Work Design Characteristics and Knowledge Sharing Behavior among Software Engineers 1
Mobashar Rehman1, Ahmad Kamil Mahmood2, Rohani Salleh3, Aamir Amin4 Faculty of Information & Communication Technology, Universiti Tunku Abdul Rahman, Kampar, Perak, 31900, Malaysia 2,4 Department of Computer and Information Sciences 3 Department of Management and Humanities 2,3,4 Universiti Teknologi PETRONAS, Bandar Seri Iskandar, 31750 Tronoh, Perak, Malaysia 1
[email protected]
Abstract - Knowledge sharing is important for Software Engineers because software is purely based on individual’s knowledge and one of the sources to learn new knowledge is through sharing. Although Software Engineering is a growing field, however research done in this profession is still not mature. One of the areas, which have not been thoroughly investigated in this profession, is working environment and its impact on knowledge sharing. Focus on this aspect of Software Engineering is important because work environment not only influences the performance of an individual but also the behavior (which in this case will be knowledge sharing behavior). Therefore, huge number of studies has been done conducted on designing work environment. One of the important works done was Job Characteristics Model (mostly used in Software Engineering field). However, it covers only five aspects of work/job. Therefore different other studies focused on more work design characteristics. This study analyzed the relationship between knowledge sharing behavior and work design characteristics among Software Engineers. Work design characteristics proposed by Morgeson and Humphrey (2006) were used for this study. This study was conducted in Malaysia and four locations were selected through geographical random cluster sampling. Results indicated that task identity, feedback from job, skill variety and received interdependence are the main work design characteristics for Software Engineers. Results also indicated that all work design characteristics has positive relationship with knowledge sharing behavior. Keywords - Work Design Characteristics, Knowledge Sharing Behavior, Software Engineers
I.
INTRODUCTION
Impact of designing work is huge on an individual’s wellbeing and organizational success [1] in any profession. Due to its importance, this area is under the spotlight from decades. One of the earlier works of how job can be designed is traced back to Adam Smith’s work where idea of breaking complex jobs into simpler jobs emerged [2]. 978-1-4799-0059-6/13/$31.00 ©2014 IEEE
Since then many researchers have proposed various methods to design a job but one of them is very popular even today among researchers and practitioners. i.e., Job Characteristics Model (JCM) proposed in [3]. Even in the field of Software Engineering, researchers have heavily used JCM. This model consists of five job characteristics namely skill variety, task identity, task significance, autonomy and feedback. Unfortunately, problem with JCM is that it covers only five job characteristics whereas there are numerous other characteristics of a job which needs to be considered. Therefore, research on job characteristics continued and one of the latest works in this field was carried out in [4]. Job/work environment has genuine and long-standing effects on individuals in every profession including Software Engineering. These effects include changes in behavior, attitude, well being and perception related outcomes. Different studies have explored the relationship between work design and work related outcomes like performance, turnover, satisfaction and stress. There is another behavioral outcome which is very crucial for the success of Software Engineers and Software Development organizations in this knowledge era but unfortunately less work is done on this behavioral outcome and that area is Knowledge Sharing Behavior (KSB). Knowledge sharing has become one of the criteria for measuring the performance of an individual. This study will fill this gap in the field of Software Engineering by analyzing the relationship between KSB and Work Design Characteristics (WDCs). WDCs used in this paper are taken from [4]. Reason for selecting these WDCs is that although job designing is very old phenomenon and research is ongoing in this field from decades but unfortunately no researcher has produced so many different aspects of work design as in [4]. Till date, JCM has been widely used as job design criteria in Software Engineering and other fields however it accommodates only five dimensions of job. To overcome this problem, WDCs were introduced and provided three main domains of work and 18 sub
