,,\V tf \ c 'o Division of Agricultural Sciences \l UNIVERSITY OF CALIFORNIA *'.ns%* CONSUMPTION PATTERNS 'OR POTATOES, RICE, AND MACARONI IN NORTHERN CALIFORNIA, 1968 0JT~<2P# ^OO <^o °0 w *b @> C® (? JERRY FOYTIK ANN HERTZENDORF CALIFORNIA AGRICULTURAL EXPERIMENT STATION BULLETIN 848 CONTENTS Page INTRODUCTION 3 Scope of the Study 3 Some Assumptions 5 Consumption Trends 6 MAJOR FINDINGS 9 POTATOES: CALIFORNIA CONSUMPTION PATTERNS 10 Misinformation and Calorie Counters 10 Selection of Potatoes 11 Serving Potatoes 13 Substitutes for Potatoes 15 FREQUENCY OF USE OF POTATOES, RICE, AND MACARONI 16 USE OF CONVENIENCE FOODS 19 Built-in Conveniences 22 General Use Patterns 22 Processed Potato Products 23 Convenience Rice Products 25 Convenience Macaroni Products 26 APPENDIX A: DEFINITION OF TERMS .... 27 APPENDIX B: THE CONSUMER SURVEY .... 28 Sample Design 28 Data Collection and Analysis 32 The Sample Drawn 32 Copy of Questionnaire 33 APPENDIX C: STATISTICAL SUPPLEMENT . . 37 ACKNOWLEDGMENTS 38 LITERATURE CITED 38 This bulletin describes survey findings concerning the prevailing use patterns for potatoes, rice, and macaroni products in Northern California during 1968, especially (1) the frequency of serving these foods in the home, (2) consumer buying habits and preferences regarding their use, and (3) consumer attitudes toward them. Attention is directed toward convenience products prepared from these three foods. THE AUTHORS: Jerry Foytik is Professor of Agricultural Economics and Agricultural Economist in the Experiment Station and on the Giannini Foundation, University of California, Davis. Ann Hertzendorf was, at the time of this study, Lecturer in Consumer Economics and Assistant Specialist in the Experiment Station and on the Giannini Foundation, University of California, Davis. JANUARY, 1971 [2] * CONSUMPTION PATTERNS FOR POTATOES, RICE, AND MACARONI IN NORTHERN CALIFORNIA, 1968 1 INTRODUCTION Food consumption patterns are not static. The types of foods people use and their method of preparation constantly change. A major change in the United States dur- ing recent years has been the acceptance and use of ready-to-cook and precooked foods, thereby starting some basic adjust- ments in the structure of the food industry and in the kinds of food products and services offered consumers. Consumption trends for basic foods have been studied to gain insight into and permit prediction of food-use patterns. Various theories are offered to explain in- creases or decreases in the use of certain foods. Such explanations are based on his- torical consumption patterns and on how supply and demand for specific foods is affected by various socioeconomic factors such as income level and distribution, prices and supply of competitive products, quality factors, cultural and religious food preferences, and degree of urbanization. For our market supply system for food to function effectively, all sectors of the food industry must recognize and even anticipate any changes in food-use pat- terns. To improve the operation of our market supply system, we need to under- stand the changes that occur in use pat- terns and the factors producing them. Potatoes, rice, and macaroni products have previously been found to be staple items in diet patterns generally prevailing in the United States. They are familiar to most consumers and are now available in many different forms. Consumers' will- ingness to use these familiar foods in processed and convenience forms may in- dicate their attitudes toward trying and 1 Submitted for publication June 8, 1970. 2 In this study, potato refers to the so-called "Irish" or white potato, excluding the sweet potato and yam; rice denotes the milled or polished grain from the rice plant; and macaroni products include all paste foods commonly known as macaroni, spaghetti, noodles, etc. [3] adopting other new or processed products, hundreds of which appear in food stores each year. Knowledge concerning the vari- ables that appear to influence consumers' purchasing decisions for the foods under study should be helpful to all sectors of the food industry. Scope of the study This report describes our findings about the prevailing purchase and use patterns for potatoes, rice, and macaroni products in Northern California during 1968. Spe- cifically, attention is focused on (1) the frequency of serving potatoes, rice, and macaroni products 2 in the home, (2) con- sumer buying habits and preferences re- garding the use of these foods, and (3) con- sumer attitudes toward them. A consider- able part of the study deals with con- venience products made from these three foods. The information presented is based upon material collected by interviews made at 1,192 California households. Our survey was limited to the nonrural area of the State north of Tehachapi Moun- tains and west of the Sierra Nevada foot- hills. Figure 1 shows the geographic loca- tion of the survey cities. The completed surveys were grouped for analysis into three geographic areas representing differ- ent urbanization segments of Northern California. These included 421 (35.3 per cent) in the five Bay Area counties, 341 (28.6 per cent) in four large inland cities (Sacramento, Stockton, Fresno, and Bakers- field) , and 430 (36.1 per cent) in 37 small cities. o INTERIOR SMALL CITIES bo Yreka Paradise Manteca Weed Oroville Tracy Dunsmuir Yuba City Modesto Redding Marysville Ceres Anderson Olivehurst Turlock Red Bluff Woodland Madera Corning Davis Hanford Willows Vacaville Delano Chico Fairfield Wasco COASTAL SMALL CITIES Areata Petaluma Eureka Napa Cloverdale Paso Robles Healdsburg Atascadero Santa Rosa San Luis Obispo LEGEND 4ft Bay Area Cities (San Francisco, East Bay, and San Jose area) ^j Inland Cities (Sacramento, Stockton, Fresno, and Bakersfield) O Small Cities— 37 (Listed above at right) Fig. 1. Location ot Survey Cities. [4] A sample of households was drawn ran- domly within each survey city. These were interviewed during October-November 1968. To provide a basis for discovering differences in use patterns among con- sumer groups, the questionnaire inquired about the household 3 as well as about the household's use of potatoes, rice, and macaroni products. Some assumptions Potatoes, rice, and macaroni have rela- tively high rates of per-capita consumption in different parts of the world and in dif- ferent cultures. Historically, rice is con- sumed in large quantities by Oriental peo- ple, macaroni products by people of Ital- ian origin, and potatoes by other Euro- peans. Food-use patterns also differ among various cultural and ethnic groups in the United States. New immigrants to the United States may adopt the existing gen- eral diet pattern or they may cling to their traditional ways for some time, even for generations. For example, macaroni con- sumption is highest in the Northeastern United States where many people of Ital- ian descent are located (U. S. Department of Agriculture, 1965«). Gray, et al. (1954) hypothesized that we must consider the influence of country of origin of our immigrants and their food- use patterns when we study changes in potato consumption. Hence, they would argue that the decrease of 50 per cent in the per-capita consumption of potatoes during the past sixty years was related to a curtailment in the number of immi- grants coming from Germany, Ireland, and the Slavic countries where potatoes had been their main food. They also believed that the carry-over of the tradition of high potato consumption among recent immi- grants and the descendants of earlier im- migrants has kept potato consumption from declining even further. M. K. Bennett (1941) proposed a potato- cereal ratio as one measure of economic development .The ratio described an evo- lutionary process wherein a nation that survives almost entirely on potatoes has a 3 Households were described in terms of location, ethnic background, education, income, family size, age, occupation, and employment status. The meanings of these terms adopted for this survey are given in Appendix A. Appendix B describes the consumer survey. The detailed data gathered appear in the tables listed in Appendix C. [5] low level of development; as economic growth occurs, people will consume more wheat products and less potato products; with further economic growth, wheat or cereal products will become the mainstay of the diet and per-capita consumption of potatoes will decline still further; and in the most advanced stage of economic growth, both wheat products and potatoes are included in a varied diet but in smaller amounts. On this basis, the United States would be classified in the most advanced stage of economic development. Both potatoes and cereals are integral parts of the diet, but neither is its primary basis. This fact, coupled with the diversified cultural back- ground of our people, might account, in part, for the varied patterns in the use of these foods. Furthermore, if we accept this explanation, then we can expect a greater use of both processed potatoes and grain products, since these processed foods may add variety and interest to our diets rather than merely serving as an inexpensive source of carbohydrates. Food growers and processors try to in- crease the consumption of their products. One way of achieving this aim is by in- ducing low-consumption families to use foods they traditionally do not use by pre- paring such foods in different forms. Many consumers are willing to try new foods and may incorporate them into their diets if the foods are attractively presented, con- venient to use, and fit into existing con- sumption patterns. The receptivity of con- sumers to new food products usually lessens the risk of food retailers who are willing to stock new food products as one method of competing for customers and sales. Processed potato, rice, and macaroni products are among the hundreds of new products offered consumers in an attempt to change traditional food-use patterns, to increase sales to present users, and to in- duce nonusers to buy the new processed items. A shift in consumption from a fresh product to its processed forms is of par- ticular interest and importance to the po- tato industry. The total consumption of potatoes would be altered, of course, only to the extent that more or less potatoes are needed for both fresh and processing markets. If the processed potatoes are easily stored, potato supply from a good year can be carried over into a year of low production. Some marketing costs may be reduced through easier handling and transporting of the processed products rather than the more bulky raw product. Parts of the crop that are not suitable for the fresh market because they are not uniform in size or shape may be proc- essed, often bringing higher total and net returns to the growers. The consumer will be able to choose between the fresh or processed product depending on the rela- tive prices, qualities, and supplies of the fresh and processed items. Consumption trends Potatoes, grown commercially in all fifty states, are now available at retail in about 70 processed forms. Yet per-capita con- sumption has had a strong downward trend during the twentieth century until about fifteen years ago when the decline became less pronounced. Figure 2 suggests that the case of potatoes is not exceptional. For example, consumption of flour and cereal products decreased more — a decline of 40 per cent since 1925 compared to 30 per cent for potatoes. During the past decade, the use of po- tatoes changed sharply (see figure 3) . Fresh consumption decreased by one- eighth from 8 million to 7 million tons, while use in processing outlets increased sharply from li/2 million to almost 5 mil- lion tons. Thus consumption of processed potatoes rose from 16 to 41 per cent of the food use of potatoes. Consumers were in- creasingly turning to processed potato products. A similar shift from fresh to processing outlets took place in the fruit and vegetable industry during and imme- diately after World War II. Rice is also a major food staple in many areas of the world. The United States is one of the major rice-producing countries. Most of the rice is grown in five states: Arkansas, Louisiana, Mississippi, Texas, and California. Over one-half of the crop is exported and about one-quarter, mostly the long-grain varieties, is consumed do- mestically. The remainder is used for seed, in brewing, and in cereal manufacturing (U. S. Department of Agriculture, 1968fr). Civilian per-capita consumption of rice averaged 7.4 pounds per year in 1910- 1920, fluctuated about a long-term average of 5.4 pounds during 1925-1960, and then increased to over 7 pounds for recent years (Appendix table C-l) . The current in- crease coincides with the rice industry's effort to increase rice usage in American homes by marketing a large number of rice specialty products. Inasmuch as there are no good statistics on the production and consumption of macaroni, spaghetti, and noodle products, it is necessary to make estimates based on consumption of durum and semolina flour, which is not used extensively for products other than paste foods. These products were made primarily from durum and semolina flour until the early 1950's but since then have used other flours also in blends, especially during of years of low domestic production of durum wheat. This means that the consumption of macaroni, spaghetti, and noodle products probably increased (on a trend basis) for the entire period since 1925 rather than dropping sharply as suggested by the bottom panel of figure 2. Recently, paste foods have been in greater demand as part of one-dish con- venience products. Noodle- and macaroni- based casserole and side-dish items have been offered in retail food markets in a wide variety of forms. Consumer accept- ance of these new products has increased the overall consumption of paste foods. Moreover, the supermarket trade is fore- casting a further fast growth in the sale of such new convenience items (Bennett, 1968) . 