Roger Clarke's Web-Site

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

Roger Clarke's 'Researcher Perspective in B08 Journals'

Researcher Perspective in the IS Discipline:
An Empirical Study of Articles in the Basket of 8 Journals

Version of 9 July 2020

Published in Information Technology & People 33, 6 (October 2020) 1515-1541
DOI: 10.1108/ITP-04-2019-0189

Roger Clarke, Robert M Davison & Wanying Jia **

© Xamax Consultancy Pty Ltd, 2018-20

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

This document is at


When observing phenomena, IS researchers mostly adopt the perspective of one of the stakeholders in the situation studied, very often that of the sponsor of the information system that is in focus. This article reports on an investigation of researcher perspectives adopted in published IS research. We sampled 659 articles published in AIS Basket journals in the years 2001, 2008 and 2015. We found that 96% of relevant articles adopted a single-perspective approach, and 93% of those were oriented towards the system sponsor. These are significant findings, because they enable the identification of opportunities for the discipline to deliver value through the adoption of the perspectives of additional stakeholders, and through dual-perspective and multi-perspective research. This has potentially far-reaching implications for the nature of the IS discipline.

1. Introduction

The phenomena studied by researchers within the information systems (IS) discipline are complex, and involve multiple actors. Sometimes those actors view the phenomena in much the same way. In most circumstances, however, their attitudes exhibit material differences. For example, managers may consider organisational policies and procedures to be the key vehicles for ensuring reliability and consistency in the performance of tasks; on the other hand, employees who engage in front-line work may regard them merely as guidelines that require reflexive interpretation as they cope with the wide variety of circumstances that naturally characterise the workplace.

Since interpretivism became mainstream within the discipline, it has been conventional to recognise that phenomena can be viewed from multiple perspectives, and that a comprehensive understanding of a phenomenon requires recognition that the different points-of-view of key actors need to be taken into account. This is illustrated by an Indian parable over two millennia old, in which several blind men have widely-varying interpretations of an elephant, because the perception of each of the observers is based on their tactile experience with a different part of the animal. Similarly, the Chinese expression Guan Zhong Kui Bao means 'to see a leopard through a narrow tube', and hence to have a limited view of something.

We contend that the person or team conducting a research project adopts the perspective of one or more of the stakeholders in the activity that they are observing. We use the term 'researcher perspective' to refer to this phenomenon. A research project may involve only a single perspective. This would be the case, for example, where the following research question is adopted:

What measures are most effective at achieving staff compliance with security policies?

A research project to address that question would be likely to adopt the perspective of the Chief Security Officer (CSO), on behalf of the organisation as a whole. Other projects adopt dual and even multiple perspectives, as for example where a research question such as the following is addressed:

Are Internet Services Providers' perceptions of their quality of service consistent with those of their customers?

In devising a research design to address this question, a researcher would inevitably adopt the perspectives of both ISPs and customers of ISPs. In the resulting report, the researcher may well prioritise the interests of the ISP, which would then be the primary beneficiary; but the depth of consideration of the customers' interests might be sufficient that customers could be reasonably regarded as co-beneficiaries of the research effort.

Authors may or may not make clear to their readers what stakeholder perspective they have adopted. Indeed, researchers may or may not intentionally choose a perspective, and may or may not even be aware of the fact that their focus is on one stakeholder's interests rather than another's. The viewpoint may be subliminal, habitual, or embedded in the research genre that the researcher is working within, such that they do not reflect on the interests they are prioritising.

One of the authors has previously conducted several exploratory studies of the researcher perspectives adopted in papers in a number of venues in which IS research is published. Both authors have been active in ethical issues, and one has served as chair of the AIS Research Ethics Committee. Perhaps reflecting that background, each of us found the results in these studies surprising, even disturbing. Based on prior knowledge of the literature, the authors had anticipated strong orientation towards managerial interests, but not to the extent of c. 90%. We reasoned that, if such dominance is actually mainstream, members of the IS disciplinary community should know about it, should have some understanding of the situation, and should debate whether it is appropriate.

We considered how we might gain insight into the extent to which the results from the several exploratory studies were indicative of the patterns within the IS discipline as a whole. Other authors have undertaken examinations of literature with the intention of identifying new departures, or philosophical gaps, which could be gainfully exploited by IS researchers. For example, Orlikowski & Baroudi (1991) argued that the then-dominant positivist approach should be complemented by the interpretive and critical research philosophies. Avgerou (2000) sought a stronger theoretical base, and a more inclusive attitude to research methods, themes that have remained active in the two decades since. Hevner et al. (2004) consolidated what they referred to as the design-science paradigm. Galliers & Whitley (2007) profiled European IS research. Orlikowski & Scott (2008) proposed moving beyond the "Discrete Entity" and "Mutually Dependent Ensemble" research streams, by recognising the "entanglement" of humans and technologies, and shifting the focus to "Sociomaterial Assemblages". Narrower frames that have been adopted include privacy (Smith et al. 2010) and electronic interaction research (Clarke & Pucihar 2013).

Our objective is to assess the extent to which the richness of the notion of researcher perspective is reflected in IS research, and to generate insights, identify opportunities, and suggest ways in which those opportunities might be exploited. The scope of our proposal is therefore significant, but far less ambitious than broad-scale conceptual shifts such as the recognition of sociomateriality.

A study that encompassed, or purported to encompass, the entire output of the discipline would require far greater resources than are available to us. We accordingly set our sights much lower, and sought instead an approach that was capable of delivering significant findings, even if they cannot be generalised to the discipline as a whole. We selected as our focus the AIS Basket of 8 journals.

The general research question that we sought to answer was:

What are the characteristics of researcher perspectives adopted in articles in Basket of 8 journals?

We commence with an outline of theory relating to researcher perspective, and a summary of the results of the exploratory research projects. This is followed by a discussion of methodological considerations and presentation of the research design and protocol that we devised for this study. The findings are then presented and discussed. The results open up a considerable number of opportunities for further investigation. We believe that exploitation of these opportunities will make significant contributions to the quality and the value of IS research.

2. Literature Review

The word 'perspective' is frequently used in articles in the IS literature to refer to the approach adopted (positivist, interpretivist, etc.), or the theoretical lens applied (such as the technology acceptance model, or actor-network theory), or the target-audience of the research (in particular, academics, students or practitioners). These senses of the term are not the focus of the research reported here. We accordingly use the more specific terms 'stakeholder perspective' and 'researcher perspective', and provide definitions for each of them.

This section draws on early outlines of researcher perspective theory in Clarke (2015, 2016), and a fuller exposition in Clarke & Davison (2020). We provide a brief outline of the theory, define terms, and distinguish single-, dual- and multi-perspective research. Our purpose is to provide only sufficient depth of explanation to support the presentation of our empirical results. We then briefly review prior empirical research, and further articulate the research question.

2.1 Stakeholder Perspective

The origins of the stakeholder notion can be traced back to the work of the Stanford Research Institute (SRI) in 1963, which defined stakeholders as "those groups without whose support the organization would cease to exist" (Freeman and Reed, 1983). Freeman (1984) later modified this to "a stakeholder in an organization [is] any group or individual who can affect or is affected by the achievement of the organization's objectives".

The term has been the subject of many further interpretations during the almost four decades since it came to prominence. One analysis distinguishes stakeholders on the basis of power, legitimacy, and urgency (Mitchell et al. 1997). Power is not a necessary characteristic to qualify as a stakeholder. If an entity perceives itself as being affected by an information system, then that entity is a stakeholder. The term 'usee' was coined (Clarke 1992, Fischer-Huebner & Lindskog 2001, Baumer 2015) to refer to entities that lack the power to affect a process and may even be unaware of it, but are nonetheless affected by it.

The notion of stakeholders has long been applied in IS research. The earliest definition we have found in the IS literature is "those who depend on the organization for the realization of some of their goals, and in turn, the organization depends on them in some way for the full realization of its goals" (Mitroff & Mason, 1980, quoted in Mendelow 1981). Seddon et al. (1998) referred to a stakeholder as "a person or group in whose interest the evaluation of IS success is being performed", and distinguished (1) an independent observer (who has no personal stake in the measure); (2) an individual user (who evaluates a system from his or her own point of view); (3) a group of users, e.g., of a group decision support system; (4) the management or owners of the organization; and (5) a country, or mankind. More recently, Sedera et al. (2004) specifically addressed "the importance of a multiple stakeholder perspective".

