The Child Development Supplement (CDS) is one research component of the Panel Study of Income Dynamics (PSID).
While the PSID has always collected some information about children, in 1997, PSID supplemented its main data collection with additional information on 0-12 year-old children and their parents.
The objective was to provide researchers with a comprehensive, nationally representative, and longitudinal data base of children and their families with which to study the dynamic process of early human capital formation.
The Original CDS was collected in three waves: CDS-I in 1997, CDS-II in 2002/2003, and CDS-III in 2007/2008.
In 2014 the CDS methodology was changed to a steady state design, collecting information on all sample children aged 0-17. For more information on the Ongoing CDS, please see the CDS-2014 User Guide and the CDS-2019 User Guide.
By nature of the CDS being a supplement to the PSID, the study takes advantage of an extensive amount of family demographic and economic data about the CDS target child's family, providing more extensive family data than any other nationally-representative longitudinal survey of children and youth in the U.S. In addition, the PSID-CDS data are "intergenerational" in structure with information contained in several decades of data about multiple family members. This rich data structure allows analysts a unique opportunity to fully link information on children, their parents, their grandparents, and other relatives to take advantage of the rich intergenerational and long-panel dimensions of the data. |
Within the context of family, neighborhood, and school environments, CDS studies a broad array of developmental outcomes including (but not limited to) physical health, emotional well-being, intellectual achievement, and social relationships with family and peers. These outcomes are measured through reliable, age-graded assessments of cognitive and behavioral development and health status indicators obtained from the primary caregiver and the sample children/youth themselves; anthropometric measures of height and weight of the sample children/youth; a comprehensive accounting of parental (or caregiver) time inputs to children/youth as well as other aspects of the way the children/youth spent their time; and other-than-time use measures of other resources for example, the learning environment in the home (using the HOME Scale measures), school resources, as reported through the National Center for Education Statistics Common Core of Data, and decennial-census-based measurement of neighborhood resources. The multi-level, interdisciplinary, and longitudinal nature of the research design facilitates analysis of the relationships between these developmental measures and changes in family structure and living arrangements, neighborhood economic and social conditions, and school resources and programs. |
| Questionnaires and supporting documentation for the PSID and its supplemental studies are located on the questionnaires and supporting documents page. |
The Original CDS-TAS sample was drawn from PSID families with children 0-12 years in 1997, and the Ongoing CDS sample was drawn from PSID families with children 0-17 years old in 2014 and 2019.
The PSID sample combines the SRC (Survey Research Center) and SEO (Survey of Economic Opportunity) samples.
Both the CDS-TAS and PSID samples are probability samples (i.e., samples for which every element in the population has a known nonzero chance
of selection). Their combination is also a probability sample. The combination, however, is a sample with unequal selection probabilities, and as a result,
compensatory weighting is needed in estimation, at least for descriptive statistics. Weight adjustments are also needed to attempt to
compensate for differential nonresponse across waves. Weights supplied on CDS and TAS data files are designed to compensate for both unequal selection probabilities and differential attrition.
In the 2002, 2007, 2014 and 2019 CDS demographic files, you will find a set of indicator variables for each module that specify (a)
if a case was eligible for that module and (b) if a record exists for that case in the corresponding data file.
These variables are helpful to merge onto your Data Center data request if you are merging variables from multiple CDS modules.
The sample weight in the Demographic file is adjusted only for the non-response in the main module, Primary Caregiver (sections A-H, J).
The module indicator variables, however, will inform you about item missing data across modules. It is up to you to then decide on your preferred
approach for addressing item missing data that results from differential response rates across modules (for example, you may leave it as missing,
impute scores, etc). The TAS data files contain wave-specific sample weights.
