Appendix A - NSCH Sampling Flags Creation Documentation

A sample frame built from administrative records for the
National Survey of Children’s Health
Keith Finlay
Center for Administrative Records
Research and Applications
US Census Bureau
[email protected]
301-763-6056
February 3, 2016
This document describes using administrative records to build a sample frame for the National
Survey of Children’s Health (NSCH).

Population of interest
The population of interest is all children residing in the US on the date of the survey. The survey
should provide oversamples of children with chronic disabilities. The sample frame should also
provide some information about household access to the Internet.

A sampling frame for all households with children
The sampling frame for all households with children comes from three data sources: the Numident,
a list of Social Security Number applicants with data updated from various administrative records;
and the CARRA kidlink file, a prototype linkage between children and parents based on Census
and administrative records. Household addresses are updated with the Master Address Auxiliary
Reference File, a file that links person identifiers with the latest location updates from a variety of
administrative data.

Using the Numident to identify children
The Numident is based on off the all individuals who have been assigned Social Security Numbers.
Demographic data from the Numident is updated from federal tax data and various administrative
records. There are 71,873,129 children in the 2015 Numident who will be aged 0–17 years on April
1, 2016 . Figure 1 shows the distribution of date of birth for these children.
The Numident is updated monthly.
1

Identifying the households containing the children in the Numident
To sample households with children, we must connect the children in the Numident to the households
in which they live. We do this with two files: the 2010 Census Unedited File and the CARRA
kidlink file.
CARRA kidlink
The CARRA kidlink file uses data from Census survey and federal administrative records to link
children PIKs to parent PIKs. We can use this file to identify the parents of children in the Numident.
The source data for the CARRA kidlink file are: the Census Numident, the 2010 Census Unedited
File, the IRS 1040 and 1099 files, the Medicare Enrollment Database (MEDB), Indian Health
Service database (IHS), Selective Service System (SSS), and Public and Indian Housing (PIC) and
Tenant Rental Assistance Certification System (TRACS) data from the Department of Housing and
Urban Development. Of these, the IRS 1040 provides the most significant information.
There are 68,519,439 unique records for children in the 2014 kidlink who will be aged 0–17 years
on April 1, 2016 .
Let us consider how many children from the Numident have been linked to a parent in the CARRA
kidlink file. Table 1 shows the number of children linked with both a mother and a father, linked
with a mother only, linked with a father only, or not linked with any parent.
Figure 2 compares the distributions of date of birth for these children against the distribution shown
in Figure 1.
The current vintage of the CARRA kidlink file is 2014. This explains the missing kidlink entries
for the youngest children in the Numident (the very right side of the distribution in Figure 2). The
CARRA kidlink file will be updated by April 1, 2016, with the newest versions of the input files for
final sample frame production.

Updating household location using the MAF-ARF
In order to update household location, we use a Census dataset called the Master Address Auxiliary
Reference File (MAF-ARF). The MAF-ARF links person identifiers to address identifiers using
Census survey data and federal administrative data. The source data for the MAF-ARF file are:
the Census Numident, the 2010 Census Unedited File, the IRS 1040 and 1099 files, the Medicare
Enrollment Database (MEDB), Indian Health Service database (IHS), Selective Service System
(SSS), and Public and Indian Housing (PIC) and Tenant Rental Assistance Certification System
(TRACS) data from the Department of Housing and Urban Development, and National Change
of Address data from the US Postal Service. Of these, the IRS 1040 provides the most significant
information.
Out of 71,873,129 children in the Numident, 55,763,902 are matched directly to a MAFID. Out
of 59,762,607 kidlink-matched mothers, 54,630,103 are matched to a MAFID. Out of 48,628,252
kidlink-matched fathers, 44,866,739 are matched to a MAFID.
2

