In June, 1997 it came to the attention of managers at the Armstrong Laboratory’s Human Resources Directorate that the Commander, Air Mobility Command (AMC) was concerned about assessing the "culture" of his command, with the intent of finding ways to improve it. Put slightly differently, he wanted to "build the rapid global mobility team" by successfully integrating new elements and encouraging his people to work well together with a sense of shared purpose, despite the existence of strong individual subcultures, different weapon systems, and varied missions. He considered this one of his most important priorities and tasked his Commander’s Staff Group with studying this issue. They needed the assistance of experts in organizational development and interview and survey methods to help them map out a plan for the Commander’s approval. The first author, experienced in these areas of R&D, was asked to assist them. With his help, the Commander’s Staff Group developed a plan which the Commander found quite acceptable. As a result, he formally requested our participation.

AMC has been in existence for only about 5 years. It was formed when the US Air Force’s aircraft-oriented major commands (MAJCOMs) were renamed and reorganized. AMC grew out of the old Military Airlift Command (MAC) with its airlift mission but incorporated tankers (refueling aircraft such as the KC-135) from the old Strategic Air Command (SAC). In April 1997 it also regained the C-130 fleet which had been in AMC but was then moved to Air Combat Command (ACC--largely formed from the old SAC) before returning to AMC. In addition, the new AMC included diverse elements from the "old MAC"--"Little MAC" and "Big MAC," the former referring to the KC-10 fleet and the latter referring to the C-141s and the very large C-5s. To add another cultural wrinkle, the command is phasing out the C-141 airlifter and has a growing fleet of new, sophisticated C-17 airlift planes. In addition, the Air Force Reserve and Guard components are key elements of the command. Strong subcultures have developed based on particular airframes, influenced by their old MAJCOM associations. In addition subcultures have formed based on function, employment category or flying versus support mission. Thus, there is not only the airframe cultures or the active duty versus guard/reserve cultures, but the flying versus support cultures, the officer versus enlisted versus civilian cultures, etc. The AMC Commander considers these cultures quite distinct and valuable and he does not want to diminish them. But he hopes to "build the mobility team" into a committed, integrated, more standardized force which interacts better together and accomplishes their diverse missions more efficiently and effectively. This is especially critical since AMC does not just prepare for war by practicing to hone its capabilities--it is continually performing hundreds of operational missions at any given time throughout the world, from transporting the President, to maintaining peace in Bosnia, to providing humanitarian aid in the third world.

This research integrates tried and true technology in the form of interviews and surveys with modern automated technology. This work is still in an early stage, although it is on a fast schedule and initial data gathering should be completed in early 1998. The approach taken first involved conducting individual and small group interviews at five representative AMC locations within the continental United States and then performing qualitative analyses of the interview data. Interview data provided a particularly rich source of information about the AMC culture and its subcultures from the perspectives of senior leaders and management/staff in the field who carry out the day to day missions. These interviews are now mostly complete and the data from them are being analyzed. These data were gathered and analyzed for two purposes: 1) provide preliminary findings and recommendations to the AMC Commander, and 2) serve as a basis for the content of a survey. This survey, which is being developed concurrently with the analysis of the interview data, will serve as a follow up to the interviews to provide a more systematic, quantifiable means of gathering perceptions about AMC culture from more people. Initial survey data will provide baseline measures of the state of the culture. However, the survey will not be used only on a single occasion. Rather, it will be administered periodically to provide longitudinal assessments. Because of its many advantages and strong user acceptance, an automated format will be used rather than a paper-and-pencil survey. More detail about the methods used for both the interview and survey portions of this research is provided below.

Interview Participants. Interview participants were selected by points of contact (POCs) at each of the five bases visited based on guidance provided by Armstrong Laboratory researchers as conveyed to them by our project officer in the Commander’s Staff Group. Members of the Commander’s Staff Group selected the bases where interviews were conducted with the intent of exposing the interviewers to people associated with or supporting a variety of different airframes and who represented the different subcultures within AMC. The bases were located in different geographic regions, and some were considered more desirable assignment locations than others. The bases selected, in the order in which they were visited were, with the number of individual (I) and group (G) interviews, and state location, are as follows: Scott, I = 7, G = 6 (Illinois), Dyess, I = 5, G = 5 (Texas), McConnell, I = 6, G = 4 (Kansas), Charleston, I = 5, G = 5 (South Carolina) , and McGuire, I = 6, G = 10 (New Jersey). Plans were also made to interview personnel at Dover (Delaware) and, as recommended by early interview participants, at a location outside the continental United States. However, these plans were canceled in the interests of time and because the interviewers got quite consistent information across the different locations. It was believed we would get diminishing returns by continuing the interviews.

