The Rasogulla-Idli-Laddoo-Vadapao Scale

The rasogulla-idli-laddoo-vadapao scale is PRISM’s adaptation of the standard semantic differential scales used in quantifying responses to questions on a survey among illiterate, less educated, rural respondents. It was developed in the course of an opinion poll to judge voting intention during assembly elections.

PRISM found that the proportion of “don’t know/ can’t say” responses was a high 40 - 50% in such polls. The reason for this was found to be the inability of large sections of the rural illiterate populace to respond to the traditionally administered scales asking a respondent to rate a candidate/ political party on a 5 or 10 point semantic differential scale. The concept of ‘rating’ or ‘assigning marks’ was alien and incomprehensible to the target audience, hence the high “can’t say” responses.

Researchers at PRISM thus adapted these rating scales to include concepts that were easier to relate to by a rural respondent, such as rasogulla in eastern parts of India and idli in the South (laddoo in the North and Vadapao in the West/ Mumbai). Respondents now had to say how many rasogullas/ idlis they would serve to a particular candidate/ political party, assuming the more they liked them the more they were to serve and vice versa.

It was noticed that using this scale immediately led to a decrease in the proportion of “don’t know/ can’t say” responses to a mere 10-12%, making election predictions much more accurate than before.