RECONSTRUCTING THE GIANT:
ON THE IMPORTANCE OF RIGOUR IN DOCUMENTING
THE LITERATURE SEARCH PROCESS
vom Brocke, Jan, Martin Hilti Chair of Business Process Management, Institute of Informa-
tion Systems, University of Liechtenstein, Fuerst-Franz-Josef Strasse 21, 9490 Vaduz,
Principality of Liechtenstein, firstname.lastname@example.org
Simons, Alexander, Martin Hilti Chair of Business Process Management, Institute of Informa-
tion Systems, University of Liechtenstein, Fuerst-Franz-Josef Strasse 21, 9490 Vaduz,
Principality of Liechtenstein, email@example.com
Niehaves, Björn, European Research Center for Information Systems, University of Muenster,
Leonardo-Campus 3, 48149 Muenster, Germany, firstname.lastname@example.org
Riemer, Kai, European Research Center for Information Systems, University of Muenster,
Leonardo-Campus 3, 48149 Muenster, Germany, email@example.com
Plattfaut, Ralf, European Research Center for Information Systems, University of Muenster,
Leonardo-Campus 3, 48149 Muenster, Germany, firstname.lastname@example.org
Cleven, Anne, Institute of Information Management, University of St.Gallen, Mueller-Fried-
berg-Strasse 8, 9000 St.Gallen, Switzerland, email@example.com
Science is a cumulative endeavour as new knowledge is often created in the process of interpreting
and combining existing knowledge. This is why literature reviews have long played a decisive role in
scholarship. The quality of literature reviews is particularly determined by the literature search proc-
ess. As Sir Isaac Newton eminently put it: “If I can see further, it is because I am standing on the
shoulders of giants.” Drawing on this metaphor, the goal of writing a literature review is to recon-
struct the giant of accumulated knowledge in a specific domain. And in doing so, a literature search
represents the fundamental first step that makes up the giant’s skeleton and largely determines its re-
construction in the subsequent literature analysis. In this paper, we argue that the process of search-
ing the literature must be comprehensibly described. Only then can readers assess the exhaustiveness
of the review and other scholars in the field can more confidently (re)use the results in their own re-
search. We set out to explore the methodological rigour of literature review articles published in ten
major information systems (IS) journals and show that many of these reviews do not thoroughly docu-
ment the process of literature search. The results drawn from our analysis lead us to call for more
rigour in documenting the literature search process and to present guidelines for crafting a literature
review and search in the IS domain.
Keywords: literature review, literature search, rigour, IS journals, IS community, research methodol-
Reviewing the literature has been proposed as a scientific profession a long time ago (Garfield 1977;
as cited in Garfield 1987, p. 113), since it represents an “essential first step and foundation when un-
dertaking a research project” (Baker 2000, p. 219). A literature review seeks to uncover the sources
relevant to a topic under study and, thus, makes a vital contribution to the relevance and rigour of re-
search: On the one hand, relevance is improved by avoiding the reinvestigation of what is already
known (cf. Baker 2000, p. 219). On the other hand, rigour is derived from an effective use of the exist-
ing knowledge base (cf. Hevner et al. 2004, p. 88). Hence, it is undisputed that literature reviews gen-
erally play a central role in scholarship (cf. e.g., Cooper 1988, pp. 104f.) and in information systems
(IS) research in particular (cf. e.g., Levy and Ellis 2006, pp. 181f.; Webster and Watson 2002,
pp. xiii f.). However, as the term ‘review’ is one of the “more ambiguous” terms in scholarship (Gar-
field 1987, p. 114), there is still a significant confusion about the structure and format of literature re-
views (Webster and Watson 2002, p. xiv). Nonetheless, it seems clear that in particular the process of
literature search plays a fundamental role in crafting a thorough review on a topic (Zorn and Campbell
2006, p. 174).
According to Webster and Watson (2002), a literature search in essence comprises the querying of
scholarly databases using keywords and backward or forward searches on the basis of relevant articles.
Whereas backward search means reviewing the references of the articles yielded from the keyword
search, forward search, in turn, refers to reviewing additional sources that have cited the article
(ebenda, p. xvi; also cf. Levy and Ellis 2006, pp. 190ff.). The search process is a challenging part of an
IS literature review, as it should include “all sources that contain IS research publications” (Levy and
Ellis 2006, p. 183). However, searching for literature is extremely complicated in an emerging field
such as IS, because an incredible (and still increasing) number of articles are published in a wide range
of sources every year: For example, in 2003 Peffers and Ya identified 326 journals that publish IS re-
search (p. 65), while the Index of Information Systems Journals lists 647 active IS journals today.
Moreover, Peffers and Hui (2003, p. 168) found that between 1997 and 2001 about 38% more articles
were published in ten “pure” IS journals (following Walstrom and Hardgrave 2001, p. 122) when
compared to 1987–1991. As a result, conducting a literature search in IS can turn into a Sisyphean
task, so that, quite necessarily, a multitude of work has to be omitted in the search process (cf. Cooper
1988, p. 114).
