Complete Site Outline

Introduction: Motivations and Backgrounds
Literature Reviews
Research Method
Discussion, Implications
Conclusion & Limitations

Write to me: animated letter


"Our ancestors were eager to understand the world but had not quite stumbled upon the method."
(Carl Sagan in Cosmos, 1980)


The main focus of this study is to examine CFOs choice of communication media under different task situations. What media do they use for different tasks? How, or on what basis, do they make their choice? Why do they choose one media over another?

To address these questions a case study methodology is adopted for this exploratory study. The details of the research design and its limitations are outlined in the following sections.


Given the research questions the case study approach is an appropriate method. Yin (1989) said that a case study is appropriate when "a `how' or `why' question is being asked about a contemporary set of events over which the investigator has little or no control."[1] This fits neatly with the above-mentioned nature of this exploratory study. A case study approach is particularly useful because it permits the collection and presentation of more details and `softer' data.

It should be noted that the case study approach adopted in this study relies primarily on the data collected through interviews with subjects, rather than direct observation of the subjects during the performance of their daily communication tasks.


4.3.1. Industry Sector

The organizations chosen for this study belong to the Over-The-Counter (OTC) financial market industry sector in Australia. These organizations include two major Australian banks and one treasury corporation. Reasons For Focusing on the OTC Markets

The OTC markets include all financial market transactions not listed on an authorised exchange. They include fixed interest, bank bill, foreign exchange and equity derivatives. This market has witnessed exponential growth in the last decade; and technologically, it is regarded as a very sophisticated industry able to handle the speed, diverse ranges of constant financial transactions. Consequently, organizations in this market are expected to have a very high level of information technology maturity such that the acceptance of modern communication media is expected to pass critical mass. This avoids one of the problems that could have contributed towards the conflicting results in prior media choice study. Where the critical mass has not been reached, the context would introduce a bias against modern communication technologies.

Moreover, by focusing on one specific market, the organizational culture and structure are more likely to be similar. The net effect is that the background conditions noted in the framework is partially controlled. This minimises the possibility that the findings are attributable to variation in the background conditions.

4.3.2. Subjects

The unit of analysis of this study is that of senior financial managers, people that perform accounting related tasks as part of their job. Five subjects were involved in this study: two from each bank and one from the treasury corporation. They are selected on the basis of location, accessibility, personal contacts of the supervisor and willingness to help with the research project.

4.3.3. Limitations

By restricting the sample to the OTC market, the generalizability of the results could be doubted. A related limitation is the small and non-random sample size. The non-random sample is, however, consistent with the specific focus and nature of this study. A smaller but focused sample permits a more detailed exploration of the research questions. The exploratory nature of this study meant that the main aim is not to prove some general proposition, rather it is to seek a better understanding of the media choice process.

Although a broader industry scope and a larger sample size would be desirable, the time constraints imposed by the honours programme prevents this. The existence of a large body of literature on media choice also help minimises the difficulties in generalising the results with such a small sample size.


The primary means of data collection was through a structured interview. It was structured in that the subjects worked through a questionnaire during the interview.

The researchers[2] took the questionnaire to the subject's office at a prearranged time.[3] The subjects were asked to complete the questionnaire in front of the researchers. They were encouraged to verbalise the reasons for their particular choice, or any other comments, opinions that they may have. Some rather insightful comments have been gained in the process. These were all recorded on tape for later analysis. After the questionnaire had been completed the researchers follow up with any additional questions necessary to clarify or elucidate the reasons behind the subject's particular choice of communication media.[4] This is particularly important where the subjects chose more than one communication media to accomplish the task. The whole process typically took 45 minutes.

In addition, observations of the general office environment of the subjects were noted. This is done particularly with respect to the technological "atmosphere" or "feel" of the organization. This provides a check on the expected high level of IT maturity of the organizations.

A questionnaire instrument was used to give structure to the interview rather than a series of verbal questions for two main reasons. First, this maintained consistency across all subjects and hence facilitated comparison. Second, the answers to the questionnaire itself provided an important source of data. The need to understand the subject's media choice process, given the myriad of task characteristics and other determinants of media choice, makes it most efficient to gather the require information through a questionnaire instrument, where their relationship can be examined systematically and succinctly.

It is acknowledged that a questionnaire alone is unable to of identify the reasons behind the choice. However, this limitation is mitigated by two additional features of this research design: (i) the presence of a comprehensive framework,[5] and (ii) the interview structure that permits the subjects to state their reasons and opinions during and after the questionnaire. The latter adds substance to the data collected from the questionnaire alone and makes the result more concrete.

4.4.1. The Questionnaire

A copy of the questionnaire is in Appendix 1. The questionnaire was self-constructed. It was based substantially on D'Ambra (1995)'s questionnaire instrument because time limits prevented the development of a new instrument from scratch. This also partially ensures the construct validity of the instrument.

The questionnaire contained a cover sheet explaining the purpose of this study, the instructions, and a sheet with the definitions of communication media under study.[6] Definitions were provided to minimise deviation in the subjects' understanding of the terms, particularly in relation to the modern communication media. The subjects tended to refer to the definitions when answering the questionnaire. This enhances the comparability between the questionnaires.

The second part of the questionnaire contained the questions. The questions are divided into four parts. The purpose of each part is:

  • Part 1: General questions to elicit availability, usage pattern and attitude towards technology or communication media. These questions and the demographic questions in part 4 indicate how the subjects stoutline sources outlined in the framework.
  • Part 2: Determination of subject's media choice under different tasks and different variations of the tasks. The communication media and the variations on the communication task were presented in a matrix form to simplify the presentation and to speed up the answering process. This format also permitted the subjects to choose more than one communication media for each task.
  • Part 3: Determination of the equivocality of the task using the instrument developed by Withey, Daft and Cooper (1983).[7]
  • Part 4: Demographics information.

4.4.2. About the Tasks

The five communication tasks were selected on the following basis:

  1. to achieve a maximum spread of equivocality;
  2. to be representative of subjects' tasks; and
  3. to ensure that all communication media has a fair chance of being selected.

Four of the communication tasks were drawn from D'Ambra (1995).[8] The wordings for two of the tasks were modified slightly to align them with the accounting nature of subjects' tasks. The other communication task (ie. Question 14) was created to introduce a group communication task dimension that was not examined in the D'Ambra study. Because this task is not pre-tested, nor is it derived through any formal approach, the validity of this question can be questioned. This said, observation during the interview indicates that questions 14(a) and (b) seem to get the desired message across; but part (c) presented certain problems related to understanding the task presented. Anyhow, question 14 did trigger in many instances some rather detailed and insightful comments. Variations Within Each Task

The framework in Chapter 3 pointed out that there are many task characteristics that could influence media choice. To test the impact of these other task characteristics on media choice, this study introduced six additional variations for each of the five tasks studied. These variations are:

  • (a) the other person(s) is/ are located on the same floor;
  • (b) the other person(s) is/ are located on the same city (but not the same building);

  • (c) the other person(s) is / are located in a different country;
  • (d)the other person(s) is/ are not in their office;
  • (e) it is not during the other person(s) working hour; and
  • (f) a time constraint aspect for each of the above variations (ie. increasing the urgency of the task).

These six variations are meant to cover the task characteristics noted in the framework, besides the task characteristic of equivocality, that have not been examined in prior studies.

4.4.3. Limitations

In addition to the limitation noted above, the validity of this research design relies upon the subjects' answers only. No attempt was made to observe directly their media choice when performing their actual day-to-day tasks. The data collected was essentially what communication media the subjects think they would choose when faced with the different task situations presented. There is, however, no guarantee that their actual choice is the same. This said, it is reasonable to say that the deviation (if any) from actual practice is unlikely to be significant.

A related limitation is that the tasks presented many not be representative of the subject's range of tasks. The lack of time prevented the use of focus groups or preliminary interviews with managers to construct more realistic tasks. The difficulty in generating appropriate tasks for the subject is further complicated by the lack of literature on the role of these subjects. To the extent that the tasks are too artificial or did not cover the subject's broad range of tasks, the ability of this research design to support the framework presented would be weakened.


The questionnaire response and the taped comments were analysed in relation to the framework in Chapter 3. First, the media chosen by the subjects for the variations of each communication task are compared. Second, comparisons were made between the different communication tasks. Given the small sample size no statistical test was performed. Rather, the result was documented in the next chapter in the format of a case study

Some may argue that the lack of a statistical test would deprive the study of objectivity, but it is submitted that the level of details provided by the subjects' comments more than compensate for any deficiency in this regard. Any potential bias is further mitigated by the systematic documentation and presentation of the data collected. Moreover, the lack of scientific (statistical) generalisation is really irrelevant in that under a case study approach it is analytical generalizability that is crucial. The design of the study and the framework provides support for this theoretical generalisation in this study.


Consistent with the exploratory nature of this study and the `how' and `why' research questions posed a case study research methodology was adopted. This methodology enabled the collection and presentation of more substantial details in relation to the subject's media choice. As with all research studies there are limitations, but the use of the interview questionnaire approach to collect the data and the systematic, detailed presentation of the analysis would nonetheless enrich our understanding of media choice.


next chapter


[1] Yin (1989), p20.

[2] Reference to researchers in this section refer to my supervisor and myself.

[3] An exception is that for one interview a copy of the questionnaire instrument was faxed through to the person beforehand (as requested by the person).

[4] Often, the comments made along the way were more than adequate.

[5] See Chapter 3.

[6] Video conference was not included in the instrument because it was felt that it is unlikely to be used by the organizations studied. However, unexpectedly, two subjects pointed out the absence of video conference among the media choice.

[7] This is the same instrument used in D'Ambra (1995). Whilst it is acknowledged that this instrument is imperfect, the exploratory nature of this research preclude the design and testing of a different instrument. Anyhow, equivocality is not the only nor the main concern of this study.

[8] In D'Ambra (1995) the tasks are generated through focus groups and content analysis. This ensure the ecological validity of the task. In that study, a large Australian insurance company was the source of the data.


next chapter



[my logo]

Copyright © 1997 Raymond Yu.