Many stakeholders are participants in the process that gives rise to the phenomenon that the researcher observes. Roles played by stakeholders include investor, data source, technology provider, system sponsor and user (Seddon et al., 1999). However, the categories of stakeholders are broader than this, as Pouloudi & Whitley (1997, p.3) make clear, comprising not only "participants in the information systems development process viz. individuals, groups or organisations who take part in a system development process" but also "any other individuals, groups or organisations whose actions can influence or be influenced by the development and use of the system whether directly or indirectly". Pouloudi et al. (2016) provided the examples of "policy-makers, activists, government agencies, [and] professional and membership organizations".

Those authors also note the importance of delving into "the deeper issues underpinning how and why identified stakeholders may become fully engaged or dis-engaged with [the relevant activity]" (p.7). Many parties may have a stake in an activity, whether they are vociferous or 'silent', and whether they are powerful or weak.

It is important to note that some players do not have themselves have a stake, investment, or 'skin in the game'. In particular, policy-makers, regulators, representative and advocacy organisations and activists are not 'stakeholders' in the literal sense, but influencers on behalf of stakeholders, most commonly on behalf of stakeholders who themselves lack the power to protect their own interests.

We accordingly adopt the following interpretation of the term:

A Stakeholder is an entity that perceives itself to be at least potentially affected by an activity.

This encompasses organisational and human entities, at any level of aggregation, and in diverse relationships including B2B, B2C, G2B, C2C, etc. Our definition intentionally excludes entities that may affect activities but are not affected by them. For that, we suggest a more descriptive term:

An Influencer is an entity that is at least potentially capable of affecting an activity.

A Stakeholder with power (in the sense in which Mitchell et al. 1997 use the term) also qualifies as an Influencer. A Stakeholder without power does not.

During the course of lengthy consultancy practice and activities with public interest advocacy organisations, the first-named author has experienced many circumstances in which organisations have considered the needs of only those stakeholders that have sufficient power to affect the success of the activity. We argue that the longstanding confounding of the notions of Stakeholder and Influencer has led to organisations excluding powerless stakeholders from consideration, and in particular from consultation processes.

Each stakeholder, whether organisational or human, whether a powerful player or a weak one, and whether a participant in the system or an outsider that is affected by the system, has a point of view and interests that may be affected, and may take action in relation to those interests:

A Stakeholder Perspective is the viewpoint adopted by a stakeholder in a particular activity, reflecting that stakeholder's perception of phenomena within the relevant context, its value-set, and the interests that it seeks to protect and advance

When applying the concept, it is important to reflect the third principle expounded in Pouloudi et al. (2016), that "different stakeholders may have different values and perspectives" (pp. 8-9). A corollary of that principle is that, in information systems that embody a degree of complexity, conflict arises among the various stakeholder perspectives.

2.2 Researcher Perspective

The design and conduct of research generally involves the adoption by the researcher of a viewpoint that privileges the interests of one or more stakeholders, in effect treating that or those stakeholders as the intended beneficiary/ies of the research. We define the concept central to this research as follows:

A Researcher Perspective is a particular stakeholder perspective that is adopted by a researcher as the, or a, viewpoint from which to observe phenomena during the conduct of a research project

The notion of 'researcher perspective' is quite distinct from the 'object of study' and the 'unit of study' that are adopted in a research project. Extending the conventional optical metaphor, the 'object of study' is the set of phenomena that the researcher observes, and the 'unit of study' refers to the level of granularity at which observation is conducted. Researcher perspective, on the other hand, refers to the direction or orientation from which the observations of the phenomena are undertaken. The relationships among these ideas are illustrated in Figure 1. The upper half of the Figure uses the abstract terms, and the lower half provides an example, in which a researcher adopts the perspective of a system sponsor, and uses the theoretical lens of the technology adoption model to observe a work-group within an organisation. In such a circumstance, the system sponsor's interests are advantaged, and the work-group and its members are objects of study but not beneficiaries of the research.

Foreground examples of stakeholders in IS research are corporations that develop and operate information systems, provide IS-related products and services, and operate facilities that support IS. Other categories of stakeholders whose viewpoints may be adopted include the many kinds of users and usees. Under some circumstances, it may be appropriate to treat abstract notions as stakeholders, such as the economy, society or the biosphere.

Figure 1: Conceptual Model of the Research Process

Reproduced from Clarke & Davison (2020)

From prior studies, it is clear that one particular researcher perspective is very common in IS research. We refer to this category of stakeholder as the system sponsor, which we define as follows:

The System Sponsor is the entity 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

A system sponsor may be conceived at various levels of granularity. It may be a single corporation, a government agency or a not-for-profit organisation (Duhan et al. 2001); or it may be a sub-organisational unit within such an organisation. It may be 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 software, or organisations that adopt a service such as a particular CRM or online office suite or accounting facility). Alternatively, it might be an individual performing a role on behalf of an organisation, such as a CEO, CIO or CSO (Ifinedo & Nahar, 2007; Singletary et al., 2003).

Many stakeholders exist other than the system sponsor. Applying the example in Figure 1, a researcher could adopt the perspective of the work-group under observation, or of an individual within the work-group. Other examples of stakeholders whose perspective researchers may adopt include upstream or downstream participants in B2B pairs, organisations subject to an information system operated by a regulator, consumers making purchases within B2C systems, and recipients of transfer payments from a welfare agency.

The framing of a research project may predetermine the choice of stakeholder perspective. For example, a project that studies employee resistance to changes in organisational structure or business processes is very likely to privilege the system sponsor; whereas an examination of how and why employees adopt 'workarounds' would be more likely to reflect the interests of those employees, or perhaps of other stakeholders, such as the customers whose needs the employees are endeavouring to satisfy despite deficiencies in official business processes.

The following two, alternative research questions adopt the perspectives of respectively the system sponsor and another stakeholder. We suggest that these questions might lead to rather different research designs and findings:

What benefits and disbenefits accrue to a business from the decentralisation of call-centre operations to sub-contractors' home-offices? What benefits and disbenefits accrue to individuals and their families from the decentralisation of call-centre operations to sub-contractors' home-offices?

The perspective that the researcher adopts has a very significant influence on the entire research undertaking, commencing with the conception of the research, and moving on through the selection and formulation of research questions, to the research design. The choice that the researcher makes effectively determines what is included among and excluded from the potential outcomes of the research, how those outcomes are expressed, and the contributions that the research can make. Because of its critical role, we contend that the choice needs to be the result of a conscious and considered decision, rather than a default.

2.3 Cardinality of Researcher Perspective

The preliminary discussion of researcher perspective above dealt with single-perspective studies, where the interests of one particular stakeholder are in focus, and the interests of all other stakeholders are secondary. This approach is perceived by reviewers as being orthodox. It avoids complexities and enables conclusions to be expressed with only limited equivocation.

However, other approaches are also possible. In dual-perspective research, the researcher has the interests of two stakeholders in mind. The research design reflects both, and new information can be delivered to both. Research of this kind can deliver insights into the impacts of each party's own actions on the other party's interests, and into the impacts of the other party's behaviour on their own interests.

An example of a research question indicative of dual-perspective research is:

What benefits and disbenefits accrue to, respectively, the employer and employees, from an information system that performs real-time monitoring of employees' activities in the workplace?

A previously-published example of dual-perspective employer-employee research is Adie & Castleman (1997), which brought industrial relations and sociology to bear on teleworking:

"There are complex and paradoxical interests among the various parties in the employment relations context. ... Satisfactory resolution of the employment relations issues requires a recognition of the contradictory pressures and interests among employers and employees" (p. 399)

Other instances of dyads for which dual-perspective research can deliver value include buyer and seller in a B2B, B2C, B2G or C2C marketspace, government service delivery agency and client, and government regulatory agency and regulatee. A recent B2C exemplar is Keith et al. (2015), which considers the interests of both app developers and consumers of their products. The following quotation indicates a degree of balance in the benefits that the authors have in mind as they explain the implications of their research for practice:

"consumers are willing to trust in the benevolence, integrity and competence of app providers, simply when the consumers deem themselves capable of coping with any of the potential risks and dangers of disclosing their personal and location data. However, ... the reality is that the consumer has little practical way of accurately evaluating the app risks ... these consumer vulnerabilities point to opportunities that need to be resolved by industry" (pp. 660-661)

In some circumstances, the perspective of the system sponsor may not be represented in dual-perspective research, because two stakeholders other than the system sponsor are studied, as would arise when addressing the research question:

What benefits and disbenefits from a government-run eHealth records system accrue to, respectively, health care professionals and patients?

