Responses to Senior Editor’s, Associate Editor’s, and Reviewers’ Comments

MIS Quarterly MS#A2626

“De-escalating Information Technology Projects:

Lessons from the Denver International Airport”

October 11, 1999

Responses to the Senior Editor’s Comments

Comment / Response or Action Taken
The AE recommends conditional acceptance of your manuscript. I concur with all the AE suggestions and the recommendation. Please address all the reviewer and AE comments. They will improve your paper. I have a few additional comments. My comments focus on the early parts of the manuscript and hence are complementary to those of the AE and the reviewers. / We thank the senior editor for conditionally accepting this manuscript. We have carefully reviewed all the reviewer and AE comments and have made appropriate revisions to address the remaining issue. Thank you for your continued encouragement on this manuscript. We look forward to its publication in MIS Quarterly.
  1. Abstract. You claim the paper reporting an "intensive case study." I would delete the word 'intensive.
/ Done.
  1. I continue to be somewhat disappointed with the lack of depth in the case. You have improved the discussion on process but I find the event facts focus overly on rational aspects and be rather superficial regarding all the political and legal aspects that must have been present in the situation. Clearly, the mayor's rerunning for an office must have influenced his decisions much more than the paper gives credit too. What ever you can do to highlight the political, legal, and work culture (e.g., civil engineering culture) aspects, the more satisfying the case study presentation and the paper.
/ Thanks for this advice. In this revision, we have enriched the case by providing more evidence of some of the conflicts that took place. Specifically, we have emphasized the following areas:
1. The political situation. As you suggest, there were political considerations in that Mayor Webb was expected to run for re-election the following year. The following paragraphs were added:
Although he expected “to be judged on more than just DIA” (Denver Post, Jul 19, 1994), after the delaying the opening of the airport four times, he was aware that his reputation was at stake. As he told the Rocky Mountain News “Politically, how can you go out and tell people that you’ re going to delay the project by a year…” (Page 12).
According to George Dougherty, who served as Denver Airport director until June 1992: “[Throughout the project, United] applied significant pressure and had previously made contributions to [Mayor Webb] political campaign and sponsored fund-raising events. He was not in a position to make a decision counter to their wishes” (Page 15).
2. Legal situation: The following evidence was added:
Immediately after Mayor Webb’s decision to authorize the construction of an alternative manual baggage handling system, United and Continental Airlines as well as BAE geared up for protracted negotiations and possible litigation. Continental maintained that the Mayor’s actions constituted a breach of contract for which it could sue the city or choose to cancel its lease of DIA gates. United urged the city to bring in mediators “because of the deteriorating relationship with BAE” (Page 14).
On August 31, 1994, the Rocky Mountain News reported that in an effort to avoid legal action, the City of Denver had proposed a “stand still” agreement whereby major parties (the City, United Airlines, and BAE) would waive certain previous agreements and rights until the new airport was opened and operational. “Of course,” the reporter emphasized, “the legal departments of these parties are going to be busy until the end of this century with this case” (Page 14).
Additionally, significant sums in legal fees had to be spent to counter the lawsuits and related investigations, which led the city auditor to remark:
“I didn’t realize when everyone talked about DIA was going to mean full employment that what that would mean was full employment for lawyers. I never dreamed that when the airport was completed that we would exchange construction workers for lawyers” (Denver Post, Feb. 28, 1995).(Page 16).
3. Conflict between the city of Denver and BAE: The following material was added.
Mayor Webb notified BAE of a $12,000-a-day penalty for not finishing the baggage system by DIA’s original October 29, 1993 completion date. Webb also demanded that BAE pay for the $50 million conventional tug-and-cart baggage system. Di Fonso, reviewing Mayor Webb’s letter, summed up the situation as follows: “We have gotten to the point with the city that we are literally not talking to each other. Consultants recommended a backup baggage system, and the minute that the decision was made, the city had to defend it. We are left out in limbo.” (Page 13)
  1. I highly encourage you to drop the tutorial like discussion on process and variance research that you now have in the beginning of the paper. There are plenty of other papers who provide this type of tutorial. I found the discussion on variance and process to be distracting and adding little to the paper.
/ The tutorial on process and variance research has been dropped. Instead, a single sentence is used to define each of these terms and a few references are provided as pointers for interested readers.
  1. Please consider including footnotes 1 and 2 as main text in the paper. These sorts of things are more important than outlining the differences between process and variance research that are easily accessible in published articles and books. There are now papers in the literature clearly arguing that the dichotomy of process and variance is an artificial division. Most IS research takes a hybrid form even the papers that claim to be taking a process perspective. The same goes for the arguments between inductive versus deductive. Your approach was hybrid.
/ These two footnotes have been folded into the main text of the paper, per your suggestion.
  1. Introduction, page 2, top paragraph. The last two sentences repeat each other. Drop the bolded sentence and simplify the sentence before it. Thank you!
/ Done.
  1. Please rewrite the first five pages. Drop the methodological stuff (e.g., page 5 and the earlier references to process versus variance, p. 8/first paragraph/last sentence, and so forth) and help the reader understand what the prior research has found that is reviewed in Table 1. The current version does not clearly articulate the contributions of the past research on de-escalation. My recommendation is driven from the point that there is nothing methodologically new in your study, but there is new in terms of your empirical findings, particularly given the previous empirical research on de-escalation. Focus your paper on where your contribution is. The reader needs to understand where the prior research stands on de-escalation so that they can more easily understand and appreciate your new contribution to this literature.
/ The first five pages have been rewritten. Most of the methodological material pertaining to process versus variance theories has been removed. We have retained only enough to motivate why our paper, which takes a process perspective, represents a contribution (in light of the prior research focus on factors). Material has been added to help the reader understand what the prior research has found. The contribution of the manuscript has been more clearly articulated.
  1. page 7. Decision 4. Is "fragment" the right word?
/ In this revision, we have substituted the word “restructure” which is a more appropriate term.
As the AE mentions, the suggested revisions should be doable. We would encourage you to complete the revision by December 31, 1999. Please follow the manuscript preparation guidelines in submitting the next version. Congratulations on your successful revision. The AE and I look forward to the next version of this paper. / Thank you again for your continued encouragement, support, and constructive feedback. We have revised the paper in the manner that you, the AE, and the three reviewers have suggested.

