Student Enrollment Projections & Methodology
(Download Adobe Acrobat Reader to view or print PDF)
Broward County Public Schools (BCPS) primary projection tool is a geographically-based Cohort Survival model, which projects future students by grade. The Cohort Survival model is considered very reliable and is used by the Florida Department of Education in their student projections and the U.S. Census Bureau for their reports. The model uses an "aging" concept that moves a group, or cohort, of students into the future and increases or decreases their numbers according to past experience through history. BCPS enrollment projections are updated semiannually, once in October and once in April.
The Cohort Survival methodology relies on historical enrollment and birth data to capture the effects of in and out-migration, housing changes, and natural trends in population. In essence, the model derives a growth factor or ratio for student survival matriculation to the next grade based upon previous survival numbers to the same grade of students in each Traffic Analysis Zone (TAZ), the basic geographic area for the model. In most cases, TAZ areas represent neighborhoods. There are over 900 TAZ areas in Broward County. TAZ areas are further divided into smaller geographic areas to account for schools that matriculate to more than one school at each grade level, (e.g. an elementary school that feeds into two different middle schools). The combination of elementary, middle and high school attendance zones and TAZ areas create a unique identifiable area called a Study Area IDentification or SAID. SAIDs capture the grade cohorts more accurately by including feeder patterns. For example, if elementary school A matriculates to two different middle schools B and C and one high school D, there would be two different SAIDs for elementary school A-one SAID to represent matriculation from elementary A to middle school B to high school D and another SAID to represent matriculation from elementary A to middle school C to high school D.
Once the model has been run for the small geographic units or SAIDs, the projections are then summarized by TAZ. In some instances, individual TAZ areas are corrected to reflect changes in growth which are not picked up in the projection model's histories. A few examples where corrections are required include areas where:
- new construction is anticipated to exceed the pace of historical construction for an area,
- an area is reaching build-out and all new construction will cease or slow down,
- an unprecedented slow-down in the economic market, or
- a boundary change has artificially increased/decreased the area.
The historical number of births is a good indicator of future kindergarten class size. Birth data is acquired from the Florida Department of Health Vital Records by U. S. Census tract. Several steps are taken to interpolate future kindergarten enrollment based on births, as not all children born will enter kindergarten. To project kindergarten enrollment, births by census tract have to be estimated for a five year period i.e., births from 2005 will potentially enter kindergarten in 2010-11. Data is then increased or decreased based on past kindergarten populations by census tract. Once the number of births is adjusted, the percentage of students that are in each census tract is broken down to the SAID level. Since the census tract may intersect more than one SAID, a unique identifier is created between the census tracts and SAIDs. The percentage of actual attending kindergarten students for the past two years is calculated for each unique SAID/census tract. This percentage is used to extrapolate the number of kindergarten from the total number of kindergarten aged students within a given unique SAID/census tract. The SAIDs are then summarized to obtain the estimated number of kindergarten students by SAID for five years.
Residential Development Data
Each year Broward County municipal planning staff provides current and forecasted certificates of occupancy to assist county and BCPS demographic staff in estimating population changes. Residential growth is also shared and monitored through the Facility Management, Planning, and Site Acquisition Department. BCPS requests city and county planning staff to estimate future certificates of occupancy over the next five years. As shown below in the sample data in Figure 1, city staff provides the location, type of residential housing, and the timing of the projected certificates of occupancy.
Figure 1: Sample City Data
Other information is analyzed to determine if the Cohort Survival rates may need to be adjusted to align with a shorter or longer historical time horizon. These data may include:
- Existing home sales (source: Florida Association of Realtors)
- Population Projections (source: U.S. Census, Broward County, Bureau of Economic and Business Research, and Florida Department of Education)
Attrition Rate of Attending Students
BCPS includes four years of attending enrollment to calculate the rate of attrition or rate of students matriculating to the next level within their SAID by grade. Attending enrollment is the total number of students within the attendance zone that are attending their geographically assigned school. Determining the attrition rate by SAID, keeps the feeder patterns intact as the grades matriculate to each specific school.
SAID 2nd grade attrition rate 2007-08 =
(# of 2007-08 attending 2nd graders by SAID) / (# of 2006-07 attending 1st graders by SAID)
SAID 2nd grade attrition rate 2008-09 =
(# of 2008-09 attending 2nd graders by SAID) / (# of 2007-08 attending 1st graders by SAID)
SAID 2nd grade attrition rate 2009-10 =
(# of 2009-10 attending 2nd graders by SAID) / (# of 2008-09 attending 1st graders by SAID)
Once the attrition rate is calculated for each grade, grades one through twelve, over the past three years, it is then averaged and used as a factor to obtain next year's projections for that grade. For example:
projected 2010-11 2nd graders by SAID =
(average SAID 2nd grade attrition rate from 2007-2010) * ( # of 2009-10 attending 2nd graders by SAID)
To calculate subsequent years of projections by grade, the model uses the projected rate of attrition based on the projected enrollment of the previous year to calculate the next projections year. For example:
projected 2011-12 2nd graders by SAID =
(average SAID 2nd grade projected attrition rate from 2008-2011) * (# of projected 2010-11 attending 2nd graders by SAID)
Projections by SAID for each grade are then reviewed school-by-school. Attrition rates can cause projections to be exceedingly high or low in which case they will have to be adjusted so as not to cause an exponential effect in outer projection years. The following are possible corrections to rates:
Out-of-Boundary Students (OOB)
Out-of-boundary (OOB) students are students attending a school from outside their attendance area (i.e. approved reassignments).
BCPS assumes that OOB students at each grade level at each school will be the same as the existing year and will have a survival rate of 100% as they matriculate through the grade levels. For example, Middle School A currently has the following OOB students: 35-6th grade, 38-7th grade, and 42-8th grade. For all projected years, Middle School A will have 35-6th grade, 38-7th grade, and 42-8th grade OOB students.
However, adjustments can be made to OOB students if enrollments naturally decline based on the calculated cohort survival rate yet economic or other conditions may suggest enrollment should increase or if schools are eligible to receive assignment transfers. Since assignment data is determined after the release of the projections and is subject to change, the OOB students typically remain constant in the model based on the current year's data.
The school-by-school Cohort Survival model projections, by grade, are compared and tested for reasonableness with other models such as the Florida Department of Education (FDOE) projections and the Broward County Planning and Redevelopment Division school-aged population projections. Accordingly, adjustments may be made to the Cohort Survival model based on the following factors:
- changes in the rate or type of new housing development within Broward county
- changes in economic conditions (e.g. the creation of jobs usually means families are moving in whereas a recession usually means families are moving out)
- natural phenomena (e.g. hurricanes)
There are also decisions made within BCPS, which may have a dramatic effect upon projections. These include:
- future placement of English Language Learners (ELL) clusters
- future placement of Exceptional Student Education (ESE) clusters
- opening and closing of magnet programs (first year projections are difficult because of the lack of a "track record")
- Adequate Yearly Progress (AYP) choice reassignments
- other approved reassignments
- opening and closing of charter and private schools throughout the year