4 Thus, there has been a considerable in- crease during the past 10 or 15 years in the use of processed potatoes, rice specialty products, and convenience items made from paste foods. Apparently this expan- sion has served (at least, in part) to check the downward trend in potato consump- tion and to increase the use of rice and macaroni products. 4 A convenience food is a food product available to consumers at retail that embodies a flavor-variety feature or a timesaving factor because it is premeasured or semiprepared. [6] 300 200 Fresh fruits, melons, and vegetables Note — Processed products included on fresh basis Potatoes (total) LViuJ ^ULUWJJ ■ Prnrpccdfl r»r»tntri£>c -^ ■■ Processed potatoes O a. I 200 a. s z o u -J 100 z z < - „ Other flour and cereal products Meat and poultry ^^ ^-~~~"^ "^ - ^^r___ Sugar ^ -" - ""^ ■ Fats and oils 1 1 1 1 1 i i i i - Rice ^s" -v -^^ - - ^«r Semolina and Durum flour - - Note — This panel has a different vertical scale 1 1 1 1 1 1 i i i - 1 925 1 930 1 935 Source: Appendix tables C-l and C-2. 1 940 1 945 YEAR 1950 1955 1960 1965 Fig. 2. U. S. Per-Capita Consumption of Certain Foods, 1925-1965, Pounds Per Year. [7] 12 FOOD USE — MILLION TONS Processed Use Fresh Use 40 - 30 20 - 10 - - PROCESSED USE— PER CENT OF TOTAL FOOD USE 1 £ Canned — Frozen Dehydrated | Chips and Shoestrings " TTTTT TTTtt - ^fWF ■-..,;,■ . '.-.-■. ,.■ 1965-1967 1956-1958 1959-1961 1962-1964 Source: Appendix table C-2. Fig. 3. Food Uses of U. S. Potatoes, 1956-1967. [8] MAJOR FINDINGS Almost all of the survey households (99 per cent) include potatoes in their basic diet. Overwhelmingly, they prefer using fresh to processed potatoes although this food is available in many different proc- essed forms. Two-thirds of the families serve baked or mashed potatoes more fre- quently than any other style of potatoes. They have a slight preference for baked over mashed potatoes. Bay Area families prefer the red potato for boiling and the California long white for other uses. Other survey families select long white and russet potatoes with equal frequency for French fries and the russet for other preparations. However, an ap- preciable number are not aware that dif- ferent varieties are better suited for some uses than for others. The choice of variety is not considered important by 10 per cent of the homemakers when baking potatoes and by 17 per cent when they serve pota- toes in other forms. Almost one-half of the homemakers (43 per cent) consider freedom from defects to be more important than any other factor when purchasing fresh potatoes. They rank price and intended use (the next two factors) much lower (18 and 12 per cent) . Generally, the respondents consider po- tatoes to be not fattening or only mod- erately so. However, one-sixth of them say that potatoes are very fattening — a con- clusion contradicted by nutrition research. Apparently, the public needs more infor- mation about the caloric and nutritional of potatoes. Sixty per cent of the households inter- viewed reported that they serve rice as a change from potatoes, 26 per cent sub- stitute paste foods, and 14 per cent, other foods. Thus, many families seem to pre- fer a starch in their meal patterns and commonly use rice when making a substi- tute for potatoes. Rice was the preferred substitute relatively more often in house- holds where family income was high or the wife had more formal education, per- haps indicating a greater awareness of ways for achieving greater variety in the diet. Potatoes were served twice as frequently as rice in the households surveyed and three times as often as macaroni products. All three foods were served more fre- quently by Mexican-American households, those with large families, those in the low income group, and those where the main earner was employed in a service or labor occupation. Households in the higher income group and in the professions served potatoes, rice, and macaroni products less often than the other groups studied. One survey also dis- closed a greater propensity among these households to buy convenience products made from these three basic foods. This tendency supports the widely held thesis that those who have an ample supply of food are likely to seek greater variety in their food purchases. Food becomes more than a way to provide needed nutrients, it is part of a desired style of life. Many prepackaged and convenience foods are associated with elegance and leisurely din- ing rather than "merely" eating. These "elegant" forms are available at a cost and in an easy-to-prepare form which saves the homemaker time and effort. The coming decade will see many new families formed from our currently large teenage population. Generally, these new households will have a higher income level than their parents, more formal education, and proportionately more employed in the "white collar" occupations. Many of them have already been introduced to a wide variety of easy-to-prepare foods in restau- rants and in their parental homes. They accept many of these convenience and processed foods and probably will continue to purchase them and to try others that become available for home use. Our findings indicate that convenience foods made from potatoes, rice, and paste foods are used in many households. They are purchased considerably more fre- quently by families with higher income, of large size, with a husband in the "white collar" occupations, and a wife who has completed a high school education. As more households acquire these character- istics, they, too, may spend more of their food dollar for products offering more variety, elegance, and convenience. In the light of expected demographic shifts in the population, per-capita "de- [9] mand" for easy-to-prepare items made from potatoes, rice, and paste foods should increase considerably. If this development occurs (as it is expected to do) , the indus- tries producing these items are likely to offer them in larger quantity (and variety) to meet the demand for variety and time- saving features expressed by the general public as well as by the institutional trade. The total consumption for these three foods, in processed and "unprocessed" forms, is likely to expand. A word of caution seems in order, es- pecially at this time when the American public is changing its views about many things — sometimes rapidly and drastically. Numerous socioeconomic factors make for differences in consumer tastes and prefer- ences. For example, the consumers' will- ingness to try new convenience items pro- duced from the three basic foods consid- ered here is also likely to lead to a willing- ness to try other new food products that may appear on the scene. Thus, attention must also be given to changes in avail- ability of substitutes, relative price pat- terns, ideas concerning a "good" diet, and other variables affecting the total picture. Consideration of these factors is beyond the scope of this study. POTATOES: CALIFORNIA CONSUMPTION PATTERNS of potatoes, and whether or not they think potatoes are fattening. Misinformation and calorie counters In a country with an overly abundant food supply and a relatively high income level, food fads and fashions can determine the success and failure of food firms. In recent years, the American public has become very interested in dietary matters and has tried to restrict the intake of starches and carbohydrates, to conform with its ideas (frequently nutritionally incorrect) con- cerning a "good" diet. Some misconceptions about potatoes were disclosed in our survey of California households. We asked the respondents, "Do you think potatoes are fattening?" 5 Almost 11 per cent of the replies were "don't know." Two-thirds of the respon- dents who gave opinions considered pota- toes to be fattening — 17 per cent said "very fattening," and 51 per cent "moderately flattening." The pattern of replies did not vary much according to the respondent's place of resi- dence — Bay Area, inland cities, or small cities. However, the replies varied substan- tially when families were grouped by demo- 5 The term "fattening'' is used in its colloquial sense. Any food is "fattening" if eaten in excess of caloric needs. On a comparative basis, potatoes contain more calories than the same quantity (in terms of weight) of most fruits and vegetables but much less than fats, oils, and sugars. Experiments are constantly undertaken to make the physical and quality characteris- tics of fresh potatoes more acceptable to consumers and processors. Information about consumers' selection and use of po- tatoes as well as processors' requirements for manufactured products can aid growers in adapting their crops better to market demands. While our primary concern in this portion of the study is with consumer purchasing patterns for potatoes and po- tato products, several other investigations indicate that consumers' choices are re- lated to certain quality characteristics of the available products. Some varieties of fresh potatoes are more suitable for cer- tain types of household preparation than are other varieties because of their differ- ent moisture and solid contents. Best re- sults are achieved by purchasing the po- tato most suited to a particular cooking method. Therefore, we asked respondents how they usually served potatoes in their households, which potato variety they used for different cooking methods, and whether they used fresh or processed potato prod- ucts in preparing meals. To increase our understanding of consumer knowledge concerning fresh potatoes, we asked about the factors that mattered in their choice [10] Table 1 DO YOU THINK POTATOES ARE FATTENING? Household characteristic Very fattening Moderately Not fattening fattening With opinion Without opinion* Total replies Per cent of replies with opinion Per cent of total replies Location Bay Area Inland cities Small cities Total Income level I (Under $3,000) II ($3,000 to $5,499).. III ($5,500 to $6,999).. IV ($7,000 to $9,999).. V ($10,000 to $14,999) VI ($15,000 and over). Total Education of wife Elementary High school HS graduate College Total Ethnic group Oriental Negro Mexican Nonwhite White Total 18.1 17.1 17.5 17.6 18.1 20.1 15.8 20.2 15.7 11.5 17.2 22.4 30.1 17.4 12.3 18.0 20.8 21.8 30.6 24.5 16.6 17.8 52.2 49.3 50.4 50.7 47.0 42.5 55.1 55.3 54.5 54.0 52.2 34.2 39.8 54.7 57.1 51.6 62.5 43.6 36.7 44.4 51.9 50.8 29.7 33.6 32.1 31.7 34.9 37.4 29.1 24.5 29.8 34.5 30.6 43.4 30.1 27.9 30.6 30.4 16.7 34.6 32.7 31.1 31.5 31.4 88.1 89.4 90.5 89.3 80.6 89.1 90.0 90.1 92.2 92.6 89.8 79.2 86.7 91.1 92.8 89.9 72.7 82.1 76.5 78.6 91.6 89.4 11.9 10.6 9.5 10.7 19.4 10.9 10.0 9.9 7.8 7.4 10.2 13.3 7.2 10.1 Number 420 340 430 1.190 103 156 141 253 269 122 1,044 165 460 334 1,055 961 1,153 * Combination of the three replies: "Don't know," "No opinion," and "Other." graphic factors (table 1) . 8 The proportion of respondents not expressing an opinion about the "fattening" effect of potatoes was much greater for nonwhite than white families and for those with lower incomes or less education. Relatively more of the respondents with opinions considered po- tatoes to be fattening (either very or mod- erately) if their education continued be- yond elementary school or if the family income was in the middle brackets. Selection of potatoes When asked, "Which factor matters most to you in selecting fresh potatoes?" (table 2) , 43 per cent of the sample indicated they wanted their potatoes to be free from defects. Another 44 per cent listed one of four other factors: price (18 per cent), in- tended use (11 per cent), cleanliness (8 per cent), and package (7 per cent) . Nonwhites and respondents with limited education and low family income attached less importance to freedom from defects and gave more weight to price than did the average interviewee. 7 For example, price was almost as important as freedom from defect among nonwhite families — 32 vs. 34 per cent. Generally, our survey consumers were much concerned about quality of the po- Each text table reports some of the data gathered during our survey of 1,192 households in Northern California. Its title is in the form of a question approximating one of those asked of respondents. The appendices contain information in greater detail. 