« 이전계속 »
1. Member interview guide. Pre-testing of the member interview guide with a series of open-ended questions,"? did not generate satisfactory
data and proved especially deficient in establishing interviewee understanding of the emergent telecommunication alternatives. The final and ultimately
successful form of the member interview guide was a set of sketches of future
congressional-constituent telecommunications potential, as reproduced earlier
in Figure Two.
This pictorial approach served to place the research within the frame
of reference of the interviewees, without having to waste words and time or use any technological language. The response of congressmen was generally
receptive to this interview approach, and seemed to confirm earlier findings
that the Journalistic or semi-structured approach is most effective for
congressional interviewing.43 Good rapport with members was readily
established, and as a result, through a process of dialogue, the yield of
useful and candid data was relatively high.
Use of pictures was also intended to help minimize problems of validity which may occur when interviewing about subjects unfamiliar to the interviewee, and problems of response bias reflecting in part the respondent's
42 Based on established procedures of interview guide design in, for example, Raymond L. Gorden, Interviewing: Strategy Techniques, and Tactics (Homewood, Ill.: Dorsey, 1969); and Charles F. Cannel and Robert L. Kahn, "Research Methods: Interviewing," in Handbook of Social Psychology, ed. Gardner Lindzey and Elliot Aronson (Reading, Ma.: Addison-Wesley, 1968), pp. 526-595.
43 Ralph K. Huitt and Robert L. Peabody, Congress: Two Decades of Analysis (New York: Harper and Row, 1969), pp. 28-34. See also Lewis Anthony Dexter, Elite and Specialized Interviewing (Evanston, ill.: Northwestern University Press, 1970).
desire to please or perhaps disappoint the researcher.44 The sketches
provided a common focus of discussion, and in many interviews elicited a more
or less "gut" response from the members. Congressmen were encouraged to give
negative as well as positive reactions to the emergent channels, and to
discuss possible problems as well as potential opportunities presented by
of course, as with any research on perceptions of potential future
use, the results can only be considered tentative. When ultimately faced with
the real rather than just projected telecommunication alternatives, actual
use may well differ from perceived use. A pilot demonstration or simulation of these future alternatives might be most desirable, although not feasible here. The sketches were designed to provide the next best approximation.
2. Staff interview guide. The member interview guide was also administered to all senior staff persons in the sample, but only after a detailed discussion with each staff person about the congressman's current constituent communication practices. The guide for this first part of the interview developed through several iterations and pre-tests into what is best described as a worksheet. Basically the worksheet provided an extensive 11st of constituent communication channels, along with various typical types of messages, as identified in the communication systems model-building. Staff persons were not asked to fill out the worksheet but only to use it as a
44See discussion in William F. Mason, Urban Cable Systems (Washington, D.C.: MITRE Corp., May 1972), p. V-12; and testimony by Weston E. Vivian in U.S., Congress, House Committee on Government Operations, Federal Information Systems and Plans, Hearings, before the Subcommittee on Foreign Operations and Government Information, 93rd Congress, 1st Session (Washington, D.C.: Government Printing Office, April 1973), pp. 51-52.
3. Data collection procedure. In selecting the sample of congressional offices (members and senior staff persons) for actual data collection, three factors were considered: definition of the sample, size of the sample,
and representativeness of the sample.45 The population universe included the
435 House offices from the 50 states. In regard to sample size, it
was chosen to be concurrently feasible in terms of available resources and
adequate in terms of the research objectives. The exploratory nature of this
study precluded the necessity for a statistically representative Sample, but
did require that the sample be judgmentally representative on specified key variables.
The initial sample size selected was 10% of the population universe (43 congressional offices), with an expectation that at least half and perhaps as many as three-quarters of the offices in the sample would participate. The actual participation rate exceeded the most optimistic expectation. Forty of the 43 offices agreed to participate, with only two declining due to lack of time and other priorities (involving House leadership activities) and one selected out for personal reasons.
The type of sampling procedure used in this research is most accurately
called a stratified judgmental procedure. 46
The population universe was first
45See Delbert C. Miller, Handbook of Research Design and Social Measurement (New York: David McKay, 1964), p. 46.
46 See Russell L. Ackoff, The Design of Social Research (Chicago: University of Chicago Press, 1953), pp. 124-25; and William A. Spurr and Charles P. Bonini, Statistical Analysis for Business Decisions (Homewood, Ill.: Irwin, 1967), pp. 343-14. For a recent example of this sampling approach, see U.S. Congress, Senate, Committee on Rules and Administration, Automated Legislative Record Keeping for the Senate, a feasibility study by the Subcommittee on Computer Services, committee print, 92nd Congress, 2nd Session (Washington, D.C.: Government Printing Office, February 1972), pp. 257-65.
stratified three ways: by seniority (1-2 terms of service, 3-6 terms of service, or 7+ terms); by major political party (Democratic or Republican);
and by geography (region, air distance from Washington, and population per
square mile of the congressional district).
Following this initial stratification, the population cells were
sampled on a judgment basis so that the total sample would approximate the
population universe as closely as possible in terms of the party, seniority,
and geography variables, while at the same time being Judgmentally representa
tive in terms of additional variables identified by the communication systems
model and technology analysis. These additional variables included: (a) level of political competition in the congressional district (safe or competitive); (D) political orientation of the congressman (11beral, moderate, or conservative); (c) age of the member; (a) socioeconomic nature of the district (degree of urbanization, median income, median education, percent Negro population, percent male white collar, and media market size); (e) key participant status (known Interest and/or leadership and/or committee membership in areas relating to the research focus); and (f) House leadership status.
Careful development of the interview Instruments, coupled with an
effective inside access strategy (using a congressional letter of introduction)
and the inherent relevance of the research subject to the job of the congress
man, resulted in the participation of 77.5% of the congressmen and 97.5% of
the senior staff persons from the final sample of 40 offices.
A "REPRESENTATIVE TIME" APPROACH TO ACCESS ALLOCATION47
Any approach to the allocation of political access to communication
media must be based on a value judgment of some sort, whether implicit or
explicit. In the view of several congressmen and staff in the present study,
and of this researcher, the basic goal during election campaigns should be to
achieve fair and balanced access to communication channels for both incumbent office-holders and challengers. A minimum level of access should be guaranteed
to all eligible and qualified candidates in order to achieve a reasonable
balance of exposure between the incumbent and challengers, and to increase the flow of information to and dialogue with voters. 48
To realize this goal with broadcast television, some have proposed
the "voters' time" concept which would require television stations to make
available specified amounts of time under clearly defined conditions for the
purpose of political broadcasts by candidates for the House or Senate.49
4 This discussion is based largely on Frederick Bruce Wood, "Politics on the Cable: A Cybernetic Approach to Access Allocation," Communication Theory in the cause of Man 3 (forthcoming).
48see the statement of Sig Mickelson, Director, Aspen Institute Project on Politics and the Media, in U.S., Congress, Senate, Committee on Commerce, Subcommittee on Communications, Federal Election Campaign Act of 1973, Hearings, 93rd Congress, 1st Session (Washington, D.C.: GPO, 1973), pp. 104-108.
49voters' time has been proposed for presidential campaigns in The Twentieth Century Fund Commission on Campaign Costs in the Electronic Era, Voters' Time (New York: Twentieth Century Fund, 1969), and in a later Fund study by Minow, Martin, and Mitchell, Presidential Television. Reps. John Anderson and Morris Udall have made a similar proposal for congressional campaigns in U.S., Congress, House, Clean Elections Act of 1973, H.R. 7612, 93rd Congress, 1st Session, Title VI.