An Introduction to Mass Media Research
Roger Wimmer, Ph.D.
While many people place the task of learning research high on the “Not to Do” list, the reality is that research is probably the only area in mass media that relates to everything. There isn’t a position in any of the mass media that doesn’t conduct or use research. That’s the reality. And if you plan to be successful in mass media, you must understand at least the basics of research.
This doesn’t mean that you must become a statistician with a plastic pocket protector for pens and pencils. What it means is that you need to strive to become a data analyst, or a person who knows how to use statistics and knows how to interpret the results of the methods. This doesn’t mean that you’ll never have to take a statistics class. You will, and probably more than one class because there are some formulas and procedures that you’ll need to know. But . . . you won’t have to become a statistics “geek.”
Three things are important before we begin the discussion about research:
Let’s get started. First, what is research? In the 10th edition of our book, Mass Media Research: An Introduction, Joe Dominick and I define research as: an attempt to discover something (Wimmer & Dominick, 2014). That’s all there is to it—an attempt to discover something. In addition, if you think about that definition, you will see why research is so important in the mass media. Consider for a moment some of the questions you could investigate:
How many times have you thought of a question while you were listening to the radio, watching TV, or reading a newspaper or magazine? For example, who decides which songs to play on a radio station? What types of articles are the most popular in your favorite magazine? When is the best time to introduce a new TV program? Where do most people go to get local news information? Why do local TV news programs spend so much time on sports reports when research studies show that sports reports are the least liked element of local TV newscasts?
Every time you develop a “who, what, when, where, or why” question when using the mass media, you have raised a question that can be investigated. And that’s what this discussion is all about— trying to find answers to mass media questions. However, before we start, we need to understand a little about business and decision-making.
The importance of research in the mass media becomes clear when we understand something about business (although this understanding also relates to everyday personal life such as relationships with family, friends, and co-workers). However, in reference to business, what ever the business may be, the goal is always very simple and the goal is always the same: Get the highest numbers. The numbers can be things such as audience size, profits, or unit sales.
That’s it. That’s the goal of any company— get the highest numbers. It isn’t more complicated than that. Now, getting the highest number may not be as easy as the goal implies. In fact, in most cases, it takes a lot of time and effort to be successful. But, in addition to finding out that every company has only one simple goal, my 25+ years of experience in both private and academic sectors has taught me that reaching this singular goal is much easier when a company’s decision-makers follow a simple approach. In this case, it’s a three-step process to reach the goal of getting the highest number:
1. Find out what the people want (customers, audience, readers, etc.).
2. Give it to them.
3. Tell them that you gave it to them.
I have seen this three-step process work successfully for both media and non-media businesses. When you think about these three steps, you should see that companies (or individuals) that follow the steps, quite simply, cannot fail. How can a business or individual fail if they give the people (customers, etc.) what they want? Frankly, they can’t. The problem is that not all companies (media and non-media) put this three-step process to work. In most cases, the decision-makers follow the axiom of not being able to see the forest for the trees. They make their business to complicated.
In addition to the problem of decision-makers not using the three steps, there is a problem with consumers or customers. The problem is that in most situations people don’t really know what they want or need. They may have an idea, but they don’t know how to verbalize it. That’s where researchers become valuable. Researchers are the people who find out what people want and need so that decision-makers can give it to them (researchers and decision-makers may be one-in-the same).
However, researchers are not limited only to getting involved in finding out what people want. Researchers also get involved in the other two steps of the process. For example, in reference to “Give it to them,” researchers find out the best way this can be accomplished. Assume for a moment that a television research project about programming uncovers an audience desire for more game shows on TV. Researchers need to find out such things as: What kind of game show? When should the game show be aired? What types of contestants should be included? Who should be the host? And so on.
Then there is the third step in the process: “Tell them that you gave it to them.” Once a new game show is developed, the viewers need to know that the show is available. Researchers find out how the program should be advertised and promoted. They test different approaches to find out which is most successful.
Research is involved in every step of the process. And regardless of whether you are involved in writing, production, talent, advertising, distribution, or any other area, you will be involved with research. There is no area of mass media that is not affected by research.