Against this backdrop, we argue that the process of excluding sources (and including respectively) has
to be made as transparent as possible in order for the review to proof credibility. Only then are readers
able to assess the exhaustiveness of a review and other scholars in the field can more confidently
(re)use the results in their own research. Therefore, we set out to explore the methodological rigour of
literature reviews in the IS domain. In doing so, we apply typical evaluation criteria, such as reliability
and validity, for analysing review articles published in the ten top-ranked IS journals, according to the
consolidated list provided by the Association for Information Systems (AIS)
. We find a surprisingly
large number of review publications that do not provide any detail on the underlying literature search.
The remainder of this paper is structured as follows. In the next section, we give an overview of our
study and spell out our research approach. In section 3, we then present the results of our analysis of
selected literature reviews published in the IS domain. In doing so, we particularly point out shortcom-
ings referring to the documentation of the literature search process. In section 4, we discuss our results
and derive three propositions that aim at providing explanations for our results. Implications from our
research are subsequently presented in the form of both guidelines for conducting IS literature re-
views, in particular the process of searching the literature, and a plea for more rigour in crafting and
documenting a literature search (section 5). We conclude with a short summary and discussion of our
work (section 6).
2 STUDY OVERVIEW
A literature review can be defined as “a summary of a subject field that supports the identification of
specific research questions” (Rowley and Slack 2004, p. 31). The primary objective of this paper is to
explore the methodological rigour in conducting literature reviews as part of IS research. More spe-
cifically, we focus on the process of literature search (cf. Levy and Ellis 2006, pp. 185ff.), which in-
volves both the identification of high quality papers and the evaluation of their applicability to the
study. Similar to ‘traditional’ research methods or studies, the term ‘rigour’ refers in particular to the
reliability and validity of the search process (cf. Levy and Ellis 2006, pp. 183ff.; Pateli and Giaglis
2004, p. 304). As such, validity characterises the degree to which the literature search accurately un-
covers the sources that the reviewer is attempting to collect. The basic question is ‘Does the reviewer
search right?’; it reflects decisions such as the selection of databases, publications (i.e. journals, con-
ferences or books) and keywords, as well as the period covered, the articles considered in the literature
search, and the application of backward or forward searches (Cooper 1988, p. 114; Levy and Ellis
2006, pp. 185ff.; Torraco 2005, p. 360; Webster and Watson 2002, p. xvi). Reliability, in turn, de-
scribes the replicability of the search process, hence, making it substantial for any review article to
comprehensively document the literature search. “The literature is the data of an integrative literature
review” and “learning about [it] and how it was obtained, including the keywords and databases used,
is of particular interest to readers, who may wonder if the literature they are familiar with was exam-
ined” (Torraco 2005, p. 360). It is also obvious that evaluating the exhaustiveness of a literature re-
view is complicated without a thorough documentation of the search process. That being said, other
researchers in the field can hardly ground their own work on the review without sufficient knowledge
on where and how authors have already searched for literature.
Consequently, the guiding research question of this paper is ‘Do IS researchers comprehensibly
document the process of literature search in their review articles?’ and, therefore, we investigate the
degree of methodological rigour, i.e. the replicability and evidence, of IS reviews. In other words, we
review IS review literature. Our own process of literature search and the results drawn from the analy-
sis are described in the next section.
3 REVIEWING IS REVIEW LITERATURE
As commonly recommended, in the following we focus on review articles of high quality (Rowley and
Slack 2004, p. 32). However, identifying high quality IS literature is complicated as there is not only a
vast amount of potentially relevant sources but also a great deal of literature of diverse quality (Levy
and Ellis 2006, pp. 183 & 185ff.). To this end, IS scholars are frequently evaluating the quality of IS
journals in order to provide rankings, e.g. Ferratt et al. (2007, p. 716), Hardgrave and Walstrom (1997,
pp. 121f.), Lowry et al. (2004, pp. 52ff.), Peffers and Ya (2003, p. 70), and Willcocks et al. (2008,
pp. 165f.). A number of those rankings have been synthesised by Carol Saunders for the AIS, which
resulted in a comprehensive list of more than 100 top-tier IS journals. We decided to include in our
literature review the ten top-ranked, peer-reviewed IS journals according to this consolidated list.
However, as Harvard Business Review (HBR) (#8) is not a peer-reviewed journal and because the
IEEE Transactions (#9) subsume various journals of differing quality and relevance for IS research
(e.g. the IEEE Transactions on NanoBioscience), these journals were not considered in our review.
This selection led us to explore the ten journals displayed in Figure 1. The figure also outlines the ac-
cessed databases as well as the searching functionality applied and the period covered in the review.
The key phrase we used in all searches was ‘literature review.’