It is important to distinguish between studies in which a dyad is considered from the perspective of one party (e.g. most studies of outsourcing) from research that takes into account the interests of both parties, and designs the research in order to identify implications for them both. This is atypical of conventional 'strategic IS', and more commonly occurs in the contexts of inter-organisational IS, business partnerships, strategic alliances, and collaborative infrastructure.

Examples also exist in which not all stakeholders are active participants, i.e. some are 'usees'. For example, Lin et al. (2015) contrast the interpretations of a system sponsor and the aboriginal people for whom a system was designed, showing that "a project that was intended to improve the education and social welfare of the aboriginal people in Taiwan" actually left them "feeling marginalized and without much of a voice" (p.697).

More broadly still, multi-perspective research can be undertaken. This approach is appropriate in multi-organisational networks and cascaded supply-chains. In such circumstances, information system designs tend to have 'win-win-win' outcomes as a design criterion. This is because, without it, such projects are doomed to failure. Although some degree of conflict between the interests and the value-sets of the two stakeholders is likely to arise in dual-perspective research contexts, such conflicts are almost inevitable in multi-perspective contexts and so there is likely to be a considerable degree of mutual non-comprehension by stakeholders of one another's interests (cf. Sedera et al., 2004).

An exemplar of multi-perspective research in a business context is Cameron & Clarke (1996), who studied multi-organisational project management in two contexts: the book trade, reflecting the interests of retailers, wholesalers, and a software provider; and international trade EDI, which encompasses many categories of business and government entities including freight forwarders, stevedores, transport firms and customs agencies. Another example is Choudhury (1997), who considered the interests of the various players in interorganisational IS in the aircraft parts industry.

An exemplar in the social policy space is Agarwal et al. (2012), who examined cyber-collective social movements such as the use of social media in the Arab Spring of 2010-2012, and reflected the activities and interests of various categories of individuals such as blogger-activists, opinion leaders, sharers and followers; and regional, national and transnational communities. Carlsson & Walden (2015) considered the use of mobile technology to enhance wellness among the young elderly, while taking into account the public health and budgetary aspects relevant to society as a whole. To the extent that IS research is to be relevant to policy-makers, it is necessary for research techniques to be adopted that enable researchers to cope with the inherently multi-perspective nature of the phenomena.

2.4 Prior Empirical Research

A first pilot study of researcher perspectives adopted in IS was performed in 2015, in support of a Conference Keynote presentation (Clarke 2015). It examined the 36 articles in Volume 18 (2014) of the Australasian Journal of Information Systems (AJIS), and a sample of 38 articles from the 2014 Australasian Conference in Information Systems (ACIS). The findings were that 83% of the papers reported single-perspective studies, and the perspective(s) adopted included the system-sponsor in 96% of cases.

A second study examined 212 papers presented at 5 years of the Bled eConference, at 3-yearly intervals from 1991 to 2015 (Clarke 2016). This found that 87% were single-perspective papers. The system sponsor's was the sole perspective adopted in 85% of papers, and was present in a further 8% of papers that adopted dual- and multi-perspective approaches, for 93% in total. In all of these cases, however, these percentages showed a downward trend from 1991 to 2015.

A third study examined 60 papers from the corpus of c. 650 published in the specialist journal Electronic Markets (Clarke 2020). One sub-sample comprised all articles on the subject of auctions, and evidenced patterns not unlike those in the previous studies.

The other sub-sample was selected on the basis of evidencing researcher perspectives other than that of the system sponsor alone. One exemplar identified in this manner is Laakso et al. (2017), which declares in its title the centrality of its "implications for stakeholders in academic publishing", and whose implications relate to:

"stakeholder perspectives for publishers, university libraries, authors, and [Academic Social Networks] themselves" (p.131)

Saravanan & Balasundaram (2013), which appeared in a Special Issue on 'Mobile Health', examines the application of mobile IT to provide timely services for needy people who are at risk while on the move. Implications are drawn for:

"doctors, patients and health care providers" (p.13)

Cabrera & Oezcivelek (2008), which was one of two research articles in a Special Issue on 'Inclusive ICT Business', concludes that:

"[ICT-based independent living services] policies should not limit their perspective to older individuals, but should include also the context of relatives, informal carers, and formal care (in its different forms)" (p.311)

This series of studies was commenced because of the impression, gained from long-term scanning of the literature, that corporate interests strongly influence IS research. Despite the existence of outliers such as those quoted immediately above, the results to date suggested that the degree of influence might be appropriately described as dominance.

It might have been expected, for example, that the research domain of 'Green IS' and/or 'Green IT' would reflect interests other than those of the system sponsor. However, the articles that have been inspected to date have not addressed such opportunities as re-cyclability of computing components, carbon trading systems, fish-stock monitoring, energy-efficiency management, or the energy profligacy of bitcoin-mining, but have instead reflected organisations' interest in the cost-savings that can be achieved through green initiatives, as well as the public relations advantages that can accrue when an organisation is perceived to be green.

The sampling frames for the three previous studies arose respectively from an invitation to present a keynote, an opportunity to conduct a review of the corpus of a venue with which the researcher has a long-term association, and an invitation to prepare a 'retrospect and prospect' essay for a journal's 30th anniversary issue. Although they are mainstream venues for the publication of IS research, the specificity of the sampling frames precludes generalisations about the discipline as a whole. This motivated us to design a broader study that could deliver deeper insights.

2.5 Research Questions

On the basis of the above theoretical discussion, we further articulated the general research question presented earlier, as follows:

  1. What researcher perspectives are evident in the AIS basket of 8 journals?
  2. To what extent are the studies single-, dual- and multi-perspective in nature?
  3. Among articles whose object of study is human, to what extent does a balance exist between the perspective of the system sponsor and the perspectives of other stakeholders?
  4. What changes in researcher perspectives are evident in the AIS basket of 8 journals over time?

3. Research Method

This section describes the sample selection process, the nature of the content analysis applied to the works, and the process undertaken in order to encode each article.

3.1 Sample Selection

The IS literature is highly diverse. The most authoritative collation, Lamp (1995-). identifies "892 IS journals, of which 684 are still active". There are many core journals and conferences, but a great many articles and papers are published in journals and conferences of cognate disciplines, and in a wide array of venues whose focus is on research domains, such as user interfaces, eCommerce, information management and green computing. It is challenging to find a suitable basis for stratified {weighted} random sampling across venues.

As discussed above, we selected the 'Basket of 8' journals as our population. These venues are projected as being 'leading', in the senses of being highly-regarded, highly-cited and much-read, and providing exemplars to entrant and early-career researchers. A wide variety of articles have used Basket of 8 journals as a population or sampling frame, including Lowry et al. (2014), Grover & Lyytinen (2015) and Tarafdar & Davison (2018). As defined by the Association for Information Systems, the Basket comprises (in alphabetical order of abbreviations):

The eight journals commenced operation at various dates, and issues have been published with varying frequencies and containing varying numbers of articles per issue. In total, the count of articles (in the order of 3,000) is sufficiently high that it exceeds the available resources to analyse. It was therefore necessary to analyse a sample rather than the entire population.

We were interested in detecting whether the measures had changed over time, and we accordingly adopted a stratified sampling approach, by first selecting different years from which to select articles. We started this research in 2016, so the articles published in the year 2015 were selected. In order to provide a spread, we then chose as the starting-point the year after the commencement of the youngest of the journals (JAIS), which was 2001. The mid-point year, 2008, was added, to provide three points at 7-year intervals across a 15-year span. This gave rise to 8*3 = 24 Volumes. The number of issues per volume varied both among journals and across years, and one journal published five issues in a relevant calendar year rather than its usual four, resulting in a total of 118 issues.

In effect, then, we defined these eight journals as the population, and selected three individual years (2001, 2008 and 2015) as the sampling frame. The focus was on refereed journal articles, and hence unrefereed contributions such as editorials, calls for papers, and book reviews were generally excluded from consideration. An exception to this was editorials that were of the nature of extended theoretical treatises, because these often have considerable impact. This produced a sampling frame of 659 articles. The number of articles per issue averaged 5.5. The number of articles per journal ranged between 49 and 124, around a mean of 82. Further aspects of the distribution are evident in Table 1.

Articles were only relevant to the study if they evidenced a researcher perspective. The primary examples of articles that did not satisfy that criterion were teaching cases, research methodology articles, and reports addressed solely to researchers. These were coded DI (Discipline Internal). The present article is an example of such a work. This removed 111 articles (17% of the 659 in the sample), leaving 548 for detailed examination. This provided a sufficiently large data-set for most of the intended purposes.

Table 1 summarises the counts of articles in the 8 journals, showing the numbers of articles that were not addressed to stakeholders external to the IS discipline, and the remaining number that are relevant to the present research.