Hypotheses and Research Questions
A typical question to ask at this point is, what kinds of questions do researchers investigate? As Wimmer & Dominick (2014) note,
Mass media researchers use a variety of approaches to answer questions. Some research is informal and seeks to solve relatively simple problems; some is based on theory and requires formally worded questions. All researchers, however, must start with some tentative generalization regarding a relationship between two or more variables. These generalizations may take two forms: research questions and statistical hypotheses. The two are identical except for the aspect of prediction-hypotheses predict an experimental outcome; research questions do not.
To make a rather complicated explanation short, research questions are used when researchers conduct preliminary research and are not interested in testing the statistical significance of their findings.
Research hypotheses are used in situations where researchers are well versed in the topic under investigation and wish to make predictions based on their data. Refer to the Internet for more information about research questions and hypotheses.
Basic Types of Research Approaches
There are two broad categories of research: qualitative research and quantitative research. Historically, qualitative research has been defined as research using small samples (oftentimes not randomly selected) where the results are not generalized to the population from which the sample was drawn. Conversely, quantitative research has been defined as research using a large, randomly selected sample of people where the results are generalized to the population from which the sample was drawn. These are the definitions that most people accept, and there are many books and articles (and even college classes) on qualitative and quantitative research.
There is nothing wrong with any of the materials or the classes. What is “wrong” is that the definitions of the two categories are not appropriate for today’s research. Let me explain.
During the past several years, the lines of distinction between qualitative and quantitative research have essentially disappeared. The real change has been with qualitative research, not quantitative. What was once considered a qualitative approach can be either a qualitative approach or a quantitative approach. A step back in time is necessary.
As mentioned, the definitions of qualitative and quantitative research have always included the type of sample used: qualitative research uses small samples, quantitative uses large samples. But what happens when a qualitative approach uses a sample size equal to that which might be used in a quantitative approach? What happens if the results from a qualitative research study are generalized to the population? Finally, what happens when both items are combined into a qualitative study: a large sample where the results are generalized to the population? Is this still a qualitative study? The answer is obviously, “no,” and it is clear that differences between qualitative and quantitative research based only on sample size or intended use of the data are inadequate.
Consider this: Why can’t a qualitative approach such as focus groups use a sample of 100 people? It can. Why can’t a one-on-one interview approach, another qualitative method, use a sample of 100 people? It can. That is what is now commonplace. Researchers oftentimes use a large sample of people with a qualitative approach. But the historical definition of the method suggests that this research isn’t as “valuable” or as “good” as it might be if a quantitative approach were used.
If qualitative research can use a large sample and results can be generalized to the population, then what happens to the lines of distinction between the two methods? If you said, “They disappear,” then you are correct. That is what has changed. Dividing qualitative and quantitative research only based on sample size and the intended use of the results no longer makes sense. That is what is wrong. We need a few new definitions. Consider these:
Qualitative research: in-depth investigations using flexible questioning.
Quantitative research: in-depth investigations using inflexible or less flexible questioning.
These definitions provide a new look at qualitative and quantitative research. These definitions mean that both qualitative and quantitative research,
The difference between the two methods relates to how the data collected, or how the questions are asked. The definitions do not include anything about sample size or the intended use of the results. This doesn’t mean that qualitative researchers “wing it” when it comes to asking questions. They don’t. In fact, there is no difference in the steps involved in designing a qualitative or quantitative research project. Qualitative research is not the “step-child” of quantitative research as some people may think.
Qualitative research is a legitimate, reliable, and valid research methodology. In a typical qualitative study, researchers start their research with a pre-determined set of questions, procedures, and plans. But that is where the similarity to quantitative research ends.
The qualitative researcher has the option to ask unique follow-up questions, change question order, or even ask new questions during the course of data collection. Flexibility is the key to qualitative research, and it takes an expert to conduct a qualitative study.
The flexibility in data collection (or questioning) isn’t possible in quantitative research. Quantitative researchers develop a measurement instrument such as a telephone questionnaire that is answered by every respondent. With the exception of certain skipped questions (questions for which a given respondent isn’t qualified to answer), each respondent answers the same questions as every other respondent in the research study.
That’s the only difference. Sample size doesn’t matter. Sample selection doesn’t matter. The intended use of the results doesn’t matter. Either approach can mix and match any of these elements. So, for example, a quantitative study may use a small sample of available respondents where the results are intended only to gather preliminary indications of a phenomenon, question, or problem.