Individual interviews were conducted with 31 people in senior leadership positions. Group interviews were conducted with an estimated 200 people in small groups (usually of about 7 or 8 persons) composed of a roughly representative sample of people in staff or management positions below the level of senior leadership. Participation was to have been voluntary, although this proved to not always be the case. Although "senior leaders" in individual interviews were all men, both genders participated in the group interviews and diverse ethnic groups were well represented. Care was taken to get a balance of officer, enlisted and civilian personnel and people in a wide range of ranks or grades. Grades or ranks of group participants ranged from airman first class to colonel and from about GS 4 to GM 15. However, to safeguard confidentiality, the interviewers intentionally focused their attention away from group participants’ ranks and names, and did not track such information. They even left participant lists in the hands of their field POCs.

Individual interviews were conducted with leaders at AMC headquarters and at each of the operational field sites. At headquarters, individual interviewees included several general officers ranging in rank from Brigadier General to Lieutenant General and included the AMC Vice Commander, the Commander of the Tanker Airlift Control Center (technically separate from headquarters), the AMC Director of Operations, and other senior officers. The Commander’s Senior Enlisted Advisor, a Chief Master Sergeant, was also interviewed individually at headquarters, as were a few Colonels. At the field locations, individual interviews were conducted with leaders ranging in rank from Major to Major General although only two general officers were interviewed at field sites. Typically, the leadership interviewed individually at operational locations were wing or squadron commanders, directors of operations, or those in charge of support functions.

Survey Participants. Survey participants have not yet been selected, and this will be done by the Commander’s Staff Group based on guidance provided by the researchers. The survey will go to a larger, more systematically selected, world-wide sample throughout AMC. We will recommend that a random sample of at least 3,500 persons be selected from AMC bases throughout the world. Depending on our research objectives and concerns about response rate, we may need to double that number. However, our experience with response rates on automated surveys gives us confidence that the response rate should be higher than with paper-and-pencil surveys. We can not be too optimistic, however, since our interviewees indicated they felt over-surveyed and many may not want to complete even an automated survey. Also, it may be more difficult getting responses back from guard and reserve than from active duty personnel. Sample composition will be partly determined by the nature of the subgroup analyses our client desires and these details have not yet been worked out. If need be, certain subgroups may be oversampled to boost the probable final subgroup size.

Interview Procedures Individual interviews usually lasted between 30 minutes to one hour. Group interviews usually lasted 45 minutes to one hour, but occasionally went longer. Although a period of at least 30 minutes between interviews was requested, they were sometimes scheduled back to back. With the exception of the interviews conducted at Scott AFB, which were conducted only by the first author, the first and second author conducted the individual and group interviews at the other bases. In most instances a representative or two from the Commander’s Staff Group introduced the interviewers, explained the purpose of interview, described their own notion of AMC culture, and discussed the reasons for enlisting the aid of researchers from the Armstrong Laboratory. Basically these reasons were twofold: because they were experts in organizational development and in interviewing, and because they could provide an independent perspective untainted by having spent years (or perhaps a career) in the air mobility business. During the introduction the confidentiality of the interview was stressed. The concept of "not for attribution" (i.e., individual comments will never be attributed back to their individual sources) was discussed. Confidentiality was reaffirmed by the interviewers when they began the interviews. Also in the interests of confidentiality, interviews were not taped. Rather, informal notes were taken but not to the extent that eye contact and rapport was lost. The two person interview team helped in this regard. The desired voluntary nature of participation was stressed by the interviewers. Because some people may have been pressured to attend, or may not have had much information on which to base informed consent, participants were told they had as much right to not speak as to speak and they could remain silent if they wished. Fortunately, even if some participants were reluctant to speak at first, most became quite vocal as they became comfortable with the interviewers.

Prior to conducting individual and group interviews, the first author developed a number of possible questions and asked the Commander’s Staff Group to review them and indicate which ones they preferred. Some of these questions were used for the group interviews at AMC headquarters. However, in the individual interviews with senior leaders at headquarters, very general questions were asked such as how would you describe your own concept of AMC culture, and what do you believe should be the focus of our investigation of AMC culture? This was usually enough to elicit an extended discourse.

Group interviews, especially at the field sites, were of the "focus group" variety and were intentionally very permissive and minimally structured. A list of questions and themes had been refined on the basis of the interviews at AMC headquarters and the Commander’s Staff Group personnel also gave us a set of desired questions. However, the interviewers discovered that it mattered little which questions were asked. Once participants became comfortable with the interviews (i.e. after we established rapport and trust), and once the nebulous concept of culture was brought down to earth, there was an outpouring of comments based largely on what participants themselves wanted to say. It was as if they had their own agenda about what we needed to know. In the permissive environment of a focus group, we largely sat back and discerned the culture from what was said.

During individual interviews with senior leaders, one or two Commander’s Staff Group representatives often sat in. This was with the consent of the interviewers since we did not believe that the senior leadership would be inhibited by their presence. During group interviews they also initially stayed for portions of the interview. The interviewers recommended that they leave the room after the introduction and the staff members agreed to do so. However, they asked to be able to return during the last 5 or 10 minutes and stay afterward to speak with participants who might wish to talk to them directly. This arrangement worked well since it allowed us to establish conditions conducive to a frank discussion early in the interview while giving our client continued involvement and a chance to hear first hand what people were saying.

