Mary F. Schraidt is a vice president at Peryam & Kroll Research Corporation. She has 19 years of experience in marketing and sen- sory research, and is an active member of several professional organizations. She has a particular expertise in consumer communications and testing with children.

Sensory testing with children Is it possible to get reliable information from children? And if so, are there special methods and protocols required? Mary F. Schraidt, Peryam & Kroll Research Corporation

Those who produce food products meant for children generally recognize that children have a say in what products they use, that there is a need for data to facilitate decisions about this important market segment and that even young children show preferences. The challenge is to reliably predict those preferences.

Testing with children has been largely based on intuition and folklore. Most approaches have at least a little success, but the literature is scant, and there are questions about the relative effectiveness of procedures. However, a consensus does exist that:

• Children can discriminate, especially in degree of liking.
• Children are able to show degree of preference.
• Children can provide useful information about products.
• Children require special handling and different procedures from those routinely employed with adults.
• It is important to gain children's confidence and communicate in a language they understand.

 

With well-written tests,
children are good respondents

 

Peryam & Kroll Research Corporation has a specialty practice in testing with children, testing several hundred children every month at three U.S. locations. The company has proven protocols in this area of research, which protocols are, themselves, the subject of continuous research.

One Peryam & Kroll study dealt primarily with the issue of scale. Based on preliminary work, three scales were selected:

• The standard, well-known hedonic scale, with the usual verbal categories, which was developed by sensory pioneer Dr. David Peryam
• A scale devised by Peryam & Kroll (P&K Scale), in which category labels are expressed in language judged more suitable to children -- from 'super good' to 'super bad'
• A picture scale in which categories are represented by a series of -- simple faces with expressions ranging from happy to sad

While initial work included children up to age 13, to get at the crucial problem two age groups, based on assumptions of verbal ability, were defined: pre-literate (with no reading skills, aged five to seven) and semi-literate (able to read but limited vocabulary, aged eight to ten). Most experiments employed a straightforward approach of reading successive categories one after another, always starting at the 'good' end. However, researchers also tried a bifurcated approach, where the child respondent first put the sample into a good/like or bad/dislike group, and then was presented with successive phrases to scale the degree of like or dislike. Another side issue was comparing one-on-one interviews with each child individually to self-administered test with instructions and the use of a printed form (eight-to-ten group only).

All the testing followed the best scientific standards and methods: the test product was validated, paired samples were tried twice, scales were used equally often and appeared equally often in first and second positions, the two test products appeared equally often in first and second positions and there was a balance of boys and girls in all age ranges.

The results were as follows:

• The P&K scale performed better than the hedonic scale or face scale.
• Reducing scale length from nine points to seven points offers no advantage.
• Five-to-seven-year-olds don't perform any better with the face scale than with verbal scales.
• The bifurcated approach does not discriminate as well as the straightforward method.
• Older children perform as well using written questionnaires as when interviewed one-on-one.

The most important conclusion? Handled properly, with well-ritten tests, children are good respondents. Perhaps because they take tests regularly in school, they know what is expected of them and are at least as reliable as adult respondents.