Broadcasting (Oct 1931-Dec 1932)

Are They Listening? A Reply to A. N. P. A. CBS Research Experts Analyze Data Accumulated by Press To Prove Radio Advertising Effective, Far-Reaching NTENDED obviously to discount he efficacy of radio advertising by howing that a relatively small percentage of radio set owners are istening at a given time, and that |;ven when listening few are conscious of the product being advertised, the telephone survey recently Conducted by the Bureau of Advertising of the American Newspaper Publishers Association has been jmalyzed by research experts of pBS to show that just the opposite Conclusions may be fairly reached, i. "Are They Listening?" was the itle of the survey, conducted by E30 newspapers in various cities, ;ach of which was requested by jthe A. N. P. A. to telephone 100 or |nore homes and ask the following ijiuestions : 1. Do you own a radio? 2. Are you listening to your radio his evening? 3. To what station are you listenng? 4. What is the program supposed to dvertise ? Translated into a simple table, >f which much has been made by mti-radio interests, notably Editor i, Publisher, newspaper trade irgan and bitter opponent of radio, he A. N. P. A. survey sets forth he following conclusions: lj 1. Fifty-one per cent of all homes n the United States have radios. 2. Twenty-four per cent of all omes in the United States are listenng to radio programs in the peak iours of the evening. 3. Thirteen per cent of all homes n the United States represent the laximum audience reached in these ours by the "best" stations. 4. Seven per cent of all homes in he United States are reached by a ngle network program at these hours. Survey Not Impartial "■ROM the standpoint of research rocedure, the CBS analysis points ut: (a) that the data were not •athered by independent investigaors but by newspapers, some of /horn feel their interests are in onflict with radio as an advertisng medium; (b) that the work was vidently not controlled directly by ,ny central source nor supervised y the personal presence of any mpartial authority, but was done ndependently and variously by ach of the 230 newspapers, who hen forwarded their own findings, nd (c) that the participation of In outside research authority, 'ercival White, Inc., was apparenty limited to tabulating and "adusting" the data and did not begin ■ ntil the actual survey was competed. Upon the conclusions reached by he A. N. P. A., the CBS experts lake the following pertinent comlents: "Comments on Conclusion 1 : j 51 per cent of all homes have adios). It is interesting to note " at the 51 per cent used as the asis for all the other figures in e survey was not the percentage f radio ownership determined by JUST AS THE DEVIL may quote Scripture for his purpose, so may almost any interpretation be applied to statistics, particularly if they are gathered to fortify rather than test a preconceived belief as in the case of the A. N. P. A. survey, which was intended to belittle the influence of radio advertising. CBS experts have turned the tables on the publishers in this article by showing that just the opposite conclusions may even more logically be reached by using the same data that the prejudiced newspapers collected. Using the A. N. P. A. conservative estimate, the network shows that any national advertiser may reach an audience of 25,000,000 at peak hours if he utilizes rebroadcasts for western zones. the newspapers which made the survey. They found radios owned in 86.0 per cent of all the homes they interviewed. This 86.0 per cent (miscalculated in the A. N. P. A. report as 82.4 per cent for 19,346 divided by 22,520 is 86 per cent. The incorrect figure of 82.4 per cent published by the A. N. P. A. is due to an error of 1,000 in subtracting 5,117 from 27,637. The result shown in the A. N. P. A. table is 23,520 instead of 22,520) has been abandoned completely in arriving at all of their conclusions, and 51 per cent has been substituted. This percentage (51) is the estimated number of sets for the entire country, including rural and desert areas where the population and markets are thin, and thus, does not represent areas which parallel the circulation of metropolitan newspapers. "Comments on Conclusion 2: (24 per cent of all homes listening). The figure of 24 per cent represents homes actually listening to their radios at the hours of the interviews. This figure, too, is not a direct conclusion from the interviews obtained, but is 'adjusted' on the basis of 51 per cent set ownership. As a matter of fact, the A. N. P. A. survey reveals that 50 per cent of all homes interviewed (or 60 per cent of all set owners) had their radio sets turned on. Juggling Statistics "COMMENTS on Conclusion 3: (13 per cent listening to 'best' station). From this point on and including Conclusion 3, the A. N. P. A. survey appears to depart still fui'ther