7 The lowest income group was not as concerned about price as the others. This strange result can be expalined by the fact that more than half the households in this income group [ii] Table 2 WHICH ONE FACTOR MATTERS MOST TO YOU IN SELECTING FRESH POTATOES? Household characteristic Defect free Price Intended use Clean- Pack- Otherf Total replies Per cent of replies Location Bay Area Inland cities Small cities Total Income level I (Under $3,000) ... II (S3, 000 to $5,499).. III ($5,500 to $6,999).. IV ($7,000 to $9,999).. V ($10,000 to $14,999) VI ($15,000 and over). Total Education of wife Elementary High school HS graduate College Total Ethnic group Oriental Negro Mexican Nonwhite White Total 41 4 38.2 44.5 42.7 30.1 41.2 48.2 42.7 49.6 44.6 44.0 31.2 43.9 43.0 45.4 42.8 33.3 35.8 30.2 33.5 44.4 42.6 14.8 18.3 20.0 17.7 15.5 25.8 18.4 16.2 14.5 9.9 25.0 22.0 17.6 15.0 18.2 21.2 33.7 34.9 31.9 14.8 17.6 11.2 12.7 11.0 11.6 12.0 10.4 8.5 11.5 13.8 11.7 9.1 8.4 6.4 7.9 12.4 11.7 7.2 7.9 7.6 9.4 11.0 6.3 5.1 6.9 3.0 11.6 11.1 10.0 7.1 7.6 10.5 4.2 6.3 7.1 9.7 5.2 7.9 8.0 6.8 3.3 6.8 5.2 4.2 8.8 7.2 7.2 18.2 3.1 7.9 7.3 7.3 7.3 11.9 18.9 10.3 13.3 16.5 13.6 10.6 12.6 11.6 14.9 12.9 18.8 10.4 12.8 13.5 13.2 15.2 7.4 9.5 9.4 14.0 13.2 Number 419 338 429 1,186 103 155 141 153 268 121 1,041 96 164 460 333 1,053 33 95 63 191 958 1,149 * Combination of two factors: "Whether they are packaged or not" and "The size of the package." Replies were about equally divided between these two factors. t Combination of "None of these" and "Other." tatoes brought. Earlier studies reported similar results. A survey conducted by the U. S. Department of Agriculture (1948) revealed that homemakers looked for po- tatoes that had a smooth skin, clean sur- face, few deep eyes, and no spots or blem- ishes. Libeau's report (1951, pp. 14-17) arrays factors affecting consumers' selec- tion of fresh potatoes in the order of quality, medium size, and price. Some varieties of fresh potatoes are more suitable for certain food preparations than others. Other studies report that con- sumers prefer to purchase one all-purpose potato rather than to buy a different va- riety for each intended use. If they pur- chased a special-purpose potato, it was for baking (U. S. Department of Agriculture, 1948, p. 4) . California consumers generally have a choice of four fresh potato varieties: round white, long white, russet, and red potatoes. The long white and russet varieties are used most often, largely because they are more available. To learn more about con- sumers' buying methods, we asked ques- tions designed to test respondents' knowl- edge of the four varieties and their cook- consisted of families whose main earner was retired, unemployed (maybe merely between jobs), a student, or a homemaker. Presumably many of these are supplementing current income by other sources of money to cover living costs and are continuing their preexisting expenditure patterns. Interpretation of data in some of the other tables may need to be modified also to take into account the unusual composition of families in the lowest income group. [12] Table 3 IF YOU WANT TO SERVE POTATOES IN EACH OF THE FOLLOWING FORMS, WHICH VARIETY OF FRESH POTATO WOULD YOU SELECT? Use and location Russet Long white Round white Red Does matter Doesn't matter Don't know Don't serve Total replies Per cent of replies ivith preference Per cent of total replies Baked Bay Area. . . Inland cities Small cities. Total Mashed Bay Area. . . Inland cities Small cities Total Boiled Bay Area. . Inland cities Small cities. Total Potato salad Bay Area. . Inland cities Small cities. Total French fried Bay Area. . Inland cities. Small cities. . Total 42.5 66.3 58.3 55.2 36.0 50.2 51.4 46 25.1 39.1 43.4 35.9 26.1 42.4 43.9 37.4 32.8 43 5 42.8 39.9 48.9 4.9 20.7 10.4 30.9 6.2 34.2 6.9 48.8 9.0 31.6 15.2 36.1 8.7 38.9 10.8 26.7 16.4 28.6 17.7 29.4 14.4 28.3 16.0 42.8 13.6 33.6 16.8 34.5 11.0 37.1 13.5 55.4 6.2 44.5 9.0 47.1 5.7 48.9 6.9 3.7 2.6 4.6 3.7 6.2 3.0 3.8 4.3 31.8 14.6 12.8 19.8 17 5 7.2 10.6 12.0 5.6 3.0 4.4 4.3 79.0 82.3 86.5 82.6 69.1 80.5 75 5 74.7 72.2 80.4 75 3 75.7 67.5 72.6 72.6 70.8 46.7 60.4 54.1 53.3 12.8 3.6 7.9 2.8 9.8 1.4 10.3 2.6 17.7 4.6 12 3.9 19.5 1.0 16.7 3.1 16.4 3.6 9.4 3.0 17.2 3.3 14.7 3.3 14.9 4.8 13.4 3.3 17.3 2.1 15.4 3 4 14.6 5.0 12.1 2.7 13.0 3.8 13.3 3.9 4.6 7.0 2.3 4.5 8.6 3.6 4.0 5.5 7.2 4.2 6.3 12.8 10.7 8.0 10.4 33.7 24.8 29.1 29.5 Number 414 427 1,169 418 334 425 1,177 414 331 425 1,170 415 328 427 1,170 418 331 423 1,172 ing qualities. The varieties were listed on the questionnaire but were not described for the respondents. Their preferences are significant because they reveal a knowl- edge about cooking qualities that conforms closely to research findings. 8 Table 3 summarizes replies to the in- quiry about respondents' varietal prefer- ences in fresh potatoes. Preferences were expressed by most homemakers (83 per cent) when they bake potatoes, by three- quarters (71 to 76 per cent) for mashing, boiling, and potato salad, and by only one- half (53 per cent) for french frying. Re- spondents to whom variety mattered had a definite preference for russet and long white potatoes, especially for baking, mash- ing, and french frying. Round white and red potatoes were preferred by 36 per cent for boiling and 25 per cent for potato salad. Bay Area housewives had a different pat- tern of varietal preferences. They selected long white over russet potatoes for all five uses, whereas the reverse situation tended to prevail in the other areas. We have no explanation for this considerable variation. Serving potatoes Two-thirds of the sample households in- dicated a preference for baked and mashed potatoes. Boiled potatoes were served most frequently in 13 per cent of the house- holds, french fries in 7 per cent, and other 8 The russet is a general- purpose "old" potato. The round white, another "old" potato, is best for boiling and french fries. Long whites, tubular "new" potatoes, are suited for boiling, mashing, and potato salad. The red potato, a block round "new" potato, is better for boiling than for other uses. [13] forms of potatoes in 13 per cent. Only 12 respondents (1 per cent of the sample) indicated using potato chips more fre- quently than any other form of potatoes. Possibly chips were not mentioned more frequently because many respondents re- gard them as a snack rather than a staple food or may not even consider chips as a potato product. These findings differ from those of ear- lier studies, mainly by indicating a higher preference for baked potatoes by today's Northern Californians and a lower pref- erence for boiled potatoes. A 1948 survey of U. S. households reported that the most frequently served form of potatoes was mashed, boiled, fried, baked, and creamed — in that order. Homemakers in Los An- geles reported that the proportion of po- tatoes used for baking was about the same as for mashing and boiling (U. S. Depart- ment of Agriculture, 1948, pp. 16-17) . Studies in 1953 and in 1957 reported that potatoes were served mainly in the mashed or boiled form (LeBovit and Clark, 1956, p. 18). A 1961 U. S. Department of Agri- culture study (Harp and Dunham, 1961, p. 1) reported that mashed potatoes rep- resented 34 per cent of the home consump- tion of potatoes. Apparently the relatively high preference for mashed potatoes pro- vided dehydrators with an excellent op- portunity to market potato granules and flakes. Preference patterns for our survey house- holds varied substantially among house- holds located in different areas, with dif- ferent income and educational levels, or in different ethnic groupings (see table 4). The major difference geographically was Table 4 WHEN YOU HAVE POTATOES AT HOME, WHAT ONE FORM DO YOU SERVE MOST FREQUENTLY? Household characteristic Baked K Mashed" 1 Sub- total Boiled * French fried Potato chips Other Total replies Location Bay Area Inland cities Small cities Total Income level I (Under $3,000) II ($3,000 to $5,499). . III ($5,500 to $6,999). . IV ($7,000 to $9,999) . . V ($10,000 to $14,999) VI ($15,000 and over). Total Education of wife Elementary High school HS graduate College Total Ethnic group Oriental Negro Mexican Nonwhite White Total 43.5 32.0 29.8 35.2 16.9 23.1 36.1 48.9 36.4 38.9 18.2 11.1 19.4 38.6 35.4 Per cent of replies 26.8 70.3 12.4 5.7 1.4 31.4 63.4 10.8 8.5 1.8 36.5 66.3 15.6 6.3 0.0 31.7 66.9 13.1 6.7 1.0 35.1 53.7 16.2 9.7 0.0 32.2 55.4 17.7 11.5 0.0 28.5 57.7 14.7 7.8 1.4 34.5 69.8 11.6 7.1 0.4 32.7 76.2 11.5 4.1 2.2 25.5 82.2 8.8 2.5 0.8 31.9 67.5 13.0 6.8 1.0 33.5 50.4 15.3 11.4 0.0 37.1 60.2 12.5 7.9 0.6 31.7 67.8 13.3 7.2 1.1 26.8 75.7 11.1 3.9 1.8 31.2 67.6 12.7 6.6 1.1 17.6 56.5 13.1 3.0 6.1 35.1 53.3 14.0 16.9 0.0 37.8 48.9 10.3 15.7 0.0 33.0 52.4 12.7 14.1 1.0 31.1 69.7 13.3 5.2 1.0 31.4 66.8 13.2 6.7 1.0 10.2 15.5 11.8 12.3 20.4 15.4 18.4 11.1 6.0 5.7 11.7 22.9 18.8 10.6 7.5 12.0 21.3 15.8 25.1 19.8 10.8 12.3 Number 421 341 430 1,192 103 156 141 253 269 122 1,044 96 165 460 334 1,055 33 95 64 192 961 1,153 * Replies "Mashed, baked as frequently' the forms indicated. and "Mashed, baked, boiled as frequently" distributed equally among [14] in the preference for baked and mashed potatoes. In the Bay Area, baked and mashed potatoes were served most fre- quently by 43 and 27 per cent of the house- holds compared to 31 and 34 per cent in other areas. There was a pronounced preference for baked potatoes with increases up the in- come and educational scale. For example, baked potatoes were the preferred form of serving potatoes in 17 per cent of the households where the wife's education stopped at the elementary school and 49 per cent if she attended college. Prefer- ences for mashed and boiled potatoes de- clined from 49 to 38 per cent and those for french fries and other potatoes from 34 to 13 per cent as the wife's education went from elementary school to college atten- dance. A similar shift was indicated by changes in income. On an ethnic basis, some differences were pronounced while others were negligi- ble. Baked potatoes were the form served most frequently by 39 per cent of the white families and by 15 per cent of the Negro and Mexican-American families; french fries by 5 and 16 per cent, respectively. Mashed and boiled potatoes were pre- ferred by 31 and 36 per cent, boiled pota- toes by 13 and 13 per cent, and potatoes in other forms by 12 and 20 per cent, re- spectively. Substitutes for potatoes Bennett (1941) stated that as a nation becomes more affluent its food pattern shifts away from a heavy reliance on po- tato or cereal products toward greater variation in its diets. This variation may be achieved in many ways. One of the most successful is to use modern technol- ogy for creating new products out of the original basic raw materials. For example, the potato industry would like to have consumers who are looking for more vari- ety in their meal patterns shift from one form of potato to another form rather than to other products. The rice industry is trying to increase consumer use of its products by marketing many specialty rice products, thereby giving the consumer a wider choice in using and serving rice. Likewise, the macaroni and noodle indus- try is encouraging creative menu planning and making available to consumers more processed dinner products with a macaroni or noodle basis. Table 5 AS A CHANGE FROM POTATOES, WHICH FOOD WOULD YOU SERVE? Household Macaroni or Salad or Other food Total characteristic Rice spaghetti Noodles vegetable or nothing replies F er cent of replit s Number Location Bay Area 69.0 10.7 9.6 5.0 5.7 419 Inland cities 52.8 22.1 5.3 14.2 5.6 339 Small cities 56.9 23.1 7.2 8.8 4.0 429 Total 60.0 18.4 7.5 9.0 5.1 1,187 Income level I (Under 13,000) 59.2 19.4 6.8 7.8 6.8 103 II ($3,000 to $5,499) 53.8 25.6 4.5 9.0 7.1 156 Ill ($5,500 to $6,999) 51.1 28.4 11.3 7.1 2.1 141 IV ($7,000 to $9,999) 60.1 15.8 5.9 13.8 4.4 253 V i$10,000 to $14,999) .... 62.9 15.2 7.8 8.9 5.2 269 VI ($15,000 and over)... 71.3 9.0 9.8 5.8 4.1 122 Total 59.8 18.4 7.5 9.4 4.9 1,044 Education of wife Elementary 60.4 18.7 4.2 11.5 5.2 96 High school 49.1 27.9 8.5 11.5 3.0 165 HS graduate 57.8 20.4 8.1 8.7 5.0 460 College 67.3 11.7 7.2 7.5 6.3 334 Total 59.7 18.7 7.5 9.0 5.1 1,055 [15] Frequency, potatoes, rice, and maca- roni products are substitutes for each other as the one starch food, frequently in addition to bread, included in many American meals; they are competitors in the American diet. Hence, we asked, "As a change from potatoes, which food would you serve?" Rice was the first choice by 60 per cent of the sample households, paste products by 26 per cent, and other foods by 14 per cent (see table 5). The pref- erence for rice as the first substitute for potatoes is much greater in the Bay Area than elsewhere — by a larger margin than can be explained merely by the Oriental population there. Table 5 also reveals that the choice of potato substitutes was affected by income and education within the household. Rice was chosen less often — and macaroni prod- ucts more frequently — by women with only a high school education than those who did not go beyond elementary or who at- tended college. There is a similar curvi- linear relation to income level. The reason for such choices is not clear. Possibly the explanation is largely in terms of relative prices at the lower income and education levels and in terms of a search for greater variety in dietary patterns by families on higher socioeconomic planes. FREQUENCY OF USE OF POTATOES, RICE, AND MACARONI Bennett's explanation (1941) of potato- cereal consumption referred to the eco- nomic development of a nation as a whole. Within a nation as diverse as ours, indi- vidual families vary and many factors may be used to explain the differing food con- sumption patterns among households. A family's identity, income level, ethnic group, educational level, age, or size may affect its use of food products. The em- ployment status of the husband, wife, or both may cause the family to set certain patterns. Meal patterns are also affected by food habits the household members ac- quired from their parents, their tastes and preferences, and their willingness to ac- cept or try new products. Within established diet patterns, some foods are so important to a family that they will be consumed daily. Others will be served occasionally and some infrequently. Potatoes and possibly rice and macaroni products are foods eaten frequently by many Californians. Variations in the use frequency among the survey households are indicated by the data reported in the tables of Appendix C. Major comparisons are given in table 6 and figure 4. Potatoes are eaten much more often than the other two foods — 3.1 times per week, on the average, compared to 1.3 for rice and 1.0 for macaroni products. The difference can be put in another way. About three-quarters (73 per cent) of the sample households use potatoes at least twice a week compared to one-third (33 per cent) for rice and one-fifth (20 per cent) for macaroni products. Consumption varies in the different parts of the survey area. This study showed that use frequency varied considerably in association with several characteristics describing the house- hold. Generally, potatoes, rice, and maca- roni products were eaten more frequently by Mexican-American families, households with large families, those with incomes be- low $7,000, those where the wife's educa- tion ended before high school graduation, and those where the main earner was in a service or manual labor occupation. Fami- lies of smaller size, at higher income levels, with more formal education, and in the professional and "white collar" occupa- tions used these foods less frequently in their diets and presumably substituted other foods for more variety. In other words, those survey families which might be put in the "blue collar" or lower socioeconomic groups used potatoes, rice, and macaroni products more frequently. Variations in use frequency were more pronounced and regular for macaroni products than for potatoes and rice. Maca- roni products were used frequently (at least twice a week) by 41 per cent of the [16] 1 1 i ' / / ' / / / / / fry ' ' /// .