Grocery Shopping Lists: What Do Consumers Write?

Susan Spiggle, The University of Connecticut
ABSTRACT - This paper presents an analysis of 129 grocery lists containing a total of 2854 items. Approximately 255 of all items were brands, most were products, few items were product classes, and one was a need. The percent of brands on lists varied across consumers. Ethnicity was the only consumer characteristic related to percent of brands. The store at which lists were collected was related to percent of brands independently of ethnicity of its customers. Inquiry into the construction of shopping lists is suggested as a potentially fruitful research area.
[ to cite ]:
Susan Spiggle (1987) ,"Grocery Shopping Lists: What Do Consumers Write?", in NA - Advances in Consumer Research Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT : Association for Consumer Research, Pages: 241-245.

Advances in Consumer Research Volume 14, 1987      Pages 241-245

GROCERY SHOPPING LISTS: WHAT DO CONSUMERS WRITE?

Susan Spiggle, The University of Connecticut

ABSTRACT -

This paper presents an analysis of 129 grocery lists containing a total of 2854 items. Approximately 255 of all items were brands, most were products, few items were product classes, and one was a need. The percent of brands on lists varied across consumers. Ethnicity was the only consumer characteristic related to percent of brands. The store at which lists were collected was related to percent of brands independently of ethnicity of its customers. Inquiry into the construction of shopping lists is suggested as a potentially fruitful research area.

The shopping list of a young, single male with an annual household income between $10,000 and $19,999, employed in a service occupation, shopping regularly two to three times a week contains the following items.

diet Pepsi-2 six packs       toothpaste

Milk                                  bandaids

bread                                shoe strings

deodorant                          T.V. dinners

cigarettes                           Spam

T.V. Guide                        mac & cheese

beans                                Old Spice

Ragu                                 Zest

spaghetti                           lunchmeat

pretzels

This list represents a typical combination of products and brand names and provides us with some insight into the consumption patterns of its author.

Consumer researchers have ignored the content of shopping lists as an area of inquiry, even though they have considered use of lists as an indicator of preshopping planning. Shopping lists represent the codified purchase intentions of consumers. Their content can suggest the extent to which consumers concretize their purchase intentions as they construct their lists, or articulate needs, the fulfillment of which remains open to choice.

The research reported here presents an analysis of the content of the lists of 129 grocery shoppers. The investigation is exploratory and reports results from a preliminary stage or an investigation of shopping styles and planning. It is directed primarily at the content of shopping lists and whether variations in content are ewident and, if so, can they be explained by consumer characteristics or marketing strategies of stores.

BACKGROUND

Consumers commonly shop for groceries with lists of items for purchase. The use of a list represents some degree of preshopping planning along with reading grocery advertisements and reduces impulse purchasing for purchase transactions exceeding fifteen items (Kollat and Willett 1967, Miller and Zikmund 1975). As Kollat and Willet (1967) note the extent of preplanning may be enumerated into five types of purchase decision making--(1) both product and brand decision made prior to store visit; (2) product decision made prior to visit, but not the brand; (3) product class decision made prior to visit, not product or brand, (4) consumer need recognized, but not product class, prior to visit; and (5) need not recognized prior to store visit

These five types represent different degrees of susceptibility to in-store stimuli However, they are not necessarily activated in a hierarchical manner Spotting a display of cookies might trigger need arousal for dessert, but once that need were aroused, the consumer might automatically purchase Peppridge Farm cookies. On the other hand, the same customer might spend several minutes deciding on which type and brand of convenience entree to purchase, having arrived at the store to purchase something quick for dinner The items on shopping list indicate the degree of preshopping planning and may be consumer needs, broad product classes, product categories, brand names, or some other categories

The existence of a brand name on a shopping list can represent one of several circumstances Many shoppers list brands when they have coupons for the brand, or the store is running an advertised special In both cases the consumer is responding to the short term promotional efforts of a firm or retailer and choice is an outcome of recent information search and brand evaluation.