dimensions thus covering far more aspects of work than JCM. Another important reason to select WDCs for this Engineering. Some of the dimensions like job complexity, information processing, problem solving, specialization, feedback from job, feedback from others and interdependence are very relevant to Software Engineering field. Therefore, these WDCs are suitable for study in the field of Software Engineering. II. LITERATURE REVIEW A. Knowledge Sharing Behavior World economy has moved from industrial to knowledgebased economy thus emphasizing on knowledge. Therefore, no organization can survive without knowledge and to make any organization successful, knowledge sharing is crucial [5]. In case of Software Engineering, knowledge sharing is even more vital because software itself is based on the knowledge and is a purely knowledge oriented output. Knowledge sharing occurs when an individual (Software Engineer in our case) shares his/her knowledge to other members of the organization [6]. This sharing of knowledge depends on the willingness of Software Engineer and the level of willingness can increase or decrease depending on various factors. One of such factors is designing of job as it can lead to job satisfaction or dissatisfaction, job performance and ultimately affecting KSB. Therefore, work design has an important role to play in increasing or decreasing KSB among Software Engineers. KSB can be categorized into knowledge donation and collection [7]. Based on this, it can be further categorized into Explicit Knowledge Donation Behavior (EKDB), Explicit Knowledge Collection Behavior (EKCB), Implicit Knowledge Donation Behavior (IKDB) and Implicit Knowledge Collection Behavior (IKCB). In this paper, relationship between all these four components of KSB and WDCs are investigated. B. Work Design Characteristics WDCs can be sub-categorized into three areas namely motivational, social and contextual characteristics [4]. Motivational characteristics are further divided into task and knowledge characteristics. Task characteristics include autonomy, task variety, task significance, task identity and feedback from job. Knowledge characteristics include job complexity, information processing, problem solving, skill variety and specialization [4].
study was the relevance of WDCs with Software Social characteristics include social support, interdependence, interaction outside the organization and feedback from others [4]. Contextual characteristics include ergonomics, physical demands, work conditions and equipment use [4]. Equipment used was tailored to Software and Tools used in order to make these dimensions more related to Software Engineering. III. RESEARCH METHODOLOGY This research was conducted using online and postal questionnaires. Instrument used for measuring WDCs was taken from [4]. However, some of the dimensions were changed according to the requirements of this study (to make the questionnaire more specific to Software Engineers). KSB was measured through EKDB, EKCB, IKDB and IKCB. Twenty-eight items were developed by the author of this study to measure KSB. Data was collected from Malaysian Software Developers (referred as Software Engineers in this study). Respondents were selected in two steps because Malaysian Software Engineers are located in various geographical areas across Malaysia. So during first step, four geographical areas were selected (Perak, Penang, Pahang and Kuala Lumpur) through simple random cluster sampling and then simple random sampling was used to identify individual Software Engineers. In total, 384 responses were obtained and the details of response rate are presented in table I.
TABLE I: Response Rate Details Online
Postal
Total
1015
222
1237
7
3
10
Total Questionnaires to Calculate Response Rate
1008
219
1227
Valid Responses
336
48
384
Refusal
30
15
45
Incomplete
0
27
27
Could not Contact/Sample Loss
34
11
45
Total Questionnaire Send Out of Scope
Majority of the respondents were male (81.7%), with Malay ethnicity (51.8%), having bachelor’s degree as their highest
qualification (82.6%), working in Penang (42%) and 2-5 years of experience (48.4%). Most of the companies in which those Software Engineers were employed were using agile methodologies (57.6%). Table II presents the details of demographic information.
TABLE II: Demographic Information Demographic Information Gender
Percentage
Male
81.7
Female
18.3 Employment Status
Permanent
8.6
Contract
91.4 Ethnicity
Malay
51.8
Chinese
29.1
Indian
19.1
Others
Highest Education
Diploma
V. RESULTS A. WDCs of Malaysian SEs Following results were obtained when Malaysian Software Engineers were asked to rate various WDCs on likert scale (1=strongly disagree, 2=disagree, 3=neither agree nor disagree, 4=agree and 5=strongly agree). SEs believed that highest WDCs are task identity (mean = 3.88, SD = 1.02), feedback from job (mean = 3.88, SD = 1.04), skill variety (mean = 3.88, SD = 1.05), received interdependence (mean = 3.88, SD = 1.04), software and tools used (mean = 3.88, SD = 1.06). Some other important WDCs are specialization (mean = 3.87, SD = 1.05), information processing (mean = 3.85, SD = 1.06), task variety (mean = 3.83, SD = 1.03), problem solving (mean = 3.82, SD = 1.07), feedback from others (mean = 3.82, SD = 1.06) and work method autonomy (mean = 3.80, SD = 1.06). Table III provides the summary of WDCs based on the feedback provided by respondents.