— t << ^ c — >. a o o CD 1 ! i > C vv o \ x w \ Vv o _c en c •T3 %- u c // / ID •'/ / Q£ / / / . i i 1 9 o CO o CN o o i 1 1 1 / / /. / a> c o £ / o / (?) en / O 'E c D o X / \ 1 f "5 / u / c u 2 _c t o Z 1 1 | 1 J *i a -^ - c 1 1 1 1 / J i / '/ / " / \y < ^ =S y\ a -? E eo — to A \ i : i V< N - ill \ sx x . \ n \ x> n^ v \\ %■ o fl D O A //i . 1 1 i i i HT 1 1 ay Area iland cities mall cities / "I *s s ^r^ ^£^ S' ^ X ^s>\ \T\- ^. S. N *s \ \ \\ D o vv a. ,\N c o o • 1 u 3 li 1 1 1 i o o o 1 — ^f SQlOHBSnOH JO 1N3D M3d Table 6 HOW OFTEN DO YOU EAT POTATOES, RICE, AND MACARONI PRODUCTS? Household characteristic Potatoes Twice a week or more Once a week Less often Rice Twice a week or more Once a week often Macaroni products Twice a week Once a week Less often Total replies Location Bay Area Inland cities Small cities Total Income level I (Under $3,000) II ($3,000 to $5,499).. III ($5,500 to $6,999).. IV ($7,000 to $9,999).. V ($10,000 to $14,999) VI ($15,000 and over) Total Education of wife Elementary High school HS graduate College Total Ethnic group Negro Mexican Oriental Nonwhite White Total 65.1 78.3 77.0 73.2 68.0 82.0 80.9 72.7 70.3 66.4 73.4 77.1 81.8 76.3 67.7 74.5 70.5 85.9 33.3 69.3 74.1 73.3 Per cent of replies 19.9 15.0 41.1 30.9 28.0 17.1 42.3 10.6 11.1 32.0 30.5 37.5 21.1 40.5 11.9 11.1 25.1 27.2 47.7 21.4 34.9 14.3 12.5 32.7 29.5 37.8 19.8 39.1 12.6 19.4 42.7 20.4 36.9 22.3 29.1 7.7 10.3 30.1 29.5 40.4 25.0 38.5 10.6 8.5 33.3 27.7 39.0 30.5 36.9 17.8 9.5 30.1 31.2 38.7 17.4 41.1 16.7 13.0 26.8 32.7 40.5 14.9 42.4 17.2 16.4 36.9 32.8 30.3 10.7 45.1 14.4 12.2 31.7 30.0 38.3 19.3 39.8 11.5 11.4 41.6 29.2 29.2 28.1 28.1 7.3 10.9 27.3 29.1 43. 6 26.1 38.8 15.0 8.7 26.3 33.9 39.8 20.0 42.0 16.4 15.9 37.4 29.7 32.9 15.9 41.3 13.9 11.6 31.4 31.4 37.2 20.4 40.0 17.9 11.6 49.5 32.6 17.9 28.4 39.0 9.4 4.7 73.4 14.1 12.5 40.6 29.7 18.2 48.5 90.9 6.1 3.0 9.0 45.5 15.1 15.6 64.6 21.9 13.5 29.2 37.0 14.0 11.9 26.7 30.8 42.5 17.9 39.3 14.2 12.5 33.0 29.3 37.7 19.8 38.9 40.6 38.4 43.7 41.1 48.6 36.5 32.6 41.5 42.7 44.2 40.9 43.8 35.1 38.0 42.8 39.6 32.6 29.7 45.5 33.8 42.8 41.3 Number 421 341 430 1,192 103 156 141 253 269 122 1,044 165 460 334 1,055 95 64 33 192 961 1,153 Note: For more detail, see Appendix tables C-7, C-8, and C- Mexican-American families compared to 28, 18, and 9 per cent, respectively, for Negro, white, and Oriental families. The percentage increased with family size — being 11 per cent for families of one or two persons, 20 per cent for three- to four- person families, and 28 per cent for larger families. It was negatively correlated with income and education. The percentage de- creased from 26 at incomes under $7,000, to 16 at incomes of $7,000 to $15,000, and to 11 at higher levels. It declined from 27 if the wife had not graduated from high school, to 20 if she had, and to 16 if she attended college. The percentage was 32 for service and labor occupations and 17 for professional and clerical workers. Potatoes had similar variations in use- frequency patterns, although the changes were of smaller magnitude. Potatoes were eaten frequently by 86 per cent of the Mexican- American families and 74, 71, and 33 per cent, respectively, by white, Negro, and Oriental families. (Note that the order of white and Negro households was reversed.) Again, frequent use was cor- related positively with household size and negatively with income and wife's educa- tion. It was higher for service and clerical workers (83 per cent) than for professional and clerical occupations (70 per cent). A different situation prevailed for rice. Over 90 per cent of the Oriental families used rice frequently. The percentages were 73, 50, and 27, respectively, for Mexican- American, Negro, and white families. [18] There were a positive correlation with fam- ily size. The correlation with both educa- tion and income was U-shaped. Thus, families with a high school educated wife ate rice at least twice a week less often than those where the wife had not gone beyond elementary school or, on the other hand, had continued with college. House- holds where the main earner was in a ser- vice or labor occupation ate rice more often but not by as wide a margin as in the case of macaroni products or potatoes. Pratt's inquiry (I960) into the domestic consumption of rice in the United States was based, in part, upon interviews made in 1957 at 560 households located in At- lanta, Dallas, and Denver (Harp and Dun- ham, 1961). His data indicate higher use frequencies for nonwhite families, larger households, and those with higher income. However, they indicate that the average family ate rice only 2.0 times per month (appendix table C-6) compared to 5.9 for our survey. This difference is much greater than can be explained by the higher per- capita consumption prevailing in Califor- nia or the upward trend that occurred. Both surveys used approximately equiva- lent questions, and we can offer no satis- factory justification for the discrepancy. The 1965 Household Food Consumption Survey for the Western Region of the United States generally reported a higher proportion of households using macaroni at all income levels than found in our study (U. S. Department of Agriculture, 19656). USE OF CONVENIENCE FOODS American consumers have a wide choice in their food shopping. The range of selection is broadening and deepening. The average retail grocery store stocked about 900 items in 1928 and 7,500 in 1968. This increase is due to several factors: growth in the size of individual stores, handling of more nonfoods, multiplication of items by a greater number of brands, introduction of new products at a rate in excess of products discontinued, etc. Gro- cers consider new items important for in- creasing sales volume and as a competitive factor. Many of these new items are convenience or specialty products that offer home- makers different foods or less time-consum- ing preparations for familiar dishes. Pota- toes, rice, and macaroni are available in canned, dehydrated, frozen, instant, and flavored food forms. Competition exists among the potato, rice, and macaroni industries and among their individual companies as they vie for consumer acceptance of their products. The amount of food that consumers eat is relatively fixed. Their purchases of starchy products is related to price, tastes and preferences, and available substitutes. The three food industries compete directly for the share of the consumer dollar that goes for starchy foods. Furthermore, there is competition within each industry among different companies as they attempt to in- duce consumers to buy their products. Each company hopes that consumers will accept its new products. Each industry hopes that consumers will buy more of the items its members produce than of substi- tutes produced by other industries. There- fore, the companies individually as sepa- rate firms and collectively as members of an industry are interested in knowing more about why consumers try new products. Some marketing studies have been di- rected toward determining the character- istics of persons who are more prone to experiment and then to accept new prod- ucts that are marketed. A study made by Cornell University found that early adopt- ers of new products usually are people who are willing to take risks, to accept change and innovation, from higher income levels, and the better educated groups. They tend to come from urban areas and to be in families where the husband is in a profes- sional or "white collar" occupation, the wife is employed ,and they have been mar- ried from four to fifteen years (Swack- hammer, 1964). In another study, Lion- berger (1968) mentions many of these same [19] factors. He also points out that late adopt- ers tend to come from small farms, to have low incomes, to be skeptical and security minded, and to be over the age of sixty. As changes occur in the demographic characteristics of the American population, we expect changes to occur in their food habits and dietary preference. The mar- keting studies pointing out characteristics of new product users also may be defining changing characteristics of the population as a whole: the geographic clustering of families in urban rather than rural settings, more married women in the labor force, the general rising level of income, the rising educational level of the population, and the increase in young people in the population who are supposedly more re- ceptive to new ideas and innovations. As our population has undergone change, the foods available on the retail- er's shelf have increased in variety and more conveniences have been offered to consumers. We wish to determine the changing use patterns for selected con- venience foods. Hence, the respondents in this study were asked about their use of 15 food items containing potatoes, rice, and macaroni. Specifically they were asked, "Have you ever purchased any of the fol- Table 7 HAVE YOU EVER PURCHASED ANY OF THE FOLLOWING CONVENIENCE FOODS? Convenience food By location Bay Area Inland cities Small cities Total By income level Under $3,000 I $3,000- $5,499 II $5,500- $6,999 III $7,000- $9,999 IV $10,000- $14,999 V $15,000 and over VI Number of Replies 421 341 430 1,192 103 156 141 253 269 122 Per cent reporting purchase* Instant mashed potatoes. . Frozen french fries Instant scalloped potatoes Frozen potato puffs Frozen potato patties Instant au gratin potatoes Instant rice Flavored rice mixes Parboiled rice Frozen rice mixes Packaged macaroni or noodle dinners Canned macaroni or spaghetti Packaged noodles almondine Frozen macaroni or spaghetti dinners Packaged beef stroganoff . 61.1 67.9 25.9 24.9 20.2 59.9 48.9 52.5 8.1 47.0 46.6 28.5 25.2 23.5 70.4 72.8 68.0 50.5 57.1 66.7 73.3 76.9 62.2 69.8 66.9 36.3 57.7 63.1 74.6 80.5 17.9 25.8 23.6 14.7 14.7 17.7 26.6 27.2 14.9 19.8 20.2 3.9 10.9 17.0 19.8 33.2 13.8 19.5 18.1 6.9 14.1 16.3 16.3 24.7 13.8 21.6 17.7 6.9 8.3 13.5 20.2 24.6 68.9 71.4 66.6 48.0 58.3 63.8 71.4 73.0 59.2 62.6 56.8 31.4 42.6 55.0 67.7 70.7 39.0 38.8 43.7 25.5 35.3 41.1 45.4 51.3 5.9 5.3 6.5 1.0 3.8 5.0 8.3 5.2 59.8 61.4 55.9 39.2 51.9 53.2 57.9 66.9 45.7 57 4 50.2 32.4 47.4 42.6 56.3 59.0 21.1 23.5 24.6 12.9 14.1 15.0 . 24.3 34.9 18.2 26.3 23.6 15.7 19.2 25.5 22.2 22.5 17.0 22.8 21.4 10.8 12.2 18.4 26.2 26.6 Average per household^ Potato items Rice items Macaroni items . . . Total 15 items 2.17 1.69 1.71 5.57 1.93 1.73 1.62 5.28 2.29 1.78 1.91 5.98 2.14 1.74 1.76 5.64 1.19 1.06 1.11 3.36 1.63 1.40 :.45 4.48 1.94 1.65 1.55 5.14 2.31 1.93 1.87 6.11 2.67 2.00 2.10 6.77 85.8 76.2 41.0 32.5 30.6 27.9 77.0 71.1 58.2 15.6 62.8 57.4 45.9 36.9 28.9 2.94 2.22 2.32 7.48 * Per cent of households reporting having purchased the item t Number of items (per household) purchased at least once. Note: For more detail, see Appendix tables C-10 to C-13. least once before the interview date. [20] Table 8 HOW MANY OF THESE CONVENIENCE FOODS HAVE YOU EVER PURCHASED? Household characteristic By food type* Potato items Rice items Macaroni items By use frequency t Three used most Four others Eight used least Average per household^ Education of wife Elementary High school . HS graduate. . . College Occupation Service and labor Crafts Clerical Professional Age of main earner Under 35 35-44 45-54 55 and over Household size 1-2 persons . . . 3-4 persons... 5 or more. . . . Location 3ay Area T .nland cities Small cities Total Income level X (Under $3,000) . . . . £1 ($3,000 to $5,499) . . III ($5,500 to $6,000) . . IV ($7,000 to $9,999) . . V ($10,000 to $14,999) VI ($15,000 and over). Ethnic group Negro Mexican — Oriental Nonwhite . White 2.17 1.93 2.29 2.14 1.19 1.63 1.94 2.31 2.67 2.94 1.24 1.42 1.64 1.42 2.28 1.37 1.76 2.28 2.71 1.86 2.13 2.34 2.65 1.99 2.61 2.27 1.68 1.66 2.39 2.39 1.69 1.73 1.78 1.74 1.06 1.40 1.65 1.93 2.00 2.22 1.30 .84 1.28 1.14 1.88 1.32 1.46 1.86 2.05 1.51 1.71 1.94 2.04 1.55 1.93 1.94 1.63 1.55 1.92 1.74 1.71 1.62 1.91 1.76 1.11 1.45 1.55 1.87 2.10 2.32 1.29 1.42 1.29 1.32 1.77 1.09 1.48 1.88 2.18 1.65 1.67 2.14 2.10 1.75 2.07 1.92 1.38 1.49 1.90 1.93 1.89 2.01 2.14 2.01 1.35 1.73 1.94 2.19 2.30 2.39 1.28 1.44 1.33 1.40 2.13 1.49 1.80 2.16 2.30 1.90 2.01 2.15 2.33 1.70 2.22 2.13 1.95 2.04 2.20 2.07 1.19 1.77 1.92 2.28 2.48 2.50 1.62 1.35 1.83 1.57 2.09 1.52 1.80 2.24 2.39 1.95 2.18 2.34 2.40 2:10 2.37 2.27 1.68 1.70 2.17 2.26 1.73 1.23 1.64 1.56 .82 .98 1.28 1.64 1.99 2.59 .93 .89 1.05 .91 1.71 .77 1.10 1.62 2.25 1.17 1.32 1.93 2.06 1.31 2.01 1.83 1.30 1.30 1.82 1.67 * Potato, rice, and macaroni items, respectively, are the first six, next four, and last five convenience foods as listed in table 7. f "Three used most" are instant mashed potatoes, frozen french fries, and instant rice. "Four others" are flavored rice mixes, parboiled rice, packaged macaroni or noodle dinners, and canned macaroni or spaghetti. "Eight used least" are the foods listed in table 7. X Number of items (per household) purchased at least once. [21] lowing convenience foods?" Their replies are given in Appendix C and summarized in tables 7 and 8. Built-in conveniences The appeal of processed and convenience foods is conditioned by numerous factors. Cost per serving may be of major im- portance to some families because their in- come is limited or they are price conscious. The saving in preparation time may be of greater concern to those with higher incomes or who do not want to spend as much time preparing meals. For still oth- ers, the introduction of greater variety into their diet is of considerable impor- tance. Each homemaker will weigh the product's cost against the