For each child observation from the Numident, we now have four possible MAFIDs: the SSI
MAFID, the kid to MAF-ARF MAFID, the child-to-kidlink-to-mother-to-MAF-ARF MAFID,
and the child-to-kidlink-to-father-to-MAF-ARF MAFID. I allocate the single MAFID using that
order. First, I assign the SSI MAFID (1,713,591 cases). If MAFID is missing, I assign the directly
identified child MAFID (54,341,170 cases). If the MAFID is still missing, I assign the mother
MAFID (4,932,199 cases). Finally, if the MAFID is still missing, I assign the father MAFID
(1,662,678 cases). That leaves 9,223,491 children from the Numident not assigned MAFIDs (a
MAFID match rate of 87.2%).
There are some MAFIDs associated with a great number of children. As an example, out of
62,649,638 children associated with a MAFID, 295,328 children are associated with a MAFID with
more than 20 child-MAFID links.
The 62,649,638 children associated with a MAFID are then collapsed down to 34,353,877 unique
MAFIDS. This implies 1.82 children per household for households assigned a flag.
We then need to scale up the MAFID list to the universe of valid MAFIDs to allow sampling of
unflagged households. A merge of the 34,353,877 unique child-flagged MAFIDS with the ACS
MAF-X file matches 30,717,480 MAFIDS with child flags, removes 3,636,397 MAFIDS with
child flags, and adds 100,608,270 MAFIDs without child flags. The sample frame file now has
131,316,961 valid MAFIDS, of which 30,717,480 MAFIDS include child flags . Compare this with
the 2011 ACS, in which 37,147,503 out of 114,991,725 households included related children.1
The MAF-ARF will be updated by April 1, 2016, with the newest versions of the input files for final
sample frame production.

A sampling frame for disabled children
A subpopulation of children with disabilities comes from the Social Security Agency’s Supplemental
Security Income (SSI) program. Children with certain disabilities from households with low-enough
income are eligible for an SSI subsidy.2
There are 1,849,126 unique records for children in the 2014 SSR who will be aged 0–17 years on
April 1, 2016 . Figure 3 compares the distributions of date of birth for these children against the
distribution shown in Figure 1.
There are a number of sampling concerns with using SSI recipients for the disability oversample:
• Using this subpopulation to create a disability sample would likely oversample children with
severe disabilities. The list of “compassionate allowances” can be found at the Social Security
site.3
• Conditioning on SSI may introduce nonrandom selection on household income for three
reasons.
1

http://www.census.gov/prod/2013pubs/p20-570.pdf
http://www.socialsecurity.gov/pubs/EN-05-10026.pdf
3
http://ssa.gov/compassionateallowances/conditions.htm
2

3

– Households with lower income are more likely to be eligible4 , but eligibility may raise
household income through the subsidy.
– These children are also more likely to get Medicaid support for healthcare expenses.
– Parental labor supply may be affected by the severity of the disability.
• Children on SSI are certainly more likely to be in the Numident.
• It may take some months for disabled children to show up in SSI. In the age distribution
figure, it’s clear that SSI recipients are older than children from the Numident.
• Our Supplemental Security Record (SSR) data file includes the variable PSTAT-CUR, which
is the payment status code for the current month. We do not have retrospective information
on the child’s payment status.

Addresses for SSI recipients
Of the children on SSR, 0.878 have been matched to a MAFID. (This compares with a MAF match
rate of 0.882 for the entire SSI data file.) We believe that the children on SSI have relatively updated
addresses, so those children can be linked directly to Master Address File IDs. The SSI file is
updated annually. The current SSR file is from 2014. We will use a 2015 vintage for the NSCH.

Sample frame construction visualization
Figure 4 shows a visualization of the sample frame construction.

Auditing the sample frame against the ACS
To examine the performance of the administrative records used to build the sampling frame, we
merge the list of MAFIDs constructed above with the American Community Survey housing-unit
sample from 2014. Currently, this audit uses unedited ACS data (i.e., item nonresponse are left as
missing and are not imputed including children’s age). If item nonresponse is random with respect
to the presence of children in the household, this should not cause any systematic bias in the audit.
All estimates are weighted with the housing-unit-level weights, which include weight for vacant
units (214,137 vacant housing units in the 2014 ACS). In vacant housing units, we assign zero
children. These estimates should reflect the NSCH survey production process.
Table 2 shows the overlap between the MAFID and ACS distributions with respect to whether any
children were present in the household.
4