Figure 1. Considered journals
As outlined in the previous section, a literature search involves both identifying and evaluating schol-
arly literature and, accordingly, we applied a similar procedure in our own review. As the search
phrase ‘literature review’ has been used in a range of contributions, many of which cannot be labelled
as review articles, the contributions identified by keyword search (‘hits,’ cf. Figure 1) have subse-
quently been evaluated, based on their abstracts, in order to assess their relevance for this study. The
then remaining articles became the basis of our review (‘reviewed’). Subsequently, we analysed
whether the identified review articles meet, i.e. document, the requirements explained in the previous
section, namely: the (number of) articles considered in the review, the period covered, the (number of)
journals and databases explored, the keywords used for the database and/or journal search, and, fi-
nally, whether a backward and/or forward search were conducted. The results drawn from our analysis
are summarised in Figure 2; they show that 6 articles – or one fourth of the examined review litera-
ture – do not provide any information on the underlying search process. At least, 15 publications accu-
rately document how many articles were included in the review. However, among these, there are only
11 reviews that state precisely, which articles were included. Whereas, more or less, half of the exam-
ined articles explicate the examined period of time, the number of databases queried has merely been
stated in 7 articles. However, in many cases it again remains unclear, which databases were actually
accessed. For example, Melville et al. (2004, p. 322) searched journal databases which “included”
Business Source Premier and JSTOR and Pateli and Giaglis (2004, p. 304) explored “several” sources
“such as” ScienceDirect, JSTOR, and InterScience. Perhaps even more notably, many reviewers con-
ducted a journal search instead of a database search, though there is “no justification for searching by
journal instead of searching by topic” (Anonymous; cited in Webster and Watson 2002, p. xvi), except
when the goal of the review gives sufficient reasoning for such a proceeding (e.g. Jasperson et al.
2002, p. 403, or the review at hand).
All in all, some inaccuracies in documenting the search process can be found in all sources. For exam-
ple, Leidner and Kayworth (2006) – whose review is still among the best-documented ones – exam-
ined “the leading journals in our field dating back to the early 1990s.” They apply search phrases
These numbers have been checked and double-checked on 2009-01-15. However, when finalising this paper and reviewing
the results once more on 2009-03-15, it turned out that some of them have considerably changed, in particular when again
searching Informs (ISR and MS) and ScienceDirect (DSS). Referring to ScienceDirect, these differences can, at least partly,
be reasoned by new issues released in the meantime. However, the results gained from the two Informs queries differ so sig-
nificantly (ISR: 30 vs. 46 hits; MS: 108 vs. 228 hits) that we are likely to assume that the underlying searching mechanisms
have somehow been changed in recent times (since our license contracts with both providers definitely stayed the same).
Therefore, please note that the numbers described in Figure 1 have been gained on 2009-01-15.
Since 2001, the MISQ provides a special review section. Therefore, keyword search has not been applied for searching
MISQ. However, all articles published in the MISQ review section have been considered in this review.
“such […] as ‘IT culture’, ‘information systems culture,’ and ‘IT values’” and conducted a search of
ABI/Inform and Business Source Premier and “similar searches” in ScienceDirect. Furthermore, they
“looked through” the references of “key articles” to not “overlook other articles” (pp. 360ff.). That
being said, though they applied a backward search (as one of only a few articles), it is not fully replic-
able, as is the entire search process. However, most striking is the fact that only 2 articles comprehen-
sibly document the underlying search phrases (Gerwin and Barrowman 2002, p. 942, and Ngai and
Gunasekaran 2007, p. 4). So even though we firmly believe that all the above listed contributions rep-
resent fundamental and way-leading reviews in their particular fields, we herein question whether it is
in fact sufficient to state that “key words from our definition of IT business value” (Melville et al.
2004, p. 322) were used or that the search was conducted “by using relevant keywords” (Xiao and
Benbasat 2007, p. 140). In the same way, Pateli and Giaglis (2004, p. 304) claim that “the selection
phase is critical, since decisions made at this stage undoubtedly have a considerable impact on the va-
lidity of the literature review results.”
(Number of) Articles Period Covered (Number of) Journals
15 1975-1990 6
1980-2007 database search only
2000-2003 database search only
– – –
– – –
– database search only
– – –
no constraint –
– – –
~200 1988-1998 > 9
– – –
early 1990s-2004 > 6
no constraint > 11
– – –
– – –
early 1990s-? –
1.485 – > 11 +/–
> 100 –
Eierman et al. (1995)
Guo et al. (2008)
Lin et al. (2008)
Ngai and Gunasekaran (2007)
Pinsonneault and Kraemer (1989)
Powell and Johnson (1995)
Tung and Turban (1998)
Pateli and Giaglis (2004)
JMIS Fjermestad and Hiltz (1998)
Elmaghraby and Keskinocak (2003)
Gerwin and Barrowman (2002)
Krishnan and Ulrich (2001)
CACM Glass et al. (2004)
Alavi and Leidner (2001)
Jasperson et al. (2002)
Leidner and Kayworth (2006)
Melville et al. (2004)
Piccoli and Ives (2005)
Wade and Hulland (2004)
Xiao and Benbasat (2007)
journal search only
journal search only
journal search only
– –– –
journal search only –– –
journal search only
journal search only
> 3 +/–
1992-2007 –90 +Petter et al. (2008) –+/– –3 +
Figure 2. Results from analysis
Concluding, even though a review article, which considers “everything that we could locate, published
in English and available by mid-1998” (Fjermestad and Hiltz 1998, p. 9) is most likely to be valid,
whenever the search process is only described very briefly, the search cannot be replicated and, thus, it
can hardly be considered rigour. Drawing on these results, the question emerges: ‘Why is it that many
IS researchers do not comprehensively document their literature search process?’