TABLE III: Work Characteristics of Malaysian SEs Descriptive Statistics N
Mean
SD
Task Identity
384
3.88
1.02
Factor Name
-
Professional Certification
3.3
Feedback from J
384
3.88
1.04
First/Bachelor Degree
82.6
Skill Variety
384
3.88
1.05
Master's Degree
14.1
Received Interde
384
3.88
1.04
-
Software & Tool
384
3.88
1.06
Specialization
384
3.87
1.05
Others Work Schedule Regular
93.5
Information Proc
384
3.85
1.06
Flexible
6.5
Task Variety
384
3.83
1.03
Problem Solving
384
3.82
1.07
Feedback from O
384
3.82
1.06
Work Method A
384
3.8
1.06
Task Significanc
384
3.79
1.09
Initiated Interdep
384
3.79
1.06
Working Condit
384
3.79
1.06
Decision Makin
384
3.78
1.06
Social Support
384
3.78
1.06
Work Schedule A
384
3.77
1.09
Job Complexity
384
2.36
1.1
Physical Deman
384
2.29
0.95
Current Work Location Perak
21.1
Penang
42
Kuala Lumpur
31.7
Pahang
5.2 Experience
Less than 2 years
24
2-5
48.4
6-9
18.7
10 or above
8.9 Methodology
Agile
57.6
Others
42.4
B. Relationship between WDCs and KSB To analyze the relationship between work design characteristics and KSB, multiple regression method was used. Results showed that all the components of WDCs
(motivational, social and contextual) showed positive relationship with KSB (EKDB, EKCB, IKDB and IKCB). Summary of results between work design characteristics and KSB is as follows: Task characteristics Æ EKDB (R-square = 0.707, Adjusted R-square = 0.701), Task characteristics Æ EKCB (R-square = 0.702, Adjusted R-square = 0.696), Task characteristics Æ IKDB (R-square = 0.702, Adjusted Rsquare = 0.696), Task characteristics Æ IKCB (R-square = 0.705, Adjusted R-square = 0.699). Knowledge characteristics Æ EKDB (R-square = 0.724, Adjusted R-square = 0.721), Knowledge characteristics Æ EKCB (R-square = 0.724, Adjusted R-square = 0.721), Knowledge characteristics Æ IKDB (R-square = 0.714, Adjusted R-square = 0.710), Knowledge characteristics Æ IKCB (R-square = 0.733, Adjusted R-square = 0.726). Social characteristics Æ EKDB (R-square = 0.725, Adjusted R-square = 0.722), Social characteristics Æ EKCB (Rsquare = 0.711, Adjusted R-square = 0.708), Social characteristics Æ IKDB (R-square = 0.713, Adjusted Rsquare = 0.710), Social characteristics Æ IKCB (R-square = 0.713, Adjusted R-square = 0.710). Contextual characteristics Æ EKDB (R-square = 0.699, Adjusted R-square = 0.696), Contextual characteristics Æ EKCB (R-square = 0.694, Adjusted R-square = 0.692), Contextual characteristics Æ IKDB (R-square = 0.704, Adjusted R-square = 0.701), Contextual characteristics Æ IKCB (R-square = 0.701, Adjusted R-square = 0.698). V. DISCUSSION Based on the Likert scale classification as in [8], (low = 1.02.33, medium = 2.34-3.67, high = 3.68-5), Software Engineers scored high on all of the WDCs except job complexity and physical demands where they scored low. Following can be the possible answers to these results. Task identity is about the degree to which one person completes a job in whole. Since each Software Engineer (architect, designer, developer, tester etc.) is assigned with his/her own task based on level of expertise so task identity is high for them because they have to finish their part of the job. Developers will be responsible for coding, testers will be responsible for testing and so on. Software Engineers scored high on feedback from job because they get immediate feedback from their own job once they complete the task. Job itself provides feedback to Software Engineers in a way that at the end of task if it is fulfilling the requirements of the client, job will be
