WEST ST. PAUL SCHOOL DISTRICT #197 ENROLLMENT PROJECTIONS

WEST ST. PAUL SCHOOL DISTRICT #197 ENROLLMENT PROJECTIONS Hazel H. Reinhardt 5/1/2012

WEST ST. PAUL SCHOOL DISTRICT ENROLLMENT PROJECTIONS Executive Summary Enrollment in the West St. Paul Public Schools is projected to increase through the ten year projection period; however, growth will be faster in the first five years o K- 4 is essentially flat but could increase by 100 students in the highest projection o Grades 5-8 will increase by 175 to 248 students in the next five years o Grades 9-12 are essentially flat in the short term (+68 to +84) but could increase by 376 students in ten years in the highest projection Enrollment patterns are atypical o The West St. Paul Public Schools enroll a higher percentage of nonresident students than a typical Twin Cities metro area district (17.1 percent) o A higher percentage of district residents open enroll out than is typical (12.2 percent) o Nonpublic enrollment is exceptionally large accounting for 21 percent of the district's school age population The combination of nonpublic schools and public options results in a low student capture rate for the West St. Paul Public Schools (62.9 percent) Because the housing stock is older, the K- 12 student yield per dwelling unit is low Very few additional housing units are projected for the district in the near term. Redevelopment may occur in the long term 1

CHAPTER 1 DISTRICT WIDE ENROLLMENT PROJECTIONS Introduction The West St. Paul School District includes parts of six northern Dakota County cities, that is, Eagan, Inver Grove Heights, Lilydale, Mendota Heights, Sunfish Lake and West St. Paul. In a school district, the school age population is closely related to the other population characteristics of the district. A prime example is the relationship between age and the number of births in a school district. A larger number of women of prime childbearing age results in more births and larger kindergarten classes five to six years later. Another example is the relationship between age and changing one s residence. Families with children under 18 years that move from one locale to another can have a major effect on school enrollment. Population turnover is ongoing in a mobile society and enrollment changes throughout the school year as families and children move. While population changes affect the total number of school age children residing in a school district, Minnesota students and their families have education choices. These choices also effect enrollment in the district s schools. Therefore, when analyzing public school enrollment, choice must be considered as well as population dynamics. Choice includes nonpublic schools, home schools, and the public options of open enrollment, charter schools and alternative schools. Two others choices exist: a) dropping out of high school, and b) attending a kindergarten alternative. Enrollment Trends Enrollment in the West St. Paul Public Schools Current Enrollment/Past Trends Total enrollment in the West St. Paul Public Schools is - 207 students or - 4.4 percent lower in 2011-12 than in 2002-03. However, during this period, enrollment bottomed out at 4,376 students in 2008-09. Since 2008-09, enrollment has inched upward, taking a bigger jump upward in the past year. K- 12 TOTAL ENROLLMENT 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 4,731 4,712 4,662 4,607 4,486 4,429 4,376 4,380 4,402 4,524 Source: West St. Paul School District, Fall Enrollment. Excludes Early Childhood and the ALC 2

To better understand enrollment change, it is important to understand the components of that change. Like all population change, school enrollment change results from two different phenomena natural increase/decrease and net migration. The difference between the size of the incoming kindergarten class and the previous year s Grade 12, called natural increase or decrease, measures the change in past birth numbers or cohort change. For example, the Baby Boom (1946-1964) and the Baby Bust (1965-1976) set in motion cycles of rising and falling enrollment that are reflected as natural increase/decrease. The life stage of suburbs is also reflected in natural increase/decrease. As the next table shows, there is no pattern in natural increase/decrease. Annual fluctuations are the norm. This phenomenon contributed to enrollment decline mid decade but had no effect in the past four years. COMPONENTS OF ENROLLMENT CHANGE Natural Increase/ Decrease Total Fall to Fall # % 2002 to 2003-19 - 0.4% 32-51 2003 to 2004-50 - 1.1% 1-51 2004 to 2005-55 - 1.2% 21-76 2005 to 2006-121 - 2.6% - 47-74 2006 to 2007-57 - 1.3% - 37-20 2007 to 2008-53 - 1.2% 8-61 2008 to 2009 4 0.1% - 1 5 2009 to 2010 22 0.5% - 14 36 2010 to 2011 122 2.8% 8 114 Net Migration The other phenomenon affecting school enrollment is migration, an indirectly derived estimate. Migration is the term used when people move across a boundary or border, in this case, the school district boundary. Net migration is calculated by the progression from grade- to- grade of public school students. For example, public school Kindergarten students are moved to Grade 1 in the following year, Grade 1 students to Grade 2, etc. Because the probability of death is very low among children, the same number of children should be in the next higher grade the following year. Therefore, if the number of students changes, migration is assumed to have occurred. A positive number indicates a net flow into the public schools and a negative number reflects a net flow out of the public schools. This method for estimating migration does not distinguish between physical movement across the district s boundaries and education choices, such as transferring from a nonpublic school to a public school, transferring to a charter school or open enrolling in another public school outside the district. Further, students who move into or out of a school district but never enroll in the district s public schools are not reflected in the migration numbers in this report. Based on the described methodology, net in migration was negative every year until 2008-09. Since then, net migration has been positive. The positive net migration resulted in a turnaround from enrollment decline to enrollment growth. 3

