Perceptual Mapping, MDS & Cluster Analysis, student travel marketing analysis

Résumé du document

We analyze the aspects of perceptual mapping, multidimensional scaling and cluster analysis for a company called Student Travel. The company is a potential new entrant into the Irish student travel market, and plans to undertake research into the needs and wants of Irish students relating to weekend breaks. The question here is to find innovative means to advertise each destination to its maximum capacity. Finding out where students are most likely to go on weekend breaks and what the driving forces behind their choices are can discover this. We need to discover what attributes are associated by the students to the different cities of Europe, their reasons for travel or otherwise and their travel companions. The following figure illustrates the factors affecting travel destination choices.

Extraits

[...] Variation in dimensionality: It is not guaranteed that respondents will use the same dimensions when perceiving an object. It is believed that there are only a limited number of dimensions that most people use for perceiving an object, but one group of respondents may use one set of dimensions and another group may use a completely different set of dimensions. Variation in importance: For 2 respondents evaluating a car, horsepower may be an evaluative dimension for both of them. However, one respondent may attach great importance to horsepower while the other respondent considers it to be relatively unimportant. [...]

[...] (Back to index) Industry Review The Irish student travel market is huge. In the 4th Quarter of 2003 there were approximately 196,700 students in Third level education between the ages of 18 and 24 alone (www.cso.ie). The student travel industry is especially significant nowadays as students are seen as having a higher budget than in previous years. An increasing number of students are choosing to travel on summer breaks. Cities in the US are experiencing an influx of Irish students looking for work with and without visas. [...]

[...] However, more recently, market researchers have become increasingly sceptical about using MDS for market research purposes. There is some doubt about its utility. There are worries relating to the controllability / manipulability of the dimensions derived from using MDS, which is primarily an exploratory data analysis tool (Carroll & Green, 1997). Individual differences and interdependence are also important factors to be considered if using MDS for market research purposes (Hair et al, 1998). The dimensions from the output of the MDS must be clearly identified and interpreted by the researcher. [...]

[...] perceptual map is a spatial representation in which competing alternatives are plotted in a Euclidian space” (Lilien, G., Rangaswamy, A., pp96). It allows the marketer to view the product in the eyes of consumers, in relation to the competition within the marketplace. This enables the marketer to view both the customer and the competition at the same time and within the same map. A perceptual map has a number of characteristics: Pair wise distances between alternatives indicate the perceived similarity between any pair of products. [...]

[...] An assumption of Non-metric MDS is that the data is all measured in the ordinal level. This gives Non- metric MDS an advantage over Metric MDS because there are no assumptions made about the underlying transformation function (Steyvers, 2001). Classical MDS Classical Multidimensional Scaling is identifiable because there is only one similarity matrix involved when using it (Young, 1985). The aim in Classical MDS is to generate a map in as few dimensions as possible such that the distances between points on the map are close in value to the observed distances (Lilien et al, 1999). [...]