http://ssa.gov/ssi/text-child-ussi.htm

4

Child flag performance by age group
We are particularly interested in the coverage of young children. In this section, we show how the
child flags perform for specific age groups. These are stricter tests since any deviation in age beyond
the age interval will cause either a Type 1 or Type 2 error.
Table 3 shows the overlap between the MAFID and ACS distributions with respect to whether any
children aged 0–2 years were present in the household. Given that the input administrative records
used to construction the child flags are 1–2 years old and that the ACS data are from 2014, it is not
surprising that the overlap for children aged 0–2 years is much lower than the overall rate shown in
Table 2.
Table 4 shows the overlap between the MAFID and ACS distributions with respect to whether any
children aged 3–5 years were present in the household. By ages 3–5, overlap between the child flag
and the ACS data is above 60%.
Table 5 shows the overlap between the MAFID and ACS distributions with respect to whether any
children aged 6–8 years were present in the household.
Table 6 shows the overlap between the MAFID and ACS distributions with respect to whether any
children aged 9–11 years were present in the household.
Table 7 shows the overlap between the MAFID and ACS distributions with respect to whether any
children aged 12–14 years were present in the household.
Table 8 shows the overlap between the MAFID and ACS distributions with respect to whether any
children aged 15–17 years were present in the household.

State-specific performance
Table 9 shows the overlap between the MAFID and ACS distributions by state. The smallest
oversample strata are in Hawaii, Maine, Vermont, and West Virginia. The largest oversample strata
are in California, Texas, and Utah. The highest rates of Type 1 error are in DC, Florida, Louisiana,
and Mississippi. The highest rates of Type 2 error are in Alaska, Hawaii, Texas, and Utah.

An Internet-accessible household flag
Here I describe the construction of tract- and block-varying Internet-accessible household flags.
The data come from American Community Survey paradata and IRS 1040 filing mode data.
Since 2012, ACS respondents have been able to submit survey forms over the Internet. ACS paradata
record whether a respondent chose the online option. The ACS paradata has been summarized at
the tract level. Our Internet-accessible household measure is equal to a weighted proportion of
the respondents that chose to submit the ACS survey over the Internet if given the option to do so.
Figure 5 shows the kernel-smoothed distribution of tract-level Internet response for the 2013–2014
ACS survey years.

5

We also get a measure of local Internet accessibility from IRS 1040 filing data. Filers have a choice
of filing electronically. We identify filers who file electronically but without a paid preparer, and
infer that these individuals file at home using the Internet. We then calculate an electronic self file
rate by Census block. Figure 6 shows the distribution of Census-block-level electronic self file for
tax year 2014.
To synthesize the information from these two measures into a single index, we use principal
components analysis (PCA). PCA is a data reduction technique that finds the linear combination
of the two input variables that maximizes the variance of a single index. PCA uses standardized
forms of the variables, so the predicted index is also in standardized form. Figure 7 shows the
distribution of the predicted Census-block-level score from the PCA. We can consider this variable
a Census-block-level Internet accessibility index.

A 150% poverty rate flag
Here I describe the construction of a block-group-varying flag for the proportion of households
below 150% of the poverty line. The data come from 2014 5-year American Community Survey
file. Figure 8 shows the distribution of block-group-level 150% poverty rate flag.

Final sample frame data layout
The three component data files are merged together based on MAFID. The data layout for this
combined file is given in Table 10.

6

List of Figures
1

Distribution of date of birth, Numident, aged 0–17 years as of April 1, 2016 . . . .

8

2

Frequency distributions of date of birth, Numident vs. kidlink entries, aged 0–17
years as of April 1, 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

8

Distributions of date of birth, Numident vs. SSI recipients, aged 0–17 years as of
April 1, 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9

4

Sample frame construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9

5

Kernel-smoothed probability distribution function of tract-level ACS Internet response rate, ACS paradata, 2013–2014 survey years . . . . . . . . . . . . . . . . . 10

6

Kernel-smoothed probability distribution function of the Census-block-level rate of
IRS electronic self 1040 filing, IRS 1040 data, tax year 2014 . . . . . . . . . . . . 10

7

Kernel-smoothed probability distribution function of the Internet accessibility index,
ACS paradata (2013–2014 survey years) and IRS 1040 data (tax year 2014) . . . . 11

8

Kernel-smoothed probability distribution function of block-group-level 150%
poverty rate, ACS, 2014 5-year file . . . . . . . . . . . . . . . . . . . . . . . . . . 11

3

7

0

5.0e−05

Density
1.0e−04

1.5e−04

2.0e−04

Figure 1: Distribution of date of birth, Numident, aged 0–17 years as of April 1, 2016

02apr1998

01jan2002

01jan2008
Date of birth

01jan2014

0

Frequency
5.0e+05

1.0e+06

Figure 2: Frequency distributions of date of birth, Numident vs. kidlink entries, aged 0–17 years as
of April 1, 2016

