Longitudinal datasets

Codebook for the 1998-2005 longitudinal/panel dataset

--- This dataset includes 8959 respondents aged 80+ plus 34 respondents whose ages were younger than 80 and were interviewed in 1998 (200 elders were interviewed in 1999), and their follow-ups at the 2000, 2002, and 2005 waves. The dataset includes changes in their health conditions, family structures, sociodemographic characteristics, healthcare, and lifestyle in addition to survival status (i.e., died, still alive, and lost to follow-up) at each subsequent wave for each re-visited respondent. The dataset also includes date at death and other nearly 40 questions (including health status, living arrangement, healthcare expense, etc. ) before dying collected from the next-of-kin for those deceased respondents who died between two adjacent waves. Only the baseline weight (i.e., 1998) is included in the dataset. Please refer to Technical Reports regarding how the weight is generated in the CLHLS and how to use weight in the analysis. Please read the specification in the codebook before you use this dataset.

Codebook for the 2000-2005 longitudinal/panel dataset

--- This dataset includes 11,199 respondents who were interviewed in 2000, and their follow-ups at the waves in 2002 and 2005. The dataset includes changes in their health conditions, family structures, sociodemographic characteristics, healthcare, and lifestyle in addition to survival status (i.e., died, still alive, and lost to follow-up) at each subsequent wave for each re-visited respondent. The dataset also includes date at death and other nearly 40 questions (including health status, living arrangement, healthcare expense, etc.) before dying collected from the next-of-kin for those deceased respondents who died between two adjacent waves. Only cross-sectional weight in 2000 is included in the dataset. Please refer to Technical Reports regarding how the weight is generated in the CLHLS and how to use weight in the analysis. Please read the specification in the codebook before you use this dataset.

Codebook for the 2002-2005 longitudinal/panel dataset

--- This dataset includes 16,064 respondents who were interviewed in 2002, and their follow-up in 2005. The dataset includes changes in their health conditions, family structures, sociodemographic characteristics, healthcare, and lifestyle in addition to survival status (i.e., died, still alive, and lost to follow-up) in 2005 wave for each re-visited respondent. The dataset also includes date at death and nearly 40 questions ( including health status, living arrangement, healthcare expense, etc.) before dying collected from the next-of-kin for those deceased respondents who died between two adjacent waves. Only cross-sectional weight in 2002 is included the dataset. Please refer to Technical Reports regarding how the weight is generated in the CLHLS and how to use weight in the analysis. Please read the specification in the codebook before you use this dataset.

If you need to get the dataset, please download the data use agreement and follow the instruction here.