4 GROWTH summa FRENCH-SECOND-LANGUAGE ENROLMENT SUMMARY STATISTICS: CHANGES IN ENROLMENT AND COVERAGE As in past years, the State of FSL Education 2004 includes an extensive set of statistical tables documenting FSL enrolment by province, programme and grade. These tables give a detailed description of the enrolment picture, within the limits of the information provided by provincial/territorial ministries of education. The inclusion of data from prior years allows readers to follow changes in enrolment patterns in each jurisdiction. In some instances enrolment figures for prior years have been changed from those published in earlier reports. These changes are based on revised figures supplied by provincial or territorial ministries. In the current report, figures for New Brunswick have been significantly revised. These tables, however, have at least three limitations. First, the large volume of numerical information sometimes makes it difficult to see the forest for the trees. A second and related problem is that the separate tables for each province/territory, made necessary by the volume of information presented, make it difficult to gain an interprovincial perspective on FSL enrolment patterns. Third, the detailed tables present yearly data as a series of snap-shots. There are no summary indicators of changes or trends in enrolment patterns. To measure growth, be it positive or negative, we simply count the gains or losses represented by current enrolment against two standards enrolment in the prior year and average enrolment over the prior two years. We calculate both the absolute loss or gain in students and the percentage change. There are two reasons for employing growth measures. First, changes from one year to the next may be affected by one-off events. Gauging year-on-year changes against a multi-year average provides insurance against over-reaction to very short-run changes. In addition, the comparison of these three growth measures can provide a useful indication of trends. If, for example, we note that the difference in enrolment between the current year and the prior year is clearly larger than the difference between the current year and the average of the three prior years, this suggests that the enrolment trend is accelerating. These growth indicators have been calculated for total enrolment (by programme as provincial/territorial data allow) and for the major curriculum divisions in each province and territory. Most jurisdictions have three major curriculum divisions. (see Curriculum Divisions in Canada on the CPF website at www.cpf.ca). These are variously termed primary, intermediate and senior, or elementary, middle and secondary, etc. The Ontario and Quebec systems have only elementary and secondary divisions. There are two reasons for calculating separate growth measures for these divisions. First, attrition from programmes is frequently high at the boundary points, particularly where these coincide with change of school. It is valuable to be able to track enrolment trends on either side of the divide. Second, we can gain valuable information by comparing trends across curriculum divisions. For example, if we see intermediate enrolment increasing while primary enrolment is stagnant or declining, this may indicate a pig in the python phenomenon. That is, growth in intermediate enrolment reflects a growth in intake in prior years that has not been sustained. COVERAGE Statistical tables in the State of FSL Education have typically included an indicator of coverage, that is participation levels in programmes. FSL enrolments have been expressed as a percentage both of total enrolment and of eligible enrolment. Here we adopt the second approach, where data are available. As with the growth indicators, we have added two measures of change. The first compares current-year coverage to prior-year coverage. The second makes the comparison to average coverage over the prior three years. Again, the three-year average provides a more stable measure than single-year results. It is less affected by either one-off errors in data provided by provinces and territories, or by isolated actual changes in enrolments. Comparison of the current-year coverage and the three-year average shows whether present coverage appears to lead or trail the average over the past two years. 32
ry statistics GROWTH AND COVERAGE TOGETHER So far as possible, indicators should form a coherent system, not simply a shopping list. We should be able to gain information by looking at the relationship between two indicators as well as from their individual values. Our very simple starter system incorporates this feature. The growth and coverage indicators measure two distinct aspects of enrolment, both important. Growth measures show absolute expansions or contractions with important implications for resourcing. Coverage measures show breadth of participation and can interpreted as indicators of accessibility. Beyond this, however, the two measures viewed together can provide a more complete picture of conditions. Declining enrolment with stable coverage suggests that demographic changes are driving enrolment trends. A declining enrolment