Rodrigue Landry, Mark Power, Marc-André Roy, Jean-Pierre Hachey, Required changes to the Canadian census, as of 2021, so that it will allow (1) the full implementation of the minority language education guaranteed by section 23 of the Canadian Charter of Rights and Freedoms, and (2) the full implementation of sections 16, 16.1, 19 and 20 of the Charter and parts III, IV and VII of the Official Languages Act, February 2017, section 8, pp. 46−62.

8. Analysis of the shortcomings in the census and the modifications and additions to the census questionnaire that are required

8.1 Mother tongue

8.1.1 Suggested modifications to question 9 of the census on mother tongue, to address the current shortcomings in the census

[1]        The current formulation of question 9 of the census, on mother tongue, communicates to respondents that the census is looking for a single response to the question. Question 9 asks what is the “language… first learned,” in the singular. The question in no way suggests that a twofold response will be accepted (although the questionnaire allows double or even multiple responses). Similarly, the three response options suggest that only one response will be accepted: “English,” “French,” or “Other language” (singular), in the English version of the question. The suggestion is the same in the French version of question 9:

Current version of question 9 of the census, on mother tongue, as it was included in the short-form and long-form questionnaires of the 2016 census

9. Quelle est la langue que cette personne a apprise en premier lieu à la maison dans son enfance et qu’elle comprend encore ? [Si cette personne ne comprend plus la première langue apprise, indiquez la seconde langue qu’elle a apprise.] 1 : Français 2 : Anglais 3 : Autre langue – précisez Cette question s’adresse à toutes les personnes inscrites sur le questionnaire. Si vous répondez pour d’autres personnes, veuillez consulter chaque personne[73].

9. What is the language that this person first learned at home in childhood and still understands? [If this person no longer understands the first language learned, indicate the second language learned.] 1:English 2:French 3: Other language – specify This question is for all persons listed on the questionnaire. If you are answering on behalf of other people, please consult each person.[74]

[2]        It is therefore important to modify question 9 on mother tongue, so as not to give respondents the false impression that they must only give one response to the question, or suggest that if a person learned more than one language at the same time in early childhood, they must choose their dominant language among those languages. Question 9 must adequately identify the mother tongue or mother tongues of every person, as well as their first official language learned and still understood.

8.1.1.1 Suggested modification to census question 9 on mother tongue – suggestion 1a

[3]        The following suggested modification communicates to the respondent, in the formulation of the question, that a person may, for the purposes of the census, have more than one mother tongue. In other words, this formulation clarifies that the census recognizes the reality of double (and multiple) mother tongues, and that respondents are not required to respond by identifying only one language, if the true response is that they have two (or in some cases three).

[4]        In order to do this, a plural option is therefore added to the question “[w]hat is the language,” in the singular, “that this person… learned.” Moreover, the mention of an “[o]ther language” apart from French or English is modified to indicate that it may be one “other language,” or more than one.

[5]        This suggested version offers three response options, like the current census, with the difference that it tells the respondent to check all the responses that apply. The formulation used to do this is one that is already used elsewhere in the census.[75] The instructions that go with the current question 9 (reproduced in full above), before and after the response options, are not included in the suggested versions below. All suggested additions to the text of question 9 are underlined.

Suggested modification to census question 9 on mother tongue – suggestion 1a
9. Quelle est la langue, ou quelles sont les langues, que cette personne a apprise(s) en premier lieu à la maison dans son enfance et qu’elle comprend encore ? [Cochez plus d’un cercle, s’il y a lieu.] 1 : Français 2 : Anglais 3 : Autre(s) langue(s) – précisez	9. What is the language, or what are the languages, that this person first learned at home in childhood and still understands? [Mark as many circles as applicable.] 1: English 2: French 3: Other language(s) – specify

[6]        It would also be possible to give respondents an exhaustive list of response options for question 9, including “English and French” / “Français et anglais,” and to ask respondents to mark one circle. Such variants could be tested.

8.1.1.2 Suggested modification to census question 9 on mother tongue – suggestion 1b

[7]        Another possibility would be to ask the mother tongue question in two parts, like the current question 8 of the census which asks firstly, “What language does this person speak most often at home?” and then, “Does this person speak any other languages on a regular basis at home?”[76] The second part of this question was added in 2001. The main question, on the language most often spoken at home, had been asked since 1971. This addition did not pose any major problems. Statistics Canada may prefer to modify the question in the following way in order to maximize the comparability of the 2021 census with earlier censuses, keeping the first part of the question unchanged and adding a second part to the question.

[8]        However, it should be noted that this case is different from that of the language most often spoken at home and the other languages spoken regularly at home. The latter are actually distinct questions, whereas in the case of the question suggested below, the second part of the question is used to collect any multiple responses that may not have been provided in the first part. It should also be noted that in the case of the question on mother tongue, adding a second part as suggested below would doubtless have an impact on the way some respondents would respond to the first question, since those who have given a multiple response to the current mother tongue question in spite of its discouraging effect, would normally change that behaviour, and indicate one of the two languages in the second part of the question. Moreover, it seems that the addition of a second part, while maintaining the current question 9 as the first part, could pose a problem for those who actually learned more than one language (e.g. French and English) at the same time and spoke both languages more or less equally often. For those persons, at least some of whom already respond that they have more than one mother tongue, despite the formulation of the current question 9, it would likely be difficult to indicate that one of their mother tongues is subordinate to the other, by indicating that it is the “other language” learned at the same time, in part (b) of the question. It therefore seems that suggestion 1a above is the preferable option. However, it may be useful for Statistics Canada to conduct tests on both options (1a and 1b) (in addition to any other option or variant Statistics Canada may consider useful to test).

[9]        Applying this structure to the mother tongue question, we might arrive at a formulation such as the following:

Suggested modification to question 9 of the census on mother tongue – suggestion 1b

9. a) Quelle est la langue que cette personne a apprise en premier lieu à la maison dans son enfance et qu’elle comprend encore ? 1 : Français 2 : Anglais 3 : Autre langue – précisez b) Cette personne a-t-elle appris une autre langue ou d’autres langues en même temps à la maison dans son enfance qu’elle comprend encore ? 1 : Non 2 : Oui, français 3 : Oui, anglais 4 : Oui, français et autre(s) langue(s) — précisez 5 : Oui, anglais et autre(s) langue(s) — précisez 6 : Oui, autre(s) langue(s) — précisez

9. a) What is the language that this person first learned at home in childhood and still understands? 1: English 2: French 3: Other language – specify b) Is there another language or languages that this person learned at the same time at home in childhood and still understands? 1: No 2: Yes, English 3: Yes, French 4: Yes, English and Other language(s) — specify 5: Yes, French and Other language(s) — specify 6: Yes, Other language(s) — specify

8.1.1.3 Suggested modification to census question 9 on mother tongue, to obtain an accurate count of persons with French as their first official language spoken

[10]      As indicated above, it is also important for the census to accurately determine the first official language learned and still understood. This is the case because it is important to count all persons outside Québec who have neither English nor French as their mother tongue, but who have a strong connection with the French language. A significant portion of these persons will be counted through the question or questions on language of education, but certainly not all. However, an increasing number of persons from Africa, Europe and elsewhere in the world receive their schooling in a language other than French, but have French as their second language spoken (and therefore their first “official” language in Canada), and as a result have a strong connection with French. Many of these persons enroll their children in a French-language school when they move to Canada. Some of these persons are not counted by the derivation method for the first official language spoken (widely known as the “FOLS”), which takes into account firstly knowledge of both official languages, secondly, mother tongue, and thirdly, language spoken at home.[77] These people must be counted, even if they speak English at home, e.g. because they live in Calgary and speak English at home in an effort to master the language required for the majority of jobs in their new city. As mentioned, many persons who fall into this category enroll their children in French-language schools outside Québec. It is therefore crucial that they be identified in the census, which is not currently the case. In order to address this problem, the census form could, when a person responds that neither French nor English is their mother tongue, ask what is the next language they learned, if any. The form could ask this question until an official language of Canada is identified, or no more languages are given (e.g., 1. Arabic, 2. French, or 1. Portuguese, 2. Spanish, 3. French). This suggestion applies to suggestions 1a and 1b above.