marginal utility of her food dollar, the time she saves in preparing meals by using the product, and its acceptance by the family. Varying degrees of processing and con- venience are built into the 15 products covered by this study. Some, such as fla- vored rice mixes, take about the same preparation time as if the product was made at home but their use offers new variations or different flavors. Others, such as packaged noodle or macaroni dinners and packaged beef stroganoff, have all of the required ingredients in premeasured quantities, are packaged in one conveni- ent container, and save the homemaker considerable preparation time. Generally, as more convenience is built into the product, its cost increases. For ex- ample, the instant mashed potatoes dis- played at supermarkets in Davis, Califor- nia, retail for about 4 cents per serving. The cost is about 9 cents for au gratin or scalloped potatoes. Flavored rice mixes cost 2 to 5 cents per serving more than instant rice while frozen rice mixes cost 6 or 7 cents more. A frozen macaroni or spaghetti din- ner, which is a complete ready-to-heat meal, costs about 35 cents and saves the homemaker a considerable amount of preparation time. Survey respondents reported substantial differences in purchase and use experi- ences for these convenience foods. These differences and the reasons for such varia- tions are examined in the next few pages. Since there are more similarities than dis- similarities in purchases by different fam- ily groups, the general patterns for all 15 convenience foods are discussed first. General use patterns About two-thirds of the sample households reported having purchased at least once each of three convenience foods: frozen french fries, instant mashed potatoes, and instant rice. Four others had been pur- chased almost as frequently (each by 44 to 57 per cent of the families): flavored rice mixes, parboiled rice, packaged macaroni or noodle dinners, and canned macaroni or spaghetti. Only 6 per cent of the re- spondents had ever purchased frozen rice mixes. The other seven convenience foods were tried by 18 to 25 per cent of the households. In other words, most families have not yet tried the newer products. Among the reasons for this situation is that they are not as readily available, are relatively unknown, and are not within the eating experience patterns of the re- spondents even as home-prepared dishes. Changes in the percentages of families that had tried using these convenience foods were associated with variations in family income, education, and other fac- tors. A strong positive correlation existed between income and the proportion of households reporting purchases for each of the fifteen items — see table 7. Families with incomes under $3,000 had purchased 22.4 per cent of the 15 foods (3.36 items). This proportion increased with succes- sively higher income levels to 50 per cent (7.48 items) for families in the $15,000 and over income bracket. Approximately the same regressional relation existed for po- tato items, rice items, and macaroni items. However, there were substantial varia- tions among individual items. These seemed to be related to purchase fre- quency. For the three items tried most often (by two-thirds of the households) the percentage of families having made pur- chases rose from 45 at the lowest income T level to 80 at the highest level. The cor- responding increase was from 30 to 62 per cent for the four foods in the second group, and from 10 to 32 per cent for the eight items tried infrequently. Thus, the rela- tive frequency of use with higher income tends to be greater for newer products, * [22] i.e., for those that have been tried less often by all households. Table 8 indicated the average number of items purchased per household classi- fied by various household characteristics. Data are given for potatoes, rice, maca- roni, and all 15 items. They are also given for the three groups classified by use fre- quency. Hence, it will be possible to com- pare variations for products used less often with those for products used fre- quently. Purchase frequency is correlated posi- tively with the wife's education as it is with income. The proportion of families making purchases increased from 25 per cent (3.78 of the 15 items) if the wife's education ended with elementary school to 46 per cent (6.94 items) if she attended college. It was much lower for families from craft and service occupations than those in professional and clerical occu- pations. Nonwhite families tried these items only two-thirds as often as did white families. Small families (one or two per- sons) had tried 4.7 of the 15 items com- pared to 6.1 for larger families. Interest- ingly, medium-sized families (three or four persons) had purchased about the same number of items as had large families (five or more persons). A definite curvilinear correlation prevailed when age (of main earner) was used as the independent vari- able. Families in the 35-54 age group pur- chased 6.4 items compared to 5.3 and 4.7, respectively, for younger or older families. Variations in purchase frequency were much greater for the eight least-used items (table 8, penultimate column) than for the others regardless of which factor is used as the independent variable. For example, wives who had attended college had pur- chased the least-used items three times as often as had those with only an elemen- tary education compared to 1.5 times as often for the most-used products. Simi- larly, professional families tried the least- used and most-used items 1.8 and 1.2 times as often, respectively, as had families in service and labor occupations. The eight newer products were tried 25 per cent less often in the inland cities than in the other two survey segments. Also the seven "older" items were purchased some- what less frequently (10 per cent) in the Bay Area. However, on the whole, these locational variations were not highly sig- nificant statistically. Differences by em- ployment status of the wife were even smaller; in fact, they were negligible. In brief, the 15 convenience foods, es- pecially those introduced most recently, were tried more often by California house- holds in the higher socioeconomic groups — as measured by family income, wife's education, husband's occupation, and the family's ethnic background. The propor- tion of users was lower among small fami- lies and for households where the main wage earner was under 35 or over 55 years of age. Thus, these findings generally support the results reported in other studies as to the basic characteristics of new product users. They also support Bennett's thesis (1941) that basic potato and cereal foods are replaced by substitutes in the econom- ically more developed countries. The sub- stitution is occurring here by households using more of the items which embody greater convenience in the form of a flavor variety feature or a timesaving factor be- cause the item is premeasured or semi- prepared. Reference was made early in this report to the definite tendency of Americans to replace fresh potatoes by processed potatoes in their diet and to use rice and macaroni specialty products in greater volume. Processed potato products The sharp increase in the processing of potatoes that occurred during the past decade or two and is still occurring does not mean that processed potato products are being substituted extensively for fresh potatoes in all households and in all cook- ing preparations. Actually, many people do not use processed potatoes except oc- casionally while others use them for some purposes and not others. Survey respon- dents were asked, "Which potato product do you buy when you want to serve potatoes in each of the following (eight) forms?" Their replies, summarized in table 9, indicated a decided preference for using fresh potatoes for each preparation except french frying. Almost all families serve potatoes in each of four ways. Fresh potatoes are pre- [23] Table 9 WHICH POTATO PRODUCT DO YOU BUY WHEN YOU WANT TO SERVE POTATOES IN EACH OF THE FOLLOWING FORMS? Item Baked Boiled Potato salad Mashed Scalloped Au gratin Patties French fries Number of replies Product purchased Fresh Dehydrated Frozen Other Total who serve*. Don't serve Total replies Bay Area Inland cities Small cities -Total Bay Area Inland cities Small cities Total 124 2 5 7 138 48 186 1,088 1,035 880 771 584 659 7 2 228 92 103 25 3 5 8 9 13 89 18 19 27 11 17 15 1,116 1,061 1,143 883 717 788 73 123 40 300 465 396 1,189 1,184 1,183 1,183 1,182 1,184 Per cent who serve\% 96.9 93.5 99.7 96.0 93.3 88.0 94.5 73.2 61.5 64.8 91.8 89.1 98.2 71.3 55.2 62.2 96.0 91.4 97.4 78.7 64.2 71.7 93.9 89.6 96.6 74.6 60.7 66.7 Per cent who buy fresh potatoes^, 97.8 99.1 99.5 96.7 96.4 80.3 86.6 81.3 80.8 97.1 97.3 73.5 88.0 82.8 87.7 98.5 99.0 76.7 87.5 80.7 83.3 97.5 97.5 77.0 87.3 81.5 83.6 543 2 393 26 964 220 ,184 80.4 82.0 82.0 81.4 53.0 62.2 54.9 56.3 * Number of sample households that buy any potato product for the indicated cooking preparation. t Number who serve the indicated cooking preparation divided by number of replies. i The percentages are derived from the replies given in the top portion of the table in the case of the "total' sample. They are derived similarly for the three locational segments from the same type of data which, however, are not given in this table. K Number who buy fresh potatoes to make the indicated cooking preparation. ferred by 98 per cent of the homemakers for baking, boiling, and potato salad, and by 77 per cent of them for mashed pota- toes. Scalloped, au gratin, and pattie po- tatoes are served in two-thirds of the homes. Most of them buy fresh potatoes. However, 13 to 18 per cent prefer proc- essed potatoes, usually frozen potato pat- ties and the dehydrated product for scalloped and au gratin potatoes. French fries are viewed differently. Fresh potatoes are preferred in only 56 per cent of the households serving french fries. Thus, California homemakers have a strong preference for fresh potatoes when they include potatoes in their meals. Frozen potatoes are used extensively for french fries and, to a much lesser extent, for potato patties. Dehydrated potatoes are purchased by 20, 14, and 10 per cent, re- spectively, by the families when they serve mashed, au gratin, and scalloped potatoes. There are locational differences among the survey families with respect to the percentage of those who serve potatoes in each of the eight ways and of those pre- ferring fresh over processed potatoes for each cooking preparation. These varia- tions are not large. There does seem to be a slight tendency, however, for potatoes to be served in each of the eight ways some- what more often by families in small cities than by those living in inland cities and the Bay Area. Users of fresh potatoes are more frequent in small cities when pota- toes are wanted for baking, boiling, and salad; in the Bay Area, when they are to be mashed; and in inland cities, for the four other uses. A consideration of the characteristics of users of processed potatoes may give a clue about the opportunities for expand- ing sales among nonusers. In the main, these follow the general patterns already [24] discussed. Some details and contrasts, how- ever, need to be mentioned (based on data represented in tables 7 and 8, and Appen- dix tables C-10 to C-13). Two items (instant mashed potatoes and frozen french fries) were purchased at least once by 67 per cent of the sample households and the other four by 20 per cent, on the average. For convenience, these two groups are referred to as the older and newer processed potato items in this discussion of user characteristics. The proportion of users was positively correlated with income, wife's education, and husband's occupation. Over the in- come and educational scales used for this study, users increased from 50 to 80 per cent of the families for the two older prod- ucts and from 10 to 30 per cent for the newer items. The proportion rose from 66 to 79 per cent for the older items and from 14 to 28 per cent for the newer ones in going from service and labor to profes- sional occupations. In conformity with the general patterns already described, the proportion of users was higher among white families, larger families, and those where the wage earner was middle-aged. White families had tried processed potatoes more often than had nonwhites — 70 compared to 49 per cent for the older products and 22 com- pared to 11 per cent for the newer items. Users represented 75 per cent of larger families and 53 per cent of small families for the older products compared to 22 and 15 per cent for the new items. The older items were tried by 77 per cent of the fami- lies aged 35-44 and by 53 per cent of those 55 and older. For the newer prod- ucts, the percentages were 27 and 15 for families in the two age groups. There were, of course, differences among the individual processed potato products. For example, the difference in the proportion of users among white and nonwhite families was about the same for frozen french fries as for instant mashed potatoes. The percentage of users varied much more for frozen french fries, how- ever, when families were grouped by each of the other characteristics: income, edu- cation, occupation, age, and size (see Ap- pendix table C-10). Convenience rice products Only 77 (6.5 per cent) of the sample house- holds had ever purchased frozen rice mixes. Since the number of users is so small, meaningful comparisons are diffi- cult to make and this item is omitted from the discussion. The limited data gathered, however, does strongly suggest that user characteristics for frozen rice mixes are similar to those for the other three rice products. Two-thirds of the households had tried instant rice compared to 57 and 44 per cent for flavored rice mixes and parboiled rice. Our comparison of user characteris- tics is largely in terms of contrasting in- stant rice with the other two rice products. Locational differences