linked to declining coverage raises questions about provincial commitment to FSL programmes, and about any formal or de facto changes in policy affecting access. What Patterns Emerge? In 2003 we reported (of the data up to 2001-02) that the most striking feature of analysis of the FSL enrolment data is the universal decline in core French enrolments across all jurisdictions in contrast to immersion where a mixed performance across jurisdictions yielded net enrolment gains in enrolment, nationally. In 2004, based on data to 2002-03, we again find core French enrolment in decline, although at a reduced rate. Changes in core French enrolment are now more closely tied to changes in coverage than in past years. Growth in immersion enrolment has strengthened in the past year. Enrolment changes continue to be linked to changes in coverage. Programmes in Saskatchewan, Prince Edward Island, Ontario and Quebec show the strongest signs of recovery. It is notable that in two of these jurisdictions there has been a substantial absolute increase in coverage since 2001-02: Prince Edward Island (2.04%) and, marginally, Saskatchewan (0.22%). In Ontario and Quebec coverage has continued to decline, although modestly. Declines are registered as well in all other jurisdictions. Proportional change in coverage is somewhat more substantial: 4.51% in Prince Edward Island and 0.55% in Saskatchewan. This measure expresses changes in coverage as a % of initial coverage, and hence is a better indicator of the degree to which a programme is gaining or losing ground. In 2003 we found little relationship (correlation.195) between changes in enrolment and changes in coverage for core French programmes. This suggested that much of the drop in core French enrolment was to be explained by declining school populations as a whole. In 2004, we find a different pattern in the case of recent changes. (Note that in both 2003 and 2004, we are looking at the relationship of our proportional measures: % change in enrolment and % change in coverage, that is the % points gain or loss in coverage divided by the total prior coverage figure.) Analysis of scatter plots examining the relationship between changes in enrolment and coverage since 2001-02 show quite a strong relationship between changes in enrolment and changes in coverage. The lower the decline in enrolment, the lower the decline in coverage, suggesting that changes in enrolment are now linked firmly to changes in coverage, with changing school In every jurisdiction except Prince Edward Island core French enrolments continued to decline. Proportionally, losses of enrolment since 2001-02, range from 6.7% in Nunavut and 5.4% for the main Newfoundland and Labrador programme to less than half a percent in Quebec (see Figure 4.1a). Losses from prior year enrolment are more severe in the Atlantic Provinces and in the west, less so in central Canada. Nationally, core French enrolment declined by just over 28,000 or 1.7%. Viewed against average enrolment over the previous three years, losses are greater almost 61,000 or 3.6%. The good news is that the rate of decline has slowed from past years. Our 2003 report cited losses nationally of almost 33,000 students or 2.8% of the prior (2000-01) year s enrolment, well above the rate for 2002-03. 33
4 FSL Enrolment Summary Statistics population playing much less of a role than previously. Where the decline in core French enrolment has slowed, this can increasingly be credited to improvements in the availability and/or popularity of programmes themselves. When the analysis is based on a three-year average as the reference point, it reveals a much weaker relationship between changes in coverage and enrolment, reflecting the greater influence of changes in school population prior to the past year. The similar rates of decline in coverage can be associated with quite different rates of decline in enrolment. Two jurisdictions might experience roughly similar declines in core French enrolment, but show very different changes in coverage depending on the extent to which the enrolment decline has been due to an overall shrinkage in the school population, rather than change in coverage. Prior to this year, core French enrolment in some jurisdictions at least, was more closely tied to changes in total school enrolment than is now the case. For more information, please refer to the scatterplots on the CPF website at www.cpf.ca. French Immersion French immersion enrolment continues to expand nationally. In 2002-03 there are gains in all jurisdictions providing data with the exception of New Brunswick where there is a marginal decline. Proportionally, gains in enrolment since 2001-02 range from 7.6% in the Yukon and 6.9% in Nova Scotia to a tenth of a percent in Manitoba (see Figure 4.2a). Enrolment growth is stronger in the Atlantic Provinces (excluding New Brunswick) and in the west (excluding Manitoba) than in Ontario. Nationally, French immersion enrolment increased by just over 5200 or 1.8%. This represents a substantial gain from 2001-02 when the increase was under 3000, about a 1% gain on 2000-01. The gains in 2002-03 compared to average enrolment over the previous