Medical employees' core competency: Application of two-stage clustering

Lo, Kuei-Hsing; Lin, Wen-Tsann; Jou, Yung-Tsan; Lin, Shu-Wei Article Medical employees' core competency: Application of twostage clustering International Journal of Management, Economics and Social Sciences (IJMESS) Provided in Cooperation with: International Journal of Management, Economics and Social Sciences (IJMESS) Suggested Citation: Lo, Kuei-Hsing; Lin, Wen-Tsann; Jou, Yung-Tsan; Lin, Shu-Wei (2018) : Medical employees' core competency: Application of two-stage clustering, International Journal of Management, Economics and Social Sciences (IJMESS), ISSN 2304-1366, IJMESS International Publishers, Jersey City, NJ, Vol. 7, Iss. 2, pp. 175-185 This Version is available at: https://hdl.handle.net/10419/180784 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by-nc/3.0/ 175 International Journal of Management, Economics and Social Sciences 2018, Vol. 7(2), pp. 175 – 185. ISSN 2304 – 1366 http://www.ijmess.com Medical Employees’ Core Competency: Application of Two-Stage Clustering Kuei-Hsing Lo1 Wen-Tsann Lin2 Yung-Tsan Jou1 *Shu-Wei Lin3 1 Dept. of Industrial and Systems Engineering, Chung Yuan Christian University, Taiwan 2 Dept. of Industrial Engineering and Management, National Chin-Yi University of Technology, Taiwan 3 The University of Melbourne, Australia In 2005, the Taiwanese government has drawn up The Talent Quality-management System (TTQS) for businesses and medical units to carry out internal human training by offering a sound education and training evaluating system. Hospital executives are more aware of the need to improve the competitiveness of clinical staff and realize that a patient-centered healthcare model requires continuous innovation and quality improvement to enhance the competitiveness of hospitals. Using two-stage clustering (Self-Organization Maps and K-means), this research identified different types of departments–including physicians, medical staff, and administrators and discussed the respective core competency of the different groups. The study result showed the importance of conveying the core competency to the various departments and units in hospitals. Based on analyses we found that hospital staff’s core competency identification is essential to help construct a human resources system, serving as a reference for hospitals and corporations in the selection, education, utilization, and retention of talents, as well as a practical benchmark for core competency applicable in the human resource management of hospitals and corporations. Keywords: Core competency, K-means, self-organization maps (SOM), Two-stage clustering, The Talent Qualitymanagement System (TTQS) JEL: I11, L15 In the last two decades, the global competition has been fierce in medical industry. Among other factors, the core competency of enterprises also lies in the quality of qualified personnel, which shows the strategic importance of human resource (Zhang, 2003). Facing the economic recession, enterprises should think optimistically on how to cope with the global competition and to be more competitive to intelligently respond to the economic fluctuations which could reduce many labor-related expenses and enterprise costs. It is a crucial issue for enterprises to find out how to make human resource grow and to create an emerging future. Human capital is one of the most important factors of production and the key to win the competition. Enterprises need to invest in the human resource, plan and train their employees, and establish a sound Manuscript received January 15, 2018; revised April 15, 2018; accepted May 2 , 2018. © The Author(s); Licensee IJMESS *Corresponding author: [email protected] International Journal of Management, Economics and Social Sciences 176 education and training system to foster the organization’ s required manpower. The Taiwanese government has, based on the concept of the UK’ s Investors in People (IIP), and the Swiss ISO10015, introduced The Talent Quality-management System (TTQS) which is a training process including Plan, Design, Do, Review, and Outcome. TTQS provides the training institution and enterprise with the system to evaluate the staff performance and process education and training that carry out from the internal training (Lin, et al ., 2010). In the process of Design, core competency is used as the assessment index. By extracting information from databases, this study identified the following core competencies: communication and coordination, innovation and development, problem-solving, learning and development, and accepting responsibility. The purpose of this study is as follows: 1. The results of the two-stage clustering analysis on how to effectively introduce TTQS is provided. By this research to enable those institutions that have implemented or are going to introduce the TTQS to effectively make their own plans and assessment indicators, thereby enhancing the overall human resource. 2. Evaluate the relationship between TTQS and core competency. By reviewing the implementation of the assessment mechanism, it shows that whether all the assessed hospitals have met the standards of each index. Within the further understanding on the TTQS assessment indicators in identification, the assessed institutions are able to observe the degree of correctness of using the core competency. 