Table 5

Panel data analysis of smoking behavior by birth cohort: long panels

Birth Cohort	1910–19	1920–29	1930–39	1940–49	1950–59	1960–69	1970–79
A. Males
Linear Model
College	0.028^*** [0.0049]	0.020^*** [0.0034]	0.013^*** [0.0030]	−0.004^* [0.0022]	−0.020^*** [0.0018]	−0.020^*** [0.0020]	−0.003 [0.0042]
Observations	423,771	604,413	533,016	561,945	508,813	264,811	46,571
Individuals	7,402	12,161	13,172	18,546	22,664	15,965	4,182
Conditional Logit Model (marginal effects)
College	0.014 [0.0176]	0.029 [0.0244]	0.0006 [0.0075]	−0.004 [0.0047]	−0.062^*** [0.0234]	−0.023 [0.0367]	0.096 [0.0597]
Observations	240,899	347,600	277,826	251,326	171,357	61,515	9,075
Individuals	4,200	6,915	6,755	8,137	7,447	3,629	817
B. Females
Linear Model
College	−0.013^*** [0.0033]	−0.0017 [0.0030]	−0.051^*** [0.0027]	−0.043^*** [0.0020]	−0.038^*** [0.0016]	−0.028^*** [0.0017]	−0.019^*** [0.0033]
Observations	781,462	885,527	688,575	692,416	640,443	343,626	60,493
Individuals	13,284	17,534	16,890	22,729	28,713	20,783	5,440
Conditional Logit Model (marginal effects)
College	−0.056 [0.0402]	−0.019 [0.0121]	−0.123^*** [0.0300]	−0.197^*** [0.0383]	−0.168^*** [0.0256]	−0.168^*** [0.0339]	−0.124^*** [0.0639]
Observations	220,317	316,512	256,349	250,788	190,848	72,549	9,294
Individuals	3,827	6,297	6,310	8,124	8,411	4,313	835
Analysis from smoking histories from the 1978, 1979, 1980, 1983,					1985, 1987,	1988, 1990,	1991, 1992,

↵Analysis from smoking histories from the 1978, 1979, 1980, 1983, 1985, 1987, 1988, 1990, 1991, 1992, 1994, 1995, 1997, 1998, 1999 and 2000 National Health Interview Surveys for individuals age 25 and older at the time of the survey. Smoking histories are reconstructed using the age at smoking initiation and cessation. See also Note 5. The age of college graduation is assumed to be 21 and the specifications include age and year dummies. Robust standard errors in square brackets. ***, **, * denote statistical significance at the 1 percent, 5 percent, and 10 percent confidence levels, respectively.