Other items are indicated by brand name because consumers find it easier, more convenient, or more natural to write the brand--Mop and Glow instead of floor cleaner with wax. Some brand names represent strong brand preferences where the item is written by the shopper or someone other than the shopper Brands listed for these reasons represent the consumer's response to the long term promotional efforts of the firm and indicate a truncation of the consumer deliberation process--the elimination of the information search and brand evaluation stages

In some cases brands may represent the product category as may be the case with 'band-aids' on the list reproduced. The consumer may not intend to purchase that brand, but uses it as a surrogate for the product category. Such use indicates a response to the long term marketing efforts of a firm, but not in a way that marketers desire

When product categories are written on lists they may reflect a delay in the brand decision. Examples are where the consumer lists a product, but plans to search for information on alternatives in the store, or locate a sale or coupon item, or Just a different brand for the sake of variety Consumers may list products when they have actually made a brand decision A product may be listed, but it stands for a brand to which the consumer is loyal, as when ketchup on the list means Heinz, or a brand which the consumer passively repurchases without a commitment to it In both cases, the product name is used, but the brand decision has been made

Finally consumers may list broad product classes-vegetables, dessert--or consumption needs--dinner stuff While it is possible that a consumer who writes such items may have a product and/or brand in mind, it does not seem likely It suggests that decision making has been deferred to allow mood and in-store stimuli to play a role in choice.

While analysis of the mere content of a grocery list, cannot demonstrate the degree of response to the long and short term marketing efforts of firms and grocery retailers, nor the degree of consumer preshopping planning, it does provide some ewidence of the labels which consumers attach to their purchase intentions A review of articles over the past twenty years in Journal of Marketing, Journal of Marketing Research, Journal of Consumer Research, and Advances in Consumer Research revealed no articles investigating the content of consumers' shopping lists, or suggesting the content of lists as a source of data for studying consumer behavior Two articles describing techniques for studying in-store customer behavior in retail settings failed to mention the existence of shopping lists, much less their content (Applebaum 1951; Granbois 1968).

Three studies using shopping lists in the research design were located. Kollat and Willett (1967) used shopping lists in their study of grocery impulse purchases. The content of shopping lists was used to assess purchase intentions as part of the measurement of impulse purchases They used the presence or absence of a shopping list as an independent variable accounting for extent of impulse purchasing. The other two studies used the content of shopping lists as part of a research design to demonstrate the efficacy of projective techniques and to assess social definitions of products Haire's (1950) study and its replication (Webster and Von Pechmann 1910) used two shopping lists constructed by the researchers as experimental stimulus variables These 'shopping list' studies provided no information about the content of actual consumers' shopping lists

The purpose of the present analysis is to provide preliminary data about the content of grocery shopping lists as a first stage in researching the dynamics of their construction. The basic research questions in this study are: (1) What are the relative frequencies of brand names, products, product classes, consumer needs, or undecipherable and miscellaneous other symbols contained on shopping lists? and (2) Are there variations across consumers in these relative frequencies and, if so, can they be explained by consumer characteristics or marketing strategies of stores?

RESEARCH DESIGN

Sample and Data Collection

No attempt was made to gather a systematic sample of a particular population since this research is exploratory. Rather, four grocery stores were picked in different locations as sites to intercept shoppers. On seven occasions at different times and on different days of the week grocery shoppers waiting in check out lines were asked if they had used a list in their shopping If they replied yes, they were asked if they would fill out a brier questionnaire and give up their list in the interest of research about what consumers write on grocery lists. Less than five percent of those approached who indicated they had used a list refused to participate. Shoppers with less than seven items and those obviously engaged in reading, conversations, or otherwise were not approached. A total of 129 usable lists and questionnaires were collected over a three week period, 51 from stores in three different chains and 78 from a single Super Stop and Shop, a recently opened, one-stop shopping, grocery, drug, and general merchandise store which is Dart of a strong regional chain

Each item on the shopping lists was coded by product category and whether it was a brand name, product name, product class, private label, need, or other Extraneous items such as "Pick up Bobby at Boy Scouts at 4:00" were ignored. Questionnaire data included basic demographic and socioeconomic data and number of shopping trips per week normally made It also included a Likert scaled question on self perception of extent of brand loyalty for ten grocery product Categories and in general

Analysis

Items from the shopping lists were aggregated by store from which they were collected The percentage of brands, products, product classes, private labels, and needs were computed for the lists as a whole and brands by product category. A Fisher's Exact test was used to test statistical significance of differences in the percentage of brand names by product category between lists gathered from Stop and Shop and those from the other stores.