Student Choices in the West St. Paul School District The number of education options available affects resident enrollment in a district's public schools. Nonpublic schools have been an option for many years. More recently, home schools became another option. Since its inception, public school options are attracting more students every year. Open enrollment allows residents of one district to attend public schools in another district. Charter schools are another public option. All these choices mean competition for a district s public schools. Nonpublic Enrollment and Home Schools Today, nonpublic enrollment falls into two categories traditional nonpublic schools and home schools. Most traditional nonpublic schools are associated with religious institutions and many home school curriculums also have religious ties. NONPUBLIC SETTINGS Traditional Year Nonpublic Schools Home Schools Total 2002-03 1,945 (1,396) 76 1,472 2003-04 2,608 (1,336) 83 1,419 2004-05 2,720 (1,336) 93 1,429 2005-06 2,498 (1,274) 93 1,367 2006-07 2,908 (1,293) 106 1,399 2007-08 2,654 (1,253) 92 1,345 2008-09 2,692 (1,240) 93 1,333 2009-10 2,666 (1,211) 82 1,293 2010-11 2,364 (1,119) 83 1,202 2011-12 2,537 (1,110) 83 1,193 Numbers in parentheses are estimates of district residents. All nonpublic elementary schools located in the district except one had a decline in enrollment Source: West St. Paul School District In Minnesota, 8.1 percent of all enrolled students were enrolled in traditional nonpublic schools and 1.9 percent of enrolled students were home schooled in 2010-11. In the West St. Paul School District, 19.2 percent of enrolled students were in traditional nonpublic schools and 1.4 percent were home schooled in 2010-11. The proportion of students in nonpublic settings is higher than the statewide percentages. Combining home school students and nonpublic students, 20.7 percent of West St. Paul district residents were in nonpublic settings. In Minnesota, 10.0 percent were enrolled in nonpublic settings. In the past ten years, traditional nonpublic enrollment decreased statewide while home schooled children increased. In the West St. Paul School District, traditional nonpublic enrollment among residents is declining as well. The number of home schooled students is flat. 4

Public Options Open Enrollment. Open enrollment allows Minnesota students to attend public schools outside their district of residence. The application to open enroll is made by the student and his/her parents and families generally provide their own school transportation. No tuition is charged. Some students attend public schools outside their home district because their home district enters into an agreement with another district, usually to provide specialized services. This is called a tuition agreement, but this arrangement is not technically a student choice. Since its beginning, open enrollment has attracted more and more students statewide and in the West St. Paul School District. In 2010-11, 754 nonresident students open enrolled into the West St. Paul Public Schools while 711 district residents attended public schools elsewhere through open enrollment. PUBLIC OPTIONS In Out Year Open Enrollment & Tuition Open Enrollment & Tuition Charter Schools Other Options (ALC only) Net 2002-03 404 488 0 n.a. - 84 2003-04 492 408 57 n.a. 27 2004-05 506 404 84 n.a. 18 2005-06 545 418 110 n.a. 17 2006-07 550 425 140 n.a. - 15 2007-08 574 457 189 n.a. - 72 2008-09 619 584 229 n.a. - 194 2009-10 669 661 232 n.a. - 224 2010-11 754 711 242 n.a. - 199 2011-12 n.a. n.a. n.a. n.a. n.a. Source: West St. Paul School District Nonresident students who open enroll into the West St. Paul Public Schools accounted for 17.1 percent of West St. Paul's total enrollment in 2010-11. Students leaving the district to attend public schools elsewhere represented 12.2 percent of district school age residents. In 2010-11, 6.4 percent of Minnesota students chose open enrollment. Charter Schools. Charter schools are another public education option. While 3.9 percent of Minnesota students attend charter schools, 4.2 percent of West St. Paul School District residents attend charter schools. Other Public Options. Other public options include alternative learning centers (ALCs), alternative learning programs, private alternative programs, and post- secondary education options (PSEOs). No West St. Paul residents are reported to be attending an ALC in 2010-11. 5

Residents in the West St. Paul School District are more likely to choose an educational choice other than their local public school than residents of other Minnesota school districts. K- 12 Capture Rate of District School Age Residents To estimate a capture rate, there must be an estimate of a district s school age population or more precisely, a district s school age population that is enrolled in school. A district s enrolled population can be constructed based on resident students in the district s schools and then adding district residents attending traditional nonpublic schools, residents being home schooled and residents opting for open enrollment out, charter schools and other public options. Based on 2002-03 and 2010-11, the estimated resident school age population decreased from 6,290 to 5,813, a decrease of - 477 students or - 7.6 percent. During this same period, resident enrollment in the West St. Paul Public Schools decreased by - 672 students or - 15.5 percent. These percentages indicate that the West St. Paul Public Schools market share decreased. A capture rate is another term for market share. Based on the estimated 2010-11 enrolled population of 5,813, the West St. Paul Public Schools captured 62.9 percent of the district s school age population. In 2002-03, the capture rate was 68.8 percent. West St. Paul's current market share is lower than that of most other suburban Twin Cities school districts. WEST ST. PAUL SCHOOL DISTRICT ESTIMATED RESIDENT SCHOOL AGE POPULATION West St. Paul Public Schools Year Resident Enrollment Nonpublic Settings Public Options Total 2002-03 4,330 1,472 488 6,290 2003-04 4,232 1,419 465 6,116 2004-05 4,162 1,429 488 6,079 2005-06 4,059 1,367 528 5,954 2006-07 3,931 1,399 565 5,895 2007-08 3,853 1,345 646 5,844 2008-09 3,766 1,333 813 5,912 2009-10 3,726 1,293 893 5,912 2010-11 3,658 1,202 953 5,813 2011-12 3,745 1,193 n.a. n.a. History of Resident Enrollment by Grade The history of public school enrollment contains several patterns with implications for the future. First, the kindergarten class has fluctuated in size including a downward trend in the mid years of the decade. However, in the past three years, kindergarten rebounded and is back to its 2002-03 and 2003-04 size. 6

