Hi IPUMS Team,
I’m working with ASEC short-panels (2 years) and ATUS data and would appreciate your advice on the best practice for linking across the two sources in a panel framework. I want to create a dataset that combines longitudinal ASEC data with time-use caregiving ATUS data at T2.
Data I’m working with:
- ATUS: 2003–2023 (n≈245,000 =; cross-sectional)
- ASEC: 2003–2024 (n≈970,571; short-panels; long format)
- Objective: Produce pooled short-panels of linked ASEC-ATUS (2003–2023) data & apply best practices for linking
So far:
-
I’ve restricted the ASEC to respondents with demographically validated links across the 2 waves (using CPSIDV). So the same individual is observed in March of year t (MISH == 4 ) and March of year t+1 (MISH == 8 ).
-
But I’m a little stuck on how I should link them. When should I use MIS=8 only vs. MIS 5-8 for ATUS-ASEC linking?
I think I understand the conceptual difference:
So essentially, I think I should use ATUS respondents whose MONTH_CPS8 == 3 (March final ASEC interview) and link ATUS to ASEC Year = YEAR_CPS8 where MISH == 8 (i.e., the follow-up ASEC wave) by CPSIDP. The final output would be both ASEC waves (T1 & T2), but attach ATUS info only to T2, since ATUS interviews are conducted after the last CPS/ASEC interview (between 2-5 months after).
Also, a few additional thoughts:
-
Am I selecting for a weird/non-random group of people if I’m restricting my sample to anyone that could be linked across ASEC short-panels + ATUS?
-
In terms of weights, I see there are weights for use with linked data (variable names start with [LNKFW]), but according to this IPUMS forum post: “There are no weights available that can be used for an analysis of ATUS linked to the CPS ASEC at two points in time, though you can make your own.” Is this still the case, that I have to make my own? Any guidance on this front would be great!
Thanks so much in advance for any light shed on these questions!
To clarify, my research goal is to build a panel to observe move-ins of adult children who are likely providing care to their older parents with health issues. I’m now realizing the IPUMS ASEC longitudinal files contain only the person records found in both of the ASEC waves, with variables for these records at both time points. Since these are balanced panels, I can’t observe any household composition changes. Is there another way to observe household move-ins? Could I accomplish this by using single-year cross-sectional ASEC files, linking households across years on CPSID + year and persons on CPSIDV, and detecting move-ins by comparing the household roster between the two years? Alternatively, can I use the basic monthly CPS files and follow households month-to-month to pick up new entrants? In both cases, I’d like to merge in ATUS data, whenever possible. Thanks for your patience & any expertise provided with this evolving project!
Based on your description, I do not see any immediate reason for restricting your links to a single month-in-sample value. Persons with a month-in-sample (MISH) value between 5-8 in the CPS ASEC are scheduled to complete their final interview in that same year and, if selected, will be interviewed for the ATUS three months after. For example, if someone were MIS 5 in March 2023, they would be MIS 8 in June 2023 and eligible to be interviewed for the ATUS sometime around September 2023. While some households in the sample will not complete their final CPS interview and therefore will not be eligible for the ATUS, you can significantly increase your sample size by retaining households in the ASEC that are eligible together with those who may become eligible to link in that year. This does not hinder your ability to construct a two-year longitudinal ASEC panel; anyone with MISH values between 5-8 in one ASEC can typically be linked to the previous year’s ASEC (provided that they are not part of the ASEC oversample). Also, note that you can link persons across any of their CPS panel months; there is no inherent reason to restrict to those who link between the ASEC samples unless there are specific ASEC variables that you require for your analysis.
Since both the CPS and ATUS are not completely random samples, you will need to carefully consider your weighting strategy when analyzing this linked sample. There is no official recommended way to do this; it may be helpful to consult the literature to see what other researchers have done. One method that uses an OLS regression with demographic characteristics to predict linking using the full ATUS sample is described in these linked CPS-ATUS supplement files.
Your suggested method of comparing household rosters over time makes sense to me. While you will not observe records of individuals before they move into the sampled household, you’ll still be able to observe these compositional changes. You can also use this method to identify cases where the parent moved in to their child’s home. Note that the ASEC includes person-level migration status variables MIGRATE1, MIGSTA1, and WHYMOVE, which may aid in your analysis.
