Thailand Law Journal 2009 Spring Issue 1 Volume 12

dummy equal to 1 if any OJT occurred in year h. OFFJT incidence: dummy equal to 1 if any OFFJT occurred in year h. OJT intensity: average duration of OJT in hours per month. OFFJT intensity: average duration of OFFJT in hours per month. Cumulated OJT: sum of OJT intensity from 1998 to year h. Cumulated  OFFJT: sum of OFFJT intensity from 1998 to year h. Father's and mother's education: higher than primary = 1, primary or less = 0. Oldest sibling's education: percentage with college degree. Previous experience in 1998: labor market experience net of tenure in 1998.    

The average age of the sampled employees was about 28 years, and their average tenure in 1998, the start of the reference period, was about 2-1/2 years for men and over 3 years for women. Previous labor market experience in 1998, defined as the time spent in the labor market from the start of the first job up to (but not including) the current job, was close to 2 years for both men and women. [FN8] The vast majority of the employees in the survey had started their current spell and their labor market experience before 1998. For these individuals, tenure and experience after 1998 are equal to tenure and experience in 1998 plus a linear trend.

The survey asked respondents to indicate their highest attained degree. We convert the answers to years of education by applying to each degree the required number of years of schooling. [FN9] Average years of attained education in the sample were 12.90 for men and 10.66 for women--statistically significantly higher, for both sexes, than attained years of education across the Thai population.

We collect information both about on-the-job training (OJT) and off-the-job (OFFJT) training provided by the firm after probation. Probation in Thailand lasts up to six months, and training during probation is typically highly structured and standardized within each firm. Therefore, we expect little variation in training incidence and intensity during probation within firms, and virtually no variation once we control for firm effects, as we do in our training regressions. [FN10] On-the-job training is administered in the workplace by senior employees, supervisors, or instructors and focused on the performance of daily tasks. By and large, this type of training tends to be more informal and ad hoc [FN11] than off-the-job training. In the survey, we explicitly excluded learning by doing and learning by watching others because it is virtually impossible to distinguish these activities from regular work.

Off-the-job training is organized by the firm either on or outside the premises, and either during or after standard working hours. While OJT, in our sample, was provided mainly by senior workers in the same unit, team leaders, and foremen (72% of the total), OFFJT was supplied mainly by instructors from professional training centers and from outside the company (86% of the total) and concerned not only daily tasks, but also instruction in the operation of machines and tools, quality control, work standards, and safety regulations.

For each type of training, we asked whether the employee had any training during the reference period, a simple (0,1) dummy, and we also asked about training intensity. We call the former variable OJT incidence for on-the-job training and OFFJT incidence for off-the-job training. For each individual, training intensity is calculated as the product of the number of training events in each month and the average intensity of each event, measured in hours. Starting with training incidence, we find that 55% of the men in the sample had undertaken some OJT in 2001, compared to 67% of the women.

Interestingly, the opposite holds for OFFJT, with 67% of the men and 58% of the women having received it. In the case of training intensity, we find similar differences by gender and type of training. The number of hours of training per month was 2.62 for men and 3.18 for women in the case of OJT and 1.57 for men and 0.87 for women in the case of OFFJT. Thus, men experienced longer OFFJT and shorter OJT than women. Similar results hold for cumulated OJT and OFFJT, obtained by adding up average monthly hours from 1998 to 2001. Further details on the training questions are in the Appendix, and the entire questionnaire is available upon application from the website http:// www.kier.kyoto-u.ac.jp (or search for Institute of Economic Research, Kyoto University.)

Turning to family background, we present in the table summary statistics on the father's and mother's education, the education of the oldest sibling (or the next to oldest if the employee was the oldest), and the number of siblings. For ease of presentation, the information on the former two variables is recoded to generate two (0,1) dummies, with 0 indicating primary education or less and 1 indicating higher education. The education of the oldest sibling is expressed as the percentage of individuals with a college degree. It turns out that male employees, who had higher educational attainment than women, also had a "better" family background, with higher average education for both parents and the oldest sibling, as well as fewer siblings.

Table 3 shows the means of the same variables separately for production workers and for team leaders, foremen, technicians, and engineers. As expected, the last of those groups was better educated than the others and received more formal OFFJT and less OJT. In the questionnaire we also asked whether OFFJT was carried out in the establishment premises or outside it, either in a training center, at a supplier, or at a parent firm. Training outside the premises of the establishment was more common for technicians and engineers than for other workers.

[FN8]. When the current job is the first job, experience is equal to zero.

[FN9]. We set 0 years for no education, 6 years for primary, 9 years for lower secondary, 12 years for upper secondary and lower vocational, 14 years for upper vocational, and 16 years for college or more.

[FN10]. Another reason for excluding training during probation is that training programs for a newly hired employee are conducted mostly before the employee's assignment to a particular section or unit in the firm. We have defined OJT and OFFJT as training taking place, respectively, inside and outside the regular workplace. This distinction does not make sense for new recruits.

[FN11]. Informal training is difficult to measure. See the discussions in Barron et al. (1997) and Lowenstein and Spletzer (1994).

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.