ENROLLMENT Grade 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08 2008-09 2009-10 2010-11 2011-12 K 363 349 346 337 342 339 338 374 348 378 1 354 357 332 338 329 326 328 321 373 352 2 310 332 346 326 322 328 335 341 327 383 3 336 307 320 341 309 326 328 321 327 340 4 363 334 316 314 331 305 311 341 331 342 5 358 348 335 297 299 326 295 299 345 330 6 377 364 366 329 308 297 323 310 294 368 7 393 390 373 357 327 325 291 342 321 336 8 401 394 385 392 374 354 333 290 331 306 9 427 425 405 414 388 378 373 343 343 370 10 367 416 414 378 408 402 370 376 345 344 11 365 351 408 395 373 393 376 360 347 329 12 317 345 316 389 376 330 375 362 370 346 Total 4,731 4,712 4,662 4,607 4,486 4,429 4,376 4,380 4,402 4,524 Source: West St. Paul School District. Excludes Early Childhood and ALC As a result of the uptick in kindergarten, the number of students per grade is larger in the elementary grades than in the high school grades. The largest number of students is in Grade 2, suggesting that enrollment is poised to increase, depending, of course, on future kindergarten size. Another way of expressing the differences by grade is to look at the average number of students per grade. For example, the average elementary grade has 359 students. The average middle school grade has 335 students, while the average for the high school grade is 347 students. The high school grades reflect a net inflow from nonpublic schools. Minnesota's largest graduating high school class since 1978 graduated in 2009. State wide, graduating classes will be getting smaller as the high school population gets smaller. Based on West St. Paul s elementary enrollment, West St. Paul will graduate larger classes in the 2020s. Enrollment Projections Projection Background Some factors affecting future school enrollment are known. However, other important factors are less clear. First, the trends around which there is confidence. Trends Where Confidence is High Aging. The population in the U.S. and Minnesota is aging. By 2020, 16-17 percent of Minnesota s population will be 65 years old or older. In 2010, the elderly made up 12.9 percent of the population. There is no historical precedent for this high proportion of older population; therefore, society is entering uncharted waters as to the effects of this change. However, we 7

know that aging will affect the housing market and reduce geographic mobility because older people move less frequently than younger people. Decrease in the school age population per household. From 2000 to 2010, the number of school age children per household decreased sharply as Baby Boomer households empty nested and started to age in place. After 2010, households with children will be headed primarily by Generation X parents who are members of a much smaller generation. Gen X (1965-1976) is only 60 percent the size of the Baby Boom (1946-1964) generation, which means the percentage of households with 5-17- year- olds will continue to decrease but more slowly. Shift in size of key adult age groups. The size of the Baby Boom generation and the Baby Bust generation, reflected in the next table, will result in significant changes in the size of adult age groups, which in turn will affect the demand for new housing units. The modest increase in the 20-34 year- old population between 2010 and 2020 is especially significant for the demand for first homes (includes apartments) and the decrease in 35-54 year- olds will affect the move up market. Growth in the 55+ year- old markets will create demand for housing for mature adults and seniors; however, these units will not yield school age children. AGE TWIN CITIES METRO AREA POPULATION (11- COUNTIES) Age 2000 2010 2020 Change 2000-2010 Change 2010-2020 20-34 yrs 629,898 693,040 725,670 63,142 32,630 35-54 yrs 902,531 981,060 952,870 78,529-28,190 55-64 yrs 217,880 359,720 460,080 141,840 100,360 65+ yrs 275,183 338,110 499,110 62,927 161,000 Sum 2,025,492 2,371,930 2,637,730 346,438 265,800 Source: Minnesota Demographic Center, 2007 These population changes by age point to a future very different from the recent past. Demand for additional housing will slow because the adult population age 20+ will increase more slowly and the 35-54 year- old age group that helped fuel the housing boom will decrease from 2010-2020. Furthermore, 60 percent of the increase in adults 20 years of age and older will be persons 65+ years of age. There may be more sellers than buyers in the housing market. Fertility. Today, completed fertility is near the replacement level. Completed fertility refers to the number of children born per woman throughout her childbearing years. In the U.S., White non Hispanic and Black women have near or below replacement fertility. (Replacement is 2.11 children per female at the end of childbearing.) Fertility rates are likely to remain at or near replacement levels. Hispanic women and immigrant women have higher fertility. Births. Births fell after 1990 in the U.S. and in Minnesota; however, since 2003, births have been increasing with the exception of the past three years. In 2007, births were higher than at any time since 1964; however, 2007 births were well below the peak Minnesota birth year of 1959 8

(88,000 resident births). Births fell in the U.S. and Minnesota in 2008, 2009 and 2010. These declines are attributed to the poor economy. As the history of resident births shows, from 1996 to 2010, resident births in Minnesota increased 7.4 percent while resident births in Dakota County decreased - 8.5 percent, although Dakota County births increased through 2006 but have fallen since. RESIDENT LIVE BIRTHS Dakota Year Minnesota County 1996 63,681 5,651 1997 64,491 5,671 1998 65,207 5,690 1999 65,953 5,738 2000 67,451 5,786 2001 66,617 5,835 2002 68,037 5,883 2003 70,053 5,931 2004 70,617 5,961 2005 70,950 5,991 2006 73,515 6,021 2007 73,675 5,592 2008 72,382 5,570 2009 70,617 5,453 2010 68,407 5,173 Source: Minnesota Department of Health Enrollment cycles. Births will increase again and a third enrollment cycle will occur in the first half of this century. Already, kindergarten classes are increasing in many districts, a sign of the beginning of this third enrollment cycle. The end of the third enrollment cycle is projected to be around 2040. (From start to finish, these cycles last about 30 years.) Unknowns The unknowns reflect recent changes such as the collapse of the housing market and tighter credit. Another unknown is the longer- term effect of the recession on domestic migration and international immigration, especially in a jobless recovery. Furthermore, will attitude and behavior changes prompted by the recession last? Collapse of the housing market and tighter credit. A high level of mobility was possible with a robust housing market with rapid appreciation and easy credit. This has now changed with the collapse of the housing market (falling home prices) and tighter credit. The recession. Although the recession is officially over, the stagnant job market slowed population movement between states and immigration from abroad. Minnesota felt the effect 9