When crafting a literature review on a topic, one should be aware of the fact that, unlike for other em-
pirical studies, there are only few explicit methods or standardized guidelines (cf. Bem 1995; Jackson
1980, p. 440; Torraco 2005, p. 359). In 2002, Webster and Watson remarked: “as the initial senior edi-
tors for MISQ Review, we quickly learned that many IS scholars are not familiar with the structure and
format of reviews” (p. xiv). This leads us to formulate a first proposition for explaining our results: (1)
IS researchers cannot refer to established guidelines for documenting the literature search process.
Our second proposition provides an alternative reasoning: (2) IS researchers are not fully aware of the
importance of rigorously documenting the literature search. Please note that it is not our intention to
imply that IS researchers are unwilling to conduct their search process in a rigorous way – in particular
referring to the outstanding contributions described above. However, conducting and documenting a
literature search are two sides of the same coin and perhaps IS authors sometimes in fact prefer ‘the
pears to the apples,’ i.e. they rather put emphasis on rigorously summarizing and synthesizing findings
gained from the literature search than on documenting the long way they walked along for uncovering
Finally, our analysis also revealed that review articles in IS research differ considerably in terms of
length, reaching from six pages (Glass et al. 2004) to 143 pages (Fjermestadt and Hiltz 1998). Fur-
thermore, there are many review articles that are not even longer than twenty pages. However, it is
undisputed that literature reviews commonly require far more pages than ‘regular’ articles (cf. e.g., the
MISQ submission guidelines). Consequently, the question arises ‘Why are so many review articles that
short (and exclude the process of searching the literature), assuming that IS researchers know how
(ad 1) and are willing (ad 2) to rigorously document their literature search process?’ We believe the
answer is most likely to be found in the submission guidelines and page limitations provided by vari-
ous IS outlets. More specifically, information on the literature search process, which exists in the be-
ginning, might be excluded during the subsequent editing process in order to free space for addressing
the reviewers’ comments. Only three of the journals we examined provide a special review section
(CACM, AI and MISQ) and only another three journals (DSCI, EJIS and ISR), more or less, welcome
review articles in their editorial statements. Please also note that – according to the results gained from
our analysis – in the case of four of the journals in our sample no review articles were published at all.
This leads us to question whether literature reviews play the key role they undoubtedly deserve in IS
research and to formulate a third proposition: (3) It is the conditions of the publication process that
often prevent a detailed description of the literature search. For example, Kari and Rozenberg (2008,
p. 83) acknowledge this perception by stating that “the upper-bound placed on the number of  ref-
erences” turned out to be a “real limitation” for their review published in the CACM.
In summary, we conclude that IS reviewers may sometimes be either (1) unable to refer to adequate
guidelines for rigorously documenting the literature search, (2) not fully aware of its importance or (3)
hindered to transparently describe their literature search in its full extent due to editorial constraints.
Here, we do not aim to evaluate our propositions, but rather formulate both guidelines for crafting an
IS literature review and search (ad 1) as well as a plea for more comprehensive documentation of the
search process (ad 2 & 3).
5.1 Guidelines for Literature Reviews
In response to proposition 1, we propose in the following a framework for conducting IS literature re-
views, with particular focus on the process of searching the literature. The framework is displayed in
Figure 3 and reflects a “circularity that exists when […] undertaking a literature review” (Baker 2000,
p. 221). That being said, as things use to change and knowledge continuously grows, literature reviews
often become out-of-vogue after a certain time, giving reason for an extension and update of the re-
view (cf. Pervan 1998, p. 158).
A major challenge in reviewing the literature lies in defining an appropriate scope and flavour of the
review (phase I). Reviews can be critical, interpretive, speculative, state of the art, and historical and
Note that this review could not be identified following our search strategy (limitations are discussed in the next section).
can vary referring to subject matter, period covered, and degree of coverage of sources (Manten 1973;
Woodward 1972; as cited in Garfield 1987, p. 114). Furthermore, literature reviews can serve a wide
range of, sometimes very different, purposes, reaching from gaining new and synthesising existing
research outcomes to identifying research methodologies or techniques commonly used in a field (cf.
