Liking and disliking make more sense if you treat them as two separate scales, rather than as opposite ends of the same scale. This can give powerful new insights for design, evaluation and marketing.
To find out how much respondents like and dislike a product, you could ask them to record their response on Likert-style scales, but you’d need to use one scale for how much the product is liked, and another for how much it’s disliked.
The image below shows six respondents’ reactions to three products A, B and C.
This representation makes it easier to see different patterns of response, and to see what needs to be done in reaction to each pattern.
Product A is loved by half the respondents and hated by the other half. It should be marketed as a ‘Marmite’ product.
Product B produces strong mixed responses in everyone. It needs to have the disliked features identified and removed.
Product C is neither particularly liked nor disliked by anyone. It needs to have positive features added.
Other methods that fit well with this approach
Visual analogue scales to get fine-grained measurements.
Think-aloud and laddering to find what is liked and disliked.
Iterative non-functional prototyping to manage the process.