University of Virginia Library

Ch 8: Is Anybody Listening? The Great Audience-size Debate

Starting in the predawn hours of each Sunday morning, the largest religious gathering in America takes place, drawing almost 130 million people to their radio and television sets. What happens is both exciting and miraculous. It involves a new approach to the problem as old as the Bible: how to introduce struggling, helpless individuals to a loving God who wants them to meet Him and be born again.
This amazing event takes place every week, all week, from early Sunday morning through the final midnight stroke on Saturday night. Making this possible is the awesome technology of broadcasting, which many consider to be one of the major miracles of modern times; and making it meaningful is the overwhelming love of a God who cares passionately about each one of the world's four billion people. I believe that God has raised up this powerful technology expressly to reach every man, woman, boy, and girl on earth with the even more powerful message of the gospel.

--Ben Armstrong, Executive Director,
National Religious Broadcasters

The ageless riddle of whether a tree falling in the forest makes a crashing sound if no one is present to hear it has an analogue in the world of broadcasting. Has a broadcaster communicated anything if there is no one out there picking up his transmitted signals?

The sociopolitical meaning of the televangelists pivots on the ques-

---

143

tion "Is anybody listening?" Obviously, people are listening. But how many? A great deal hinges on the answer.

A small but steadily growing audience might suggest that the potential social and political influence of the televangelists is on the ascent. And, similarly, a declining audience would suggest that the potential for the video vicars to shape American politics is waning.

On the surface, nothing would seem to be simpler. Two major organizations, the A. C. Nielsen Company and the Arbitron Company, monitor America's radio and television listening and viewing habits. Theoretically, Arbitron and Nielsen have the technological capability to measure all radio and television audiences in the nation on a continuous basis. But for the large majority of programs, they don't do so; the cost does not warrant the value of the information.

Three times a year-February, May, and November-Nielsen and Arbitron study the entire country in what is called the "sweeps," assembling great quantities of data about audiences to serve the needs of stations, networks, producers, advertisers, and other interested parties. This information is published in a separate volume for each of the 200 "base" markets in the country. In addition, both Nielsen anti Arhitron publish summary volumes for all syndicated programs. In recent years, both companies have commenced publishing separate volumes on syndicated religious programs.

So why not just ask Arbitron or Nielsen how many people are watching religious television? Or, better yet, why not borrow their studies from a local library and study the data?

The seemingly simple question of the audience size of religious programs is actually complex-to the point of defying an answer. It is, nevertheless, a question that must be addressed in order to assess the potential of religious television as an instrument for mobilizing conservative Christians to social action.

Television audiences have to be estimated based on sampling, a science that is reasonably sophisticated. The difficulties in measurement arise from variations in human behavior that no sampling technique can monitor adequately.

First, since people may be watching one moment and doing somethillg else the next reliable measurement requires more-or-less-continuous sampling, which can be done with meters hooked up to a carefully selectecl sample of television sets. The meters automatically

---

144

read whether a set is on or off, as well as the channel to which it is tuned. But the meter cannot tell you who is watching, if anyone.

To measure who is watching, Arbitron and Nielsen ask a sample of Americans to complete a diary (log) of their viewing habits. While diaries can provide an approximation of the demographic characteristics of viewers, this method is notoriously flawed. Most people do not keep an ongoing log, and when they do sit down to record their viewing from the previous hours or days, they tend toward systematic inclusion or exclusion of certain types of programs. Further, cable television has significantly increased the number of available channels, and this, in turn, has increased the possibilities for error.

The methods used to measure audiences has been called "a mix of science and voodoo." It is, nevertheless, a "quirky but agreed-upon ratings system" that has been altered very little in commercial television over the last thirty-seven years. [1]

In the fall of 1987, Nielsen introduced an innovation in audience measurement that created a storm in the industry. Called a "people meter," the new method combines a meter and a diary. It monitors the set, as with the old meter system, while providing a hand- held device individual household members can use to electronically log in or log out when they watch television.

In theory, the new method should produce more accurate and reliable data than the old tried-and-true methods. But William Rubens, vice president for research at NBC, thinks the people meters will produce "chaos." We may never know for certain whether the people meters produce better or merely different statistics.

What has upset network officials is that preliminary tests of the new technique produced smaller audience numbers. CBS's "Dallas," for example, had 11 percent fewer viewers with the people-meter method than with the old system. And NBC's "Cosby Show" lost two rating points, which translates roughly into 1.7 million households.

