An example comes from scales used to assess the price sensitivity of different demographic groups. You could be asking if regular customers are willing to pay more than infrequent customers or if those who live downtown consider a $10 price point to be expensive, while those who live uptown don’t.
Shopping frequency and residence area are demographic variables. A general rule of thumb on consumer surveys is to put the demographic questions toward the end, since respondents become cautious in answering as soon as potentially identifying information is requested. However, because asking about prices is even more sensitive, you’d probably ask about the demographics first.
In asking about shopping frequency, you could, for instance, have a scale with intervals “1-3 times per month; 4-6 times per month” and so on or have a scale with “1-6 times per month; 7-12 times per month” and so on.
For the price sensitivity items, you might adapt the four questions pioneered by Dutch behavioral economist Peter van Westendorp and used by a great many retail pricing specialists. Your adaptations might be, each beginning with the words “At what price would you consider this product or service to be…”
…inexpensive enough that you’d think hard before deciding not to purchase it?Again, the response scale for each item could have a fewer number of broad categories or a greater number of narrow categories.
…expensive enough that you’d think hard before deciding whether to purchase it?
…so inexpensive that you’d think there might very well be something wrong with the item?
…so expensive that you’d immediately reject the idea of buying the item?
Research at New York University, University of Southern California, London School of Economics and Political Science, and survey firm Vision Critical indicates that you’ll obtain more useful information from the pricing items if you make the demographic item response categories broad rather than narrow.
When a retailer or consulting firms is designing the scale intervals, there’s a natural tendency to make them narrow. The thought is that the intervals can always be combined later during data analysis, but you can’t afterwards divide up a broad category. The research evidence argues for questioning that natural tendency.
Click below for more:
Answer Van Westendorp Pricing Questions
Decide on Your Segmentation Objectives Early