8.1.2 Suggested modifications to the instructions for the mother tongue question provided in the census guide – suggestion 2

[11]      The current formulation of the instructions for question 9 on mother tongue contributes to the impression given to respondents that the census is looking for a single response to this question. It is important to modify the wording of question 9 and/or add a new question to correctly identify the mother tongue or mother tongues of each respondent. Clarifications in the instructions are not sufficient, as there is every indication that the majority of respondents do not consult the instructions. That being said, the instructions are still important and must therefore be clarified.

[12]      The following formulation of the instructions regarding question 9 would inform the respondent, in the applicable situations, that multiple responses are accepted (suggested additions are underlined and suggested deletions are struck out):

Suggested modification to the instructions for question 9 on mother tongue – suggestion 2

Si une personne a appris deux langues ou plus à la maison en même temps dans sa petite enfance, indiquez lales langues qu’elle parlait le plus souvent à la maison avant d’aller à l’école et qu’elle comprend encore. Indiquez deux langues ou plus si la personne les utilisait aussi souvent les unes que les autres et si elle les comprend encore. Dans le cas d’un enfant n’ayant pas encore appris à parler, indiquez la langue utilisée ou les langues utilisées le plus souvent à la maison pour communiquer avec l’enfant. Les personnes sourdes ou les personnes ayant des troubles de la parole doivent indiquer leur connaissance du français ou de l’anglais s’il y a lieu, en cochant l’option appropriée. Pour d’autres langues, incluant la langue des signes, il faut les inscrire dans la case « Autre langue – précisez ». Soyez précis lorsque vous inscrivez d’autres langues. Par exemple, les personnes qui déclarent le chinois devraient plutôt mentionner la langue chinoise précise, soit le cantonais, le mandarin, le chaochow, le fou-kien, le hakka, le shanghaïen, le taïwanais, etc.

For a person who learned two or more languages at home at the same time in early childhood, report the languages this person spoke most often at home before starting school and that are still understood by this person. Report two or more languages only if those languages were used equally often and are still understood by this person. For a child who has not yet learned to speak, report the language or the languages spoken most often to this child at home. For people who are deaf or for people who have a speech disability, report knowledge of English or French as applicable, by marking the appropriate option. Other languages, including sign language, should be entered in the box labeled “Other language - specify”. When reporting other languages, be specific. For example, people who report Chinese should instead report the specific Chinese language: Cantonese, Mandarin, Cheochow, Fukien, Hakka, Shanghainese, Taiwanese, etc.

[13]      It goes without saying that like the possible changes to the questions, the instructions could also be tested by Statistics Canada.

8.1.3 Response to concerns on the part of Statistics Canada regarding the variability of twofold responses to the census questionnaire on mother tongue

[14]      During his testimony before the Standing Senate Committee on Official Languages on December 5, 2016, in response to a question from Senator Raymonde Gagné (independent) on multiple mother tongues, Jean-Pierre Corbeil indicated that Statistics Canada considers multiple responses to be “extremely unstable” from one census to another:

What you need to know is that multiple responses are extremely unstable from one census to another. In recent years, we have paired files to examine the extent to which people who provided multiple responses in one census provided exactly the same type of response in the next census.

We have concluded that the information over the past 25 or 30 years has been very consistent, as almost 80 p. 100 of respondents who provide multiple responses in one census do not provide the same responses in the next census. This does not mean that those answers are not valid, but Statistics Canada tries to ensure that those who give more than one response have actually learned both languages ​first; people tend to report more languages ​than the first language learned, even the languages ​they use outside the home.

Our goal is to get the best quality possible. However, in all census data, 1.5 to 2.5 per cent of Canadians still provide multiple responses to the mother tongue question.[78]

[15]      It would appear, however, that there are a great many reasons why responses regarding the mother tongue could vary over the years, that would not suggest that the multiple responses given at one of the times in question are false.

[16]      First, the same person does not always respond to the census for a given individual. One questionnaire – electronic or written – is completed per household. Often one person completes the questionnaire for the household without discussing the responses given with the other members of the household.[79] If a person with two mother tongues is in a relationship with an Anglophone partner, and they only use English when speaking to each other, the Anglophone partner is very likely to respond for herself and her partner that English is their mother tongue. If that person with two mother tongues completes the census questionnaire himself during a subsequent census (regardless of whether he is still in a relationship with the same partner, single, or in a relationship with a new partner – Anglophone, Francophone or Allophone), he is liable not to give the same response.