Lower ratings made a tempest inevitable because ratings determine how much broadcasters can charge for advertising. Network advertising rates range from $80,000 to $400,000-more for such extravaganzas as the Super Bowl-for thirty seconds. Take away advertising and you take away the major reason for audience ratings.

Gathering audience data is very expensive. While very limited sum-

---

145

mary data are published weekly in the entertainment sections of major newspapers' the details of the data-gathering process are carefully guarded industry secrets. Arbitron and Nielsen sell the results of their studies to the networks, to advertisers, and to syndicated programmers for a very handsome price. Their data books are not found in libraries because that would make the results public information.

Prior to 1980, there was little interest in the audience ratings for syndicated religious programs. Since there is no commercial advertising during these broadcasts, the information on audience size seemed to have limited intrinsic worth. Only a few of the producers of religious programs subscribed to either the Arbitron or the Nielsen service. Many were even unaware that a service existed.

The general public's lack of interest in the audience size for religious broadcasters changed sharply during the summer of 1980, when candidate Ronald Reagan made appearances at the National Affairs Briefing in Dallas and the eastern regional meeting of the National Religious Broadcasters in Lynchburg. The latter took place on the campus of Jerrv Falwell's Liberty University. Reagan's appearances greatly spurred the media's discovery of the potentially large voting bloc religious broadcasters were attempting to mobilize. It was a matter of straightforward political arithmetic: the larger their audiences, the greater the potential political influence of televangelists.

At the time this "need" to know how many people were listening arose, it would not have been easy to predict the storm of controversy that would later ensue. The great audience-size debate continues, although the broad parameters are better understood today than they were in 1980.

The first round in the audience-size controversy was a direct outgrowth of the recognition that TV preachers might exercise political clout from their airwave pulpits. In the beginning, many prominent print and broadcast media, including the New York Times, cited Ben Armstrong's estimate of a weekly audience of 130 million for religious broadcasting without raising an eyebrow. [2]

Armstrong, executive director of the National Religious Broadcasters, based this figure on an extrapolation of a study conducted by a sociologist. It was an unrealistic figure, and Armstrong knew that. But it was a powerful attention-getter for the opening paragraph of

---

146

his book, The Electric Church. Later in the book he cited as more reasonable an A. C. Nielsen estimate of 14 million weekly television viewers. [3]

Armstrong, who initially expressed serious reservations about the movement of religious broadcasters into the political arena, inadvertently contributed to that process with his audience- size hyperbole. So did Jerry Falwell, the most visible of the televangelists who were speaking out from their video pulpits on political and social issues. His public-relations promotion packet claimed an audience of 17 million for the "Old Time Gospel Hour," and on numerous occasions he claimed even larger audiences.

Then, in July 1980, at the Republican National Convention in Detroit, Falwell told at least two reporters that the "Old Time Gospel Hour" really had an audience of 50 million viewers! It was as if R2D2, who was starring in The Empire Strikes Back that summer, had fixed the hyperdrive on his Old Time Gospel Ship so it could travel at astronomical speed to spread the Good News.

Armstrong and Falwell, of course, were not the only ones with a tendency to get carried away when talking about their audiences. There were many others. One televangelist (who realistically could barely claim an audience of six figures) frequently boasted that while he didn't know the size of his audience, he was sure that it measured "in the millions." It may be an occupational carryover from the days of urban revivalism, when preachers were known to exaggerate the number of souls they saved. Jerry Falwell is now more cautious about throwing around grossly exaggerated audience figures. When he does use unsubstantiated figures, he often quips, "ministerially speaking," as if to say, "Preachers don't lie, but sometimes they remember big."

After a while, reporters became skeptical of the audience claims made by television ministries, but the doubters seemed content to deal with their misgivings by simply writing, "So and so claims X million viewers," rather than digging around to see if any bonafide statistics were available.

In the months following the 1980 election, there were literally scores of articles and broadcast reports attempting to assess the role of the New Christian Right in the outcome. Audience size of the television ministries was intensely debated. Meanwhile, no one seemed aware

---

147

of the fact that Arbitron and A. C. Nielsen had been routinely collecting "hard data" on audience size for many years.

