Alec Gallup of The Gallup Organization, Princeton, New Jersey, is the son of George Gallup Sr., the firmís founder and one of the originators of the opinion poll.
He is a co-author of The Gallup Poll Cumulative Index: Public Opinion, 1935-1997.
New River Media Interview with: Alec Gallup
QUESTION: How did your late father, George Gallup, Sr., get involved in public opinion polling?
ALEC GALLUP: Well, he actually had a sort of heavy background in a way. It is sort of a natural, because when he had been teaching - actually his Ph.D. is in psychology - but he was teaching journalism at Northwestern. He was interested in what people read in the newspapers. And so he in effect was doing survey sampling very early on, as early as the early 1920s, where he did two surveys that I can remember at that time. One was for the St. Louis Post-Dispatch about the readership of the newspaper, and the other one he did for a department store in Des Moines, Iowa. Also, his Ph.D. thesis was on the reading of the newspaper and reading of magazines. So in a funny way it was sort of a natural extension. It was what he knew and what he was interested in.
He had been teaching at Northwestern, at the Medill School of Journalism. He was the acting dean there. And Ray Rubicam of Young and Rubicam, the advertising agency in New York, invited him to come to New York to be the first research director of an advertising agency. And when he went to Young and Rubicam, the sampling methods are very similar. The only difference is you are talking about marketing and advertising as opposed to political and social issues. He worked at Young and Rubicam for about three or four years, and in the course of that decided that this would be a terrific method to poll the public on issues rather than just marketing subjects. So he actually started work on the Gallup poll experimentally in, I think as early as 1932.
One other influence was his mother-in-law. She was the secretary of state for the state of Iowa, and it was very unusual to have a female at that point who was the secretary of state of any state. And he did some surveys to see how she would do in the election. He did an early poll in which he found that she had an excellent chance and probably would win, and he was right on the button, right on the mark. But of course that was an unofficial poll. It was done for her and never published or anything like that. So he was experimenting in the very early 1930's to use his sampling procedures.
QUESTION: What are some of the anomalies of his early survey findings in the mid-thirties, for example that people found the government spending too much on relief?
ALEC GALLUP: Well, given the fact there was a Depression, they always felt as too much spending and, given today's climate, today, they keep talking about spending. Nothing has changed, and here it is sixty-five years later and they're still talking about government spending. You could ask it at any time - the middle of the Depression or in good times - they are always saying the government is spending too much. So that wasn't in a way looking back, given the benefit of hindsight, that's not strange.
There's another rather unusual situation too, that even during the Depression, when you asked people if they were conservatives or liberals or moderates or whatever, no matter how you asked it or what scales you used, people are conservative. And that, now, that does throw me. That surprises me. But if you look back, they always do. It never changes. I think the word "liberal," a lot of people associate it with permissiveness almost.
QUESTION: Was your father politically unaligned?
ALEC GALLUP: Yeah, I think he was. He's very hard to read. He worked very hard to be detached. And it was even difficult for myself and my brother and my mother to try to figure out where he was. He was very much in the middle - always has been kind of in the middle. He was upset about, as a lot of people were, even on the left, or in the middle, about the court packing in 1936. That got him a little uptight. But for the most part he's hard to read.
QUESTION: What led him to "bet" on FDR in the 1936 presidential election against Landon?
ALEC GALLUP: Well, I think that he felt that - well, the mood was one of - I think there was no question there was huge support for FDR for what he was doing. And as I say even though they might say they were spending too much, that if you took it act by act, they were supportive almost entirely. FDR had a very strong base of support throughout the 1930s.
I guess - maybe he figured he had nothing to lose, because he wasn't going anywhere - (laughs) - I mean at the time. He was positive he was right, and he felt that this was the right thing to do. It would have been more of a challenge for The Literary Digest in a way. What is not known is that they had a pretty good record going back into the early part of the century actually. They had done very well. One of the reasons they had done well [was that] there wasn't this split on the basis of economics back then. But liberals, I mean Democrats and Republicans were on both sides of the upper-income, lower-income and what have you. But with the Depression you had that split, and that changed everything.
He knew because, remember, the Literary Digest was using car registration lists and telephone books. Well, back then in the 1930s, if you had a car or a telephone you were definitely up market and usually a Republican. So that's what happened. And he knew that was going to happen, and he was absolutely sure he was right. So he was able to do that. And our numbers were very good too, so although we were still six or seven points too low on the Democratic side. However, the main thing is he called the election, and he called the right side, whereas the Literary Digest called the Republican side.