The percentage of total items which were brand names was calculated for each respondent Means for percentage of brand names on lists and for brand loyalty self perception were calculated separately for categories of sex, marital status, ethnicity, age, household income, number of income earners, occupation, household size, number of shopping trips per week, and store where data was collected An ANOVA procedure was used to calculate statistical significance for difference in the means. Differences in consumer characteristics between stores were tested by Chi-square

RESULTS

Shopping List Content

The 129 lists yielded a total of 2854 shopping items for an average of 22 items per list. The shortest list contained five items, while the longest contained 112 The modal number of items was 17.5. of the total shopping items 25.7% were brand names. Only nine items were private labels, and 49 were product classes One need was listed--"baby needs". Eight of the items were not decipherable.

The percentage of brand names varies by product category as one would expect. Product categories with few brand alternatives such as meat and fish and produce have very small percentages of brands. Within product categories where a number of brand alternatives exist, the percentages are much higher, and they vary considerably. Table 1 presents the percentage of brands by product category for all shopping items and for those on lists collected at Super Stop and Shop and the other stores.

The differences in percentage of the brand names between lists obtained at Stop and Shop and those from other stores suggests different potential explanations One is that the store's marketing strategy--its product assortment, its quality levels in different product lines, its carrying a private label, or its sales promotional strategy induces consumers to list fewer or more brands. Another potential explanation is that consumer characteristics, socio-demographic or shopping orientations, account for these differences. To assess the efficacy of these competing explanations the characteristics of consumers shopping at Stop and Shop versus those at other stores were compared.

Results of Consumer Characteristics

A comparison of the socio-demographic characteristics of Stop and Shop respondents versus those from the other stores revealed no differences in age, household income, number of wage earners, size of household, or occupation. However, the ethnic composition of respondents from the two samples was significantly different. The Stop and Shop respondents were 98.7% (77) White and 1.3% (1) Black with no Hispanics. The respondents from other stores were 76.5% (39) White, 19.7% (11) Black, and 3.9% (2) Hispanic (X2 = 16.9, P < .01). Differences in the percentage of brand names on lists from stores may then be a function of the subcultural affiliations of their customers.

TABLE 1

PERCENTAGE OF BRANDS BY PRODUCT CATEGORY FOR ALL ITEMS AND BY STORE

The mean percentage of brands on the lists of the three subcultural groups varied considerably with Whites at 21.5%, Blacks at 35.9% and Hispanics at 57.0% (F = 8.6, P <.001). The respondent numbers were very limited for Blacks and Hispanics with 11 and 2 respectively. Thus, their small sampling argues for caution in interpreting these figures as estimates of the mean number of brands per list for these two subcultural groups. The magnitude of the differences does suggest that there may likely be subcultural differences in relative incidence of brand names on shopping lists.

Since the use of brand names results from different shopping strategies and sensitivities to long versus short term promotional efforts, the differences found across the subcultural groups may be explained by these consumer behaviors. While the present study lacks data on subcultural variations across all or the potentially relevant consumer behaviors, such as deal proneness and coupon usage, data was collected on self perception of brand loyalty across product categories. There were no significant differences in brand loyalty self perception between ethnic groups for bakery products, canned goods, beer, household cleaning products, pet products, or in general. Table 2, however, reports differences in self perception of brand loyalty measured on a five point Likert scale anchored by 1 as very and 5 as not at all for rive product categories.

TABLE 2

BRAND LOYALTY SELF PERCEPTION BY SUBCULTURE

Blacks consistently considered themselves to be more brand loyal than Whites across these five product categories The greater incidence of brand names on Blacks' shopping lists could be a function of their stronger brand loyalties, consistent with previous findings (Wellington 1981).

Results of Store Differences

In order to remove the effects of the differential ethnic composition of the stores as a source of variation in the shopping list content and brand loyalty self perception, these two variables were compared for respondents from Stop and Shop versus other stores for Whites only Table 3 demonstrate 3 that differences exist in brand loyalty for three product categories and for percentage of brands on shopping lists between Stop and Shop, Beits (a local chain), and two other stores.

Table 3 indicates that the brand loyal self perceptions for some product categories and the number of brands on consumer shopping lists vary between stores when ethnic composition is controlled. While the brand loyalty self perceptions and percent of brands on lists are similar between respondents from Stop and Shop and Beits, they differ widely from the respondents at the two other stores Thus, the store's marketing strategy appears to have an effect on consumer tendency to use brand names independently of its customers' characteristics

TABLE 3

BRAND LOYALTY SELF PERCEPTION AND PERCENTAGE OF BRANDS ON LIST BY STORE FOR WHITES

SUMMARY OF FINDINGS

Data collected from shoppers at four stores indicated that the items on grocery shopping lists are approximately 25% brand names. The percentage of brand names varies from none to 68.9% with a standard deviation of 16.9% and mode of 17.5% Variations in the percent of brand names occurs across consumers, between stores and across product categories. Half or more of the items in the categories of beer, cereals, household cleaning products, and personal care products are written as brand names. Less than two percent of the items on lists were product classes, needs were virtually nonexistent , with the majority of items being product categories.

Variations between stores could not be explained by the socio-demographic composition of their customers with the exception of ethnicity Blacks and Hispanics used higher percentages of brand names on lists than Whites and indicated higher levels of brand loyalty for certain product categories. No other socio-demographic characteristics were associated with use of brand names or brand loyalty. Respondents from different stores demonstrated differing tendencies to use brand names on lists when the effects of ethnic composition were removed

Limitations

There are two sources of limitations to the interpretations made of this data (1) The limited number of Black and Hispanics in the sample made analysis of the interaction between store shopped, ethnicity, and other consumer characteristics impossible Also their limited numbers suggests caution in using the brand loyalty self perception data and percentage of brand names on lists as estimates for Blacks and Hispanics This data does, however, suggest that ethnic subcultural affiliation is related to use of brand names on lists and brand loyalty and invites an investigation to explain this.

(2) The use of a brand name on a shopping list may represent intense brand loyalty--Coke for soda, use of a brand to represent a product category--Kleenex, or deal proneness--inclusion of brands on sale and those for which the consumer has coupons These are different types of consumer behaviors which reflect very different responses to the marketing efforts of firms. They cannot be distinguished by mere enumeration of the percent of brands. Thus, while the percent of brands on lists does not vary across the socio-demographic categories, except for ethnicity, the relevant consumer behaviors which are reflected in use of brand names may vary

While this research is exploratory and has limitations, it suggests that variations in the content of grocery store shopping lists may reflect different types of shopping orientations. Several hypotheses for future research are suggested by the findings reported here.

H1 The content of shopping lists varies by consumer characteristics, specifically ethnicity.

H2 The content of shopping lists varies across grocery retailers, reflecting competitive strategies.

Further research should attempt to discern the extent to which items on lists reflect prestore decision making as an indicator of shopping orientations and susceptibility to the long and short term promotional efforts of firms.

DISCUSSION

The analysis of 129 shopping lists has several implications of interest to consumer researchers and marketers. If we assume that the vast majority of brands on lists represents a brand choice (as opposed to brand used as surrogate for product category), then at least one fourth of grocery store purchases involve prestore decision making That a substantial percentage of brands should be listed is not surprising, given the large marketing expenditures of the food and package goods industries and their promotional strategy of preselling products. The attempt to create intense brand loyalty to the firm's brand or to dislodge established loyalties to and purchase patterns for other brands through coupons and advertised specials are designed to shape consumer choice prior to shopping and insulate consumers from in-store stimuli. As some consumers write products on lists even where brand choice is made, we may assume that the relative frequency or preshopping brand choices are even greater

The relative frequency of brands on shopping lists is also indicative of the pervasive impact of commercialism on American culture. Recent research has demonstrated the increasing incorporation of brand symbols in cultural products from novels, plays, popular songs and comic art (Friedman 1985a and 1985b; Spiggle 1986). Artists are obviously not attempting to solve consumption problems, nor planning purchases when they incorporate brand symbols in their work Rather, they use brand symbols to identify, satirize, idealize, or concretize the reality of their art. However, consumers and artists are both involved in activities in which they symbolize objects in the process of externalizing consciousness.

For both the artist communicating to an audience and a grocery list maker indicating a purchase intention to self or household shopper, brand symbols represent widely known, stable, concrete objects whose images are constructed by advertising (Cf. Schudsen 1984). In the absence of brands, only symbols of a more generic nature would be available for use- their referent objects more abstract, inexact, and specific attributes less precisely definable. For the consumer the existence of brands results in the availability of a wide range of choice objects to satisfy needs. At the same time brands permit consumers to truncate the information search and alternative evaluation stages of choice, or to simplify information processing in these stages.

