USER NOTESPD - 3rd LONGITUDINAL FILEA revised data dictionary and data file were released on March 7, 2006. In the previous SPD 3rd longitudinal file, people who permanently left the universe were incorrectly classified as "longitudinally non-interviewed." Zero weights were given to people who were qualified to be classified as longitudinally interviewed until they died or permanently moved outside the survey universe, and through the non-interview adjustment, others in the sample were weighted up to account for them. This error creates a bias by transferring the weights of these dead and out-of-universe people to the living, longitudinally interviewed sample people. Therefore, on the previous SPD 3rd longitudinal file, the longitudinally interviewed sample people represented incorrectly both the living people and the dead and out-of-scope people in the SPD/SIPP universe. From the estimation standpoint, the final weights on the previous SPD 3rd longitudinal file would produce substantial over-estimation of living people. On the current SPD 3rd longitudinal file, we have corrected this error by reweighting the file and classifying these people as longitudinally interviewed up until they came out of scope. Last updated: 5/22/06
|
|
Contact: (dsd.survey.program.dynamics@census.gov)
| Introduction to SPD |
Survey Design & Content |
Data Editing |
Finding
SPD Info | Sampling &
Weighting |
|
|
Census
2000 | Subjects
A to Z | Search
| Product
Catalog | Data
Access Tools | FOIA
| Privacy
· Policies | Contact
Us | Home
|