GC : MOST CENTRAL SKILLS AND KNOWLEDGE IN ENGINEERING JOBS: A NETWORK ANALYSIS OF THE O*NET DATABASE

Transcription

1 GC : MOST CENTRAL SKILLS AND KNOWLEDGE IN ENGINEERING JOBS: A NETWORK ANALYSIS OF THE O*NET DATABASE Panagiotis Asteris, School of Pedagogical & Technological Education (ASPETE) Panagiotis G. Asteris, Ph.D, P.E., is Assistant Professor at Department of Civil & Construction Engineering and Director of the Computational Mechanics Laboratory of the School of Pedagogical & Technological Education (ASPAITE), Athens, Greece. He received the B.S., M.S., and Ph.D. degrees in Civil Engineering from the National Technical University of Athens (NTUA), and joined the Institute of Structural Analysis & Aseismic Research, NTUA, as a research associate. Prof. Asteris is member of the American Society of Civil Engineers (ASCE), American Society of Engineering Education (ASEE) and New York Academy of Sciences (NYAS). Address: School of Pedagogical & Technological Education (ASPETE), N. Heraklion GR 14121, Athens, Greece; Tel.: (+30) , Fax: (+30) , E mail: asteris@aspete.gr. Christos Athanasopoulos, University of Crete Dr. Christos Athanasopoulos is currently Visiting Assistant Professor with the Applied Mathematics Department of the University of Crete in Greece. He received his Ph.D. in experimental nuclear Physics from Temple University, conducting research at the Los Alamos National Laboratory in the USA. He teaches Physics, Mathematics, and Entrepreneurship courses since His research interests include the application of mathematical methods for the analysis of finance and climate data as well as network analysis in complex networks. He has several publications in scientific journals and in conference proceedings and has participated in European and national projects in Education and Information and Communication Technology. Address: Department of Applied Mathematics, University of Crete, Heraklion, Crete, Greece; Tel.: (+30) , Fax: (+30) , E mail: athanaso@csd.uoc.gr. George Neototistos, University of Crete George Neofotistos, Ph.D. Research Assistant Professor with the Physics Department of the University of Crete, Greece. He obtained his Ph.D. in Physics from Temple University (Philadelphia PA, USA). He then joined the Electrical Engineering Dept. of Purdue University (West Lafayette, IN, USA) as a researcher, the Physics Dept. of Temple University as a visiting Assistant Professor, and the Physics and Space Sciences Dept. of the Florida Institute of Technology (Melbourne, FL, USA) as an Assistant Professor. Address: Physics Department, University of Crete, Heraklion, Crete, Greece; Tel.: (+30) , Fax: (+30) , E mail: neofotistos@physics.uoc.gr. American Society for Engineering Education, 2009

2 Most Central Skills and Knowledge in Engineering Jobs: A Network Analysis of the O*NET Database Abstract In this paper, we identify the most central skills and knowledge in the engineering jobs by analyzing the characteristics of the weighted networks of the 35 skills and 33 knowledge categories for the 39 jobs comprising the Architecture and Engineering job family of the O*NET (Occupational Information Network) database, USA's primary source of occupational information. We apply methods of network analysis, starting with the identification of the Maximum Spanning Tree (MST), which connects all nodes with the maximum total weight and dominates transport in weighted networks. The MST is obtained from the weighted network using Prim s algorithm and can be partitioned into two distinct components, namely, the superhighways, which comprise the infinite incipient percolation cluster, the nodes of which exhibit high centrality values, and the roads, which contain the remaining, low centrality nodes. Recent advances in network theory suggest that transport in networks can be enhanced significantly by improving the superhighways. Thus, the strengthening of the central skills and knowledge can lead to better transport of skills and knowledge to the engineering jobs in the contemporary world. Introduction There is a widespread belief that workers' skills and education are not adequate for the demands of jobs in the current economy 1. Journalistic reports, employer surveys, popular and policy debates on school quality and education reform, sociological writings on the economy, and economic accounts of the recent growth of wage inequality all suggest a mismatch between the skills workers possess and what jobs require, what economists call an imbalance between the supply of and demand for human capital. Many believe that the problems will become even more serious because the pace of change is accelerating and the workplace is becoming increasingly high tech, service-oriented, and reorganized to involve greater employee participation in the workplace 2. For the engineering domain, the Special Report on the Research Agenda for the New Discipline of Engineering Education 3 states that: rapid changes in the worldwide engineering enterprise are creating a compelling rationale for us to rethink how we should educate future generations of engineers 4-7. According to The Engineer 8 of 2020, tomorrow s graduate will need to collaboratively contribute expertise across multiple perspectives in an emerging global economy that is fueled by rapid innovation and marked by an astonishing pace of technological breakthroughs. Deteriorating urban infrastructures, environmental degradation, and the need to provide housing, food, water, and health care for eight billion people will challenge the analytical skills and creativity of engineers. The Special Report puts emphasis on the need for conducting research in engineering education, in order to become the engine that drives change to improve the technical fluency of students and teachers, increase interest in engineering and awareness of the social impact

