摘要：Retail food demand studies are becoming increasingly concerned with the role of nutrition and health, yet consumer perceptions and attitudes are often ignored. The purpose of this pilot study is to determine consumer perceptions involving nutrition levels for selected foods. The influence of demographics and information about nutrition and health on perceptions toward meat items are determined. Results generally indicate that consumer perceptions toward fat and cholesterol levels in meats are based on the comparison of animal sources, not the comparison of individual cuts or preparation techniques. Recent efforts in the study of retail food demand have moved toward the role of nutrition and health. Several attempts have been made to measure the role that nutrition plays in food value or purchase habits (LaFrance (1983), Huffman (1988), Brown and Shrader (1990)). These studies use actual nutritional content of foods consumed to estimate demand impacts. However, it is possible that consumers perceive the nutritional elements of certain foods to be significantly different than actual levels. Differences between actual and perceived levels represent measurement error in these variables. Such errors may adversely affect the results of our demand studies. Consumer misperceptions may be an especially important issue when a utility maximization model such as Lancaster's Consumer Goods Characteristics Model (CGCM) is used. In such a model, the utility function arguments are the characteristics of the goods not the goods themselves. If consumers misperceive the nutritional value of food products, such models should include the perceived levels of nutrition, not the actual levels. The CGCM has been used extensively in recent years. In particular, CGCM was used by Ladd and Suvannant (1976) to test if food prices were a sum of the values of certain nutrients; by Adrian and Daniels (1976) to estimate nutrient demand based in part on demographic variables; by Morgan, Metzen, and Johnson (1979) to estimate hedonic prices for breakfast cereal characteristics; and by Terry, Brooker, and Eastwood (1986) to estimate the demand for nutrients. Each of these studies used actual nutrition levels. If, however, perceived nutrition levels are different than the actual levels, the results and conclusions may be affected. Results from these models vary widely. In the case of some nutrients, the implicit values can switch from significantly positive to significantly negative across models. Some of the variability may be associated with specification and differences in time periods. However, some variation may result from differences in perceptions which also change over time. Models which do not directly specify nutrient levels may fall prey to another problem. Work by Brown and Schrader (1990) and later by Capps and Schmitz (1990) utilize an index of nutritional awareness. Models of this nature allow for consumer perceptions to be included. However, when results of these models are reviewed, the results are compared to actual data, not perceptions. These results may be compared to the wrong benchmarks. Thus perceptions need to be considered, regardless of the approach used.