Hart 1998, pp. 27ff.). In order to clearly define the scope of a review, we propose to draw on an estab-
lished taxonomy for literature reviews presented by Cooper (cf. Cooper 1988, pp. 109ff., and Figure 4
in the follow-up). Cooper’s taxonomy is comprised of six constituent characteristics, each containing
certain categories, some of which are mutually exclusive (perspective and coverage), while others can
be combined (audience, organisation, goal, and focus). The focus (1) of a literature review is con-
cerned with what is of utmost importance to the reviewer. Most literature reviews focus on research
outcomes, research methods, theories, and/or applications (cf. Bem 1995; Torraco 2005, p. 361). Com-
mon goals (2) of literature reviews include summarising, criticising, and/or integrating findings (cf.
Jackson 1980, p. 438). For organising a literature review (3), Cooper suggests a historical, conceptual
or methodological structure. The perspective (4) of a review reflects whether a certain position is es-
poused or not; the audience (5) particularly determines the writing style of the author(s) (cf. Bem
1995, pp. 173f.).
Figure 3. Framework for literature reviewing
This paper is based on the perception that in particular the degree of coverage of sources (6) is crucial
for reviewing the literature on a topic. According to Cooper, four levels of coverage can be distin-
guished, namely: exhaustive (including the entirety of literature on a topic or at least most of it), ex-
haustive with selective citation (considering all the relevant sources, but describing only a sample),
representative (including only a sample that typifies larger groups of articles), and central (reviewing
the literature pivotal to a topic) (Cooper 1988, pp. 110f.).
Figure 4. Taxonomy of literature reviews (following Cooper 1988, p. 109)
While the above framework does not provide immediate answers to the questions of literature search,
its application is a necessary first step of clarification in any literature review, which bears implica-
tions for the later search process. An exemplary application of Cooper’s taxonomy is also given in
Figure 4 by highlighting categories that characterise our own literature review presented in section 2:
We did not focus (1) on the research outcomes or theories described or applied in the above analysed
articles, but rather on their underlying research methodology. We criticized (2) the way these reviews
document the literature search process, since it was our position (4) that such documentation is crucial
to ensure the replicability of the literature search. As the purpose of our review was not to summarise
or synthesise, but rather to evaluate scholarly literature, the categories proposed by Cooper relating to
organisation (3) can hardly be applied to our case. It is hoped that our results are of some value for the
whole IS community (5) – even though we did not consider all IS review articles ever published in our
study, but restricted the analysis to a small sample of literature reviews only. However, this proceeding
can be reasoned by the fact that it was our explicit objective to analyse high-quality review articles
that may be considered as representative (6) for the IS domain.
In a next step, attention should be paid to the fact that a review must begin with “a broad conception of
what is known about the topic and potential areas where knowledge may be needed” (phase II) (Tor-
raco 2005, p. 359). Therefore, working definitions of the key terms should be provided at this point
(Zorn and Campbell 2006, p. 175). Baker (2000, p. 222) suggests that one should firstly consult “those
sources most likely to contain a summary or overview of the key issues relevant to a subject,” such as
seminal textbooks, encyclopaedias, or handbooks. A reasonable way for identifying key concepts is
represented by concept mapping, which also provides the opportunity to uncover relevant search terms
(in particular related concepts or synonyms and homonyms) that can be applied in the subsequent lit-
erature search (cf. Rowley and Slack 2004, p. 36). Accordingly, we began our study by consulting the
seminal textbooks on literature reviews by Arlene Fink (2005) and Christopher Hart (1998) and intro-
duced a working definition of the term ‘literature review’ in section 2. However, the various synonyms
of the term ‘literature review’, such as ‘meta-analysis’ or ‘research synthesis,’ have been disregarded
in our study. Therefore, it is possible (or indeed even likely) that several articles which may be la-
belled as ‘literature reviews’ have not been included in our study, which again shows that our search
strategy can hardly be labelled as exhaustive – even referring to the small sample of IS journals exam-
As we already noted, the search process (phase III) involves database, keyword, backward, and for-
ward search, as well as an ongoing evaluation of sources (cf. Figure 5).
