Thailand Law Journal 2009 Spring Issue 1 Volume 12

 Table 4. Distribution of Employees by Education.    

--------------------------------------------------------
                           Years of Schooling          

                 ---------------------------------------
No. of Siblings         <9  9         12        >12    

--------------------------------------------------------
0-1                   6.18  14.61     40.73     38.48  

2                      8.21  19.57     33.09     39.13  

3                      8.84  24.39     36.59     30.18  

>3                  19.34  23.43     33.96     23.27  

--------------------------------------------------------

Results

Training Incidence and Intensity

We estimate a bi-probit model for OJT and OFFJT incidence and a tobit model for OJT and OFFJT intensity, after pooling all available observations over the period 1998-2001 and including in the vector Y in (1) and (2) gender, a third order polynomial in age, firm dummies, and a linear trend and quadratic trend, which capture the influence of aggregate effects.[FN14] We present our estimates in Tables 5 and 6, which are organized in four columns. The first two columns treat years of education as exogenous, and the latter two test the exogeneity of education using the procedure devised by Blundell and Smith (1986).

 Table 5. Bivariate Probit Estimates of Training Incidence, Full Sample, 1998-2001. (dependent variables: OJT and OFFJT incidence; robust standard errors in parentheses)  

-------------------------------------------------------------------------------
                  Years of Education Treated as     Testing for Exogeneity of  

                            Exogenous                       Education          

                 -------------------------------  -----------------------------
Variable               OJT            OFFJT             OJT           OFFJT    

-------------------------------------------------------------------------------
Sex              -0.160 [FNaaa1]   0.092 [FNa1]       -0.031          0.020    

                     (.047)           (.050)          (.107)          (.110)  

Age              0.305 [FNaaa1]   0.213 [FNaa1]   0.397 [FNaaa1]      0.141    

                    (.102)           (.099)          (.133)          (.128)  

Age2           -0.010 [FNaaa1]  -0.005 [FNa1]   -0.013 [FNaaa1]     -0.003  

                     (.003)           (.003)          (.004)          (.003)  

Age3 [FNa1]    0.001 [FNaaa1]    0.000 [FNa1]   0.001 [FNaaa1]      0.000    

  10                                                                          

                               (.000)           (.000)                  (.000)          (.000)  

                                                                              
Tenure in 1998       -0.009       0.022 [FNaaa1]      -0.010          0.021    

                                                                                              [FNaa1]  

                               (.007)           (.008)                   (.008)          (.008)  

                                                                              
Previous         -0.013 [FNaa1]       0.005        -0.013 [FNa1]      0.004    

  Experience in                                                                

  1998                                                                        

                               (.006)           (.006)                    (.007)          (.007)  

                                                                              
Years of         -0.055 [FNaaa1]  0.032 [FNaaa1]  -0.140 [FNaa1]      0.079    

  Education                                                                    

                               (.008)           (.009)                    (.069)          (.068)  

                                                                              
First Step                     --               --                         0.084          -0.048  

  Residuals                                                                    

                                                                                  (.068)          (.068)  

                                                                              
<<rho>>                 0.334 [FNaaa1]                          0.326 [FNaaa1]              

                                    (.025)                                          (.022)                  

                                                                              
Observations                    5,525           5,525                   5,465   5,465    

-------------------------------------------------------------------------------
   Notes: Each regression includes firm dummies, family background variables, and a linear and a quadratic trend. <<rho>> = correlation between training error terms. Sex: men = 1, women = 0.                                        

[FN14]. These trends also capture the time-varying effects of experience and tenure.

This article is published with the kind permission of Kenn Ariga and Giorgio Brunello. The article originally appeared in Volume 59, Issue 4, July 2006, of the Industrial and Labor Relations Review. Copyright Cornell University.