8. More complete answers were received for this part of the question on the second wave of interviews, so that they could be coded in more detail. On this second wave, 47 percent (P) and 45 percent (S) specifically mentioned Greece's importance as a military base. 9. While responses on this point may have been affected by the prevailing political atmosphere, the pro-Communists would be more apt to give no answer than to give a false one. Even if the percentage not answering these questions is considered to be (in part, at least) hostile to the United States, the over-all picture is still extremely favorable.

10. To simplify the presentation, the statistics for the panel are not reported here. On some points the percentages are somewhat different for the two samples. In part this is due to sampling variations. In part it stems from a "panel effect" often noted in studies of this kind, which comes about because respondents have been asked the same or similar questions previously and are sensitized to the subject matter and the interview situation. The data for the panel group are examined in detail later.

11. As opposed to those who said that the U.S wants Greece to remain free and independent. 12. The rise in Patras may be due to independent mentions of the Declaration in the Athens and Patras papers, or simply to a difference in interviewing technique between the two waves of the study. Knowledge of the Declaration, rather than the claim to knowledge, is the major consideration here.

13. Although the jump from two percent to ten percent in Patras is also statistically significant, it simply reflects the increase in the number who claim to have heard of the Declaration, whereas in Salonica there was a great rise in the proportion as well as the number who knew what it was.

14. The actual, rather than the sex-weighted totals are used here, since the booklet sample is so predominantly male.

15. Anti-American respondents who feared to express their views openly might be expected to give the weaker of the two responese.

RADIO LIBERTY'S AUDIENCE IMPACT AND

EFFECTIVENESS*

BY JOSEPH G. WHELAN

This discussion of Radio Liberty's attempts to obtain some feedback from its audience indicates the political and societal limitations on audience analysis in closed societies.

* *

*

I. POSITIVE MEANS FOR EVALUATING RADIO LIBERTY'S EFFECTIVENESS

A. Interviews with Soviet Citizens

1. Some Sources for Judging Audience Response

Despite known disadvantages, which by the nature of things cannot be easily corrected, Radio Liberty (RL) attempts, nonetheless, to establish some positive basis for judging audience impact, however imprecise it may seem compared with the reasonably accurate techniques in American public opinion research. It does this by collecting evidence on audience reaction, (1) in interviews with Soviet visitors to the West, legal Soviet expatriates, Soviet defectors, and with some Soviet listeners actually living in the Soviet Union itself; and (2) from letters received from Soviet listeners through the indirect method of a mail drop in the

*Excerpts from Chapter VI, "Radio Liberty-A Study of Its Origins, Structure, Policy, Programming and Effectiveness," The Library of Congress, Congressional Research Service, Washington, D.C., February 29, 1972.

West. The substitute audience panel and the special program auditions panel provide another imput, though limited, into assessing possible audience response.

Through these efforts [the Audience Research Division of Radio Liberty] (ARD) attempts to fill the gap between the speaker and the audience, to determine an accurate image of the listener and the listener's image of RL (both essentials for success in communications), and to build the foundations for a continuing dialogue with the audience. The Director of ARD files a quarterly report containing an analysis of and references to listener mail, reactions to RL emanating from the media of the Soviet Union and other Communist countries, conversations with Soviet listeners at home and abroad, and related background material on attitudes and conditions in the Soviet Union.

2. Difficulties in Interviewing

Interviews, or "conversations" as RL prefers to term them, presumably since they lack the formal structure and statistically-workable ingredients of professional poll-taking known in the United States, are regarded as a prime indicator of audience response. During FY 1971 interviews were conducted with hundreds of Soviet citizens, of whom well over one-half were foreign radio listeners. But, interviews are very difficult to conduct. Polling by Soviet Government agencies has only begun in recent years in the Soviet Union, and the people do not regard them as scholarly efforts but rather as instruments of the KGB. The problem is especially acute in the Soviet Union where the climate of suspicion impedes a genuine exchange of views, especially with a foreigner. Even interviews conducted with Soviet visitors in the West are carried out with difficulty. It is estimated that only one out of eight contacts will yield a fruitful interview.2

The results of interviews are sent to RL headquarters in Munich by field correspondents where they are compiled in a final Target Area Listeners Report that is prepared by ARD and distributed throughout the organization. These reports describe the interviewee, his occupation, nationality, age, and language facility. They include such details as place, listening times, language of broadcast, audibility, jamming effectiveness, and specific programs of interest. In addition, the conversation is summarized, and this may include statements about public reactions, comments on recent events and expectations about future developments. 3. Analysis and Use of Interview Data

RL approaches analysis of audience research data conservatively. ARD does not claim to have enough data at its disposal from interviews to speak about a "sample"-that is, as interpreted in a statistical sense. It does not go beyond the claim of having only "bits and pieces of samples that could be indicative of some certain groups." Yet, Mr. [Howland] Sargeant [President of the RL Committee] has written that such interviews have been collected and properly coded, transferred to cards and

zia and Armenia in the South. [sic] Most listeners whose nationality ascertainable were Russians, followed by Ukrainians, Belorussians, elians, Georgians, Kazakhs, Latvians, Estonians, Lithuanians, ars, Bashkirs, and others from Eastern Europe. 10

ther categories within the report either directly or indirectly provide he indication of RL's effectiveness, but perhaps none so much as adio Liberty's Impact and Image in the Eyes of Its Listeners." Under is category numerous comments were made about reasons for listening RL and what its impact was judged to be.