[17]      Second, the instructions associated with question 9 on the mother tongue, according to which a respondent should only indicate more than one mother tongue if they spoke both languages equally often before starting school,[80] is itself liable to cause variations in the responses given. Quite simply, many persons with English and French as the mother tongue would not give the same response to the question depending on whether they consulted the instructions associated with the question.

[18]      Third, as pointed out above (see paragraphs [74] to [76]), different responses were given to the mother tongue question when the respondent completed the short-form questionnaire for one census and the long-form questionnaire for the other, or vice versa. When the mother tongue question was asked in the context of other language questions, i.e. preceded by questions on knowledge of official languages (and knowledge of other languages in the long-form questionnaire) and languages spoken at home (long-form questionnaire for the 2001 and 2006 censuses and short-form questionnaire for the 2011 census), questions that clearly allow multiple responses, the percentage of multiple responses was relatively low for the country as a whole: 1.3% (2001), 1.3% (2006) and 1.9 % (2011). These results are very different from those of respondents to the 2001 and 2006 short-form questionnaires, for which the respective percentages of multiple responses to the mother tongue question were 4.9% and 3.6%.[81] It would seem therefore that the questions preceding the mother tongue question (questions that clearly invite multiple responses, whereas the mother tongue question suggests that only one language should be identified) have an effect on the percentage of multiple responses. We have also already pointed out that the rates of multiple responses are much higher among “Francophones” outside Québec (10.6% on average in 2011) than in the Canadian population as a whole.

[19]      Fourth, the respondent or respondents (who, as pointed out above, are not necessarily the same person for different censuses, in relation to a particular individual) may interpret the question differently during different censuses. In other words, the respondent (whether or not it is the same person) may draw different conclusions during different censuses as to whether multiple responses are allowed.

[20]      Fifth, a person’s level of awareness regarding the minority language may change considerably over the years as a result of changes in their personal circumstances. For example, a person with English and French as mother tongues living with an Anglophone spouse in a region with a large Anglophone majority, not living near their family and with no children, would be less likely to think about their French mother tongue as the same person if they have children, particularly if they have enrolled them (or are considering enrolling them) in a French-language school.

[21]      Sixth, a person’s language skills may change over the years. First, a person who has two mother tongues may, later in their life, no longer understand one of those languages, in which case that language is no longer considered to be one of their mother tongues by the census. However, even a much less extreme decline in their skills may cause a person to change their response (or the response of someone else on that person’s behalf) regarding their minority mother tongue. Such a person may, particularly if they do not use their minority mother tongue often, decide at some point that they no longer speak it “well enough” to say it is their mother tongue. That same person may, later in their life, particularly if their children attend a minority language school, regain a closer connection with that language and improve their skills in that language, with the result that the person once again considers the minority language to be a true mother tongue for them. Indeed, a parent who has received two mother tongues from exogamous parents may have indicated only English as their mother tongue on one census, because that was the language they spoke most often, but indicate two mother tongues (English and French) on the next census, once they have become aware of their status as a rights holder under section 23 of the Charter, for example, because they enrolled their child in a French-language school outside Québec, or plan to do so when the child reaches school age.

8.2 Language of education – of parents and children

8.2.1 Analysis of the “language of education” questions suggested by Statistics Canada that were tested in 1993 and 1998

[22]      During his testimony before the Standing Senate Committee on Official Languages on December 5, 2016, in response to questions posed by members of the committee, Mr. Corbeil mentioned two tests conducted by Statistics Canada in the past, in which a question on “language of education” was added. These tests were carried out in 1993 and 1998. In 1993, the question tested was “Since September 1993, in which language was this person taught most of the courses taken at school, college or university?” The results of the tests showed that this question did not provide the desired information, since the ideas of immersion and second language instruction programs created confusion among respondents.[82] Indeed, this question clearly does not even suggest to the respondent that there are Francophone programs and French immersion programs, and that the differences between these programs is important. Moreover, the question combines all levels of education, which causes problems in terms of the usefulness of responses for enumerating rights holders under section 23 of the Charter and their children. Finally, the suggested question attempted to collect data only on elementary, secondary, college and university education that was ongoing at the time when the question was asked. The question would therefore necessarily have failed to identify rights holders under paragraph 23(1)b) of the Charter (due to the parent’s elementary-level education), since all the data collected regarding elementary-level education in French would have concerned the schooling of children who were attending elementary school at the time when the question was asked; the question did not attempt to collect data on the schooling of parents who had received their elementary-level education many years before the time specified in the suggested question (“[s]ince September 1993,” i.e. the current year when the question was tested).

