Current Population Survey Design and Methodology - Census Bureau
Current Population Survey Design and Methodology - Census Bureau
Current Population Survey Design and Methodology - Census Bureau
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
have completed their eighth <strong>and</strong> final CPS interview (i.e.,<br />
CPS MIS 8). There is a two-month lag from the last CPS<br />
interview, to initial contact with the household for an ATUS<br />
interview. For example, households selected from the<br />
January 2005 CPS sample would be part of the March<br />
2005 ATUS sample.<br />
The CPS households from which the ATUS sample is<br />
selected are stratified by race/ethnicity <strong>and</strong> presence of<br />
children. Non-White households are sampled at a higher<br />
rate to insure that valid comparisons can be made across<br />
major race/ethnicity groups. Half of the households are<br />
assigned to report on Saturdays <strong>and</strong> Sundays, <strong>and</strong> half are<br />
assigned to report on weekdays. One person 15 years or<br />
older is selected from each ATUS sample household for<br />
participation in the ATUS.<br />
Weighting procedure<br />
The basic weight for each ATUS record is the product of<br />
the CPS first-stage weight, an adjustment for the CPS<br />
state-based design, the within-stratum sampling interval<br />
<strong>and</strong> a household size factor. The ATUS basic weight is then<br />
adjusted by an ATUS noninterview factor, plus a set of<br />
population control adjustments; the population control<br />
adjustments are based on sex, age, race/ethnicity, education,<br />
labor force status, <strong>and</strong> presence of children in the<br />
household. A day adjustment factor is computed separately<br />
for weekdays, Saturdays, <strong>and</strong> Sundays. This factor<br />
accounts for increased numbers of weekend interviews,<br />
<strong>and</strong> for varying frequencies of each day of the week for a<br />
particular month.<br />
Annual Social <strong>and</strong> Economic Supplement (ASEC)<br />
Description of supplement<br />
The ASEC is sponsored by the <strong>Census</strong> <strong>Bureau</strong> <strong>and</strong> the BLS.<br />
The <strong>Census</strong> <strong>Bureau</strong> has collected data in the ASEC since<br />
1947. From 1947 to 1955, the ASEC took place in April,<br />
<strong>and</strong> from 1956 to 2001 the ASEC took place in March.<br />
Prior to 2003, the ASEC was known as the Annual Demographic<br />
Supplement or the March Supplement. In 20012 ,a<br />
sample increase was implemented that required more time<br />
for data collection. Thus, additional ASEC interviews are<br />
now taking place in February <strong>and</strong> April. Even with this<br />
sample increase, most of the data collection still occurs in<br />
March.<br />
The supplement collects data on family characteristics,<br />
household composition, marital status, education attainment,<br />
health insurance coverage, the foreign-born population,<br />
work experience, income from all sources, receipt of<br />
2 The exp<strong>and</strong>ed sample was first used in 2001 for testing <strong>and</strong><br />
was not included in the official ADS statistics for 2001. The statistics<br />
from 2002 are the first official set of statistics published<br />
using the exp<strong>and</strong>ed sample. The 2001 exp<strong>and</strong>ed sample statistics<br />
were released <strong>and</strong> are used for comparing the 2001 data to the<br />
official 2002 statistics.<br />
<strong>Current</strong> <strong>Population</strong> <strong>Survey</strong> TP66<br />
U.S. <strong>Bureau</strong> of Labor Statistics <strong>and</strong> U.S. <strong>Census</strong> <strong>Bureau</strong><br />
noncash benefit, poverty, program participation, <strong>and</strong> geographic<br />
mobility. A major reason for conducting the ASEC<br />
in the month of March is to obtain better income data. It<br />
was thought that since March is the month before the<br />
deadline for filing federal income tax returns, respondents<br />
were likely to have recently prepared tax returns or be in<br />
the midst of preparing such returns <strong>and</strong> could report their<br />
income more accurately than at any other time of the year.<br />
The universe for the ASEC is slightly different from that for<br />
the basic CPS. It includes certain members of the armed<br />
forces in the estimates. This requires some minor changes<br />
to the sampling procedures <strong>and</strong> to the weighting methodology.<br />
The ASEC sample consists of the March CPS sample, plus<br />
additional CPS households identified in prior CPS samples<br />
<strong>and</strong> the following April CPS sample. Table 11−2 shows the<br />
months when the eligible sample is identified for years<br />
2001 through 2004. Starting in 2004, the eligible ASEC<br />
sample households are:<br />
1. The entire March CPS sample.<br />
2. Hispanic households—identified in November (from all<br />
month-in-sample (MIS) groups) <strong>and</strong> in April (from MIS<br />
1 <strong>and</strong> 5 groups).<br />
3. Non-Hispanic non-White households—identified in<br />
August (MIS 8), September (MIS 8), October (MIS 8),<br />
November (MIS 1 <strong>and</strong> 5), <strong>and</strong> April (MIS 1 <strong>and</strong> 5).<br />
4. Non-Hispanic White households with children 18 years<br />
or younger—identified in August (MIS 8), September<br />
(MIS 8), October (MIS 8), November (MIS 1 <strong>and</strong> 5), <strong>and</strong><br />
April (MIS 1 <strong>and</strong> 5).<br />
Prior to 1976, no additional sample households were<br />
added. From 1976 to 2001, only the November CPS<br />
households containing at least one person of Hispanic origin<br />
were added to the ASEC. The households added in<br />
2001, along with a general sample increase in selected<br />
states, are collectively known as the State Children’s<br />
Health Insurance Program (SCHIP) sample expansion. The<br />
added households improve the reliability of the ASEC estimates<br />
for the Hispanic households, non-Hispanic non-<br />
White households, <strong>and</strong> non-Hispanic White households<br />
with children 18 years or younger.<br />
Because of the characteristics of CPS sample rotation (see<br />
Chapter 3), the additional cases from the August, September,<br />
October, November <strong>and</strong> April CPS are completely different<br />
from those in the March CPS. The additional sample<br />
cases increase the effective sample size of the ASEC compared<br />
with the March CPS sample alone. The ASEC sample<br />
includes 18 MIS groups for Hispanic households, 15 MIS<br />
groups for non-Hispanic non-White households, 15 MIS<br />
groups for non-Hispanic White households with children<br />
18 years or younger, <strong>and</strong> 8 MIS groups for all other households.<br />
<strong>Current</strong> <strong>Population</strong> <strong>Survey</strong> Supplemental Inquiries 11–5