By modifying the question on the languages ​​of use, StatCan compromised the results of the 2021 census

The new formulation of the questionnaire on the languages ​​of use suppresses the declaration of regularly spoken second languages ​​and has compromised any real comparability with the data of previous years. Here is an analysis by Calvin Veltman, Professor (retired) of ÉSG-UQAM and sociolinguist at the University of Quebec in Montreal.


ANALYSIS – Statistics Canada dropped a little bombshell last week by releasing its latest data on the language used by the country’s citizens.

Compared to data from censuses from 2001 to 2016, there is a regression of French across the country.

However, the wording of some key questions has changed. This caused a drastic drop, for example, in the francization rate of immigrants.

Comparability has become impossible

One of the fundamental principles guiding the decisions of census managers is the concern to preserve the comparability of data from one census to another.

When I worked as director of a project on linguistic mobility at the National Center for Education Statistics (USA), we developed the project to revise an invalid question concerning the mother tongue of Americans. Developed in collaboration with researchers from the US Census Bureau, the project went through all the stages of approval before being refused by the director himself. The reason ? We would lose the comparability of the data with those of the censuses already carried out.

This logic also guides Statistics Canada, which has maintained a question on mother tongue since 1941 which encourages the most assimilated respondents not to declare their true mother tongue, under the pretext that they have “forgotten” it.

What is the language that this person first learned at home in childhood and still understands?

However, Statistics Canada has departed from this well-established tradition by modifying the questions on the language of use in 2021.

Changes in questions

From 2001 to 2016, Statistics Canada asked the following questions:

has) What language does this person speak most often at home?

  • French
  • English
  • Other language — specify

b) Does this person regularly speak other languages ​​at home?

  • Nope
  • Yes French
  • Yes, English
  • Yes, other language — specify

In 2021, the questionnaire was modified as follows:

has) What language(s) does this person speak regularly at home?

  • French
  • English
  • Other language(s) — specify:

If this person indicates only one language in the question, go to question 10.

b) Which of these languages ​​does this person speak most often at home?

Indicate more than one language only if they are spoken equally often at home.

  • French
  • English
  • Other language — specify:

The results

Without going into the details of my data compilation method, here is the linguistic evolution from 2001 to 2021 for three linguistic groups, leaving aside the small group of people declaring French/English bilingualism as their mother tongue.

The differences observed between the regular evolution of the data from 2001 to 2016 suggest what we should find in 2021. Here are the data for people whose mother tongue is French.

French mother tongue. Calvin Veltman, Author provided

This table shows the fairly sustained incursion of the English language into French-speaking households from 2001 to 2016. Regularly spoken English as a second language gained 130,000 followers. According to the 2021 census, only 244,140 Francophones spoke English as a second language at home, bringing us back to the situation observed in 2001.

The adoption of English as a home language with French as a second language increased by about 10,000 people from 2001 to 2016. But almost all of it is lost from 2016 to 2021. Worse, English, without the French, grew quietly from 2001 to 2016, but jumped by more than 16,000 people in 2021.

The 2021 census therefore suppresses the declaration of second languages, English in the case of people who speak mainly French, and French for those who speak mainly English.

We see the same evolution among Anglophones.

English mother tongue. Calvin Veltman, Author provided

French without English jumped by more than 13,000 people, while the category of French speakers who retained English as a second language decreased by more than 14,000 individuals.

Among those who spoke primarily English but had adopted French as a regularly spoken language, a group that has been growing steadily since 2001, the 2021 census shows a loss of more than 30,000 people. While English unilingualism is growing slowly from 2001 to 2016, it adds nearly 62,000 people from 2016 to 2021.

We conclude that the new wording of the questionnaire suppresses the declaration of regularly spoken second languages.

Third mother tongues (allophones). Calvin Veltman, Author provided

Fewer French-speaking allophones

Our analysis of the linguistic evolution of Quebec society is largely based on the linguistic behavior of the allophone group. However, several aspects of this table leave me perplexed.

French without English increases by about 78,000 from 2001 to 2006, by 58,000 from 2006 to 2011, then by 69,000 from 2011 to 2016. The 2021 census adds only 5,000 people. Can we really believe that a process that has been so well established for 15 years can be completely reversed in a single five-year period?

Part of the expected growth seems to have landed more in bilinguals: thus, French with English as a second language increases by 27,000 people rather than the 5,000 to 7,000 people expected; English with French as a second language adds about 21,000 people, rather than the 2,000 or 3,000 people expected according to the previous evolution.

A surprising decision

In fact, Statistics Canada has made the data from 2001 to 2016 non-comparable to the newly published data. The 2021 data are so discordant with the trends observed from 2001 to 2016 that we must conclude that Statistics Canada erred in agreeing to modify a questionnaire that had already been used for the four previous censuses.

In my opinion, this is a surprising decision given the importance of the language issue in Quebec and, in particular, the study of its evolution. At best, Statistics Canada has started a new series of data that can be compared from the 2026 census — but any real comparability with the past is now compromised. Unless we restore comparability by imposing the 2016 questionnaire for the 2026 census.

Calvin Veltman, Full Professor (retired), ÉSG-UQAM, sociolinguist, University of Quebec in Montreal (UQAM)

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Subscribe to our newsletter and receive a daily summary of Montreal news.

The conversation

Follow by Email
Telegram
WhatsApp
FbMessenger