For a scientist, RL had become a "most important source of informaion and a link between our more free-thinking intelligentsia and the Soviet working people." A musician considered that "normalization" of Soviet life could only come through the creation of a public opinion which would exert pressures for change. Since there was no freedom of information at home, the initiative had to come from abroad-and RL, he said, was the only possible source. "Thanks to Radio Liberty's broadcasts," said a Moscow engineer, "I have learned to think and develop a 'free relationship' with the outside world." This theme was reiterated by a Soviet musician who said that "free voices from abroad, especially the voice of Radio Liberty, are the only forces which can wake people up and open their eyes."

Listeners were aware of regime attacks on RL and expressed concern for the radio's future. According to a Moscow engineer, rumors were being circulated to the effect that the station would soon cease broadcasting.

RL was of special interest ot a Leningrad scientist since, in his words, "It is what a Russian radio station would have been like if we had had freedom of speech in our country." A traveler from Sverdlovsk considered RL to be "a free and international" station, not government operated like the BBC or VOA, and this enabled it to devote so much time to Soviet problems. He appreciated this. A Kiev intellectual who previously thought RL was an "American propaganda station" had changed his opinion when he became more familiar with its broadcasts. He now felt that, although financially supported by the United States, it was working for the good of Russia. Whenever an official announcement was made in Soviet media, said one engineer, it was very common to hear people remark: "That's what 'ours' said. But we still have to know what 'they' will say." "They" was RL. 11

Scattered evidence of impact and effectiveness like this and other evidence included in the quarterly report cannot constitute a "sample" according to the statistical requirements of public opinion research. RL is the first to acknowledge this limitation. Yet, it would seem to be a commonsense judgment that comments by opinion leaders such as scientists, engineers and others within the intelligentsia could be reflective of important preferences and predilections of others within the Soviet int ligentsia and perhaps even in broader segments of Soviet society.

can be processed by machines to show "an increasingly accurate picture of the listeners, and of potential listeners." "

Accordingly, ARD is able to give RL programmers some perception of their audience's image and scattered evidence of impact and effectiveness. In recent years this perception and measure of effectiveness have no doubt become more reliable because the percentage of ARD interviews has been progressively increasing from 17 percent in 1969, to 23 percent in 1970 and up to 41 percent during the first quarter of 1971.5 The ages of RL's listeners were estimated in mid-1971 to be 35 percent in their 20s; 31 percent in their 30s; and 16 percent in their 40s. The preferred listening time was cited at 2001-2400 for 57 percent of listeners, with the next highest percentage preference at 16 percent from midnight to 0400.7 The distribution of audience occupation in 1970 was concentrated heavily among the intellectuals, including university students: 72 percent of the listeners were said to be from the intellectual professions.8

Clearly, audience research data of this nature confirms RL's judgment on audience structure, policy content, and program design. This is especially true of ARD studies on audience reaction to samizdat [the private publication and circulation of one's own works in the Soviet Union], presently the main staple of RL's programming. Again, RL's judgment was reaffirmed. During the first quarter of 1971, 87 percent of the interviewees mentioned samizdat favorably, 3 percent with mixed or neutral reaction, 10 percent hostile reaction. During the fourth quarter of 1970, 100 percent of the interviewees mentioned samizdat in a friendly manner; for the third quarter of 1970, there was 90 percent with 10 percent mixed or neutral. The total overall percentage for this mini-month timeframe from July 1, 1970 to March 31, 1971 was 85 percent friendly, 4 percent mixed or neutral, and 11 percent hostile. Moreover, RL's focus on the intelligentsia was reaffirmed. The majority of listeners mentioning samizdat (70 percent) belonged to various segments of the intelligentsia." Undoubtedly the most comprehensive assessment of empirical evidence of RL's effectiveness drawn from interviews is contained in ARD's quarterly reports. The report for the second quarter of 1971 generalized on the data collected in an effort to determine listener profile, effects of jamming, specific program interests, programming suggestions, RL's impact and image in the eyes of its listeners, attitudes on specific problems such as samizdat and Jewish emigration, criticism of RL, and responses from the nationality areas.

On the basis of empirical evidence the listener profile for the second quarter of 1971 looks like this: RL respondents came from all walks of life, although the number of those holding positions in the technological, scientific, and cultural fields outweighed those in other professions. Many listeners were under 40 and an "overwhelming majority" supported RL and its aims. Listening took place mainly in large industrial cities such as Moscow, Leningrad, and Kiev and their surroundings, though a sizeable number in Siberia, Uzbekistan and Kazakhstan in Central Asia, to