Table 1: The Article Counts

EJIS 10, 17, 24
ISJ 11, 18, 25
ISR 12, 29, 36
JAIS 2, 9, 16
JIT 16, 23, 30
JMIS class=font9>17-18, 24-25, 31-32
JSIS 10, 17, 24
MISQ 25, 32, 39

3.2 Content Analysis

In order to encode the articles and generate statistical summaries, it was necessary to analyse the contents of each article. Studies of previous publications take several forms. Reviews of relevant literature on theory and prior empirical research are commonly found early in articles that report the results of empirical studies. Further techniques include systematic literature reviews and several forms of content analysis. A review of these techniques was conducted as part of the overall research project (Clarke 2017), in order to inform the design of the study reported on in this article.

In order to conduct the empirical studies reported earlier, a suitable form of content analysis had been developed, trialled and applied. This project adopted a modified version of that coding protocol. Iterative refinements were made during the coding process, with revisions made primarily with respect to descriptions of how the guidelines were to be interpreted in particular contexts. These refinements of the protocol were necessitated by the complexities involved in applying the guidelines consistently and systematically to a highly diverse set of articles. The approach adopted is a customised version of a technique that is usefully described as 'the critical analysis of published works'. Clarke (2019) describes the basis on which the technique was developed, and its key characteristics.

The examination of each article commenced with a search for passages that indicated the stakeholder interests that are being represented and/or the audience to which the outcomes are addressed. Explicit statements were sought about which stakeholder interest(s) the researcher intended to privilege, but were seldom found. Even where found, however, they were evaluated within the context of the whole text. Key passages were noted.

Within each paper, the most informative sources of insight into the nature of the article were generally the title and abstract, which were first considered, and the introduction, the implications for practice, and the conclusions.

An example of text that is strongly indicative of a system sponsor perspective is:

"Recommendations are provided as to how online managers and web designers can use web personalization cues to positively influence website stickiness and to strengthen their digital business model" (Benlian 2015, p.225, emphasis added)

The following text, in contrast, indicates a researcher perspective other than that of the system sponsor:

"we focus on understanding how social media empowers communities during crisis response" (Ling 2015, p.174, emphasis added)

We were initially unsure whether the available resources would be sufficient to analyse the sample of 548 articles. The assessment process 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. At the outset, it was unclear to us how resource-intensive the assessment of each article would prove to be. We accordingly adopted a two-phase approach.

In the first phase, we selected a 20% sample (135 articles) of the articles within each volume 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 then selected, in the order in which they appeared in that journal in each of 2001, 2008, and 2015. We encoded articles on the basis of the full text, with particular reference to the title, abstract, introduction and conclusions, seeking out information relevant to the research method, intended beneficiary and target audience.

We then piloted an alternative approach, whereby we encoded articles based on the title and abstract alone. We compared the two approaches and established that differences were minimal, and that in all cases in which a potential difference in coding could have occurred, sufficient uncertainty existed that it was apparent that fuller assessment was needed.

We accordingly encoded the remaining 80% of the sampling frame (524 articles) on the basis of the title and abstract, with the proviso that, where doubts arose, the first-phase method of checking the full text was to be applied. This was the case with 47 articles (9%). Checks established that a high level of confidence in the efficacy of the two-phase approach was warranted.

3.3 The Coding Process

The project team comprised three people. 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 guidelines 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 achieve complete agreement on all codings, thereby satisfying the requirement for inter-rater reliability.

In order to be able to encode each article, a number of preliminary steps were necessary. A sufficiently deep understanding of an article depends on first appreciating the author's purpose. Depending on the type of research, this may be expressed in the form of a Research Question (for empirical work, both positivist and interpretivist), or an Objective (particularly for design science research), or a topic-area (in the case of theoretical discussions). The first two steps were accordingly to categorise the type of research, and to then isolate the purpose, wherever possible by extracting the author's own words but where necessary by inferring the purpose from the text as a whole.

The content was then analysed as described in the previous sub-section. Key passages were noted. Two aspects of researcher perspective were gathered. A cardinality measure was recorded, to identify respectively single-, dual- and multi-perspective research. Secondly, each specific stakeholder was recorded. Sufficient data was recorded to support not only review and audit, but also deeper analyses where required.

To enable RQ3 to be addressed, one further attribute of each article was encoded. The object of study was allocated to one of three categories. Some research has as its specific focus a Technology, by which is meant an artefact such as an information system or an element of an information system, such as hardware, software, user interfaces, methods or manual procedures. The category 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 CIO and CISO. The third category, which is relevant to this study, is Human(s), by which is meant social collectives irrespective of scale (i.e. groups, communities and societies) and individuals, but excluding humans operating on behalf of formal organisations.

4. Findings

Of the 659 articles in the sampling frame, 548 (83%) were relevant to the study, in that they addressed the interests of real-world stakeholders. On the basis of the research process described above, we now report the results of our analysis, answering the specific Research Questions identified earlier.

4.1 Researcher Perspectives Evident in the Basket of 8 Journals

Of the 548 articles, 528 (96%) adopt a single-perspective approach rather than seeking a balance between the interests of two or more stakeholder groups. Single-perspective research is dominant in all of the journals (range 93-99%).

The large majority of single-perspective research is concerned with the interests of the system sponsor: 491 articles (93% of the category, and 90% of all articles).

Of the 528 single-perspective articles, the perspective of a stakeholder other than the system sponsor is adopted in 37 (7%). Over half of the 37 are participants in online markets, in 14 cases sellers and in 8 buyers. The remaining 15 comprise 4 instances of online communities, 8 of participants in online communities, and 3 of 'meatspace' communities.

All journals evidence dominance of the system sponsor perspective (range 82-98%). Over the 3 years, 1 journal published no single-perspective articles in which the researcher perspective was any stakeholder other than the system sponsor. 3 journals published 1 each, and the others 4, 6 and 9, with JMIS as an outlier, with 15.

Table 2 identifies some common examples of the system-sponsor and other-than-system-sponsor categories. An example of a passage indicating a system-sponsor perspective was "We discuss implications of these findings for both IT management and future research" (Rutner et al. 2008 p.635). A passage indicating that the researcher perspective adopted was online communities was "sociability and knowledge contribution behaviors as well as structural social capital lead to being identified as a leader by members of the online community" (Faraj et al. 2015, p.393).

Table 2: Researcher Perspective

System Sponsor
Supply Chain

Government Agency
Senior Staff
IT Professionals

Cha et al., 2008
Baldwin, et al. 2001; Randall et al.,

Jones & Hughes, 2001
Earl, 2001
Rutner et al., 2008

Other than
System Sponsor
Online Community
Faraj et al., 2015
Singh et al., 2015

Gao, et al., 2015

4.2 The Incidence of Single-, Dual- and Multi-Perspective Research

Of the 548 articles, the 20 that were not single-perspective were in 15 cases (3%) dual-perspective and in 5 cases (1%) multi-perspective. Of these 20 articles, the system sponsor was included among the perspectives in 17 instances, whereas 3 articles considered only stakeholders other than the system sponsor (e.g. the interests of buyers and sellers, rather than that of the marketspace provider).

The distribution across the journals is shown in Table 3. Cumulatively over the 3 sampled years, 5 journals each published precisely 1 dual-perspective article, EJIS and ISR each published 3, and JMIS published 4.

Table 3: Perspectives in the Sample

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

>1 Perspective

We did not find any straightforward expressions by authors about adopting the perspectives of two stakeholders within the research domain, such as 'Implications were drawn from the research for respectively e-marketers and e-consumers: ...'. Examples of text that is indicative of dual-perspective research are as follows:

The 5 multi-perspective articles during the 3 years in the sample appeared in only 2 journals, i.e. no such articles appeared in 6 of the journals. Many of the articles of this kind that have come to light in earlier studies lack precision about the particular stakeholders whose interests are being reflected in the work. This characteristic was also apparent in the relevant, small sub-sample from the Basket of 8.

For example, Whitley & Hosein (2008) analysed UK parliamentary debates on the technological and scientific aspects of a proposed ID scheme, but did not delineate the stakeholders in public policy decision-making. Only a little more specific are "family, community and other concerns" (Cordoba & Midgley 2008, p.125), and "individuals, organizations, and society" (Chatterjee et al. 2015, p.49). The sole instance of real clarity is "recommendations for firms ..., for education providers ..., and for users" (Matook et al., 2015, p.278).

4.3 The Incidence of Researcher Perspectives where the Object of Study is Human

We distinguished three categories of object of study. The most common was Organisations, with 323 articles (59%), and the least common in the sample was Technologies, with 27 (5%).