Responses to the Associate Editor’s Comments

Comment / Response or Action Taken
As all three reviewers have noted, the authors have done a commendable job of addressing concerns about the clarity of objectives and organization of the paper. Generally, the paper reads well, and the authors have focused clearly on presenting and supporting a process model of de-escalation. The paper now represents a useful contribution to theory and practice, and I congratulate the authors on their efforts. I tend to agree, however, with reviewer 3 that the paper still needs some revision. I believe the revisions necessitate only a few hours' work but can attack the concerns that reviewer 3 raises and can greatly strengthen the paper. Specifically, I have three concerns: / We thank the associated editor for conditionally accepting our manuscript. In this revision, we have followed the many excellent suggestions you provided. Specifically, we have:
1)clarified the distinction between phase 1 and phase 2
2)removed the propositions
3)clearly separated the case facts from our own analysis and interpretation.
  1. While the four phases help describe events at DIA and a more general process model of de-escalation, I find phase 2 to be vague. In particular, the description of "clarifying the magnitude of the problem," refers to the high cost of keeping DIA closed. It appears, however, that this recognition of cost was happening at about the same time as other events in Phase 1, and that it is the basis for much of the external pressure that the city was experiencing. In addition, it did not appear to result from a purposeful "reexamination of prior course of action," which is how the authors label this phase. Thus, both temporally, and characteristically, the clarifying of the magnitude of the problem, as described here, appears indistinguishable from Phase 1. The second part of the description of phase 2-"redefining the problem"-however, appears to be distinct in nature from Phase 1. This appears to be the legitimate focus of Phase 2.
/ When the problems with the baggage system were first detected, Mayor Webb announced that he would delay the opening of the airport until the baggage system was fully functioning. It was at this juncture that DIA came under investigation and external pressure mounted. Although it can be said that problem recognition occurred in phase 1, the true magnitude and nature of the problem was not apparent to the decision makers until phase 2. It was in this phase that an outside risk management consulting firm was hired to re-examine the prior course of action and to evaluate whether the airport opening delays and added cost would materially affect the airport’s ability to meet operating and debt services when it opened. We have added relevant case facts to clarify the distinction between these two phases.
  1. The propositions are repetitive and do not contribute to the analysis. As Reviewer 3 notes the propositions in almost all cases restated other statements by the authors. In fact, the authors would typically state a point as a conclusion of their analysis prior to stating it as a proposition. (See, for example, page 22: "Given the psychological, social, and organizational forces that can promote and reinforce escalation behavior, feedback must be unambiguously negative and there must often be external pressure in order for problem recognition to occur. Thus we state the following proprosition: Proposition 3. Problem recognition is most likely to occur under conditions of unambiguous negative feedback and external pressure toward withdrawal.")
/ Given that the propositions, as pointed out by you and reviewer 3, were redundant and were not adding to the analysis, we
have removed them in this revision.
  1. The presentation of the case facts actually mixes in analysis, conjecture, and opinion. This is why, I believe, two reviewers found the analysis section to be repetitive. Overall, the authors have done an excellent job of adding evidence, but this section undermines the evidence by also boldly stating the authors' analysis and opinions. The authors do not provide evidence for some of the statements in the case description. For example, the first statement under "Recognizing negative feedback" on page 11 appears to represent author conjecture-there is no corroborating evidence. The authors opine, on the bottom of page 11, that "this failed test represented unambiguously negative feedback." Similarly, the first paragraph on page 12 surmises the reasons for erosion of confidence in BAE but provides no support. Note, too, that at the top of page 14, the authors state that "By redefining the problem, Webb encouraged the exploration of alternative courses of action." This is analytical in nature and leads to the problem of repetitiveness in the analysis section. On page 15 under "Managing Impressions" the authors discuss how Mayor Webb saved face but there is no corroborating evidence that that was either intention or outcome. Again, this is clearly appropriate as analysis, but it isn't a case fact. This practice is repeated on page 16 under "De-institutionalizing the project."
/ We have edited the manuscript to achieve a clear separation between case facts (which are presented on the front-end) and our analysis and interpretation (which is presented on the back-end). Thank you for this excellent suggestion. In making these changes, we found it useful to present a simplified version of our model on the front-end as a means of organizing the case narrative and the key decisions that marked the boundaries between phases. On the back-end, we concentrate on further development of the model by examining the key de-escalation triggering activities in each phase. Here, we move beyond the case facts, offering the results of our own analysis and interpretation as well as relevant information from the literature.
My sense is that this paper would be significantly strengthened if the authors addressed the above three points. The first two are very easy, the third requires some rethinking but again does not represent a major departure from the existing paper.
In more clearly separating case facts from analysis, I believe the authors could address the repetition that the reviewers observed. Specifically, they can note in pages 1-5 that they observed in their data four distinct phases. They can then provide the facts for those four phases without breaking them down into their subparts as the current text does. They can eliminate all analysis and conjecture from the case facts. I would encourage the authors to again add any quotes or paraphrasing of interviewees' comments in the case facts, but I would discourage their own assessments.
Following the case description, they can explain that they found "triggers" and "outcomes" associated with each phase. Currently, they explain this twice. Then they can rely on both existing literature, as they do, and case facts, which they also do, adding in their own analysis of the sequence of events.
I believe the paper will then be ready for publication and I am anxious to see it published in MISQ. / The extensive comments that you provided have been very helpful in further strengthening the manuscript. Thank you for your continued encouragement. We believe we have addressed all of the remaining issues that were raised and look forward to the paper’s publication.