3 of the engineering profession, increase diversity in the engineering student body, and increase the U.S. contribution to the global engineering workforce. In this paper, we study the jobs-skills and the jobs-knowledge bipartite weighted networks, which consist of the 39 jobs, 35 skills and the 33 knowledge categories comprising the Architecture and Engineering job family of the O*NET (Occupational Information Network) database, USA's primary source of occupational information 9. In particular, we identify the most central skills and knowledge by applying methods of network analysis, starting with the identification of the Maximum Spanning Tree (MST) in each network. The MST connects all nodes with the maximum total weight and dominates transport in weighted networks. The MST is obtained from the weighted network using Prim s algorithm and can be partitioned into two distinct components, having significantly different transport properties. One component, the superhighways, is the Infinite Incipient percolation Cluster (IIC), for which nodes with high centrality 10. The other component, that is roads, comprises the remaining, low centrality nodes. Recent advances in network theory suggest that transport in networks can be enhanced significantly by improving the superhighways 10. Thus, the strengthening of the central skills and knowledge can lead to better transport of skills and knowledge to the engineering jobs in the contemporary world. Our work contributes to the areas of Engineering Epistemologies (that is, research on what constitutes engineering thinking and knowledge within social contexts now and into the future) and Engineering Learning Mechanisms (that is, research on engineering learners developing knowledge and competencies in context) 3. In particular, our research utilizes solid occupational datasets in order to provide answers to significant questions such as a) what knowledge and skills characterize engineering as a unique field?, and b) how do elements such as critical thinking, mathematics, science, engineering sciences, design, analysis, judgment, and communication relate to each other to characterize the core of engineering as a profession? The paper is organized as follows. Section 2 presents the O*NET database and the particular dataset we have used. Section 3 presents the methodological approach. In Section 4, results are presented. Finally, Section 5 ends the paper with discussion and conclusions. The O*NET Database The Occupational Information Network (O*NET) database contains information on standardized and occupation-specific descriptors, and is continually updated by surveying a broad range of workers from each occupation. Based on the Standard Occupational Classification (SOC), the O*NET-SOC taxonomy includes 812 occupations which currently have, or are scheduled to have, data collected from job incumbents or occupation experts. The most recent O*NET-SOC 2006 taxonomy includes 949 occupational titles, 812 of which represent data-level occupations. The O*NET Program is collecting and disseminating updated data for the 812 data-level occupations. Data are gathered on approximately 200 occupations each year, with the goal of replenishing the database every five years.

4 The O*NET jobs-knowledge network is a weighted bipartite network. A bipartite network has two kinds of nodes, say, J (denoted as such for jobs) and K (denoted as such for knowledge), in which there are only links between two nodes of different kinds. Table 1 and Table 2 present the weights of the links between skills, knowledge and selected jobs from the O*NET Architecture and Engineering job family. For each one of the 39 jobs in this job family, the architectural and engineering professionals who have been surveyed have graded the 35 skills and the 33 knowledge domains with respect to the requirements of their particular job. Methodological Approach In large complex networks, not all nodes are equivalent Centrality measures address the question, Which is the most important or central node in this network? The simplest of centrality measures is the degree centrality, also called simply degree. The degree of a node in a network is the number of links attached to it. We view the jobs-knowledge and jobsskills networks in terms of their transport properties, that is the ease with which skills and knowledge are linked to the engineering jobs. Links with greater weights can be interpreted as more important ( wider ) roads in connecting the nodes, and nodes with higher centrality are more important in spanning the network. From this perspective, transport in the aforementioned weighted networks is dominated by the Maximum Spanning Tree (MST), that is, the tree connecting all nodes with the maximum total weight. The MST is obtained from the weighted network using Prim s algorithm and can be considered as partitioned into two distinct components 10, having significantly different transport properties. One component, the superhighways, is the Infinite Incipient percolation Cluster (IIC), for which nodes with high centrality dominate 10. For the other component, that is the roads, which includes the remaining nodes, low centrality nodes dominate. The global (network-wide) transport can be enhanced significantly by improving the small fraction of the network, the superhighways 10. To identify the IIC of the network, we start with the fully connected network and remove links in ascending order of their weights. After each removal of a link, we calculate κ <k 2 >/<k>, (where <k> is the average degree that is, the average number of links- and <k 2 > is the average squared degree), which decreases with link removals. When κ < 2, we stop the process because at this point, the largest remaining component is the IIC 13, the knowledge nodes, which comprise the superhighways in the network. Results We start by calculating a very simple measure of ranking the skills and knowledge, namely, the total sum of weights, for all jobs in the O*NET Architecture and Engineering job family. Table 3 and Table 4 present the rankings. As can be seen, the top 10 ranked skills are: Reading Comprehension, Critical Thinking, Active Listening, Complex Problem Solving, Active Learning, Time Management, Writing, Mathematics, Speaking, Judgement and Decision Making. The top 10 ranked knowledge categories are: Engineering and Technology, Mathematics, Design, Computers and Electronics, English Language, Mechanical, Customer and Personal Service, Physics, Administration and Management, Production and Processing.