Among the 548 articles, Humans, including not only individuals but also collectives such as groups and communities, were the object of study in 198 articles (36%). We reasoned that researchers would be considerably more cognisant of stakeholder perspectives other than that of the system sponsor in the Humans category than in relation to Organisations and Technologies.

Of the 198 articles in which humans were the object of study, 167 (84%) adopted solely the system sponsor perspective, with the individual journals ranging 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 a perspective other than that of the system sponsor.

Of the other 31 articles, 23 (12%) adopted a single perspective other than that of the system sponsor. In most cases, the perspective was that of (usually some segment of) individual participants in social media or online communities, or in online markets.

The remaining 8 articles (4%) were dual- or multi-perspective in nature. Of those, 5 (3%) included the system sponsor among the perspectives, bringing the total that reflected the system sponsor's interests to 172/198 (87%).

Table 4 provides a comparison between these counts and those for the sample as a whole. This provides evidence in support of the proposition that only a marginally larger proportion of researchers reflect the interests of other stakeholders when the object of study is people. The proportion of research that is dominated by the interests of the system sponsor remains very high.

Table 4: Cognisance of Other Stakeholder Perspectives

%age of the Sample>
(n = 548)
%age with Humans
as Object of Study
(n = 198)
Single-Perspective / System-Sponsor
Dual- & Multi-Perspective incl. SS
Single-Perspective / Other Stakeholder
Dual- & Multi-Perspective excl. SS

4.4 Changes in Researcher Perspectives Over Time

Single-perspective studies were dominant in all three years, 2001 (99%), 2008 (95%) and 2015 (96%).

On the other hand, the proportion of articles that considered exclusively the system sponsor perspective was lower in the later years in the sample, respectively 94%, 91% and 85%, with other single-perspective research at 5%, 4% and 11%. (The remaining few percentage-points are of articles that are dual- or multi-perspective in nature).

The Objects of Study were considerably different in the three years, with the focus on humans c. 20%, c. 30% and c. 50%. The level of system-sponsor orientation in studies of humans was 88% in both 2001 (22 articles) and 2008 (52), but 82% in 2015 (93).

As shown in Table 5, of the 37 articles in which a single-perspective other than that of the system sponsor was adopted, 6 were published in 2001 (5% of the relevant articles that year) and 7 in 2008 (3.5%), whereas 24 appeared in 2015 (11%). The sub-sample is small, but the possibility exists that researchers have become more aware of the interests of stakeholders others than the system sponsor. Of those 24 articles, 92% were in just 4 of the 8 journals. The surge was a result of the interest in social media and online community topics.

Table 5: Single-Perspective Research Other Than System-Sponsor


Only a single dual-perspective article was published in 2001, in JMIS, but 7 dual-perspective articles appeared in 4 journals in 2008, and 7 in 6 journals in 2015.

Only 2 journals published multi-perspective articles - EJIS (3, all in 2008) and JMIS (2, both in 2015).

5. Discussion

This section first discusses the aspects of the study that limit the scope for generalisation of the results. It then identifies opportunities for research that this work has opened up. It concludes with some thoughts regarding the implications for the IS discipline.

5.1 Limitations of the Research

This research examined all articles published in 8 journals in each of 3 years within a 15-year period. It therefore comprises a series of cross-sectional snapshots, and does not represent a random sample of the population of 15 years of the Basket of 8 journals. The results are therefore authoritative with respect to the years in question, but are not generalisable even across the population of Basket of 8 articles, far less to the IS literature as a whole. All observations made on the basis of the data reported in this article must be interpreted with that caveat in mind.

Because the study's focus was the discipline's leading journals, it has delivered insights into researcher perspective in high-quality publications which are widely used as exemplars. The life-cycle of research reporting in high-quality venues is slow. A disadvantage of this choice of venues is therefore that the results of the study describe past practices. This limitation was addressed to some extent by selecting years spaced across two 7-year intervals, and noting differences in the patterns in those three years. It would not be reasonable, however, to either interpolate within, or project beyond, that period in order to postulate trends in the various measures. A better approach to detecting change, and particularly non-linear and disruptive change, would be to study samples of venues in which more rapid adaptation can occur, such as the leading conferences, and particularly conferences that have a reputation for welcoming novel approaches.

The coding guidelines have been applied by the same team that developed them. Appropriate care was taken, and procedures were adopted to ensure that intra-team consistency was achieved. Nonetheless, the protocol will benefit from examination and application by other teams, and to articles and papers in further publishing venues.

A further limitation of this research is that its primary focus is, by design, on broad questions about researcher perspectives, within the publishing venues included in the sample. That approach cannot generate much insight into researcher perspectives in specific topic-areas, such as inter-organisational systems, outsourcing, marketer-consumer relationships, and C2C and social media interactions. Nor can the research reported here reliably distinguish the nature of researcher perspectives adopted by adherents to particular research approaches (i.e. positivism, interpretivism, critical theory, design science) and particular research techniques (e.g. surveys, laboratory experiments, action research). However, the coding protocol is applicable to samples that are selected on the basis of criteria other than publishing venue, and to research genres. It has already been applied to two topic-areas - electronic markets, and IS research into privacy topics - and it required very little modification.

Clearly, the problem of limited diversity in researcher perspectives could afflict other disciplines as well as IS. Our preliminary searches in other literatures did not unearth any examinations along the lines of the work reported here. It would be valuable to perform deeper searches, particularly in other business school disciplines, in order to gain an appreciation of how, for example, organisational psychology and industrial relations deal with such issues, and what lessons from those disciplines might be applied to the IS discipline.

5.2 Implications for Research

The outcomes of this research shine a spotlight on a number of opportunities for IS researchers. This section defines those opportunities, and offers suggestions as to how they might be exploited.

The findings reported in this article at s.4.1, in common with previous studies of other publishing venues, were that the single-perspective / system-sponsor (SP/SS) researcher perspective dominated all of the alternative approaches.

Superficially, this could be seen as the natural approach to adopt, to the extent that an IS researcher perceives their role as being to serve the interests of organisations that operate information systems. Even given that assumption, however, the quality of conclusions drawn from such research is limited by its one-dimensionality.

A foundational postulate of strategic information systems theory is that each organisation is subject to a range of competitive forces. The classical model (Porter 1980, 1985) has its focus on industry participants external to the organisation, comprising existing firms, suppliers, customers, new entrants, and providers of substitute goods and services. Regulatory regimes also embody competitive threats and opportunities. Sources of internal competition exist as well, most crucially for capital access and for executive attention. Also not to be overlooked is the competition between return on capital and return to labour. Employees may or may not be sufficiently coordinated to exercise market power against their employer; but their levels of productivity, and their creativity, may well reflect their perception of the nature of the deal that they have with their employer.

Given that rich palette of challenges to the power of each system sponsor, it is uncommon for an organisation's interests to be best served through single-minded analysis of activities from that organisation's perspective alone. To achieve depth of understanding, the behaviour of other entities within the research domain needs to be considered as more than just an object of study or an exogenous variable. That behaviour is guided by each stakeholder's perceptions of its own best interests. System sponsors need IS research to encompass those interests, because that enables it to deliver insightful conclusions and recommendations based on a broad model of the relevant world, rather than offering mere reinforcement of their own view of their world.

This article has distinguished three alternative approaches to researcher perspective that can make such contributions. The first is single-perspective research that adopts the perspective of a stakeholder other than the system sponsor, e.g. employees, suppliers or customers. The results of such research are of legitimate interest to system sponsors. However, such research by definition privileges the interests of the particular stakeholder whose perspective is adopted, which opens up new opportunities for meaningful services to the world, and, with that, to new sources of data, and new sources of research funding.

Undertaking research from a perspective other than that of management might seem counter-cultural, at least for those IS researchers used to adopting a narrow interpretation of a business school's mission. On the other hand, suppliers and corporate customers are corporations within such researchers' field of view. Further, some business schools, and most IS departments outside business schools, define the interests of employees and of consumers as being within-scope. Moreover, although it might seem counter-intuitive to regard other-stakeholder researcher perspectives as serving the needs of system sponsors, they too can gain from such research, because it is capable of delivering much clearer understanding about the perceptions of the system sponsor's competitors than can conventional SP/SS research.