of these changes as population growth slowed in the Twin Cities metro area. The recession also increased public school enrollment as some families decided that nonpublic schools were beyond their current financial resources. Cohort Survival Method The most common and most robust model for projecting school enrollment is the cohort survival method. The first step in the cohort survival method is aging the population. In a standard cohort survival model, aging the population involves estimating the number of deaths expected in an age group before it reaches the next older age group. When the cohort survival method is applied to school enrollment, the first step is to move a grade to the next higher grade. However, because mortality is so low in the school age population, the entire grade is assumed to survive to the next higher grade in the following year. Once a grade or cohort has been aged to the next grade, net migration is added to or subtracted from that grade. Using survival rates accomplishes both aging and migration in a single step. Over time, the size of a cohort will increase or decrease as a result of migration as its progresses through the grades. For example, the 2002-03 kindergarten class had 363 members. This same cohort had 370 members in Grade 9 in 2011-12. The projection of future kindergarten class size is important in long- term enrollment projections because these students will be in school over the life of the projection. If a school census exists, it is a resource for short- term kindergarten projections, i.e., a couple of years. However, school censuses are notoriously inaccurate for children less than four years of age. To project kindergarten, the best theoretical approach, but the least practical, is to project births based on the age of the female population. These birth projections then must be survived to age five and then adjusted for migration to yield kindergarten projections. Determining the age of females in a school district is the first challenge and then, many assumptions must be made, making this approach impractical. A simpler approach is to use resident births as a proxy for kindergarten five years later. Of course, not every child born in the district will enter the district's kindergarten classes five to six years later. However, some "native born" children who move out before enrolling in kindergarten will be replaced by children born elsewhere who move into the district before entering kindergarten. If the number of "ins" and "outs" are equal, the net effect is zero and the kindergarten class would be 100 percent of resident births. However, no public school district captures all its potential. Some resident kindergarten students attend private schools or are home schooled. Others may attend a charter school or open enroll at another district. Therefore, a public school's capture rate should be less than 100 percent. If a public school capture rate is 100 percent or higher, more preschool children are moving into the district than leaving (net in migration). Although births five years earlier are a good proxy for a kindergarten class, kindergarten students must be 5 years- old by September 1. This requires an estimate of the number of births occurring after September 1 in any calendar year. As a rule of thumb, about one- third of births occur in the last third of a calendar year. This means that about one- third of the kindergarten class is born six years earlier not five years earlier. For example, one- third of the 2011-12 kindergarten class was born in 10

2005 and two- thirds were born in 2006. Adjusting birth years to fit the age requirements of kindergarten creates a kindergarten pool. If births are used as a kindergarten proxy, kindergarten projections are available for only a few years into the future. To extend kindergarten projections another five years, West St. Paul's kindergarten will be projected based on the Minnesota Demographic Center s projection of Dakota County resident births. Kindergarten Assumptions Upon special request, the Minnesota Department of Health will provide resident births by address so the births can be geocoded to a school district s boundaries. Some out- of- wedlock births may be withheld because unmarried parents may choose whether to make birth information by address public. As long as this segment s behavior does not change, a history of a district s resident births should be reliable. (All resident births are reported in city and county data.) The next table shows resident births for the West St. Paul School District based on the months corresponding with the school year. As the birth data show, births fluctuated over the period; however, a trend line shows that district births declined slightly. DISTRICT RESIDENT LIVE BIRTHS SEPTEMBER 1 TO AUGUST 31 1996-1997 495 1997-1998 424 1998-1999 468 1999-2000 458 2000-2001 434 2001-2002 414 2002-2003 453 2003-2004 457 2004-2005 454 2005-2006 468 2006-2007 522 2007-2008 452 2008-2009 479 2009-2010 450 2010-2011 417 Source: Minnesota Department of Health Two kindergarten pools were constructed. The West St. Paul district pool is based on resident births in the district. The Dakota County pool is based on resident births in Dakota County. When West St. Paul s kindergarten is compared to a kindergarten pool, a kindergarten capture rate can be calculated. With county birth data available through 2010, the kindergarten classes through 2015-16 can be projected from actual births. 11

The next table shows that the West St. Paul pool has fluctuated but has been between 450 and 470 most years. The district pool is projected to peak based on the births of 2006 and 2007 and then drop to its usual level. However, the births of 2010 and 2011 bring the district pool to a low of 417. The Dakota County pool peaked with the births of 2005-2006 and is declining. The table also shows West St. Paul s kindergarten as a percentage of each of these pools. In the past ten years, West St. Paul s kindergarten class fluctuated between 73.3 percent and 82.3 percent of the West St. Paul district pool. In the past four years, the capture rate ranged between 74.6 percent and 81.8 percent. Calculating an average of the kindergarten capture rates for two or more years smooths out annual fluctuations and produces a more typical capture rate for that period. For example, the average of the past four years is 78.5 percent while the average of the past three years is 79.8 percent. The average of the past two years is 78.8 percent. West St. Paul s kindergarten as a percentage of the Dakota County pool fluctuates between 5.8 percent and 6.4 percent. While the range is small, so is the district's proportion of county births, which is a weakness of this data set. The average of the past four years is 6.0 percent while the average of the past three years is 6.1 percent. KINDERGARTEN POOLS AND WEST ST. PAUL PUBLIC SCHOOLS KINDERGARTEN AS A PERCENT OF EACH POOL Birth Years West St. Paul District Pool Percentage Dakota Co. Pool Percentage Kindergarten Year 1996; 1997 495 73.3% 5,665 6.4% 2002-03 1997; 1998 424 82.3% 5,683 6.1% 2003-04 1998; 1999 468 73.9% 5,723 6.0% 2004-05 1999; 2000 458 73.6% 5,770 5.8% 2005-06 2000; 2001 434 78.8% 5,818 5.9% 2006-07 2001; 2002 414 81.9% 5,867 5.8% 2007-08 2002; 2003 453 74.6% 5,915 5.7% 2008-09 2003; 2004 457 81.8% 5,951 6.3% 2009-10 2004; 2005 454 76.7% 5,981 5.8% 2010-11 2005; 2006 468 80.8% 6,011 6.3% 2011-12 2006; 2007 522 5,734 2012-13 2007; 2008 452 5,577 2013-14 2008; 2009 479 5,492 2014-15 2009; 2010 450 5,265 2015-16 2010; 2011 417 2016-17 These data suggest there is a range in which future kindergarten class size is likely to fall. To capture this expected range, three kindergarten projections were made. Two projections are based on the West St. Paul district pool and the other projection is based on the Dakota County Pool. The first projection is designed to capture the low end of the recent range in the West St. Paul district pool. For the projections, the low kindergarten assumption will be based on 78.5 percent of the West St. Paul Area pool and the high assumption will be based on 79.8 percent of the West St. Paul Area pool. A third projection will be based on the Dakota County pool. A percentage of 6.1 will be used. 12