02apr1998

01jan2002

01jan2008
Date of birth
Numident

8

kidlink entries

01jan2014

0

Density
5.0e−05 1.0e−04 1.5e−04 2.0e−04 2.5e−04

Figure 3: Distributions of date of birth, Numident vs. SSI recipients, aged 0–17 years as of April 1,
2016

02apr1998

01jan2002

01jan2008
Date of birth
Numident

01jan2014

SSI recipients

Figure 4: Sample frame construction
Numident: children
in population
SSI recipients:
kids to MAFIDs

MAF-ARF: kids
to MAFIDs

Kidlink: kids
to moms and dads

MAF-ARF: moms
to MAFIDs

Set of PIK-MAFID links
(four possible)

Set of child-flagged MAFIDs
(collapse from PIK-level to MAFID-level;
prioritizing SSI, then child MAFID,
then mother MAFID, then father MAFID)
Complete set of ACS household MAFIDs
(append valid MAFIDs,
exclude invalid MAFIDs)

9

MAF-ARF: dads
to MAFIDs

0

1

Density

2

3

Figure 5: Kernel-smoothed probability distribution function of tract-level ACS Internet response
rate, ACS paradata, 2013–2014 survey years

0

.2
.4
.6
.8
ACS Internet response rate, weighted, by tract

1

0

.5

1

Density
1.5

2

2.5

Figure 6: Kernel-smoothed probability distribution function of the Census-block-level rate of IRS
electronic self 1040 filing, IRS 1040 data, tax year 2014

0

.2
.4
.6
.8
Rate of IRS 1040 electronic self filing, weighted, by Census block

10

1

0

.1

Density
.2

.3

.4

Figure 7: Kernel-smoothed probability distribution function of the Internet accessibility index, ACS
paradata (2013–2014 survey years) and IRS 1040 data (tax year 2014)

−6

−4
−2
0
2
Internet accessibility index, weighted, by Census block

4

0

1

Density

2

3

Figure 8: Kernel-smoothed probability distribution function of block-group-level 150% poverty
rate, ACS, 2014 5-year file

0
.2
.4
.6
.8
1
Proportion of individuals below 150% of poverty line, weighted, by block group

11

12

List of Tables
1

Child-parent links in the CARRA kidlink file relative to the Numident population,
aged 0–17 years as of April 1, 2016 . . . . . . . . . . . . . . . . . . . . . . . . . 14

2

Comparison of NSCH child flags and ACS data, any children in household, 2014
ACS, housing unit weights including vacants . . . . . . . . . . . . . . . . . . . . 14

3

Comparison of NSCH child flags and ACS data, any children in household aged
0–2 years, 2014 ACS, housing unit weights including vacants . . . . . . . . . . . . 14

4

Comparison of NSCH child flags and ACS data, any children in household aged
3–5 years, 2014 ACS, housing unit weights including vacants . . . . . . . . . . . . 15

5

Comparison of NSCH child flags and ACS data, any children in household aged
6–8 years, 2014 ACS, housing unit weights including vacants . . . . . . . . . . . . 15

6

Comparison of NSCH child flags and ACS data, any children in household aged
9–11 years, 2014 ACS, housing unit weights including vacants . . . . . . . . . . . 15

7

Comparison of NSCH child flags and ACS data, any children in household aged
12–14 years, 2014 ACS, housing unit weights including vacants . . . . . . . . . . 15

8

Comparison of NSCH child flags and ACS data, any children in household aged
15–17 years, 2014 ACS, housing unit weights including vacants . . . . . . . . . . 16

9

Comparison of NSCH child flags and ACS data, any children in household, 2014
ACS, housing unit weights including vacants, by state . . . . . . . . . . . . . . . . 17

10

NSCH population data file layout . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

13

Table 1: Child-parent links in the CARRA kidlink file relative to the Numident population, aged
0–17 years as of April 1, 2016
Type of link
Frequency Percent
Mother and father
45,918,616
64%
Mother only
13,843,991
19%
Father only
2,709,636
3.8%
No link
9,400,886
13%
All children in Numident 71,873,129
100%

Table 2: Comparison of NSCH child flags and ACS data, any children in household, 2014 ACS,
housing unit weights including vacants
Observed ACS households
NSCH child flags
No children Any children
Total
No children
90.3%
9.7%
100.0%
Any children
22.7%
77.3%
100.0%
Total
74.6%
25.4%
100.0%
N (ACS households) 2,322,722