three years are higher: 8167 or 2.9%. There have been small absolute gains in coverage from 2001-02 in all jurisdictions, even marginally in New Brunswick despite the slight drop in enrolment. Absolute changes in coverage range from 0.98% in Quebec and 0.74% in Nova Scotia to a miniscule 0.02% in Ontario. Proportionally, gains in coverage are greater, exceeding 5% in five jurisdictions: Yukon (7.48%), Saskatchewan (6.81%), Nova Scotia (8.97%), Prince Edward Island (7.42%) and Newfoundland and Labrador (8.47%). Central Canada is notably absent from the list. Growth in immersion enrolments continues to be strongly associated with increasing coverage, whether we compare year-to-year changes or use three-year averages as our reference point. For more information please refer to the scatterplots on the CPF website at www.cpf.ca. Not Staying On: Enrolment Loss in Programmes After the Compulsory Period In seven provinces, French is currently a compulsory subject for some range of grades. (In British Columbia, a second language rather than specifically French is compulsory, however, most students take French.) In this section we look at what happens to enrolment in core French programmes following the end of the compulsory period. Here we use much the same approach as was used in our 2003 Report to examine attrition in general. The enrolment statistics for succeeding years are used to calculate net loss of enrolment in programmes with progress through the grades. The following example explains the process. In British Columbia (see Table 1, part A) in the 1999-2000 school year, there were 34,927 students enrolled in the core French programme at Grade 8. A year later, in the 2000-01 school year, 21,248 students were in Grade 9 core French classes. The difference between these two figures represents a net loss of 39.2% of the Grade 8 enrolment. In 2001-02, there were 15,956 students still in the programme at Grade 10, representing a net loss of 24.1% from Grade 9. In 2002-03, 12,422 attended at Grade 11, a drop of 22.1% from the Grade 10 enrolment a year earlier. We can also calculate a summary net attrition figure for this cohort s experience between Grades 8 and 11. The 12,422 students enrolled in Grade 11 in 2002-03 signify a net loss of 64.4% of the cohort of 34,927 students at Grade 8 in 1999-2000. It should be stressed that we are defining attrition as net loss, that is, outflows of previously registered students, counterbalanced in part by inflows of new students. Our enrolment statistics do not allow us to separately track inflows and outflows. We can only see the net result. Table 1, Parts A-G show the net loss in regular and alternate core programmes for single and multiple grade levels by province. Tables 2 and 3 present summary comparisons, based on the same data. Regular Programmes: Variation Across Provinces Attrition rates in regular core programmes after the end of compulsory French are summarized in Table 2. These programmes differ in two ways: what the last grade for compulsory French is, and where this grade is situated in terms of the transition to high school. In British Columbia and the Yukon, French ends well 34
Table 1 Attrition in Programmes After End of Compulsory French A. British Columbia* 8 36,110 36,296 36,397 9 21,248 21,124 21,462 10 18,012 15,956 16,262 11 14,636 14,149 12,422 12 5,108 5,399 5,446 Grade 8>9 39.2 41.5 40.9 Grade 9>10 24.1 24.9 23.0 Grade 10>11 21.4 21.4 22.1 Grade 11>12 64.2 63.1 61.5 Grade 9>11 40.4 41.5 Grade 9>12 77.1 Grade 8>11 64.4 Grade 8>12 projected 86.3 * A second language is compulsory.this is usually, but not necessarily, French. B. Yukon 8 260 275 248 9 205 175 129 10 150 109 99 11 105 105 99 12 34 23 22 Grade 8>9 31.0 32.7 53.1 Grade 9>10 28.9 46.8 43.4 Grade 10>11 34.0 30.0 9.2 Grade 11>12 73.6 78.1 79.0 Grade 9>11 50.2 51.7 Grade 9>12 89.6 Grade 8>11 66.7 Grade 8>12 projected 87.2 C. Ontario 2000-2001 2001-2002 2002-2003 9 71,185 70,795 68,448 3,419 3,218 2,863 10 16,192 14,820 15,714 2,638 2,304 2,144 11 10,431 6,706 6,644 1,941 1,725 1,498 12 814 652 4,016 64 67 1,029 OAC/13** 5,996 5,374 1,289 1,149 0 Grade 9>10 77.6 79.2 77.8 23.8 32.6 33.4 Grade 10>11 56.7 58.6 55.2 21.3 34.6 35.0 Grade 11>12 42.5 42.2 40.1 29.2 37.4 40.3 Grade 9>11 90.7 90.7 50.2 56.2 Grade 9>12 94.4 70.3 E. Nova Scotia 2000-2001 2001-2002 2002-2003 9 9,718 9,303 9,220 827 771 671 10 3,854 3,510 2,891 553 450 465 11 1,776 1,486 1,382 445 387 309 12 1,014 918 806 260 227 298 Grade 9>10 61.0 63.9 68.9 33.8 45.6 39.7 Grade 10>11 58.4 61.4 60.6 22.7 30.0 31.3 Grade 11>12 47.9 48.3 45.8 38.2 49.0 23.0 Grade 9>11 85.0 85.8 53.7 62.6 Grade 9>12 91.8 64.3 G. Newfoundland 2000-2001 2001-2002 2002-2003 9 5,819 5,727 5,418 10 2,640 2,564 2,417 112 77 127 11 1,557 1,333 1,212 61 99 111 12 526 396 347 40 27 71 Grade 9>10 57.1 55.9 57.8 Grade 10>11 45.2 49.5 52.7 18.7 11.6-44.2 Grade 11>12 70.6 74.6 74.0 11.1 55.7 28.3 Grade 9>11 78.3 79.2 Grade 9>12 94.4 Grade 10>12 86.1 86.9 64.0 36.6 D. New Brunswick 10 5,080 4,955 4,781 11 1,097 921 901 12 508 483 478 Grade 10>11 78.3 81.9 81.8 Grade 11>12 55.8 56.0 48.1 Grade 10>12 90.5 90.6 F. Prince Edward Island 9 1,525 1,451 1,406 10 551 562 535 11 286 216 301 12 156 115 206 Grade 9>10 60.6 63.1 63.1 Grade 10>11 54.0 60.8 46.4 Grade 11>12 34.2 59.8 4.6 Grade 9>11 84.6 80.3 Grade 9>12 85.3 35