To most people, the word “statistics” congers up many negative thoughts—hate, disgust, fear, and so on. Why? It may be that most people don’t like the word statistics because the word just sounds like a difficult area. Some people actually are afraid of statistics and will do anything to avoid anything “statistical.” (There may be some credence to this fear. Have you ever noticed how closely the word “statistics” relates to “sadistics?”) However, in reality, learning statistics is just like learning anything else. The key is to understand the language of the statistics. There is no reason to fear statistics. With that in mind, let’s move on.
First, what is statistics? One generally accepted definition of statistics is: the science that uses mathematical methods to collect, organize, summarize, and analyze data. In reality, statistics are merely numerical operations. There is nothing magical about the procedures. Statistics alone will not “correct” a misdirected, poorly phrased, or ambiguous research question or hypothesis, or a study that uses sloppy measurement and design and contains numerous errors. Statistics provide valid and reliable results only when the data collection and research methods follow established scientific procedures. Statistics are needed in most cases to allow researcher to discover something (Wimmer & Dominick, 2014).
Types of Statistics
There are two broad types of statistics: descriptive statistics and summary statistics. Descriptive statistics allow researchers to reduce data sets to allow for easier interpretation. For example, if you asked a random sample of 100 people which television programs they watched last night and recorded all 100 answers on a sheet of paper, it would be difficult to draw conclusions by looking at that list. Analysis of this information is much easier if the data are organized in some meaningful way. This is the function of descriptive statistics.
These statistical methods allow researchers to take random data and organize them into some type of ordered fashion (see Wimmer & Dominick, 2014).
Summary statistics help make data more manageable by measuring two basic qualities of data distributions: central tendency and dispersion (variability). Central tendency statistics answer the question, what is a typical score? These statistics provide information about the grouping of the numbers in a distribution by giving a single number that is characteristic of the entire distribution. Central tendency statistics include such things as the mode (the score or scores that occur most frequently), the median (the midpoint of a distribution), and the mean (average score). These singular numbers are used to describe an entire data set.
Descriptive statistics are used to measure dispersion, or variation. Where measures of central tendency determine the typical score of a distribution, measures of variation describe the way the scores are spread out about a central point. Dispersion statistics are valuable when researchers compare data from different distributions, such as people who read newspapers every day to those who read newspapers only on Sunday. The common measures of dispersion are the range (the difference between the highest and lowest scores in a distribution), variance (the index of the degree to which scores are different from the mean), and standard deviation (the square root of the variance). For more information on these topics, refer to the Internet and consult the references listed at the end of this chapter.
Nonparametric and Parametric Statistics. Statistical methods are further commonly divided into two additional categories: parametric statistics (inferential) and nonparametric statistics (non-inferential). Historically, researchers have recognized three primary differences between parametric and nonparametric statistics:
1. Nonparametric statistics are appropriate with only nominal and ordinal data. Parametric statistics are appropriate for interval and ratio data. (The nominal level is the weakest form of measurement where numerals or symbols are used to classify persons, objects, or characteristics. Ordinal data are usually ranked along some dimension, such as from smallest to largest. Interval data have equal distances between points, but lack a true zero point, such as a thermometer. Ratio data have equal distances between points and also a true zero point, such as weight.)
2. Nonparametric results cannot be generalized to the population. Generalization is possible with only parametric statistics.
3. Nonparametric statistics make no assumption about normally distributed data; parametric statistics assume normality.
Some nonparametric statistics include the chi-square goodness of fit, the Kolomogorov-Smirnov test, and contingency table analysis (cross-tabulations). Some parametric statistics include the t-test, analysis of variance (ANOVA), basic correlational methods such as the Pearson Product-Moment Correlation, linear regression, and multiple regression. All of these topics are beyond the scope of this discussion. However, keep in mind that all of the procedures can be learned very easily. See the Internet for more information about parametric and nonparametric statistical procedures.
Univariate and multivariate statistics. The final distinction among statistical procedures categorizes methods in terms of the number of dependent variables tested in a study. If one dependent variable is tested (there is no limit on the number of independent variables), then the procedure is a univariate statistical study (uni = one, variate = dependent variable). If two or more dependent variables are tested simultaneously, then the study is classified as multivariate.