To extend the kindergarten projections beyond 2015-16, West St. Paul's kindergarten will be based on the Minnesota Demographic Center s projections of Dakota County resident births. However, the birth projections for Dakota County are higher than actual births beginning in 2007. To compensate for this over projection, the Demographic Center's projections will be adjusted downward by 5 percent to reflect the growing variance between actual births and projected births. To create a West St. Paul Area kindergarten pool, district births must be projected. In the past three years, the district's share of Dakota County resident births averages 8.4 percent. Future district resident births were projected to be 8.4 percent of the county's resident births. RESIDENT BIRTHS DAKOTA COUNTY Births Original Projection Actual Difference Adjusted Projection Year 2005 5,593 5,991-6.6% 2006 5,612 6,021-6.8% 2007 5,632 5,592 0.7% 2008 5,651 5,570 1.5% 2009 5,671 5,453 4.0% 2010 5,690 5,173 10.0% 2011 5,738 5,451 2012 5,786 5,497 2013 5,835 5,543 2014 5,883 5,589 2015 5,931 5,634 2016 5,961 5,663 Adjusted projection is 5 percent lower than original projection in 2011 and thereafter. (Assumes that births in 2010 were exceptionally low) Source: Minnesota Demographic Center KINDERGARTEN POOLS Year West St. Paul County 2016-17 417 5,359 2017-18 460 5,482 2018-19 464 5,528 2019-20 468 5,574 2020-21 472 5,619 2021-22 475 5,653 Applying the kindergarten capture rates discussed earlier to the projected kindergarten pools result in the kindergarten projections shown in the next table. As this table shows, the kindergarten projections based on the Dakota County pool are lower than the projections based on the district's pool. 13

Because the district is such a small proportion of the county, only the kindergarten projections based on the district pool will be used to make enrollment projections. The kindergarten projections show smaller kindergarten classes for several years. Because the kindergarten pool increases in the later projection years, so do the kindergarten projections. The housing market should recover by 2016 making this increase reasonable. Furthermore, the large Gen Y population will be in its prime childbearing years, especially after 2020. KINDERGARTEN PROJECTIONS Dakota West St. Paul District Pool County Pool Year @78.5% @79.8% @6.1% 2011-12 378 378 378 2012-13 410 417 350 2013-14 355 361 340 2014-15 376 382 335 2015-16 353 359 321 2016-17 327 333 327 2017-18 361 367 334 2018-19 364 370 337 2019-20 367 373 340 2020-21 371 377 343 2021-22 373 379 345 Net Migration Assumptions The concept and method of calculating migration was explained earlier in this report. However, the limitations of the methodology are worth repeating. The method of calculating migration does not distinguish between physical movement across a district s boundaries and education choices, such as transferring from a nonpublic school to a public school, transferring to a charter school or open enrolling in another public school. Further, students who move into or out of a school district but never enroll in the district s public schools are not reflected in the migration numbers in this report. In the past ten years, annual net migration fluctuated from year to year but was negative for all but the past three years. In the past year, net in migration was significantly greater. The large positive number resulted from larger net in migration in the elementary and middle school grades. The next table shows net migration aggregated by the elementary grades (Kindergarten- Grade 4), the middle school grades and the high school grades. Kindergarten to Grade 4 net migration was negative until the past two years. Net in migration at Grade 9 is typical in Minnesota, however, in the past two years net in migration to Grade 9 was sufficiently large enough to produce net in migration at the high school grades. 14

NET MIGRATION SCHOOL YEAR TO SCHOOL YEAR 02 to 03 03 to 04 04 to 05 05 to 06 06 to 07 07 to 08 08 to 09 09 to 10 10 to 11 K- 4-33 - 31-25 - 51-17 - 17-5 1 42 5-8 5 23-15 11 37-11 21-1 49 9-12 - 23-43 - 36-34 - 40-33 - 11 36 23 Total - 51-51 - 76-74 - 20-61 5 36 114 In the West St. Paul Public Schools, net out migration occurs between Kindergarten and Grade 1, which is unusual. Net in migration between Kindergarten and Grade 1 is typical in Minnesota's public schools. The progression from grade to grade in the remaining elementary grades fluctuates from year to year but has been positive in recent years. Migration patterns by grade also fluctuate from year to year in the middle school grades but are usually positive. Between Grade 8 and Grade 9, West St. Paul has an influx of students. This is the time when nonpublic students transfer into the public schools. After Grade 10, the high school grades show losses. This pattern is typical of Minnesota's public high schools. NET MIGRATION BY GRADE SCHOOL YEAR TO SCHOOL YEAR 02 to 03 03 to 04 04 to 05 05 to 06 06 to 07 07 to 08 08 to 09 09 to 10 10 to 11 K to 1-6 - 17-8 - 8-16 - 11-17 - 1 4 1 to 2-22 - 11-6 - 16-1 9 13 6 10 2 to 3-3 - 12-5 - 17 4 0-14 - 14 13 3 to 4-2 9-6 - 10-4 - 15 13 10 15 4 to 5-15 1-19 - 15-5 - 10-12 4-1 5 to 6 6 18-6 11-2 - 3 15-5 23 6 to 7 13 9-9 - 2 17-6 19 11 42 7 to 8 1-5 19 17 27 8-1 - 11-15 8 to 9 24 11 29-4 4 19 10 53 39 9 to 10-11 - 11-27 - 6 14-8 3 2 1 10 to 11-16 - 8-19 - 5-15 - 26-10 - 29-16 11 to 12-20 - 35-19 - 19-43 - 18-14 10-1 Total - 51-51 - 76-74 - 20-61 5 36 114 Percent - 1.1-1.1-1.6-1.6-0.4-1.4 0.1 0.8 2.6 For making projections, migration is converted into survival rates. Survival rates show the percentage change from grade to grade each year. For example, 1.00 indicates no change or 100 percent of the grade progressed to the next highest grade. Any number over 1.00 reflects the 15

