). While the consistency of the concepts between censuses permits the calculation of historical comparisons, users should be aware that underreporting of high school completions in censuses prior to 2006 contributed to elevated results for this category in censuses prior to 2006. Based on Census results, the proportion of Canadians aged 15 and over without a high school diploma or equivalent and without further schooling showed a sizable decrease from 33.2% in 2001 to 23.8% in 2006. This decrease was more pronounced than in the Labour Force Survey where the rate went from 28.8% to 24.9% over the same period. The convergence of results between the Census and the Labour Force Survey in 2006 is attributed in large part to changes made to the Census questionnaire in 2006.

• The five university categories which consist of: 'With high school certificate or equivalent – with a university certificate, diploma or degree' – from a Bachelor's degree to an Earned doctorate (2006) are comparable to 'With high school graduation certificate, with university certificate, diploma or degree' from a Bachelor's degree to an Earned doctorate (2001). The university categories are comparable because questionnaire changes resulted in only minimal impact on highest university credentials and editing rules were applied similarly in 2001 and 2006. Specifically when a university certificate, diploma or degree was reported without a high school certificate, one was assigned.

Major Field of Study

For the first time with the 2006 Census, major field of study data were coded with the Classification of Instructional Programs (CIP), Canada, 2000.

Prior to the 2006 Census, the Major Field of Study Classification (MFS) was used to classify major field of study. We recommend that users not make historical comparisons between the MFS and CIP classification systems on the basis of field of study descriptions. Even though some entries in the two classifications are similar, direct comparison would be inappropriate, given the much more detailed character of the new classification.

A theoretical concordance table between the Classification of Instructional Programs (CIP) and the Major Field of Study Classification (MFS) showing the definitional relationship between the two classifications was developed. This table is available in the 2006 Census Dictionary (Appendix N). This type of concordance allows users to see the relationship between the two classifications based on the definitional aspects of each. However, users are cautioned that this type of concordance cannot be used to convert counts from one classification system to another.

In order to make historical comparisons possible, the 2006 major field of study variable was recoded using the MFS classification that was used in censuses prior to 2006. This allowed the creation of empirical concordance tables that enable field of study comparisons over time by recoding 2006 data with the MFS classification, or previous censuses with the CIP Canada 2000 classification. For more information on the CIP Canada 2000 – Major Field of Study empirical concordance tables, please see Appendix S and T in the 2006 Census Dictionary.

While versions of the field of study variable coded to MFS at the most aggregated level (13) are deemed historically comparable (see note regarding Coding change – Removal of 'M482 – No specialization'), the following changes to the questionnaire and coding practices have influenced the variable at the most disaggregated level.

Questionnaire change – Request to 'Please be specific':  Added to the major field of study question in 2006 was text requesting that respondents 'Please be specific' when reporting their major field of study. The general impact of this change was to increase the tendency of respondents to report specific branches of a given field of study and lessen reporting of more general fields of study. For example, in the field of engineering, counts of individuals whose major field of study was in narrow fields of specialization such as 'M281 Instrumentation engineering', 'M282 Power engineering', 'M289 Marine engineering' and 'M270 Aeronautical and aerospace engineering' recorded growth rates from 2001 to 2006 of over 100% while the much larger and more general field 'M293 Engineering n.e.c.' recorded a 22% decline due to decreased reporting of 'Engineering'. When looking at trends in any field of study at the most disaggregated level, users are advised to consider this shift in reporting behaviour and look at trends in any specific field in the context of the larger field of study to which it belongs.

Questionnaire change – Change to write-in examples: The major field of study question on the census questionnaire includes a number of examples of typical fields of study. The presence of these examples, while beneficial in clarifying the nature and specificity of the response desired, leads to a degree of response bias as respondents are more likely to respond with one of the examples provided. In the 2006 Census questionnaire, a new set of examples of typical fields of study was introduced. The general impacts of this change in 2006 were: (a) to lessen the tendency to respond using a 2001 write-in example and, conversely, (b) to increase the tendency to respond using one of the new 2006 write-in examples.

The following list the fields of study used as examples in both 2001 and 2006:

Fields of study appearing as write-in examples in the 2001 Census questionnaire

• Accounting
• Carpentry
• Civil engineering¹
  
• History
• Legal secretary
• Welding

Fields of study appearing as write-in examples in the 2006 Census questionnaire

• Automobile mechanics
• Civil engineering¹
  
• Dental technology
• Aircraft mechanics
• Medical laboratory technology
• Day-care
• Agricultural economics

Fields of study appearing as examples of Registered Apprenticeship or other trades certificates or diplomas OR college, CEGEP or other non-university certificates or diplomas in the 2006 Census questionnaire

• Hairdressing
• CNC machinist
• Accounting technology
• Real estate agent
• Industrial engineering technology

The presence of two specific mechanic examples in 2006, along with the request to 'Please be specific' resulted in a significant decline in reporting of 'Mechanic', leading to a 23% decline from 2001 to 2006 in MFS category 'M349 Mechanical engineering technology – General.'

Coding change – 'Administration':  One of the more common non-specific write-ins in the field of study question is 'Administration' (and other similar variations such as 'Admin' or 'BA Admin'). In the 1996 Census, such write-ins were coded to 'M198 Public Administration' by default. In the 2001 Census, a correction was made to instead transfer the majority of these write-ins to 'M189 Business Administration'. For 2006, to coincide with the CIP interpretation of this term, all write‑ins of 'Administration' (and other similar variations) were assigned to 'M189 Business Administration'. Users should keep this in mind when observing the decrease in Public Administration and the strong corresponding increase in Business Administration over the 1996‑2006 time period.

Coding change – Removal of 'M482 No specialization':  For reasons related to the processing of the major field of study question by CIP prior to MFS, the MFS category 'M482 No specialization' was dropped from the 2006 MFS dissemination codeset. Instead, records originally coded to M482 were reassigned another MFS code through donor imputation based on the distribution of the already-assigned CIP code. For the majority of the 9,421 cases involved (approximately 45,000 weighted records), this resulted in a reassigned code of 'M120 Humanities – General Arts'. A lesser but still significant number were also assigned to 'M187 Social science and related n.e.c.'. Users should keep this change in mind when observing trends for these MFS codes. When comparing 2006 MFS results with previous census years at the highest level of aggregation (MFS12), we recommend that users combine 'No specialization (M481-M482)' with 'Humanities and related fields (M080-M124)' to make this category more conceptually consistent over time.

Attendance at school

As was noted above in the section ‘Comparisons with other data sources’, counts for those aged 15 to 19 attending school in 2001 were lower than expected and not replicated in other sources. In 2006, the changes to the questionnaire, brought results for those aged 15 to 19 back in to line with previous censuses.

The data for those aged 20 to 44 are comparable over time. In 2006, school attendance for the population aged 45 and over showed a large increase and the data for this age group are not comparable with previous censuses. This increase is believed to be due in large part to the absence of a note in the 2006 questionnaire instructing respondents to report attendance only for courses towards a certificate, diploma or degree, coupled with the change to a question focused on type of institution. It appears that some respondents, especially those aged 45 and over, may have reported non-credit courses in this category which resulted in attendance rates being higher than expected. It is not recommended to compare data for the age group 45 and over from 2006 with previous censuses.

---

Note:

1. Appears in both 2001 and 2006.