PopStats Data Resource Manual. Demographic and Socio-Economic Data

Transcription

1 PopStats Data Resource Manual Demographic and Socio-Economic Data

2 This page is intentionally left blank.

3 Table of Contents 1 Introduction and How to Use this Guide 2 Using PopStats 3 Data Sources and Methodology 8 Unprecedented Accuracy 12 STI BlockPoint

4 This page is intentionally left blank.

5 Introduction and How to Use this Guide Welcome This Resource Manual provides all the information you ll need to get started and to use PopStats demographic and socio-economic data. PopStats is the demographic and socio-economic data produced by Synergos Technologies, Inc. (STI). STI is one of the nation s leading data providers. STI data achieves unprecedented levels of accuracy. Today, market research demands precision. It requires keen market knowledge to make the most profitable location-focused decisions, such as where to open new stores, when to close existing stores, and how to identify new business opportunities. Every research variable needs to be dependable especially the current year and projected year population estimates. Since its launch in 2001, a growing number of progressive companies have come to depend on the accurate, current, and robust population estimates presented in PopStats. They use PopStats to capture real population growth as it occurs, estimate population counts accurately at small geographic levels, gain in-depth knowledge about specific populations, access data for historic growth trends and future forecasts, and much more. PopStats data delivers the leading annual updated population estimates. Populations across the U.S. are constantly in flux. PopStats data captures these changes, whether up or down, as they occur, in small geographic areas. Updated estimates include current year estimates and five year forecast. PopStats data consistently delivers the highest level of accuracy. PopStats data uses a combination of unique source data from the U.S. Postal Service ZIP + 4 records and the U.S. Census, its own proprietary bottom-up methodology, and field tests to deliver the most dependable population estimates available. The ZIP + 4 source data adds a householdlevel quality to the estimates. PopStats data gives companies an expanding range of consumer data. PopStats data has grown from 21 variables in 2001 to over 1,200 today. All new variables are based on STI s dedication to innovation and to fulfilling its clients specific data requests. 1

6 Using PopStats Why Use PopStats PopStats data delivers a range of benefits to organizations across industries: retail, food service, real estate, communications, healthcare, non-profits, and economic development and other government agencies, and others. PopStats data informs market decisions, supports smarter market investments, and offers users a competitive advantage. Features Gain greater research confidence Accurate Track current population changes Pinpoint high-growth markets Robust data Powerful Block-level calculations Household-level Distinctive Open book Benefits Elevate your research to the highest level of insight with accurate, robust, timely, and fieldtested population estimates. Field-tested to ensure the highest level of accuracy. See population changes as they occur, including trade area upswings and downturns, new growth, seasonal growth, and other vital changes. Find new areas of growth before your competitors and be the first to secure customer loyalty. More information about your markets. Access hundreds of data variables to conduct a wide range of research including mortgage-risk, seasonal, transient, home values, and much more. Captures market upswings and downturns with equal precision. Behind the scenes data is calculated using censusblock level data to assure that all geographies have the most accurate possible information. Analysis conducted at the ZIP+four level, which effectively represents the household level. Incorporates unique sources of data, such as transient population counts, mortgage risk, seasonal, and unemployment. STI shares fundamentals of its proprietary product methodology. 2 PopStats Data Resource Manual

7 Data Sources and Methodology Data Sources for PopStats PopStats uses the following data sources: United States Census Bureau United States Postal Service (USPS) United States Department of Defense (DMDC) National Center for Education Statistics (NCES) National Center for Health Statistics (NCHS) Federal Financial Institutions Examination Council (FFIEC) Internal Revenue Service (IRS) Bureau of Economic Analysis (BEA) Bureau of Labor Statistics (BLS) Office of Federal Housing Enterprise Oversight (OFHEO) Department of Defense (DOD) PopStats Data s Innovation Overcomes Traditional Population Data Challenges PopStats data s unique approach to population estimates includes four primary innovations over traditional population data methodologies. ZIP + 4 and 2000 Census vs. Census-Only Source Data STI was the first data provider to realize the value of ZIP + 4 postal data and to envision a way to leverage this household-level source data. PopStats data factors in ZIP+4 calculation methods to data provided at higher level geographies such as census tracts. The ZIP + 4 level leads to more accurate population estimates for five reasons: 1. It is extremely detailed 2. It contains over 28 million records 3. It includes all major population centers 4. It can be manipulated statistically 5. It is easily consolidated into any geography 3