At Dyess, groups were mixed in terms of grade or rank and category of employee (officer, enlisted, civilian). This may have been intimidating and at other locations groups were more homogeneous. Thus, lower or higher graded/ranked civilians, enlisted personnel and officers were assigned to their own separate interview groups. This approach worked well.

Survey Construction. The notes taken during the interviews provided a typed transcription of over 100 pages and a content analysis of the notes identified approximately 30 themes and a large number of subthemes. After some integration of related themes, seven were considered to be major. The data contained in interview notes and the themes and subthemes will serve as the basis for item development. The target survey length is between 60 and 100 items. However, past experience has shown that a good survey requires the generation of two to three times as many "item ideas"(crude preliminary items) to get the targeted number of good final items. Redundancies will be culled and the most promising item ideas will be refined into good items; the rest will be not be used. The survey will contain at least three parts: a demographic section, a section containing items related to AMC culture, and a comments section. In automated format, respondents will also be able to comment on each item and may branch to different items based on subgroup affiliation (demographic information). In iterative fashion, the draft survey will be shared with the Commander’s Staff Group and revised based on their comments until the best possible draft is developed in paper-and-pencil format. This will be pretested at Dover Air Force Base on a diverse group of at least 60 people and revised one more time on the basis of their feedback. The final survey will be sent to the survey office at Air Force Personnel Center (AFPC) headquarters for approval and for a survey control number with expiration date.

Survey Automation for Initial Administration. Under the guidance of the third and fourth authors, the survey will then be automated. Although automation plans are not finalized, automation will probably be done under contract. Although we have not yet approached them, other Air Force organizations experienced with surveys might also assist us with automation and other aspects of the initial survey. Such organizations include the USAF Occupational Measurement Squadron and the Air Force Personnel Center with whom we have worked well in the past. With regard to possible contracting for survey automation, the Armstrong Laboratory has a contract with Metrica, Inc. to perform research on survey automation and delivery via the internet. If the internet is a suitable vehicle, the AMC culture survey may become an initial testbed under this contract. It may be that automation on computer disk is the best method for survey delivery or that both disk and internet administration should be an option to be compared and contrasted.

Longitudinal Survey Administration. The quality office at headquarters AMC will be given a copy of the initial survey in both paper-and-pencil and automated form. They will have the option of adding a few items and will be responsible for the administration and analysis of subsequent waves of survey administration. The longitudinal survey will be used to assess progress in building the AMC culture. Armstrong Laboratory personnel will serve only as consultants during these later project phases.

Analyses. As mentioned in the discussion of survey construction, interview data analyses involved a content analysis of themes from interview notes. Notes have been transcribed and visually scanned to identify themes and subthemes, allowing development of a taxonomy to guide item development.

Just who will conduct the initial survey analyses is uncertain at this time as we are considering several options. It has been our experience that Air Force clients like data analyses kept simple and easy to interpret. Therefore, initial survey data will be analyzed using simple descriptive statistical methods. Means, standard deviations, frequencies and percents will be generated and displayed both numerically and visually (via histograms) in a manner which is very user friendly. An average pattern of responding across all attitudinal items or within items of similar content will also be calculated to serve as a benchmark against which to compare individual item responses. If desired by the client, or if needed for more academic presentations, inferential analyses will be conducted.

Results Reporting. Results will be reported in simple narrative, numerical and visual form to the AMC Commander and his senior staff in briefings and executive summaries. On 1 Oct 97, the first author and the Chief, AMC Commander’s Staff Group gave a preliminary brief of the status of the project and interview results. This brief was for the AMC Commander and approximately 200 other AMC senior leaders, such as wing commanders. Plans call for the AMC Commander to be briefed on the initial survey results next year. Interview participants indicated a strong desire to receive feedback. AMC headquarters is aware of the need to provide a summary of results to people throughout the command, with a discussion of any actions taken based on such results. Specifically how this will be done is uncertain. However, possibilities include use of command and base magazine and newspaper articles, commander videos, AMC web pages, or via site visits by senior leaders. The AMC quality office will handle results reporting for subsequent survey administrations.

Just as we must protect the confidentiality of the communications of individual respondents, we are also ethically obliged to hold the interview results in confidence for now. At this time the only appropriate recipients of this information are the AMC Commander and his staff. The Commander will decide when and how to share the findings with others.

We can make a few recommendations concerning future research. It would be beneficial to gather data from family members on their perceptions of the AMC culture and its impact on their lives. It would also be wise to explore differences between the culture perceptions of AMC personnel stationed within and outside the continental United States at both urban and more remote locations. This work could also be extended to other commands for cross-command comparisons. Additional research is also needed in the area of automation--how best to develop and administer automated surveys.
________

The authors thank Maj Sue Ross, our project officer, and Col Tom Berry and Maj Scott Chesnut of the AMC Commander’s Staff Group for their contribution to this research.

Next Page 

Back to Symposium Table of Contents

Back to IJOA Home Page