[23]      Mr. Corbeil did not cite the text of the suggested questions on “language of education” that were tested in 1998, but he did indicate that it was a two-part question, to attempt to distinguish French immersion programs from minority French-language programs. He said the results were overall much better than those of 1993, and that following the qualitative tests conducted by Statistics Canada, the recommendations received were of interest. Mr. Corbeil indicated that a quantitative survey should have been carried out later with a large sample of 75,000, in order to proceed with the addition of the questions. At that time, it was too late to conduct that type of survey.[83]

[24]      The text of the two variants of these questions that were tested in 1998, Statistics Canada’s analysis of those questions and the results they yielded, and the text of the modifications suggested by Statistics Canada following that analysis, are contained in an internal Statistics Canada report (prepared in French only), Test du recensement national de 1998: analyse des variables linguistiques 1998 national census report: analysis of linguistic variables.[84] The text of the questions tested, along with the text of the variants suggested in the report Test du recensement national de 1998 is reproduced and analyzed below. It should be noted from the outset that the questions tested in 1998, like the one tested in 1993, were aimed at collecting data on education being received at the time when the questions were asked (here, “[s]ince September 1997”). However, although the purpose of these suggested questions was therefore different from that of a question on language of education in the 2021 census, should such a question be added, the progress made by Statistics Canada in 1998 and 1999 on the issue of clearly identifying the difference between a French-language school and a French immersion school is relevant to the formulation of a question or questions on language of education for inclusion in the census questionnaire beginning in 2021. It should also be noted that the context of minority language schooling has changed considerably since the 1990s, and that Canadians’ knowledge about the existence of, and differences between, French immersion programs offered by Anglophone school boards, on the one hand, and French-language schools outside Québec, on the other, has improved significantly since that time.

[25]      The following two variants of the language of education question were tested in 1998 (the bold type is in the original version):

Version 1 :	Since September 1997, what was the language used in the majority of subjects taught to this person at school, college or university? You should mark “French – Immersion in an English school” in the case of students who attend an English school in which a number of subjects are taught to them in French. Mark one circle only.
		French – French school, college or university French – Immersion in an English school English – English school, college or university Other – Specify
Version 2 :	a) Since September 1997, what was the language used in the majority of subjects taught to this person at school, college or university?
		French English Other – Specify
	b) Was this person enrolled in an immersion program in an English school in which they are taught a number of subjects in French?

[26]      Statistics Canada concluded that the second version was preferable, since respondents found it easier to understand, whereas the first version contained too much information in a single question.[85] Statistics Canada summed up the weaknesses in the question, based on its analysis of the test results, including telephone follow-up, as follows:

The problems with the understanding of the concept of French immersion observed during the telephone follow-up seem to indicate that there is some risk of obtaining biased results with regard to the distribution of students in the French and English school systems, and that this bias would favour the English system.[86]

[27]      Statistics Canada had indeed concluded that there was some confusion around the concept of French immersion, but mainly in Québec. After applying partial corrections in this regard, the results outside Québec were essentially unchanged:

As expected, the bias is in favour of the English school system, and it is greater in all sites in Québec than in those outside Québec as the understanding of the concept of immersion is more problematic there. Thus, following the partial correction made, the proportion of young people attending [schools in] the English system in all sites in Québec fell from 14.9% to 10.8%, while in all sites outside Québec it only fell slightly, from 87.9% to 87.6%.[87]

