Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

A field test for estimating sweetness preferences to improve estimates of sucrose intakes in individuals

Ellen Messer
Visiting Lecturer, International Food and Nutrition Planning Program, Massachusetts Institute of Technology, Cambridge, Massachusetss, USA

There is evidence generally for increasing consumption of refined sugars, with related health problems in developing countries (1-3), but existing methods of recording dietary intake fail to capture accurately the quantities of sugar consumed and preferences for sweetness. This paper describes a simple procedure for measuring sweetness that can be used, in combination with standard dietary intake records, to represent more accurately the sucrose intakes of individuals. Independently, the data can also be used to check sweetness levels and preferences within and across communities to describe geographic and socio-cultural patterns in sweet "tastes," dietary trends, and their possible nutritional/health implications.


Taste psycho-physicists have developed experimental procedures to measure individual perceptions and preferences for sweetness (4). One method is to have individuals taste an experimental liquid and then match it to another liquid of perceived equivalent sweetness. Such taste tests, in which sweetness perceptions and preferences of individuals are precisely recorded (5), are ordinarily carried out in laboratory settings in which the temperatures, other flavours, and textures of the substances are carefully controlled.

"Taste matching" standardized sucrose/water solutions to local, very weakly flavoured sugar-water beverages, such as weak teas, coffees, and herbal preparations that are widely consumed, particularly by the poor, offers an easy way to measure how sweet individuals (and societies) consume and prefer such beverages. Knowledge of the amounts of sucrose/water in beverages, multiplied by the quantities of the beverage used, provides an accurate estimate of sucrose ingested in such liquids. Such figures can also be acquired and checked by observing recipes used in beverage preparations, and total sucrose purchases over a standard period of time.

TABLE 1. Six Sugar/Water Solutions Used to Test Perception of Sweetness

Molar Concentration Sucrose per Litre

Sucrose per 500 ml

0.032 M 10.94 g 5.47 9 = 1tsp
0.056 M 19.15 g 9.58 9 = 1 3/4 tsp
0.1 M 34.2 g 17.1 9 = 1 Tbsp + 1/4 tsp
0.18 M 61.56 g 30.789 = 1/8 CUP + 11/4 tsp
0.32 M 109.44 g 54.72 9 = 1/4 CUP + 1/2 tsp
0.56 M 191.50 g 95.76 9 = 1/3 CUP + 1/4 tsp

Sucrose 1 mole = 342.0 g/litre.

These molar concentrations correspond to natural thresholds of sweetness taste perception, determined experimentally by Linda Bartoshuk and colleagues.


Six solutions can be prepared with locally available refined sugar and purified water (table 1). Water and sugar can be measured with standard volumetric equipment. Sugar is weighed and then measured and converted into the volumes indicated. Solutions should be made up and labelled in advance in sanitary plastic or glass containers that are easily transported from house to house. Individual tasting is efficiently carried out with disposable clear plastic cups, such as those used in urban areas for gelatines and other liquids or semi-liquids. They are inexpensive, sanitary, and provide uniform containers, so that participants will not be influenced by colour or other characteristics of the receptacle.


Taste-testing is carried out as follows. Each of the solutions is presented to a subject, one at a time, in irregular order. After tasting it, the subject says whether it is the preferred amount of sweetness, not sweet enough, or too sweet. Responses are recorded for each solution. Then the subject is allowed to taste a bit of the sugar-water that the household ordinarily prepares and consumes as part of its daily intake, and repeats the tasting of the various solutions. The responses are again recorded. Solutions are tasted until a consensus is reached on which solution "matches" the home beverage.

The procedure may be carried out with all household members to establish (a) whether all perceive the "match" equally, and (b) their consensus on how sweet they like their beverages. Each household member is also asked how much of the home beverage he/she consumes each day. Where possible, the volumes should be computed by measuring the vessels from which the beverages are consumed. All such intakes are summed.


In many developing countries, people drink quite sweet beverages. In one Mexican community, for example, the author found that the "norm" for coffee-flavoured sugar water for 16 households sampled was the 0.18 M solution, although six households ordinarily drank coffee sweeter than that, and the two households who drank less sweet coffee recognized that they drank less sweet than the "norm." Individuals drank in the range of 0.6 to 1.2 litres of such coffees per day, thereby taking in 37.74 grams of sugar just in beverage consumption. Youngsters two to five years old also consumed coffee from the family pot in the range of 1 litre per day.

While the data do not indicate that there is a connection between intensity of sweetness preferred and sucrose intakes, attention to both sweetness and quantities of sweet beverages consumed can begin to establish whether such linkages exist, as well as whether individuals are consuming a disproportionate percentage of total calories in the form of sucrose. The degree of sweetness, extent of sweet beverage consumption, and percentage of calories provided by sucrose are also factors that can, in the future, be checked to see how they relate to incidence of dental caries, diabetes, and atherosclerotic heart disease (6). Additional studies might include examination of sweetness levels in weaning foods as a first step in understanding what degree of sweetness makes them palatable, how sweetness preferences are formed, as well as possible later dietary and health consequences of such sweet preferences.


1. J. Yudkin, J. Edelman, and L. Hough, Sugar (Peter H. Wyden, New York, 1972).

2. J. Yudkin, "Patterns and Trends in Carbohydrate Consumption and Their Relation to Disease," Proc. Nutr. Soc., 23: 149 ( 19641.

3. J. Yudkin, "Sugar and Disease," Nature, 239: 19711972).

4. L. Bartoshuk, "The Psychophysics of Taste," Amer. J. Clin. Nutr., 31: 1068 (1978).

5. L.M. Bartoshuk, K. Rennert, J. Rodin, and J.C. Stevens, "Effects of Temperature on the Perceived Sweetness of Sucrose," Physiol. Behav., 28: 905 (1982).

6. E.L. Bierman, "Carbohydrates, Sucrose, and Human Disease," Amer. J. Clin. Nutr., 32 (12) (Supplement): 2712 (1979).

Contents - Previous - Next