Figure 5. Literature search process
As they have typically been peer-refereed before publication, it is commonly recommended to focus
on articles published in scholarly journals (Rowley and Slack 2004, p. 32) or proceedings of renowned
conferences (Webster and Watson 2002, p. xvi). However, one should note that the quality of contri-
butions in conference proceedings is usually considered lower and less mature than those in journals
(cf. Levy and Ellis 2006, p. 187). Thus, authors who intend to include conference articles in their re-
views should concentrate on the better ones (e.g., Walstrom and Hardgrave (2001, p. 121) and Will-
cocks et al. (2008, p. 166) provide rankings for IS conferences). Consequently, the identification of
journals makes up the first sub phase of our framework – even though we would agree that it rather
makes sense to query scholarly databases allowing for a topic-based search (cf. Webster and Watson
2002, p. xvi). However, one challenge definitely lies in identifying proper databases. We propose to
search those databases (sub phase 2) providing access to the leading IS journals (identified in sub
phase 1) – a proceeding which finally allows for ensuring that all the top-tier sources are included in
the review. The identified databases have then to be queried on the basis of a keyword search (sub
phase 3). It is commonly recommended to use a precise (set of) search phrase(s) in order to exclude
contributions covering topics or research questions which are not necessarily relevant (Rowley and
Slack 2004, p. 35). Querying EBSCOhost using the search term ‘literature review,’ for example,
reveals hundreds of possible information sources; when adding the term ‘information systems’
however, the amount of literature is significantly reduced. By also adding the supplementary search
phrase ‘Europe,’ the huge pool of potentially relevant articles is finally restricted to merely a handful
of papers. This example illustrates that keyword search is crucial, since the selection and combination
of the search phrases sets “the parameters of the research itself” (Baker 2000, p. 222). Thus, particu-
larly the applied keywords have to be documented precisely, so that other scholars can evaluate
whether they sufficiently match the topic under investigation. According to Webster and Watson
(2002, p. xvi), the process of backward search refers to reviewing older literature cited in the articles
yielded from the keyword search and forward search means reviewing additional sources that have
cited the article (sub phase 4), e.g. by querying the ISI Web of Knowledge by Thomson Reuters (note
that Levy and Ellis (2006, pp. 190ff.) discuss additional forms of backward and forward search in de-
tail). Evaluation in all phases means limiting the amount of literature identified by keyword search as
well as backward and forward search to only those articles relevant to the topic at hand. We therefore
propose an evaluation of the articles’ contents, which may mean to analyse their titles, abstracts or
even full texts. As described above, the review at hand is based on applying a keyword search (‘litera-
ture review’) to a sample of ten journals. That being said, though these journals were accessed via
online databases, a ‘pure‘ database search has not been conducted. Moreover, backward or forward
searches were also not part of our literature search strategy.
After collecting sufficient literature on a topic it has to be analysed and synthesised (phase IV). As we
focus on the search process, we herein only briefly touch upon this phase. For the analysis, a concept
matrix, as developed by Salipante et al. (1982) and adapted for IS literature reviews by Webster and
Watson (2002, p. xvii), can be used, which subdivides topic-related concepts into different units of
analysis (cf. Figure 6). This allows for arranging, discussing, and synthesising prior research.
Figure 6. Concept matrix (Webster and Watson 2002, p. xvii)
However, as underpinned by several IS reviews (e.g., Ahuja 2002, pp. 30ff.), the synthesis of literature
is further expected to result into a research agenda (phase V), comprised of sharper and more insight-
ful questions for future research (Webster and Watson 2002, p. xix). The research agenda provides the
basis for extending the review in order for the IS community to keep up-to-date and it may be devel-
oped based on the proposed concept matrix. That being said, certain fields of the matrix, which remain
‘blank’ during a literature study, often highlight research areas that are significantly under-researched.
5.2 A Plea for More Rigour in Searching the Literature
We hope that the above section provides some insights on how to rigorously conduct a literature re-
view and search in the IS domain. However, we did not intend to present ‘yet another guideline’ for
reviewing the literature – we rather consider our explanations as both a rationale for our own literature
study and a response to scholars who are likely to estimate our first proposition to be the most striking
one (‘IS researchers cannot refer to adequate guidelines for rigorously documenting the literature
search’). Nonetheless, we think that most IS scholars – and in particular those very renowned ones
cited above – indeed have a good understanding of how to conduct a literature search. Moreover, we
firmly believe that the literature search process underlying (at least most of) the analysed articles is
certainly a highly rigorous one. However, it is simply not documented rigorously enough, leading us to
question the role the literature search process plays in the IS domain. Thus, one may also consider this
paper as ‘yet another call’ for more rigour in IS research; in this case, for documenting a literature
search. In the following, we appeal to the authors of IS articles (proposition 2) as well as to the editors
of IS journals (proposition 3) to not only conduct and write (or support the writing of) literature re-
views that are of high quality, i.e. reviews that are thoroughly crafted, useful and interesting, but also
to allow for adequate documenting of the review process, including the specifics of the search process.
When writing a literature review, we advocate authors to invest in planning and describing the litera-
ture search process in detail. In the majority of cases, literature reviews serve as the means to reveal
open research gaps and are part of a larger research endeavour. Planning and accomplishing the litera-
ture search process in a rigorous manner will help to discover similar research endeavours early and
prevent the researcher from doing redundant work. Moreover, it will allow for a better understanding
of where to find relevant articles in the jungle of different sources available and how to retrieve them
later. Rigorously documenting the literature search in particular means to share these experiences with
the community. As a result, a well-documented review will be gladly used, extended and cited by
other researchers, which will in turn positively affect scientific impact.