Responses to the Reviewer 1’s Comments

Comment / Response or Action Taken
Congratulations on an excellent revision! Your methodology is much more suited to the type of data you have, and has resulted in a much richer and more insightful paper.
The only comment I have is that you may want to tighten the wording in some of the propositions – words such as "scapegoat", "uninterested third party", etc. - could be stated more formally.
Otherwise, great job. / Thank you for the kind words. We appreciate the positive feedback on our paper.
Given that the propositions were found to be redundant, we have removed them in this revision.

Responses to the Reviewer 2’s Comments

Comment / Response or Action Taken
The authors have done an excellent job re-writing the paper from a process-model perspective. The paper is clear and well written and seems to meet the criteria for an applications article. The section which describes the case (The computerized baggage handling system at DIA) and the section which presents the model (Discussion: revisiting the DIA findings in Light of the de-escalation literature) could probably be combined..... But in the scheme of things, this is trivial since the paper is well organized and reads well.
Congratulations to the authors for a job well done. / Thank you for the kind words. We appreciate the positive feedback on our paper. In this revision, we have tried to address the one remaining issue that you noted.

Responses to the Reviewer 3’s Comments

Comment / Response or Action Taken
The author has responded well to the suggestions made by previous reviews. Obviously a lot of new work has gone into the revised paper, and the paper is now well written, very clearly structured and provides an explicit, explained, detailed process model of de-escalation arising from the single case study. The paper is now informed by a much better sense of clear, limited objectives. All this is significant gain from the previous review process, and the author is to be commended on this achievement. At this stage it would be unhelpful to suggest any fundamentally new direction and I will restrict myself mainly to assessing the degree to which the paper has responded to the reviewers' suggestions, especially those of the SE and AE. However, because the paper is now different in several major ways, it is important to also to assess how far these new departures are sufficient to merit publication in present form. My view is that the paper does still need certain revisions and I will detail these below. / Thank you for the kind words. We appreciate the positive feedback on our paper.