Quantitative research constructs marketing issues in terms of numbers and uses those numbers for statistical analysis. Typical sources of such data are surveys with scalable items, experiments with measurable independent and dependent variables, and purchase information obtained from scanner data. The goals of quantitative research may be descriptive (e.g., profiling market indicators in percentages), correlational (e.g., assessing associations between such factors as age and brand purchase) or causal-predictive (e.g., testing one direct mail piece against another).
Qualitative research is composed of the insights, themes and basic understanding of a situation. Though numbers may be used in conjunction with qualitative research, it is usually framed in verbal terms.
In comparing the two broad approaches of qualitative and quantitative research, the following issues are important:
- Objectivity vs. subjectivity: Qualitative research is widely considered to be less objective and scientific than quantitative research. By contrast, quantitative research is usually considered objective and scientific since it is formulated in terms of numbers and is subject to rigorous statistical analysis. Sometimes these two approaches are framed in terms of being positivistic (objective-scientific) vs. post-positivistic (subjective or even anti-scientific). However, since most quantitative research relies on good qualitative research, the argument over scientific validity is for all practical purposes moot.
- Degree of structure: Quantitative methods are considered to be structured with definite measures and types of analyses, while qualitative methods are said to be unstructured because their approaches to data collection and interpretation are much less defined and programmatic. Quantitative questions are sometimes referred to as closed-ended questions and qualitative ones as open-ended. For example, a closed-ended question might involve asking a person to rate his feelings about a commercial on a scale of one to seven. An open-ended question might be, "Tell us whether you like or dislike this commercial and why."
- Method of analysis: Quantitative methods employ a wide variety of statistical methods that look for statistically significant and behaviorally meaningful differences in predicting behavior. Statistical significance means that a variable has achieved a certain probability level of having an effect. Often that level is 95% or more. Behavioral meaningfulness refers to the general size or applicability of that effect. The analysis of qualitative research proceeds in different ways, deals with different questions and provides different answers. Such analysis would be more interpretive than predictive and seeks to provide fuller insight.
An interesting hybrid of quantitative and qualitative research is known as cognitive response analysis. Often used in advertising and other studies, study subjects think aloud while engaging in an activity, including watching commercials. Often such data are coded in terms that can be quantified, such as product-related thoughts concerning the product shown, message-related thoughts concerning the ad itself, source-related thoughts concerning the source of the message and unrelated thoughts. These thoughts can be scaled in numbers and further analyzed statistically in relation to other variables.
- Final use or disposition: The most frequent application of quantitative and qualitative research has been to use qualitative research as an exploratory step leading to eventual quantitative study and confirmation. To study image perception change evoked by a commercial, for example, researchers might conduct focus groups to see what effects are taking place and then use those results to construct a survey to confirm those effects on a larger and more quantified scale. Qualitative research is sometimes used as the primary research mode. Its findings are then used to answer research questions and formulate ad strategy without any accompanying quantitative research.
Ultimately the relationship of various modes of research should be one of triangulation. Triangulation means that researchers use different types of research and then compare the results for convergence. Many large companies follow this philosophy to varying degrees, using many methods and researchers. If the results of triangulation are convergent, then it is likely that the research is pointing in the right direction; if the results are not convergent, that could reflect validity problems with the modes of research, or it could mean that deeper issues exist that have not been addressed.
Focus groups are the most commonly used type of qualitative research. In advertising, focus groups have been widely used at all stages of campaigns from creation to recall testing. Examples of the influence of focus groups on advertising include Hamilton Beach creating an ad campaign that stressed a more youthful appeal for its home appliances; the Lucent Technologies developing its name and logo; and the reformulation and glamorization of an industrial product, the insecticide Durban Pro, by DowElamco.
The main idea of a focus group is to get people, typically a group of six to 12, to talk in depth about a particular topic. The role of the moderator is to keep the conversation flowing, keep it on topic and try to involve all participants. The moderator must probe for details when necessary and ensure that the proper degree of depth is achieved. The sessions are usually recorded (audio or video), and often one-way mirrors are used so that the marketing team can observe the group and, in particular, watch the body language of the participants.
Incentives are usually given for focus group participation, and these vary depending on the participants. Focus groups involving doctors or professionals often require larger incentives, while everyday consumers may be paid as much as $50 to $100, depending on geographic locale and the amount of time spent.
Conducting focus groups is a managerial process that includes planning, execution, analysis and reporting to the client. Planning includes determining the type of research to conduct, what to ask, where to do it and who is to do the research. The development of a strong focus group interview guide or protocol in which questions are laid out in advance is a critical part of planning.
Execution depends on the moderator, who is responsible for what happens in the focus group. Often the moderator is an outside consultant who participates throughout most or all of the process and is also responsible for the analysis and interpretation of the groups.
When the process of conducting focus groups is finished, the analysis begins. Various forms of analysis can be used. Often a form of coding is employed. For example, statements may be coded in terms of agreement-disagreement among group participants. Thematic analysis, which develops the main points of a focus group, may also be used. Whatever analysis is applied, it is useful for more than one person to assess the results for triangulation purposes because two analysts often interpret the same texts differently. A final report is then written for the client in which the results are summarized and recommendations are made.
Recent developments in both quantitative and qualitative research have followed the trend toward more online activity. Focus groups are conducted online through the use of focus group software. So-called online, or wired, groups operate in a more anonymous environment than do face-to-face focus groups, and participants are thought to be freer to express their opinions. Each person's opinion appears to carry equal weight. Opinions are registered on the computer screen and people can see each other's opinions while they are writing their own.
The use of online focus groups likely will grow for several reasons: They are cost-efficient; they are flexible and allow marketers and advertisers to reach broad audiences both nationally and worldwide from one central location; follow-ups by e-mail are relatively easy; and new technological developments, such as live video and increased bandwidth, may reduce the alienation.