A Forecasting Solution for Enrollment Planning

In an environment of rapidly changing demographics, the ability to predict student populations is crucial for effective planning. This is especially true where growing enrollment has forced school districts to make significant adjustments with portable structures and overcrowded classrooms. FORECAST/DSTM brings the future into focus, allowing you to predict student enrollment for up to twenty years into the future, and confirm the accuracy or uncertainty of the predictions.With this powerful tool you can simulate multiple scenarios, and see how your decisions accommodate growth or decline.

A probabilistic model

Typically, single-number projection techniques do not account for uncertainty. With FORECAST/DS, enrollment projections are presented in terms of a probability distribution. The planner can take into account the possibilities of events such as drastic shifts in birth rates, catastrophes or the relocation of a large industrial plant, by estimating the probabilities of these events. The forecaster must give three estimates for each variable; a “high,” a “low” and a “most likely” estimate, with the difference between the “high” and “low” estimate representing a 90 percent confidence interval. A Monte Carlo procedure combines these various estimates into probability distributions of enrollment predictions.

Decision support system

FORECAST/DS is more than an enrollment forecasting system. It is a Decision Support System. After the enrollment is forecasted, decisions must be made as to how the growth or decline will be managed. Alternatives that the planner can choose from are:
Open a new school
Build a new wing or classroom
Change attendance boundaries
Increase class size
Add portable buildings
Change to double session
Change grade level configuration
Change to year-round school
Exclude inter-district transfers
Send back intra-district transfers
FORECAST/DS links directly to ONPASS® for simulation of opening a new school, changing the capacity of one or more schools, changing attendance boundaries or changing grade level configuration. Student enrollment forecast can be used to run ONPASS planning simulations for up to 20 years in the future.

A multivariate system

“This technology has brought consistency and professionalism to our school district while providing immediate results to a vital process that used to be time consuming and many times inaccurate.”

Cindy Hepting
Program Coordinator
West Linn-Wilsonville School District

The cohort method of enrollment prediction was modified to accommodate probability input and output. Five years of historical student data are used to calculate a cohort probability distribution for each school and grade level. In addition, new housing data—by planning area, year and dwelling type—combined with student generation factors provide forecasted student counts from new construction. New housing tracts can be followed over time to validate student generation factors. Migration, retentions, skips, transfers, births and dropouts are also part of the forecasting logic.

New housing construction

One of the most important factors in forecasting school enrollment is new students resulting from new housing construction. This is a relatively easy statistic to track, record and use with the FORECAST/DS software. FORECAST/DS provides data input screens to enter subdivision new housing data, generation factors by grade and housing types. New housing counts—by planning area, year and dwelling type—combined with student generation factors, provide forecasted students from new construction in the planning area where the construction occurs. New housing tracts can be followed over time to validate student generation values.

Data check reports

By using the MasterPlan historical enrollment data the user can go back in time and forecast a year where the enrollment counts are already known. One of the Data Check Reports shows a comparison of the actual and forecasted enrollment. Adjustments can then be made to calibrate the forecasting parameters. Over thirty reports are available for reporting input data and forecasting results.

Integrated systems

FORECAST/DS is one of the MasterPlan software applications from Planware Systems, LCC. It is part of the ONPASS program for boundary planning and student demographics. FORECAST/DS expands the utility of the ONPASS modeling system. ONPASS allows you to vary parameters of space and population. By adding another dimension to the ONPASS model, the effectiveness of ONPASS is dramatically increased. This dimension is one of time. By modeling future populations, planners can study where new growth or migration may have large impacts on school populations, facilities and programs. After working with FORECAST/DS, the projected student data can be immediately accessed in ONPASS from a pull-down menu showing the forecasted years.

Build-out studies

The ONPASS Land Use layer and County Assessor parcel records are important features in the forecasting of future enrollment. The import of County Assessor parcel records allows ONPASS to calculate the number of dwellings by dwelling type in each residential planning area. Planning areas are identified by land use categories and displayed in a Land Use layer. Student generation values by dwelling type are calculated by dividing the number of students by the number of dwellings in each planning area. Future dwelling growth is calculated in undeveloped planning areas by year as a density projection by dwelling type. The number of dwellings and their student generation values are read by FORECAST/DS to project future student enrollment.

Navigation, status and help

FORECAST/DS is a robust application that is very easy to use. Where to go and what to do is outlined in the navigation bar, procedures are run by “wizards” and reported in the status screen. A help button, describing the activity, is included with each navigation step. As steps are completed, the date is entered under the status section.