The second round of the audience-size controversy commenced in mid-1981, when Arbitron and Nielsen data were made public for the first time. In June, Jeffrey K. Hadden and Charles Swann presented Arbitron data in Prime Time Preachers [4] while sociologist William Martin reported Nielsen data in an article entitled "The Birth of a Media Myth" in The Atlantic. [5]

The two data sets were not entirely comparable, but the rankings and audience-size estimates agreed remarkably. Similar conclusions coming from independent sources lent credibility to the craftsmanship of both organizations.

The most important conclusion to be drawn from these two independent surveys was that the audience size for the syndicated religious broadcasters was much smaller than the claims that had been made by many of the ministries. Martin reported an audience of 13.8 million for the top ten programs in the Nielsen ratings for November 1980. [6] Hadden and Swann reported a total of 20.5 million for Arbitron's sixtysix syndicated programs, with a total of 14.9 million for the top ten programs-a discrepancy of a mere 8 percent for data gathered nine months apart. [7]

The infusion of hard (or at least harder) facts into the controversy did not settle the debate about audience size and trends. If anything, it added fuel to the flames. The reason: Lots of people have vested interests in how many people are watching and how the numbers are interpreted.

Enter the political factor.

It should not have come as a surprise that religious broadcasters themselves were not happy with these numbers-particularly those who had publicly claimed vastly larger audiences than the Arbitron and Nielsen analyses showed.

Ironically, some of the electric church's strongest adversaries also challenged these smaller audience estimates. Their reason for doing so was understandable. If the number of persons watching Jerry Falwell and other politically minded televangelists was much smaller than they claimed, then the rationale for the liberal People for the American Way organization, for example, was diminished. In short, the greater

---

148

the audience size of the right-wing televangelists, the greater the threat and' hence, the greater the urgency to support countermovement organizations such as "People For," which had been formed to do battle with the New Christian Hight.

Meanwhile, another group of adversaries argued that the Arbitron and Nielsen figures ocerestimated the "real" audience size of the televangelists. [8] These figures, they argued, do not take into account the duplication of audience when the same person watches more than one religious program. The point is well taken, but the data regularly reported by Arbitron and Nielsen provide no basis for determining what proportion of the measured audience involves duplication. Thus, any attempt to adjust the Arbitron or Nielsen data downward to account for duplicated audience involves unverifiable assumptions. In short, it's back in the guessing game.

One of us, having been a party to the controversy, felt no particular need to defend either the Arbitron or the Nielsen figures if a more satisfactory method of measuring audience size came along. So he listened to the arguments of broadcasters who claimed larger audiences and the critics who claimed audiences were smaller.

Although both sides raised interesting and possibly legitimate arguments in behalf of their case, there was no compelling reason to conclude that the factors that would produce a larger audience number were more significant than the overestimation that resulted from duplication. Hence, in the absence of clearly superior data, a reasonable conclusion seemed to be that the Nielsen and Arbitron figures were not too far off the mark. Furthermore, they had the merit of being standardized measures that could reveal trends over time. Comparing data from the two companies would call attention to any great discrepancies. In a word, the Arbitron and Nielsen data seemed much preferable to nonempirical "guesstimates."

Debate about how many people are watching religious programming was one factor leading to the creation of an ad hoc committee to investigate the effects of religious television, which grew out of the Consultation on the Electronic Church sponsored by the National Council of Churches in February 1980. From the beginning, the project was conceived as bipartisan, and the committee consisted of a coalition of evangelical religious broadcasters and the mainline church traditions. It was co- chaired by William Fore, director of the National

---

149

Council of Churches Communication Commission, and Ben Armstrong. [9]

The ad hoc committee raised more than $150,000 and commissioned the Annenberg School of Communications and the Gallup Organization to conduct the research. Even before research had commenced, the Religion and Television Research Project was heralded as "the definitive" investigation of religious broadcasting.

Throughout the study, the two investigative teams worked independently. George Gerbner, dean of the Annenberg School, was responsible for coordinating and integrating a final report. In the spring of 1984, the long-awaited report appeared. [10]

As it turned out, there was little integration of the two investigations. De facto, they became two studies, with the Gallup study appearing as an appendix to the Annenberg study. [11] The findings of the two studies were at odds on several critical points-including the important question of how many people are watching.

