Question

Experimental research is not used very often in marketing research. Why is that? Why are most marketing research studies descriptive instead of experimental? Briefly explain how a new soda might try to create a study using an experimental design to determine if consumers like the taste of its new cola product. What challenges would the soda company face in creating that design? Could the company use a quasi-experimental design instead? If so, which design might it use and why? Why would it be much easier for the firm to create a descriptive study? If the company does a descriptive study, why does it need to be careful not to suggest a cause-and-effect relationship when reporting its

Answer 1

Experimental researches are aimed at establishing a cause and effect relationship between variables. The two main variables of any experimental research are independent and dependent. The Independent Variable (IV) is manipulated by the researcher and its effect on the Dependent Variable (DV). The researcher engages in two types of such manipulation- instructional or situational. This means that either the respondents are put in experimental situations and their responses are assessed or they are instructed to behave in a particular manner and then their behavior is manipulated. Field experiments or Lab experiments may both be conducted to see the cause-effect relationship between the IV and DV. In market research, if the company wants to launch a new product, then they have no choice but to use a descriptive research. They may conduct an experiment but that would require a lot of planning as the experimental research entails the maximization of the experimental variables, minimization of the error variance and controlling the effects of extraneous variables. Extraneous variables are difficult to identify in field experiments. In such a scenario, controlling them is difficult.

If the soda company wants to launch a new soda, in that case they may use experimental design to figure out the comparisons between various flavors. They need to randomly assign participants in various tasting situations and compare their results or choices for a particular flavor using ANOVA. Experimental design are at times not feasible as they require a lot of time, effort and money is required to conduct them.

I don’t think that quasi-experimental design will suit the purpose. Quasi-experiments are those experiments that cannot classified as true experiments.

In such a scenario, with little efforts and resources, the company may conduct a survey of large number of respondents and analyze the data to assess the most preferred flavor and act accordingly.

In the given case, descriptive research is better than experimental research.