5 Subsequently, we identify the MST of the jobs-skills and jobs-knowledge networks and calculate the degree centralities of the nodes. Table 5 and Table 6 present the ranking of the skills and knowledge with respect to their degree centralities as nodes of the MST. As can be seen, the top-ranked skills are: Reading Comprehension, Active Listening, Critical Thinking, Management of Material Resources, Active Learning, Complex Problem Solving, Science, Troubleshooting, Technology Design, Writing, Judgment and Decision Making. The topranked knowledge categories are: Engineering and Technology, Design, Computers and Electronics, Mathematics, Building and Construction, Geography, Law and Government, Production and Processing. Finally, we identify the IIC of each network, a process, which, as described before, starts with the fully connected network, removing links in ascending order of their weights. Table 7 presents the ranking of the skills with respect to their degree centralities as nodes of the IIC. As can be seen, the top-ranked skills are: Mathematics, Reading Comprehension, Active Listening, Critical Thinking, Judgment and Decision Making, Science, Writing, Active Learning, Complex Problem Solving, Equipment Selection, Quality Control Analysis, Speaking, Time Management. The top-ranked knowledge categories are: Engineering and Technology, and Design. Discussion Top-ranked skills such as: Mathematics, Reading Comprehension, Active Listening, Critical Thinking, Judgment and Decision Making, Science, Writing, Active Learning, Complex Problem Solving, Equipment Selection, Quality Control Analysis, Speaking, and Time Management, comprise the most central skills in the superhighways leading to engineering jobs. Engineering and Technology, and Design, comprise the top-ranked knowledge leading to engineering jobs. Transporting skills and knowledge to the engineering jobs can be enhanced significantly by improving the superhighways in the networks. Thus, improving the central skills and knowledge constitutes a significant and strategic contribution in enhancing engineering education. Our findings contribute to a better understanding of skills-formation and knowledgeconstruction paradigms in engineering education. Our methodological approach can also be used to systematically monitor the coupling between education systems and the evolution in the worldwide engineering enterprise. Bibliography 1. Skills mismatch in the labor market. Handel M. J. Annual Review of Sociology, 29: (2003). 2. Information technology, workplace organization, and the demand for skilled labor: Firm-level evidence. Bresnahan T. F., Brynjolfsson E, and Hitt L. M. The Quarterly Journal of Economics. 117: (2002). 3. Special Report: The Research Agenda for the New Discipline of Engineering Education. Journal of Engineering Education. October Restructuring Engineering Education: A Focus on Change, The National Science Foundation, 1995, 5. Rising Above the Gathering Storm, Committee on Science, Engineering, and Public Policy, 2006, 6. Educating Engineers for the 21st Century: The Industry View, The Royal Academy of Engineering, 2006, 7. Changing the Culture: Engineering Education into the Future, Review Report, Institution of Engineers, Australia, The Engineer of 2020, National Academy of Engineering, 2004, 9. O*NET (Occupational Information Network).

6 10. Transport in weighted networks: partition into superhighways and roads. Wu Z., Braunstein L. A., Havlin S., and Stanley H. E. Physical Review Letters. 96(14): (2006). 11. Statistical mechanics of complex networks. Barabasi A. L. Reviews of Modern Physics. (74):47-97 (2002). 12. Exploring complex networks. Strogatz S.H. Nature. 410(6825): (2001). 13. Resilience of the Internet to random breakdowns. Cohen R., Erez K., Ben-Avaraham D., and Havlin S. Physical Review Letters 85, 4626 (2000).