were negligible between inland cities and small cities. However, families living in the Bay Area had tried each rice product in substan- tially different proportions. More had pur- chased parboiled rice at least once (53 versus 39 per cent), while fewer had used the other two items — 60 versus 70 per cent for instant rice and 49 versus 61 per cent for flavored rice. The positive correlation of the percen- tage of users to income and education was fairly similar for the three items, especially with respect to the increase over the range of income and education used. However, some differences might be mentioned. Most of the increase in the proportion of users of instant rice and flavored rice occurred over the income range from under $3,000 to S7,000-$9,999, whereas the proportion continued upward for parboiled rice. Users of the three products increased in much the same way over the educational scale, except that the proportion was rela- tively low for instant rice bought by wives who continued their education into high school but did not graduate. Noticeable differences showed up among users when classified occupationally. For parboiled rice, percentages of users were 33 in service and labor occupations, 42 in the crafts occupation, and 52 among both clerical and professional families. In con- trast, about 57 per cent of the families from the first three groups (service-labor, crafts, and clerical) had purchased flavored rice compared to 69 per cent among pro- [25] fessional families. Instant rice was in an intermediate position. The percentage of users rose more or less steadily through the occupational groups from 58 among service and labor families to 76 for profes- sional families. All three rice products had been pur- chased more frequently by white than nonwhite families: 73 versus 42 per cent for instant rice; 62 versus 36 per cent for flavored rice; and 46 versus 32 per cent for parboiled rice. Some significant differ- ences existed among minority ethnic households. Users of the three products were almost identical among Negroes and Orientals — the proportion ranged from about 39 to 43 per cent for both groups. Mexican-Americans had tried instant rice to the same extent but only 25 and 17 per cent had ever purchased flavored rice and parboiled rice. In conformity with the general patterns for all the convenience food studied, the proportion of users of the three rice prod- ucts varied in a curvilinear fashion with age and size of the household. For each product, users were about equally frequent among families in the 35-44 and 45-55 age brackets. Substantially fewer younger and older families were users. These products had been tried more often by households with 3-4 persons than by larger or smaller families (see table 10). Convenience macaroni products For convenience, the five macaroni prod- ucts are put into two groups according Table 10 PERCENTAGES OF USERS FOR CONVENIENCE RICE PRODUCTS Characteristic Age of main earner Under 35 35-44 45-54 55 and over Size of household 1-2 persons 3-4 persons 5 or more In- Fla- Par- stant vored boiled rice nee rice Aver- age Per cent 51.8 68.9 71.4 63.9 63.3 72.0 64.6 57.2 41.5 63.5 51.5 63.2 51.5 45.3 39.6 47.7 36.9 64.1 49.1 59.0 45.0 50.2 61.3 62.0 49.6 49.3 61.7 56.2 to use frequency by the sample households. About half of the families have tried packaged macaroni or noodle dinners (56 per cent) and canned macaroni or spa- ghetti (50 per cent). Only one-quarter have purchased the other three: frozen macaroni or spaghetti dinners (25 per cent), packaged noodles almondine (24 per cent), and packaged beef stroganoff (21 per cent). The discussion is oriented mainly to comparing user characteristics for these two product groups, the older and newer macaroni products, respec- tively. Although there are, of course, dif- ferences among the individual items, the similarities are stressed. Locational differences prevailed but the pattern was not clear cut. Generally, one- quarter fewer families in one part of the survey had purchased a given item than was the case elsewhere. Users were less frequent (by this proportion) in the Bay Area for packaged macaroni or noodle dinners, in the Bay Area and inland cities for canned macaroni or spaghetti, and in inland cities for the three newer products. There were no sharp differences between older and newer products with respect to user characteristics associated with income, education, or occupation. Over the income range (from under $3,000, to $15,000, and over), the proportion of users increased from 36 to 60 per cent of the families for the two older products, and from 13 to 37 per cent for the three newer ones. The number of users increased to the same extent in going from families with only an elementary education to those with some college — from 40 to 59 and from 10 to 33 per cent, respectively, for the two product groups. Occupational differ- ences were small between families of ser- vice and labor workers and craftsmen on the one hand or between those of clerical and professional classes on the other. For all five products, variations in users were confined mainly to differences between "blue collar" and "white collar" classes. For the older products, the per- centages of users Avere 54 and 61 for the two groups of families, respectively, com- pared to 20 and 30 for the newer products. User frequency was correlated by a curvilinear relationship to age of the main [26] earner. Families in the 35-44 age group used the five products most frequently, followed in order by families in the 45-54 age group, those under 35, and, finally, the oldest group (55 and over). The older products were purchased by 61 per cent of the families in the 35-44 bracket and newer products by 28 per cent. The pro- portion of users decreased to 42 and 18 per cent, respectively, for the two product groups for families where the main earner was 55 years or older. These products had been tried less frequently by small families (1-2 persons) than by others. For the two older products, users represented 43 per cent of the small families compared to percentages of 57 for 3-4 person families and 61 for larger ones. The user pattern was less clear for the newer products. But, on the average for the three items, the percentages were 21, 25, and 24, respectively, for small, medium, and large families. APPENDIX A: DEFINITION OF TERMS Meanings of several demographic fac- tors were modified somewhat for this con- sumer survey. The precise definitions adopted are listed below together with such supplemental comments as appear ap- propriate. Household unit: A person living alone or a group of persons living together (in a house, an apartment or other group of rooms, or a single room which constitutes a housing unit) who pool their incomes and draw from a common fund for their major expenses. This is a modification of the BLS definition of a family or consumer unit. Thus, the sample includes one-per- son households and those with a widow, widower, or a divorced person, but ex- cludes a group of students or others living together in a boarding house or coopera- tive. Income level: The respondent was asked to indicate the yearly income of his house- hold by identifying one of six income cat- egories listed on a card: Income level I: under $3,000 Income level II: $3,000-$5,499 Income level III: $5,500-$6,999 Income level IV: $7,000-$9,999 Income level V: $10,000-$ 14,999 Income level VI : $ 1 5,000 and over Ethnic group: After completing the inter- view, the enumerator classified the house- hold ethnically without asking the re- spondent to identify the family's ethnic background. Hence, the classification rep- resents the enumerator's opinion about the respondent (and other persons of the household that may have been seen). Since only three households were classed as In- dian, these were included with Mexican- Americans and only four ethnic groups are used: 1 . Negro 2. Oriental 3. Mexican — Mexicans-Americans and Indians 4. White — Caucasians except Mexican-Americans Educational level: The respondent was asked to state the last grade of school com- pleted by the husband and by the wife. For most purposes four educational levels are used: 1. Elementary — completed 1-8 years of school. 2. High school — completed 9-1 1 years of school. 3. High school graduate — completed 12 years of school. 4. College — completed one or more years at any collegiate institution of higher learning. Study or training at a business or technical school was not counted toward attendance at college. Occupational class: The respondent's de- scription of the occupation of the husband (or main income earner if there was no husband in the household) provided the basis for putting the household in one of the occupational categories defined by the U. S. Bureau of the Census (1960 Census of Population, Classified Index of Occu- [27] pations and Industries). Six occupational classes are established: 1. Professional — professional, technical and kindred workers, officials, plus farm and nonfarm managers and pro- prietors. 2. Clerical — clerical, sales, and kindred workers. 3. Crafts — craftsmen, foremen, kindred skilled workers, and semiskilled op- eratives. 4. Service and labor — service workers, private household workers, and labor- ers (including those on farms and in mines). 5. Retired — retired workers. 6. Other — military personnel, home- makers, students, and unemployed persons. Employment of wife: The wife in the household (if any) is classified as 1. Employed, if she was part of the labor force, working either full time or part time for pay. 2. Homemaker, if she was not employed regularly outside the home. Household size: Replies to "How many persons are there currently living in your household?" identify three family-size groups: 1. 1-2 persons 2. 3-4 persons 3. 5 or more persons Age of main earner: Replies to "What is the age of the main earner?" specify four age groups for the sample families: 1. Under 35 years 2. 35-44 years 3. 45-54 years 4. 55 years or more Note: See Appendix table B-l for list of survey cities and Appendix table C-10 (or text table 7) for an enumeration of the 15 convenience foods covered by the survey. APPENDIX B: THE CONSUMER SURVEY' It was expected that the information secured by this survey of households in Northern California would be affected by variations in demographic factors such as ethnic classification, level of income, oc- cupation of the main earner, education of the husband and wife, size of family, etc. It was assumed, furthermore, that some consumer activities, preferences and atti- tudes might vary substantially from com- munity to community. Hence, it was de- cided to make interviews in households of varying socioeconomic classes in several cities of different population size rather than conducting a more intensive study in only a few metropolitan markets. Sample design A modified two-stage area sampling method was adopted. The first stage con- sisted of the selection of the cities sur- veyed; the second of the selection of house- holds to be interviewed within each city. Figure 1 shows the geographical distribu- tion of these survey cities. The number of the sample households and their loca- tion are listed in table B-l. Sample cities. The survey was limited to the portion of California north of the Tehachapi Mountains. However, house- holds located in 25 northern counties (in- cluding all of those along the eastern border of the State) were excluded as were those in rural areas and in population cen- ters with less than 2,800 inhabitants. Thus, interviews were made in the nonrural area north of the Tehachapi Mountains and west of the Sierra Nevada foothills. This survey region is referred to as Northern California in our study. Three different urbanization segments were chosen. Bay Area includes the cities in the urbanized territory of the five Bay Area counties, centered around three population-concentration areas: San Fran- 9 The same survey was used to obtain a better understanding of consumers' attitudes towaid, and the use of, credit cards. Results of this part of the inquiry are reported in DeLoach and O'Rourke, 1969. [28] cisco (plus nearby Daly City and Bayshore), East Bay (consisting of Richmond, San Pablo, El Cerrito, Albany, Berkeley, Em- eryville, Piedmont, Oakland, Alameda, and San Leandro), and the San Jose region (comprised of San Jose, Santa Clara, Campbell, Sunnyvale, and Mountain View, plus adjoining urbanized areas). Other urbanized areas are represented by the four inland cities of Sacramento, Stockton, Fresno, and Bakersfield, plus sur- rounding populated environments. Small Table B-l THE SAMPLE: NUMBER AND LOCATION OF SAMPLE HOUSEHOLDS; POPULATION IN SAMPLE AREA Sample city* Bay Area San Francisco East Bay San Jose Total Bay Area IrJand Cities Sacramento Stockton Fresno Bakersfield Total Inland Cities . Small Cities Areata Eureka Cloverdale Healdsburg Santa Rosa Petaluma Napa Paso Robles Atascadero San Luis Obispo. . Coastal Yreka Weed Dunsmuir Redding Anderson Red Bluff Corning Willows Chico Prradise Oroviile Upper Interior Yuba City Marysviile Olivehurst Woodland Davis Yeeaville Fairfield Middle Interior 1960 Population 1,000 Persons 740.3 754.3 358.8 1,853.4 191.7 86.3 133.9 56.8 468.7 5.2 28.1 2.8 4.8 31.0 14.0 22.2 6.7 6.0 20.4 141.2 1.8 3.2 11.5 9.6 4.8 13.5 8.9 10.9 15.0 74.2 Per cent of 1950 95 100 291 139 122 146 163 140 140 122 220 148 173 136 163 138 174 144 148 147 118 127 126 + + 147 118 137 120 % 114 154 146 122 135 144 251 344 480 192 1960 Households (1,000) 292.0 257.8 105.1 654.9 66.6 27.7 42.5 18.1 154.9 1.5 9.5 .9 1.7 10.9 4.7 7.3 2.2 1.6 6.9 47.2 1.6 1.1 1.0 4.4 1.3 2.3 1.1 1.3 4.9 3.3 2.2 24 5 3.7 3.2 1.4 4.3 2.6 3.1 4.1 22.4 Sample sizet Drawn 200 200 200 600 160 70 100 70 400 14 138 6 6 16 8 6 94 18 14 6 18 18 18 18 110 Used 144 99 178 421 133 54 100 54 341 9 24 6 6 28 9 20 9 10 14 135 10 6 6 14 6 8 6 6 17 8 6 93 15 14 6 12 17 18 17 99 [29] 1960 Pop ulation 1960 Households (1,000) Sample sizef Sample city* 1,000 Persons Per cent of 1950 Drawn Used Ratio* Manteca Tracy 8.2 11.3 36.6 4.4 9.1 14.4 10.1 11.9 6.8 112.8 399.9 2,722.0 217 134 210 188 146 138 101 137 122 155 158 121 2.5 3.5 12.3 1.3 3.1 4.3 3.2 3.3 1.9 35.4 129.5 939.3 10 14 30 6 18 12 10 12 6 118 460 1,460 6 14 25 5 16 12 10 9 6 103 430 1,192 Ceres Turlock Madera 291 Total Small Cities 332 GRAND TOTAL 127 * Nearby urbanized territory is included in the Bay Area and in the inland cities, t Sample size is described by three figures which have the following meanings: Drawn — The number of households specified by the sample design. Used — The number of completed questionnaires obtained. Ratio — The number of completed questionnaires per 100,000 households (as of 1960). In five small cities (Yreka, Chico, Areata, Paso Robles, and Atascadero), extra interviews were made in con- nection with the training program and hence for these cities the number used exceeds the number specified. t Population data reported separately only since 1950 by geographical area indicated. Source: U. S. Bureau of the Census [1960] plus data on this survey. cities denotes the 37 other survey cities in- tended to represent the remaining area. They were selected by a cluster sampling of city groups. 