Conjoint Analysis: Choice Models

Download the PDF Version 

Background

  • Conjoint analysis is typically used to measure consumers’ preferences for different brands and brand attributes. Conjoint analysis revolves around one key idea; to understand the purchase decision best.
  • This methodology was developed in the early 1970’s. It has become one of the most widely used quantitative tools in marketing research.
  •  It is described in many published journal papers. Green & Rao (1971)1 first introduced the research concept and Batsell & Elmer (1990)2 discuss its application to pricing and demand forecasting.
  •  There are many different conjoint methods; adaptive conjoint analysis (ACA), full profile conjoint analysis (CVA) and choice based conjoint (CBC).

Description of How it Works:

  • Respondents in a market research interview are asked to make either choices or rankings of preference regarding hypothetical product profiles.
  • A group of products are presented to a respondent. The respondent is asked to select or rank the products.
  • This process is repeated several times with the levels of each product attribute (and sometimes price) varying in each scenario.
  • The respondent’s choice data derived from each scenario enables the analyst to decompose their preferences for the different profiles and individual attributes.

Strengths

  • Conjoint testing holds all extraneous real world factors constant. (i.e. advertising, new product entries, stock outs, promotions, distribution, etc.,)

Weaknesses

  • Conjoint testing holds all extraneous real world factors constant. (i.e. advertising , new product entries, stock outs, promotions, distribution, etc.,)

Source: 

1. Green, P. & Rao, V.R. (1971), Conjoint Measurement for Quantifying Judgmental Data. Journal of Marketing Research (August), p. 355.

2. Batsell, R.R. & Elmer, J.B. (1990), How to Use Market Based Pricing to Forecast Consumer Purchase Decisions. Journal of Pricing Management. (Spring), p.5.