3. This study aims to bring the positive impact on Taiwan's medical centers, by improving employees’ core competency. Core competency is a general term for behavior, motivation, and knowledge related to work success (Byham and Moyer, 1996). Healthcare behavior involves high complexity and uncertainty. When medical employees perform healthcare behavior-based work, they must possess highly professional medical knowledge and skills to ensure the provision of safe patient care services. The importance of employees’ core competency allows the country to compete with the international community and continuously provide the research results for future implementation. LITERATURE REVIEW Talent Quality-Management System (TTQS) Lo et al. 177 The Talent Quality-Management System for Talent Development was specially introduced by the International Organization for Standardization (ISO) in “ISO10015 Quality Management - Training Guide” and the British Investors in People (IIP) promulgated in December 1999. Workforce Development Agency (WDA), Ministry of Labor, Taiwan planned to develop the Talent Quality-Human Resource Management System based on the five major facets i.e. Plan, Design, Do, Review and Outcome (PDDRO) to ensure the reliability and correctness of the training process and construct good systematic training environment for human capital investments (Lin et al ., 2010). Started in 2005, the TTQS specifications began to take shape with the efforts of many experts, aiming at the training of reviewers, developing grading criteria with TTQS score cards and completing the pretest of ten organizations. Through constant modifications, the government aims to make TTQS to ensure the effects of Taiwanese enterprises’ investment in human resource in a more complete manner. The systematic development of this system can enhance its application value and implementation quality to continuously review and amend the standards for training quality assessment. With the introduction of the TTQS system and following the PDDRO standard (see Figure 1), it evaluates the results of employees’ and enterprise’s performance after the training program and integrate the results into the systematic plan that makes the education and training be more aligned with the business demands (Lin et al ., 2011). In addition, core competency is a crucial category of TTQS index. Source: Corporation Training Network, Taiwan Figure 1. TTQS Cycle International Journal of Management, Economics and Social Sciences 178 Core Competency Spencer and Spencer (1993) contended that competency represents the underlying characteristics of a person, which not only relate to the person’s work position, but also facilitate understanding the person’s expected or real reactions and expressions influencing behavior and performance. The core competency varies according to the organizational strategy and culture and the circumstances, such as customer orientation, innovation, integrity, and so on. Vazirani (2010) refers it as capability to compete with the market for unique intellectual processes or product capabilities that refer to the collective learning or performance ability of the entire organization. Chen et al . (2012) further point out that enterprises should give priority to their core competency (such as traits and motivations of employees) based on the selection of appropriate candidates for their functions because the core competency is not easy to modify and develop. Two-Stage Clustering Cluster analysis is the procedure of objective classification based on the similarity and difference. The purpose of the classification is to facilitate the identification of the similarity between certain research subjects and group them in the same cluster on the basis of similar features. Subjects stay in the same cluster shows the high level of homogeneity. Sharma (1996) recommends the categorization method like stratification and non-stratification. In the first stage of stratification, self-organization Map (SOM) method was implemented for clustering to decide the number of clusters. Then, those clusters were substituted in K-means in the stage two. The main reason of adopting the two-stage clustering is the drawback that once two subjects are clustered together in the 1st stage, they will always stay in the same cluster. K-means offset the drawback and reach the best number of clusters being homogenous in the clusters and heterogeneous clusters. While using the K-means, users have to keep trying locating the most appropriate number of clusters. As a result, using two-stage clustering reduces the cost of calculation and saves the time. -Self-Organizing Map (SOM) SOM is a type of unsupervised learning network proposed by Kohonen (1990). It adopts the idea that brains have the feature of birds of a feather flock together . While, network learning is completed, adjacent output Lo et al. 179 processing units have similar functions for clustering. As SOM makes clustering in the concept of nearest neighbor, the resultant clusters are seen with overlapping partitions. The resultant clusters of non-level-type clustering can produce non-overlapping partitions. Vesanto and Alhoniemi (2000) identify self-organizing map patterns by matching the vector values of the pattern elements and classifying the patterns. - K-Means K-means cluster algorithm was proposed by J. B. MacQueen in 1967, which is used to deal with the problem of data clustering. The relatively simple algorithm is widely used in the scientific field research and industrial applications. K-means method is frequently used in non-stratified cluster analysis (Buttrey and Karo, 2002). It requires predetermined number of clusters. Inappropriate number of