> mdata  <-read.table("wage.csv", header=T, sep=",")
> work   <-mdata[,1]
> hour   <-mdata[,2]
> child  <-mdata[,3]
> age    <-mdata[,4]
> edu    <-mdata[,5]
> wage   <-mdata[,6]
> hinc   <-mdata[,7]
> year   <-mdata[,8]
> mydata <-data.frame(work,hour,child,age,edu,wage,hinc,year)
> library(sampleSelection)
> mysampsel_2step <-selection(work~edu+year+age+child+hinc, wage~edu+year, method="2step")
> summary(mysampsel_2step)
--------------------------------------------
Tobit 2 model (sample selection model)
2-step Heckman / heckit estimation
100 observations (50 censored and 50 observed)
12 free parameters (df = 89)
Probit selection equation:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  1.63565    1.17287   1.395 0.166616    
edu          0.16231    0.07152   2.269 0.025663 *  
year         0.07934    0.02016   3.936 0.000164 ***
age         -0.09014    0.02342  -3.849 0.000223 ***
child       -0.86373    0.39121  -2.208 0.029827 *  
hinc        -0.07682    0.03743  -2.053 0.043048 *  
Outcome equation:
            Estimate Std. Error t value Pr(>|t|)  
(Intercept) -0.78297    2.64560  -0.296   0.7680  
edu          0.40795    0.17326   2.355   0.0207 *
year         0.01658    0.03875   0.428   0.6698  
Multiple R-Squared:0.1865,      Adjusted R-Squared:0.1335
   Error terms:
              Estimate Std. Error t value Pr(>|t|)
invMillsRatio  -1.2190     0.9514  -1.281    0.203
sigma           2.2182         NA      NA       NA
rho            -0.5495         NA      NA       NA
--------------------------------------------