A simple example will help. Assume that a researcher wants to try to find out the affect of light and heat on taking a written vocabulary test. One group of respondents takes the test in a room that has adequate lighting; a second group takes the test in which the lighting consists of only one small candle. With all else being equal, the differences between the two group’s vocabulary test scores can be attributed to the amount of available light. One measurement is made, and this is a univariate study.
Now assume that a researcher wants to do the same type of research, but in addition to the vocabulary test, the researcher also measures the respondents’ ability to memorize a list of random numbers. Two measurements are made and this is a multivariate study. Some multivariate statistics include factor analysis, cluster analysis, discriminant analysis, and canonical correlation. Once again, these topics are beyond the scope of this chapter. Search the Internet for “univariate statistics” and “multivariate statistics” for numerous examples of both statistical categories.
We now have a basic understanding of what research is, and the various types of statistics. The next step for a researcher is to understand the elements of the scientific approach—the method that researchers use. Why the scientific method? What differentiates the scientific method from other ways of learning things or other ways of investigating? We need to understand this because it’s important in understanding the overall concept of research.
Methods of Investigation
As discussed by Wimmer & Dominick, there are a variety of ways to investigate a research question or hypothesis. In 1986, Kerlinger, using definitions provided nearly a century ago by C. S. Peirce, presents four approaches to finding answers, or “methods of knowing.” They are tenacity, intuition, authority, and science. As Wimmer & Dominick (2014) state:
A user of the method of tenacity follows the logic that something is true because it has always been true. An example is the storeowner, who says, “I don’t advertise because my parents did not believe in advertising.” The idea is that nothing changes-what was good, bad, or successful before will continue to be so in the future.
In the method of intuition, or the a priori approach, a person assumes that something is true because it is “self-evident” or “stands to reason.” Some creative people in advertising agencies resist efforts to test their advertising methods because they believe they know what will attract customers. To these people, scientific research is a waste of time.
The method of authority promotes a belief in something because a trusted source, such as a parent, a news correspondent, or a teacher, says it is true. The emphasis is on the source, not on the methods the source may have used to gain the information.
The scientific method approaches learning as a series of small steps. That is, one study or one source provides only an indication of what may or may not be true; the “truth” is found through a series of objective analyses. This means that the scientific method is self-correcting in that changes in thought or theory are appropriate when errors in previous research are uncovered. For example, in 1984 Barry Marshall, a medical resident in Perth, Australia, identified a bacterium (Helicobacter pylori or H. pylori) as the cause of stomach ulcers (not an increase in stomach acid due to stress or anxiety). After several years, hundreds of independent studies proved that Marshall was correct, and in 1996, the Food and Drug Administration (FDA) approved a combination of drugs to fight ulcers-an antacid and an antibiotic. In space exploration, NASA disclosed in early 1998 that water had been found on Earth’s moon, changing the centuries-old idea that water could not exist there.
The scientific method is the choice of mass media researchers because it includes all the steps that allow for the advancement of knowledge. The five qualities that differentiate the scientific method from the other methods of knowing or learning, adapted from Wimmer & Dominick (2014), include:
1. Scientific research is public. The advancement of science and scientific information depends on freely available information. Scientific researchers cannot claim that their methods, data, or findings are private knowledge. With some restrictions such as private sector research conducted by individual companies, institutions, or governments, scientific research must be available to the public.
2. Science is objective. Scientific research attempts to exclude the judgments of researchers. In other words, everything is done to ensure that objectivity is the guiding principle for design, data collection, and especially interpretation of results. As Wimmer & Dominick (2014) state: “Objectivity requires that scientific research deal with facts rather than interpretations of facts. Science rejects its own authorities if their statements conflict with direct observation.” The data and interpretations must fall where they may.
3. Science is empirical. Scientific researchers are interested in measuring concepts and phenomena (the term empiricism comes from the Greek word for “experience.”) They must be able to perceive and classify what they study and to reject metaphysical and nonsensical explanations of events. Scientific researchers reject knowledge that cannot be perceived, classified, or measured (Wimmer & Dominick, 2014). However, “this does not mean that scientists evade abstract ideas and notions; they encounter them every day. But they recognize that concepts must be strictly defined to allow for observation and measurement. Scientists must link abstract concepts to the empirical world through observations, which may be made either directly or indirectly via various measurement instruments. Typically, this linkage is accomplished by framing an operational definition (Wimmer & Dominick, 2014). Search the Internet for more information about operational definitions.
