Rise and Fall of an Information Technology Outsourcing Program: a qualitative Analysis of a Troubled Corporate Initiative
Download 1.05 Mb. Pdf ko'rish
|
Rise and Fall of an Information Technology Outsourcing Program A
Data Collection
I selected interview participants and documents for review based upon my knowledge of who was involved in IT outsourcing strategy development, execution, and impact at “Icarus,” the pseudonym used throughout this study for the case organization (see the latter part of the upcoming chapter for a broader overview of Icarus). Maxwell (2005) refers to this category of selection methods as purposeful sampling. Additionally, I used snowballing (Bogdan & Biklen, 44 2007) as a means to augment my interviewee population, thus accounting for individuals falling outside my awareness of those closest to the events of the study. The interviewee initial solicitation email is included in Appendix A. All interviewees signed and received a copy of a consent form (Appendix B) which I also read to each subject before beginning the interview process. Interviews with directors and vice presidents were conducted in their offices at Icarus; all other interviews were held in my office at Icarus. Interviewees were given the option to select a different interview location, although none elected to do so. A categorization of notable interviewee characteristics is in Figure 3.1 below. Figure 3.1. Interviewee Characteristics In order to triangulate my data, my interview sample included IT executives (directors and vice presidents), senior managers, line managers, and non-management IT workers including engineers, business analysts, and project managers. Finally, as I am also a participant in my research, I was able to gather rich ethnographic data through my own observations and notes as an additional, reflexive source of data and further means of triangulation and validation. 45 I interviewed thirty-six unique subjects once and conducted follow-up interviews with fifteen of these subjects to capture any changes or differences in their experiences from the beginning of my data collection to the later phases. In total, I conducted fifty-two semi- structured interviews (see Figure 3.1), each lasting between thirty and sixty minutes over the eighteen-month period between September 2012 and February 2014. I employed a third-party service to transcribe the interviews. I added memos and observer comments (OCs) as I reviewed and reflected on the interviews. Additionally, I used memos to capture my experiences in the outsourcing work and added memos and OCs as I reviewed internal documents, images, and other non-human artifacts relevant to the study. The transcribed interviews yielded over seven hundred pages of single-spaced text. I performed open, focused, and axial coding on this material as my primary means of data analysis. I conducted open coding throughout the eighteen months of data gathering. The follow- up interviews added new sets of codes as interviewees described different experiences since our initial conversations. Given the volume of data gathered, I developed an initial list of over one hundred and fifty codes that I later condensed to fewer than sixty. I reached data saturation as I analyzed data from both the original round of interviews and the later follow-up sessions. Individual interviews occurring near the end of the data collection cycles rarely generated new codes, but did provide data aligned to previously identified codes. As suggested by Charmaz (2009), I used action verbs or gerunds to establish my initial codes. Rather than trying to discretely fit actors into Rogers’s (2003) adopter categories (innovators, early adopters, early majority, late majority, and laggards) I used these labels as guides for coding the sequence of actions grounded in the data. Furthermore, I addressed McMaster’s (2001) criticism of diffusion theory and flattened the strata of interviewees into a 46 binary pair of “executives” and “employees” (or innovators and recipients). As my coding progressed, I compared and contrasted comments made by employees with those made by executives. Evaluating employee-to-employee, executive-to-executive, and employee-to- executive comments provided insights into how these different groups understood their situations. I began to develop larger categories and units of analysis during focused coding. Examining the most significant and frequent data from initial coding, I categorized data into themes that would form major analytical sections of this research. These focus areas included company culture, communication and decision making systems, executives’ career aspirations, bureaucratic power structures and struggles, and individuals’ beliefs about why the outsourcing project was or was not successful. Next, during axial coding, I discerned the codes into a concept map representing relationships between subcategories of data and the major units of analysis. The concept map formed the general outline of this grounded theory and suggested the relevant theories (discussed in Chapter Two) for analyzing the data. Chapter Five covers the Icarus habitus, digital retail field, the Icarus IT Department organizational structure, and the Icarus IT taxonomy. Chapter Six addresses the factors that lead to the creation of the outsourcing program and provides an overview of the outsourcing project. In Chapter Seven, I explore back stage and front stage decision-making rituals and communication practices. Finally, in Chapter Eight, I address the impacts that both executives’ career aspirations and anomalies to executives’ plans had on the outsourcing project. I used pseudonyms for a handful of principal actors because they were central to the events that influenced SSP and activities where referred to by several interviewees. However, in 47 most instances throughout the data analysis, I do not name or provide pseudonyms for specific individuals. Instead, I only note if the quotation came from an “employee” or an “executive.” The “executive” group includes directors and vice presidents; any employee who is a Senior Manager or below is included in the “employee” group. There are four reasons for using these monikers. First, as already mentioned, I followed McMaster’s (2001) criticism of diffusion-theory and flattened the strata of interviewees into a binary pair of “executives” and “employees.” Second, interviewees participated in the research with explicit commitments to their confidentiality. Third, this research is a grounded theory study. Presenting the events from these different positions within the IT department was my way of harmonizing disparate perspectives into a quasi-singular narrative of how and why SSP unfolded as it did. Finally, the “executive” versus “employee” distinction also became apparent as a layer of the Icarus IT Department taxonomy as I applied Lincoln’s (1989) concept (discussed in Chapter Five). Although I did not directly interview any ComTech contractors for this study, given my role at Icarus and the need to avoid any conflicts of interest, their voices are partially represented in the comments shared by Icarus employees included in this research. Download 1.05 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling