Orange County NC Website
Browse       Search
DRAFT :. APPENDIX <br />2. US Census Bureau <br />NC State Certified July 1, 2005 County Population Estimates <br />NC Office of Budget, Planning and Management, Demographic Section <br />Methodology - -- Cogpty Totals <br />The county estimates are averages of two sets of estimates, a set of modified Census Bureau <br />estimates, and a set of Ratio /Correlation estimates produced by this office. <br />Modified Census Bureau Estimates <br />In March of 2006, the Population Division of the United States Bureau of Census released <br />their 2005 state and county population estimates for North Carolina. They used an administrative <br />records technique similar to the 1990's technique of the same name. In making their 2005 <br />estimates, the Census Bureau assumed that the July 2005 institutional population of each county <br />would be the same as that of July 2004. By April of 2006, I was able to obtain the July 2005 data <br />for all of our military bases and for the institutions possessing the majority of the other <br />institutional population in North Carolina. As a result, I adjusted the Census Bureau's 2005 <br />county and state estimates for the differences between the July 2004 and July 2005 populations <br />for these institutions. A second approximation used by the Census Bureau was to assume that <br />July 1, 2000 institutional populations were identical to the corresponding April 1, 2000 values. I <br />made further corrections where this assumption was false. <br />Ratio /Correlation Estimates <br />Two basic procedures were used to build these estimates. First, I used changes in the numbers of <br />MEDICARE enrollees to estimate the population 65 years of age and older for each county. <br />MEDICARE enrollee counts from July 1, 1999 and July 1, 2000 were used to estimate the value <br />for April 1, 2000. The change in the number of MEDICARE enrollees from July 1, 2003 to July <br />2004 was assumed to equal the change in the number of MEDICARE enrollees from July 1, <br />2004 to July 2005. Then, the 5.25 year change in the number of MEDICARE enrollees from <br />April 1, 2000 to July 1, 2005 was multiplied by the ratio of the April 1, 2000 population count <br />for persons 65 years of age and older to the estimated number of MEDICARE enrollee counts for <br />April 1, 2000. The resulting value was assumed to approximate the change in the numbers of <br />persons aged 65 and older between April 1, 2000 and July 1, 2005. This change was added to the <br />April 1, 2000 population count for persons 65 years of age and older to create the final <br />population estimate for this age group. <br />Second, I used a standard ratio /correlation method to estimate the population aged 0 -64. The data <br />series used were automobile and truck registrations (Xl), school enrollment in grades 1 through <br />8 (X2), and a three -year sum of births (X3). For 2005, the three -year sum of births was the sum <br />of final calendar year values for 2003, 2004, and 2005; for 2000, the sum of final calendar year <br />values for 1998, 1999, .and 2000. The prediction equation for each North Carolina county is <br />given by <br />y =. 0.00061057 + 0.51853 * X1 + 0.27341 * X2 + 0.20135 * X3, <br />where y represents the estimated ratio of percentage shares of nongroup quarters population and <br />each of the series indicators (Xl, X2, and X3) represents the ratio of percentage shares of the <br />2 <br />8/16/20( <br />