8 Bottom-Up vs. Top-Down Methodology The decennial U.S. Census is based upon a traditional top-down construction, which takes macro-level data and extrapolates it down to a micro-level: from U.S., to state, to county, to tract, to block group. The Census Bureau s nationalto-local direction was copied by demographers to generate population estimates at the block-group level. However, there are significant problems with this direction. Foremost, macro-level data is unsuitable for use at a microlevel, like block groups, which are greatly influenced by singular local events, such as a new apartment complex or a building demolition. To compensate, demographers developed population-spreading techniques that broad-stroke areas of growth and decline at the sub-county level. This is an improvement, but still retains limitations. Specifically, it can mask block-group level growth or decline. PopStats data delivers a more accurate population count on a micro level by starting at the ZIP + 4 level then moving from the bottom-up to the block group, tract, county, ultimately state levels. Quarterly Updates Leads to More Accurate Data Traditional population data is updated only once every 12 months (typically in May or June). As a result, the data is chronically out-of-date and adjustments to account for seasonal change are lagging. This puts researchers at a serious disadvantage. STI created the industry s first population data to be updated on a quarterly basis - every January, April, July, and October. While these quarterly updates are done behind the scenes, when applied annually, the PopStats data is better adjusted for national and local trends, resulting in very accurate updates. Expanding vs. Static Variables Unlike many population data providers, STI continually expands into new data territory. PopStats data launched in 2001 with 21 variables and now has over one-thousand variables. Recent variables include mortgage risk, home values, employment, five- and ten-year forecasts, and much more. New variables are based on a combination of STI innovations and client data requests. For example, a leading grocery store chain requested seasonal data, a QSR (quick serve restaurant) client requested transient (i.e., hotel, motel, and RV park) population counts, and a drug store chain requested Puerto Rico population counts. Every new data variable is available to all PopStats clients. 4 PopStats Data Resource Manual

9 Estimating Methodology The overall PopStats model is a collection of sub-models that calculate the population estimates. The methodology consists of the following three steps. STEP 1: Estimate Households STI s research has shown that a unique and quantifiable relationship exists between USPS (United States Postal Service) data and U.S. Census Bureau household counts. Due to this relationship, STI can model population shifts quickly and accurately using a proprietary technique leveraging the correlation between the two. The process is initiated by base-lining the ZIP + 4 data and its associated. Then, as new ZIP + 4 data is made available (monthly), STI models and derives a growth factor for every ZIP + 4 in the country. This application occurs via a proprietary model including other pertinent factors to generate a current estimate. To limit bias in the data due to extraneous figures, such as errors in the raw data, PopStats methodology includes automated processes for overcoming anomalies including ZIP + 4 inaccuracies, data smoothing issues, conversions (lofts), and overrides. STEP 2: Estimate Household Populations A variety of U.S. Census Bureau and private studies have shown that the relationship of persons-to-households remains relatively stable over time. STI takes the Census 2010 persons-per-household-per-block group figures, and adjusts the ratio to reflect any changes in the county estimated personsper-household generated by the U.S. Census Bureau. These new figures are then applied to the estimated households to derive an estimated household population. STEP 3: Apply Controls To further ensure accuracy and limit estimate bias, STI uses a series of checksand-balances to validate the results. One of these steps is to compare its estimates to the U.S. Census Bureau s annual population estimates released every spring. If any major discrepancies occur between the two numbers, the model applies a set of heuristics to determine the most probable population figure. STI also consults with multiple state and federal agencies whose data is independently gathered and calculated. In addition, selected cities throughout the U.S. are field-surveyed to further validate our model s results. 5

10 Methodology for Key PopStats Data Variable Break Outs Once the base population has been estimated, the PopStats model breaks out several demographic estimates, such as age and sex, race and ethnicity, group quarters, incomes, and housing values. Many more data variables are available in the ever-expanding PopStats data product. Age and Gender Age and gender are determined through a traditional cohort survival analysis. This sub-model to the main model looks at each age distribution within a race category and applies the appropriate birth and survival rates as determined by the NCHS (National Center for Health Statistics). These results are then balanced back to the base population using an iterative approach. In addition, information from the NCES (National Center for Education Statistics) is applied to validate the age distribution of school-age children. U.S. Census estimates are used to validate all other age ranges. Race (Ethnicity) Race is calculated using a ratio analysis of April 2010 observed and annual U.S. Census estimates. In areas of high growth we use race information gathered by the FFIEC (Federal Financial Institutions Examination Council) which collects information from financial institutions concerning loans and race issues. It is a reasonable source for understanding race percentages in high-growth areas. As a final check for race, the STI model also consults with NCES race data for elementary school children and checks NCES data against PopStats figures. Income Estimates Income estimates are based on a two-step process. First, household incomes at the county level are estimated using a blend of information from the IRS s Survey of Income, U.S. Census Bureau s ACS dataset (American Community Survey) income estimates, and personal income estimates from the BEA (Bureau of Economic Analysis). Once the county estimate is derived, STI estimates the block group level. This is done in two parts. First, STI separates existing households from new-growth households, because research has found that in high-growth areas existing households are not a good indicator for determining the income of the new households entering the area. Therefore, STI uses a typical income growth approach that resembles the growth of county income. Then STI adds a separate income growth for new households modeled on the FFIEC s mortgage data transactions. 6 PopStats Data Resource Manual