A second, more challenging alternative is dual-perspective research. The foreground variant involves the selection of a dyad (such as employer-employee, supplier-customer, outsourcer-serviceprovider, or government-agency-and-its-client), and embodying both stakeholders' perspectives within the research conception, design and execution. This has a long tradition in strategic theory, associated with collaboration, business partnerships, and the 'win-win' notion. Projects of this nature are more challenging, both intellectually and in terms of the harnessing of research techniques to address the research questions. In return, dual-perspective research offers greater depth of understanding to each stakeholder whose interests are privileged, and can contribute to the development of designs that are more reliable and stable, because both parties stand to lose if they withdraw from or flout the rules of a win-win game.

Dual-perspective research is valuable in other circumstances as well. One is where a system sponsor wishes to understand and accommodate the needs of some other stakeholder because its disapproval of the project could lead to project failure. This may arise because of the other entity's market power (e.g. where it is a major supplier, customer, or trade union) or institutional power (e.g. in the case of a regulator, financier or insurer).

Alternatively, there may be a legislative or contractual requirement that the system sponsor reflect the other stakeholder's interests. This arises, for example, in health care, and where the activities are intended to benefit, or may harm, a relatively weak actor, such as people who suffer particular kinds of disabilities, and members of other disadvantaged or minority groups.

The dyad that is chosen in dual-perspective research does not necessarily have to include the system sponsor. For example, joint studies of the interests of buyers and sellers in marketspaces can deliver value to stakeholders on both sides of the bargain. At the same time, they can deliver the marketspace operator / system sponsor deeper understanding of the interests, behaviour and needs of its clients.

Dual-perspective research is likely to be of greatest value where two stakeholders between them dominate a landscape. That has been a common pattern in many industries, but in far from all; and contemporary patterns of repetitive disruption may be rendering it less common than it used to be. In such circumstances, it may be essential to take up the greater challenge of multi-perspective research in order to deliver real value.

The field of inter-organisational information systems (IOS or IOIS) is well-established, and multiple industry topologies have been much-studied, particularly 'hub-and-spoke' patterns, industry supply chains and cascades, and increasingly also industry networks. Studies can deliver greater insights for all parties by going beyond the drawing of inferences for an individual stakeholder, and instead directly addressing the factors that underlie 'win-win-win' strategies.

A further opportunity exists, although it appears to be unpalatable to many within the IS discipline. Only a small proportion of published IS research to date addresses the needs of public policy-makers. Research of this nature must be much more open-textured than conventional IS research. This applies not only to the choice of research domain, object of study and unit of study, but also researcher perspective. Exemplars exist, including an AIS Award-winner (Agarwal et al. 2012). Meanwhile, guidance on research techniques is available (Majchrzak & Markus 2014), and an ICIS panel session has addressed the issues (Niederman et al. 2017).

At a more reflective level, this article also raises questions about whether the current patterns of dominance are accidental or intentional. If they are accidental, then demonstration that there are valuable opportunities to be exploited may be enough to stimulate greater investment in projects and programs adopting alternative researcher perspectives. If, on the other hand, the dominance of system sponsor interests is an intentional and entrenched choice among all, many or even some researchers, then there will be resistance to the propositions advanced in this article.

Wall et al. (2015) refer to accidental adoption as 'unconscious hegemonic participation'. Research is needed to establish the extent to which SP/SS dominance is a result of a bandwagon effect, with unreflective adoption by later researchers of approaches and attitudes embedded in the work of their predecessors. Alternatively, intentional adoption, in Wall's terms 'conscious hegemonic participation', might arise from single-minded implementation of what individual researchers might perceive as an appropriate 'business school' commitment to the needs of business enterprises. Understanding of these patterns would be assisted by surveys designed to elicit the views of early career researchers, on the one hand, and senior members of the discipline who perform 'gatekeeper' functions, on the other.

The findings in s.4.3 showed that single-perspective / system-sponsor-viewpoint studies dominate IS research even where the object of study is human individuals or groups. Surveys could establish whether this is perceived by various categories of IS researchers to give rise to ethical challenges for the discipline. There would be benefit in devising parallel surveys of the views of business leaders about the appropriateness and usefulness of research that primarily treats their employees as objects of study rather than as stakeholders in corporate activities.

A complementary approach would be to investigate the reasons why other stakeholder perspectives are represented in such a small proportion of research published in the sampled venues. Reasons that could be postulated include the following:

We suspect that most, if not all, of these factors play a role in the single-mindedness of so much published research in IS. Formal research into these issues could be conducted, e.g. by means of structured interviews and focus groups.

A further plausible explanation for the paucity of dual-perspective research in IS may be an absence of research techniques, or of research techniques that are acceptable by, or perceived as being acceptable by, IS discipline gatekeepers. A survey of relevant techniques used within and outside the IS discipline could be undertaken. If appropriate techniques are unable to be found, the opportunity exists to formulate research protocols, and trial them.

Similarly, in the area of multi-perspective research, projects could be undertaken to establish why so little is published. This may identify the need to adopt, adapt or develop research techniques suitable for application to inter-organisational systems in which multiple parties have sufficient market power that progress is dependent on collaboration and hence on the adoption of 'win-win-win' strategies.

Research techniques for addressing more open-textured public policy questions can be sought out and developed, as proposed by Majchrzak et al. (2016). Researchers can also seek to partner with government agencies that have policy responsibilities. The scope also exists for researchers to work with other organisations that could benefit from the efforts of IS academics, such as advocacy organisations for the interests of employees, consumers, welfare recipients and special interest groups.

5.3 Speculative Implications for the IS Discipline

This research project was conducted, and this article was composed, in the expectation that it would encounter resistance from reviewers and editors. That expectation was fulfilled, with the article's lengthy journey including combative responses from some of the senior editors and reviewers who considered it. Moreover, some of the more positive reviewers who agreed that the work was well-performed and well-described, felt that the article was uninteresting. A reason for this, we feel, is the strenuous efforts we made to painstakingly convey the still-new theory, to explain the somewhat novel research method that we adopted, and to express the analysis and arguments carefully, even punctiliously.

We accordingly offer the following, much more speculative thoughts, with the intention of conveying that the impacts of this work may be far less bland than they have appeared to be, at least to some reviewers. The work has potentially very substantial and even explosive implications for the IS discipline. However, because the area is new, much remains to be done before forthright propositions can be formulated.

Critical review of publications in many domains may well conclude that IS research has been of far lower quality than it could be, and should be. It may be failing to fulfil the needs even of those intended as the primary beneficiaries, that is to say executives and managers, primarily in for-profit corporations.

The reasoning underlying this suggestion is that researchers are wearing blinders (in US terms) or blinkers (in English terms), and as a result delivering simplistic, uni-dimensional views of complex realities. Single-perspective research is appropriate only to simple contexts in which other stakeholders are little-impacted by interventions intended to benefit the stakeholder in focus. Such circumstances were commonplace during the early decades of computing. Even in the contemporary era of digitalisation and roboticisation, some simple circumstances continue to exist, and represent suitable training grounds for researchers. In the large majority of contexts that demand attention from IS researchers, on the other hand, the complexity of the domain makes single-perspective research appear pitifully inadequate. Some evidence for this might be found in the ongoing parade of IS failures, and of unsuccessful impositions of technologies, despite the person-centuries invested into their study. One potential explanatory factor would be IS research designs that match poorly to on-ground realities.

As just one example, we draw attention to the problem of organisations excluding powerless stakeholders from consideration, and in particular from consultation processes, which we associated earlier with a failure to separate out the roles of Stakeholder and what we have called Influencer. We believe that has in turn been a factor in many instances of project failure, to the detriment of both return on investment and economic and social outcomes.

If our speculative interpretation of single-perspective research leading to low-quality research outcomes should prove to be justified, the antidote would be to shift the norm from single-perspective to at least dual-perspective research. Only by this means could executives and managers be provided with insights into the interests and likely behaviour of other stakeholders whose market or institutional power is capable of preventing them from achieving their goals.

A large proportion of contemporary research, review and publication resources are dedicated to small refinements in models that represent only very constrained segments of real-world problem domains. The overwhelming commitment to rigour might feel satisfying inside ivory towers, but it ensures that only very limited contributions can be made to practice. Switching to inherently more challenging research techniques would inevitably have negative consequences for research rigour, at least during the early years during which errors are usefully made, and expertise is developed; but those consequences would have to be endured if it became clear that IS research has been generating low-quality outputs and failing to deliver the insights that decision-makers need.

The single-perspective approach is particularly incongruous in research into long industry-chains, industry networks, and collaborative infrastructure. Results are inevitably of unacceptably low quality where the power of the players is shared, or is continually shifting as the relationships reconfigure themselves. In such contexts, even dual-perspective research misses the point. So the challenge must be taken up to develop multi-perspective techniques that can reflect the perspectives of many stakeholders rather than just one or two powerful players.