FSL Enrolment Summary Statistics Table 2 Comparison of Attrition in Programs Aftrer End of Compulsory French Last grade of compulsory French First Grade of High School % Net loss in first non-compulsory year (Average) % Net loss in first % Net loss to first % Net loss to last British Columbia 8 11 40.5 21.7 64.4 86.3 Yukon 8 11 38.9 24.4 66.7 87.2 Ontario 9 9 78.2 N/A N/A 94.4 New Brunswick 10 9 80.7 N/A N/A 90.6 Nova Scotia 9 10 64.6 64.6 64.6 91.8 Prince Edward Island 9 10 62.3 62.3 62.3 85.3 Newfoundland & Labrador 9 10 56.9 56.9 56.9 94.4 Net loss in the first year in which French is not compulsory varies enormously across programmes. In British Columbia and the Yukon, where compulsory French ends several grades before high school, there is a comparatively low attrition rate, with net losses of less than 25% of enrolment. In contrast, we find single year net losses in the range of 55-65% for the three Atlantic Provinces where the end of compulsory French also marks the transition to high school. Finally, in Ontario and New Brunswick where compulsory French ends after the transition to high school, single-year net losses in the first year of optional French are approximately 80%. In British Columbia and the Yukon, the transition year to high school is similarly characterized by low attrition rates compared to the three Atlantic Provinces making both transitions (to optional French and to high school) at once. However, in the two western jurisdictions, the cohort entering high school has already experienced two prior years of attrition from optional French programmes. This raises the question of how the cumulative rate from the end of compulsory French to the beginning of high school compares to the single-year rate in the Atlantic jurisdictions where both changes happen at once. Table 2 also shows cumulative, multi-grade attrition rates for the first and last years of high school. In fact, the three-year cumulative attrition rates in British Columbia and the Yukon between Grades 8 and 11 (when high school begins) are similar to the single year net attrition rates in Nova Scotia, Prince Edward Island, and Newfoundland and Labrador. In other words, making French compulsory to the beginning of high school essentially compresses a slower attrition rate, which would have taken place over several years, into a single year big bang. Making French optional at a later grade level does not appear to encourage students to continue in the programme past the compulsory period. 36 4 before the start of high school (Grade 11). In Nova Scotia, Prince Edward Island, and Newfoundland and Labrador compulsory French ends just before the start of high school. The transition to high school corresponds to the transition from compulsory to optional French. In Ontario and New Brunswick, French is compulsory through the transition, being compulsory for one year of high school in Ontario and two years in New Brunswick. Table 2 shows year-on-year attrition rates at two junctures. The first is the point at which French is no longer compulsory for students; the second is the transition to high school. We see a similar pattern in comparing the net attrition to the last. The rates are remarkably similar across jurisdictions, regardless of the different schedules of grade-specific attrition rates. By the last grade of high school, attrition rates in British Columbia, the Yukon, and Prince Edward Island are virtually identical. Cumulative attrition is somewhat higher in Nova Scotia and New Brunswick, but only Ontario and Newfoundland and Labrador stand out as having a notably higher attrition rate. These results are disturbing. They suggest that making French compulsory to later grades and in particular, into the first years of high school, simply produces a pent-up demand to leave at the first opportunity, rather than encouraging more students to stay till the end of high school. Alternate core programmes are considerably more successful in retaining students to the end of high school. Regular Core and Programmes in Ontario and Nova Scotia: Comparative Net Attrition Rates Table 3 shows that in Ontario alternate core programmes accounted for less than 5% of total core enrolment in Grade 9 in 1999-2000. Four years later, alternate core students in the last year of high school (Grade 12) represent 20% of total core enrolment. We find the same pattern in Nova Scotia. Alternate core represents less than 10% of Grade 9 core enrolment in 1999-2000, but over 25% of Grade 12 core enrolment four years later. These graduate outcomes demonstrate that these alternate core programmes are considerably more successful in retaining students to the end of high school. Table 3 Comparison of Attrition in Regular Core and Special Core Programs in Ontario and Nova Scotia Core % Net loss to last year of high school Special Core as % of All Core Ontario Grade 9 (1999-2000) 72,193 3,461 94.4 4.6 Grade 12 (2002-2003) 4,016 1,029 70.3 20.4 Nova Scotia Grade 9 (1999-2000) 9,878 835 91.8 7.8 Grade 12 (2002-2003) 806 298 64.3 27.0