completed otherwise Software Engineers have to make changes accordingly. Software Engineers also scored high on skill variety (because in this constantly changing era they have to update themselves with multiple tools and methodologies). Received interdependence is also high because developing software requires involvement from multiple specialists like software architect, designer, coder, tester, so job interdependence is high in Software Engineering profession. Some other important WDCs in which Software Engineers scored high includes specialization (software development requires multiple professionals with expert skills in their field so specialization is high in this profession). Information processing is also high because Software Engineers has to process lots of information from the client and their colleagues as well in order to develop software. Problem solving characteristic is also high because each software development project brings its own challenges and requirements. Task, knowledge, social and contextual characteristics increases the motivation and performance of Software Engineers on job. Thus higher performance leads to higher KSB as knowledge sharing is performance related outcome [9]. Also evident from the results is that task characteristics have positive relations with all components of KSB (EKDB, EKCB, IKDB and IKCB). Similarly, knowledge, social and contextual characteristics also have significant relationships with EKDB, EKCB, IKDB and IKCB because all these characteristics increases the performance of Software Engineers so increase in performance means increase in KSB as both are related. VI. CONCLUSION & FUTURE DIRECTION This paper analyzed the work design characteristics of Malaysian Software Engineers. In addition, relationship between work design characteristics and KSB was also analyzed. Based on the findings, task identity, feedback from job, skill variety, received interdependence, software and tools used, specialization, information processing, task variety, problem solving, feedback from others and work method autonomy are the main WDCs in the field of Software Engineering. Work design characteristics also showed positive relationship with EKDB, EKCB, IKDB and IKCB. In future, other studies should also be conducted across different countries in order to generalize the work design characteristics of Software Engineers and to see their relationship with KSB.
VII. PRACTICAL IMPLICATIONS Software Engineering industry can benefit from the findings of this study. This research will help top management to provide a better working environment in the light of results obtained in this study. This will not only lead to motivated Software Engineers but will ultimately increase KSB among them. Increase in KSB will result in more learning opportunities and ultimately performance of Software Engineers and Software Development organizations will improve both financially and non-financially.
REFERENCES [1] F.P. Morgeson, & M.A. Campion. “Work design.” in W. C. Borman, D. R. Ilgen, & R. J. Klimoski (Eds.), Handbook of psychology: Industrial and organizational psychology, 12, pp. 423-452, 2003 [2] B. Jelstad. “Job characteristics and its outcomes. - A comparative work design study of non-profit and profit organizations.” Internet: http://www.polis.no/Paper/Aarhus05/PJELSTAD.PDF, 2005, [May 29, 2013] [3] J. Hackman, & G. Oldham. (1976). “Motivation through the design of work: Test of a theory.” Organizational Behavior and Human Performance, 16, pp. 250-279, 1976. [4] F.P. Morgeson, & S.E. Humphrey. “The Work Design Questionnaire (WDQ): developing and validating a comprehensive measure for assessing job design and the nature of work.” The Journal of Applied Psychology, 91(6), pp. 1321-1339. doi: 10.1037/0021-9010.91.6.1321, 2006. [5] A. Ellahi & R. Mushtaq. “Probing Factors Affecting Knowledge Sharing Behavior of Pakistani Bloggers.” Electronic Journal of Information Systems in Developing Countries, 45(6), pp. 1-14, 2011. [6] S. Ryu, S. Ho, & I. Han. “Knowledge sharing behavior of physicians in hospitals.” Expert Systems with Applications, 25, pp. 113-122. doi: 10.1016/S0957-4174(03)00011-3, 2003. [7] B. van Den Hooff, & J.A. De Ridder. “Knowledge sharing in context: the influence of organizational commitment, communication climate and CMC use on knowledge sharing.” Journal of Knowledge Management, 8(6), pp. 117-130. doi: 10.1108/13673270410567675, 2004. [8] K. Landell. ”Management by Menu.” London: Wiley dan Sons Inc. 1997. [9] L. Rabbiosi, K. Makela, L. Rabbiosi. “Organizational climate and knowledge sharing : an individual- level perspective.” in Proc. DRUID Summer Conference. CBS - Copenhegan Business School, Denmark, June 17-19, 2009.