The IS discipline has long failed to make meaningful contributions to public policy-making. The maturation of research techniques that support multi-perspective approaches would enable not only the primarily economic interests in industry networks to be examined, but also the mixed economic and social interests inherent in the broader policy field.

We also raise the question as to whether the dominance of the system sponsor researcher perspective in IS is inherently unethical. Larsen (1979) referred to the necessary "normative dimension" of a profession as covering "the service orientation of professionals, and their distinctive ethics, which justify the privilege of self-regulation granted to them by society" (p.x). The Codes of Ethics of the relevant professional associations assert that safety, health, and welfare of the public is paramount (IEEE), and that ICT and IS practitioners are to contribute to society and to human well-being, avoid harm and respect privacy (ACM) and enhance quality of life (ACS). Until December 2019, the AIS Code was only concerned with ethical behaviour by researchers to the extent that it affects other researchers. The Association then adopted a Code of Ethics (AIS 2019) modelled on that of the ACM. It is arguable that requirements "1.1 Contribute to society and to human well-being, acknowledging that all people are stakeholders in computing" and "1.2 Avoid harm" mean that a failure by AIS members to consider the interests of other stakeholders in their research and their reports on their research is unethical behaviour. Moreover, it would also be unethical behaviour on the part of reviewers and editors if they respectively recommended publication and published such articles.

In the sample studied here, 84% of the c.200 articles whose object of study was humans reported single-perspective research for the benefit of, and only of, the system sponsor. It would appear that the absence of ethical constraints on IS academics has permitted corporate interests to dominate the interests of individuals and communities, even though the students who those academics teach are required to respect the interests of people affected by the systems that they develop and apply.

If the proposition that the IS discipline is inherently unethical were to be shown to be more than mere speculation, a major contribution of this research might transpire to be the surfacing and articulation of irresponsible behaviour by the IS academic community. Might it be that a future, genuinely academic discipline of IS will incorporate the perspectives of consumers, employees, users, usees and affected communities into the majority of the projects that they undertake? That might represent an inversion of the present situation, to the extent that single-perspective research continues to be practised, but adopting a greater diversity of perspectives. Alternatively, a switch could occur to greater balance in the delivery of benefits to stakeholders, to the extent that, as we propose, dual and multi-perspective research come to dominate the field.

The research reported here might lead to a further, yet more radical outcome. The IS discipline might respect and encourage differences between the business school culture and its tendencies towards single-minded, single-perspective, system-sponsor dominated research, and alternative philosophies. Work outside business school norms would include some single-perspective research that prioritises the interests of stakeholders other than the system sponsor; but much greater economic and social value would be achieved through dual-perspective and multi-perspective approaches.

It might be that the IS discipline has developed sufficient pluralism and tolerance of diversity in approaches, methods and topics, and hence can embrace such distinct philosophies, and the distinct research questions they will give rise to, and the distinct techniques that they will apply. On the other hand, if business school culture represents, in the terms of Wall et al. (2015), 'conscious hegemonic participation', such tolerance may be in short supply. An alternative scenario would therefore be schism between a narrow 'MIS' specialisation within the business disciplines and a genuine academic discipline of IS.

6. Conclusion

Our objective was to assess the extent to which the richness of the notion of researcher perspective is reflected in IS research, and to generate insights, identify opportunities, and suggest ways in which those opportunities might be exploited. This article has examined the researcher perspectives adopted by over 500 articles in 3 years of the IS discipline's leading journals. We selected as our focus the AIS Basket of 8 journals because of the influence they have, particularly on entrant and early-career researchers. Our study has revealed two dominant characteristics of such research: adoption of a single perspective (rather than a balance between that and dual- and multi-perspective research); and adoption of the system sponsor's perspective (rather than a balance between that and the perspectives of other stakeholders).

The content analysis technique adopted in this research provides a basis for the examination of researcher perspective in other journals and in conferences, in particular sectors of industry and government, and in particular research domains.

The choice of researcher perspective pervades research activities, strongly influencing conception, design, formulation of research questions and expression of results. We contend firstly that choices of researcher perspective should be conscious rather than accidental, and secondly that those choices should be much more diverse.

This study has enabled the identification and delineation of gaps and opportunities. Many opportunities for research exist that have been under-utilised, and that are capable of delivering value to industry, government and people, of both an economic and a social nature. The first area is single-perspective research that adopts the perspective of a stakeholder other than the system sponsor, such as suppliers, customers and employees, and non-participant usees. A second is dual-perspective research, reflecting the interests of both sides of dyads, and delivering insights into the needs and behaviour of both rather than of just one of them. Another is multi-perspective research into supply chains and industry networks, directly addressing the factors that underlie 'win-win-win' strategies. A fourth is multi-perspective research in support of public policy decision-making.

Dual-perspective and multi-perspective research are more demanding than single-perspective research. To enable adequate rigour to be achieved, and hence the delivery of reliable rather than haphazard insights, there is a need for development, application and refinement of a richer range of research techniques, beyond those currently regarded as being in the mainstream. That represents an opportunity for innovative research into research methods.

We have speculated on the broader and longer-term implications of our work for the IS discipline. Greater adoption by researchers of alternative perspectives, dual perspectives and multiple perspectives may expose a significant proportion of contemporary IS research as representing a low-grade, echo-chamber service that few stakeholders beyond academia value highly. A shift in approaches and techniques may come about, which might be accommodated within the ever-changing scope and style of of the discipline, or may lead to a bifurcation, with the business school culture withdrawing even further inside its walls, and the broader IS discipline pursuing its own course.

A final consideration is the freshly-minted AIS Code of Ethics. This expressly requires the consideration of the interests of stakeholders, not just clients; and a breach of the AIS Code is subject to AIS Bylaw No. 4, with remedial measures including "member censure, suspension or expulsion".

The analysis presented in this article has therefore not only identified important opportunities for IS researchers, but is also directly relevant to IS researchers' obligations. It provides a basis for the work of the AIS Member Conduct Committee in investigating accusations of non-compliance by IS researchers with their professional Code.


Adie J. & Castleman T. (1997) 'Employment relations issues for telework: occupational, organisational and public policy factors' Proc. Bled eConference, June 1997

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

AIS (2019) 'AIS Code of Ethics and Professional Conduct' Association for Information Systems, December 2019, at

Avgerou, C. (2000) Information systems: what sort of science is it?, Omega, 28, 5, 567-579

Baldwin, L.P., Irani, Z. & Love, P.E.D. (2001). Outsourcing Information Systems: Drawing Lessons from a Banking Case Study, European Journal of Information Systems, 10, 1, 15-24.

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

Benlian A. (2015) 'Web Personalization Cues and Their Differential Effects on User Assessments of Website Value' Journal of Management Information Systems 32, 1, 225-260

Cabrera M. & Oezcivelek R. (2008) 'From Applications to Policies: ICT-based Independent Living Services in the EU' Electronic Markets 18, 4, 304-314

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.

Carlsson C. & Walden P. (2015) 'Digital Wellness for Young Elderly: Research Methodology and Technology Adaptation' Proc. Bled eConference, June 2015, at$File/1_Carlsson.pdf

Cha, H.S., Pingry, D.E. & Thatcher, M.E. (2008). Managing the Knowledge Supply Chain: An Organizational Learning Model of Information Technology Offshore Outsourcing, MIS Quarterly, 32, 2, 281-306.

Chatterjee S., Sarker S. & Valacich J.S. (2015) 'The Behavioral Roots of Information Systems Security: Exploring Key Factors Related to Unethical IT Use' Journal of Management Information Systems 31, 4, 49-87

Choudhury V. (1997) 'Strategic Choices in the Development of Interorganizational Information Systems' Information Systems Research 8, 1, 1-24

Clarke R. (1992) 'Extra-Organisational Systems: A Challenge to the Software Engineering Paradigm' Proc. IFIP World Congress, Madrid, September 1992, PrePrint at

Clarke, R. (2015). Not Only Horses Wear Blinkers: The Missing Perspectives in IS Research, Proc. ACIS'15, Adelaide, December 2015, at

Clarke, R. (2016). An Empirical Assessment of Researcher Perspectives, Proc. 29th Bled eConf., Slovenia, June 2016, PrePrint at

Clarke, R. (2017). Content Analysis in Support of Critical Theory Research: How to Deliver an Unwelcome Message Without Being Shot' Proc. 30th Bled eConference, June 2017, PrePrint at

Clarke R. (2019) 'The Challenges Involved in Establishing a Research Technique' Australasian J. of Infor. Syst. 24 (March 2020) 1-22, PrePrint at

Clarke R. (2020) 'Researcher Perspective in Electronic Markets' Electronic Markets 30,1 (Mar 2020) 15-27, PrePrint at

Clarke R. & Davison R.M. (2020) 'Through Whose Eyes? The Critical Concept of Researcher Perspective', Journal of the Association for Information Systems, 21, 2, 483-501, PrePrint at

Clarke R. & Pucihar A. (2013) 'Electronic Interaction Research 1988-2012 through the Lens of the Bled eConference' Electronic Markets 23.4 (December 2013) 271-283, at

Cordoba J.-R. & Midgley G. (2008) 'Beyond organisational agendas: using boundary critique to facilitate the inclusion of societal concerns in information systems planning' European Journal of Information Systems 17, 2, 125-142

Duhan, S., Levy, M. & Powell, P. (2001). Information Systems Strategies in Knowledge-based SMEs: The Role of Core Competencies, European Journal of Information Systems, 10, 1, 25-40.