QUESTION: What improvements in methodology his he use that the Literary Digest's poll did not use?
ALEC GALLUP: Well, his method was random selection. He realized that there was a systematic bias to use automobile registration lists and telephone books, and he discovered that very early. In sampling them he got an overwhelming Republican kind of responses. And also, I guess more important the political party ID was very heavily Republican. So he knew. He said, this is crazy, this isn't right. And he worked really from that. He was able to get as random a sampling he could possibly get given the situation.
How it was done was, you went into the field. You had interviewers around the country, and the sampling points were selected totally at random. He would go in, and - it became much more systematic later on, but in those early days the one thing they did do was try to get the interviews all done in the household, so you would actually go to somebody's house and knock. It was a systematic procedure.
You will see a lot of pictures we have around with people interviewing farmers in the field and all that. That was mostly PR, that they did not like them doing that. They weren't supposed to interview people on the street, because there's a tendency when you interview people on the street is to get the most accessible people. So you are introducing a systematic bias when you do that. You may have a sample of people who like to talk to people, and they can be atypical say. And so what they would do is go house to house. The idea was to make it as systematic as possible. If you make it systematic you avoid bias.
QUESTION: What was your father's reaction when the election went for FDR, the way he had predicted?
ALEC GALLUP: He was so convinced he was right that there wasn't this great, Hooray, we got it right! - or like winning the lottery - it wasn't one of those situations. It was, Hey, I knew it - it had to be that way, because that was the way it was going to happen. And one of the interesting - this is sort of a sideline on this - he thought [the opinion poll] was a great contribution to democracy and all, but he didn't think it would have a lot of staying power. He thought that it would probably run its course in three or four or five years, which is very not generally known; that he felt that the media would get bored with it after a while, see, and the newspapers would drop him. Never happened. A little sideline.
As a matter of fact, if you look back, Gallup was virtually alone. Elmo Roper of the Roper Survey for Fortune magazine was around, but for the most part Gallup was alone - you know, naked and alone most of the time - until I would say about 1975 when the media - when the New York Times and the big networks got into it, got into polling its competitors. Gallup was alone. As a result he got a lot of credit, and it was great for Gallup, but at the same time if you were wrong, as in 1948, it was pretty brutal, but he really took it.
QUESTION: What was the legacy of the 1936 election success for Gallup and scientific polling?
ALEC GALLUP: Well, what it did was it made it not only acceptable but made it credible. I mean, when he got that one right, the poll just took off and the papers, it was a syndicated newspaper service, as you know, and he signed the papers up, and they paid pretty good money back then. One of the reasons that we actually weren't as active in market research in the early years as we might have been, was there wasn't enough money in it. The papers were paying big money.
I once asked my father why was it so successful - why did it take off like it did? He said - it's something I wasn't even aware of - he said the decade of the 1930s for a very short period of time, from let's say 1932 until the war - was sort of the era of the Sunday newspaper, and they needed filler back then for that Sunday paper. And so he cranked out these full pages - you know, surveys and things - and it was great filler. And this was, for a very short period, was the era of the Sunday newspaper.
QUESTION: What were some of the reasons why Gallup was wrong in the 1948 presidential election when he picked Dewey over Truman?
ALEC GALLUP: One, we stopped interviewing too early, about two and a half weeks, I think, before. And you can't do that. But there's also another factor that's not mentioned very often, and that was that the Wallace vote collapsed. As is often the case in an election, the third-party candidate will collapse. And Wallace had been getting a fairly high - I think as much as eight or 10 percentage points - and that collapsed completely and went back into the Democratic column. And that probably, in many respects, was the most important reason. I don't think it was a failure of sampling so much as it was those two factors.
But certainly the probability-based kinds of sampling, which came into being sort of in the early 1950s was definitely a superior method. And the difference is that before, the interviewer had some leeway into which households he approached. He had a given territory, but he could hit different households. Well, I don't think there's any question there's a tendency to go to the households that don't have dogs or don't have mean-looking people sitting on the porch, or they're going to go to places that are more accessible. And that does introduce bias. And so this new approach, probability sampling, was a definite improvement on the sampling procedures that we used.