11 Group Quarters Group quarters are a collection of unrelated people where no one individual can claim head of household, such as college students and military personnel. Generally speaking, group quarters data can be divided into three categories: colleges, military bases, and institutions (i.e., state homes, hospitals, and prisons). STI estimates each category individually, and then combines them for a total estimate. College student dormitory information is derived from the NCES annual college survey. Military group quarters are determined based on a direct data feed received from the DOD (Department of Defense) Manpower Data Center. Institutionalized persons are estimated using historical trends from the U.S. Census. Housing Values Housing values are determined in a fashion similar to income estimates. Housing and associated values that existed as of April 2010 are updated using data from the OFHEO (Office of Federal Housing Enterprise Oversight). The STI model performs a detailed analysis of same-home selling prices that occur over time. STI uses the resulting growth factors and applies them to existing April 2010 owner-occupied homes. New home values (homes built after April 2010) are determined by ratio analysis of the FFIEC s mortgage values and actual selling prices. 7

12 Unprecedented Accuracy Background Understanding the accuracy of an estimate is fundamental to the use of that estimate. After all, a lack of faith in that estimate makes its use moot. The best way to prove the accuracy of an estimate is, of course, to compare it to a known quantity. In our case that known quantity is the United States Census. With the April 2010 Census data, STI was able to provide irrefutable evidence that PopStats is the leading demographic product on the market. STI s analysis was performed for population at five levels of geography: national, state, county, census tract, and block group, using Census 2000 boundaries as the boundary definition standard. For national through county levels, virtually no data conversion was necessary to convert Census 2010 boundaries to Census 2000 boundaries (note: four 2010 counties in Alaska had to be consolidated back into their original 2000 boundaries). However, extensive data conversion did take place for both census tract and block group boundary data. The methodology to derive an accuracy measure for populations was performed in two steps. First, STI converted the Census 2010 tract and block group boundaries into Census 2000 format. Then, STI performed the variance calculation known as Mean Absolute Percentage Error, or MAPE. STI took the data for the analysis from commercially release CD s. STI pulled the April 2010 population estimate from the CD released in July 2010; and it took the Census 2010 data converted from 2010 to 2000 boundaries from the CD released in July Data Conversion The Census Bureau provides relationship files so that a data user can translate data from 2000 to 2010 boundaries, and vice versa. Using these relationship files, STI converted the Census 2010 data at the block level to Census 2000 blocks. Once the conversion was completed, the block data was aggregated to the block group level, the standard level of geography at which most data comparisons is performed. When this conversion is performed, a certain amount of rounding error is introduced. In this case, the original total U.S. population changed from 308,745,538 to 308,745,517, a difference of only 21 people. The converted data was only used at the tract and block group levels; STI was able to use the actual Census numbers for national through county levels. 8 PopStats Data Resource Manual

13 MAPE The Mean Absolute Percentage Error, or MAPE is the standard that most data vendors use to measure the accuracy of their data. The calculation is straightforward: subtract the estimate from the actual, take the absolute value of the result, and divide it by the actual value. This yields the Absolute Percentage Error (APE). Perform this calculation for every entity in a geographic layer, sum those results and then divide by the number of entities in that geographic level. This final value is the MAPE. A newer method to judge accuracy was recently introduced and is referred to as MAPE-R or Mean Absolute Percentage Error - Rescaled. This method compensates for extreme outliers. This method accounts for the notion that if the method being judged gets 99.9% of the values accurately, is it fair to allow one extreme outlier to cloud the results? STI decided not to use MAPE-R for two reasons. First, the industry as a whole has traditionally used MAPE as the standard to judge all estimates, changing those rules now does not allow for comparisons of results to earlier years. Secondly, STI strongly believes that all entities in a geography should have a equal weight in the result, discounting extreme outliers does the researcher little good if his business happens to be located in those outlying areas, the most notable of which are high growth areas. These high growth areas are the ones most likely to be inaccurately estimated by an estimating methodology. If a researcher wants to compare a MAPE with a MAPE-R value, they must take into account that the MAPE-R method has potentially artificially lowered the methodology s true error rate, and therefore the difference between the two measures may actually be greater. STI Results STI results are extraordinary. At every level of geography, the PopStats methodology demonstrate an unparalleled level of accuracy. National The first level is national. If the method fails to achieve a high degree of accuracy here, then by definition all of the lower geographic levels cannot be more accurate then the level of above it. STI results show: Geography Census 2010 PopStats MAPE National 308,745, ,748, % The PopStats estimate differed from the actual national census by only 3,422 people. 9

