Conjoint Analysis: Methods, Applications, and Business Use

Introduction to Conjoint Analysis

Like qualitative researchers, quantitative researchers have a number of approaches available to them today. The selection of a research approach depends on what type of information is being sought, what type of information is available, and the goals of the researcher. One research methodology that is gaining increasing popularity is conjoint analysis, a quantitative methodology that is discussed in detail below, followed by a summary of the research and important findings in the conclusion.

Description and Business Applications

Conjoint analysis is a quantitative methodology that measures the perceived values of various possible product designs (Calantone & Di Benedetto, 1990). Respondents participating in conjoint analyses view several variations of product concepts and then assign ranks according to their individual preferences (Calantone & Di Benedetto, 1990). The analysis of these responses can be used to identify the respective utility associated with each of the attributes ranked by respondents, and these findings can then be used to identify the combination of attributes that consumers would most prefer (Calantone & Di Benedetto, 1990).

According to Orme (2009), "Conjoint analysis has become one of the most widely-used quantitative methods in marketing research. When used properly, it provides reliable and useful results" (p. 1). Although there are a number of different conjoint analytical methods that employ different pairings of example products, the general goal is to "measure the perceived values of specific product features, to learn how demand for a particular product or service is related to price, and to forecast what the likely acceptance of a product would be if brought to market" (What is conjoint analysis?, 2011, para. 1).

Types of Management Problems Addressed

A number of researchers have used conjoint analysis to examine different aspects of management problems. Wind, Grashof, and Goldhar (1978) applied conjoint analysis to scientific and technical information services, while Pekelman and Sen (1979) used the method to improve the predictability of product acceptance and concept choice. In addition, conjoint analysis appears to have the potential to play an increasingly useful role in the identification of consumer preferences with respect to environmental issues (Bateman & Willis, 1999).

Detailed Methodology and Protocol

According to Bateman and Willis (1999), "Conjoint analysis is based on judgments on a carefully constructed set of hypothetical cases that allow for capture of the 'partworths' of each cue-utility at each of its levels. So linearity of cue use is not assumed, but simplification of cues into different levels is necessary" (p. 346). A sample protocol for developing a conjoint analysis is outlined in the following four steps, adapted from Mora (2011).

Step 1: Identify Key Attributes and Levels

This is probably the most important step in the design of a conjoint analysis study, since the selection of appropriate attributes directly affects the ability to reflect how buyers make purchase decisions. Qualitative research or other research sources should be used to provide realistic attributes. The presentation of attribute levels is not restricted to text—images may be used when appropriate. According to Orme (as cited in Mora, 2011), attributes or factors used in this step should:

1. Cover the full range of possibilities for existing products; 2. Be independent from each other with no overlapping meaning; 3. Be mutually exclusive; and 4. Have a balanced number of levels across attributes whenever possible.

Step 2: Create an Experimental Design

This step should provide: (1) frequency balance, meaning each attribute appears the same number of times; (2) orthogonality, meaning each item is paired with other items the same number of times; and (3) position balance, meaning each item appears the same number of times in each position.

The experimental design is then used to generate a set of choice tasks for respondents. Using this approach—for example, in the case of online market research courses—would not only provide a more realistic scenario for respondents but would also prevent them from focusing solely on price, decreasing the natural tendency to provide low estimates when price becomes the center of attention, as commonly occurs in price research approaches using direct questions about willingness to pay and purchase intent.

Step 3: Estimate Utility Coefficients

Utility coefficients are measures of desirability for each of the attribute levels. They can be estimated using different methods, including aggregated multinomial logit, latent class analysis, or Hierarchical Bayes estimation, the last of which is the most common approach today.

Step 4: Develop a Market Simulator

This step uses the utility coefficients to predict which product configurations are more likely to be chosen among many possible configurations, including those not presented to respondents. This is the most valuable tool for managers, as it allows them to conduct "what-if" analyses to predict shares of preferences for different product configurations.

Conclusion

The research showed that quantitative researchers have a number of methods available to them, the selection of which depends on a wide range of factors, the most salient of which is the type of question that is involved. Conjoint analysis is typically used to evaluate product preferences, but the approach has been applied to a number of other management settings in recent years as well, including its use as a decision-making tool in formulating improved patient care practices that promote patient satisfaction.

References

Bateman, I. J. & Willis, K. G. (1999). Valuing environmental preferences: Theory and practice of the contingent valuation method in the U.S., EU, and developing countries. New York.

Bowling, A. & Ebrahim, S. (2005). Handbook of health research methods: Investigation, measurement and analysis. Maidenhead, England: Open University Press.

Calantone, R. J. & Benedetto, C. A. (1990). Successful industrial product innovation: An integrative literature review. New York: Greenwood Press.

Mora, M. (2011). Conjoint analysis and realism in price research. Relevant Insights. Retrieved from http://relevantinsights.com/conjoint-analysis.

Orme, B. (2009). Which conjoint analysis method should I use? Sawtooth Software Research Paper Series. Retrieved from http://www.sawtoothsoftware.com/download/techpap/whichmth.pdf.

Pekelman, D. & Sen, S. (1979). Improving prediction in conjoint measurement. Journal of Marketing Research, 16, 211–220.

What is conjoint analysis. (2012). Sawtooth Software. Retrieved from

Wind, Y., Grashof, J. F. & Goldhar, J. D. (1978). Market-based guidelines for design of industrial products: A new application of conjoint analysis to scientific and technical information (STI) services. Journal of Marketing, 42(3), 27–37.