[The election result for Truman] didn't have much impact in our newspaper business. I think we only lost one newspaper, believe it or not. I think we may have had as many as 150 to 200 newspapers that subscribed to the Gallup poll. And as I understand it, we only lost one. Now, maybe there were a few down the line after that, but it had almost no impact there. But it had a definite impact on our own commercial market research [and] on public opinion polling. It set us back; there wasn't any question about it. And as a result, when 1952 came, and even again in 1956, we were very, very cautious and worried, and we came up in 1952 with about three sets of figures to cover ourselves, and everybody was very sort of paranoid as a result of it for a while. And really, it damaged the industry as a whole. But we recovered and we came back. I think it took a good four to six years, I really do, before it completely came back. And by then it was okay. And then we did very well in 1960. We were right smack on target in one of the closest races, if not the closest race, in history. We were right on, and that helped. And from that point on, we've been very good, for the most part.
QUESTION: What was your father's view of the importance of the opinion poll and how would he feel about its use today?
ALEC GALLUP: Well, he felt that it was critical. It was the one dimension that was missing. Before it was, you know, smoke-filled rooms. And no matter how you felt about it, you needed to know you didn't have to slavishly follow what the public wanted or what your polls showed the public wanted, but you at least needed to know where they stood. How could you make an informed decision without it?
You'd have to say he was a zealot. He was convinced this was the way of the future; this was the right thing to do, even though he took heat lots of times. And right now you hear an awful lot about the fact that Clinton won't make a move or the Republican candidates won't make a move without checking their polls out. And it's kind of cynical, and they use them for marketing and positioning and all the rest. And I think he'd still say, if he were around, he'd say although it's being overdone now, you still have to know where the public is. And this is what a democracy is about. Without it, what do you have?
Overall he'd be very proud. He'd think it's doing what it's supposed to do. And on balance, it's the right thing and it's the wave of the future and it's going to go this way. It's a little unfortunate some of it is a little overdone sometimes.
QUESTION: What was your father's personality like?
ALEC GALLUP: Extraordinarily laid back. I've never known anybody who was as calm. You almost never even saw him angry. I mean, he was almost phlegmatic. He was an intense guy and, you know, it was constant. His mind was moving all the time. But he never lost his cool. He never got angry. He was a moderately religious guy, I would say. And as I said, I would place him very much in the middle. I think he worked very hard to be - for instance, he would never vote. He never voted. This was one of the things people couldn't understand. The reason he didn't vote is because he didn't have people asking him how he voted. He said, "I don't. It's easier not to." He said, "Everybody should vote but me." He said, "I don't want to do it, because then people ask me, and then they say, 'Oh, you're a Republican or you're a Democrat.'"
QUESTION: What about some of the criticisms of polling, such as the bandwagon effect, manipulation of opinion and so on?
ALEC GALLUP: Oh, in terms of bandwagon, he didn't believe that that was true. All you had to do was look at 1948, I mean, in places and situations where there was no bandwagon there. And he felt that that always had been sort of exaggerated, that that really wasn't the case.
Well, he always felt that that was a danger [of manipulation] and that did happen, but that knowledgeable people and people who are experts in the field could spot that. You can tell loaded questions, and it happens all the time. And there are certain organizations in America, who will go nameless, who do spin off loaded questions, and they're very obviously loaded. And if the question is objective, the sampling is done correctly, you're okay. And obviously, it can be used cynically by certain organizations who try to put a spin on, but there's less of it than you'd think.
QUESTION: Was there ever pressure put on Gallup by political partisans?
ALEC GALLUP: We would get calls from the . . . what would it be? What's the organization, the Republican organization? You would get calls wanting to know. And you're caught, because you don't want to give [it to] them. The data's in the public domain once it's released. And they would want you to release a poll three or four hours earlier, or earlier if you had it the next day. And we constantly were saying, "No, we can't do that. We can't do that." So the Nixon administration did bombard us in a way. They would suggest questions we should ask. We didn't pay attention. We wouldn't do them, because usually they were loaded and they were obvious what they were doing. And we'd just politely say, "That's an interesting question." Click.
QUESTION: Has Gallup become synonymous with polling?
ALEC GALLUP: It's a generic word in a way. And what's interesting, it's used in Scandinavia as a generic term. It's the word for survey. So you'd have a Harris gallup or a Roper gallup. The word for poll is a gallup, with a small 'g', I guess. And so it is. And, as a matter of fact, it's a little spooky, the generic part of it. It causes us a lot of problems, because in Scandinavia you really can't say, "Hey, you can't use our name," because they can use it, because it means survey.