Hi Ivan & IPUMS Team,
Thanks so much for your thorough response! I’ve given more thought on how to identify individuals moving into households, but I’m still unclear on the best approach – how do I know, or at least approximate, who’s a new entrant to the household?
Currently, I’m using the Basic Monthly CPS (2021-2023) to observe household compositional changes over time. I’ve been using the first observed year-month in my extract (regardless of MISH) to flag potential move-ins by comparing each person’s first observed date (pers_first_date ) to their household’s first observed date (hh_first_date ).
Based on the following, I flag the individual as a potential mover-in with their parents:
pers_first_date > hh_first_date- MOMLOC or POPLOC is non-zero or non-missing
(coreside_flag==1)
AGE>=25 (to avoid college kids, since my target sample are adult kids)
However, I’ve noticed:
- In each HH,
PERNUM == 1 will never have pers_first_date > hh_first_date since hh_first_date is based on the first person’s survey info (usually PERNUM == 1).
- Since
PERNUM == 1 is typically the household head, I’ve assumed an adult child moving in with a parent typically wouldn’t assume headship immediately, so this might be a minimal source of bias, but I plan to check this systematically.
Alternatively, I have also considered:
- Identifying movers-in by simply restricting each individual to MISH 2+, but because some people only appear in MISH 5-8 in my extract (their MISH 1-4 occurred before my 2021 start date), but they weren’t necessarily move-ins. I got this idea from an IPUMS Forum post: “I would expect anyone who joins (or replaces) a household in MISH 2 or later to have moved in the past year.”
- Would checking the first 6 digits of CPSID vs CPSIDP, as suggested in this IPUMS Forum post, help identify movers-in?
- Checking for changes in
METAREA or the most granular geography as a proxy for moves, but this won’t work because the CPS surveys residences and doesn’t follow households/individuals as they move.
Long story short, initially, I chose the BMS for its 16-month panel with monthly detail, but now I see that migration variables (MIGRATE1, WHYMOVE) come from the ASEC. I could merge these into the BMS, but that limits me to the ASEC subset, and only a certain proportion of BMS respondents have matched ASEC data.
Given these tradeoffs, is it better to continue using the BMS to observe household compositional changes for detecting movers-in, or is manually linking cross-sectional ASEC files to utilize migration variables a cleaner approach for identifying movers-in in this context?
Apologies for the long-winded reply and thanks in advance for any help!
The person designated the household head (i.e., the person with RELATE = 101) does not change between survey panels unless that person moves out of the household. An adult child moving in with a parent will not replace their parent as the household head. More broadly, the Census Bureau discontinued the use of the term “head of household” beginning with the 1980 CPS and replaced it with the concept of a “householder”. A householder is a person in whose name the housing unit is owned or rented. The householder is determined when the household enters the panel; they are listed first in their household roster (assigned PERNUM = 1) and are designated the “reference person” to whom the relationship of all other household members is recorded (see the Census Bureau’s subject definitions glossary).
My recommendation is to begin by obtaining the month that each person first appears in the CPS panel. With your dataset that includes all monthly samples from 2021-2023, you can group matching CPSIDV values and sort them by YEAR, MONTH. Generate a new variable for each group of matching CPSIDV values that equals the first YEAR, MONTH combination that they are observed in. You can then compare this value to the first six digits of that person’s CPSIDV value (which is the same as their household’s CPSID value); these first six digits (or entirety of the variable if using CPSID) index the year and month that the household entered the CPS. Therefore, you will be looking for cases where the date the household entered the panel does not match the date of the first appearance of the person’s CPSIDV value. For example, an individual that first appears in the February 2022 sample in a household with a CPSID value that begins with “202201” would have moved in between January and February 2022 (i.e., they became a usual resident; see table 3-2.4 in technical paper 77 for how household membership is determined). Note that you will only be able to identify these movers for households whose first month-in-sample is within the years included in your data (i.e., MISH = 1 in your earliest observation period). If your sample begins in January 2021, you will not have the full history of households who entered before this date.
The forum post that you link to that suggests checking the first 6 digits of CPSID vs CPSIDP is inaccurate; the first digits of CPSID and CPSIDP/V will match regardless of when the individual first appears in the sample. Thank you for bringing our attention to this! We will revise our team’s response to that post to ensure it doesn’t cause confusion.