Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Consumer Expenditure Surveys
PRINT:Print
CE CES Program Links

Using the State Weights on the Public-Use Microdata

Jimmy Choi and Taylor J. Wilson

Division of Consumer Expenditure Surveys

The variables included in the state-level weight datasets are: NEWID, STATE, QINTRVYR, QINTRVMO, STRTYEAR, STRTMNTH, PSU, and STATEWGT. To use the state-level weights with the public-use microdata (PUMD), merge the state-level weight dataset and the PUMD file of interest by NEWID and then proceed as usual in investigating weighted sample statistics.

Although the state weights are designed to generate weighted sample statistics at the state-level, they cannot be used to generate weighted sample statistics at any other geographic level. It is not valid to use national weights at the state level since their weights are calibrated to population totals at either the Census Region (prior to 2017) or the Census Division (2017 and later) level, and likewise it is not valid to use state weights at the national level either.

Also, although the new weights are appropriate for estimating weighted sample statistics for the given state, 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, but it is not ready yet.

Table I. Variable Definitions
VariableDescriptionFormat

NEWID

Consumer unit (CU) identification number. Digits 1-7 (CU sequence number, 1 through 9999999) uniquely identify the CU. Digit 8 is the interview number, 1 through 4. It is possible for a CU to skip an interview. For example, a CU could have a 1st, 2nd, and 4th interview but no 3rd interview. Values of NEWID contain a leading zero. Therefore it will appear that the NEWIDs are 7 numbers long, when they are in fact 8 numbers. BLS derived.CHAR(8)

STATE

State identifier (see Section IV.A. and Section X.D. in the PUMD documentation for important information regarding suppression and topcoding of the STATE variable): 01 Alabama, 02 Alaska, 04 Arizona, 05 Arkansas, 06 California, 08 Colorado, 09 Connecticut, 10 Delaware, 11 District of Columbia, 12 Florida, 13 Georgia, 15 Hawaii, 16 Idaho, 17 Illinois, 18 Indiana, 19 Iowa, 20 Kansas, 21 Kentucky, 22 Louisiana, 24 Maryland, 25 Massachusetts, 26 Michigan, 27 Minnesota, 28 Mississippi, 29 Missouri, 30 Montana, 31 Nebraska, 32 Nevada, 33 New Hampshire, 34 New Jersey, 35 New Mexico, 36 New York, 37 North Carolina, 38 North Dakota, 39 Ohio, 40 Oklahoma, 41 Oregon, 42 Pennsylvania, 45 South Carolina, 47 Tennessee, 48 Texas, 49 Utah, 50 Vermont, 51 Virginia, 53 Washington, 54 West Virginia, 55 WisconsinCHAR(2)

QINTRVYR

Interview yearCHAR(4)

QINTRVMO

Interview monthCHAR(2)

STRTYEAR

Diary start date – yearCHAR(4)

STRTMNTH

Diary start date – monthCHAR(2)

PSU

S11A Boston-Cambridge-Newton, MA-NH; S12A New York-Newark-Jersey City, NY-NJ-PA; S12B Philadelphia-Camden-Wilmington, PA-NJ-DE-MD; S23A Chicago-Naperville-Elgin, IL-IN-WI; S23B Detroit-Warren-Dearborn, MI; S24A Minneapolis-St. Paul-Bloomington, MN-WI; S24B St. Louis, MO-IL; S35A Washington-Arlington-Alexandria, DC-VA-MD-WV; S35B Miami-Fort Lauderdale-West Palm Beach, FL; S35C Atlanta-Sandy Springs-Roswell, GA; S35D Tampa-St. Petersburg-Clearwater, FL; S35E Baltimore-Columbia-Towson, MD; S37A Dallas-Fort Worth-Arlington, TX; S37B Houston-The Woodlands-Sugar Land, TX; S48A Phoenix-Mesa-Scottsdale, AZ; S48B Denver-Aurora-Lakewood, CO; S49A Los Angeles-Long Beach-Anaheim, CA; S49B San Francisco-Oakland-Hayward, CA; S49C Riverside-San Bernardino-Ontario, CA; S49D Seattle-Tacoma-Bellevue, WA; S49E San Diego-Carlsbad, CA; S49F Honolulu, HI; S49G Anchorage, AK. Note: Only “S” size PSUs are identified on the public-use microdata.CHAR(4)

STATEWGT

Calibrated full sample final weight for given state population. BLS derived.NUM(11,3)

The following tables show how the weights alter simple sample means when applied for the existing state identifiers in the Interview Survey. They are calendar year estimates using all five quarters of data. The state weights change the sample estimates and provides a more accurate value that represents state population expenditures and characteristics.