percentage increase while a number below 1.00 reflects the percentage decrease. For example, 0.98 indicates a 2 percent decrease. SURVIVAL RATES SCHOOL YEAR TO SCHOOL YEAR 02 to 03 03 to 04 04 to 05 05 to 06 06 to 07 07 to 08 08 to 09 09 to 10 10 to 11 K to 1 0.983 0.951 0.977 0.976 0.953 0.968 0.950 0.997 1.011 1 to 2 0.938 0.969 0.982 0.953 0.997 1.028 1.040 1.019 1.027 2 to 3 0.990 0.964 0.986 0.948 1.012 1.000 0.958 0.959 1.040 3 to 4 0.994 1.029 0.981 0.971 0.987 0.954 1.040 1.028 1.046 4 to 5 0.959 1.003 0.940 0.952 0.985 0.967 0.961 1.012 0.997 5 to 6 1.017 1.052 0.982 1.037 0.993 0.991 1.051 0.983 1.067 6 to 7 1.034 1.025 0.975 0.994 1.055 0.980 1.059 1.035 1.143 7 to 8 1.003 0.987 1.051 1.048 1.083 1.025 0.997 0.968 0.953 8 to 9 1.060 1.028 1.075 0.990 1.011 1.054 1.030 1.183 1.118 9 to 10 0.974 0.974 0.933 0.986 1.036 0.979 1.008 1.006 1.012 10 to 11 0.956 0.981 0.954 0.987 0.963 0.935 0.973 0.923 0.954 11 to 12 0.945 0.900 0.953 0.952 0.885 0.954 0.963 1.028 0.997 COMPARISON OF SURVIVAL RATES AVERAGED Grade Past 2 years Past 4 years First 5 years K to 1 1.004 0.982 0.968 1 to 2 1.023 1.029 0.968 2 to 3 1.000 0.989 0.980 3 to 4 1.037 1.017 0.992 4 to 5 1.005 0.984 0.968 5 to 6 1.025 1.023 1.016 6 to 7 1.054* 1.054* 1.017 7 to 8 0.961 0.986 1.034 8 to 9 1.067* 1.067* 1.033 9 to 10 1.009 1.001 0.981 10 to 11 0.939 0.946 0.968 11 to 12 1.013 0.986 0.927 *Average based on 3 years to smooth huge fluctuations One of the advantages of the cohort survival method is that it produces projections for every grade. However, this requires migration assumptions for every grade. To illustrate possibilities, three migration assumptions were constructed. The desired outcome was a low end and a high end of recent experience. The average of the survival rates for the past two years (fall 2009 to fall 2011) is higher than the average of the survival rates for the past four years (fall 2007 to fall 2011). These averages are higher than the averages for the first five years (2002-2007). Therefore, the loss of students is less now than earlier in the decade. 16

At first glance, the rates look very similar. However, the higher rates do make a difference in the enrollment projections. Only two migration assumptions will be used in the enrollment projections. PROJECTED SURVIVAL RATES Grade Low (Past 4 Years) High (Past 2 Years) K to 1 0.982 1.004 1 to 2 1.029 1.023 2 to 3 0.989 1.000 3 to 4 1.017 1.037 4 to 5 0.984 1.005 5 to 6 1.023 1.025 6 to 7 1.054 1.054 7 to 8 0.986 0.961 8 to 9 1.067 1.067 9 to 10 1.001 1.009 10 to 11 0.946 0.939 11 to 12 0.986 1.013 Because net migration will be projected based on survival rates by grade, the percentage change will be the same each year while the actual number of students added or subtracted by grade may change from year to year. Projection Results The kindergarten and net migration assumptions are trend lines, which remove annual fluctuations. However, the future, like the past, will be characterized by annual fluctuation, sometimes large. Because there is no reasonable way to forecast when fluctuations around trend lines will occur, it is arbitrary to project them. Furthermore, long- term projections are designed to approximate a future point in time not to yield the best projection for each intervening year between the present and the projection end date. For this reason, long- term projections should not be used for annual budgeting purposes. The district should continue to use its version of the cohort survival methodology for annual enrollment projections. Four cohort projections are shown in the next table. In ten years, there is a 279 student difference between the lowest projection and the highest projection. The migration assumptions result in a difference of 213-216 students over ten years while the kindergarten assumptions result in only a 63-66 student difference. Therefore, in these projections, the migration assumptions make the biggest difference in projected enrollment. In all the projections, enrollment increases from 2011-12 throughout the projection period. However, growth is fastest in the first five projection years. This reflects the larger increase in K- 4 during the first five years. 17