Models of consumer choice and marketing principles emphasize that consumers do not seek products or brands, but seek to fulfill needs, solve problems, and realize the intangible benefits provided by market offerings. The virtual absence of needs on shopping lists and the dearth of product classes suggests that consumers may, in fact, seek products and brands. That is, for package goods and grocery products consumers may experience need arousal (Assael 1984) or problem recognition (Engel, Blackwell, and Miniard 1986), the process of purchase initiation in which consumers become aware of deprivation, as a defined need for a specific product or brand. Thus, it may be that consumer behavior is driven more by pursuit of concrete products and offerings than by diffuse motivations and needs whose goal objects are ill defined in the initial stage of purchase choice.

Whether or not consumers are loyal to specific brands, the marketing efforts of grocery product and package goods firms has enabled consumers to equate needs, for which precise, widely shared symbols, labels, and cultural images are absent, with concrete objects-market offerings for which these are present. Thus, the consumer when thirsty may actually experience the need as one for Coke, and not as a general motivated state whose goal object may be one of many product types, or more specifically, not even as a caffeine based, sweet, carbonated readily available chilled can of soda.

As Pollay (1985) has demonstrated, advertising has undergone a cycle in which print ads became more focused on consumer benefits, portrayed more use situations, and used more emotive rhetorical styles in the 1950s, followed by a resurgence in the 1960s and 1970s of ads with the product only, focusing on product attributes and using rational, informative textual styles. In an era in which the marketing concept is firmly established we might expect both consumers and marketers to focus on consumer needs and benefits. There is ewidence from content analysis of advertisements and shopping lists that a continuing product orientation exists on both sides of exchange transactions.

REFERENCES

Applebaum, William (1951), "Studying Customer Behavior in Retail Stores," Journal of Marketing 16 (October), pp. 172-178.

Assael, Henry (1984), Consumer Behavior and Marketing Action 2nd edition, Boston: lent. Engel, James F., Roger D. Blackwell, and Paul W. Miniard, (1986), Consumer Behavior, Hinsdale, Illinois: The Dryden Press.

Friedman, Monroe (1985a), "Are Americans Becoming More Materialistic? A Look at Changes in Expressions of Materialism in the Popular Literature of the Post-World War II Era," in Advances in Consumer Research, Vol 11, eds Morris B. Holbrook and Elizabeth C Hirschman, Ann Arbor, MI: Association for Consumer Research, 385-387.

Friedman, Monroe (1985b), "The Changing Language of a Consumer Society: Brand Name Usage in Popular American Novels in the PostWar Era," Journal of Consumer Research, 11 (March), 927-938.

Granbois, Donald H. ( 1968) s "Improving the Study of Customer In-store Behavior," Journal of Marketing 32 (October), pp . 28-33.

Haire, Mason (1950), "Protective Techniques in Marketing Research," Journal of Marketing 14 (April), pp. 649 - 56.

Kollat, David B and Ronald P Willett (1967), "Customer Impulse Purchasing Behavior," Journal of Marketing Research 4 (February), pp. 21-31.

Miller, Stephen J and William G. Zikmund (1975) "A Multivariate Analysis of Prepurchase Deliberation and External Search Behavior," in Mary J. Schlinger (Ed.), Advances in Consumer Research, Vol. 2, Ann Arbor: The Association for Consumer Research, pp. 187-96.

Pollay, Richard W. (1985), "The Subsiding Sizzle: A Descriptive History of Print Advertising, 1900-1980," Journal of Marketing, 49 (Summer), pp.24-37.

Schudson, Michael (1984), Advertising, The Uneasy Persuasion: Its Dubious Impact Upon American Society, New York: Basic Books.

Spiggle, Susan (1986), "Measuring Social Values: A Content Analysis of Sunday Comics and Underground Comix," Journal of Consumer Research, 13 ( June), pp.100-113.

Webster, Frederick E. Jr. and Frederick Von Pechmann (1970), "A Replication of the 'Shopping List' Study," Journal of Marketing 34 (April), pp. 61-65.

Wellington, Alphonzia (1981), "Traditional Brand Loyalty," Advertising Age (May 18), p. S-2.

----------------------------------------