The two research teams used different methods to measure audience size. and, with hindsight, it is not particularly surprising that they came up with radically different conclusions. Piggybacking on Arbitron "sweeps" data, the Annenberg team came up with a figure of 24.7 million viewers. Then turning to viewing diaries kept by a sample of Arbitron respondents, they concluded that the nonduplicated audience was 13.3 million. In contrast, the Gallup survey asked people if they had watched a religious television program during the previous month. Thirty-two percent, which would represent approximately 70 million viewers, claimed that they had.

The Annenberg measure was for one week, the Gallup measure for one month. Certainly, if there was duplication during a week, the amount of duplication would increase over the course of a month. Thus, the discrepancy of more than 500 percent-13.3 million versus approximately 70 million-was even greater than it appears.

Regrettably, the Annenberg team took an imperialistic view, assuming the superiority of their method and ignoring the Gallup data in the Executive Summary. So also did the news release, which was prepared by the National Council of Churches.

Furthermore, the text of the study was treated as a quasi-secret document. Only the Executive Summary was made available to the press. Naturally, the audience figures were disputed by the religious

---

150

broadcasters. So the great audience-size debate was back to ground zero.

But not quite. There remained the puzzle of why some adversaries of the electric church, who had earlier accepted the myth of huge audiences, abruptly changed their minds and became adamant in pressing the case for even smaller audiences than those measured by Arbitron and Nielsen.

Most vociferous in this group of liberal churchmen was William Fore, the chief spokesperson of the National Council of Churches. In a July 1980 TV Guide article, Fore wrote, "Some 47 percent of Americans see at least one religious program a week on TV." [12] Later, when the Annenberg team reported the weekly "unduplicated" viewing audience for religious programs at 13.3 million, Fore whittled the number down to a mere 7.2 million. Precisely how he did this calculation was not clear. [13]

Why the great reversal? The most plausible explanation is that the mainline Protestant communities now see their chances of tapping into free airtime if they can "prove" that the current syndicated broadcasters are not reaching nearly as many people as they claim. Given the confusing nature of the data, and the absurdity of the claims made by some, whv not claim that the electric pews are empty-or nearly so? If hardly anyone is watching the current offerings, so this logic goes, then stations and networks should give the liberal churches free time so they can produce programs that are more in keeping with the tastes of "mainline Protestants" and thus will attract much larger audiences.

Pursuing this strategy, the liberal church leaders have created another myth, which is just as misleading as the myth of exaggerated audience size debunked in 1981.

This new fabrication might be dubbed "the myth of the tiny and insignificant remnant." It seeks to create a self-fulfilling prophecy. If nobody believes that anybody is watching, then perhaps nobody else will tune in. And if nobody is watching, how can stations justify selling airtime to the televangelist charlatans, who are, supposedly, getting that money by ripping off poor little ole ladies?

If this is indeed a conscious strategy for returning mainliners to a significant place in religious broadcasting, it is doomed from the beginning, less a strategy than a futile, head-in-the-sand response. The

---

151

liberal church tradition in America has lost every major battle for access to the airwaves in the past three decades. Its leaders are bankrupt of ideas for getting back into the competition. Furthermore, they have duped their liberal church constituencies into half-believing that the airwaves are inherently evil and that they shouldn't be there. Unless, of course, they get a chance to do so-in which case they will do better than evangelicals and commercial broadcasters. In this way the mainline leadership exhibits a schizophrenia that staggers the imagination, reminding one of Humpty Dumpty's proclamation that words mean just exactly what he chooses them to mean.

Part-though certainly not all-of the debate about the audiences of religious television results from using different measures. Comparing the Gallup and the Annenberg statistics is rather like comparing apples and oranges. The figures don't lend themselves to comparison. It is impossible to obtain a comparable figure by multiplying the Annenberg figure by four (the number of weeks in a month) because part of the audience is duplicated-i.e., individuals who watched more than once. But neither does dividing the Gallup figure by four produce a comparable figure. Whereas the Gallup figure is a cumulative one (the total number of people who watched), the Annenberg figure represents an average quarter-hour.

How do we get out of this quagmire? We begin by recognizing that measuring a television audience is not as simple and direct as counting how many people go through a turnstile at a football game or attend church services in a local congregation, where people are either present or not present. Most people do not watch television the way they attend a church service or an athletic contest. They tune in late and depart early; they switch channels and then leave the room. In front of a television set, people generally feel at greater liberty to talk with one another than they do in church. They may also do lots of other things-take a phone call, prepare breakfast, eat, do dishes, get ready to go out, and so on.

