Global Information Systems

Dorothy E Leidner, Tim Kayworth

• Publisher: Routledge
• Year: 2008
• Print ISBN: 9781138435315
• eBook ISBN: 9781136400599
• Topic: Business
• Subtopic: Business General
• Index: Business

Information technology has had a major impact on individuals, organizations and society over the past 50 years. There are few organizations that can afford to ignore IT and few individuals who would prefer to be without it. As managerial tasks become more complex, so the nature of the required information systems (IS) changes - from structured, routine support to ad hoc, complex enquiries at the highest levels of management. Global Information Systems aims to present the many complex and inter-related issues associated with culture in the management of information systems. The editors have selected a wide range of contemporary articles from leading experts in North America and Europe that represent a wide variety of different national and cultural environments. They offer valid explanations for, rather than simply pointing out cultural differences in articles that cover a variety of national cultures, including: China, Egypt, Finland, Hong Kong, Hungary, India, Jamaica, Peru South Korea, Kuwait, Mexico, Singapore, Sweden, the United Arab Emirate, the UK, and the US.

Part One

The Role of Culture in IS Development

This first part of the book focuses on the role played by culture in the development of IS. Research on IS development considers the challenges of IS design and development, the different development approaches used as well as the outcomes of the methods, the motivations and incentives of IS developers, and issues concerning IS developer productivity. Considerable work has focused on the cultural differences observed across different software development teams and approaches. A common finding across studies is that cultural variation explains different approaches towards software development and that an approach that works well in one cultural context may not work so well in another.

For example, Dagwell et al. (1983) found that systems designers’ approach towards end users varied across four nationalities (USA, UK, Australia, and Sweden): Australian and Swedish designers favored a more people-oriented approach to IS development whereas US and UK designers favored a more process and efficiency orientation. Likewise Kumar et al. (1990) found that the Danish designers emphasized people-related issues in ISD projects more than their Canadian counterparts, who emphasized technical issues.

This part comprises four papers on the role of culture in IS development. The first three of these examine culture from a national perspective whereas the final paper examines culture from an organizational perspective. Of the articles looking at national culture, the following nations are included: China, Hungary, India, Jamaica, Singapore, and the United States.

The first article, by Perlow, Gittell, and Katz, describes the approaches to software development employed by software development teams in India, China, and Hungary. Having spent time observing the different teams onsite, the authors were able to get detailed information on how much time the team members spent interacting with one another, and for what purpose, as well as how much time, and for what purpose, they spent interacting with the leader. The article demonstrates the dramatic difference in how the project leader interacts with the team members in the different cultural environments as well as the differences in how the team members interact with one another. Moreover, the reward structures and work hours are shown to be quite different.

The Walsham (2002) article – the second in this part – considers the consequences of cross-cultural software development teams. In Walsham’s paper, the interactions of Indian and Jamaican software developers are described, as the team members try in vain to complete an urgently needed software application. The differences in leadership style, in work habits, and in the very meanings associated with timetables and deadlines are shown to be stark across the two cultures. As a result, the developers from both cultures are challenged to adapt to the other.

In the third article in this part, Tan, Smith, Keil, and Montealgre (2003) address the question of information gathering and dissemination during software projects. Specifically, they are curious as to whether national culture influences the degree to which unfavorable project information (such as being behind schedule and or over budget) is shared upwards in the hierarchy. Such information has important implications for whether doomed software projects are continued, or recognized as troubled and stopped or adjusted. The cultures they consider are Singapore, and the United States where the United States is an ‘individualistic’ culture and Singapore is considered to be ‘collectivist’.

The final paper in this section examines organizational culture and its influence on CMM. The capability maturity model (CMM) has been part of a trend towards software development improvements over the past decade and is used by companies around the globe. Yet software development projects continue to fail at an alarming rate.

Ngwenyama and Nielsen (2003) uncover the core assumptions about organizational culture that are embedded into CMM and explain why CMM might not be successful in all oganizations.

With the increasing use of offshore development practices (Carmel and Agarwal, 2002; Kaiser and Hawk, 2004) as well as cross-cultural software development teams, it is essential to understand how value differences in culturally diverse software development teams may influence the systems development process and outcomes. Moreover, as software process improvements, such as CMM, become common, it is likewise important to understand how these are accepted, or rejected by, an organization’s culture. The four articles in this part help expand our understanding of these points.

References

Carmel, E. and R. Agarwal (2002), “The Maturation of Offshore Sourcing of Information Technology Work,” MIS Quarterly Executive, Vol. 1, No. 2, June, pp. 65–78.

Dagwell, R., R. Weber, and R. Kling (1983), “Systems Designer’s User Models: A Comparative Study and Methodological Critique,” Communications of the ACM, Vol. 26, No. 11, November, pp. 987–997.

Kaiser, K. and S. Hawk (2004), “Evolution of Offshore Software Development: From Outsourcing to Cosourcing,” MIS Quarterly Executive, Vol. 3, No. 2, June, pp. 69–82.

Kumar, K., N. Bjorn-Anderson, and R. King (1990), “A Cross-Cultural Comparison of IS Designer Values,” Communications of the ACM, Vol. 33, No. 5, May, pp. 528–538.

Ngwenyama, O. and P. A. Nielsen (2003), “Competing Values in Software Process Improvement: An Assumption Analysis of CMM From an Organizational Culture Perspective,” IEEE Transactions on Engineering Management, Vol. 50, No. 1, February, pp. 101–111.

Perlow, L., J. H. Gittell, and N. Katz (2004), “Contextualizing Patterns of Work Group Interactions: Toward a Nested Theory of Structuration,” Organization Science, September, pp. 520–536.

Tan, B. C. Y., H. J. Smith, M. Keil, and R. Montealegre (2003), “Reporting Bad News about Software Projects: Impact of Organizational Climate and Information Asymmetry in an Individualistic and Collectivist Culture,” IEEE Transactions on Engineering Management, Vol. 50, No. 1, February, pp. 65–77.

Walsham, G. (2002), “Cross-Cultural Software Production and Use: A Structurational Analysis,” MIS Quarterly, Vol. 26, No. 4, December, pp. 359–380.

1 Contextualizing Patterns of Work Group Interaction: Toward a Nested Theory of Structuration

Leslie A. Perlow, Jody Hoffer Gittell and Nancy Katz

Abstract: The focus of this article is the patterns of interaction that arise within work groups, and how organizational and institutional factors play a role in shaping these patterns. Based on an ethnographic study of groups across three national contexts, we describe the variation in patterns of interaction that we observed. We further suggest how different patterns of interaction form mutually reinforcing systems with aspects of the organizational context. In addition, we suggest how these mutually reinforcing systems are perpetuated by aspects of the broader institutional context. Our findings point toward a nested theory of structuration, expanding structuration theory to multiple levels simultaneously. In turn our findings have theoretical and practical implications for better understanding and managing interaction patterns among group members.

Introduction

How does work get done in groups? Why does similar work get done differently in different places?

To address these questions, we must explore what people actually do at work. As Stephen Barley and Gideon Kunda (2001, p. 90) aptly describe: “The dearth of data on what people actually do – the skills, knowledge, and practices that comprise their routine work – leave us with increasingly anachronistic theories and outdated images of work and how it is organized”. Moreover, if we want to understand how work gets done, we cannot strip away the context. Rather, we must contextualize our findings to better understand the phenomenon we observe (e.g., Johns 2001, Rousseau and Fried 2001). Also, we must consider how factors at multiple levels of analysis shape and constrain the phenomenon we study (Kozlowski and Klein 2000, Hackman 2003).

Back in the 1950s, researchers at the MIT Group Networks Laboratory established that variation in patterns of communication among group members affects group functioning and performance (Bavelas 1950, Leavitt 1951, Shaw and Rothschild, 1956). However, by the 1970s, this stream of research had been largely abandoned (Monge and Contractor 2001). Along with several other researchers (i.e., Argote et al. 1989, Brown and Miller 2000, Sparrowe et al. 2001, Cummings and Cross 2003), our purpose is to revive research that explores patterns of work group interaction. Moreover, we are interested in expanding the scope of this research to include both the national and organizational context in which these patterns emerge.

Prior research that explores the effects of contextual factors has tended to focus on either macrocontextual variables and how they affect organization structure (e.g., Hamilton and Biggart 1988, Maurice et al. 1980), or on organization-level factors and how they affect aggregate measures of group functioning and performance (e.g., Stokols 1981, Seers et al. 1995, DeMatteo et al. 1998). Some researchers have further explored how differences in cultural profiles, based on individuals’ value orientations (Hofstede 1980, House et al. 1999), affect how people behave in groups (Mann 1980, Earley 1993, Bond and Smith 1996). Missing in these studies is an exploration of the patterns of work group interaction. Moreover, missing in these studies is an exploration of how the multiple different levels of context affect each other, as well as the patterns of work group interaction.

Our research is unique in its multilevel focus on interaction patterns and both the organizational and national context in which these patterns exist. We further chose to focus on the patterns of helping as a type of interaction pattern because of the central role of helping behavior in completing the work of the software engineering teams we studied. Using what Hackman (2003) calls “informed induction,” we further identified the structures located at each level that appeared to most powerfully explain our local phenomenon, patterns of helping. We found that the reward structures can both explain and be explained by the patterns of helping. In turn, this mutually reinforcing relationship between the reward structures and helping patterns can itself be explained by elements of the larger institutional context.

Our analysis suggests a nested theory of structuration, extending structuration theory across multiple levels of analysis simultaneously. This nested theory of structuration further sheds light on alternative theories of fit – contingency, configuration, and congruence – suggesting how the underlying mechanism of mutual influence associated with structuration theory may explain how fit is achieved in these alternative theories. Our analysis also illustrates the value of unpacking aggregate measures, be they cultural or group aggregates, to better understand how work really gets done. In the end, our findings have theoretical and practical implications for better understanding and managing interaction patterns among group members.

Methods

Given how little is known about patterns of interaction within work groups and their relationship with the organizational and national contexts in which they exist, we engaged in close observation and built grounded theory. We studied software engineering as a type of work because, characteristic of knowledge work, software engineering includes both an individual and an interdependent component (Perlow 1999). Furthermore, the work is open-ended, creative, individually styled, and very demanding (Barley and Orr 1997). It cannot be standardized or fully planned out in advance (Bell 1973). Moreover, as it turned out, individuals – even on the components of their work for which they were individually responsible – often got stuck and needed help. We could therefore isolate a particular type of interaction – getting help on individual work – and explore how patterns of helping varied across groups.

We sought to hold constant the nature of the work as much as possible, while varying the context. Toward this end, we explored three teams of software engineers doing similar work at similar stages in the work process, but in very different cultural contexts.¹ We studied software engineers working for three partnerships, each a venture with the same American-headquartered multinational corporation. These partnerships were located in three distinct national contexts – Bangalore, India; Shenzen, China; and Budapest, Hungary. We refer to the three sites studied as follows: (1) Cco (all company names are pseudonyms), the joint venture located in Shenzhen, China; (2) Ico, the joint venture located in Bangalore, India; and (3) Hco, the strategic partnership, with longer-term plans to form a joint venture, located in Budapest, Hungary.

To further choose the three sites, we looked for teams who were comparable in terms of performance. These regions were home to some of the fastestgrowing, most technically advanced software development industries in the world (IDC 1995, NASSCOM 1997). Moreover, each of the local organizations we studied had been chosen by the headquarters for an alliance because of its reputation for being technically advanced in its region. The teams we studied were further chosen by senior managers at each location as being among their highest-performing teams of software engineers. Additionally, members of the U.S. headquarters helped to select the three teams studied to ensure highly comparable work assignments.

Research Sites

Physically, the offices looked quite similar. The Cco office was immaculate, air conditioned, and well lit. Cco engineers sat in cubicles in wide-open spaces with managers around the edges in closed offices. Like Cco, Ico was clean, well lit, and air conditioned. Ico engineers also sat in cubicles; however, their workspace was configured in honeycombs, rather than rows. At Hco, the entire company was not all c...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Contents
• List of Contributors
• Introduction: An overview of culture and IS
• Part One: The Role of Culture in IS Development
• Part Two: The Role of Culture in IS Adoption and Diffusion
• Part Three: The Role of Culture in IT Use and Outcomes
• Part Four: The Role of Culture in IT Management
• Index