Table II. California 2017 Interview Survey Comparison
Interview Variables California Unweighted Sample Estimates California Weighted Sample Estimates

Average Annual Expenditures

$64,327 $64,333

Average Annual Housing Expenses

$23,778 $23,876

Average Age of Reference Person

51.5 50.2

Average Household Size

2.6 2.7
Table III. Florida 2017 Interview Survey Comparison
Interview Variables Florida Unweighted Sample Estimates Florida Weighted Sample Estimates

Average Annual Expenditures

$47,913 $50,157

Average Annual Housing Expenses

$16,988 $17,847

Average Age of Reference Person

53.7 53.1

Average Household Size

2.3 2.4
Table IV. New Jersey 2017 Interview Survey Comparison
Interview Variables New Jersey Unweighted Sample Estimates New Jersey Weighted Sample Estimates

Average Annual Expenditures

$69,047 $74,047

Average Annual Housing Expenses

$26,138 $27,410

Average Age of Reference Person

54.2 53.1

Average Household Size

2.4 2.6
Table V. New York 2017 Interview Survey Comparison
Interview Variables New York Unweighted Sample Estimates New York Weighted Sample Estimates

Average Annual Expenditures

$60,433 $59,323

Average Annual Housing Expenses

$22,074 $22,864

Average Age of Reference Person

53 51.7

Average Household Size

2.3 2.4
Table VI. Texas 2017 Interview Survey Comparison
Interview Variables Texas Unweighted Sample Estimates Texas Weighted Sample Estimates

Average Annual Expenditures

$53,884 $54,908

Average Annual Housing Expenses

$17,815 $18,014

Average Age of Reference Person

49.6 48.5

Average Household Size

2.6 2.7

The following tables show the sum of the weights in the files for which the state-level weights are available. These sums equal the number of consumer units in the given state in each of the quarters. Minor differences between total populations computed using the weights and Census population estimates are due to non-response adjustment and calibration.

Table VII. 2016 Estimated Number of CUs (Sum of Weights)
File Sum of Weights – California Sum of Weights – Florida Sum of Weights – New Jersey

FMLI161x/FMLD161

14,027,341 8,345,518 3,396,891

FMLI162/FMLD162

14,165,407 8,471,073 3,485,309

FMLI163/FMLD163

14,132,369 8,451,935 3,433,119

FMLI164/FMLD164

14,005,236 8,546,045 3,447,063

FMLI171

14,143,974 8,521,142 3,361,656
Table VIII a. 2017 Estimated Number of CUs (Sum of Weights)
File Sum of Weights – California Sum of Weights – Florida Sum of Weights – New Jersey

FMLI171x/FMLD171

14,143,974 8,521,142 3,361,656

FMLI172/FMLD172

14,231,899 8,693,464 3,429,084

FMLI173/FMLD173

14,155,851 8,727,188 3,411,203

FMLI174/FMLD174

14,021,341 8,793,148 3,400,273

FMLI181

14,192,553 8,637,217 3,374,599
Table VIII b. 2017 Estimated Number of CUs (Sum of Weights)
File Sum of Weights – New York Sum of Weights – Texas

FMLI171x/FMLD171

7,884,149 10,203,817

FMLI172/FMLD172

7,970,722 10,305,476

FMLI173/FMLD173

7,850,671 10,232,306

FMLI174/FMLD174

7,889,376 10,308,378

FMLI181

7,992,264 10,357,787

 

Last Modified Date: May 24, 2019