[28]      Statistics Canada formulated two modified versions of part (b) of the question on language of education, in light of the preliminary analysis of the results of the national test conducted in October 1998. Qualitative tests were carried out on these two questions, but no quantitative tests were done. In this modified version of part (b), Statistics Canada tried to “reduce certain difficulties noted in the analysis of the results and during the telephone follow-up, such as:”

• respondents had not realized that it was a French immersion program;
• it did not seem clear that it was a program offered in an English school system;
• respondents believed they could indicate their situation in the past, if they had been enrolled in an immersion program.[88]

[Bold type in the original version]

[29]      The variants on part (b) of the question on language of education were formulated as follows:

Version 1 :	b) Is this person currently enrolled in an immersion program at an English school in which several subjects are taught in French?
		No Yes
Version 2 :	b) Is this person currently enrolled in an immersion program in French at an English school?
		No Yes[89]

[30]      Statistics Canada concluded that version 2 was preferable as it was clearer, but that a quantitative test would be required to confirm that it would yield reliable results throughout the country:

Based on the comments gathered during the interviews, version 2 seems clearer and easier to understand. It is clear that we are referring to French immersion, whereas in version 1 it is referred to in a less direct way. Furthermore, the bold type seems to help emphasize what we are seeking to determine. Thus in Québec, where the understanding of the concept of immersion was most problematic, a number of people noticed that it was a program offered in the English school system. However, although this version seems promising, only a quantitative test would allow us to state with certainty that this wording would yield plausible results both in Québec and outside Québec. The results obtained should be similar to the data suggested by administrative records.[90]

[31]      It seems that version 2 of part (b) reworked by Statistics Canada clearly identified what is meant by a French immersion program. However, three comments are called for with regard to the questions suggested by Statistics Canada, from the perspective of collecting the data required for the full implementation of section 23 of the Charter, and the full implementation of sections 16, 16.1, 19 and 20 of the Charter and Parts III, IV and VII of the OLA.

[32]      First, both parts of the question focus on education that is ongoing at the time of the census. The question should enumerate the persons who received instruction in French regardless of when they received that instruction.

[33]      Second, the first part of the question indicates that the question is about instruction received “at school, college or university.” The question thus combines too many categories, which need to be distinguished in order to provide data that is truly useful. Data on schooling cannot be combined with data on post-secondary education. Given that the suggested question was about education being received at a specific time, it may be that Statistics Canada planned to be able to separate the data by age, which would allow it to be divided, with an acceptable degree of precision, between responses regarding persons at school and those at college or university.

[34]      However, given that data must be collected on education regardless of when it was received, as indicated, it will be impossible to ask a question about multiple levels of schooling and education in this way. First, data on elementary education must remain separate from all other data, since the right guaranteed in paragraph 23(1)b) of the Charter specifically depends on the parent’s elementary-level education. Second, data on secondary-level education must remain separate from data on college or university, since the right conferred by subsection 23(2) of the Charter depends on the schooling – at the elementary or secondary level – of a parent’s child, and not on their post-secondary education. If it is possible to ask a question about the language of post-secondary education, that would certainly be desirable. However, if it is not possible to ask three questions, or three parts of questions, about elementary, secondary and post-secondary education respectively, post-secondary education is the category that must be eliminated, so that at least the data required for assessing the universe of rights holders under section 23 of the Charter and their children can be collected.

8.2.2 It will be very important to deal with French immersion in the question on language of education, both to ensure the quality of the data on language of education and because of the importance of French immersion for Canadian bilingualism and the composition of French-language communities, including exogamous couples

[35]      It will be very important to deal with French immersion in the question on language of education. First, it is important that the data clearly distinguish between education received in French-language schools and that received in immersion programs offered by English-language school boards, since only the first of these gives parents rights under paragraph 23(1)b) of the Charter (with regard to their own education), or subsection 23(2) (with regard to the education of one of their children).[91] Second, it is important to know the distribution of those who have attended immersion programs, since they represent a significant proportion of bilingual Canadians, including Francophiles who are an integral part of French-language communities, and parents whose children attend French‑language schools – either because the other parent is a rights holder under section 23 of the Charter, or because the child was admitted to a French-language school outside Québec, or because the child attends a French-language school in Québec (where all children are eligible to attend a French-language school). It is also important to collect data on persons who received their education in French immersion programs, because this would provide a better understanding of the composition of the growing category of exogamous couples, in which the rate of transmission of French as a mother tongue has improved in recent decades. It should be noted that the number of students in French immersion is very high and growing:

Enrolments in French immersion programs totalled 409,893 in 2014/2015, up 4.5% compared with 2013/2014 when 392,430 students were enrolled. Increases in these program enrolments were seen in virtually every province and territory.[92]

[36]      The distinction between French-language schools and immersion programs is important for another reason. It would provide the number of children with at least one parent who is a rights holder under section 23 of the Charter outside Québec who are enrolled in an immersion program rather than the French-language school. The Survey on the Vitality of Official Languages conducted by Statistics Canada following the 2006 census showed that around 15% of children eligible to attend a French-language school were attending an immersion program at an English-language school.[93] The importance of this number is underlined by the fact that only around 50% of children with at least one rights holder parent outside Québec attend a French-language school[94] and around 41% of parents whose child attends the English-language school would have preferred for their child to attend a minority school.[95] These results would be used to organize campaigns to raise awareness among rights holders outside Québec of the effects of the various education programs on their children’s bilingual development.[96] Many parents are under the mistaken impression that a bilingual school program (e.g. 50% of classes in French and 50% of classes in English, a formula similar to the one used in immersion programs) will provide their child with stronger bilingual language skills than the French-language school.[97] Both for the children of rights holders in exogamous relationships and for those of endogamous Francophone couples in a minority setting, the French-language school provides stronger bilingual language skills.[98] Recently, the Commissioner of Official Languages for Canada, in a report on early childhood, recommended that campaigns of this kind be organized to raise awareness among rights holders outside Québec, recognizing the period of early childhood as crucial for the vitality of Francophone communities.[99]

8.2.3 Suggested additions with regard to language of education, to address the current shortcomings in the census

[37]      The following suggested question on language of education would be a new question 10 on the census, following the mother tongue question. It would be included in the short-form questionnaire, and would thus be asked of 100% of the population, since as indicated above, questions that figure in the short-form census are also included in the long-form census.

[38]      Two formulations for a new question 10 are suggested below. No words are in bold print in the suggested questions below. Statistics Canada may wish to emphasize certain words, or test different variants, with certain words emphasized using bold print in some of them. Although this is a suggested addition, since the entire question is an addition, in the interest of legibility, the text is not underlined.

8.2.3.1 Addition of a new census question 10 on language of education – suggestion 3a

[39]      The suggested question recognizes that, as the Supreme Court of Canada confirmed in Solski[100] and Nguyen,[101] a person is not required to have had all their schooling in the language of the minority in order for section 23 of the Charter to give them rights under paragraph 23(1)b) or subsection 23(2), and section 23 does not “require that the time spent in the minority language education program be greater than the time spent in the majority language program.”[102] The suggested question therefore asks about “a substantial part” of the person’s education, at the primary level in part (a) of the question, and at the secondary level in part (b) of the question. The suggested instructions between the question and the response options (for both part (a) and part (b)) tell the respondent that they may mark more than one circle, if applicable. This instruction communicates to the respondent that there may be more than one substantial part of a person’s primary or secondary education. This instruction therefore communicates that it is not necessary to give just the response that corresponds to the largest number of years of education. Moreover, multiple responses will be useful since they will make it possible to identify and quantify school careers that involve a combination of programs – e.g., the French-language school and French immersion, or French immersion and the regular Anglophone program in an English-language school.

[40]      It would also be possible to ask a similar question about post-secondary education, either as an added part (c) of the question proposed below, or as a separate question. It seems that it would be preferable to ask such a question separately, as there are significant differences between primary and secondary education on the one hand and post-secondary education on the other, including the fact that minority language programs in the meaning of section 23 of the Charter do not exist beyond the secondary level.

