
An official website of the United States government
CE Research ProductsThis page provides an introduction to special tabulation tables and other research products being developed by the Division of Consumer Expenditure Surveys. New research products will be added and updated periodically. Information on this page includes information on how to get the most from CE Tables, the state weight files, the state-level expenditure tables by income, the All Consumer Unit Prepublication Means, Variances, Percent Reporting Tables (All CU Prepublication MVP), new Census Division and Census Region Variance Tables, cross-tabulated tables with higher incomes, Generational tables that sort expenditures by generation/cohorts, and the 2014 Higher Income table which expanded the top income range to $200,000 and over. Note that the 2014 through mid-2016 Generational tables were research work, and had not been produced using BLS production methods and standards. Starting with 2016 data, the Generational tables are now included in the annual expenditure tables. How to get the most from CE tablesThis document provides information to help one delve deeper into the published as well as the very detailed non-published (prepublication) tables. This short article describes how one can use consumer unit characteristics to uncover useful underlying information about the data. State Weight FilesGuidance to the user: Care should be taken when analyzing Public-Use Microdata using the State Weights, as the small number of households for some expenditures can cause the mean dollar estimate to be imprecise. The more aggregated summary variables will produce more precise estimates. Additionally, it should be noted that these weights are only for their respective states and cannot be used to make inferences about any other geographic areas. Similarly, the national weight is unable to provide state level estimates. The provided data must be used in conjunction with the Public-Use Microdata to obtain state level estimates. The State Weights initiative by CE is an effort to produce research microdata products that can allow users to explore consumer expenditure data at the state level, a feature previously unavailable in the data. We intend to explore the viability of the CE sample to support weight creation for as many states as possible. The first available states are California, Florida, New Jersey, New York, and Texas. If others become available, they will be added to this page for download and use. The 2019 state weight files will be released in the spring of 2021. DocumentationCaliforniaYears available: 2016-2018 FloridaYears available: 2016-2018 New JerseyYears available: 2016-2018 New YorkYears available: 2017-2018 TexasYears available: 2017-2018 State-Level Expenditure Tables by IncomeSimilar to the regular CE Tables, the research tables below provide annual means by income quintiles for selected states using the state weight files and the public-use microdata. These tables cover two years of data to increase the reliability of the data. Please note that the tables do not include standard errors, and coefficients of variation as they are not appropriate for estimating their variances or standard errors. The variance estimation technique using the 44 replicate weights on the PUMD’s national-level FMLI and FMLD files is appropriate at the national, regional, and division levels, but it is not appropriate at the state level for various technical reasons. We are in the process of developing a statistically valid variance estimation technique for state-level expenditures. 2017-2018All CU Prepublication Means, Variances, and Percent reporting (MVP) TablesAll consumer units: Annual detailed expenditure means, standard errors, coefficients of variation, and weekly (D) or quarterly (I) percents reporting, Consumer Expenditure Survey, 2013-19Guidance to the user: Care should be taken when analyzing detailed expenditure subcategories in the prepublication tables, as the small number of households reporting an expenditure can cause the mean dollar estimate to be imprecise. Users should consider the coefficient of variation (CV). Expenditures with CVs of 25 percent and over are generally considered unreliable. For further information on standard error and variance in the CE, see Standard Errors in the 2016 Consumer Expenditure Survey. This table was developed to show the large amount of additional detailed expenditure data calculated by the CE. The table contains Means, Variances, and Percent reporting (MVP). The CE regular tables are an aggregate of the integrated data, and contain only higher level summary rows. This Prepublication table contains data pulled from three different tables. The three underlying tables summarized in this table can be obtained by contacting us. Available Prepublication tables in this disaggregated format are based on the same set of sixteen regular annual tables on our tables page. Displaying the disaggregated rows at the lowest level allows for the addition of either weekly or quarterly percent reporting statistics. Furthermore, one can calculate the weekly or quarterly means expenditure for only those who pay for particular items. The three types of tables are available upon request for the same demographics as our regular tables and contain the following information. The MVP tables displaying these data can be viewed in pdf or xlsx formats. Census Division and Census Region Variance TablesThe new tables below are for the current period (2017 - 2018) and include means, standard errors, and coefficient of variations. These tables cover two years to increase the reliability of the data. 2017-2018Archived Research ProductsCross-Tabulated Tables, 2015-2017The four 2-year research tables with rebalanced income categories and higher income ranges were produced on a research basis using 2015-2016 and 2016-2017 data. This work was done to increase the usefulness of CE data. Rebalancing was undertaken to consolidate and balance income ranges. As income rose when the economy recovered, the number of consumer units in the highest income range had become unbalanced and top heavy, while the lowest income ranges contained fewer and fewer CUs. For example, the current 2-year income for single men table has seven income ranges, but the $40,000 and over top income range has 36 percent of all CUs in it, and contains more CUs than the lowest three income ranges combined. Starting in 2018, these tables are now included in the cross-tabulated tables. 2016-20172015-2016Generational Tables, 2014 through mid-2016The midyear generational table for July 2015 through June 2016 with an All consumer units column and five additional columns using the birth year of the reference person grouped into generations has been added to this section. The source for the years used in the generational groupings comes from the following report. Starting with the 2016 annual data, this table is now included in the regular tables. The five generational columns are: The birth year of the reference person was approximated based on the collected age of the reference person variable in the CE data. Higher Income Table, 2014CE evaluated the income ranges on the regular income table, and found that over time the number of consumer units in the lowest income columns had decreased. This research table divided the existing $150,000 and over income column into two new columns: $150,000 to $199,999, and $200,000 and over. In 2015, these columns were integrated into the annual “Income before taxes” table.
Last Modified Date: September 9, 2020 |