Perceptual Map using binary outcomes

I have 10 classes of drugs and within each there are a series of characteristics, e.g., color, aroma, strength and then for each the respondent notes if wanted or not, a binary response and need to display a response map by drug thus 10 maps. The data received looks like this for each drug (Yup! That's what it looks like and I had no control as to design):

Color Aroma Strength ..............
Yes No No
No No No
Yes Yes Yes


Would like to map the results Characteristic by Want (with focus on the Yes responses)

Any thoughts are appreciated


TS Contributor
Since you are referring to 'perceptual map', I understand you are looking for a dimensionality reduction method.
Let's assume you have a cross-tabulation, with the respondents listed in rows, and each drug characteristic is listed in columns. Let's assume further that for each respondent you have his/her preference as to each characteristic (coded as yes/no).

If your goal (it is not so clear from your description) is to understand which respondent is more similar to which on the basis of the preferred characteristics (and, by the same token, which characteristic goes hand in hand with which characteristic), you may want to use Correspondence Analysis to get a perceptual map. The only modification that you have to do is to recode 'yes' with 1, 'no' with 0.

There are many facilities to perform CA. It is routinely provided in commercial stat program (SPSS, STATISTICA, JMP, Minitab, XLSTAT), or in free program (R, PAST).

If you want to give R a go, there is a package (FactoMineR) which is very handy, and there are instructional video on Youtube.

I have built a website focused on Correspondence Analysis. It is tailored for archaeologists, but the underlaying principles may be applied to any research field (

Hope this helps.