Addition of a new census question 10 on language of education – suggestion 3a

10. a) Dans quelle sorte de programme éducatif cette personne a-t-elle fait une partie substantielle de sa scolarité au niveau primaire ? [Cochez plus d'un cercle, s’il y a lieu.] [Si la personne est toujours inscrite à l’école primaire, répondez par rapport à la partie des études primaires faite jusqu’à ce jour.] 1 : Programme francophone, dans une école de langue française 2 : Programme d’immersion en français, dans une école de langue anglaise 3 : Programme anglophone, dans une école de langue anglaise 4 : Autre – précisez 5 : Cette personne n’a pas fait des études au niveau primaire 10. b) Dans quelle sorte de programme éducatif cette personne a-t-elle fait une partie substantielle de sa scolarité au niveau secondaire ? [Cochez plus d'un cercle, s’il y a lieu.] [Si la personne est toujours inscrite à l’école secondaire, répondez par rapport à la partie des études secondaires faite jusqu’à ce jour.] 1 : Programme francophone, dans une école de langue française 2 : Programme d’immersion en français, dans une école de langue anglaise 3 : Programme anglophone, dans une école de langue anglaise 4 : Autre – précisez 5 : Cette personne n’a pas fait des études au niveau secondaire

10. a) In what type of educational program did this person receive a substantial part of his or her primary school education? [Mark as many circles as applicable.] [If the person is still enrolled in primary school, answer regarding the portion of primary school instruction completed to date.] 1: English program in an English-language school 2: French immersion program in an English-language school 3: Francophone program in a French-language school 4: Other – specify 5: This person has not received instruction at the primary school level 10. a) In what type of educational program did this person receive a substantial part of his or her secondary education? [Mark as many circles as applicable.] [If the person is still enrolled in secondary school, answer regarding the portion of secondary-level studies completed to date.] 1: English program in an English-language school 2: French immersion program in an English-language school 3: Francophone program in a French-language school 4: Other – specify 5: This person has not received instruction at the secondary school level

8.2.3.2 Addition of a new census question 10 on language of education – suggestion 3b

[41]      The following is an alternative formulation for a new census question 10 on language of education, taking into account the location where the person received their schooling (in Canada or outside Canada):

Addition of a new census question 10 on language of education – suggestion 3b

10. Dans quelle sorte de programme éducatif cette personne a-t-elle fait une partie substantielle de sa scolarité au niveau primaire au Canada ? [Si la personne est toujours inscrite à l’école primaire, répondez par rapport à la partie des études primaires faite jusqu’à ce jour.] 1 : Programme francophone, dans une école de langue française 2 : Programme d’immersion en français, dans une école de langue anglaise 3 : Programme anglophone, dans une école de langue anglaise 4 : Combinaison de 1 et 2 5 : Combinaison de 2 et 3 6 : Combinaison de 1 et 3 7 : Cette personne n’a pas fait des études au niveau primaire au Canada : Spécifiez la langue principale de la scolarisation : __________________ 8 : Cette personne est trop jeune et n’a pas fait des études au niveau primaire 9 : Cette personne n’a pas fait des études au niveau primaire

10. In what type of educational program did this person receive a substantial part of his or her primary education in Canada? [If the person is still enrolled in primary school, answer regarding the portion of primary school instruction completed to date.] 1: English program in an English-language school 2: French immersion program in an English-language school 3: Francophone program in a French-language school 4: Combination of 1 and 2 5: Combination of 2 and 3 6: Combination of 1 and 3 7: That person has not received instruction at the primary school level in Canada: Specify the primary language of schooling:______________ 8: This person is too young and has not received primary school level instruction 9: This person has not received instruction at the primary school level

[42]      If this option (suggestion 3b) is tested by Statistics Canada, a second question, or a second part of the question, would also need to be added, asking about the educational program in which the person received their secondary-level schooling.

[43]      It should be noted that the suggested question 3a does not ask where the instruction in question was received. The question could ask for this additional information, as is the case in suggested question 3b. In addition, the question could ask respondents to specify where the instruction was received, in the case of instruction received outside Canada, which is not done by suggestion 3b (if, for example, it is considered to be useful to determine whether instruction in French outside Canada was received in the Democratic Republic of the Congo, Haiti, Belgium, etc.). Even without asking where the instruction was received, it can be expected that responses identifying French immersion would refer to instruction received in Canada. Similarly, it would be possible to conclude that most responses regarding instruction received in a language other than French or English, identified under category 4, “Other,” of suggestion 3a refer to instruction received outside Canada.