We hope for more editors to consider establishing a dedicated literature review category within their
journals. Moreover, the rigorous accomplishment of a literature review should not be impeded by arti-
cle length or reference restrictions. One possible solution may be the publication of two versions of the
same review – a shorter one that contains the major findings and is published in the printed version
and a comprehensive one that outlines the whole literature search process and states how and where
the sources were discovered and which is published online (see Kari and Rozenberg (2008) for an ex-
ample). Moreover, we suggest demanding the documentation of the literature search process as an im-
portant review criterion for judging the quality of literature reviews.
Research is a collaborative endeavour, since each researcher builds on what has been developed and
worked out before. Thus, literature reviews play an important role in scholarship. In this paper we ar-
gued that documenting the literature search process is a crucial part in any review article. However,
drawing on the results gained from an analysis of literature reviews published in ten of the most im-
portant IS outlets, we showed that many of these reviews do not thoroughly document the process of
literature search, e.g. by not comprehensibly explicating the queried databases, the applied keywords
or the examined period of time. This is why we presented both guidelines for conducting a literature
search and review as well as a plea for more rigour in documenting the search process.
Certainly, most IS scholars know what it actually takes to craft high-quality reviews, and we would
agree that our suggestions are likely to further impede this challenging task. But our guidelines do not
intend to imply that conducting a literature search means to uncover and analyse all sources ever pub-
lished. A review that considers only five research papers, but sufficiently states which ones were cho-
sen for whatever sensible reasons, may be of more value to both its authors and the community than a
review that analyses a broad range of contributions, without providing sufficient information on
where, why and what literature was obtained, hence, making it hard to judge its quality and the scope
of its contribution. In addition, a well documented literature search offers the potential to being ex-
tended and transferred to, for instance, other domains, additional journals, or newer/older volumes.
Against this background, search results are better comparable and a well-documented search process
thus provides the basis for a review article to contribute to a cumulative effort of reviewing literature.
Ahuja, M.K. (2002). Women in the information technology profession: a literature review, synthesis
and research agenda. European Journal of Information Systems, 11 (1), 20-34.
Alavi, M. and Leidner, D.E. (2001). Review: Knowledge Management and Knowledge Management
Systems: Conceptual Foundations and Research Issues. MIS Quarterly, 25 (1), 107-136.
Baker, M.J. (2000). Writing a Literature Review. Marketing Review, 1 (2), 219-247.
Bem, D.J. (1995). Writing a Review Article for Psychological Bulletin. Psychological Bulletin,
118 (2), 172-177.
Cooper, H.M. (1988). Organizing knowledge syntheses: A taxonomy of literature reviews. Knowledge
in Society, 1, 104-126.
Eierman, M.A., Niederman, F. and Adams, C. (1995). DSS theory: A model of constructs and rela-
tionships. Decision Support Systems, 14 (1), 1-26.
Elmaghraby, W. and Keskinocak, P. (2003). Dynamic Pricing in the Presence of Inventory Considera-
tions: Research Overview, Current Practices, and Future Directions. Management Science, 49 (10),
Ferratt, T.W., Gorman, M.F., Kanet, J.J. and Salisbury, W.D. (2007). IS Journal Quality Assessment
Using the Author Affiliation Index. Communications of the Association for Information Systems,
19 (34), 710-724.
Fink, A. (2005). Conducting Research Literature Reviews: From the Internet to Paper, 2nd edition,
Sage Publications, London.
Fjermestad, J. and Hiltz, S.R. (1998). An assessment of group support systems experimental research:
methodology and results. Journal of Management Information Systems, 15 (3), 7-149.
Garfield, E. (1977). Proposal for a new profession: scientific reviewer. Essays of an Information Sci-
entist, 3, 84-87.
Garfield, E. (1987). Reviewing Review Literature. Part 1. Definitions and Uses of Reviews. Essays of
an Information Scientist, 10, 113-116.
Gerwin, D. and Barrowman, N.J. (2002). An Evaluation of Research on Integrated Product Develop-
ment. Management Science, 48 (7), 938-953.
Glass, R.L., Ramesh, V. and Vessey, I. (2004). An analysis of research in computing disciplines.
Communications of the ACM, 47 (6), 89-94.
Guo, H., Zhang, J. and Koehler, G.J. (2008). A survey of quantum games. Decision Support Systems,
46 (1), 318-332.
Hardgrave, B.C. and Walstrom, K.A. (1997). Forums for MIS scholars. Communications of the ACM,
40 (11), 119-124.
Hart, C. (1998). Doing a literature review: releasing the social science research imagination, Sage Pub-
Hevner, A.R., March, S.T., Park, J. and Ram, S. (2004). Design Science in Information Systems Re-
search. MIS Quarterly, 28 (1), 75-105.
Jackson, G.B. (1980). Methods for Integrative Reviews. Review of Educational Research, 50 (3),
Jasperson, J., Carte, T.A., Saunders, C.S., Butler, B.S., Croes, H.J.P. and Zheng, W. (2002). Review:
Power and Information Technology Research: A Metatriangulation Review. MIS Quarterly, 26 (4),
Kari, L. and Rozenberg, G. (2008). The many facets of natural computing. Communications of the
ACM, 51 (10), 72-83.
