Robert J. Gould on "Quantitative Research"

So I just wanted to make sure people know that you go search the literature, find out what others have done, find out about as much as you can because you’ll save an awful lot of time, an awful lot of money. That said, you’re then in a place to figure out, “Well, where are we?” in terms of answering all these questions about what we’re going to say and who we’re going to say it to. Very often if you have the opportunity to do or find or do quantitative research, it can really help you answer the question of, “Who are the best groups to go after? How many of them are there who would really be responsive to our message or need to hear what we’re trying to do that could create the best public health benefit?” And quantitative research, because it can be projected to a larger audience, a larger population, is extremely useful for that; whereas, qualitative research doesn’t achieve that. So if you’re really trying to get who are the big groups, a good example would be when we did the Five-a-Day campaign about ten years ago, the quantitative research helped us identify the big group, and those were people who were eating three to four servings of fruits and vegetables and we felt getting to that group and helping them get to that next level would give us the greatest “bang for the buck” in social marketing terms. So that’s a good example of the use of quantitative research.

There are existing databases that of extensive research that you can tap into without even asking any additional questions that can provide insights, so obviously those are the least expensive ways to go because you’re really just paying for an analysis of data that’s already there. An NHANES study, for example, is a very, was where we got our information about how many fruits and vegetables people were eating and a bit of other information about their demographics and other information about their lifestyle that was very helpful in initiating that campaign. There are other databases. In fact, the one I think our audience really needs to know about is the research that, database that Porter Novelli has established for this, primarily for the CDC but for other clients as well. It’s called the Healthstyles Survey. It’s conducted since 1995, and there you have a quantitative survey that already exists. You can ask additional questions that are particularly relevant to the program that you’re starting to initiate and want to find out more information about. But what’s important here is that much information is already in that database and, therefore, the cost of going in and saying, “Well, look, could you tell us about” -- and it’s the people who are this kind of demographic category: women of child-bearing age; or this kind of behavior: they’re smokers who would like to quit; or their attitudes: they are people who are very into health information seeking and would be the people we would want to encourage to do some other health information seeking. So there’s so much in that survey already and in the database that getting someone to tab, cross-tab, do a cross-tabulation and then gaining the insights that you need from the survey is relatively expensive. Or doing a recontact survey, is what it’s called to ask some more questions is inexpensive as well.

I think the practicalities are really to find one of these suppliers, including us, and give them a call and say, “Here’s what we’re trying to do. Can you show us what the survey has and what, what it would cost to do X?” If they do a recontact survey or do an analysis or what have you, or call an omnibus survey supplier -- there are many of them -- and jump on their survey. So those, the quantitative is very good also I will say -- and maybe I should emphasize this -- very often in the work we do in social marketing, we are being looked at very carefully to see what our programs really achieve. Once we decide what our strategy is, what are we trying to influence and include -- if we’re trying to raise awareness of a health issue, if we’re trying to change an attitude or a social norm about a health behavior, or if we’re trying to eventually change the behavior itself. Sometimes that last question can only be answered by very, very different kinds of surveys and different kinds of research; but if it’s intentioned to change behavior or reported behavior change, very often all three of those can be benchmarked and tracked through quantitative surveys and only through quantitative surveys. So I want to make sure people know that evaluation -- the summative evaluation or the progress kinds of monitoring; not just the planning and how do we do it, but the marking your achievements or lack of progress -- is very important.

I’m trying to think of an example for you. Probably one of the most famous recent social marketing programs that most people know about is the Truth campaign, the youth prevention campaign -- and there’s one nationally now, but the original campaign was in Florida -- and here, you know, there’s a great qualitative research story to tell. I’ll save that. But in the quantitative survey, once we decided that we wanted to change the social norms surrounding smoking, and that awareness we needed to change was an awareness that the tobacco industry really exploited youth, the attitude we wanted to change was an attitude of saying that, “We don’t like being manipulated by tobacco companies, and we believe that that’s not a good thing and they are bad people for doing that, bad grown-ups.” And the behavior obviously we wanted to change was and then the other attitude was there’s nothing particularly cool or uncool about smoking. It’s not related to being cool. And then the behavior we wanted to change was obviously the initiation of smoking behavior. So those were the things that our research -- that only quantitative research -- can track, and once we decided these are what we want to retrack all three. We have been tracking all three.

And what really created the insight was qualitative research done on the street -- sending young people out to hang out with what we consider to be edgy or high sensation seeking youth -- and just talking and hearing how they felt about smoking and how they felt about smoking, what they knew about the tobacco companies, which was nothing. They didn’t know, they wouldn’t know Philip Morris from a whole in the wall. They’d understand the brand, and they were into the brands. So everything that came out of the Truth campaign was based on an insight that you really could not talk about health and you could not talk from adults to youth. It had to be, let’s replace all the excitement and adventure and risk taking by getting a rebellion going against the tobacco companies, and they are the bad guys, and never talk about smoking. Never say don’t smoke. Ever, ever.

If you know absolutely nothing, as we knew absolutely nothing when we started the March of Dimes campaign on prematurity, there was no literature on what, who knew what about prematurity.

So there’s nothing to do quantitatively. We couldn’t figure out how to dimensionalize the audience beyond that we knew it’d probably be women would be a primary audience; but we knew nothing about the issues, so we conducted qualitative research to explore it and found a great insight there was that people felt prematurity -- they knew what it was and they knew it was very common -- but they had absolutely no idea it was dangerous. They felt that prematurity, they felt, “Well, it’s prematurity, but then most kids survive and it’s fine.” And when they found out that actually the incredible risks of disability and death from premature birth, that was a tremendous revelation and very motivating and very engaging, but that only, that was a great, that was a great example of qualitative research being the first thing that we did and then we did a national survey to corroborate and dimensionalize and identify the audience segments based on that qualitative research.

My tip would be that for surv, for survey research, it’s very easy to, it’s a higher risk proposition to do surveys, to do surveys because it’s a higher cost. So my tip would be, first of all, don’t do it ‘til you’re absolutely sure the question’s not been asked somewhere. It’s pretty rare that you find something that’s so new like the prematurity question that there’s just nothing on it; so don’t do that. And then try to find a survey that’s, again, as rich as possible in terms of other information because what happens very often is that you, you end up getting the question answered to your, how much do you care about X, or how much do you know about Y, or here’s your behavior that we care about. And then you ask your question and pay, say, “We’ve got to save money. We’ll put it on an omnibus survey.” And what you get is a whole lot of demographic cross-tabulations. You can cross-tabulate it against sex or gender and age and income, and it’s a pretty thin gruel of information; and if you really want to get the rich tapestry that can help you plan a communications campaign, try to figure out how to put it onto a richer survey of, you know, that asks people what their lifestyle is, what other health behaviors they’re engaged in, their beliefs, their attitudes, their media habits -- because very often you’re going to want to find out how do we reach them, what are the credible vehicles, who are the credible spokespersons, who is going to be the, you know, how do they respond to different audiences? So even if you don’t get that out of your survey, frankly, you’ve got to sort of then find out, “Well, what do we know from other research that could help answer those questions?” If you can get it all in one place, you’ve really got something going. And that’s why we’re so happy about this, again, for the health style survey because that’s what we’ve tried to do is get the attitudes, the behavior, the lifestyle, the demographics, the media habits, and all of this in one place in one survey with one group and, therefore, you can cross-tabulate to your heart’s content.