Earl, M. (2001). Knowledge Management Strategies: Toward a Taxonomy, Journal of Management of Information Systems, 18, 1, 215-233.

Faraj, S., Kudaravalli, S. & Wasko, M. (2015). Leading Collaboration in Online Communities, MIS Quarterly, 39, 2, 393-412.

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

Freeman, R.E. & Reed, D.L. (1983). California Management Review. 25, 3, 88-106.

Freeman, R.E. (1984). Strategic Management: A Stakeholder Approach. Cambridge: Ballinger.

Freeman, R.E. (1994). The Politics of Stakeholder Theory: Some Future Directions, Business Ethics Quarterly, 4, 409-422.

Galliers R.D. & Whitley E.A. (2007) 'Vive les differences? Developing a profile of European information systems research as a basis for international comparisons' Euro. J. of Infor. Syst. 16, 1, 20-35

Gao, G., Greenwood, B.N., Agarwal, R. & McCullough, J.S. (2015). Vocal Minority and Silent Majority: How Do Online Ratings Reflect Population Perceptions of Quality, MIS Quarterly, 39, 3, 565-589.

Grover, V., Jeong, S.R. & Segars, A.H. (1996). Information Systems Effectiveness: The Construct Space and Patterns of Application, Information and Management, 31, 177-191.

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

Hevner, A., S. March, J. Park and S. Ram (2004) Design Science in Information Systems Research, MIS Quarterly, 28, 1, 75-105.

Ifinedo, P. & Nahar, N. (2007). ERP systems Success: An Empirical Analysis of How Two Organizational Stakeholder Groups Prioritize and Evaluate Relevant Measures, Enterprise Information Systems, 1, 1, 25-48.

Jones, S. & Hughes, J. (2001). Understanding IS Evaluation as a Complex Social Process: A Case Study of a UK Local Authority, European Journal of Information Systems, 10, 4, 189-203.

Jung, Y. (2014). What a Smartphone is to me: Understanding User Values in Using Smartphones, Information Systems Journal, 24, 4, 299-321.

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, 637-667.

Laakso M., Lindman J., Shen C., Nyman L. & Bjoerk B.-C. (2017) 'Research output availability on academic social networks: implications for stakeholders in academic publishing' Electronic Markets 27, 2, 125-133

Lamp J. (1995-) 'The Index of Information Systems Journals' Deakin University, since 1995, at

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, 119-150.

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

Ling C.L.M., Pan S.L., Ractham P. & Kaewkitipong L. (2015) 'ICT-Enabled Community Empowerment in Crisis Response: Social Media in Thailand Flooding 2011' Journal of the Association for Information Systems 16, 3, 174-212

Lowry, P.J., Moody, G.D., Gaskin, J., Galletta, D.F., Humpherys, S., Barlow, J.B. & Wilson, D.W. (2014). Evaluating Journal Quality and the Association for Information Systems Senior Scholars' Journal Basket Via Bibliometric Measures: Do Expert Journal Assessments Add Value? MIS Quarterly 37, 4, 993-1012.

Lyytinen, K., Mathiassen, L. & Ropponen, J. (1998). Attention Shaping and Software Risk - A Categorical Analysis of Four Classical Approaches, Information Systems Research, 9, 3, 233-255.

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, 447-464.

Majchrzak A. & Markus M.L. (2014) 'Methods for policy research: Taking socially responsible action' 2nd ed., Sage, 2014

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, 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, 278-310.

Mendelow, A.L. (1981). Environmental Scanning - The Impact of the Stakeholder Concept . Proc. ICIS 1981, 20.

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, 853-886

Mitroff, I.I. & Mason, R.O. (1980). A logic for strategic management, Human Systems Management 1, 2, 115-126.

Niederman F., Clarke R., Applegate L., King J.L., Beck R. & Majchrzak A. (2017) 'IS Research and Policy: Notes From the 2015 ICIS Senior Scholar's Forum' Commun AIS 40, 5 (February 2017), PrePrint at

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, 245-259.

Orlikowski, W. and J. Baroudi (1991) Studying information technology in organizations: Research approaches and assumptions, Information systems research, 2, 1, 1-28.

Orlikowski, W. and S. Scott (2008) Sociomateriality: Challenging the Separation of Technology, Work and Organization, Annals of the Academy of Management, 2, 1, 433-474.

Porter M.E. (1980) 'Competitive Strategy' The Free Press, New York, 1980

Porter M.E. (1985) 'Competitive Advantage' The Free Press, New York, 1985

Pouloudi N., Currie W. & Whitley E.A. (2016) 'Entangled Stakeholder Roles and Perceptions in Health Information Systems: A Longitudinal Study of the UK NHS N3 Network' Journal of the Association for Information Systems 17, 2, 107-161, at

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, 1-14.

Randall, D., Hughes, J., O'Brien, J., Rouncefield, M. & Tolmie, P. (2001). 'Memories are Made of This': Explicating Organisational Knowledge and Memory, European Journal of Information Systems, European Journal of Information Systems, 10, 2, 113-121.

Rutner, P.S., Hardgrave, B.C. & McKnight, H.D. (2008). Emotional Dissonance and the Information Technology Professional, MIS Quarterly, 32, 3, 635-652.

Saravanan A. & Balasundaram S. R. (2013) 'Integrated location aware medical care services' Electronic Markets 23, 1, 5-14

Seddon, P.B., Staples, D.S., Patnayakuni, R. & Bowtell, M.J. (1998). The IS Effectiveness Matrix: The Importance of Stakeholder and System in Measuring IS Success, Proceedings of the 1998 International Conference on Information Systems (165-176), Atlanta: Association for Information Systems Atlanta.

Seddon, P.B., Staples, D.S., Patnayakuni, R. & Bowtell, M.J. (1999). Dimensions of Information System Success, Communications of the AIS, 2, 20, 1-39.

Sedera, D., Gable, G.G. & Chan, T. (2004). Measuring Enterprise Systems Success: The Importance of a Multiple Stakeholder Perspective. 12th European Conference on Information Systems, AISeL.

Shankar, V., Urban, G.L. & Sultan, F. (2002). Online Trust: A Stakeholder Perspective, Concepts, Implications, and Future Directions, Journal of Strategic Information Systems, 11, 3-4, 325-344.

Singh, H., Aggarwal, R. & Cojuharenco, I. (2015). Strike a Happy Medium: The Effect of IT Knowledge on Venture Capitalists' Overconfidence in IT Investments, MIS Quarterly, 39, 4, 887-908.

Singletary, L., Pawlowski, S. & Watson, E. (2003). What is Applications Integration? Understanding the Perspectives of Managers, IT Professionals, and End Users, Proceedings of the 9th Americas Conference on Information Systems (486-493), AISeL.

Smith H.J., Dinev T. & Xu H. (2011) 'Information Privacy Research: An Interdisciplinary Review' MIS Quarterly 35, 4, 989-1015

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, 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, 128-145.

Tarafdar, M. & Davison, R.M. (2018). Research in Information Systems: Intra-disciplinary and Inter-disciplinary Approaches, Journal of the Association for Information Systems 19, 6, 523-551..

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

Whitley E.A. & Hosein I.R. (2008). Doing the politics of technological decision making: due process and the debate about identity cards in the U.K., European Journal of Information Systems 17, 6, 668-677.

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 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.

Wanying Jia 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: 16 December 2018 - Last Amended: 9 July 2020 by Roger Clarke - Site Last Verified: 15 February 2009
This document is at
Mail to Webmaster   -    © Xamax Consultancy Pty Ltd, 1995-2022   -    Privacy Policy