Table 3: Comparison of NSCH child flags and ACS data, any children in household aged 0–2 years,
2014 ACS, housing unit weights including vacants
Observed ACS households
NSCH child flags
No children 0–2 Any children 0–2
Total
No children 0–2
98.3%
1.7%
100.0%
Any children 0–2
62.0%
38.0%
100.0%
Total
97.5%
2.5%
100.0%
N (ACS households)
2,322,722

14

Table 4: Comparison of NSCH child flags and ACS data, any children in household aged 3–5 years,
2014 ACS, housing unit weights including vacants
Observed ACS households
NSCH child flags
No children 3–5 Any children 3–5
Total
No children 3–5
96.7%
3.3%
100.0%
Any children 3–5
38.2%
61.8%
100.0%
Total
92.9%
7.1%
100.0%
N (ACS households)
2,322,722

Table 5: Comparison of NSCH child flags and ACS data, any children in household aged 6–8 years,
2014 ACS, housing unit weights including vacants
Observed ACS households
NSCH child flags
No children 6–8 Any children 6–8
Total
No children 6–8
96.6%
3.4%
100.0%
Any children 6–8
35.4%
64.6%
100.0%
Total
92.3%
7.7%
100.0%
N (ACS households)
2,322,722

Table 6: Comparison of NSCH child flags and ACS data, any children in household aged 9–11
years, 2014 ACS, housing unit weights including vacants
Observed ACS households
NSCH child flags
No children 9–11 Any children 9–11
Total
No children 9–11
96.6%
3.4%
100.0%
Any children 9–11
33.0%
67.0%
100.0%
Total
92.1%
7.9%
100.0%
N (ACS households)
2,322,722

Table 7: Comparison of NSCH child flags and ACS data, any children in household aged 12–14
years, 2014 ACS, housing unit weights including vacants
Observed ACS households
NSCH child flags
No children 12–14 Any children 12–14
Total
No children 12–14
96.7%
3.3%
100.0%
Any children 12–14
31.7%
68.3%
100.0%
Total
92.1%
7.9%
100.0%
N (ACS households)
2,322,722

15

Table 8: Comparison of NSCH child flags and ACS data, any children in household aged 15–17
years, 2014 ACS, housing unit weights including vacants
Observed ACS households
NSCH child flags
No children 15–17 Any children 15–17
Total
No children 15–17
96.6%
3.4%
100.0%
Any children 15–17
31.1%
68.9%
100.0%
Total
91.9%
8.1%
100.0%
N (ACS households)
2,322,722