14 State through Block Group The extraordinary results at the national level followed through at every lower level of geography: Geography MAPE State 1.01 County 3.40 Tract Block Group For example, on average, the PopStats estimate varied at the state level by approximately 1.01%. In fact, a detail analysis showed that 33 states varied by less than 1%. Texas, one of the fastest growing states in the country, had differed by only 0.034%. Finally, over 1,750 counties, representing over 215 million people, differed by no more than 2.5% to the actual census. Vendor Comparisons Recently, other data vendors have released their results as well. STI chose to compare its data results to Esri, which published its own results on May 21, 2012 via a press release and supporting white paper entitled Vendor Accuracy Study Estimates versus Census Esri hired one of its partners, Cropper GIS, to analyze Esri s own results compared to four other unnamed vendors. As mentioned in the press release, four demographers were assembled (three of whom are part of the management team at Cropper GIS) and were given data obtained by Esri. It is important to note that PopStats was NOT included in Esri s analysis. Since the other vendors are not known, the following table only shows STI s results compared to those reported by ESRI in the above cited white paper. According to Esri s press release, all of the other vendors included in the study scored more poorly than ESRI on an overall basis. The following table displays a comparison of PopStats estimates using MAPE with Esri estimates using MAPE-R. As displayed in the table below, PopStats has significantly greater accuracy than Esri s estimate across all geographic levels. Geography PopState with MAPE Esri with MAPE-R State County Tract Block Group Total PopStats Data Resource Manual

15 Esri also performed an analysis at the household level. The following table compares those results 1. As displayed in the table below, PopStats has significantly greater accuracy than Esri s estimates across all geographic levels. Geography PopState with MAPE Esri with MAPE-R State County Tract Block Group Total Note: Households were ignored in the initial analysis because Census 2010 Household data was not available to the STI user community in the CD s mentioned earlier. Census 2010 Household data was not released by the Census to the public until September 2011 and was incorporated in the October 2011 release of PopStats, in the 2010 boundary format. PopStats More Accurate than Esri As displayed in the results shown above, PopStats, in overall accuracy proved to be 3.77 times more accurate than Esri. Esri s overall MAPE-R aggregated score is (the sum of and from the two tables above). PopStats overall MAPE aggregated score is 65.55, significantly lower than Esri s. 11

16 STI BlockPoint BlockPoint Block groups have been the de facto geographic measurement instrument in trade area analysis for decades. But extracting the most current population estimates within subsections of the nation s 209,000 block groups has been difficult, if not impossible. While most population estimating products simply spread population growth across block groups based on historical averages, an innovative STI product called BlockPoint brings a finer level of granularity to the process, resulting in significantly higher accuracy. There are over eight million census blocks nationwide. BlockPoint leverages the power of these blocks to give end-users population estimates that represent true population changes within all block groups. BlockPoint is built into the data methodology, so it s not something the enduser will actually see or access. Estimating Populations with Finer Granularity BlockPoint gets its precision from the same calculation methodology described above: combining Census data with residential ZIP+4 postal codes. However, BlockPoint takes the population calculation one step further. It allocates population growth to subsections within block groups based on actual changes. For example, if 1,000 people have moved into a block group that has ten block points, but 70 percent of the new residents have moved into only the southwest quadrant of the block group, BlockPoint can identify those specific growth locales and calculate both population estimates and growth percentages for them. Ideal Solution for an Aging Census BlockPoint overcomes one of the leading trade area research problems of our times: the aging population. One of the downsides to this decennial Census is that population growth has changed significantly over the last 15 years. Today many previously low-density block groups are among the country s fastest growing areas. Conversely, many areas that had large populations have lost residents. But even more relevant is that growth and decline has rarely occurred uniformly across block groups. BlockPoint overcomes the disparity with its unique methodology. As a result, businesses gain another valuable tool at their desktops to more accurately assess population changes in new and existing markets. 12 PopStats Data Resource Manual

17 This page is intentionally left blank. 13

18 CIVICTechnologies provides a full suite of web services, customized applications and consulting services to meet the complete market research and analytical needs of libraries and businesses from small and medium size enterprises to Fortune 500 companies. CIVICTechnologies enables its customers to visualize data in new and innovative ways in order to deliver more effective products and services, make better informed decisions, build consensus faster, allocate resources more wisely, and implement organizationwide solutions more effectively. Contact CIVICTechnologies Toll Free: (888) civictechnologies.com support@civictechnologies.com 2015 Futterman and Associates, Inc. prm PopStats Data Resource Manual