Projective Mapping

 
Many techniques exist to explore differences among a set of products: paired comparisons, similarity scaling, descriptive analysis and sorting tasks, to name a few. However some of these techniques have major drawbacks. Performing paired comparisons on a large set of products can be very time consuming: a set of 10 products would require each panelist to perform 45 paired comparisons! Sorting tasks, while quick and easy to conduct, do not yield much information on the degree of difference between products.
 
Projective mapping serves as a simple and quick technique to obtain product inter-distances. It produces better product differentiation than sorting1 and is faster than similarity scaling2. Additionally, product maps obtained from projective mapping have been demonstrated to have a high degree of similarity to maps obtained by descriptive analysis.2 It has been suggested that this technique can be applied to product sets with as many as 12 samples3. The speed with which projective mapping can be conducted is due to the fact that training is not required. The task is simple enough that it can be performed by consumers. However, a major drawback of projective mapping is that the task may become difficult or yield poor results if the products are too similar.

Methodology

 
To perform projective mapping, each panelist is presented with a large sheet of paper, usually 60cm x 60cm, however any size can be used as long as it provides ample space for panelist to fully separate products. Each panelist is also provided with a full set of samples and given instructions on how to create a map. An example of the instructions provided to panelists would be:
 
    Please taste all the samples in front of you and arrange them on the paper in such a way that similar samples are located near one another and different samples are placed far apart. You are free to evaluate the samples according to any criteria that you choose, and you do not need to specify your criteria. Feel free to use as much of the paper as is necessary to express the differences you may perceive. When you are finished, please mark the location of each sample with the corresponding number.
 

Analysis

 
Additionally panelists can be instructed to write a few descriptors which are characteristics of each product or groups of products. Integration of descriptive data into the analysis lends interpretation to the product map generated. Data from the maps is then collected as the x- and y-coordinated of each product. The entire set of product locations can be analyzed by Multi Factor Analysis (MFA), General Procrustes Analysis or the MDS-INDSCAL model (if the X and Y coordinates are converted to distances).
Video Demonstration
 
Click here for a video demonstration of the Projective Mapping method “from the consumer’s perspective.” In this scenario, the consumer is tasting citrus juices one at a time and placing them on the surface. Notice how the consumer’s dimensions get adjusted and evolve as the product space becomes more defined. The consumer is then asked to write down attributes that relate to each sample (which can be later correlated to the group product space with multiple regression).

Notes

 
1 King, M. C., M. A. Cliff and J. W. Hall (1998). Comparison of projective mapping and sorting data collection and multivariate methodologies for identification of similarity-of-use snack bars. Journal of Sensory Studies, 13: 347-358.
 
2 Risvik, E., J. A. McEwan and M. Rodbotten (1997). Evaluation of sensory profiling and projective mapping data. Food Quality and Preference, 8: 63-71.
 
3 Pages, J. (2005). Collection and analysis of perceived product inter-distances using multiple factor analysis: Application to the study of 10 wines from the Loire Valley. Food Quality and Preference, 16: 642-649.