個人績效評估與時序變動之動態資料包絡分析: 人力資本效率模型的模式發展與實證分析

A Dynamic Data Envelopment Analysis of Individual Performance Appraisal and Sequential Change: Model Development and an Empirical Study of the Human Capital Efficiency Model

吳國鳯(Kuo-Feng Wu)

指導教授 ： 李誠 ； 邱皓政

國立中央大學/管理學院/人力資源管理研究所/博士(2014年)

摘要

績效評估是企業組織針對員工的工作表現進行考核評估的正式活動，員工工作績效是組織與個人投入諸多資源與努力所得到成果，在人資實務或組織行為研究中對於績效評估的討論，多未納入投入與產出的生產效率函數概念，也甚少處理連續跨期績效表現的變動趨勢分析，忽略勞動成本效益的實證分析。因此，本研究導入資料包絡分析（data envelopment analysis）建立評估員工績效衡量的效率導向數量模型。首先，本研究整理關於資料包絡分析的概念與原理，說明資料包絡分析方法應用於績效評估的可行性；其次，在實證分析部份，本研究採用個案公司所提供的12期員工績效與薪資資料，以人力資本相關要素與工作投入相關要素作為投入項與連結變數；員工各期的績效分數作為產出項，進行無導向性動態差額變數模型與無導向性超效率差額變數模式分析。
本研究的實證樣本為某個案公司製造課68位員工，效率分析的結果辨識出18位效率前緣員工（效率值達1.0），藉由超效率差額分析得以進行效率前緣員工的效率排序。在模式部分，本研究發現月資歷越高效率值越好，平均月資歷在18個月左右的員工效率值表現最佳。採動態差額變數分析得到的效率值較單期差額變數模式效率值來得高，符合實務上因工作經驗累積而提高工作效率的假設。人資部門可根據效率分析結果鑑別員工工作效率，進一步擬定管理策略方案，例如效率佳的員工可進行工作輪調或納入接班人計畫；效率較差的員工可藉由降低加班、減低技術津貼、增加教育訓練、員工輔導等方式提升效率值。
本研究將資料包絡分析之應用從總體層次延伸到個人層次，藉由差額變數對投入、產出項非等比例的調整，應用於個別員工效率評估與診斷，除能提供直線主管客觀數據作為與人力成本與績效評估的參考依據，並具體提出績效評估中成本與效率概念的數量模型與衡量技術，未來建議研究者可將此方法進一步的進行延伸與應用至人力資源管理領域中。
關鍵字: 動態資料包絡分析、動態差額變數模型、超效率差額變數、人力資本、工作投入、績效分數

關鍵字

動態資料包絡分析 ； 動態差額變數模型 ； 超效率差額變數 ； 人力資本 ； 工作投入 ； 績效分數

並列摘要

Performance appraisal is the formal activity of organizations to assess employee’s performance, which achieved by large amount of resources and efforts invested by individuals as well as organizaitons. In human resource practices and empirical studies, however, the concepts of input and output related to individual performance have not been taken into account in performance evaluation procedures yet. Moreover, there are also very few research conducting dynamic trend analyses in performance overtime, as the result of ignoring the cost-benefit aspect and dynamic change in empirical analysis. Therefore, the aim of the research is to build a quantitative model for employee performance evaluation by introducing data development analysis. Firstly, this research presents the concept and principle of data development analysis and explains the feasibility of applying this approach to performance evaluation. Secondly, in model development, the personnel data related to performance and salary information over twelve months period provided by the researched company are used. The factors of human capital and job involvement are considered as inputs and linking variable; the employees’ performance score over twelve-month period is considered as output. This research used non-oriented dynamic slacks-based measure model and non-oriented super slacks-based measure model to test the research purpose.
Sixty-eight samples were used to test the model. They are seniority between two-one hundred and fifty one months, and eighteen of them with efficiency scores equal to 1. By using Super Slacks-Based Model, the researcher can sort the order of efficiency from the employees whose efficiency score equal to 1. The study found the employee seniority longer the efficiency score well. The employees with average Eighteen months seniority have the highest efficiency; Furthermore, efficiency of dynamic slacks-based omeasure model is higher than slacks-based measure model. This supports the hypothesis that work efficiency increases with the accumulation of working experiences. Lastly, the company can further develop human capital solutions based on employee efficiency. The employees with high efficiency can be included in job rotation or succession planning; the employees with lower efficiency can be improved by reduce overtime; reduce skill allowance, increase training and coaching.
The study extends the application of data development analysis from macro level to micro level. By non-radial adjustment of the differences in input variable, output variable, increasing input or reducing performance score improves the overall efficiency. The result not only provides objective data in HR cost and performance evaluation for line managers, but also provides a specific quantitative model and measurement technique for performance evaluation cost and efficiency. It is suggested that the approach can be further extended and applied to the field of human resources management in the future.
Keyword: data envelopment analysis, dynamic slacks-based measure, super slacks-based-measure, human capital, job involvement, performance score

並列關鍵字

dynamic data envelopment analysis ； dynamic slacks-based measure ； super slacks-based-measure ； human capital ； job involvement ； performance score

參考文獻

黃芳玫(2011)。台灣人力資本與薪資不均之研究，台灣經濟預測與政策，42(1)，1-37。

諸承明、蔡美玲(2012)。薪酬設定與員工服務品質關係之研究-以餐飲業第一線服務人員為例，中原企管評論，10(2)，75-98。

歐陽彥慧(2009)。金融從業人員在工作不安定下對工作績效之影響－以工作壓力與工作投入為中介變項，華人經濟研究，7(2)，78-95。

黃美珠、陳玉麟(2009) 。薪酬與生產技術效率的解釋:性別與層級在台灣會計師產業扮演的角色。當代會計，10(2)，163-188。

李俊賢、許惠媚、黃芳銘(2013)。知覺屈就（高才低就、才非所用）與轉職行為之研究，人力資源管理學報13(2)，45-72。