No one would contend that the only legitimate audience count for the electric church covers those who arrive on time, listen attentively in a fixed location, and stay until the program is over. Even if someone did want to apply the standard of attentiveness, it simply is not possible. Thus, all measures are inexact and arbitrary.

The question is, what counts? Abstractly, the answer is that there

---

152

are many ways to count, and no single way of counting is necessarily better than another. All ways involve some degree of arbitrariness and some degree of error. Furthermore, what constitutes a reasonable or fair way of counting an audience for one purpose may not be a fair measure for some other purpose.

On the practical level, it is helpful to understand how the professionals count. Pollsters, such as the Gallup Organization, ask people to recall whether they watched television. The questioner may specify a time period-such as last night, last week, or last month. And the interviewer may further inquire as to whether the viewer watched a specific program, some category of programming, or just any program. Nielsen and Arbitron use combinations of diaries and meters. And both Nielsen and Arbitron supplement their national surveys with individual market surveys. As with the case of nationwide estimates, local market surveys are used to set advertising rates. Larger markets have larger surveys because there are higher advertising revenues at stake. Smaller samples are drawn in smaller communities, producing a greater error ratio in the findings.

The most frequently cited Nielsen and Arbitron statistics are quarterhour averages, a statistic they use for at least two very good reasons. First, the quarter-hour unit is the basic building block and common denominator for accumulating all statistics. Second, the quarter-hour average has commonsense utility for marketing airtime. For the potential advertiser it answers the question, "If I buy advertising time on this program, what size audience am I going to be reaching?"

Many different types of information are gathered beyond quarterhour averages. Household data are the most easily accessible and the most reliable. But there are also important reasons for obtaining detailed demographic information about individuals, such as sex and age characteristics. Advertising is geared to demographics. To cite an obvious example, it makes no sense to advertise a product for children in the late evening when most children have already gone to bed.

There are also important reasons for comparing one program with another. One way of doing this is by examining audience share-what proportion of the television sets that are turned on are tuned to a particular program? Audience rating, on the other hand, indicates what proportion of all the households in the survey are tuned to a particular program.

---

153

For all of these statistics, the industry builds upon the average quarter-hour data. But there is no intrinsic reason why other statistics should not be used. It all boils down to what a statistic is used for. Statistics may be valuable in making economic decisions (e.g., setting advertising rates), or in creating public- relations images, or in establishing that a certain program has the potential to communicate to a particular audience.

The average quarter-hour rating may not be a particularly useful figure for assessing the audience of an event lasting several hours. A professional football game, for example, will have fluctuations in its audience size depending on the ebb and flow of the contest. If the score of the game is close, the audience may build toward the end, whereas a lopsided score may result in a thinning-out of the audience.

Another reason the average quarter-hour audience may not be useful is the simple fact that programs are not all the same length. "Oral Roberts and You," for example, is a thirty-minute program, while "The 700 Club" airs ninety minutes a day, five days a week, with reruns on Saturday and Sunday in some markets. In terms of average quarter-hour audiences, Oral Roberts has a much larger audience than "The 700 Club," but his cumulative audience is much smaller, because "The 700 Club" has fifteen times more airtime per week. Comparing the simple quarter- hour averages, then, distorts the market significance of daily broadcasters.

And this brings us back to the politicized nature of statistics on the religious television audience. In 1980 it was the perceived potential of mobilizing viewers for political purposes that made us all conscious of audience size. Now audience size has taken on a political significance of its own. As noted earlier, religious programming is governed by free-enterprise market principles even though advertising does not figure in the scheme of their programming. As more programs have entered the market, competition for airtime and audiences has increased.

Every producer has a stake in presenting his or her program to its audience and other interested parties in the most favorable light. What now exists is the creative use of statistics to achieve this goal. One religious television program after another claims to be Number One. And there is a fair chance that they are all telling the truth. Everything depends on the statistic being cited. Clearly this is a case where the

---

154

phrase caveat emptor applies. Viewers need to develop judgment and consumer sense in listening to the competing claims.

Whereas the audience statistics have become politicized as a matter of market competition, Pat Robertson's candidacy for the presidency has renewed interest in religious television audiences as a potential political bloc. Who is listening? Do evangelical Christians represent an important new force in the political process in America?