Krishnan, V. and Ulrich, K.T. (2001). Product Development Decisions: A Review of the Literature.
Management Science, 47 (1), 1-21.
Leidner, D.E. and Kayworth, T. (2006). Review: A Review of Culture in Information System Re-
search: Toward a Theory of Information Technology Culture Conflict. MIS Quarterly, 30 (2),
Levy, Y. and Ellis, T.J. (2006). A Systems Approach to Conduct an Effective Literature Review in
Support of Information Systems Research. Informing Science, 9, 181-212.
Lin, C., Standing, C. and Liu, Y. (2008). A model to develop effective virtual teams. Decision Support
Systems, 45 (4), 1031-1045.
Lowry, P.B., Romans, D. and Curtis, A. (2004). Global Journal Prestige and Supporting Disciplines: A
Scientometric Study of Information Systems Journals. Journal of the Association for Information
Systems, 5 (2), 29-77.
Manten, A.A. (1973). Scientific literature review. Scholarly Publishing, 5, 75-89.
Melville, N., Kraemer, K. and Gurbaxani, V. (2004). Review: Information Technology and Organiza-
tional Performance: An Intergrative Model of IT Business Value. MIS Quarterly, 28 (2), 283-322.
Ngai, E.W. and Gunasekaran, A. (2007). A review for mobile commerce research and applications.
Decision Support Systems, 43 (1), 3-15.
Pateli, A.G. and Giaglis, G.M. (2004). A research framework for analysing eBusiness models. Euro-
pean Journal of Information Systems, 13 (4), 302-314.
Peffers, K., Hui, W. (2003). Collaboration and Author Order: Changing Patterns in IS Research.
Communications of the AIS, 11 (10), 166-190.
Peffers, K. and Ya, T. (2003). Identifying and evaluating the universe of outlets for information sys-
tems research: Ranking the journals. Journal of Information Technology Theory and Application,
5 (1), 63-84.
Petter, S., DeLone, W. and McLean E. (2008) Measuring information systems success: models, di-
mensions, measures, and interrelationships. European Journal of Information Systems, 17 (3), 236-
Pervan, G.P. (1998). A review of research in group support systems: leaders, approaches and direc-
tions. Decision Support Systems, 23 (2), 149-159.
Piccoli, G. and Ives, B. (2005). Review: IT-Dependent Strategic Initiatives and Sustained Competitive
Advantage: A Review and Synthesis of the Literature. MIS Quarterly, 29 (4), 747-776.
Pinsonneault, A. and Kraemer, K.L. (1989). The impact of technological support on groups: an as-
sessment of the empirical research. Decision Support Systems, 5 (2), 197-216.
Powell, P.L. and Johnson, J.E.V. (1995). Gender and DSS design: the research implications. Decision
Support Systems, 14 (1), 27-58.
Rowley, J. and Slack, F. (2004). Conducting a literature review. Management Research News, 27 (6),
Salipante, P., Notz, W. and Bigelow, J. (1982). A Matrix Approach to Literature Reviews. Research in
Organizational Behavior, 4, 321-348.
Te’eni, D. (2001). Review. A Cognitive Affective Model of Organizational Communication for De-
signing IT. MIS Quarterly, 25 (2), 251-312.
Torraco, R.J. (2005). Writing integrative literature reviews: Guidelines and examples. Human Re-
source Development Review, 4 (3), 356-367.
Tung, L.-L. and Turban, E. (1998). A proposed research framework for distributed group support sys-
tems. Decision Support Systems, 23 (2), 175-188.
Wade, M. and Hulland, J. (2004). Review: The Resource Based View and Information Systems Re-
search Review, Extension, and Suggestions for Future Research. MIS Quarterly, 28 (1), 107-142.
Walstrom, K.A. and Hardgrave, B.C. (2001). Forums for information systems scholars: III. Informa-
tion & Management, 39 (2), 117-124.
Webster, J. and Watson, R.T. (2002). Analyzing the Past to Prepare For the Future: Writing a Litera-
ture Review. MIS Quarterly, 26 (2), xiii-xxiii.
Willcocks, L., Whitley, E.A. and Avgerou, C. (2008). The ranking of top IS journals: a perspective
from the London School of Economics. European Journal of Information Systems, 17 (2), 163-168.
Woodward, A.M. (1972). Review literature: characteristics, sources and output in 1972. Aslib Pro-
ceedings, 26 (9), 367-376.
Xiao, B. and Benbasat, I. (2007). E-Commerce Product Recommendation Agents: Use, Characteris-
tics, and Impact. MIS Quarterly, 31 (1), 137-209.
Zorn, T. and Campbell, N. (2006). Improving the Writing of Literature Reviews Through a Literature
Integration Exercise. Business Communication Quarterly, 69 (2), 172-183.