邁入21世紀，高科技產業已成為台灣最重要的支柱產業，產業間製程技術不斷日新月異，即便企業擁有最先進的儀器設備，也要有懂得操作的人員才能達到最大的生產效率，因為企業無法直接取得營運所需的知識與技能，必須間接透過個人投資人力資本來累積，而在職訓練是完成正式教育後持續累積人力資本的重要途徑，其對個人及企業的重要性與日俱增。以往國內在職訓練的文獻多集中於廠商面的分析，本文嘗試使用橫斷面的個體資料，深入探討個人面在職訓練的報酬率與決定因素，實證分析所使用的資料來源為行政院主計處所發布的「97年人力運用調查」，經過處理後共有817筆樣本，由於無法明確的判斷哪些勞工有投資在職訓練，哪些勞工沒有投資在職訓練，必須使用未知狀態的內生轉換回歸模型來取得薪資方程式的參數估計，再透過經驗-薪資曲線的形態來觀察訓練成本的分攤與訓練的專屬性，並計算出Overtaking發生的時點。實證結果符合人力資本理論的假設，有投資在職訓練者的起薪較低，但工資率成長較快，每多增加一年的工作經驗會提高3.5至3.7%的工資率，而未投資在職訓練者的起薪較高，但工資率不會成長，兩者的起薪差距不大，這也表示台灣高科技產業大多從事廠商特定型訓練。考慮不同轉換點之下，計算出發生Overtaking的年數介於0.7年至6.25年，稍微早於國外的研究，也發現個人的特性變數確實會顯著影響投資在職訓練的機率，且廠商規模越大越重視人才的培育。

Step into the 21st century. High-tech industry has become the most important pillar industry in Taiwan. Through the inter-industry continued innovation of processing technology, the companies not only require the most advanced equipment, also need the optimal employees to maximize the production efficiency. Because companies can’t get the knowledge and skills directly, so need a person to accumulation indirectly in the human capital. Moreover, On-the-job training is one of the most important ways of human capital accumulation when formal education over. OJT has become more and more important to individuals and companies. Most of domestic literatures focused on the company side. This paper attempts to use cross-sectional individual data to study the rate of return in OJT and the decision factors, empirical analysis of sources of information used in the premises for the Executive Yuan published "97-year survey of human use", remaining the 817 samples after treatment. As we have no prior information on who invests in OJT and who does not, we use an endogenous switching regression model with unknown regimes to obtain parameter estimates of the wage profiles, and through the wage profiles form to understand cost-sharing and training specificity, further calculated the number of overtaking years. The empirical results confirm the hypothesis of human capital theory. The workers engaged in OJT have lower initial salary, but growth of wage rate is faster. Each additional year of working experience can increase 3.6% to 3.7% wage rate. On the other hand, the workers not engaged in OJT have higher initial salary, but the wage rate will not grow. Smaller salary gap between the two groups said the most training of high-tech industries in Taiwan are firm-specific training. This study considers different switch points. The estimation of overtaking level is ranging from 0.7 to 6.25, slightly earlier than the study abroad. Our study also found that the variables of individual characteristics are significantly affects the probability of investment in OJT, and more scale firm more emphasis on foster talent.

中文部分：
[1] 單 驥 (1986),“台灣職業訓練現況分析及廠商辦理職業訓練
因素之決定,”台灣人力資源論文集, pp 363-391.
[2] 王雅鈴 (1999),“員工在職訓練之效益分析-以新竹科學園區廠
商為例,”國立中央大學產業經濟研究所碩士論文
[3] 鄭琇仁 (1996),“在職訓練對廠商行為決策之影響-台灣地區營
造業之實證分析,” 國立中央大學產業經濟研究所碩士
論文
[4] 黃慈蕙 (1998),“高科技產業正規在職訓練決定因素之研究-新
竹科學工業園區積體電路產業實證分析,” 國立中央大
學人力資源管理研究所碩士論文
英文部分：
[1] Barron, J.M. , Berger, M.C. and Black, D.A. (1999),
“Do Workers Pay On-The-Job Training,” The
Journal of Human Resources, 34(2), pp 235-252.
[2] Becker, G.S. (1962), “Investment in Human Capital: A
Theoretical Analysis,”The Journal of Political
Economy, 70(5), pp 9-49.
[3] Linder, A.S. (1976), “On Dogmatism in Human Capital
Theory,” The Journal of Human Resources,
11(1), pp 8-22.
[4] Dickens, W.T. and Lang, K. (1985), “A Test of Dual
Labor,” American Economic Review, 75(4),
pp 792-805.
[5] Duncan, G.J. and Hoffman, S. (1979), “On-The-Job
Training and Earnings Difference by Race and
Sex,” The Review of Economics and Statistics,
61(4), pp 594-603.
[6] Dutoit, L. (2007), “Heckman’s Selection Model,
Endogenous and Exogenous Switching Models, A
survey,” The Selected Works of Laure C
Dutoit, Available at:
http://works.bepress.com/laure_dutoit/3
[7] Ericson, T. (2008), “The Effect of Wage Compression
on General and Firm-Specific Training,”Applied
Economics Letters, pp 165-169.
[8] Groot, W. and Mekkelholt, E. (1995), “The Rate of
Return to Investments in on-the-job Training,”
Applied Economics, 27, pp 173-181.
[9] Hans van Ophem, (1993), “A Modified Switching
Regression Model for Earnings Differentials
Between the Public and Private Sectors in the
Netherlands,” The Review of Economics and
Statistics, 75(2), pp 215-224.
[10] Heckman, J.J. (1979), “Sample Selection Bias as a
Specification Error,”Econometrica, 47(1),
pp 153-161.
[11] Hotchkiss, J.L. (1991), “The Definition of Part-Time
Employment: A Switching Regression Model with
Unknown Sample Selection, ” International
Economic Review, 32(4), pp 899-917.
[12] Lazear, E.P. (1998),“ Personnel Economics for
Managers,” New York: John Wiley & Sons, Inc.
[13] Lee, L.F. (1978), “Unionism and Wage Rates: A
Simultaneous Equations Model with Qualitative
and Limited Dependent Variables,”International
Economic Reviews, 19(2), pp 415-433.
[14] Loewenstein, M.A. and Spletzer, J.R. (1997),“Delayed
Formal On-the-Job Training,” Industrial and
Labor Relations Review, 51(1), pp 82-99.
[15] Lokshin, M. and Sajaia, Z. (2004), “Maximum
Likelihood Estimation of Endogenous Switching
Regression Models,” The Stata Journal, 4(3),
pp 282-289.
[16] Maddala, G.S. (1983), “Limited-Dependent and
Qualitative in Econometrics,” Cambridge
University Press.
[17] Mincer, J. (1962), “On-The-Job Training: Cost,
Returns, and Some Implications,” Journal of
Political Economy, 70(5), pp 50-79.
[18] Mincer, J. (1973), “Schooling, Experience, and
Earnings,” NBER working paper, pp 1-144.
[19] Mincer, J. and Ofek, H. (1982) , “Interrupted Work
Careers: Depreciation and Restoration of Human
Capital,” The Journal of Human Resources,
17(1), pp3-24
[20] Rosen, H. S. (1982), “Taxation and On-The-Job
Training Decisions,” The Review of Economics
and Statistics, 64(3), pp 442-449.
[21] Owen, J.D. (1978), “Why Part-Time Workers Tend to be
in Low-Wage Jobs,” Monthly Labor Review 101,
pp 11-14.
[22] Van der Gaag, J. and Vijverberg, W. (1988), “A
Switching Regression Model for Wage
Determinants in the Public and Private Sectors
of a Developing Country,” The Review of
Economics and Statistics, 70(2), pp 244-252.
[23] Willis, R.J. (1986), “Wage Determinants: A Survey
and Reinterpretation of Human Capital Earnings
Functions,” Handbook of Labor Economics,
pp 525-602.