Day 6 TEAM- A

Permap – A new dimension

Data can be deceiving, a hypothesis that has been proved correct by PERMAP. What is Permap? It is simple perceptual mapping software. The program, PERMAP, uses conventional metric multidimensional scaling techniques. That is, it uses pairwise numerical values (correlations, proximities, dissimilarities, etc.) to construct a map showing the relationship between objects. A unique feature of PERMAP is that it embeds the mapping techniques in an interactive, graphical system that minimizes several difficulties associated with multidimensional scaling practices.

A major advantage of perceptual maps is that they deal with problems associated with substantiating and communicating results based on data involving more than two dimensions.

Permap usage can be of two types based on data and variable we use for e.g.:

1.) Overall percep mapping

2.) Attribute based percep mapping

Difference between Overall and Attribute is that in case of overall we take the variables and plot on Percep Map based on no such specific attribute but on general perception. For eg: Which carbonated drink you like? Your answers would be plotted on percep map but nothing specific can be drawn from percep map, since no such attribute is given preference. Whereas if we ask you to rate your carbonated drink based on some specific attributes then the result of percep map would be quite different. Literally it can be said all about the way to perceive!!

So how do you enter data?

Dissimilarity information can be entered as either dissimilarity or similarity values and can be based on any linear scale. They can be entered as half matrices (lower triangle) or whole symmetrical matrices. The whole matrix option is useful when one is downloading information from another program. It contains a significant amount of redundant information that one would not want to enter manually, but that need not be eliminated if the mechanics of data transfer cause it to be present. PERMAP deduces from the data context which kind of data is being entered (similarities or dissimilarities, values shifted or not shifted from zero, values expanded or contracted by a constant multiplier) and whether or not a full matrix is being entered. If PERMAP cannot make a safe deduction concerning the nature of the input data, it asks for more information.

After a perceptual map has been constructed, the hard part starts. Additional progress is dependent on the ability to forma clear, unequivocal definition of the nature of the object groupings. The identification task is often difficult and demands creativity and a deep understanding of the subject matter.

by 

Abhik Chakraborty