More documentation on the CDS and TAS weights can be found on the documentation page. |
| Every individual in the PSID - including the children - has both an "ID68" (1968 Family Identifier - ER30001) and "PN" (Person Number - ER30002) that combine to uniquely identify that individual. As a user of the CDS data, you can use these identifiers to find information about the CDS targeted child and caregivers in the PSID data files. Background information about the CDS target child, such as birth date, sex, and relationship to the PSID family household Reference Person (starting with the 2017 wave, the term ‘Reference Person’ has replaced ‘Head') can be obtained from the PSID individual and sampling variables files. Use the ER30001 and ER30002 combination to select the PSID variables for just the CDS target child sample, or, when you get to the "Output Options" page in the Data Center, after selecting the variables you want, select "CDS Children" at the bottom. |
There are two PSID family history files that may be of particular interest to CDS users: the Childbirth and Adoption History File and the Parent Identification File.
Childbirth and Adoption History File:
The Childbirth and Adoption History File is specifically designed to facilitate access to detailed information collected since 1985 regarding histories of childbirth and adoption. Variables on this file include the identifiers for each parent and child, month and year of birth for both parent and child, birth order, birth weight and date of death for a child, year of most recent report and number of births/adoptions, etc. Data on this file are structured in a one-record-per-event format, with each record representing a specific childbirth or adoption event.
Parent Identification File:
The Parent Identifier File synopsizes information collected from various sources since the 1983 wave of PSID about parent-child relationships. This file consists of identifier variables that link children with their parents. The file is intended to be used to facilitate linking children's and parents' data records from the Individual File. Linkages can be done from either the child's or a parent's standpoint.
|
There are a large number of variables in the PSID that can be used along with CDS.
Demographic, health, economic, and other family data about PCG (primary caregivers) and OCG (other caregivers) can be found in the PSID data files. Every individual in the PSID has
both "ID68" (1968 Family Identifier - ER30001) and "PN" (Person Number- ER30002) that combine to uniquely identify that individual. As a user of the CDS data, you can use these identifiers
to find information about the CDS targeted child and caregivers in the PSID data files. These identifier variables are available through a Child to Caregiver Map, provided with each Data Center download. |
The child to caregiver map, provided with each CDS data download, provides "1968 INTERVIEW NUMBER" (ID68) and "PERSON NUMBER 68" (PN) for CDS individuals.
These CDS individuals are the target child, the target child's primary caregiver (PCG) in both the Original and Ongoing CDS, as well as the target child's other caregiver (OCG) in the Original CDS,
if one exists. Missing data means that the child did not have an OCG for the CDS interview year.
All CDS files, by default, contain variables ER30001 (1968 INTERVIEW NUMBER) and ER30002 (PERSON NUMBER 68). Since these variables are also in the map file, the map file can be used to merge PCG and OCG data from PSID Individual data to CDS Child level data in a two step process. |
There are two steps to locating data for siblings in the CDS data files:
In the Demographic Data File, there is a sibling indicator variable that tells you if a CDS target child had a sibling who also participated in the CDS data collection.
Automatically appended to your data download is the "Family Interview Identification Number" for the corresponding PSID main interview. This variable uniquely identifies the family.
Using these two variables, you can locate data on a wide range of information about the target children and their siblings in the CDS.
See also the codebook explanation text for the family identification number. There is a variable for any year in the PSID in both
the individual and family files. |
Original CDS: In CDS-I, height of the child was measured by the interviewer and weight was reported by the parent. In CDS-II and CDS-III, both height and weight were measured by the interviewer.
Ongoing CDS: In CDS-2014 and CDS-2019 height and weight were measured by the interviewer for families that participated in the in-home module. For children not included in the in-home module, height and weight were reported by the parent. |
The Behavior Problem Index was originally developed by James Peterson and Nicholas Zill from the Achenbach Behavior Problems Checklist to measure in a survey setting the incidence and severity of child behavior problems. The BPI scale is based on responses by the primary caregiver as to whether a set of 32 problem behaviors is often, sometimes, or never true of the targeted child.
These items are then divided into two subscales: 1) a measure of externalizing or aggressive behavior and 2) a measure of internalizing, withdrawn or sad behavior. The User Guide specifies the individual items that map into the internalizing and externalizing subscales.