10 For example, the North Coast includes two cities: Eureka with 28,- 000 people (in 1960) and Areata with 5,200. These 37 small cities had a 1960 popula- tion of 393,900. The five largest ranged from 37,000 to 20,000, eleven had 15,000 to 9,000, and the other 21 varied from 9,600 to 2,800. n Sample households. A set of dwellings was drawn randomly for each survey city. The first step in this selection consisted of specifying the sample street and block. A street map of the city was marked off with equally spaced parallel, vertical, and horizontal rulings giving about ten times as many intersection points as the number of dwellings to be selected. These grid points were numbered in sequence (from the upper left-half to lower right-hand corner and the required number was drawn at random. Each such selected grid point gave the approximate location of a sample street. Its exact location was deter- mined by taking the street lying nearest the grid point. The portion included be- tween the two intersecting streets (or other "natural" boundary, such as railroad, 10 Cluster sampling is a procedure for dividing the universe into several groups or clusters and then drawing a number of these clusters for the sample. The clusters in our case were groups of cities along (or near) preselected segments of principal highway routes. This method was used in preference to a random sample because it was assumed that the cost savings realized from the reduction in travel time for interviews more than offset the additional cost involved in using a larger sample which is needed for getting results of equal precision. 11 Sample households were randomized within each urbanization segment but were not distributed proportionately among the three segments. Presumably the sample information gives unbiased estimates of parameter values for the segments considered separately. Values for the entire region surveyed can be estimated better by using household numbers (or some other factor) as weights for combining sample values for the three segments than by merely "blowing up" the sample information. This refinement was not introduced even though the sampling ratio was less (maybe by one-third) in the Bay Area than elsewhere. Its omission is serious only in those cases in which the parameter estimate is much different for the Bay Area than for the other two segments. [30] tracks, river, end of street, etc.) lying on either side of the grid point specified the sample block of this sample street. The second step designated the sample dwelling for the sample street and block. A number from 1 to 10 followed by the letter E or U was assigned randomly to each sample street and block. The letter denoted the side of the street to be used, with "E" representing an even numbered address and "U" referring to an uneven (i.e., odd) numbered address. The digit de- noted the dwelling count on that side of the street going from lowest to highest numbers for the block. Thus, 3E referred to the third even number, 1U to the first uneven (odd) address, etc. 12 Finally, a location sheet was prepared to identify the sample dwellings for each city. Column 1 gave the grid point on the marked street map. Column 2 designated the sample dwelling (e.g., 3E). Columns 3 and 4 identified the street and particular block for this sample dwelling. These dwellings were listed (separately for each city) numerically by grid number rather than in the order randomly drawn. It was estimated that about 10 per cent of the households contacted would not re- sult in completed questionnaires even after substitutes were made because an adult was not home or refused to cooperate or the sample block drawn had no residences. This was a mere guess. It proved to be very ample. The number of sample dwellings to be selected from the various cities was not in proportion to the number of households. Relatively more dwellings were chosen for the sample from smaller cities. The sample had 92 dwellings per 100,000 households (in 1960) for the Bay Area, 258 for the four large inland cities, 261 for the eleven larger small cities, and 444 for the twenty-six small cities with fewer than 4,000 house- holds in 1960. Dwellings were drawn in pairs rather than individually as is often done. For each sample dwelling, a nearly paired household (selected in a definite manner) was included for a second interview. It was estimated that 50 per cent more inter- views could be completed per day by the pairing method because of the consider- able reduction in travel time. Each sample dwelling was assigned an uneven ques- tionnaire number and its paired dwelling the next integer so that statistical tests could be made to determine whether this pairing method was more efficient. This determination has not yet been made. Evaluation. No "callbacks" were made to contact housewives not home at the time of the initial call. Instead interviewers con- tacted a substitute dwelling. No evening or weekend interviews were attempted in an endeavor to include more working wives. Presumably this procedure resulted in an undersampling of households with housewives away from home a relatively large part of the day because of work or for other reasons. Since sample streets and blocks were drawn on an area basis, dwellings on larger lots had a higher probability of being selected. This means a bias toward lesser inclusion of families living in congested areas — i.e., generally, those with less edu- cation, lower income, etc. The selection of small cities was guided in considerable part by the geographic distribution of available enumerators. However, there is no reason for assuming that the information gathered from the cities used is different to any material 12 Procedures were developed for handling exceptions that might and did arise. Here are a few examples. If the designation was 41st Street between J and K. Streets whereas, in reality, K Street did not intersect 41st Street, the interviewer was instructed to use the block between J and L. If the sample duelling did not yield a completed questionnaire because occupants were not at home or refused to cooperate, the interviewer selected a substitute dwelling in a prescribed manner. Procedures were also specified for handling other exceptional situations: the selected block had no residences, all residences were on one side of the street, more than one family lived in a single dwelling, the address located a multiple dwelling, the number sequence for the dwelling exceeded the number of dwellings on that side of the street, the street had irregular intersections, etc. For unspecified exceptions, interviewers were asked to make a common sense adjustment and proceed since it was felt that giving rigid rules to cover all possible cases was unnecessary since a few small deviations probably would not be of statistical consequence. [31] extent from what would have been ob- tained for a sample taken more randomly. Data collection and analysis The questionnaire was constructed about three months before the survey began. It was checked by various personnel at the University of California to determine what changes, if any, were needed in the order of questions or in their wording. A pilot study was conducted in Davis and Sacra- mento mainly to obtain an estimate of the time required for completing an interview. The allocation allowed seemed ample. (A copy of the questionnaire is at the end of this appendix.) The interviews were made in October and early November, 1968. To achieve greater consistency in the interviewing process, a training session was held in September for those who were to supervise the field enumeration. It was devoted to explaining the objectives and design of the survey and the method to be used by enumerators for locating sample dwellings and for filling out questionnaires. The three-hour meeting also seemed to allow ample time for answering questions raised by those in attendance. Each field interviewer received a loca- tion sheet listing the dwellings to be con- tacted and a marked city map to be used for planning an orderly and efficient rOute to be followed in making interviews. She was cautioned to contact all assigned sam- ple and paired dwellings (or their sub- stitutes) since omissions in one section of the city might introduce a systematic statis- tical bias. She was instructed to enter ques- tionnaire numbers in the prescribed man- ner so that paired dwellings could be identified. The only check on the work of the in- terviewers was that done by field super- visors. A few minor discrepancies were dis- covered and corrected. Editing was limited to reviewing the questionnaires to remove inconsistencies and to coding the informa- tion for punch cards. One card was partly mispunched as to occupation and, hence, this observation is omitted from some tabulations. The chi-square test was used to deter- 13 Nonresponses are significantly different proportions at even the 0.005 probability level for three of the cross-classifications in table B-2: wife's education X husband's education, wife's mine whether the observed differences in results associated with variations in demo- graphic factors were due to random sam- pling fluctuations. Generally, the probabil- ity level of 0.05 was used as the boundary between significance and nonsignificance. It was not necessary, however, to apply the test in many cases since mere examination of the differences indicated that they were or were not sufficiently large to be statis- tically significant. The sample drawn Table B-2 (see note on page 37) classifies respondents by household characteristics. This information indicates that several factors are highly intercorrelated. For ex- ample, a great deal of assortative mating by education existed. This means that women tended to have a husband whose education ended with about the same grade as their own. Thus, of the women who completed eight or less years of school- ing, 68 per cent were married to men with the same limited education, while 81 per cent of those who went to college had husbands with some college education. Similarly, income was affected by various demographic characteristics. To illustrate, 57 per cent of the Negro and Mexican families had incomes under $5,500 and 6.5 per cent over $10,000 compared to 20 and 43 per cent, respectively, for white families. „. Table B-2 also reveals that nonresponses did not present a serious problem. Their number was quite small for some factors. For example, only 15 households (1.3 per cent of the 1,192 in the sample) were rot classified by the number of persons cur- rently living in the household. The figure was 3.3 per cent by ethnic group, about 4 per cent by occupation and the husband's education, and 9 to 12 per cent by the wife's education, income level, and age of the main income earner. More nonresponses were given in the two-way classifications. Generally, they were distributed randomly. In a few cases, however, a definite bias was indicated. For example, if the data are sorted by the wife's and husband's education, the pro- portion of nonresponses decreases steadily from elementary to college. 13 [32] We assumed the sample would be large error was 0.283 or 2.12 per cent of the de- enough to give results which would be rived mean. This means that for a sample good approximations for the true facts of this size, the chances are 2 out of 3 that describing the universe, whatever those the true mean lies within 2.12 per cent may be. To judge whether this was the of the observed mean and 19 out of 20 that case, the standard error was computed for it is within 4.24 per cent (i.e., two standard several means. 14 These computations sug- errors). The standard errors were 3.21 and gest that the sample size was adequate for 2.68 per cent of the sample means (5.90 giving reasonably reliable estimates of true and 4.34 times per month) for rice and means, particularly where the data were macaroni products. 15 Sample results had not distributed among an excessive num- about the same reliability for purchases of ber of cross-classifications. seven convenience foods (Appendix table Potatoes were eaten 13.33 times per C- 10) and were half as reliable for the other month (on the average), by the 1,192 sam- eight. 16 pie households. The computed standard Copy of Questionnaire 17 HOUSEHOLD SURVEY— NORTHERN CALIFORNIA, 1968 Selected Consumer Shopping and Credit Patterns Questionnaire Number We are interested in learning about consumer shopping practices and use of certain foods. May we ask you a few questions? PART I: We would like to ask you about your use of some specific foods — potatoes, rice, and macaroni products. education x occupation, and income x occupation. They are significant at approximately the 0.05 level for wife's education x ethnic group and income x wife's education. In all other cases, the computed value of chi-square is smaller than the tabulated value at the 0.10 level. In brief, variations in the number of nonresponses arise primarily from sampling fluctuations. 