16

Table 9: Comparison of NSCH child flags and ACS data, any children in household, 2014 ACS,
housing unit weights including vacants, by state
NSCH frame
Any children (a)
No children (b)
N (c)
ACS obs. children Any (d) None (e) Total Any (f) None (g)
Total
(d)/(a)
(e)/(a)
(a)/(c) (f)/(b)
(g)/(b) (b)/(c)
(c)
State
×100
×100
×100 ×100
×100
×100
Alabama
71.9
28.1
21.2
9.9
90.1
78.8 37,511
Alaska
72.4
27.6
19.8
15.6
84.4
80.2
9,534
Arizona
74.5
25.5
21.3
10.0
90.0
78.7 44,646
Arkansas
71.6
28.4
21.6
11.0
89.0
78.4 22,495
California
80.1
19.9
27.3
10.9
89.1
72.7 217,111
Colorado
82.7
17.3
22.8
9.9
90.1
77.2 37,691
Connecticut
79.6
20.4
23.2
8.2
91.8
76.8 23,385
Delaware
76.1
23.9
22.3
7.9
92.1
77.7
7,367
District of Columbia
68.6
31.4
16.6
6.9
93.1
83.4
4,693
Florida
69.0
31.0
19.7
7.8
92.2
80.3 121,828
Georgia
74.4
25.6
25.5
11.7
88.3
74.5 57,019
Hawaii
72.4
27.6
14.8
18.4
81.6
85.2
9,856
Idaho
80.2
19.8
23.0
10.6
89.4
77.0 11,545
Illinois
78.2
21.8
24.6
8.7
91.3
75.4 97,583
Indiana
78.4
21.6
24.5
8.5
91.5
75.5 48,569
Iowa
83.3
16.7
23.9
7.0
93.0
76.1 34,025
Kansas
79.3
20.7
25.6
8.3
91.7
74.4 26,961
Kentucky
75.6
24.4
22.3
10.6
89.4
77.7 34,115
Louisiana
68.0
32.0
24.2
11.1
88.9
75.8 31,206
Maine
77.3
22.7
15.7
6.3
93.7
84.3 17,636
Maryland
79.2
20.8
25.0
9.0
91.0
75.0 39,331
Massachusetts
82.3
17.7
22.7
7.5
92.5
77.3 43,395
Michigan
80.0
20.0
22.4
6.2
93.8
77.6 100,990
Minnesota
83.9
16.1
23.5
6.7
93.3
76.5 72,611
Mississippi
70.1
29.9
24.1
11.9
88.1
75.9 18,761
Missouri
76.4
23.6
22.4
8.3
91.7
77.6 50,595
Montana
77.5
22.5
17.2
8.5
91.5
82.8 11,567
Nebraska
84.1
15.9
24.1
8.4
91.6
75.9 21,002
Nevada
72.6
27.4
20.7
11.4
88.6
79.3 18,288
New Hampshire
82.4
17.6
19.5
6.9
93.1
80.5 11,239
New Jersey
81.0
19.0
24.4
9.5
90.5
75.6 57,087
New Mexico
71.7
28.3
18.9
12.6
87.4
81.1 16,173
New York
75.0
25.0
19.9
11.6
88.4
80.1 138,735
North Carolina
76.0
24.0
22.2
9.7
90.3
77.8 68,857
North Dakota
81.1
18.9
20.0
9.8
90.2
80.0
9,642
Ohio
78.1
21.9
24.0
7.2
92.8
76.0 90,191
Oklahoma
71.9
28.1
22.6
12.3
87.7
77.4 46,397
Oregon
82.3
17.7
21.6
7.9
92.1
78.4 26,748
Pennsylvania
79.8
20.2
21.6
7.0
93.0
78.4 120,084
Rhode Island
79.4
20.6
21.4
8.0
92.0
78.6
6,819
South Carolina
72.4
27.6
21.6
8.8
91.2
78.4 32,989
South Dakota
76.3
23.7
20.6
10.0
90.0
79.4
9,957
Tennessee
75.0
25.0
23.3
10.0
90.0
76.7 44,043
Texas
76.1
23.9
26.5
14.3
85.7
73.5 146,897
Utah
82.5
17.5
31.2
13.8
86.2
68.8 18,761
Vermont
83.1
16.9
15.9
7.8
92.2
84.1
9,097
Virginia
79.8
20.2
24.7
9.4
90.6
75.3 54,668
Washington
80.6
19.4
22.9
9.0
91.0
77.1 47,839
West Virginia
74.3
25.7
14.4
11.2
88.8
85.6 15,434
Wisconsin
82.5
17.5
22.2
6.9
93.1
77.8 75,291
Wyoming
76.7
23.3
19.0
11.7
88.3
81.0
4,458
17

Table 10: NSCH population data file layout
Variable name
Label
Format
mafid
Master Address File ID
long integer
maf_curstate
State
str2
maf_curcounty
County
str3
maf_curblktract
Tract
str6
maf_curblkgrp
Block group
str1
kids_00_02
Number of children aged 0–2 years
integer
kids_03_05
Number of children aged 3–5 years
integer
kids_06_08
Number of children aged 6–8 years
integer
kids_09_11
Number of children aged 9–11 years
integer
kids_12_14
Number of children aged 12–14 years
integer
kids_15_17
Number of children aged 15–17 years
integer
any_ssi
Any children in household on SSI?
byte
acs_tract_net_response Tract-level ACS Internet response
float
block_elf_self_rate
Block-level 1040 electornic self file rate float
block_net_access_index Block-level Internet accessibility index float
blockgroup_150povrate Block group-level 150% poverty rate
float
Filename: nsch_pop_file.sas7bdat
Population: all MAFIDs with valid housing unit types
Unit of observation: household (MAFID)
Number of observations: 131,316,961
Filesize: 9GB

18

Domain
9 digits

≥0
≥0
≥0
≥0
≥0
≥0
{0, 1}
[0, 1]
[0, 1]
(−∞, ∞)
[0, 1]