Kamhon Kan and Myoung-jae Lee, "Lose Weight for a Raise Only if 
Over-Weight: Marginal Integration for Semi-Linear Panel Models", Journal 
of Applied Econometrics, Vol. 27, No. 4, 2012, pp. 666-685.

In the paper we use two different samples of the National Longitudinal
Survey of the Youth 1979 data. The two samples are in two separate files:

  wf2yr.csv: used in the paper's marginal integration analysis.
  wf2yr_iv.csv: used in the paper's instrumental variable estimation. 

In the file wf2yr_iv, we have a few more variables (i.e., the instruments,
which pertain to information about a respondent's siblings) and fewer
observations because we delete dropped respondents for whom there is 
missing information on their siblings.

Both data files are ASCII files in DOS format. They are zipped in the file
kl-data.zip. Unix/Linux users should use "unzip -a".


Description of wf2yr.csv:

Number of Observations: 2604

Variable names:

id, lnw1, bmi, everborn youngkid, nokidhh, wcbc, miss_wc,  marr_sp,
marr_ot,  hgc, miss_hgc, enrolled, expwks miss_e, tenure, miss_t,
i_20hrs, miss_i, age, lur_low, lur_high, miss_lur, year, reg_ne, reg_nc,
reg_s, ht1, bmi_re, ht_re, wt_re, ht1,
momhgc, miss_mh, dadhgc, miss_dh,  gpcs1, miss_g

Variable definitions:

id: individual identification number
lnw1: log wage per hour 
everborn: number of children ever born
youngkid: age of youngest kid
nokidhh: no children in household 
wcbc: white collar job (dummy variable)
miss_wc: wcbc missing (dummy variable)  
marr_sp: married with spouse in household (dummy variable)
marr_ot: married but spouse in the household (dummy variable)
hgc: years of schooling
miss_hgc: hgc missing (dummy variable)
enrolled: currently enrolled in school (dummy variable)
expwks: work experience (years) 
miss_e: expwks missing (dummy variable)
tenure: tenure at current job (years)
miss_t: tenure missing (dummy variable)
i_20hrs: work more than 20 hours per week (dummy variable)
miss_i: i_20hrs missing (dummy variable)
age: respondent's age
lur_low: unemployment rate in county of residence below 6%  (dummy variable)
lur_high: unemployment rate in county of residence above 9% (dummy variable)
miss_lur: lur_low and lur_high missing 
year: year of interview
reg_ne: live in Northeast (dummy variable)
reg_nc: live in North Central (dummy variable)
reg_s: live in South (dummy variable)
ht1: height in inch in 1986
bmi_re: measurement error corrected body mass index 
ht_re: height in inch 
wt_re: measurement error corrected body weight (pounds) 
ht1: height (inch) in 1986
momhgc: mother's years of schooling  
miss_mh: momhgc missing (dummy variable)
dadhgc: father's years of schooling
miss_dh: dadhgc missing (dummy variable)
gpcs1: a measure of general intelligence
miss_g: gpcs1 missing (dummy variable)

Sources:

The data come from the National Longitudinal Survey of the Youth 1979 
(http://www.bls.gov/nls/nlsy79.htm). We obtained a cleaned version of the
data from Professor Cawley, who used the data in his paper (Cawley J. 2004.
"The impact of obesity on wages," Journal of Human Resources 39:
451-474.). We extracted the 1986 and 2000 waves of Professor Cawley's data
for our paper.


Description of wf2yr_iv.csv:

Number of Observations: 1074

Variable names:

id, lnw1, bmi, everborn youngkid, nokidhh, wcbc, miss_wc,  marr_sp,
marr_ot,  hgc, miss_hgc, enrolled, expwks miss_e, tenure, miss_t,
i_20hrs, miss_i, age, lur_low, lur_high, miss_lur, year, reg_ne, reg_nc,
reg_s, ht1, bmi_re, ht_re, wt_re, ht1,
momhgc, miss_mh, dadhgc, miss_dh,  gpcs1, miss_g,
sib1_bmi_re, sib1_female, sib1_age   

Variable Definitions:

sib1_bmi_re: body mass index of a respondent's sibling 
sib1_female: sibling is female (dummy variable) 
sib1_age: sibling's age   

For questions about the data, please contact:

Kamhon Kan
Institute of Economics, Academia Sinica
Taipei, Taiwan 11529
kk [AT] sinica.edu.tw