4. Science is systematic and cumulative. A major difference between scientific research and the other methods of knowing is that in science, a single research study does not stand alone, nor is it considered to be the final answer to any question. There are always additional studies (replications) to recheck or verify (or refute) what has already been discovered. If changes in beliefs, thinking, or procedures are needed based on the new studies, then that is what is done.
5. Science is predictive. Science is concerned with relating the present to the future (Wimmer & Dominick, 2014). While many mass media researchers investigate questions to gather information for immediate use, many studies are conducted to allow decision-makers to predict what may happen if certain decisions are made.
Now let’s see how all this information relates to mass media research.
Areas of Mass Media Research
As mentioned in the beginning of this chapter, research is used extensively in all of the mass media. The following discussion relies heavily on information from Wimmer & Dominick, 2014. For additional information, consult the references and search the Internet for information about all of the topics discussed in the following sections.
Electronic Media. Electronic media research studies today fall into two main categories: ratings and nonratings research. Two companies conduct the bulk of electronic media ratings in the United States: A. C. Nielsen (www.acnielsen.com and www.nielsenmedia.com) conducts local market and network TV and cable TV, and The Arbitron Company (www.arbitron.com) conducts local market radio ratings.
As Wimmer & Dominick state:
The research methodologies used by Arbitron and Nielsen are complex; each company publishes several texts describing its methods and procedures that should be consulted for specific information (listed in the references at the end of this chapter). The data for ratings surveys are currently gathered by two methods: diaries and electronic meters (commonly called people meters). Each method has its own advantages and disadvantages.
Search the Internet for more information about electronic media ratings. The websites by Arbitron and Nielsen offer a wealth of information. In addition, go to other sites such as www.variety.com and www.adage.com. There are many types of nonratings research used by the electronic media. This type of research provides information about what the audience likes and dislikes, analyses of different types of programming, demographic and lifestyle information about the audience. Some of the procedures include:
Print. While there are many types of research conducted by the print media, these are the types of research that have gained most attention in the past several years:
Advertising and Public Relations. Just as with the other media, research in advertising and public relations includes a variety of topic, some of which include:
Internet. The Internet has quickly become a mass medium and it is changes every day. However, one thing is certain about the Internet, and that is it will provide a countless number of research possibilities for mass media researchers. Consider some of the possibilities:
References and Further Learning
Hurlburt, R. T. (1998). Comprehending behavioral statistics (2nd ed.). Pacific Grove, CA: Brooks/Cole.
Jaccard, J., & Becker, M. A. (1996). Statistics for the behavioral sciences (3rd ed.). Belmont, CA: Wadsworth.
Kerlinger, F. N. (2001). Foundations of behavioral research (4th ed.). New York: Holt, Rinehart & Winston.
Langer, S. K. (1967). Philosophy in a new key: A study in the symbolism of reason, rite, and art (3rd ed.). Cambridge, MA: Harvard University Press.
Tukey, J. W. (1986). The collected works of John W. Tukey. Belmont, CA: Wadsworth and Brooks/Cole.
Williams, F. (1992). Reasoning with statistics (2nd ed.). New York: Holt, Rinehart & Winston.
Wimmer, R. D., & Dominick, J. R. (2014). Mass media research: An introduction (10th ed.). Boston, MA: Cengage Learning.
Further Learning. Two good websites provide unique opportunities to learn research and more. The first is OnlineCourses.com, described as:
OnlineCourses.com aggregates the best free and open courseware for students and professionals, all in a conveniently searchable online courseware platform. Search, track, and share progress on over 500 university and college level classes globally. Formerly known as LectureFox.com, we continue to carefully select and share with you new courses every week. Our passion for developing and sustaining OCW projects drives us to curate the best. From biology to accounting, foreign languages to science, OnlineCourses.com ensures you get a world-class education from the world's top schools and scholars anytime, anywhere.
A second website for further learning is Online University, described as:
Discover thousands of free online courses from top universities around the globe. We’ve compiled a list of quality online courses from more than 35 countries, allowing you to access the world’s knowledge with the click of a mouse.
Note: This article was updated on September 16, 2013.
If you arrived here via a search engine, click here.