Elsevier

Food Quality and Preference

Volume 15, Issue 6, September 2004, Pages 541-548
Food Quality and Preference

Consistency of sensory testing with 4- and 5-year-old children

https://doi.org/10.1016/j.foodqual.2003.11.006Get rights and content

Abstract

Rank-order and paired comparison tests are widely used methods to assess sensory perception of young children. Small age differences could, however, influence the ability of children to carry out such tasks. This study compared rank-order and paired comparison tests for consistency in 4- and 5-year-old children. During four sessions, 22 young adults, 21 4-year-old and 47 5-year-old children carried out rank-order and paired comparison tests to measure discriminatory ability (0.22 M, 0.25 M, 0.29 M, 0.34 M, 0.39 M sucrose in orangeade) and preference (0.14 M, 0.20 M, 0.29 M, 0.42 M, 0.61 M sucrose in orangeade). Young adults and 5-year-old children were able to discriminate between all solutions and showed a high consistency between the rank-order and pair-wise tests for discriminatory ability (>76% consistency) and preference (>71% consistency). In contrast, 4-year-olds detected differences in sweetness during the preference tests, but failed to distinguish sweetness intensities during the discriminatory ability tests. It is concluded that the dissimilarity between 4-and 5-year-olds in performing sensory tests was due to a difference in their cognitive skills rather than their sensory perceptual differences.

Introduction

Children ranging in age from 0–10 years prefer higher concentrations of sucrose in foods than adults (De Graaf & Zandstra, 1999; Desor & Beauchamp, 1987; Desor, Greene, & Maller, 1975; Enns, Van Itallie, & Grinker, 1979; Zandstra & De Graaf, 1998). Furthermore, it has been suggested that children are less sensitive to sweet taste than adults (De Graaf & Zandstra, 1999; James, Laing, Oram, & Hutchinson, 1999; Kimmel, Sigman-Grant, & Guinard, 1994; Zandstra & De Graaf, 1998). In order to investigate what children prefer, and which levels of sweet they are able to discriminate, reliable sensory tests are needed. Tests that are reliable for adults do not always give reliable information when used with young children. In general, children tend to answer in the affirmative way, and have less experience with scales than most adults (Guinard, 2001).

Rank-order and paired comparison tests are the two most widely used methods to measure sensory perception and preferences in young children (Birch, 1979, Birch, 1980a, Birch, 1980b; Birch & Marlin, 1982; Guthrie, Rapoport, & Wardle, 2000; Kroll, 1990; Leon, Marcuz, Couronne, & Koster, 1999; Liem & Mennella, 2002; Liem & Mennella, 2003; Olson & Gemmill, 1981; Sullivan & Birch, 1990). Most sensory studies reported in the existing literature, grouped 4- and 5-year-old children together (Birch, 1980b; Birch & Marlin, 1982; Kimmel et al., 1994; Olson & Gemmill, 1981; Sullivan & Birch, 1990). However, children in this age range are rapidly changing. Their attention spans lengthen and their ability to categorize, analyze, and synthesize information rapidly becomes stronger and more efficient (Krug & Mikus, 1999). In the Netherlands, children normally enter kindergarten at the age of four, therefore it is important to asses differences between 4- and 5-year-olds.

Sensory taste tests with young children typically consist of two types: preference tests and discrimination tests. Most studies in children focus on preference rather than discrimination. Stimuli that are commonly used are not only different in taste, but are also different in color and appearance (e.g. different fruits, different sandwiches). It has been suggested that with these stimuli, rank-order and paired comparison tests give consistent data in young children (Leon et al., 1999).

However, it remains unknown whether rank-order and paired comparison tests give consistent data when the presented stimuli differ only in taste (e.g. orangeade with different concentrations of sucrose). It is also unknown if rank-order and paired comparison tests give consistent data when they are used to measure discriminatory ability in young children (i.e. which stimulus is more sweet) instead of preference.

The present study investigated the consistency between rank-order and paired comparison tests as measures of discriminatory ability and preference, using stimuli that differed only in sweetness. Subjects of the study were young adults, 4-year-olds and 5-year-olds.

Section snippets

Subjects

Twenty-one 4-year-old children (4.4 ± 0.3 years, 12 girls and 9 boys), 47 5-year-old children (5.4 ± 0.3 years, 23 girls and 24 boys) and 22 young adults (22.6 ± 1.8 years, 17 females and 5 males) participated in the study (see Table 1). The children attended a kindergarten in Bennekom, a village 4 km from Wageningen. Since the research study took place in October, and children in The Netherlands start kindergarten in September, the 5-years-olds (second graders) had almost one year more experience

Discriminatory ability, children and young adults

Fig. 1 shows the results of the rank-order and paired comparison tests for discriminatory ability. Children, compared to young adults, were less able to discriminate between the five different sugar concentrations. This was evident during the rank-order tests (t(88df)=−5.2; p<0.0001) and during the paired comparison tests (t(88df)=−6.2; p<0.001) (see also Table 2). No significant differences were observed between girls and boys during the rank-order tests (t(66df)=−0.48; p=0.63), or during the

Discussion

The present study suggests that 5-year-old children were able to carry out rank-order and paired comparison tests for discriminatory ability in a consistent manner. However, 5-year-olds were less able to discriminate between different sugar concentrations in an orange drink than adults. Four-year-olds failed to carry out both discriminatory ability tests in a consistent way. This was in contrast with the tests for preference. In these tests, young adults and both 4-and-5-year-old children,

References (33)

  • F Leon et al.

    Measuring food liking in children: A comparison of non-verbal methods

    Food Quality and Preference

    (1999)
  • B Turnbull et al.

    Taste sensitivity to 6-n-propylthiouracil predicts acceptance of bitter-tasting spinach in 3-6-y-old children

    American Journal of Clinical Nutrition

    (2002)
  • M Vazquez et al.

    Flavor preferences in malnourished Mexican infants

    Physiology and Behavior

    (1982)
  • E.H Zandstra et al.

    Sensory perception and pleasantness of orange beverages from childhood to old age

    Food Quality and Preference

    (1998)
  • G.K Beauchamp et al.

    Development of sweet taste

  • L.L Birch

    Effects of peer models' food choices and eating behaviors on preschoolers' food preferences

    Child Development

    (1980)
  • Cited by (0)

    View full text