ACS Summary Tables Compared to Microdata

The microdata is a sub-set of the ACS sample.  As a subset, one would expect that it would not produce the same results as shown in the summary tables.  However, I would at least expect it to not vary significantly and to be within the margins of error of the summary table.  However, that is not what I am finding when I use the data.census.gov microdata tool.  Is this an issue with the microdata tool since it is beta or something else causing the large discrepancies? For example, I was comparing the 2022 ACS 1-Year class of worker data for Virginia using summary table S2408 and a microdata query of the same data, and it shows significantly different numbers for each class of worker as you will see if you enter the following URL's.

S2408 Table: https://data.census.gov/table/ACSST1Y2022.S2408?q=class%20of%20worker%20Virginia

Microdata: https://data.census.gov/mdat/#/search?ds=ACSPUMS1Y2022&rv=COW&nv=ucgid&wt=PWGTP&g=0400000US51

Parents
  • Dear Jill,

    You need to use PUMS variables ESR and COW in combination as follows:

    Note the universe for S2408 is the Employed Civilian population 16 years old and over

    # ESR: Employment status recode
    # b [0].N/A (less than 16 years old)
    # 1 .Civilian employed, at work
    # 2 .Civilian employed, with a job but not at work
    # 3 .Unemployed
    # 4 .Armed forces, at work
    # 5 .Armed forces, with a job but not at work
    # 6 .Not in labor force

    # for S2408 Employed ESR== 1 or 2

    # COW Character 1
    # Class of worker
    # b [0].Not in universe (less than 16 years old/NILF who last worked
    # .more than 5 years ago or never worked)
    # 1 .Employee of a private for-profit company or business, or of an
    # .individual, for wages, salary, or commissions
    # 2 .Employee of a private not-for-profit, tax-exempt, or
    # .charitable organization
    # 3 .Local government employee (city, county, etc.)
    # 4 .State government employee
    # 5 .Federal government employee
    # 6 .Self-employed in own not incorporated business, professional
    # .practice, or farm
    # 7 .Self-employed in own incorporated business, professional
    # .practice or farm
    # 8 .Working without pay in family business or farm
    # 9 .Unemployed

    For S2408
    # COW == 2
    "Civilian employed population 16 years and over : Private not-for-profit wage and salary workers"
    # COW == 5
    "Civilian employed population 16 years and over : Federal government workers";
    # COW == 3
    "Civilian employed population 16 years and over : Local government workers";
    # COW == 1 or 7
    "Civilian employed population 16 years and over : Private for-profit wage and salary workers"

    # state=51 (Virginia)

    # 2022 1 year PUMS and subject tables
    #
    # Deviation (MoE)
    # S2408
    # Local government workers 347176 (9635) PUMS=351504 (-347175)=4329
    # Federal government workers 354980 (9837) PUMS=358209 (-354980)=3229
    # Private not-for-profit wage and salary workers 359054 (10818) PUMS=359002 (-359054) = -52
    # Private For-profit wage and salary workers 2839546 (22977) PUMS=2840360 -2839546=814

    You can review the other categories (e.g. state workers and family workers/self employed not incorporated

    The difference between the PUMS estimate and S2408 value/estimate falls within the S2408 MoE

    Best

    Dave Dorer

  • Dave,

    I am new to using the PUMS data. This is immensely helpful! The ESR is what I was unfamiliar with, but now I got it working and can move forward with additional analysis.  I will be sure to keep an eye on the table universe going forward.

    Thank You!

    Jill Kaneff

Reply
  • Dave,

    I am new to using the PUMS data. This is immensely helpful! The ESR is what I was unfamiliar with, but now I got it working and can move forward with additional analysis.  I will be sure to keep an eye on the table universe going forward.

    Thank You!

    Jill Kaneff

Children
  • Dear Jill,

    I'm curious about what stat software you use.  Also I don't have details on your project and the entity supporting your analysis.  But I have a foundation that does pro bono consulting for 501(c) (3) and government entities. (dorerfoundation.org).  Dave Dorer

  • Dear Jill,

    I just "googled" you and it appears that you work for

    NVRC which is a Commonwealth of Virginia entity and would be eligible to receive services from DCSF.  I did some work using ACS data with Erik Beecroft at the Virginia JLARC.

    Best Dave