-
Which method is used for the projections?
-
We use interregional cohort component projection methods. This is a standard cohort component projection method; except that it uses in-out county-to-county migration rates instead of county net-migration rates. The foundation of the projection method is described in Andrew M. Isserman, "The Right People, The Right Rates: Making Population Estimates and Forecasts with an Interregional Cohort Component Model," Journal of the American Planning Association, Vol.45, 1993, pp. 45-64.
-
What is the base year of your projections?
-
The base year of the projections is now 2000. The base numbers are counts from the 2000 Census.
-
What is the difference between short-term projections and long-term projections? Which projection is better?
-
Short-term projections use one unit of 5-year in-out migration data, which in this case is the 1995-2000 in-out migration. Long-term projections, on the other hand, use the average of 3 units of migration data: 1985-1990; 1990-1995 (estimates); and 1995-2000 (actual) in-out migration. Which one is more suitable depends on the assumption of in-out migration trends. In general, if a county has only mild fluctuations in migrations in the last two decades, then the average long-term migration will be the preferred measure of the future migrations in the county. However, if steady trends or stable rates of migration are observed during the last three or four migration periods, then the most current migrations (short term migrations) should predict future migrations better. West Virginia has both types of counties.
-
Why are births and deaths rates used in your projections?
-
We used 2000 birth rates and 1990 death rates. To improve the quality of projection and to minimize the noise in the data, we use 2000 birth rates. We will apply the 2000 death rates and when they become available from the Census Bureau.
-
Tell me more about migration rates.
-
The 95-00 migration data are the most recent migration data available (based on the 2000 U.S. Census). Around the year 2000 many counties in West Virginia started to experience approximately zero natural growth. This makes in-out migration numbers become very critical elements that determine population changes, such that small over/under estimation in the base year would magnify into large projection errors in the following years. Our method does estimate migration rates for the 90-95 and 95-00 periods, by combining projections based on 85-90 migration data with population counts for the year 2000. We only report the long-term projections because we are reluctant to use only the long out-of-date 85-90 rates.
-
When will you update your projections?
-
We expect the next update to occur when new death rate data are released. This is not expected to occur for several years.
-
Do you have projections by cities as well as by counties?
-
No, we don't. The available data do not support projections at the sub-county level.