7 Table 1: Weights of the links between Knowledge and selected Jobs comprising the O*NET Architecture and Engineering Job Family Electrical Industrial Mechanical Architects Civil Engineers Engineers Engineers JOBS Engineers KNOWLEDGE Administration and Management Biology Building and Construction Chemistry Clerical Communications and Media Computers and Electronics Customer and Personal Service Design Economics and Accounting Education and Training Engineering and Technology English Language Fine Arts Food Production Foreign Language Geography History and Archeology Law and Government Mathematics Mechanical Medicine and Dentistry Personnel and Human Resources Philosophy and Theology Physics Production and Processing Psychology Public Safety and Security Sales and Marketing Sociology and Anthropology Telecommunications Therapy and Counseling Transportation

8 Table 2: Weights of the links between Skills and selected Jobs comprising the O*NET Architecture and Engineering Job Family Electrical Industrial Architects Engineers Engineers SKILLS JOBS Civil Engineers Mechanical Engineers Active Learning Active Listening Complex Problem Solving Coordination Critical Thinking Equipment Maintenance Equipment Selection Installation Instructing Judgment and Decision Making Learning Strategies Management of Financial Resources Management of Material Resources Management of Personnel Resources Mathematics Monitoring Negotiation Operation and Control Operation Monitoring Operations Analysis Persuasion Programming Quality Control Analysis Reading Comprehension Repairing Science Service Orientation Social Perceptiveness Speaking Systems Analysis Systems Evaluation Technology Design Time Management Troubleshooting Writing

9 Table 3: Ranking of Skills according to the descending order of the total sum of weights, for the Jobs-Skills network of the O*NET Architecture and Engineering Job Family. RANKING OF SKILLS TOTAL SUM OF WEIGHTS Reading Comprehension 3014 Critical Thinking 2948 Active Listening 2947 Complex Problem Solving 2762 Active Learning 2726 Time Management 2703 Writing 2675 Mathematics 2663 Speaking 2639 Judgment and Decision Making 2625 Coordination 2545 Monitoring 2339 Troubleshooting 2271 Learning Strategies 2228 Equipment Selection 2189 Instructing 2165 Science 2116 Operations Analysis 2096 Technology Design 2071 Quality Control Analysis 2039 Systems Analysis 2023 Persuasion 1987 Social Perceptiveness 1923 Systems Evaluation 1921 Negotiation 1918 Management of Personnel Resources 1828 Service Orientation 1788 Management of Financial Resources 1574 Management of Material Resources 1567 Operation Monitoring 1560 Operation and Control 1492 Installation 1431 Equipment Maintenance 1322 Repairing 1196 Programming 1085

10 Table 4: Ranking of Knowledge according to the descending order of the total sum of weights, for the Jobs-Knowledge network of the O*NET Architecture and Engineering Job Family. RANKING OF KNOWLEDGE TOTAL SUM OF WEIGHTS Engineering and Technology 3139 Mathematics 2736 Design 2724 Computers and Electronics 2517 English Language 2497 Mechanical 2023 Customer and Personal Service 2001 Physics 1979 Administration and Management 1976 Production and Processing 1888 Public Safety and Security 1728 Building and Construction 1635 Clerical 1577 Education and Training 1574 Law and Government 1537 Chemistry 1403 Communications and Media 1223 Telecommunications 1138 Personnel and Human Resources 1129 Sales and Marketing 1075 Geography 1071 Transportation 1063 Economics and Accounting 1046 Psychology 957 Biology 693 Sociology and Anthropology 453 History and Archeology 443 Foreign Language 354 Medicine and Dentistry 332 Philosophy and Theology 297 Therapy and Counseling 258 Fine Arts 214 Food Production 185

11 Table 5: Ranking of Skills according to the descending order of the total sum of non-zero MST degree centralities, for the Jobs-Skills network of the O*NET Architecture and Engineering Job Family. RANKING OF SKILLS MST Degree Centrality Reading Comprehension 9 Active Listening 7 Critical Thinking 5 Management of Material Resources 4 Active Learning 2 Complex Problem Solving 2 Science 2 Troubleshooting 2 Technology Design 2 Writing 2 Judgment and Decision Making 1 Table 6: Ranking of Knowledge according to the descending order of the total sum of non-zero MST degree centralities, for the Jobs-Knowledge network of the O*NET Architecture and Engineering Job Family. RANKING OF KNOWLEDGE MST Degree Centrality Engineering and Technology 20 Design 6 Computers and Electronics 5 Mathematics 3 Building and Construction 1 Geography 1 Law and Government 1 Production and Processing 1 Table 7: Ranking of Skills according to the descending order of the total sum of non-zero IIC weights, for the Jobs-Skills network of the O*NET Architecture and Engineering Job Family. RANKING OF SKILLS IIC WEIGHTS Mathematics 3 Reading Comprehension 3 Active Listening 2 Critical Thinking 2 Judgment and Decision Making 2 Science 2 Writing 2 Active Learning 1 Complex Problem Solving 1 Equipment Selection 1 Quality Control Analysis 1 Speaking 1 Time Management 1