clustering will lead to vague difference among clusters. In selecting clusters, it is recommended selecting different number of clusters for more algorithms to have reasonable explanation. Both Abidi and Ong (2000) and, Vesanto and Alhoniemi (2000) proposed the technique of two-stage clustering as the clustering strategy for conducting data mining. This study adopts a two-stage clustering method (SOM and K-means) to carry out scientific and confirmatory comparison among groups of physicians, nurses and general administrative staff. METHODOLOGY This study focused on the relevance and correctness of one Taiwan’s case hospital in promoting core competency through the TTQS-accredited medical institutes. This research urges other medical research institutes to import TTQS research index and advises the medical centers to create a database which can be used for academic research purposes. It is hoped that the database of the medical hospital can be helpful to get the optimal results, generate feedback and reference. In this study, the status of training and operation of human resource in medical institutions was mainly learned from the database of medical institutions. The research was divided into two parts. First, a discriminant analysis was conducted for each cluster. The medical center introduced core competency to physicians, nurses and general administrations in order to explore the various groups of surgical, medical and administrative support during the review in the hope of finding out the core competency that were most International Journal of Management, Economics and Social Sciences 180 distinguishable among different levels of groups. By using SOM and K-means separately identifies the doctors, nurses and general administrative units of each department in the medical institution’s database. In this study, we hope to see whether the core competency of the TTQS have been effectively internalized and implemented from the perspective of the department that has been imported into TTQS, so as to translate the content of the database into information with reference value. Two-Stage Clustering Algorithm The algorithm of two-stage clustering is as follows: 1. SOM: Kohonen used self-organization to conduct graphic recognition, where vector values in consistence with the graphic elements are used to classify the graphs. 2. K-means: In terms of economy, simplicity and effectiveness, K-means is a method worth applying. It is a non-layer type clustering, not subject to the effects of outliers, errors in distance measurement and selection of method of distance calculation. The results of cluster are better if the initial point of the cluster is known. The K-means algorithm criterion function adopts square error criterion be defined as Equation 1: 2 (1) Where, E is total square error of all the objects in the data cluster, p bellows to data object set, mi is mean value of cluster Ci. Comparison of Two Groups by Grouping Method This study used two-stage clustering to conduct a confirmatory comparison. In order to determine the target of this study to explore the “promotion of the core competency of the medical center” and to understand the distribution of the medical library, this study explored and analyzed the existing field data to divide the functional units into several clusters, obtained similarities from the same cluster, and through the results of the analysis, via the unsupervised self-organizing map (SOM) network. In the drawing for the distribution of SOM clusters, being the numbers of initial clusters in the first stage, the numbers were eight, as shown in Figure 2, which shows the distributions of group numbers of SOM grouping. Using the two-stage clustering analysis results of K-means, the study obtained the distribution of each cluster composition, as shown in Figure 3. Lo et al. 181 Source: This study Figure 2. SOM Grouping Scatter Diagram Source: This study Figure 3. K-Means Cluster Composition Distribution From “K-Means Distribution” in Figure 3, the blue portion of bar represents the Department of Surgery (D1), the red portion of bar represents the Department of Internal Medicine (D2), and the purple portion of bar represents the general administration (D3). The size of the cluster became more relevant to each other. In Figure 4, the size of the cluster became more explicit as it relates to each analysis. The distribution is as follows: Each cluster shows the distribution proportion and it can be clearly observed. In Cluster-1, it includes the “ doctors” , “ nurses” and “ administration staff” in medical department of hospital, which accounts for 10.48 percent. In Cluster-2, it involves “ doctors” , “ nurses” and “ administration staff” in surgical International Journal of Management, Economics and Social Sciences 182 department of hospital, which makes up 85.3 percent. In Cluster-3, it includes “ doctors” , “ nurses” and “ administration staff ” in general Administration, which occupies 4.22 percent. Source: This study Figure 4. K-Means Cluster Composition Distribution Then, the initial population of SOM was substituted into K-means. The K-means was used for the secondstage grouping of nodes on the map and is plotted (see Figure 5) as a two-stage clustering model of SOM-K- means cluster scatter. Source: This study Figure 5. Distribution of SOM-K-Means Clusters As can be seen from Figure 5, the distribution of SOM-K-means scatter plots showed that the degree of consistency and the extent to which core competency is promoted. The analysis results are as follows: