Roger Clarke's Web-Site

© Xamax Consultancy Pty Ltd,  1995-2024
Photo of Roger Clarke

'Research Perspective in the Basket of 8'

Researcher Perspective in the 'Basket of 8' IS Journals:
A Weakness in Disciplinary Scope?


Review Version of 7 March 2018

Roger Clarke, Robert Davison and Wanying Jia **

© Xamax Consultancy Pty Ltd, Robert Davison and Wanying Jia 2016-18

Available under an AEShareNet Free
for Education licence or a Creative Commons 'Some
Rights Reserved' licence.

This document is at


In this research, we explore the perspectives that IS researchers adopt in their research designs, considering articles published in AIS Basket journals in the years 2001, 2008 and 2015. By researcher perspective, we refer to the viewpoint from which phenomena are observed. This usually represents the interests of one or more stakeholders, and strongly influences the research design. Following a review of the literature, we develop a coding scheme that we apply to a corpus of 548 articles. 97% of the articles adopted a single-perspective approach rather than seeking a balance between the interests of two or more stakeholder groups. Further, 90% of the articles were committed solely to the interests of the entity central to the research design and 93% of articles considered only economic aspects of the research.

The choice of researcher perspective has considerable influence on the design, conduct and outcomes of any research endeavour. Our examination of the data concludes that most recent IS research reflects the selection of a single beneficiary, i.e. just one among the stakeholders, who stands to benefit most from the research. We conclude that this is both unhealthy and likely to cause missed opportunities. We contend that the principle of triangulation needs to be applied not only to data sources and research methods, but also to researcher perspectives, and that a consequent broadening of the IS discipline's scope is essential.


1. Introduction

In IS research, we commonly study phenomena in various contexts that include information, systems that handle that information, and technology that supports that information-handling. The organisations and individuals within our field of view are stakeholders, with interests that they wish to protect and advance. The purpose of the research may be exploratory or confirmatory. It may also involve the development, testing, modification or refutation of theory. In empirical research designs, it is common that the interests of one or more stakeholders are privileged. These stakeholders may include: the system sponsor or organisational entity that is central to the research context; the employees who work in the organisational context that is being studied; the customers or consumers who benefit from or are affected by the information system. The stakeholder notion can also be interpreted to include non-human entities such as local, regional or planetary ecosystems.

In this article, we explore the nature of stakeholder perspective in IS research. We believe that this is a significant issue for the IS discipline because the impacts that IS research may exert are largely determined by the perspective that the researcher takes. In other words, if IS researchers predominantly privilege the perspective of a single stakeholder, then the impact that IS research has will also be limited to contexts and problems where that single stakeholder has an interest. Further, that single stakeholder may be served sub-optimally because the researcher's appreciation of the problems will lack the richness that would have resulted from a more holistic exploration of the phenomenon.

A flavour of the arguments that we develop in this article is provided by consideration of a particular research genre: personal data markets. The research published in this area commonly treats the interests of marketing corporations as objectives, whereas those of consumers are conceptualised as constraints on the interests of the corporate players, and as challenges to the corporations' business models. Much of the research undertaken in this genre comprises the playing of laboratory games designed to help understand how corporations can minimise the cost of inveigling consumers into trading their privacy off for a service, for convenience, or for a token amount of money. This process actively stimulates an arms race, in which those sympathetic to consumers' interests produce and distribute means of combatting the consumer marketers' techniques.

A much more constructive approach to research on personal data markets would reflect the perspectives of not only marketers and marketspace operators, but also consumers. This would lead to a deeper understanding of the various stakeholders' interests and needs, and a holistic grasp of the market as a gestalt. The foundations would be laid for win-win-win strategies, sustainable market models, and active and informed participation by consumers, rather than sullen capitulation to marketers' power by some consumers, and opposition and interference by others.

In order to investigate the nature of perspective in IS research, we frame our overarching research question as: What are the most common perspectives adopted by IS researchers, as evidenced by the IS literature? Table 1 contains the more detailed questions that we subjected to empirical testing. These emerged from the theoretical analysis presented in the following sections.

Table 1: Detailed Research Questions

  1. What researcher perspectives are evident in the AIS 'basket of 8' journals?
  2. What changes in researcher perspectives are evident in the AIS 'basket of 8' journals during the recent past?
  3. To what extent are the adopted researcher perspectives aligned with the economic, the social and the environmental dimensions?
  4. To what extent are the studies single-, dual- and multi-perspective in nature?
  5. To what extent does a balance exist between the perspective of the system sponsor and the perspectives of other stakeholders?
  6. Among studies whose object of study is people, to what extent does a balance exist between the perspective of the system sponsor and the perspectives of other stakeholders?

In order to investigate which perspectives are adopted by IS researchers, we analyse a representative sample of articles published by our top journals (AIS Basket of 8) over the last 15 years. As we describe below, we develop a coding scheme to classify the nature of researcher perspective as presented in each of the 548 relevant articles in our sample.

Following this introduction, we first investigate the notion of researcher perspective through an extensive review of the relevant literature. This is followed by an explication of the research method, including both a detailed account of our sampling technique, a description of our coding scheme and how we developed it, and the delineation of more detailed questions aligned with the overarching research question as expressed above. The findings follow before we discuss the outcomes and assess the limitations, implications for future research and conclusions.

2. Theory Relating to Researcher Perspective

This section introduces the notion of 'researcher perspective' and how it relates to the IS research process. Consideration is then given to the stakeholders whose perspectives may be adopted, the dimensions on which the interests of stakeholders lie, and the nature of single-perspective, dual-perspective and multi-perspective research.

2.1 The Concept

The concept that is central to this research is 'researcher perspective'. We use this expression to refer to the viewpoint from which phenomena are observed by the researcher. A wide variety of viewpoints are possible. This means that the researcher, in designing and conducting the research, adopts a viewpoint that privileges the interests of one or more stakeholders. The acceptance of interpretivism within the IS discipline brought with it the recognition that "the phenomenon of interest [is] examined ... from the perspective of the participants" (Orlikowski & Baroudi 1991 p.5). However, this is better expressed as 'from the perspectives [plural] of the participants', to avoid the presumption that all participants share the same view. Further, phenomena are subject to multiple interpretations and so the perspective adopted by any one party is not definitive but simply one among many.

Searches for terms associated with the concept of perspective have been largely unsuccessful, in both text-books on research and in journal articles dealing with research process. Searches in leading IS journals identify limited usage in the sense described here. Perspective is frequently used to refer to the theoretical lens, but seldom relates to the interests of a stakeholder.

It is important to appreciate the role of the concept of perspective within the research process. A perspective is quite distinct from an 'object of study' or a 'unit of study'. The 'object of study' is the set of phenomena that the researcher observes, and the notion of a 'unit of study' refers to the level of granularity of the observation. Perspective, on the other hand, refers to the direction or orientation from which the observations of the phenomena are undertaken. Hence, a researcher adopts a perspective from which observations are made of an object of study, at a level of abstraction called the unit of study, together with a method that enables the collection and analysis of data. In the diagrammatic representation in Figure 1, the first segment depicts the inter-relationships among the methodological concepts, and the second segment provides an example, whereby a particular research project adopts the perspective of the system sponsor, with the researcher observing the activities of a work-group through the lens of the Technology Acceptance Model (TAM) (Davis, 1989).

Figure 1: Conceptual Model of the Research Process

2.2 Stakeholders

A commonly used term for the categories of entity that have an interest in particular phenomena is 'stakeholders' (Freeman & Reed 1983). Many stakeholders are participants in the process or intervention, in such roles as investor, data source, technology provider, system sponsor and user (Seddon et al. 1999). However, the categories of stakeholders are broader than this (Pouloudi & Whitley 1997, p.3), comprising not only "participants in the information systems development process ... viz. individuals, groups or organizations who take part in a system development process" but also "any other individuals, groups or organizations whose actions can influence or be influenced by the development and use of the system whether directly or indirectly".

Contrary to this inclusive approach, there is a strong tendency in industry and government practice, recognised in academic analysis, to include, as a qualifying condition for a category of entities to be recognised as a stakeholder, a requirement that the party be recognised as being capable of significantly affecting the success of the project. This might derive from market power, or, in the case of a regulator, for example, institutional power. This approach has the effect of marginalising all but the most powerful participants (Mitchell et al. 1997, Achterkamp & Vos 2008).

The foreground example of less powerful participants is what are commonly referred to as 'users'. In B2B and G2B contexts, these may take the form of business enterprises that are smaller or otherwise less powerful than the system sponsor. A common example is a large 'hub' or downstream corporation (such as a motor vehicle assembler or a retail chain) dealing with smaller providers of components or shelf-stock. In some circumstances, less powerful players may have genuine choice, but in others they may be forced to comply with what are tantamount to instructions. Thus, Walmart (Narsing, 2005) requires all of its suppliers to be compliant with its RFID labelling requirements. Similarly, in B2C contexts, the value proposition offered by a vendor or platform is typically non-negotiable: consumers may have little prospect of acquiring goods or services unless they fall in line with the system imposed on them by the platform, as well as any associated legal environment, that together enable and govern the transaction (Muzellec et al., 2015).

Contemporary information systems have very substantial reach. As a result, there may be entities that reside in the background yet are nonetheless materially affected by the implementation of these systems. For example, an online travel booking system has the potential to disintermediate one or more companies, resulting in the cessation of business operations, layoffs, and dislocation and economic hardship for employees' families, and perhaps for others in the regions in which the disintermediated company operated. Another example relates to information systems that store personal data about third parties, as arises with databases relating to credit reporting, tenancy, and criminal intelligence. The agreement[1] between the US and the EU with respect to the collection and storage of air passenger data also comes into this category (Mitsilegas, 2015).

The term 'usees' is descriptive of entities that are affected parties who are not participants. The term has been in casual use since the mid-1980s in IFIP Technical Committee TC9 (ICT and Society), and especially Working Group WG9.2 (Social Accountability and Computing). See also Clarke (1992), Fischer-Hübner & Lindskog (2001), Baumer (2015). The entities may be organisations or individuals, and may fall into various categories, defined by, for example, attribute, function or location. In some cases, usees are largely or even entirely unaware of the existence of the system or their entry in it. This applies, for example, to criminal intelligence databases and the data-holdings of companies that surreptitiously gather data about Web-users' online behaviour. Examples of record systems that usees are more likely to be aware of include credit bureau holdings (which exist for both consumers and corporations), pooled records of insurance claims, and tenant databases. In some circumstances, usees might benefit from such systems, but it is more common for their interests to be harmed by them. The interests of usees can only be recognised and appreciated if researchers select perspectives that accommodate their viewpoints.

2.3 The Significance

The perspective that a researcher adopts has a very significant influence on the entire research undertaking. The researcher perspective dictates the framing of the research, it drives the selection and formulation of research questions, it provides the criteria based on which alternative research designs are evaluated, and it effectively determines what is included within or excluded from the potential outcomes of the research, as well as how those outcomes are expressed. In effect, the choice of researcher perspective reflects the choice of a beneficiary, i.e. the stakeholder who stands to benefit most from the research. It also frames how any discussion of the contributions of the research can proceed, since the beneficiary of those contributions is likely to be closely linked to the perspective that the researcher takes.

In order to provide insights into the depth of the impact of the choices made in relation to researcher perspective on the formulation of research questions, the authors devised a research question that is reasonably pertinent and not unduly unrealistic: "What are the impacts of the withdrawal of the customer option of receiving printed invoices through the post?". Table 2 presents a set of alternative pairs of researcher perspective and research question.

Table 2: Alternative Research Questions from Different Perspectives

2.4 Dimensions

Stakeholders exhibit considerable diversity in their weltanschauung. In business contexts, most system sponsor interests are ultimately financial or economic in nature. However, although the economic dimension has tended to dominate discussions, tension has always existed between economic and social interests.

There has long been a tradition within some segments of the IS discipline of reflecting the social interests of individual users within organisations, through the socio-technical approach and participative analysis of requirements and system design (e.g. Land & Hirschheim 1983). As systems reached beyond organisational boundaries, organisations other than the system sponsor were recognised as having a stake in their design and implementation (Pouloudi & Whitley 1997). Then, as individuals outside the organisation became users of networked computing facilities, the systems became extra-organisational in nature (Clarke 1992), and the individuals' actions and interests, both economic and social, also needed to be encompassed within the frame of reference.

During the last 50 years, as computing became a major consumer of electricity and as cathode-ray tubes mounded up in scrapyards, impacts on the environment came to the fore. More recently, as the economic value of, for example, coltan[2] has come to be realised for the supply chains associated with the production of tantalum capacitors used in a wide variety of electronic devices, so the exploitation of mineral rich deposits located in developing countries (primarily Brazil and DR Congo), and the associated environmental degradation, has also reached our collective attention. As a result of developments such as these, it is now reasonable to treat the environment as having joined economic and social needs as a third dimension.

Economic and social interests are associated with reasonably definable entities such as people and organisations of various kinds. Environmental interests might also be treated that way (e.g. by associating them with nature and wilderness reserves, national parks and environmental trusts). On the other hand, it may be more appropriate to treat the abstract notion of the environment as being a third dimension along with the economic and social dimensions. The emergence of 'triple-bottom-line'[3] reporting ostensibly reflects this kind of thinking (Elkington 1994), although it is also critiqued as being inadequate to ensure genuine sustaining of the earth's ecology (Milne and Gray, 2013). Nevertheless, this tripartite notion has become a precept underlying the notion of corporate social responsibility (CSR) (Hedman and Heningsson, 2016). It is notable that a large proportion of 'Green IS' research focuses on the economic dimension, given an interest in corporate wellbeing in an era of rising energy costs. Research on the environmental dimension is, however, also on the rise (Watson et al. 2010, Elliot 2011, Deng et al. 2015, Gholami et al. 2016).

2.5 Levels of Abstraction

In categorising the range of alternative perspectives that can be adopted, it is useful to combine the concept of dimensions with the 'unit of study' notion. Phenomena may be observed at different levels of abstraction. For example, a corporation or government agency has multiple sub-organisations and individual roles within them; and the interests of humans affected by information systems can be considered at the level of each individual employee, of work-groups, or of the employed workforce as a whole; or at the level of each individual external to the organisation, the communities with which they identify, or society, variously at the level of a region or a nation. Understanding of environmental phenomena is similarly tiered, reflecting, for example, individual-species-within-context, local bio-communities, regional bio-communities, or the biosphere as a whole.

In Table 3, a wide range of perspectives is presented each of which sometimes is, or could be, and, we contend, should be, adopted by IS researchers under some circumstances. The perspectives are allocated to the dimension to which they belong, and organised in descending order of abstraction.

Table 3: Dimensions and Levels of Abstraction of the Three Perspectives

One actor is central to the operation of information systems, and is accordingly often foremost in the minds of IS researchers. We use the term system sponsor in this article to refer to the organisation (or entity or unit) that develops, implements or adapts a system, process or intervention, causes it to be developed or implemented, or for whose benefit the initiative is undertaken. In some cases, however, the perspective adopted may be that of a category of organisations, such as those that install a particular ERP package, or adopt a particular category of application, such as electronic health record systems or cloud-based customer relationship management (CRM) services. In such contexts as joint ventures and collaborative inter-organisational schemes, the system sponsor may be a collective. Our informal observation, which we further investigate in this article, is that, in a great deal of IS research, the research questions (or, in design science and action research, the objectives) are formulated with the system sponsor's interests at least as the primary focus, and even as the sole set of interests that is recognised by the research design. The interests of all other stakeholders are then treated as constraints on the achievement of the system sponsor's interests.

The focus on the system sponsor as a central player in an IS is evident in positivist empirical research, where observational studies, experiments and surveys are conducted in order to understand the impacts of interventions, from the system sponsor's perspective. In addition, a great deal of interpretivist research is also performed with the intention of providing the system sponsor with an understanding of those impacts.

In design science, the sense in which the term 'design' is used is "the purposeful organization of resources to accomplish a goal" (Hevner et al. 2004). Most commonly, that goal is formulated in order to serve the interests of the organisation(s) by or for which the intervention is made, or the design activities are performed. Action Researchers, meanwhile, aim to ameliorate organisational problem situations for all relevant stakeholders (Davison et al. 2004). Action research that is informed by a critical or emancipatory epistemology may go further and prioritise the interests of the individual employees or usees (Ledwith 2016).

Business enterprises define objectives specifically in terms of the organisation's own interests. This is 'in the DNA', because the nature of the joint stock company is such that directors have a legal obligation to act in the interests of the company. In some cases, social and/or environmental interests may be directly commensurate with economic interests. In other circumstances, regulatory mechanisms as diverse as statutory obligations, activism among investors or employees, and public opinion may cause an organisation to compromise its own economic interests in order to provide social and/or environmental benefits (Hedman & Henningson 2016). This mode of thinking has migrated beyond the business context, because mission statements and corporate objectives are now the driving force in the public sector and not-for-profit organisations as well. The system sponsor may recognise that other entities also have interests, yet those interests are seldom treated as objectives, and generally operate only as constraints on the organisation's achievements of its own objectives.

2.6 Single-Perspective Research

It is common for IS researchers to adopt the perspective of only one of the many possible stakeholders. There are a number of advantages of this approach. It simplifies many aspects of the research, such as the formulation of the research questions or objectives, the design of the research process, and the expression of the findings. The results of the research are likely to be at least comprehensible to that stakeholder, and quite possibly also of interest, relevance and value to them.

In practice, as discussed above, the system sponsor is seldom the only party that has an interest in the system's operation. The multiple stakeholders in information systems often have distinct interests. Moreover, it is likely that the interests of some of the pairs of stakeholders will be at least to some extent and even diametrically opposed, in patterns that approximate a zero-sum game. In those circumstances, the decision by the researcher to privilege the perspective of one of the parties is tantamount to a political act, whether the researcher has made a conscious choice or is merely following the established lines of a particular research genre. The latter is referred to as unconscious hegemonic participation and refers to the state where an author adopts and takes for granted a dominating ideology (Wall et al. 2015). Unconscious hegemonic participation can be manifested in the dominance of common research topics, questions, theories and methods, as well as beliefs about how research should be undertaken and research results interpreted (Wall et al., 2015). According to the emancipatory precepts of critical theory, it is likely that the choice most commonly exercised will be that of the most powerful stakeholder, whose ideological hegemony is unlikely to be challenged, especially if it is also this stakeholder that underwrites the research funding (cf. Stahl et al., 2008).

Whether or not selection bias exists in the researcher's choice of perspective, questions arise about the value that single-perspective research can offer to the party whose perspective is selected. Research undertaken from the perspective of any one participant may in effect grasp one part of a large elephant, without gaining much of a feel for the remainder of the pachyderm. The conclusion might be reached that a particular system feature is beneficial to the system sponsor's interests, but it is unlikely that much insight can be offered into the intensity of opposition that the feature might engender among other stakeholders, how that opposition may be manifested, or how the impacts of that opposition might be mitigated. In short, the principle of triangulation needs to be applied not only to data sources and research methods, but also to researcher perspectives.

We offer the following example of how richer understanding might be achieved in the research domain of social media funded by advertising. Research from the system-sponsor's perspective might adopt the following research question: "What proportion of social media users need to authorise the provider to exploit their data to ensure that advertising-based business models are viable?". On the other hand, research conducted from the perspective of social media users might ask the following, rather different, research question: "What techniques and tools are available to social media users to enable them to obfuscate, subvert or falsify their identities and locations, in order to prevent the provider from exploiting their data?". This might be further extended to investigate perceptions of the ethicality of such activities, or the understandability and practicality of such techniques and tools.

2.7 Dual-Perspective and Multi-Perspective Research

As an alternative to the single-perspective approach, we conceive of a research question that internalises the tensions and enables the emergence of insights of value to at least two protagonists. A research question that corresponded to these expectations could read as follows: "How do the views of social media users and providers compare with respect to providers' Terms of Service, privacy features and policies, user identity, and obfuscation or falsification of data by users related to their location or other privacy-sensitive information?". An example of research that recognises this potential is Fletcher (2015), in which consumer power, exercised for example through the use of ad-blocking software, has negatively influenced the impact of online advertising and business attempts to benefit from CRM systems.

Even where a competing interest lacks power (such as the capacity not to adopt a feature, or to misuse it), the system sponsor may nonetheless benefit from dual-perspective research, because it can provide a more comprehensive understanding of the attitudes and likely behaviours of the various actors. This may also apply to the impacts on and behaviour of indirectly-affected parties, referred to above as 'usees'.

Many inter-, multi- and extra-organisational systems involve far more than two stakeholders. Beyond dual-perspective research, multi-perspective approaches need to be understood, their benefits appreciated, and appropriate research techniques adopted and matured. Examples of research questions that would deliver value to policy-makers include: "What are the social and economic impacts of the current business model for social media?" and "What benefits and disbenefits would accrue to which stakeholders if regulatory measures were imposed in order to achieve balance between the interests of providers and users?".

These kinds of multi-perspective research questions do not represent the mainstream in articles published in IS research journals. However, we feel that the value and impact of IS research may be significantly weakened if the discipline fails to develop techniques that facilitate the study of situations characterised by multiple, partly conflicting interests. For example, in the context of international trade EDI, Cameron & Clarke (1996) addressed the research question: "What are the critical success factors for a project management framework for collaborative inter-organisational systems, from the viewpoints of each of the players?". More recently, Agarwal et al. (2012) examined cyber-collective social movements (CSMs) such as the use of social media in the 'Arab Spring' of 2010-2012. After surveying the available research methods literature, the authors developed an analysis based on Individual Perspective, Community Perspective, and Transnational Perspective. The article featured no system-sponsor, but rather three levels of abstraction of the social rather than of the economic dimension discussed earlier in this article. Another example is Selander and Jarvenpaa (2012), who expressly adopt the perspective of social movement organizations that work for changes in societies. Both Agarwal et al. (2012) and Selander & Jarvenpaa (2012) are noteworthy in being among only a small number of papers located by the research team in which the term 'perspective' is used in a manner similar to that proposed in the present paper.

The opportunity exists to extend research techniques whose focus is on the construction of artefacts (which we refer to here as 'constructivist') into multi-perspective research. The way has already been shown by action research, because it adopts "the idiographic viewpoint [whereby] any meaningful investigation must consider the frame of reference and underlying social values of the subjects" (Baskerville 1999). Thus, Olesen and Myers (1999), in documenting the adoption of Lotus Notes by senior management at a university in New Zealand, experienced an action research failure when the interests of the system sponsor (the university) clashed with those of the personal assistants (users) of the senior university managers (usees): the users refused to apply the system as it eroded their power to control their managers' diaries.

A similar approach can be adopted with the other major constructivist research technique, design science. The accumulated understanding of socio-technical thinking (Emery & Trist 1960, Mumford 2000) can be applied in order to articulate what might be usefully described as 'participative design science'. Beyond asking 'What is a feasible and effective process for the design of a particular system or category of systems?' (cf. Guideline 3 of Hevner et al., 2004), research questions of the following form can be investigated: 'What is a feasible and effective process for reflecting the perspectives of all parties in the design of a particular system or category of systems?'.

Each of these opportunities may benefit from consideration of recent discussions in the literature relating to 'phronesis'. This Aristotelian notion injects an ethical flavour into research design, by requiring deliberation about the appropriate values that need to be applied, which in turn depends on appreciation of the various perspectives involved and accommodation of the various stakeholder interests (Harrison & Zappen 2003, Constantinides et al. 2012). In addition to seeking relevance to practitioners, Constantinides argues that IS researchers should consider "a whole range of entirely different and relevant ends for IS research with different audiences" (p.2). These audiences include professionals, managers, executives and company directors; and policy-makers in parliamentary, governmental and advocacy contexts. Once again, the ground has been prepared, in this case by critical theory research. This is inherently multi-perspective in nature, being directly concerned with conflicts among the interests of the various actors, and with the power-structures that determine the outcomes. It is also inherently instrumentalist, because it is conducted with the express intention of influencing the phenomena that are under observation (Klein & Myers 1999, McGrath 2005; Cecez-Kecmanovic 2001, 2005; Cecez-Kecmanovic et al. 2008).

In Table 2 above, we provided a set of examples showing how the choice of researcher perspective affects the formulation of the research question, using the example of: "What are the impacts of the withdrawal of the customer option of receiving printed invoices through the post?". We contend that an alternative approach can deliver even more value. We use the term 'instrumentalist research' to refer to undertakings that begin with a perceived problem, are goal-oriented, and involve a search for a solution to that problem. Techniques which are essentially instrumentalist in nature include action research (Avison et al., 2001; Davison et al., 2004) and design theory research (Land & Hirschheim 1983, Hevner et al. 2004). For each of the research questions in Table 2, an alternative, more useful formulation is possible, along the lines of the following: "How can an organisation manage the negative impacts on [insert stakeholder(s)] resulting from the withdrawal of the option of receiving printed invoices through the post?".

3. Research Background and Method

The research reported in this article investigates the current and recent practices in the IS discipline in relation to researcher perspective. This section outlines prior research, expresses the detailed research questions, and explains the method adopted in order to address them.

3.1 Prior Research

We found little evidence of prior work on the nature of researcher perspective, its consequences, and its application within IS research. We searched the literature in a number of ways. First, we searched using the key terms `perspective' and `research(er) perspective' both conjoined and as separate terms, variously with and without the qualifying term `information systems'. Our primary search vehicle was Google Scholar, in order to cover as large a catchment as possible, although we also searched the AIS eLibrary, in order to complement the generic approach with the discipline-specific collection of well over 30,000 papers.

We encountered difficulties in identifying further terms likely to be correlated with the notions being examined, especially terms that were not highly ambiguous. One such term was 'stakeholder', particularly when used in ways that could reflect at least recognition of complexity, and perhaps even an endeavour to address it. For example, the use of the plural form ('stakeholders') was considered more likely to indicate a relevant approach, particularly in conjunction with 'interests', and perhaps also with 'conflict'.

In the current study, we deliberately elected to focus on a set of articles that represents high quality research conducted on a variety of topics in IS and published in high quality journals that are readily accessible to IS scholars around the world. For these reasons, we selected articles from the AIS basket of 8 journals. This approach is consistent with Grover and Lyytinen's (2015) and Tarafdar and Davison's (2018) recent meta-analyses that also examined articles published in the same set of eight journals. Further, while there are overlaps in coverage across these journals, each has developed particular areas of primary focus. We suggest that the articles published in these journals will be representative of high quality IS research, where we would expect to see a variety of researcher perspectives presented.

Most uses of the term 'research(er) perspective' merely distinguish between the interests of researchers on the one hand and professionals, educators or students on the other. Most uses of the term 'perspective' alone relate to the theoretical lens adopted, and some to interests of an actor within the research domain. The few cases found where the term relates to the angle of view adopted by the researcher are cited in the appropriate places within this article. The concept of researcher perspective can, however, be detected in the literatures on project success (DeLone & McLean 1992, Seddon et al. 1999), project failure (Lyytinen & Hirschheim 1987, Sauer 1993, Heeks 2002), and soft systems methodology (Checkland 1981, Avison & Wood-Harper 1990). The concept can be inferred from the focus of these genres on the extent to which practitioners do or do not recognise and manage the interests of stakeholders in an information system.

Action research necessarily involves careful attention by the researcher to the interests of all stakeholders involved in the project. Searches of literature on action research techniques suggest that the perspective of the researcher is seldom explicitly discussed; implicitly, it is assumed that the approach is holist in nature, and hence agnostic to the values of specific participants and to the resolution of conflicts among them (Baskerville & Wood-Harper 1996, Davison et al. 2004). Nevertheless, a careful reading of action research papers suggests that in some cases the values of different groups of participants are identified and hence that multiple perspectives are taken by the action researcher. Thus Wong and Davison (2018), in their action research investigation into knowledge exchange practices in a global logistics firm, identify the values of both employees (the knowledge exchangers) and management (the system sponsor). Further, we suggest that action research projects are more likely to be successful when they do consider the interests of all significant stakeholders, and not only those of the system sponsor. As Davison et al. (2012) observed, an AR-based change that is designed exclusively from the perspective of management (the system sponsor) but is of little interest to employees, is likely to fail when it depends on the same employees' cooperation for implementation.

Guidelines in relation to critical theory research also include evidence of the 'researcher perspective' notion. For example, Principle 2 of Myers & Klein (2011, p.25) states that "Critical theorists advocate values such as open democracy, equal opportunity, or discursive ethics", and 4 and 5 argue for an orientation towards "individual emancipation" and "improvements in society".

A key exemplar in both the social dimension and the multi-perspective approach is Agarwal et al. (2012). Despite being an AIS prizewinner, this article has accumulated only 27 citations in its first 5 years, suggesting that it sits outside the disciplinary mainstream. This is consistent with our initial finding from the literature review that there is a limited literature on the specific matters at the heart of the present research. A second exemplar, with a similar number of citations, which also involves the social dimension and is dual-perspective in nature, is Lin et al. (2015). This contrasts the interpretations of a system sponsor and the aboriginal people who the system was designed for, but upon whom it was in effect imposed.

3.2 Research Method

In the Introduction, we presented our overarching research question as: What are the dominant perspectives adopted by IS researchers, as evidenced by the IS literature? In Table 1, we articulated the general question into a set of six more specific research questions.

In order to gain an insight into changes in researcher perspective that have occurred during the recent past, three years were selected, and samples extracted from each. We started this research in 2016, so the most recent complete year (2015) was selected as an end point. 2001 was chosen as a starting-point, on the basis that JAIS, the most recently established of the 8 journals, commenced publication in 2000. The mid-point year, 2008, was added, to provide two 7-year intervals. The sampling frame accordingly comprised a total of 8 x 3 = 24 volumes. In seven cases, the journal volume numbering corresponded with calendar years throughout. In one case (JMIS), the journal switched from an overlapping arrangement in 2015, and as a result published five issues rather than four in that calendar year. The number of issues per volume varied both among journals and across years, with the result that the sample extends across a total of 118 issues. The focus was on refereed journal articles, and hence unrefereed contributions such as book reviews, calls for papers and editorials were generally excluded from consideration.

Even though some previous meta-analytical studies had been conducted (e.g. Grover and Lyytinen, 2015; Tarafdar and Davison, 2018), it remained unclear how resource-intensive the assessment of each article would be. The evaluation of articles in order to enable coding requires familiarity with topics, methods and styles across the breadth of the discipline. It is also slow and fatigue sets in reasonably quickly, which increases the elapsed time to complete the project and/or reduces data quality. A three-phase process was accordingly devised.

In the first phase, a 20% sample was selected from the sampling frame: a total of more than 100 articles. In each article, the full text was evaluated, and the article coded. In the second phase, a sub-sample was selected from the full sample. In this case, only the article's title and abstract were evaluated, and the article was coded on that basis. The number of articles and the volume of text involved were such that the analyst could no longer remember the full detail of each article, and hence bias based on deeper knowledge acquired in the first phase was limited and controllable. Comparisons were then made between the whole-article and title-and-abstract-only assessment processes and results. The conclusion was reached that articles could, with only limited qualifications, be as confidently coded on the basis of title and abstract alone as they could be on the basis of the full text. The third phase comprised evaluation of the remaining articles in the sampling frame, a total of over 500 further articles. The protocol specified that, where doubts arose, the full article was to be checked.

For Phase 1, a technique was devised to select a 20% sample of the articles within each journal in such a manner as to achieve reliability, reproducibility and pseudo-randomness. Within each journal, the first article was selected into the sample; in turn every fifth article thereafter was selected in the order they appeared in the journal in 2001, 2008, and 2015. The technique is fully specified in Appendix 1. This resulted in a sample of 135 articles, with the count per journal ranging from 10 to 25.

Each of the selected articles was then examined in accordance with a coding protocol. The Coding Guidelines are provided in Appendix 2. The protocol was established prior to the commencement of coding, but iteratively refined by the authors throughout the coding process, with revisions being made in particular with respect to descriptions of how the protocol should be interpreted in particular contexts. This refinement of the protocol was necessitated by the complexities involved in applying the protocol consistently and systematically. To be relevant to this research, an article needed to have substantial and direct relevance to at least one real-world stakeholder. Articles that did not satisfy this requirement were coded 'Discipline Internal'. Examples include teaching cases, research methodology articles, and reports addressed solely to researchers. The 19 articles that we coded as `Discipline Internal' in Phase 1 were excluded from the later analysis, with the result that the Phase 1 data set was reduced to 116 articles.

The essential components of the protocol involved identifying the: category for each article (Constructivist; Theoretical; or Empirical); dimension on which the perspective lies (Economic, Social or Environmental); object of study (Organisation, Human or Technology); cardinality of the article (single, dual or multiple perspectives); and perspective taken (system sponsor, other, or greater than one perspective). The purpose of coding for category was to assist in identifying the article's purpose. Articles that are strongly Empirical in nature have 'research questions', whereas Theoretical articles may focus on a topic or domain, and Constructivist research has an objective. (Because all 'strongly Empirical' articles have theory content as well, and many 'Theoretical' articles have some empirical content, typically 'illustrative' in nature, boundary cases exist, and hence some degree of arbitrariness arises).

During this phase, the examination extended across the whole article, with particular reference to the title, abstract, introduction and conclusions, seeking out information relevant to the research method, intended beneficiary and target audience. Where an express statement about the researcher's perspective was found, this was noted but checked against other evidence within the article, to ensure that the coding reflected the actual characteristics of the research. In the large majority of cases, however, the perspective needed to be inferred by the coder. Key quotations were identified and the article was categorised in accordance with the protocol, in order to enable the research questions to be addressed.

The articles were first coded by a trained Research Assistant, then reviewed and if necessary re-coded by one of the lead-researchers, then further reviewed by the other lead-researcher. Where discrepancies arose, they were discussed and resolved. Respect for the protocol ensured that sufficient material was recorded about each article to show the basis on which codes were assigned, and to support review, audit, and replication. It was important that we achieved complete agreement on all codings, thereby satisfying the requirement for inter-rater reliability. The detailed coding sheets are provided in the Supplementary Materials.

In Phase 2, a sample-size issue was confronted. The count of 116 relevant articles was satisfactory for addressing questions about the sample as a whole, but the counts in many of the cells were sufficiently small that drawing inferences from them would be fraught with danger. An investigation was accordingly undertaken to establish whether a larger sample-size was feasible within acceptable resource-constraints. During the Phase 1 coding process, the clear impression had emerged that the code-assignment process was in most cases well served by the title and abstract. A test was accordingly devised of the proposition that the same degree of confidence in the data-coding process could be achieved based on the abstract and title rather than the full-text. A 15% sample of the initial sample of 135 articles was selected. A spread of the 20 articles was achieved, across journals, issues and years, by commencing with the 2nd article of the 1st journal and selecting every 7th article up to the last in the sample.

Once again, text was sought that was indicative of the research method, intended beneficiary and target audience, relying on explicit statements wherever they were available, and drawing inferences only where necessary. Key quotations were identified, and the article was coded. The coding generated from the title-and-abstract evaluation was then compared with that from the full-text. With one exception, the coding was equivalent, and could be achieved much more quickly considering only title and abstract. In one case, the title and abstract were not sufficiently clear to enable coding of all aspects with confidence. (The full text had also presented difficulties, but was marginally clearer). In two further instances, the coding of category as either Theoretical or Empirical was consistent across the two methods, but the confidence in the coding was somewhat lower using title and abstract alone.

In Phase 3, all remaining articles in the sampling frame of 118 issues across the 8 journals were evaluated on the basis of title and abstract. Where any uncertainty arose, the full text was evaluated. This was the case with 47 articles (9%). Of the further 524 articles, 92 were Discipline-Internal, and hence usable data was captured for a further 432 articles. The remainder of this article reports on the total of 548 (116 + 432) relevant articles. The full set of 548 relevant articles from the 24 volumes provided a satisfactory data set for most of the intended purposes.

Table 4 summarises the counts of selected and relevant articles in the eight journals, including the numbers of articles which were not addressed to stakeholders external to the IS discipline (coded Discipline-Internal, DI), and the remaining number that are relevant to the present research.

Table 4: The Article Counts

4. Findings

In this section, we report the results arising from the assessment. The detailed coding sheets, including statistical summaries of the whole data set and within years and across journals, are provided in the Supplementary Materials.

Of the 659 articles, 111 (17%) were coded as being `Discipline Internal' (DI), in that they were not dealing with any external phenomena, were not addressed to real-world stakeholders but instead primarily to other researchers, and/or their focus was on research methods or teaching methods. Of the 8 journals, only 4 (EJIS, ISJ, JIT, MISQ) scored close to the mean of 17%. Three journals had very few DI articles (JSIS, JMIS; ISR), while 1 evidenced relatively high proportions of such articles (JAIS). Over the three selected years, no pattern was apparent, with the proportions in each journal varying considerably. Figure 2 presents a detailed distribution of the DI articles.

Figure 2: Distribution of Discipline-Internal Articles

That left 548 articles (83%) relevant to the study, with the proportion from each journal varying from a low of 71% (JAIS) to a high of 93-94% (JMIS and JSIS). The number of articles per journal ranged from 46 (JSIS) and 49 (ISJ and JAIS) to 84 (ISR) and 115 (JMIS). The following sub-sections report the results of the analysis, commencing with the two more general questions, and then moving on to the other four, more specific questions.

4.1 What Researcher Perspectives Are Evident in the AIS 'Basket of 8' Journals?

The data showed strongly skewed distributions. The vast majority of articles reported single-perspective research (97%). Single-perspective research was dominant in all of the journals (range 93-99%). This is examined in detail in sub-section 4.4. The large majority of single-perspective research was concerned with the interests of system sponsors (90%), with only 37 of 528 single-perspective articles adopting the perspective of some other stakeholder (7%). The system sponsor perspective was dominant in all journals (range 82-98%). See Table 5.

Table 5: Perspectives in the Sample

(SS = System Sponsor, O = Other, D = Dual, M = Multi)

4.2 What Changes in Researcher Perspectives Are Evident in the AIS 'Basket of 8' Journals During the Recent Past?

Sub-samples across the eight journals were too small to draw meaningful conclusions at the level of individual journals. Across the complete sample, on the other hand, a change is readily identifiable. Single-perspective studies were dominant in all three years, 2001, 2008 and 2015.

Figure 3: Distribution of Perspectives in 2001, 2008 and 2015

(SS = System Sponsor, O = Other, D = Dual, M = Multi)

The proportion of articles considering exclusively the system sponsor perspective declined somewhat over the time period, from 94% (range 93-100%), to 91% (range 84-100%), to 85% (range 77-94%), and in most of the 8 journals individually. See Figure 4.

Figure 4: System Sponsor Perspective Distribution in 2001, 2008 and 2015

As Table 6 shows, of the 37 articles in which a single perspective other than that of the system sponsor was adopted, 6 were published in 2001 and 7 in 2008, with a leap to 24 in 2015. However, almost 80% of the 2015 articles were in just 3 journals: JMIS (8), ISR (7) and MISQ (4). As regards multi-perspective studies, 6 journals published none whatsoever in the sampled volumes. (The only two that did were EJIS, only in 2008, and JMIS, only in 2015). In the case of dual-perspective studies, 5 journals each published precisely 1 such article (ISJ, JAIS, JIT, JSIS and MISQ).

Table 6: Detailed Perspectives in the Sample

(SS = System Sponsor, O = Other, D = Dual, M = Multi)

4.3 To What Extent Are the Adopted Perspectives Aligned with the Economic, the Social and the Environmental Dimensions?

In most cases, the perspective(s) adopted could be readily coded to a single dimension. On the modest number of occasions where multiple dimensions were evident, preference was given to the dimension other than economic (which was, in all cases, social). The reason for this was the importance of not overlooking any of the relatively small numbers that were on some dimension different from, or additional to, the economic.

The distribution was strongly skewed, in that 93% of articles were on the economic dimension, with only 7% on the social dimension, and not a single article detected on the environmental dimension. Within the individual journals, the range was from a low of 87% economic (ISR) to 96-98% (ISJ, JSIS, JMIS). The details are in Figure 5.

The situation in 2001 was extreme, with 7 journals exhibiting 100% and ISR's 2 articles on the social dimension (out of ISR's total of 17 papers in 2001) bringing the total across all 8 journals down to 98%. The skew was almost as pronounced in 2008 (95% economic). In 2015, however the figure was 89%, with a range across individual journals from 83-85% (JAIS, ISR, MISQ) to 94-95% (EJIS, JIT, JMIS).

Figure 5: Distribution of the Economic Dimension

The proportion on the social dimension was 7%, with a range from 2% (JMIS) to 13% (ISR). The proportion grew from 2% in 2001 (2 articles in ISR), via 5% in 2008 (5 articles in EJIS, 2 each in ISR and JIT and 1 in MISQ), to 11% in 2015 (25 articles, of which 7 were in ISR, 5 in MISQ and 4 in JAIS). See Table 7.

Table 7: Distribution of the Social Dimension

The 37 articles on the social dimensions include the following examples:

4.4 To What Extent are the Studies Single-, Dual- and Multi-Perspective in Nature?

96% of the 548 articles were single-perspective, with only 15 (3%) dual- and 5 (1%) multi-perspective (see Figure 6). Every journal published at least 1 instance of dual-perspective research, although 5 of the 8 published precisely 1 each. Only 2 journals published any multi-perspective research in the sample: EJIS (3) and JMIS (2).

Figure 6: Distribution of Cardinality of Perspective

Examples of dual and multi-perspective research include the following:

EJIS users and the organisation that employs them (Maier et al. 2015)

ISJ mobile phone providers and users (Keith et al. 2015)

ISR individual buyers and sellers (Hinz & Spann 2008)

JAIS manufacturer and retailer (Legner & Schemm 2008)

JIT platform provider and app developers (Oh et al. 2015)

JMIS individuals, organizations and society (Fadel et al. 2015)

JSIS care institution and the elderly (Spagnoletti et al. 2015b)

MISQ service providers and service-disadvantaged (Srivastava & Shainesh 2015)

4.5 To What Extent Does a Balance Exist Between the Perspective of the System Sponsor and the Perspectives of Other Stakeholders?

In addressing Research Question 1, reported on in 4.1 and Figure 3 above, the numbers and proportions of system sponsor perspective articles were distinguished from 'Other Single-Perspective' and 'Greater than One Perspective' articles. The intention of Research Question 5 was to adopt a more fine-grained approach, by examining the sub-categories within the 'Other' and 'Greater Than One' classifications. Because the system sponsor perspective dominates, with 90% of the sample, the numbers in these sub-categories are small: 37 Other and 20 Greater Than One. The proportions of the whole that they represent are very small: 7% and 4%. Across the journals, the dominance of the system sponsor perspective ranged from lows of 82% (JMIS) and 86% (ISR) to highs of 96-98% (JIT, JSIS, ISJ).

The share of System-Sponsor Perspective fell over the 3 years from a high of 94% (2001) to a low of 85% (2015). Other Single-Perspective articles rose from 5% to 11%. On the other hand, Dual- and Multi-Perspective articles never exceeded 5%. The increase in Other Single-Perspective articles from 2008 to 2015 was much more marked in ISR (by 6 articles) and JMIS (by 5) than in the other 6 journals (by 6 articles combined).

In analysing the sub-categories, a first consideration is the extent to which the interests of the system sponsor are represented in 'Greater Than One Perspective' articles. (The authors postulate that, where the system sponsor is intended as the primary beneficiary of the research, higher quality outcomes will be achieved by means of dual- and multi-perspective research than is possible with single-perspective research). Of the 20 such articles, only 6 included user-organisations' interests.

In 7% of the sample (37 articles), a single perspective was adopted which was other than that of the system sponsor. An inspection of these articles resulted in sub-categories being identified (see Table 8).

Table 8: Sub-Categories of the 37 Other Than System Sponsor Single Perspective

In 3% of the sample (15 articles), a dual-perspective approach was adopted; and in 1% of the sample (5 articles), a multi-perspective approach was adopted. Inspection of these articles resulted in sub-categories being identified (see Table 9).

Table 9: Sub-Categories of the 15 Dual-Perspective and 5 Multi-Perspective Articles

In total, 57 articles (10%) were not limited to the system sponsor perspective (37 single-perspective, 15 dual-perspective and 5 multi-perspective). Among these articles, the economic dimension was very strong (38/57 = 67%). Of the 38 articles on the economic dimension, 37% were committed to the seller perspective and 21% to the buyer perspective, while 42% reflected interests on both sides of the market. Of the 19 articles on the Social dimension, 42% considered the interests of individuals, and 58% the interests of various categories of community. A listing of all articles that adopted a perspective other than that of the system sponsor is available in Appendix 3.

4.6 Among Studies Whose Object of Study is People, to What Extent Does a Balance Exist Between the Perspective of the System Sponsor and the Perspectives of Other Stakeholders?

The most common object of study was organisations (59%). Technologies were the focus in 5%. Humans were the object of study in 198 articles (36%). That includes individuals, and informal collectives such as groups and communities. The proportions were considerably different in the three years within the sample, however, with the focus on humans growing from 20% to 30% to 50% (see Figure 7). The change from 2001 to 2008 appeared to reflect the latter part of the rise in consumer eCommerce and the early years of social media. The change from 2008 to 2015 appeared to relate primarily to social media, both as a marketing tool and within the enterprise.

Figure 7: Distribution of Object of Study

Of the 198 articles in which humans were the object of study, the proportion that adopted solely the system sponsor's perspective was 84%, with a range from 72% (JMIS) to 94-95% (ISJ, JSIS, JAIS). Of the 24 journal volumes in the sample, 10 contained not a single article in which a human object of study was researched from other than the perspective of the system sponsor (see Table 10). To deal with the strong skew, the system sponsor data is shown as a percentage of the sample, whereas the other two categories are displayed as counts.

The remaining 31 of the 198 articles were 23 (12%) Other-Single-Perspective and 8 (4%) Dual- and Multi-Perspective. Single-perspective articles that adopted a perspective other than that of the system sponsor numbered only 23 in total - 2 and 3 in 2001 and 2008, but 18 in 2015. The significant increase in 2015 is 80% attributable to 3 journals - JMIS (6), ISR (5) and MISQ (4). Of the other 5 journals, 4 published 1 such article in the sample and 1 published zero articles.

Articles that reflected more than one perspective (i.e. dual- and multi-perspective articles combined) numbered only 8 of the 198. Of these, 4 were in JMIS, 2 in EJIS, and 1 each in ISJ and ISR. There were no dual- or multi-perspective articles in the sampled volumes of JAIS, JIT, JSIS and MISQ. Of these 8 articles, 3 included the system sponsor's perspective, adding a percentage-point to the 84% represented by single-perspective articles.

Of all of the circumstances in which it would seem reasonable to expect dual- and multi-perspective research to be commonplace, the most prominent is in studies of people. The tiny proportion that is evident in the sample (4%), and the dominance of the system sponsor perspective (85%), raise questions about both the ethicality of IS research and, at least to the extent that individuals may have or been able to mobilise market or institutional power, about the extent to which IS research is effectively serving even the needs of system sponsors.

Table 10: Distribution of the 198 Articles Where the Object of Study was Human

(Perspectives: HSS = System Sponsor, HO = Single, Other, HG = Greater than 1)

5. Discussion

In this investigation into the perspectives adopted by researchers, as published in the AIS basket of 8 IS research journals, a number of distinctive patterns have emerged. In this section, these patterns are summarised and some key implications identified. The most significant aspect is that the dominant form of research is single-perspective, reflecting the interests of the system sponsor, and is on the economic dimension. This dominance raises questions about the ethicality of research behaviour. The current situation might be interpreted as a failure to adapt to two very substantial shifts that have occurred over the last 35 years: the application of computing outside organisational boundaries, and the broadening and strengthening of IT's impacts not only on economic but also on social and environmental values. There is a clear need to revisit the calls that have been made over the years for the broadening of the IS discipline's scope.

Commencing with one of the most salient findings, the articles in our sample were overwhelmingly (97%) single-perspective in nature, and very strongly oriented towards the system sponsor (90%). This singularity of purpose seems quite remarkable. Secondly, the economic dimension dominates the sample (93%) with the social dimension visible in a very small (7%) number of articles, and the environmental dimension entirely absent. These findings are alarming, for they suggest that IS researchers, for whatever reason, are either scarcely interested in non-economic, non-system sponsor research perspectives; alternatively, if this research is being undertaken and submitted then it is either published elsewhere or consistently rejected.

In our earlier theoretical discussion, we considered the vexed question of unconscious hegemonic participation, i.e. the tendency for researchers to adopt uncritically and unreflectively a perspective that is consistent with that of the hegemonic stakeholder or power broker. Normatively, we would expect this stakeholder to be evident as the primary financial underwriter and/or as the primary beneficiary of the research. However, the perspective that the researcher adopts (whether consciously or unconsciously) is closely connected with not only the research question(s) that are posed, but also the conception of the research problem and the selection of the phenomena to which the research question(s) pertain. The research perspective is thus intimately related to all other aspects of the research. Indeed, the identity of the primary beneficiary of the research and hence the likely research perspective can be more or less directly inferred from the abstract of a article. Thus, Sabherwal & Chan (ISR) (2001) write this article "examines the impact of alignment on perceived business performance" and Chan & Pan (JSIS) (2008) describe how their article "intends to contribute to the research and practice of e-government systems implementation".

In order to trace the rationale for the selection of perspective, we need to look further back to the motivation to undertake the research in the first place. If it is system sponsor-funded research, with the research question identified in collaboration with that system sponsor, then the logic is clear. However, much research is premised on the existence of so-called gaps in the literature. These gaps should, in principle, be addressable from multiple perspectives and indeed from more than one perspective at a time. Nevertheless, it is almost invariably the case that a) a single perspective is taken, b) this perspective is that of a real or hypothetical system sponsor, i.e. the organisation or industry that will most likely and to the greatest extent benefit from the research, and c) an economic aspect will prevail over social or environmental concerns.

We suggest that there is a significant ethical dimension to the current situation, at least as it pertains to research that purports to provide insights to practitioners. The dominance of system sponsor and economic perspective research has several implications. Firstly, it is clear that other perspectives (e.g., customer, social, environmental) are scarcely reflected in research designs. Secondly, it is likely that non-economic and non-system sponsor perspectives are either wilfully ignored or are acknowledged as existing yet neglected. Thirdly, and as a corollary, the omission of non-economic perspectives in research designs is directly linked to the failure of these neglected perspectives to exert any significant impact on real world practice. Taken together, these three implications firmly entrench the current economic and system sponsor hegemony. Further, the findings suggest that we are failing to grasp the opportunity to learn from other perspectives. Since different stakeholders will have different insights into phenomena, it is essential to ensure that research designs are sensitive to these different perspectives if research outcomes are to offer greater value to all stakeholders. In contrast, single perspective research (whether of dimension or cardinality) will inevitably offer lower quality findings than dual or multiple perspective research because it is in its very nature blind to alternative perspectives.

A further consideration relates to the role of university-based researchers and the funding arrangements that support their research. Some research is industry funded, or funded by research grants that specify industry partners as intended beneficiaries. In such circumstances, taking a systems sponsor perspective is reasonable and appropriate. However, research that is funded by public universities may be expected to contribute to society more generally, since a large proportion of their cash flow comes from the public purse. If this contribution to society does not occur, then the discrepancy between source of funding and identity of the beneficiaries of research looms large: why should industry be the primary beneficiary of government-funded research, with society receiving limited direct benefits? It is for this reason that it is so troubling to see so few articles in our sample premised on a social perspective, and none that considered an environmental perspective. The data suggests that university-based researchers may be essentially setting themselves up as publicly funded business consultants, which should be a matter of great ethical concern for society.

In the current state of affairs, system sponsor oriented research that takes an economic perspective has achieved far more than a simple critical mass, being well represented in research journals across the board. The notion of shifting baseline syndrome could be applied (Pauly, 1995; Ortmann, 2010). This posulates that research norms migrate over time, and that previous 'baseline states' quickly fade away. One reasonable interpretation of the current state of affairs in IS might be that research that favours the system sponsor and adopts an economic perspective is the new norm, and previous baseline states favouring a broader range of perspectives have been largely forgotten.

The opposite interpretation appears, however, to be a better fit to the evidence, i.e. the baseline has failed to shift, despite contextual changes. IS began by studying systems in which data was the raw material and work-in-process, and information was the product. MIS in the USA and Wirtschaftsinformatik (WI) in German-speaking countries began by studying such systems specifically from the viewpoint of managers and executives. All of these threads emerged in the mid-1960s, about a decade after the first applications of computing to administrative data (Clarke 2008, Hirschheim & Klein 2012). The early development of the discipline occurred in parallel with very rapid progress in computing capabilities. Until the mid-1970s, such systems were accessible only by government agencies and corporations of some size. By 1980, end-users within such organisations had become one of the focal points for the IS discipline. Hence, during the discipline's formative years (1965-1985), there was little incentive or scope to consider perspectives other than that of the system sponsor, or dimensions other than the economic.

Gradually, the concept of computer users outside organisations began to make sense, as first ATMs and then EFTPOS systems made progress. IS and computing entirely outside organisations gradually became tenable from about 1980, but only became significant from the early-to-mid-1990s, initially in independent form through the installation of software products on individuals' own devices. This independence was quickly reversed commencing in about 2005, as consumer products were replaced by consumer services.

In short, the IS discipline's foundational paradigm (in the sense popularised by Kuhn 1970) was based on addressing the interests of the system sponsor, was of necessity single-perspective research, and was inherently on the economic dimension. In the last three decades, the discipline has had the opportunity to move beyond that paradigm. Instead, much of the discussion has been about a withdrawal by the discipline even further into its shell. Orlikowski and Iacono (2001) and Benbasat & Zmud (2003) proposed that the IS discipline be defined in terms of 'the IT artefact', effectively removing from scope any non-IT-based activities. Alter (2003) counter-proposed as the core concept the performance of 'IT-reliant work', which would remove from scope those aspects of IS that lie outside working contexts, such as domestic, hedonic and consumer-based arrangements. The long thread of discussion continues. We find it extraordinary that a discipline should consider marginalising elements that are (a) vital parts of real-world systems, and (b) major factors in the success and failure of endeavours to intervene in those real-world systems.

We argue that the continued dominance of the longstanding (system sponsor and economic dimension) paradigm constitutes an ossification of the discipline, because the framework within which IS research is undertaken remains frozen within a context that has long since passed on. As an intellectual community, we have failed to reflect the far broader applications and implications of IT-enabled information systems, and failed to embrace the far larger and more diverse sets of stakeholders. We accordingly support proposals that have been made to adopt a more open interpretation of the role of IS. For example, Clarke (1988), writing at a time when the environmental impacts of IT had yet to gain prominence, called for consideration of economic, legal and social implications to be integrated with research in and the application of IT, not segregated from it. Clarke (ibid, p.519) also argued that "the closing sections of our papers must not be confined to 'implications for further research', but must also directly address 'implications for people'".

Further, DeSanctis (2003) argued for an inclusive approach: "a path toward improvement via boundary enhancement" (p.360). Meanwhile, Galliers (2003) suggested that "an appropriate locus of IS study is more broadly based than organizations or individuals. Societal, policy and ethical issues might reasonably be included within the ambit of the IS field" (p.342). In Nunamaker & Briggs (2012, p.20:1-20:7), the IS discipline was exhorted to "expand our vision to embrace information needs and uses in all kinds of people and teams. Systems exist in a rich milieu of economic, social, political, cognitive, affective, and physical values, and are designed to create value for humans along all these dimensions. Studies of these perceptions of value are therefore also equal in importance to studies of technology".

Casual observation of IS literature suggests that, in recent years, more IS researchers have become interested in both the social dimension and the conduct of dual-perspective studies. A noteworthy example is the MISQ special issue on 'ICT and Societal Challenges'. The special issue editors report in their editorial (Majchrzak et al. 2016) that a large number of papers were submitted and a substantial number accepted for publication. Further, the guest editors suggest that "IS researchers should relabel their practical implications section to policy implications. There is no reason to presume that managers of businesses are the only practitioners who can benefit from our insights" (Majchrzak et al. 2016, p. 275).

If a narrow MIS / Wirtschaftsinformatik worldview is adopted, then the dominance of the system sponsor perspective and the economic dimension might be seen as being natural, appropriate and desirable. On the other hand, even from that narrow worldview, it would not be rational to applaud the dominance of single-perspective research, because dual-perspective and in some cases multi-perspective research can deliver superior quality information to the system sponsor. Adopting a broader worldview, we readily reach the conclusion that the continuing dominance of system-sponsor, single-perspective, economic-dimension research represents not merely a missed opportunity, but a serious flaw in the contemporary IS discipline.

Given the longstanding paradigm's dominance, it is difficult to avoid the conclusion that a hegemony (Wall et al., 2015) exists. IS researchers participate in and are unconsciously and disproportionately influenced by this hegemony as they read prior literature to identify materials to support their arguments. Breaking the shackles of this hegemony requires a dramatic shift in values away from the current, overwhelming focus on the economic perspective towards social and environmental concerns, and away from system sponsors to other legitimate stakeholders such as individuals, teams and non-human entities such as ecosystems and the planet. Within the current mainstream of IS, the likelihood of such a revolution in values seems farfetched, though punctuated equilibrium theory (Eldredge and Gould, 1972; Gersick, 1991) provides some clues as to how it might be achieved. If stable but dysfunctional behaviours, such as the current systems sponsor-economic narrative, are punctuated with radical stimuli, then dramatic changes may ensue. However, the punctuation agent needs to offer a new narrative that is more persuasive than the current one (Monbiot, 2017). In this way, it may be possible to engineer a shift in the baseline (Pauly, 1995) away from the status quo towards a new equilibrium.

Such a new narrative needs to be communicated to, accepted and enacted by a critical mass of early adopters who accept its legitimacy and push for its adoption as the new mainstream narrative. Alternatively, a group of researchers might break away to form a new sub-discipline, most likely one with its own conference(s) and journal(s) that will publish this kind of work. Existing sub-disciplines do exist, for instance under the IFIP umbrella, Technical Committee 9 (ICT and Society) with working groups such as WG9.2 (Social Accountability and Computing), WG9.4 (Social Implications of Computers in Developing Countries), and perhaps most promisingly WG9.9 (ICT and Sustainable Development). For instance, IFIP WG 9.9's website[4] encourages the investigation of "interaction among social, environmental and economic issues in the development of ICTs and their applications". These or other communities of practice (Wenger, 1988), perhaps in conjunction with a special issue organised at one of the more liberally minded journals, may provide the venue where a new sub-discipline can emerge.

Working to the extent feasible within the current paradigm, we suggest that greater value can be offered by IS research if it applies the principle of triangulation not only to data sources and research methods, but also to researcher perspectives. Thus, the economic perspectives need to be complemented by social and environmental perspectives that examine the interests of a variety of stakeholders, not only those of the system sponsors. Such an approach will ensure the realisation of deeper insights by all parties into the needs of all stakeholders, and lead to more effective designs of systems, applications and interventions.

The dominance of single-perspective research, of the interests of the system sponsor, and of the economic dimension, is unhealthy for both society and the IS discipline, and represents a missed opportunity. Like any hegemony, it will be challenging to overcome this mind-set and achieve a shift in the discipline's baseline.

6. Conclusions

In preparing for this research, we noted the dominance of the system sponsor perspective adopted by researchers in IS. We also noted the apparent absence of any notion of researcher perspective from the debates about the 'core' of the IS discipline, stimulated by such senior scholars as Benbasat & Zmud (2003) and Walsham (2012). We found both the dominance and the absence to be remarkable. Why should one perspective dominate all others? Why has this fact escaped the notice of senior scholars in the field and as a result not appeared on the wider research radar? We have suggested reasons for the dominance of the systems sponsor perspective, which are entangled with the current fetish for an economic justification for research, manifested increasingly frequently in econometric analyses of what are only notionally IS problems.

Explaining the absence of debate about the phenomenon is probably fruitless. We suggest that it is more productive to start the debate now and challenge the IS community in general, and journal editors in particular, to increase their awareness of researcher perspective, seeking ways to encourage what can only be described as abnormal perspectives that lie outside the current hegemony, such as those of users, usees, ecosystems and other non-human stakeholders. We need to challenge the current dominance of the economic dimension, since despite its flaws, triple bottom line reporting is already well established and so the legitimacy of non-economic values is already recognised.

Notwithstanding the current dominance of the economic and system sponsor perspectives, we are personally aware of articles that are on the environmental dimension and that do consider beneficiaries at the level of ecosystems, whether local, regional or planetary. However, the absolute number of such studies is tiny and we would not expect that a study of all articles published in the basket of 8 journals since their inception would reveal a very different pattern of results. One practical step that we can take is to ensure that all submitted research articles explicitly identify the implications for practice, society and the environment. Clarke (1988) argued that "Economic, legal and social implications of information technology must all be considered together, to enable the various factors to be seen in perspective". By documenting these different implications, we can raise awareness of their importance more generally and thereby encourage researchers to consider their own research designs more carefully. Future meta-analyses could consider how journals and their editors have approached this challenge.

In order to guide researchers who wish to diversify their research designs yet remain within the mainstream, we offer the following prescriptive advice. We suggest that researcher perspective should be recognised as an important element of research conception, design, conduct and reporting. This will require researchers to deliberate on the alternative perspectives that may be relevant in the specific context under investigation. As a result of that deliberation, researchers will need to determine which perspectives are to be adopted. At various points in an article, but particularly motivation, method, discussion and conclusion, the perspective taken will need to be referred to. Formal structuring of this kind may seem unwieldy and laboured at first, but in time will be accepted as simply part of how research is undertaken and reported. Researchers who have less patience to wait for the mainstream to change may focus more of their attention on sub-communities, such as the various IFIP groups, where the interests and perspectives of multiple stakeholders are more evident. Their departure will be a loss for the IS discipline, so this is an outcome to be avoided.

Appendix 1: Sample Selection Method

This note explains the means whereby the sample was selected.

The Sampling Frame

The sampling frame was defined as the refereed papers published in the IS 'Basket of 8' journals in 2001, 2008 and 2015.

The purpose was to select a 20% sample from that sampling frame.

There were 659 papers in the sampling frame, and hence a notional sample-size of 131.8.

The Sampling Process

Many possible ways exist to apply the "20% sample" notion.

The criteria applied in order to determine an appropriate sampling technique were reliability, reproducibility, pseudo-randomness, understandability and ease of application.

Each journal was considered separately.

Within each Journal:

Hence in each journal, the papers selected were numbers 1, 6, 11, 16, 21, 26, 31, 36, 41, 46, ... 121.

Selecting the first paper in each Journal had the disadvantage of slightly over-sampling the whole.

The significant advantage was, however, that it ensured that at least 20% of the papers in each journal were included.

The Resulting Sample

The sampling frame and the sample are shown in Tables 1 and 2 below.

The smallest number of papers in the sampling frame was 49 (JSIS), and the largest was 124 (JMIS).

The process delivered sample-counts of between 10 (JSIS) and 25 (JMIS), and a total of 135 papers.

Table 1: The Sampling Frame

Table 2: The Sample

Appendix 2: Coding Guidelines

1. Purpose

This document describes the codes applied to articles (section 2), and the process whereby the codes are assigned (section 3).

2. The Codes

The purpose of the coding process is to document the researcher perspective adopted by each article, together with additional information of value to the analysis. This section explains the codes.

* Discipline-Internal

Articles are coded DI if they do not have substantial and direct relevance to any real-world stakeholder, and hence the concept of Researcher Perspective does not apply.

Articles coded DI do not have any other codes assigned.

Examples include papers on:

Development, calibration and adaptation of an artefact, process or method that is for use by researchers falls within this classification. If, on the other hand, the artefact, process or method is for use by a real-world entity (such as a business process analysis technique, or a prototype intended for trial deployment by a business), then it is not DI, and needs to be allocated codes as follows.

* Researcher Perspective

This is the primary focus of the research. It is defined as

"the viewpoint from which phenomena are observed by the researcher"

The concept is distinct from the object of study, although of course a particular entity (such as an organisation, a work-group or a category of individuals such as consumers or employees) may be both the object of study and the stakeholder whose viewpoint is being adopted.

Most commonly, the viewpoint adopted is that of one of the stakeholders in the activities within the research domain.

A particular perspective that is frequently adopted in IS research is that of the System Sponsor, which is defined as

"the organisation (or entity or unit) that develops, implements or adapts a system, process or intervention, [or] causes it to be developed or implemented, or for whose benefit the initiative is undertaken"

Rather than being a single entity, it may be a category of entities (e.g. 'corporations using ERP systems') rather than one specific entity.

Two other circumstances arise, however. One is where the interests that the work addresses are not those of any stakeholder in the phenomena, but instead the purpose is entirely related to academic endeavour. If so, then the article is coded Discipline-Internal (DI), as discussed above.

The other circumstance is where the researcher recognises and addresses the interests of more than one stakeholder.

One such pattern arises where the object of study is a dyad and the researcher actively considers the interests of both sides of that dyad. Examples include buyer and seller in a B2B or C2C marketspace, marketing organisation and consumer, employer and employee, government service delivery agency and client, and government regulatory agency and regulatee.

Another pattern involves more than two stakeholders. For example, in cascaded supply-chains, and in networks of organisations such as those involved in international trade, a study of information system designs to achieve 'win-win-win' outcomes is multi-perspective in nature. Another multi-perspective context is public policy settings, which inherently involve multiple stakeholders with interests that are at least partly in conflict, and in which pareto optimality is difficult or impossible to achieve.

In order to capture all relevant aspects of researcher perspective, two kinds of codes are applied to each article. The first of the two codes records the cardinality. The possible values are:

The second code records the specific perspective(s) adopted. The possible values are:

The system sponsor may be a single corporation, government agency or not-for-profit; or a sub-organisation within one of those entities; or a collective of organisations such as a joint venture, a supply chain, or an industry segment or sector; or a category of organisations (such as organisations that use ERP systems, or organisations that adopt a service such as a particular CRM or on-line office suite or accounting facility)

* Dimension

This documents the nature of the reflected interests. The distinctions to be drawn broadly reflect the 'triple bottom line' and 'corporate social responsibility' literature.

In general, a business enterprise's interests are Economic. Exceptions arise, however, where an action is taken that potentially compromises the normal business objectives of profit, revenue and/or market share. One example is affirmative action employment on the basis of gender, ethnicity, socio-economic deprivation or disability. Another is pro bono work, such as that undertaken by a law firm.

The interests of government agencies, not-for-profit organisations and individuals are also in many cases on the Economic dimension, where the objectives relate to operational efficiency, or effective achievement of an economic objective. Rather more exceptions arise in the case of these categories of organisations, however, because the purpose of their mainstream activities is commonly a social or environmental service or outcome.

Where a paper examines how social intentions or activities fulfil an economic purpose, it should be coded as Economic.

Where a paper exhibits both Economic and Social dimensions (e.g. a Dual Perspective paper that considers not only pricing issues but also consumer wellbeing, e.g. privacy or equity), then it should be coded as Social, because the intention is to distinguish papers that have at least a material degree of orientation to other-than-economic interests.

The possible codes and associated keywords are:

Trust is most likely Economic (e.g. trust in a supplier), but in some cases may be Social (e.g. trust in an online counsellor). Gender may be Economic (e.g. differences in workplace performance) or Social (e.g. differences in personal responses to stress). Communities may be Economic (e.g. training for the workplace) or Social (e.g. services to the physically disabled).

* Object of Study

This refers to the entity about which research is performed.

The possible codes are:

The level of abstraction or 'unit of study' is not relevant to the coding.

Hence Organisation includes all sizes of formal organisation, and any kind of organisational sub-unit, such as department, section or team. It also includes organisational roles that make decisions on behalf of formal organisations, including for example CEO and CIO. This code is also applied where the object of study comprises more than one organisation, as occurs with dyads (inter-organisational systems), and chains and networks (multi-organisational systems).

The Human category includes not only individuals, but also social collectives of small, medium or large scale (i.e. groups, communities and societies) that are formed on any basis of some commonality (e.g. of interest, language, belief or location), but excludes humans operating on behalf of formal organisations (e.g. corporations, business partnerships, incorporated associations). Hence, in a study of the behaviour of CIOs in relation to decisions about business, technology or management, the Object of Study is Organisation not Human. In a study of employee compliance with organisational policy, on the other hand, the Object of Study is Human.

Technology includes information systems, elements of them, software tools, user interfaces, and hardware, but also methods and processes. This code is applied where the focus is specifically on the technology. If the focus is on the strategic or operational use of technology by an organisation, it should be coded 'Org'. If the focus is on the operational use of technology by an individual, the paper should be coded as 'H'.

* Type of Research

This refers to the approach adopted by the researcher.

The possible codes are:

The reason for coding articles in this manner is that the purpose of the research is expressed differently in each case, in particular:

Of course, many articles embody elements of two or more of these approaches, e.g.:

Of particular significance in distinguishing between theoretical and empirical research is the degree of significance of the empirical component. In empirical research, the purpose of the empirical component is to answer the research question or test the hypotheses. In theoretical research, the purpose of any empirical component is to illustrate the application of the theory.

All forms of model-building, and simulation using artificial data (including Monte Carlo simulation even if the weighted random selections draw on empirical data), are treated here as theoretical not empirical.

* Researcher Perspective where the Object of Study is Human

This relates to an aspect of particular relevance to the research that this coding is intended to support. It is a derivative code, based on the Researcher Perspective and the Object of Study.

The possible codes are:

3. The Process

There are several possible modes in which the coding may be performed:

The process commences by examining the target text for relevant passages. The term 'passage' is used in its conventional sense, to refer to a text-segment of any length, including word, word-group. phrase, clause, sentence, paragraph or paragraphs. (For example, the Macquarie Dictionary defines a passage as "an indefinite portion of a writing, speech, or the like, usu. one of no great length ...").

In the case of title and abstract only examination, the whole of those texts is carefully examined. To facilitate validation within the research team, and to enable replication and audit, it is beneficial to highlight key passages that influenced the characterisation of the article.

Where the entire article is considered, the title, abstract, introduction, implications and conclusions sections are commonly the most informative sources of insight into the nature of the article for the purposes at hand. In this case, it is preferable to extract the key passages into the coding sheets.

Codes can be assigned in whichever order supporting evidence is gathered. However, it will commonly be most helpful to categorise in the following sequence:

Where any material uncertainty remains, highlight the entry to indicate the probable need for detailed discussion among the members of the research team.

Appendix 3: List of Articles with a Perspective other than System Sponsor

Single-Perspective, Other-Than-System-Sponsor - O (6, 7, 24)

EJIS (0, 0, 1)

Zhou, 24, 3, 247-261 - Second Life participants

ISJ (0, 0, 0)
ISR (1, 1, 7)

Butler, 12, 4, 346-362 - groups of people

Bapna, 19, 4, 400-416 - consumers in C2C

Johnson, 26, 1, 165-187 - online communities

Sabnis, 26, 2, 301-319 - marketers

Kourandi, 26, 2, 320-338 - content consumers

Yan, 26, 3, 496-512 - patients

Chen, 26, 3, 513-531 - musicians

Krasnova, 26, 3, 585-605 - social media users

Choi, 26, 4, 675-694 - SNS users

JAIS (0, 1, 3)

Zhang, 9, 2, 73-94 - online communiities

Ling, 16, 3, 174-212 - communities during crisis response

Bock, 16, 6, 418-447 - virtual communities

Chiu, 16, 11, 947-979 - members of virtual communities

JIT (1, 0, 0)

Howcroft, 16, 4, 195-204 - investors

JMIS (4, 3, 8)

Tam, 17, 4, 97-124 - IT vendors

Aron, 18, 2, 65-88 - Sellers of information goods (in e-markets)

Gundepudi, 18, 2, 107-131 - Sellers of information goods (in e-markets)

Kauffman, 18, 2, 157-188 - Online sellers

Lau, 25, 1, 79-104 - traders in e-markets

Wu, 25, 1, 167-198 - monopoly manufacturer

Dedrick, 25, 2, 41-72 - online sellers

Abbasi, 31, 4, 109-157 - users in household and enterprise settings

Adomavicius, 31, 4, 182-212 - sellers of digital goods in digital markets

Johnson, 31, 4, 311-342 - participants in negotiations

Zhao, 32, 1, 261-290 - sellers in e-markets

Dissanayake, 32, 2, 8-39 - virtual team members

Clemons, 32, 2, 40-70 - users of educational applications

Berger, 32, 3, 105-128 - content providers

Lee, 32, 4, 109-143 - sellers over social commerce systems

JSIS (0, 0, 1)

Gleasure, 24, 4, 219-233 - (non-)users of crowdfunding

MISQ (0, 2, 4)

Hsieh, 32, 1, 97-126 - the socio-economically disadvantaged

Gregg, 32, 3, 653-670 - sellers using eBay

Faraj, 39, 2, 393-412 - online community leaders

Gao, 39, 3, 565-589 - consumers

Lin, 39, 3, 697-712 - Taiwanese aboriginals (usees)

Singh, 39, 4, 887-908 - investors in IT startups

Single-Perspective, Other-Than-System-Sponsor - O (6, 7, 24)

Sellers in Online Markets, incl. Prosumers (14)

Sabnis - ISR 26, 2, 301-319 - marketers

Chen - ISR 26, 3, 513-531 - musicians

Tam - JMIS 17, 4, 97-124 - IT vendors

Aron - JMIS 18, 2, 65-88 - Sellers of information goods (in e-markets)

Gundepudi - JMIS 18, 2, 107-131 - Sellers of information goods (in e-markets)

Kauffman - JMIS 18, 2, 157-188 - Online sellers

Lau - JMIS 25, 1, 79-104 - traders in e-markets

Wu - JMIS 25, 1, 167-198 - monopoly manufacturer

Dedrick - JMIS 25, 2, 41-72 - online sellers

Adomavicius - JMIS 31, 4, 182-212 - sellers of digital goods in digital markets

Zhao - JMIS 32, 1, 261-290 - sellers in e-markets

Berger - JMIS 32, 3, 105-128 - content providers

Lee - JMIS 32, 4, 109-143 - sellers over social commerce systems

Gregg - MISQ 32, 3, 653-670 - sellers using eBay

Buyers in Online Markets, incl. Content Consumers, Patients, Investors (8)

Bapna - ISR 19, 4, 400-416 - consumers in C2C

Kourandi - ISR 26, 2, 320-338 - content consumers

Yan - ISR 26, 3, 496-512 - patients

Howcroft - JIT 16, 4, 195-204 - investors

Abbasi - JMIS 31, 4, 109-157 - users in household and enterprise settings

Gleasure - JSIS 24, 4, 219-233 - (non-)users of crowdfunding

Gao - MISQ 39, 3, 565-589 - consumers

Singh - MISQ 39, 4, 887-908 - investors in IT startups

Online Community Participants (8)

Zhou - EJIS 24, 3, 247-261 - Second Life participants

Krasnova _ ISR 26, 3, 585-605 - social media users

Choi - ISR 6, 4, 675-694 - SNS users

Chiu - JAIS 16, 11, 947-979 - members of virtual communities

Johnson - JMIS 31, 4, 311-342 - participants in negotiations

Dissanayake - JMIS 32, 2, 8-39 - virtual team members

Clemons - JMIS 32, 2, 40-70 - users of educational applications

Faraj - MISQ 39, 2, 393-412 - online community leaders

Online Communities (4)

Butler - ISR 12, 4, 346-362 - groups of people

Johnson - ISR 26, 1, 165-187 - online communities

Zhang - JAIS 9, 2, 73-94 - online communiities

Bock - JAIS 16, 6, 418-447 - virtual communities

Communities (3)

Ling - JAIS 16, 3, 174-212 - communities during crisis response

Hsieh - MISQ 32, 1, 97-126 - the socio-economically disadvantaged

Lin - MISQ 39, 3, 697-712 - Taiwanese aboriginals (usees)

Dual-Perspective - DP (1, 7, 7)

Underlining identifies instances where the system sponsor is included (6)

EJIS (0, 2, 1)

Kietzmann, 17, 3, 305-320 - manufacturers, their org'l clients and their respective mobile workers

Weedman, 17, 5, 476-488 - Users (earth science staff) and IS staff (CS researchers)

Maier, 24, 5, 447-464 - organizations, SNS providers, and SNS users

ISJ (0, 0, 1)

Keith, 25, 6, 637-667 - mobile phone providers and users

ISR (0, 1, 2)

Hinz, 19, 3, 351-368 - buyers and sellers

Ravindran, 26, 2, 379-397 - outsourcers and contractors

Chen, 26, 4, 754-772 - advertisers and advertising service providers

JAIS (0, 1, 0)

Legner, 9, 3/4, 119-150 - manufacturer and retailer

JIT (0, 0, 1)

Oh, 30, 3, 245-259 - platform provider and app developers

JMIS (1, 3, 0)

Jarvenpaa, 18, 1, 151-183 - systems sponsor and employed knowledge workers

Abbasi, 25, 1, 49-78 - buyers and sellers

Zhu, 25, 1, 199-232 - database creators and users

Bolton, 25, 2, 145-170 - market operators and traders

JSIS (0, 0, 1)

Spagnoletti, 24, 2, 128-145 - care institution and the elderly

MISQ (0, 0, 1)

Srivastava, 39, 1, 245-267 - service providers, service-disadvantaged segments of society

Economic Dimension

Kietzmann - EJIS 17, 3, 305-320 - manufacturers and clients

Weedman - EJIS 17, 5, 476-488 - Users and IS staff

Maier - EJIS 24, 5, 447-464 - SNS providers and users

Keith - ISJ 25, 6, 637-667 - mobile phone providers and users

Hinz - ISR 19, 3, 351-368 - buyers and sellers

Ravindran - ISR 26, 2, 379-397 - outsourcers and contractors

Chen - ISR 26, 4, 754-772 - advertisers and advertising service providers

Legner - JAIS 9, 3/4, 119-150 - manufacturer and retailer

Oh - JIT 30, 3, 245-259 - platform provider and app developers

Jarvenpaa - JMIS 18, 1, 151-183 - systems sponsor and employed knowledge workers

Abbasi, 25 - JMIS 1, 49-78 - buyers and sellers

Zhu - JMIS 25, 1, 199-232 -- database creators and users

Bolton - JMIS 25, 2, 145-170 - market operators and traders

Spagnoletti - JSIS 24, 2, 128-145 - care institution and the elderly

Social Dimension

Srivastava - MISQ 39, 1, 245-267 - service providers, service-disadvantaged segments of society

Multi-Perspective - MP (0, 3, 2)

Underlining identifies instances where the system sponsor is included (0)

EJIS (0, 3, 0)

Cordoba, 17, 2, 125-142 - family, community and other concerns

Gal, 17, 3, 290-304 - multiple organisations

Whitley, 17, 6, 668-677 - political debate

ISJ (0, 0, 0)
ISR (0, 0, 0)
JAIS (0, 0, 0)
JIT (0, 0, 0)
JMIS (0, 0, 2)

Chatterjee, 31, 4, 49-87 - individuals, organisations and society

Matook, 31, 4, 278-310 - employers, education providers, users

JSIS (0, 0, 0)
MISQ (0, 0, 0)

Economic Dimension

Gal - EJIS 17, 3, 290-304 - multiple organisations

Matook - JMIS 31, 4, 278-310 - employers, education providers, users

Social Dimension

Cordoba - EJIS 17, 2, 125-142 - family, community and other concerns

Whitley - EJIS 17, 6, 668-677 - political debate

Chatterjee - JMIS 31, 4, 49-87 - individuals, organisations and society

Supplementary Materials 1: Detailed Coding Sheets

Available at

Supplementary Materials 2: Sub-Sample Coding Sheets

Available at


Achterkamp, M.C. & Vos, J.F.J. 2008. "Investigating the Use of the Stakeholder Notion in Project Management Literature: A Meta-Analysis," International Journal of Project Management (26), pp. 749-757.

Agarwal, N., Lim, M. & Wigand, R. 2012."Raising and Rising Voices in Social Media: A Novel Methodological Approach in Studying Cyber-Collective Movements," Business & Information Systems Engineering (4:3), pp. 113-126.

Alter, S. 2003. "Sidestepping the IT Artifact, Scrapping the IS Silo, and Laying Claim to 'Systems in Organizations'," Communications of the AIS (12), pp. 494-526.

Avison, D.E. & Wood-Harper, A.T. 1990. Multiview: An Exploration in Information Systems Development, Blackwell.

Avison, D.E., Baskerville, R. & Myers, M. 2001. "Controlling Action Research Projects," Information Technology & People (14:1), pp. 28-45.

Baskerville, R.L. & Wood-Harper, A.T. 1996. "A Critical Perspective on Action Research as a Method for Information Systems Research," Journal of Information Technology (11), pp. 235-246.

Baskerville, R.L. 1999. "Investigating Information Systems with Action Research" Communications of the AIS (2:19).

Baumer, E.P.S. 2015. "Usees," Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (CHI'15), April.

Benbasat, I. & Zmud, R.W. 2003. "The Identity Crisis Within the IS Discipline: Defining and Communicating the Discipline's Core Properties," MIS Quarterly (27:2), pp. 183-194.

Butler, B.S. 2001. "Membership Size, Communication Activity, and Sustainability: A Resource-Based Model of Online Social Structures," Information Systems Research (12:4), pp. 346-362.

Cameron, J. & Clarke, R. 1996. "Towards a Theoretical Framework for Collaborative Electronic Commerce Projects involving Small and Medium-Sized Enterprises," Proceedings of the 9th International Conference on EDI-IOS, Bled, Slovenia, 10-12 June.

Cecez-Kecmanovic, D. 2001. "Doing Critical IS Research: The Question of Methodology," Chapter VI in 'Qualitative Research in IS: Issues and Trends: Issues and Trends' (ed. Trauth E.M.), pp. 141-163, Idea Group Publishing, 2001, at

Handbook of Critical Information Systems Research: Theory and Application, pp.19-27, Edward Elgar.

Cecez-Kecmanovic, D., Klein, H.K. & Brooke, C. 2008. "Exploring the Critical Agenda in Information Systems Research," Information Systems Journal (18:2), pp. 123-135.

Chan, C.M.L. & Pan, S.L. 2008. "User Engagement in E-government Systems Implementation: A Comparative Case Study of Two Singaporean E-government Initiatives," Journal of Strategic Information Systems (17:2), pp. 124-139.

Checkland, P. 1981. Systems Thinking, Systems Practice, Wiley.

Clarke, R. 1988. "Economic, Legal and Social Implications of Information Technology," MIS Quarterly (12:4), p. 517-519.

Clarke, R. 1992. "Extra-Organisational Systems: A Challenge to the Software Engineering Paradigm," Proceedings of the IFIP World Congress, Madrid, September.

Clarke, R. 2008. "A Retrospective on the Information Systems Discipline in Australia," Chapter 2 of Gable G.G., Gregor, S., Clarke, R., Ridley, G. and Smyth, R. (Eds.) (2008) The Information Systems Discipline in Australia, ANU e-Press, 47-107.

Constantinides, P., Chiasson, M.W. & Introna, L.D.(2012. "The Ends of Information Systems Research: A Pragmatic Framework," MIS Quarterly (36:1), pp. 1-19

Davis, F.D. 1989. "Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology," MIS Quarterly (13:3), pp. 319-340.

Davison, R.M., Martinsons, M.G. & Kock, N. 2004. "Principles of Canonical Action Research," Information Systems Journal (14:1), pp. 65-86.

Davison, R.M., Martinsons, M.G. and Ou, C.X.J. 2012. "The Roles of Theory in Canonical Action Research," MIS Quarterly (36:3), pp. 763-786.

DeLone, W.H. & McLean, E.R. 1992. "Information Systems Success: The Quest for the Dependent Variable," Information Systems Research (3:1), pp. 60-95.

Deng, Q., Wang, Y. and Ji, S.B. 2015. "Why Organizations Adopt Green IT: A Comprehensive Review," Proceedings of CONF-IRM, 40

DeSanctis, G. 2003. "The Social Life of Information Systems Research: A Response to Benbasat and Zmud's Call for Returning to the IT Artifact," Journal of the Association for Information Systems (4:1), 16.

Eldredge, N. and Gould, S.S.J. 1972. "Punctuated Equilibria: An Alternative to Phyletic Gradualism," in Schopf, T.J.M. (Ed.), Models in Paleobiology. San Francisco: Freeman Cooper. pp. 82-115.

Elkington, J. 1994. "Towards the Sustainable Corporation: Win-Win-Win Business Strategies for Sustainable Development," California Management Review (36:2), pp. 90-100.

Elliot, S. 2011. "Transdisciplinary Perspectives on Environmental Sustainability: A Resource Base and Framework for IT-Enabled Business Transformation," MIS Quarterly (35:1), pp. 197-236.

Emery, F.E. & Trist, E.L. 1960. "Socio-technical Systems," in: Churchman, C.W. and Verhurst, M. (Eds.) Management Sciences Models and Techniques, London: Pergamon, Vol. 2, pp. 83-97.

Fadel, K.J., Meservy, T.O. & Jensen, M.L. 2015. "Exploring Knowledge Filtering Processes in Electronic Networks of Practice," Journal of Management Information Systems (31:4), pp. 158-181.

Fayard, A.-L. & DeSanctis, G. 2008. "Kiosks, Clubs and Neighborhoods: The Language Games of Online Forums," Journal of the Association for Information Systems (9:10/11), pp. 677-705.

Fischer-Hübner, S. & Lindskog, H. 2001. "Teaching Privacy-Enhancing Technologies," Proceedings of the IFIP WG 11.8 2nd World Conference on Information Security Education, Perth, Australia.

Fletcher, K. 2015. "Consumer Power and Privacy: The Changing Nature of CRM," International Journal of Advertising (22:2), pp. 249-272.

Freeman, R.E. & Reed, D.L. 1983. "Stockholders and Stakeholders: A New Perspective on Corporate Governance," California Management Review (25:3), pp. 88-106.

Galliers, R.D. 2003. "Change as Crisis or Growth? Toward a Trans-disciplinary View of Information Systems as a Field of Study: A Response to Benbasat and Zmud's Call for Returning to the IT Artifact," Journal of the Association for Information Systems (4:6), pp. 337-351.

Gersick, C.J.G. 1991. "Revolutionary Change Theories: A Multilevel Exploration of the Punctuated Equilibrium Paradigm," Academy of Management Review (16:1), pp. 10-36.

Gholami, R., Watson, R.T., Hasan, H., Molla, A. & Bjørn-Andersen, N. 2016. "Information Systems Solutions for Environmental Sustainability: How Can We Do More?," Journal of the Association for Information Systems (17:8), pp. 521-536.

Grover, V., and Lyytinen, K. 2015. "New State of Play in Information Systems Research: The Push to the Edges," MIS Quarterly (39:2), pp. 271-296.

Harrison, T.M. and Zappen, J.P. 2013. "Methodological and Theoretical Frameworks for the Design of Community Information Systems," Journal of Computer Mediated Communication, (8:3), at

Hedman, J. and Henningsson, S. 2016. "Developing Ecological Sustainability: A Green IS Response Model," Information Systems Journal (26:3), pp. 259-287.

Heeks, R. 2002. "Information Systems and Developing Countries: Failure, Success, and Local Improvisations," The Information Society (18:10), pp. 1-112.

Hevner, A.R., March, S.T. & Park, S. 2004. "Design Science in Information Systems Research," MIS Quarterly (28:1), pp. 75-105.

Hinz, O. & Spann, M. 2008. "The Impact of Information Diffusion on Bidding Behavior in Secret Reserve Price Auctions," Information Systems Research (19:3), pp. 351-368

Hirschheim, R. & Klein, H.K. 2012. "A Glorious and Not-So-Short History of the Information Systems Field," Communications of the AIS (13:4), pp. 188-235.

Keith, M.J., Babb, J.S., Lowry, P.B., Furner, C.P. & Abdulatt, A. 2015. "The Role of Mobile-Computing Self-Efficacy in Consumer Information Disclosure," Information Systems Journal (25:6), pp. 637-667.

Klein H. & Myers, M. 1999. "A Set of Principles for Conducting and Evaluating Interpretive Field Studies in Information Systems," MIS Quarterly (23:1), pp. 67-93.

Kuhn, T.S. 1970. The Structure of Scientific Revolution, 2nd Edition, University of Chicago Press: Chicago.

Land, F. & Hirschheim, R. 1983. "Participative Systems Design: Rationale, Tools and Techniques," Journal of Applied Systems Analysis (10), pp. 91-107.

Ledwith, M. 2016. "Emancipatory Action Research as a Critical Living Praxis: From Dominant Narratives to Counternarratives," in: Rowell, L.L., Bruce, C.D., Shosh, J.M. and Riel, M.M., (eds.) The Palgrave International Handbook of Action Research. Palgrave Macmillan, New York, USA, pp. 49-62.

Legner, C. & Schemm, J. 2008. "Toward the Inter-organizational Product Information Supply Chain: Evidence from the Retail and Consumer Goods Industries," Journal of the Association for Information Systems (9:3/4), pp. 119-150.

Lin, C.C., Kuo, F.-Y. & Myers, M.D. 2015. "Extending ICT4D Studies: The Value of Critical Research," MIS Quarterly (39:3), pp. 697-712.

Lyytinen, K. & Hirschheim, R. 1987. "Information Systems Failures: A Survey and Classification of the Empirical Literature," in Southon G., Sauer C. & Dampney K., Oxford Surveys in Information Technology Vol 4, pp. 257-309.

Maier, C., Laumer, S., Eckhardt, A. & Weitzel, T. 2015. "Giving Too Much Social Support: Social Overload on Social Networking Sites," European Journal of Information Systems (24:5), pp. 447-464.

Majchrzak, A., Markus, M.L. & Wareham, J. 2016. "Designing for Digital Transformation: Lessons for Information Systems Research from the Study of ICT and Societal Challenges," MIS Quarterly (40:2), pp. 267-277.

Matook, S., Cummings, J. & Bala, H. 2015. "Are You Feeling Lonely? The Impact of Relationship Characteristics and Online Social Network Features on Loneliness," Journal of Management Information Systems (31:4), pp. 278-310.

McGrath, K. 2005. "Doing Critical Research in Information Systems: A Case of Theory and Practice Not Informing Each Other," Information Systems Journal (15:2), pp. 85-101.

Milne, M.J. and Gray, R. 2013. "W(h)ither Ecology? The Triple Bottom Line, the Global Reporting Initiative, and Corporate Sustainability Reporting," Journal of Business Ethics (118:1), pp. 13-29.

Mitchell, R.K., Agle, B.R. & Wood, D.J. 1997. "Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts," Academy of Management Review (22:4), pp. 853-886

Mitsilegas, V. 2015. "The Transformation of Privacy in an Era of Pre-emptive Surveillance," Tilburg Law Review (20:1), pp. 35-57.

Monbiot, G. 2017. Out of the Wreckage: A New Politics for an Age of Crisis, Verso: London.

Mumford, E. 2000. "A Socio-Technical Approach to Systems Design," Requirements Engineering (5:2), pp. 125-133

Muzellec, L., Ronteau, S. and Lambkin, M. 2015. "Two-sided Internet Platforms: A Business Model Lifecycle Perspective," Industrial Marketing Management (45), pp. 139-150.

Myers, M.D. & Klein, H.K. 2011. "A Set of Principles for Conducting Critical Research in Information Systems," MIS Quarterly (35:1), pp. 17-36.

Narsing, A. 2005. "RFID and Supply Chain Management: An Assessment of its Economic, Technical, and Productive Viability in Global Operations," Journal of Applied Business Research (21:2), pp. 75-80.

Nunamaker, J.F. & Briggs, R.O. 2012. "Toward a Broader Vision for Information Systems," ACM Transactions on Management Information Systems (2:4), pp. 20:1-20:12.

Oh, J., Koh, B. & Raghunathan, S. 2015. "Value Appropriation between the Platform Provider and App Developers in Mobile Platform Mediated Networks," Journal of Information Technology (30:3), pp. 245-259.

Olesen, K. & Myers, M.D. 1999. "Trying to Improve Communication and Collaboration with Information Technology: An Action Research Project which Failed," Information Technology and People (12:4), pp. 317-332.

Orlikowski, W.J. & Baroudi, J.J. 1991. "Studying Information Technology in Organizations: Research Approaches and Assumptions," Information Systems Research (2:1), pp. 1-28.

Orlikowski, W. & Iacono, C.S. 2001. "Research Commentary: Desperately Seeking the 'IT' in IT Research - A Call to Theorizing the IT Artifact," Information Systems Research (12:2), pp. 121-134.

Ortmann, H. 2010. "On Drifting Rules and Standards," Scandinavian Journal of Management (26), pp. 204-214.

Pauly, D. 1995. "Anecdotes and the Shifting Baseline Syndrome of Fisheries," Trends in Ecology and Evolution (10:10), p. 430.

Pouloudi, A. & Whitley, E.A. 1997. "Stakeholder Identification in Inter-Organizational Systems: Gaining Insights for Drug Use Management Systems," European Journal of Information Systems (6:1), pp. 1-14.

Sabherwal, R. & Chan, Y. 2001. "Alignment between Business and IS Strategies: A Study of Prospectors, Analysers and Defenders," Information Systems Research (12:1), pp. 11-33.

Sandeep, M.S. & Ravishankar, M.N. 2015. "Social Innovations in Outsourcing: An Empirical Investigation of Impact Sourcing Companies in India," Journal of Strategic Information Systems (24:4), pp. 270-288.

Sauer, C. 1993. Why Information Systems Fail: A Case Study Approach. Henley, UK: Alfred Waller.

Seddon, P.B., Staples, S., Patnayakuni, R. & Bowtell, M. 1999. "Dimensions of Information Systems Success," Communications of the AIS (2:20).

Selander, L. & Jarvenpaa, S. 2016. "Digital Action Repertoires and Transforming a Social Movement Organization," MIS Quarterly (40:2), pp. 331-352.

Son, J.Y. & Kim, S.S. 2008. "Internet Users' Information Privacy-Protective Responses: A Taxonomy and a Nomological Model," MIS Quarterly (32:3), pp. 503-529.

Spagnoletti, P., Resca, A. & Lee, G. 2015a. "A Design Theory for Digital Platforms Supporting Online Communities: A Multiple Case Study," Journal of Information Technology (30:4), pp. 364-380.

Spagnoletti, P., Resca, A. & Saebo, O. 2015b. "Design for Social Media Engagement: Insights from Elderly Care Assistance," Journal of Strategic Information Systems (24:2), pp. 128-145.

Srivastava, S.C. & Shainesh, G. 2015. "Bridging the Service Divide Through Digitally Enabled Service Innovations: Evidence from Indian Healthcare Service Providers," MIS Quarterly (39:1), pp. 245-267.

Stahl, B.C., Doherty, N.F. and Shaw, M. 2012. "Information Security Policies in the UK Healthcare Sector: A Critical Evaluation," Information Systems Journal (22:1), pp. 77-94.

Tarafdar, M. and Davison, R.M. 2018. "Research in Information Systems: Intra-disciplinary and Inter-disciplinary Approaches," Journal of the Association for Information Systems - forthcoming.

Wall, J.D., Stahl, B.C. and Salam, A.F. 2015. "Critical Discourse Analysis as a Review Methodology: An Empirical Example," Communications of the AIS (37:11), pp. 257-285.

Walsham, G. 2012. "Are We Making a Better World with ICTs? Reflections on a Future Agenda for the IS Field," Journal of Information Technology (27:2), pp. 87-93.

Watson, R.T., Boudreau, M.-T. & Chen, A.J. 2010. "Information Systems and Environmentally Sustainable Development: Energy Informatics and New Directions for the IS Community," MIS Quarterly (34:1), pp. 23-38.

Wenger, E. 1998. Communities of Practice: Learning, Meaning and Identity, Cambridge: Cambridge University Press.

Wong, L.H.M. and Davison, R.M. 2018. "Knowledge Sharing in a Global Logistics Provider: An Action Research Project," Information & Management



Author Affiliations

Roger Clarke is Principal of Xamax Consultancy Pty Ltd, Canberra. He is also a Visiting Professor in Cyberspace Law & Policy at the University of N.S.W., and a Visiting Professor in the Research School of Computer Science at the Australian National University.

Robert Davison is a Professor of Information Systems at City University of Hong Kong. He is Editor-in-Chief of both the Information Systems Journal and the Electronic Journal of Information Systems in Developing Countries, and Chair of IFIP WG 9.4.

Wanjing Yia holds an M.Sc. from the City University of Hong Kong in Electronic Business and Knowledge Management and a B.Mngt from Shandong University of Science and Technology.

xamaxsmall.gif missing
The content and infrastructure for these community service pages are provided by Roger Clarke through his consultancy company, Xamax.

From the site's beginnings in August 1994 until February 2009, the infrastructure was provided by the Australian National University. During that time, the site accumulated close to 30 million hits. It passed 65 million in early 2021.

Sponsored by the Gallery, Bunhybee Grasslands, the extended Clarke Family, Knights of the Spatchcock and their drummer
Xamax Consultancy Pty Ltd
ACN: 002 360 456
78 Sidaway St, Chapman ACT 2611 AUSTRALIA
Tel: +61 2 6288 6916

Created: 17 November 2016 - Last Amended: 7 March 2018 by Roger Clarke - Site Last Verified: 15 February 2009
This document is at
Mail to Webmaster   -    © Xamax Consultancy Pty Ltd, 1995-2022   -    Privacy Policy