We performed a confirmatory factor analysis on our two expected subscales. The results showed that the items grouped into these two factors quite readily, with one variable overlapping on both subscales, as did in CDS-I, and two variables not loading at all. We constructed an overall or total BPI score, using all 32 items, as well as separate scores for each of the two subscales, internal or withdrawn and external or aggressive. Before scoring, the individual items are recoded such that a score of "1" becomes "0" and a score of "2" or "3" become a "1". Scores for the total BPI and Externalizing and Internalizing are sum scores. Higher scores on these measures imply a greater level of behavior problems. Cases were included if they had data approximately 75% valid data on the variables contributing to the BPI Indices. |
The Home Observation for Measurement of the Environment-Short Form from the Caldwell and Bradley HOME Inventory is used as a measure of cognitive stimulation and emotional support that parents provide to their children. The particular items used in the PSID Child Development Supplement were taken directly from the National Longitudinal Survey of Youth, Mother-Child Supplement so that the scales would be as similar as possible. The HOME-SF items include both parent/caregiver-reported items and interviewer observations of the home and neighborhood environment. The HOME-SF is divided into four parts:
- Infant/Toddler (IT) HOME, designed for use during infancy (birth to age three);
- Early Childhood (EC) HOME, designed for use between 3 and 6 years of age;
- Middle Childhood (MC) HOME, for use between 6 and 10 years; and
- Early Adolescent (EA) HOME, designed for use from 10 to 15 years old.
We have included three scores for HOME-SF for each age module appropriate for CDS-II and CDS-III data: 1) a total raw score, 2) an emotional support subscale raw score, and 3) a cognitive stimulation subscale raw score. The total and subscale raw scores for the HOME-SF are a summation of the recoded individual item scores and varies by age group, as the number of individual items varies according to the age of the targeted child / youth.
Additional information about the HOME-SF in the CDS can be found in the CDS User Guide.
|
The Woodcock-Johnson Psycho-Educational Battery-Revised (WJ-R) provides a normed set of tests for measuring cognitive abilities and academic achievement. In the Original CDS, CDS-I-III, we selected three subtests as a measure of reading and match achievement: the Letter-Word, the Passage Comprehension, and the Applied Problems tests (the Calculation test was additionally administered in CDS-I. These scales can be used individually, or in the case of the four subscales, combined to create scores for Broad Reading and Broad Math. When applicable, the Spanish version of the WJ-R (Batería-R, Form A), was used for children whose primary language was Spanish.
The Woodcock-Johnson Revised (WJ-R) tests of achievement have standardized administrative and scoring protocols. The tests are designed to provide a normative score that shows the CDS target child's reading and match abilities in comparison to national average for the child's age. The normed scores are constructed based on the child's raw score on the test (essentially the number of correct items completed) and the child's age to the nearest month. Raw scores are charted on normative tables based on the child's age and what percentile the child falls into. More information on scoring is provided in the
CDS User Guides. |
| In CDS I, we included two Woodcock Johnson - Revised math-skill tests: Calculations and Applied Problems. A broad math score was constructed based on these two tests. In CDS II, we only included the Applied Problems; hence, no broad math score can be constructed - just a score for applied problems. |
| File release information is available through the News section of our website. You can also sign up to have the news delivered to your
email by logging in and selecting to receive updates on the "Settings" page. |
| Only one file is allowed in your cart if CDS Time Diary is selected. To add CDS Time Diary variables to your cart, you must select variables from just one file, and there cannot be any variables from other files in your cart. Time diary data are not at an individual or family level (like other data in the data center), so the data center cannot merge them automatically. |
To open the .txt files into Stata (SAS, or SPSS), save both the .txt and .do (.sas. or .spss) file from your download to your machine and take note of the path where the .txt file is located.
Once you save the .txt file to your computer you will open the read-in statements (.do, .sas, or .spss) into your statistical program. Replace the section of the read-in statements “[path]” with the path of where you saved your data text file, for example, C:/yourcomputer/yourfiles/. Once you have identified the path (the location of the .txt file) you will run the read-in statements and the program will read-in the data and label your variables accordingly. You can then save the resulting data set as a data file.
For step-by-step instructions, please see the web tutorial Accessing and Downloading PSID Data. |
|