from the data on which it is based. Quoting from the report in Editor and Publisher of March 26, 'We may assume that 52.4 per cent of the 24 homes with radios turned on were found to be listening to the 'best' station. Thus it is indicated that slightly less than 13 out of 100 homes consciously are listening to the most popular nearby station. "Since the A. N. P. A. survey at this point falls back, in its own language, on 'we may assume' and 'thus it is indicated,' it can hardly be considered to be statistically significant from here on. Overwhelming and authoritative evidence is available to show that the proportion, of people who listen to any one station at any given hour is preponderantly affected by the program broadcast over that station. It is known for instance, that the least popular of the four big stations in New York which ordinarily commands a good deal smaller audience than any of the other three, had 65 per cent of the entire listening audience at a certain Sunday hour when the programs on the other stations were restricted in their aprjeul. It has likewise been determined by surveys identical with the A. N. P. A. telephone method, but conducted bv advertisers and agencies, that a single station may command at a given hour as much as 80 or 90 per cent of the total listening audience in its area at that time. Thus the A. N. P. A. conclusion that a maximum audience reached by the 'best' station is a little less than 13 out of 100 homes is hardly significant. "Comments on Conclusion U : (7 per cent listening to a single program). The same fallacies which betray Conclusion 3 reduce Conclusion 4 to a complete statistical paradox. The A. N. P. A. bureau is to be congratulated on its frankness in prefacing this conclusion with the phrase. 'The bureau believes, however, that if its methods were applied on a sufficiently wide scale to determine the coverage of a single program, it would be shown not to exceed 7 homes in each 100.' No competent radio broadcasting authoritv would deny, of course, that a single net work might command an audience as low as 7 per cent of all homes in the United States (representing over 6,000,000 listeners.) Another program, on the other hand, may attract an audience as high as 20 to 25 per cent of all homes (an audience of 20,000,000 to 25,000,000 people) — the exact size of the audience between such limits depending obviously on program appeal." Re Listener-Consciousness THE A. N. P. A. survey then goes on to what it calls the subject of "listener-consciousness of the advertising program." In reply to the question, "What is the program sunnosed to advertise?" it states that only one-third of those who were listening were able t > mention the product advertised. There are several obvious factors, says the CBS analysis, to be considered in appraising the soundness of the conclusion: "A. The commercial announcement tends overwhelminglv, in network programs to be placed in the second half of the -program. It is upon this advertising announcement that the advertiser depends in using radio as a medium, not upon the mere mention of his name in the introduction of the program. Since obviously half of the calls during each 15-minute period occurred before the commercial announcement in the program (assuming that the interviews were made consecutively), the A. N. P. A. was asking listeners to identify the product advertised before, in many cases, it was even mentioned, before they could have heard it — at least before they could have heard the principal advertising announcement contained in the program. Thus the question, 'What product is it supposed to advertise?' might be compared, in this instance, to concealing the lower half of an advertisement containing the signature of the advertiser, showing it to someone, and asking whose advertisement it was. "Further exception might well be taken to the wording of the question asked. Many programs stress an institution rather than a product. The question, 'What advertiser is sponsoring the program?' would seem to recommend itself if impartial results were desired. "Both of these two factors — asking the question before the commercial announcement was made, and asking for the name of the product instead of the advertiser — would inevitably tend to depress the results. "B. What about sustaining programs? The A. N. P. A. report states that two-thirds of those who were listening were unable to mention the product advertised. How could they identify the name of a product on a sustaining program? This oversight needs no stressing. Suffice it to say that even in the eastern time zone, sustaining programs are broadcast over the network on various days of the week between 8:30 and 9:30 p.m., and in the western territories this would occur even more frequently. This is (Continued on page 20) lay 1, 1932 • BROADCASTING Page 9