The lowest projection is based on the low kindergarten and low migration assumptions. In this projection, enrollment increases by 467 students by 2021-22 or 10.3 percent. In five years, enrollment increases by 289 students or 6.4 percent. ENROLLMENT PROJECTIONS Year Low K Low Mig High K Low Mig Low K High Mig High K High Mig 2011-12 4,524 4,524 4,524 4,524 2012-13 4,608 4,615 4,634 4,641 2013-14 4,657 4,670 4,700 4,713 2014-15 4,738 4,757 4,801 4,821 2015-16 4,768 4,793 4,850 4,875 2016-17 4,813 4,844 4,914 4,946 2017-18 4,865 4,902 4,994 5,033 2018-19 4,874 4,918 5,032 5,077 2019-20 4,917 4,967 5,100 5,152 2020-21 4,952 5,008 5,152 5,211 2021-22 4,991 5,054 5,204 5,270 The highest projection, based on the high kindergarten and high migration assumptions, shows an enrollment increase of 746 students or 16.5 percent between 2011-12 and 2021-22. The intervening years show the same pattern as the lowest projection. In between the highest and lowest projections are two other projections. Because the migration assumptions have the biggest effect, the projections with the same migration assumptions are closest to each other. As a group, the four projections reflect a range of possibilities. Looking at the projections based on the elementary, middle school and high school grades is instructive. K- 4 enrollment increases more in the first five projection years than in the last five years. The first five years are based on actual births while the second five years are based on projected births. Middle school enrollment is significantly higher than today. In 2016-17, grades resulting from the kindergarten assumptions have not yet reached middle school so we see the effects of the current larger elementary grades as they age and the migration assumptions. ENROLLMENT PROJECTIONS K- 4 5-8 9-12 Total 2011-12 1,795 1,340 1,389 4,524 2016-17 Low K/Low Mig 1,825 1,515 1,473 4,813 High K/Low Mig 1,856 1,515 1,473 4,844 Low K/High Mig 1,869 1,588 1,457 4,914 High K/High Mig 1,901 1,588 1,457 4,946 2021-22 Low K/Low Mig 1,839 1,471 1,681 4,991 18

High K/Low Mig 1,869 1,496 1,689 5,054 Low K/High Mig 1,881 1,567 1,757 5,204 High K/High Mig 1,912 1,593 1,765 5,270 High school enrollment shows the largest increase over the ten year projection period. Today's larger elementary cohorts predict a larger high school enrollment by 2021. In 2021-22, the 2011-12 kindergarten class will be in Grade 10, which means that all the grades below Grade 10 are products of the projection assumptions. Detailed grade by year projections are at the end of this report. Housing Unit Method The housing unit method is another way of projecting population and school enrollment. While the number of dwelling units (housing units) is related to the number of school age children, dwelling units alone do not determine the number of school age children. The number of school age children per unit is a key component in the equation of projecting the school age population. The chief reason to use the housing unit method is to understand the effect of additional housing units. It could be said that housing stock is like DNA. It determines the size and characteristics of the resident school age population. Year built and value emerge as important housing characteristics. Year built reflects how families lived in a particular era and is a proxy for square feet and characteristics such as number of bedrooms, number of bathrooms and number of garage spaces. The presence of a master suite, walk- in closets, etc. can also be inferred from year built. Value implies some of these same characteristics plus lot size, location and interior amenities such as kitchen and bathroom appointments and finishes. The relationships between housing unit characteristics and student numbers and characteristics have been established by work in three states. Findings based on school districts in three states follow. The findings for West St. Paul are not like those of other districts. The most likely reason for these differences is that the student data represents only 63 percent of students living in the district. Dwelling unit type affects the school age child per unit yield. Single- family detached units have the highest school age child per unit yield. Single- family attached, such as townhouses, have significantly fewer children per unit than single- family detached units while apartment units have even fewer school age children per unit, although there are some local exceptions. For enrollment projection purposes, the change in single- family detached housing units is what affects the number of school age children in a district. Seventy- one (71) percent of West St. Paul Public School resident students come from the district's single- family detached units. WEST ST. PAUL PUBLIC SCHOOLS RESIDENT K- 12 BY HOUSING TYPE Resident Students Single- Family Single- Family 19

Attendance Total Detached Attached Multi- Family Area # % # % # % # % District wide 3,704 100% 2,619 71% 196 5% 889 24% Source: Dakota County Geographic Information System and Student Information System For the West St. Paul Public Schools, the student yield by unit type is low, which is not unusual for a district with older housing stock. However, the yield does not vary much by unit type. This, too, most likely results from the student data representing only 63 percent of district resident students. WEST ST. PAUL PUBLIC SCHOOLS HOUSING TYPE BY RESIDENT YIELD Units Resident K- 12 Resident K- 4 Resident 5-8 Resident 9-12 Unit Type # # Yield # Yield # Yield # Yield Single- Family 9,658 2,619 0.27 984 0.10 801 0.08 834 0.08 Detached Single- Family 1,113 196 0.18 91 0.08 56 0.05 49 0.04 Attached Multi- Family n.a. 889 n.a. 451 n.a. 271 n.a. 167 n.a. Total n.a. 3,704 n.a. 1,526 n.a. 1,128 n.a. 1,050 n.a. Source: Dakota County Geographic Information System and Student Information System Newer single- family detached units yield more students per unit than older single- family detached units. For West St. Paul, student yield is essentially the same regardless of unit age. Again, the students are only those attending the West St. Paul Public Schools. WEST ST. PAUL PUBLIC SCHOOLS SINGLE- FAMILY DETACHED RESIDENT STUDENT YIELD BY YEAR BUILT Resident K- 12 Resident K- 4 Era Built Units # Yield # Yield Pre 1950 2,051 591 0.29 241 0.12 1950-1959 2,315 644 0.28 260 0.11 1960-1979 2,626 709 0.27 263 0.10 1980-1999 2,357 603 0.26 192 0.08 Post 2000 309 72 0.23 28 0.09 Total 9,658 2,619 0.27 984 0.10 Source: Dakota County Geographic Information System and Student Information System As single- family detached units sell (turnover) the student yield increases, especially in the newer units. 20