14 The standard error measures the precision of the sample mean by specifying the extent to which averages for new samples (of the same size) Avould be likely to vary solely because of chance fluctuations in sampling at random from the same universe. About 68 per cent of the sample means will lie within one standard error of the true mean, 95 per cent within two standard errors, etc. The standard error of the mean is computed as the standard deviation of the sample items divided by the square root of the sample size used for deriving the mean. 15 If sample means are computed separately for the three geographic areas, the standard errors are larger (by about 70 per cent), as is to be expected. Averages of the three standard errors (each expressed relative to the relevant sample mean) are 3.54, 5.41, and 4.63 per cent, respectively, for potatoes, rice, and macaroni products. Even for rice (the least reliable case), averages by areas lie within 10.8 per cent (twice 5.41) of the true means with 95 per cent confidence. 16 In computing average purchases (and standard errors) for these 15 convenience foods, weights of 0, 1, and 2 were used for the three answers (have not purchased, have purchased once, and have purchased more than once). On this basis, the standard errors (expressed as percentages of the sample means) were about 2.2 per cent for three foods (french fries, instant mashed potatoes, and instant rice), 3.0 per cent for four more (flavored rice, parboiled rice, canned macaroni or spaghetti, and packaged macaroni or noodle dinners), 5.9 per cent for seven others, and 11.6 per cent for frozen rice. It should be indicated that the weights used here were selected arbitrarily. However, another set (within reason), would give essentially the same comparisons. For example, using weights of 0, 1, and 3 increases the standard errors only slightly. [33] 1. How often do you eat potatoes at home? 1. More than once a day 4. Twice a week 2. Every day 5. Once a week 3. Every other day 6. Twice a month 7. Once a month 8. Seldom 9. Never 2. When you have potatoes at home what one form do you serve most frequently? 1. Mashed 4. French fried 7. Mashed, boiled, baked as frequently 5. Potato chips 2. Boiled 3. Baked 6. Au gratin 8. Mashed, baked as frequently 9. Other 3. Which one of these factors matters most to you in selecting fresh potatoes? 1. Price 5. The size of the package 2. Cleanliness of potatoes 6. The way you will use the potatoes 3. Freedom of the potatoes from defects 7. None of these 4. Whether they are packaged or not 8. Other 4. If you want to serve the potatoes in each of the following forms, which type of fresh potato would you select? Form White round White long Russet Red Doesn't matter Don't know Don't serve in this form Boiled Mashed Baked French fried Potato salad 5. If you want to serve potatoes in each of the following forms, which type of potato product would you buy? Form Fresh Frozen Instant or dehydrated Canned Don't serve in this form Other Boiled Mashed Baked French fried Au gratin Scalloped Patties Potato salad 17 This version differs somewhat from the questionnaire actually used. It omits questions on PART II (the section dealing with the distribution and use of customers' credit cards) and various rulings which were added to facilitate editing the questionnaires and preparing answers for punch cards. Also, the material is given in a more condensed form to permit pre- senting it on fewer pages. [34] 6. Do you use fresh potatoes more often than potatoes in a convenience form? 1. Yes 2. No 7. Do you think plain potatoes are fattening? 1. Very fattening 3. Not fattening 5. Don't know 2. Moderately fattening 4. No opinion 6. Other 8. As a change from potatoes which of the following would you serve first? 1. Rice 3. Macaroni or spaghetti 5. Nothing 2. Noodles 4. A salad or vegetable 6. Other 9. How often do you eat rice at home? 1. More than once a day 4. Twice a week 7. Once a month 2. Every day 5. Once a week 8. Seldom 3. Every other day 6. Twice a month 9. Never 10. When you buy rice is it usually 1. Plain regular rice 4. Brown rice 2. Parboiled or converted rice 5. Flavored rice mixes 3. Minute or instant rice 6. Other 11. How often do you eat macaroni, spaghetti or noodle products at home? 1. More than once a day 4. Twice a week 7. Once a month 2. Every day 5. Once a week 8. Seldom 3. Every other day 6. Twice a month 9. Never 12. Have you ever purchased any of the following products? If so, have you purchased them more than once? Do you have any in your home at this time? Product Have not purchased Have purchased once Have purchased more than once Have now Do not have now Have now Do not have now Dehydrated instant mashed potatoes Dehydrated scalloped potatoes Dehydrated au gratin potatoes Frozen french fries Frozen potato patties Frozen potato puffs Parboiled or converted rice Minute or instant rice Flavored rice mixes Frozen rice mixes Frozen macaroni or spaghetti dinners Canned macaroni or spaghetti Packaged macaroni or noodle dinners Packaged beef stroganoff 'ackaged noodles almondine [35] PART III: We would now like to get some general background information about your household. 1. How many persons are there currently living in your household? a. How many adults aged 19 or over? b. How many teenagers aged 13 to 18? c. How many children aged 12 or under? 2. What is the age of the main earner? a. What is the occupation of the husband (or main earner if no husband in house- hold)? (Note to enumerator — have respondent describe occupation.) b. Does he work full time or part time ? c. What is the occupation of the wife (include homemaker as an occupation)? (Note to enumerator — have respondent describe occupation.) d. If employed outside the home, does she work full time or part time ? 4. a. What was the last grade of school completed by the husband? b. What was the last grade of school completed by the wife? 5. Could you tell us which of these categories would best describe your yearly house- hold income? (Show card to respondent.) 1 . Under $3,000 4. $7,000-$9,999 7. No answer 2. $3,000-$5,499 5. $10,000-$ 14,999 3. $5,500-$6,999 6. $15,000 and over PART IV: For enumerator to answer — Do not ask respondent 6. Was the respondent 1. Caucasian 3. Oriental 5. Indian 7. Other 2. Negro 4. Mexican-American 6. Don't know 7. Was the interview completed in 1. Rural farm 3. Urban 2. Rural nonfarm 4. Large metropolitan area 8. Would you describe the area as 1. Low income 2. Middle income 3. High income 9. Was the residence 1. A private home 4. Mobile home 2. An apartment unit 5. Rooming or boarding house 3. A duplex or attached house 6. Other 10. Would you rate the appearance and upkeep of the house as 1. Superior 2. Standard 3. Inferior 4. Very inferior [36] APPENDIX C: STATISTICAL SUPPLEMENT This appendix consists of 13 large tables, the titles to which appear below. Photocopies of the tables, plus table B-2 may be obtained at their actual cost of reproduction ($1.50) from: Agricultural Publications University of California Berkeley, California 94720 Make checks or money orders payable to The Regents of the University of California. Send payment with order to the above address — not to the Regents. Photocopies of the tables may be seen at the following libraries: National Agricultural Library, Beltsville, Md. Agricultural Reference Library — Giannini Hall, University of California, Berkeley Bio-Agricultural Library — University of California, Riverside University of California Library, Davis Table Titles B-2. Household Characteristics of Survey Respondents Cross-Classified by Income Level, Education of Wife, and Ethnic Group, Northern California, 1968. C-l. U. S. Per Capita Consumption of Certain Foods, 1910-1965, Pounds Per Year. C-2. Utilization of U. S. Potatoes, 1956-1967 Crops. C-3. Views About Whether Potatoes Are Fattening. Northern California, 1968. C-4. Substitutes Used to Replace Potatoes in the Diet, Northern California, 1968. '-5. Frequency of Eating Potatoes, Rice, and Macaroni Products at Home, Northern California, 1968. C-6. Frequency of Serving Rice in Atlanta, Dallas, and Denver, 1957. C-7. Frequency of Using Potatoes, by Household Characteristics, Northern California, 1968. -8. Frequency of Using Rice, by Household Characteristics, Northern California, 1968. ,-9. Frequency of Using Macaroni Products, by Household Characteristics, Northern California, 1968. C-10. Purchases of 15 Convenience Potato, Rice, and Macaroni Products, by Household Characteristics, Northern California, 1968. C-l 1. Purchases of Eight Convenience Potato, Rice, and Macaroni Products, by Income Level and Ethnic Group, Northern California, 1968. C-l 2. Purchases of Eight Convenience Potato, Rice, and Macaroni Products, by Income Level and Education of Wife, Northern California, 1968. C-l 3. Purchases of Eight Convenience Potato, Rice, and Macaroni Products, by Income Level and Household Size, Northern California, 1968. Tables C-l, C-2, and C-6 contain secondary data taken from the sources indicated. Source references are omitted for the other tables since they summarize information gathered by the interviews conducted for this consumer survey. Terms used in this survey and these tables are defined in Appendix A. Replies to all questions were not given by all 1,192 respondents. The number of nonresponses is indicated on some tables. Nonresponses are omitted, however, when their number is already given elsewhere or can be determined easily from the data tabulated. When percentage distributions are computed, their sums do not always total 100.0 cactly because of rounding. The practice followed here was to modify the data for the components (slightly) to get "accurate" totals. [37] ACKNOWLEDGMENTS An investigation of this type, based on detailed analysis of hundreds of field interviews, requires the cooperation of many persons. The authors are indebted to all who helped. Three colleagues, professors of Agricultural Economics, University of California at Davis, gave us valuable assistance. D. B. DeLoach helped throughout the study and offered many suggestions regarding the analysis and presentation of the material. Sylvia Lane and G. W. Dean read our final draft and suggested improvements. Field work on the survey was done by county home advisors, community cooperators, and students from several California state colleges. This phase of the work was the primary responsibility of Gaylord P. Whitlock, Program Leader, Family and Consumer Sciences, Agricultural Extension Service, University of California, Davis. He also served as the liaison person between that organization and the Department of Agricultural Economics. Finally, this study would have been impossible without the active cooperation of those persons who generously furnished the detailed information requested for com- pleting the questionnaires. LITERATURE CITED Bennett, M. K. 1941. Wheat in national diets, Wheat studies of the Food Research Inst. Stanford: Stanford University, Food Research Inst., 18 (2): 37-72. 1968. Chain store age. 1968 Supermarket Sales Manual, Mid-July, pp. 215-17, New York: Lebhar-Friedman Publishers, Inc. DeLoach, D. B., and A. D. O'Rourke 1969. Use of bank credit cards in grocery stores. Berkeley: University of California. Agric. Exp. Sta. Bui. 843. Gray, Roger W., Vernon L. Sorenson, and Willard W. Cochrane 1954. An economic analysis of the impact of government programs on the potato industry of the United States. St. Paul: University of Minnesota, Agric. Exp. Sta., Tech. Bui. 211:8-21. Harp, Harry, and Denis Dunham 1961. Market potential for processed potato products. Washington, D.C.: U.S. De- partment of Agriculture, Econ. Research Serv., Marketing Econ. Div., Market- ing Research Rept. 505. Hiemstra, Stephen J. 1968. Food consumption, prices, and expenditures for 1909-1966. Washington, D.C.: U. S. Department of Agriculture, Econ. Research Serv., Agric. Econ. Rept. 138. 192 pp. LeBovit, Corinne, and Faith Clark 1956. Household practices in the use of foods, three cities, 1953. Washington, D.C.: U. S. Department of Agriculture, Agric. Research Serv., Agric. Inform. Bui. 146:59-64. Libeau, Clayton P. 1951. Marketing potatoes for consumer approval. Moscow: Western Regional Re- search Publication, Western Regional Potato Marketing Rept., University of Idaho, Exp. Station Bui. 285. Lionberger, H. F. 1968. Adoption of new ideas and practices, in Wasson, C. R., et al, Competition and human behavior. New York: Appleton-Century-Crofts, pp. 55-59. [38] Pratt, Parley M. 1960. Rice: domestic consumption in the United States. Austin: University of Texas, Bur. of Business Research. SWACKHAMMER, GENE L. 1964. A study of consumer acceptance and rejection of new food products, Syracuse, New York, 1963. Ithaca: Cornell University, unpubl. thesis for a Master of Science, February. U. S. Bureau of the Census 1960. U. S. Census of population: 1960, California, number of inhabitants and gen- eral population characteristics. Final Repts. PC (1)-6A and PC (2)-6B. U. S. Department of Agriculture 1948. Potato preferences among household consumers. Washington, D.C.: Miscell. Publ. 667. U. S. Department of Agriculture, Agricultural Research Service 1965a. Household food consumption survey, 1965-66, food consumption of house- holds in the United States. Washington, D.C.: Rept. 1, Spring. 19656. Household food consumption survey, 1965-66, food consumption of house- holds in the west. Washington, D.C.: Rept. 5, Spring. U. S. Department of Agriculture, Consumer and Marketing Service 1968a. 1969 spring potatoes: acreage marketing guides. Washington, D.C.: Publ. AM G-65, November. U. S. Department of Agriculture, Economic Research Service 19686. Rice situation. Washington, D.C.: RS-12, January. 5m-l,'71(Pl302L) J.F. •