nsw74psid1               package:DAAG               R Documentation

_L_a_b_o_u_r _T_r_a_i_n_i_n_g _E_v_a_l_u_a_t_i_o_n _D_a_t_a

_D_e_s_c_r_i_p_t_i_o_n:

     This data frame contains 2675 rows and 10 columns. These data are
     pertinent to an investigation of the way that    earnings changed,
     between 1974-1975 and 1978, in the absence of training.  Data for
     the experimental treatment  group (NSW) were combined with control
     data results from the  Panel Study of Income Dynamics (PSID)
     study.

_U_s_a_g_e:

     nsw74psid3

_F_o_r_m_a_t:

     This data frame contains the following columns:

     _t_r_t a numeric vector  identifying the study in which the subjects
          were enrolled (0 = PSID, 1 = NSW).

     _a_g_e age (in years).

     _e_d_u_c years of education.

     _b_l_a_c_k (0 = not black, 1 = black).

     _h_i_s_p (0 = not hispanic, 1 = hispanic).

     _m_a_r_r (0 = not married, 1 = married).

     _n_o_d_e_g (0 = completed high school, 1 = dropout).

     _r_e_7_4 real earnings in 1974.

     _r_e_7_5 real earnings in 1975.

     _r_e_7_8 real earnings in 1978. 

_S_o_u_r_c_e:

     http://www.columbia.edu/~rd247/nswdata.html

_R_e_f_e_r_e_n_c_e_s:

     Dehejia, R.H. and Wahba, S. 1999. Causal effects in
     non-experimental studies: re-evaluating the evaluation of training
     programs. Journal of the American Statistical Association 94:
     1053-1062.

     Lalonde, R. 1986. Evaluating the economic evaluations of training
     programs. American Economic Review 76: 604-620.

_E_x_a_m_p_l_e_s:

     print("Interpretation of Regression Coefficients - Example 6.6")

      nsw74psid1.lm <- lm(re78~ trt+ (age + educ + re74 + re75) +
        (black + hisp + marr + nodeg), data = nsw74psid1)
      summary(nsw74psid1.lm)$coef
     options(digits=4)
     sapply(nsw74psid1[, c(2,3,8,9,10)], quantile, prob=c(.25,.5,.75,.95,1))
     attach(nsw74psid1)
     sapply(nsw74psid1[trt==1, c(2,3,8,9,10)], quantile, 
     prob=c(.25,.5,.75,.95,1))
     pause()

     here <- age <= 40 & re74<=5000 & re75 <= 5000 & re78 < 30000 
     nsw74psidA <- nsw74psid1[here, ]
     detach(nsw74psid1)
     table(nsw74psidA$trt)
     pause()

     A1.lm <- lm(re78 ~ trt + (age + educ + re74 + re75) + (black +
           hisp + marr + nodeg), data = nsw74psidA)
     summary(A1.lm)$coef
     pause()

     A2.lm <- lm(re78 ~ trt + (age + educ + re74 + re75) * (black +   
           hisp + marr + nodeg), data = nsw74psidA)
     anova(A1.lm, A2.lm)