Turnover (sale of existing units) does not bring more students to the West St. Paul Public Schools. WEST ST. PAUL PUBLIC SCHOOLS SINGLE- FAMILY DETACHED UNITS BY SALES STATUS (July 1, 2008 - June 30, 2011) Sales Status Units K- 12 Yield K- 4 Yield Not Sold 9,164 0.27 0.10 Sold 494 0.23 0.11 Total 9,658 0.27 0.10 The market value of single- family detached units affects the school age child per unit yield. Moderately priced to higher priced units yield more school age children than the lowest priced units. West St. Paul Public School students are slightly more likely to come from less expensive homes. (Nonpublic students may reside in more expensive units.) WEST ST. PAUL PUBLIC SCHOOLS SINGLE- FAMILY DETACHED RESIDENT STUDENT YIELD BY MARKET VALUE Estimated Single- Family Resident K- 12 Resident K- 4 Market Value* Units # Yield # Yield $0- $199,000 3,948 1,086 0.28 438 0.11 $200,000- $399,000 4,509 1,270 0.28 467 0.10 $400,000+ 1,201 263 0.22 79 0.07 Total 9,658 2,619 0.27 984 0.10 Different racial/ethnic groups and/or major language groups have different housing patterns by unit type. Among West St. Paul Public School students, the proportion of students living in single- family detached units (48 percent) is virtually the same as the proportion living in apartments (45 percent). WEST ST. PAUL PUBLIC SCHOOLS HOUSING TYPE BY RESIDENT MINORITY STUDENTS Resident Minority Students Single- Family Single- Family 21

Attendance Total Detached Attached Multi- Family Area # % # % # % # % District wide 1,489 100% 714 48% 107 7% 666 45% Source: Dakota County Geographic Information System and Student Information System As the population ages, more dwelling units are being built for mature adults (55+ years) and for seniors. These units will have zero school age children per unit. One- half of the district s single- family detached units contain at least one person age 55+. The district is "aging in place." WEST ST. PAUL SCHOOL DISTRICT SINGLE- FAMILY DETACHED HOMES WITH WEST ST. PAUL PUBLIC SCHOOL K- 12 STUDENTS AND REGISTERED VOTERS AGE 55+ With K- 12 West St. Paul Public School Students Percentage with K- 12 West St. Paul Public School Students With Registered Voter 55+ Percentage with Registered Voter 55+ Attendance Areas Single- Family Detached District wide 9,658 1,579 16% 4,875 50% Source: Dakota County Geographic Information System and Student Information System Versions of the Housing Unit Method The Housing Unit Method has two versions. One version is based on adding the projected number of dwelling units to the existing stock and then applying a child per dwelling unit estimate to the total dwelling unit count. The other version of the method or the housing starts method is based on estimating the school age children per new unit and adding these students to the student population from existing units. Both versions of the Housing Unit Method face some of the same challenges. Historically, the weakness of both versions was the difficulty in quantifying the effect of housing turnover and the demographic change that occurs when existing housing units are sold. With yield data from the Housing and Enrollment study, some of these problems are overcome. A unit s status, i.e., new; existing but recently sold; and existing not sold recently, becomes an important component in the housing unit method projections. The student yield for each status type is different and the student yield also differs by geographic area within the district. Even with these major improvements, the method does not yield projections by grade. Projections Dwelling Unit Growth As of March 2012, the West St. Paul School District is estimated to have approximately 17,000 dwelling units. Of these estimated units, 9,658 are single- family detached units, 1,113 are single- family 22

attached units and the remainder are multi- family units. Some of these units may be vacant, but for the purposes of this report, all dwelling units will be treated as occupied. Only 35 single- family detached units were built in the past three years. Based on projections from the cities in the district, only 60 twin homes are projected for the Mendota attendance area and even there are on hold. New development will have virtually no effect on enrollment in the West St. Paul Public Schools in the next five years. Projections based on the housing unit method are in Chapter 2, which focuses on resident K- 4 projections by attendance area. 23

CHAPTER 2 ENROLLMENT PROJECTIONS BY ELEMENTARY SCHOOLS AND ELEMENTRY ATTENDANCE AREAS Projecting K- 4 enrollment by school or attendance area is fraught with potential errors because the enrollment at any one school or in any one attendance area is small, which magnifies annual fluctuations. For this reason along with the short time that existing students are part of the K- 4 student body, projections will be made for five years rather than ten years. This chapter focuses on the five West St. Paul elementary schools and the district's five elementary geographic attendance areas. Past Trends The following table shows a five year history of Grade K- 4 enrollment. Since 2007-08, Grade K- 4 enrollment increased by 171 students or 10.5 percent. The largest enrollment increases occurred at Garlough Elementary School (99), Somerset Elementary School (93) and Moreland Elementary School (38). Enrollment declined at Mendota Elementary School (- 57), while enrollment at Pilot Knob declined but then came back to its 2007-08 level. ENROLLMENT GRADES K- 4 School 2007-08 2008-09 2009-10 2010-11 2011-12 Moreland 326 299 291 317 364 Mendota 363 356 332 317 306 Garlough 316 383 409 404 415 Somerset 318 327 376 376 411 Pilot Knob 301 275 290 291 299 Total 1,624 1,640 1,698 1,705 1,795 KINDERGARTEN School 2007-08 2008-09 2009-10 2010-11 2011-12 Moreland 60 48 47 75 88 Mendota 76 66 72 56 66 Garlough 76 106 94 79 80 Somerset 68 70 85 67 88 Pilot Knob 59 48 76 71 56 Total 339 338 374 348 378 24