Pat Robertson, far more than any other religious broadcaster in America, has utilized market research to understand his audience. Thus, it was not surprising to learn in 1985 that he had commissioned the A. C. Nielsen Company to conduct special research to better assess the size of his audience.

A cumulative study of "The 700 Club" audience should inevitably produce much larger numbers than the average quarter-hour figures. Thus, CBN's Nielsen project seemed self-serving at first glance. But CBN was interested in a number of vexing issues about audience size as well. For several years, CBN marketing personnel had been convinced that the average quarter-hour ratings were themselves an underestimation. Their analysis of volume of audience response showed a higher proportional response from households receiving "The 700 Club" on cable than from regular syndicated broadcasts, suggesting that cable households were underenumerated.

A third issue of interest was the matter of audience duplication. In order to measure cumulative audience, to check out their suspicions regarding underenumeration of cable audiences, and to examine the extent of audience duplication, CBN had Nielsen do special tabulations, utilizing their national meter sample, during the period of the February 1985 sweeps. Data were to be collected for the other top ten syndicated religious programs in addition to "The 700 Club."

Having followed the technical issues in the religious-televisionaudience debate for several years, the authors immediately recognized that this study would provide unprecedented data and insight. When the study was initiated, Hadden was president of the Society for the Scientific Study of Religion, an international organization of scholars from thirty-five nations, and Shupe was program chair for the organization's annual meeting.

The authors asked David Clark, CBN's vice president for marketing,

---

155

to share the findings with the international community of social scientists that would be gathering for the annual meetings of SSSR in Savannah, Georgia, in October 1985. The authors specified only that Nielsen send a representative to the meetings to certify that the data had been gathered in a manner consistent with the organization's high standards of scientific data collection and to answer questions from scholars regarding the research methodology. Both CBN and Nielsen agreed to participate. [14]

Before summarizing the findings, the authors should note that a few critics have claimed the statistics are inflated and that they were produced purely for political purposes. The authors do not agree with this conclusion, nor have we seen evidence that would give it plausibility.

Since one of us had been an early skeptic of the audience claims of televangelists, we approached these new results with great caution. We examined the data carefully before concluding that the statistics are accurate within the parameters of error normally expected with sample survey methods.

The highlights of the CBN/Nielsen study follow the three broad areas of concern described above.

1. The cumulative audience sizes were significantly larger for the ten programs studied than were the average quarter-hour audiences.

The average quarter-hour audience size for "The 700 Club," using Nielsen's conventional measure, was 424,500 households. The cumulative ninety-minute daily audience was 2,547,000 households, the weekly total was 7,216,500, and the monthly grand total of viewers came to 16,300,000. Thus, the daily cumulative audience is six times larger than the average quarter-hour, the weekly audience thirty-eight times larger.

The top ten television programs had a combined average quarterhour audience of just under 8 million. The cumulative figure for a week came to 27 million and for the month 68 million.

2. The national meter sample, which included households receiving the broadcasting via cable, revealed significantly higher average quarter

---

156

hour audiences than had been measured by the conventional method of averaging the diaries from 200 separate markets.

The data reveal that all quarter-hour estimates obtained by the conventional method are undercounted. Falwell's quarter-hour estimate increased from 594,300 to 1,358,400 and Jimmy Swaggart's from 1,443,300 to 2,631,999. For the top ten programs, the meter method produced an average quarter-hour audience 82 percent higher than the conventional method.

3. A significant proportion of the audience estimates, regardless of the method of measurement, consists of duplication, that is, counting the same viewer more than once. Still, the total cumulative audience of religious television is far from insignificant.

The cumulated monthly audience-cume, in the trade-for the top ten religious programs was 67.7 million households; the unduplicated cume, 34.1 million households. On a monthly basis, then, the duplicated cume is inflated by 98 percent. Since Nielsen estimates a total of 85.9 million TV households in America,l5 the unduplicated figure indicates that, on average, 40 percent of all American households watch at least one segment of religious television each month. The unduplicated figure for an average week is 21 percent.

The credibility of these figures is supported by a Gallup survey, conducted in April 1987, in which 39 percent of the American public reported viewing a religious program within the previous thirty days and 25 percent said they had watched during the previous seven days. 16 That weekly figure of 25 percent is up sharply from the 18 percent reported in 1983.

Furthermore, televangelism is a regular part of the religious diet of evangelical Christians. In the Gallup survey, 79 percent of all evangelicals reported watching religious broadcasting; 46 percent reported viewing in the week preceding their interview.

CBN has continued to have Nielsen monitor its own "700 Club"

---

157

audience with the more sophisticated cumulative audience surveys. Regrettably, there are no data for other religious programs, so there will not be an opportunity to study trends.

The cabling of television in America has increased the complexity of measuring audience trends. Each year sees expanded cable penetration. From October 1983 through February 1987, the number of households in the United States receiving television via cable grew from approximately 31 million to 42 million, representing an increase in cable penetration from 38 percent to 48 percent. [15]

With cable television has come a decline in the networks' share of television viewing. From 1981 through early 1987, the networks experienced a decline from 83 percent to 76 percent of the audience share during prime-time viewing. [16] During this same period, religious broadcasters moved increasingly into cable. The syndicated program analysis books published by Arbitron and Nielsen largely miss the cable audience, to the point where the utility of the syndicated reports is dubious.

Consider the following: From 1980 through 1985, Arbitron figures show an increase in audience size from 20.5 million to 23.9 million, for a net growth of 16 percent. [17] Averaging just a little over 3 percent annually, this is not a torrid pace of growth, but still an upward trend. Then, in 1986, the combined Arbitron figures show a sharp drop to 19.2 million, a decline of almost 20 percent.

Nothing in the collective experience of syndicated religious broadcasters would confirm a calamity of this magnitude. But data prepared by Nielsen for CBN suggest the decline may be substantially attributable to the crossover to cable broadcasting. The CBN data do show a dramatic decline of 17 percent in the average quarter-hour audience for their syndicated broadcasts of "The 700 Club." [18] But the daily cumulative audience for syndicated and cable programming combined shows a slight decline of 2 percent, while the total cumulative monthly figures show an increase of 2 percent.

If caution is the best advice in dealing with any data on religious broadcasting, it is evidently difficult to communicate that advice. Reporters hungry for statistics, and critics anxious to make a point, continue to pick up whatever they can find, or whatever makes a point that supports their case.

The feature stories that appeared in the days following the Bakkers'

---

158

resignation from PTL communicated lots of misinformation regarding audience size. Perhaps none was more confusing than the "Guide to Sunday Morning Services" published by People magazine. [19] Capsule summaries of six leading religious broadcasts indiscriminately mixed together average quarter-hour figures with cumulative weekly figures with household access via cable, terribly distorting any attempt to make an intelligent comparison.

The sharp declines in the average quarter-hour figures are certain to be picked up and heralded as evidence that the electric church is collapsing. Further, someone is likely to conclude that the evidence of a collapsing audience before the Bakker scandals lends credibility to the claim that Jimmy Swaggart or Jerry Falwell or someone was attempting a hostile takeover to buoy his own sagging audience size and revenues.

In the final analysis, the answer to the simple question, "How many people are watching?" is simply that we don't know. Furthermore, the inherent flaws in the data are likely to cloud rather than clarify the issue in the years ahead.

But the answer to the question, "Is anybody listening?" is an unequivocal YES. The audiences are not as huge as some of the televangelists and their supporters were maintaining prior to the addition of Nielsen and Arbitron data to the debate. But, just as certainly, the audiences are not as small as some of their adversaries would like the American public to believe.

The best assessment of the data would indicate that the general trend in audience size has been upward. But evidence of this growth is obscured by the inadequate measure of cable viewing, an everincreasing proportion of all religious viewing.

As the 1988 presidential primary process unfolds, there can be no mistaking the fact that the audience watching religious broadcasting is very large indeed. And the show Pat Robertson started, as measured by cumulative audience size, is the top drawing card. The Nielsen data indicate "The 700 Club" is daily reaching approximately 2.5 million viewers, and, on a monthly basis, more than 16 million tune in. That is a very substantial base on which to build a grass-roots political organization. But the major question remains-do television audiences translate into voter constituencies?

The powerful medium of television-aided by space satellites and

---

159

cable networks-has provided the vehicle for the consolidation of conservative Christian feeling and opinion. Television, in a word, can mobilize that most crucial resource for a social movement in American democracy: committed voters.

Slowly but surely, the electric church appears to be transforming its audiences. The process is far from finished, but it is underway.

Notes