Can u do a case analysis for me? I need it quick. Let me know.

Family Life Cycle Segmentation for Foodservice Marketing: An Exploratory Case Study

Robert E. Frash, Jr. John M. Antun

Harland E. Hodges

ABSTRACT. As U.S. families continue to allocate more of their food dollar to the foodservice industry, it becomes increasingly important to explore market segmentation strategies. Family life cycle (FLC) modeling, which examines the implications of lifestyle and expenditure patterns brought about by family role transitions, has demonstrated prior utility in hospitality and tourism. This exploratory research was the first to see if FLC staging was also related to consumers' specific dining preferences (e.g., reservations and food type). The data revealed, however, only a weak association between FLC staging and dining preferences. Nonetheless, when examined singularly, age, marital status, and children, which are FLC component variables, some were found to have potentially important

Robert F. Frash Jr., PhD, is an Assistant Professor in the Department of Hospitality and Tourism Management, School of Business and Economics at the College of Charleston.

John M. Antun, PhD, FC, RVIP, CHF, is an Assistant Professor in the Department of Retail, Hospitality, and Tourism Management, College of Education, Health, and Human Sciences at the University of Tennessee, 1215 W. Cumberland Ave- nue, 200A Jessie Harris Building, Knoxville, TN (F-mail: [email protected]).

Harland E. Hodges, PhD, is a Senior Instructor in the Department of Manage- ment and Decision Sciences, School of Business and Economics at the College of Charleston, 66 George Street, Charleston, SC 29424-0001 (E-mail: hodgesh® cofc.edu).

Address correspondence to: Robert E. Frash Jr., College of Charleston, 66 George Street, Charleston, SC 29424-0001 (E-mail: [email protected]).

Joumal of Foodservice Business Research, Vol. 11(4) 2008 © 2008 by The Haworth Press. All rights reserved.

382 doi: 10.1080/15378020802519728

Frash, Antun, and Hodges 383

roles in suggesting particular guest preferences. These findings should help foodservice operators better understand the most efficient methods for gauging their patrons' particular dining preferences.

KEYWORDS. Family life cycle, segmentation, expenditure patterns

INTRODUCTION

The foodservice industry is forecast to cross the half-trillion dollar sales threshold in 2007. It is estimated that the average restaurant will spend roughly 2% of this revenue, over $10 billion, on marketing dis- bursements (National Restaurant Association, 2006). Targeting those expenditures with cogent consumer segmentation is essential to yielding optimal return. The family life cycle (FLC) concept, which captures life- style and expenditure pattern differences brought about by family role transitions, has been shown to be a practical marketing research design (Murphy & Staples, 1979; Wells & Gubar, 1966).

FLC staging, previously evidenced to be well suited to tourism seg- mentation because of the discretionary nature of its expenditures (Cai, Hong, & Morrison, 1995; Lawson, 1991), should similarly be well paired with discretionary restaurant spending behavior. In fact, the literature sup- ports a connection between demographic characterization (e.g., family size, and age of adults/children) and food purchases made away from home (Cai et al., 1995; McCracken & Brandt, 1987). The importance of this connection continues to grow as Americans are now estimated to spend nearly 50% of their food dollar away from home, which is up from 25% in 1955 (National Restaurant Association, 2006).

To date, though, FLC research in the tourism and hospitality literature has principally focused on leisure travel and vacation expenditure patterns (Bojanic, 1992; Cai et al, 1995; Hong, Fan, Palmer, & Bhargava, 2005; Lawson, 1991). Although this might address foodservice operations peripher- ally, no research has yet explored FLC staging implications for consumers' explicit restaurant dining preferences. This study seeks to address this lacuna in the research through an exploratory investigation of the relationship between F'LC staging and diners' preferences, including restaurant category, service style, food type, reservation's policies, and others. The current study should help foodservice operators to better understand their patrons' expecta- tions and to improve their ability to target particular products and services.

384 JOURNAL OF FOODSERVICE BUSINESS RESEARCH

LITERATURE

Family Life Cycle (FLC)

Murphy and Staples (1979) rightly point out that the family is one of the great features of American society. Accordingly, FLC has become one of the most established concepts in marketing segmentation and consumer research (Lawson, 1988). Dominguez and Page's (1981) research on cor- rect use of social stratification technique affirms that FLC has particularly important contributions to make in market segmentation, supporting its common use today.

The concept of identifying FLC transitions was first employed by soci- ologists back in the 1930s (Derrick & Lehfeld, 1980). Rowntree (1903), an English sociologist, is credited with initially conceptualizing the FLC. In Figure 1, Rowntree shows how the relationship between the head of household's income and the needs of the family produces a life cycle refiecting alternate periods of "want" and "plenty." Using the head of household as proxy, Rowntree's life cycle begins at birth and is then defined around a series of decisive occasions, including first job, marriage, impact of children, and retirement. This notation of behavioral changes, corre- sponding to critical life events, is central to the FLC concept and bears

FIGURE 1. Rowntree's life cycle of head of household income (adapted from Lawson 1988, p. 16).

Begins to earn

Marries Children begin to earn

Chiidren ieave home and marry

Retires

Primary Poverty Line

AGE

Frash, Antun, and Hodges 385

general support in the literature (Andreasen, 1984; Dominguez & Page, 1981; Lawson, 1988).

Thus, FLC is a multidimensional construct accounting for age, mar- riage, the presence of children, and implied income shifts. As Lawson (1988) reports, FLC ". . . reflects many ofthe basic needs and constraints that influence the purchasing patterns of a family over its history. With regard to income it is implicit in concept that absolute levels of income will rise over the life cycle unit until retirement. However, the differences between the stages are especially important in describing that amount of income which is likely to be uncommitted after meeting tue basic needs of the family and is therefore available for expenditure on more discretion- ary items."

FLC modeling first began to be found in marketing research in the mid-1950s. Lansing and Morgan (1955) studied income, assets, and hous- ing expenditures using a seven-stage FLC model based on age and marital status of the head of household, as well as the age of the youngest child. They found that the pattern of family income followed an inverted U-shape; increasing over the life cycle, peaking when there were depen- dent children still at home.

Wells and Gubar's (1966) FLC model augmented Lansing and Morgan's (1955) staging to include the employment or retirement of the head of household. In brief, the nine family role transition stages in Wells and Gubar's (1966) FLC model are designated bachelor stage, young, single people of either sex; newly married stage, young couples with no chil- dren; three/«// nest stages—married couples, of various age cohorts, with dependent children under or over 6 at home; two empty nest stages— employed or retired mature married couples with no dependent children at home; two solitary survivor stages—employed or retired mature single peopleof either sex.

FLC Staging

Kotier and Lillien (1983) suggest that a market segmentation model should be as parsimonious as possible. That is, the model should produce homogenous categories, maximize between-group variance, classify the majority of households, and yield a reasonably small number of adequate size categories (Schaninger & Danko, 1993). Achieving these goals became more convoluted as the 1970s ensued. Family demographics began to change as a result of increasingly more divorced parents and older childless couples, creating "nontraditional" family households. Some

386 JOURNAL OF FOODSERVICE BUSINESS RESEARCH

were concerned (Murphy & Staples, 1979) that these nontraditional family profiles would render traditional FLC staging less effective because note- worthy market segments might be left out as unclassifiable.

Indeed, tourism researchers, employing traditional FLC taxonomies, met with troubling classification issues. Lawson's (1991, p. 13) use of Wells and Gubar's FLC model resulted in almost 40% of the sampled tourists being unclassifiable. Small numbers of qualifying respondents in certain categories forced Fodness (1992) to collapse Wells and Gubar's original nine stages into five. In response, modernized family life cycle models (Gilly & Enis, 1982; Murphy & Staples, 1979) were created. Murphy and Staples (1979) conceptualization of the FLC included 13 stages to address these nontraditional family profiles.

Bojanic (1992) reported significantly improved efficacy with this mod- ernized approach when he added single parents and middle-aged couples without children to Wells and Gubar's (1966) staging. These two classifi- cations represented a substantial portion of his sample (15.3%). Bojanic's (1992) modernized approach yielded only 11% unclassifiable, which is a marked improvement over others using traditional FLC approaches. Of note, Bojanic (1992) suggested that subsequent research utilize all 13 stages of Murphy and Staple's (1979) modernized FLC design. However, Lin and Lehto's (2006) more recent attempt to incorporate all 13 stages was less than successful. Small cell sizes forced them to collapse Murphy and Staple's modernized taxonomy down to nine categories.

Food Service and FLC

Thus far, discussions in the literature of restaurant dining preferences relating to FLC staging have been dealt with indirectly, under an umbrella of tourist behavior. Discussions of restaurant dining behavior focused only on categorizing FLC stages by vacation food purchase amounts or the general importance of foodservice quality versus other vacation attributes; for example, tourists' entertainment venues and outdoor activity options.

Lawson (1991) examined the relationship between FLC and tourist behavior in New Zealand in the 1980s. Lawson constructed FLC stages around the essential scheme devised by Wells and Gubar (1966). The sample was based on questionnaires completed at airports, as visitors left the country. FLC staging was engaged over other segmentation schemes (e.g., age, sex, or purpose of visit) because it was thought to provide mul- tidimensional criteria that addressed New Zealand's international and multicultural mix of tourists. Significant relationships were found between

Frash, Antun, and Hodges 387

the tourists' stage in the FLC and elements of vacation expenditures, which included meal spending. Tourists in the full nest stages spent sig- nificantly more on restaurant purchases. This is not surprising, however, because it is likely, first, that tourists eat more meals out and, second, full nesters with children would have more persons dining than other stages, such as bachelors, empty nesters, or solitary survivors without children.

Bojanic (1992) examined vacation attributes preferred by U.S. resi- dents when traveling overseas. Questionnaires were mailed to individuals living in major metropolitan areas throughout the United States. Respon- dents reported on trip behavior and preferences regarding destination characteristics. These attributes were then correlated to FLC stages. Bojanic added single parents and middle-aged couples without children to Wells and Gubar's (1966) basic FLC model to include nontraditional households in the segmentation. Bachelors reported thinking nightlife and entertainment activities more important than food service and lodging. Middle-aged couples without children also did not consider food service and lodging as most important. Instead, their top priorities were the his- torical appeal of the vacation destination and the ability to experience local customs. However, all other stages included food service and lodg- ing in their most important travel considerations. In general, it seems that food service and lodging were more important to those with children than those without dependent child responsibilities.

Cai, Hong, and Morrison (1995) utilized secondary data from the 1990 Consumer Expenditure Survey to investigate how vacation spending behavior for tourism products varied by changes in household and family characteristics. FLC staging was employed, as well as other social class, cultural, and geographic independent variables. Dependent variables included expenditures on food, lodging, transportation, and sightseeing/ entertainment. Cai et al. used the head of household as proxy to segment the FLC into age groupings, married and unmarried, as well as house- holds with and without dependent children. The 25 to 34 age group dem- onstrated a negative relationship with vacation food purchases, tending to spend less than those over 65 years of age; married households spent more on food than unmarried households. The number of children was shown to have a negative relationship with food purchases. Cai et al. theorized that this reflected the time constraints nictitated in caring for more children.

Corresponding to Cai et al. (1995), Hong, Fan, Palmer, and Bhargava (2005) utilized secondary data from the 2000 Consumer Expenditure Survey to examine the impact of FLC on travelers' spending patterns. Bojanic's

388 JOURNAL OF FOODSERVICE BUSINESS RESEARCH

(1992) adaptation of the Well's and Gubar FLC model was used for stag- ing. Hong et al. found that marriage seemed to have had a positive impact on vacation food purchases. The married with no children stage and the empty nester stage spent more than those in the singles stage on food. It was also found that single parents do not spend significantiy different amounts, statistically, than other FLC stages with dependent children (i.e., the full nest stage).

In summary, although some evidence is available on vacation food pur- chase behavior, no research was found seeking to target FLC staging onto restaurant dining preferences. However, the tourism behavioral literature clearly supports the ability to identify specific market segments through FLC categorization. As such, the fT..C concept should help to better illustrate consumer behavior patterns and psychological preferences for restaurant attributes, products, and services.

Theory and Application

Families now spend 48% of their food dollar dining out, which represents a near doubling over the last 50 years (National Restaurant Association, 2006). When one considers this prodigious growth, it follows that the foodservice industry needs to better target consumers' dining predilections. Because of this study's seminal and exploratory nature, explicit hypotheses about FLC stage dining preferences are not proffered. However, it is theo- rized that FLC staging will be statically associated with one or more of consumers' dining preferences. The researchers conjectured, for example, that those falling FLC stages with younger adults might be less likely to attend fine dining restaurants because of their nascent income. Or, that those FLC stages with children might be more adverse to restaurants that do not prohibit smoking.

The literature supports that FLC staging has successfully allowed retail marketers to discriminate their promotions across family role transitions. This study hopes to find evidence of utility in the FLC concept for the foodservice industry. If so, FLC segmentation might provide foodservice marketers more targeted information about their guests' preferences.

METHODOLOGY

With consideration of the sample demographics and previous procedural approaches (Bojanic, 1992; Lin & Lehto, 2006; Murphy & Staples, 1979),

Frash, Antun, and Hodges 389

this research operationalized a modified modernized approach, which loosely parallels Bojanic's (1992) FLC staging. Bojanic's eight stages were bach- elor, newly married, full nest 1 & 2, empty nest, solitary survivor, single parent, and middle-aged couples without children. The term "bachelor" was replaced by "young single" because it is gender specific when the stage criterion is not. Further, as Schaninger and Danko (1993) suggested, the stages were modified so that the model herein produces homogenous categories, maximizes between-group variance, classifies the majority of households, and yields a reasonably small number of adequate size catego- ries. To that end, single parent, solitary survivor, and middle-aged couple without children were replaced with a newly defined stage, mature single. Table 1 contains classification criteria and frequency percentages for each of the six stages used in the study.

This seminal study of FLC segmentation strategies for foodservice marketers draws on data outcomes from an annual survey sponsored by the restaurant association of a moderate sized city in the southeast region of the United States. The city was thought particularly appropriate for this study because of its restaurants' reputations. It was recently listed as one of the top 10 global culinary destinations by Travelocity (Gaines, 2007) and forecasted to be in one of the top five restaurant growth regions in the United States for 2007 (National Restaurant Association, 2006). Pedestri- ans were interviewed throughout the downtown area, where many of the city's restaurants are located. Though care was taken to gather a random set of responses, collecting data from passersby on the street typically will not result in a true random sample. The method for this study, however, seems sufficient because the principle aim was to obtain information on the respondent's behavior to illustrate the potential use of value-based

TABLE 1. Modernized FLC criteria and frequency percentages

stage

Young single Newly married Full nest 1 Full nest II Empty nest Mature single

Age

18-34 18-39 18-39 40-49

50 and over 35-55

Marital status

Single Married Married Married Married Singie

Children

No No Yes Yes No No

Frequency %

23.8 13.8 15.3 10.0 13.2 10.9

Note. 10.6% were unclassifiable.

390 JOURNAL OF FOODSERVICE BUSINESS RESEARCH

proflles for restaurant marketers, rather than to provide point and interval estimates of population parameters (Blose & Litvin, 2005; Calder, Phillips, & Tybout, 1981).

Each person was asked to respond in the context of their evening "dinner" meal. Responses, therefore, exclude breakfast or lunch behavior. Survey items were created through a series of meetings with a Delphi panel (a.k.a. focus group) of local restaurant proprietors to ensure content validity. A similar version of the survey had been employed by the local restaurant association, annually, over the last 3 years; albeit, without the inclusion FLC staging variables. Criteria needed for FLC classification items were established given the FLC literature. The survey included both categorical response items, given in Table 2, and Likert-type importance scale items, given in Table 3. There were 20 items in the survey overall, and on average it took less than 5 minutes for respondents to complete. Data collection

TABLE 2. Categorical Delphi panel survey items and response options

Categorical items Response options

Restaurant category Fine dining, Upscale casual, Dinnerhouse, Grill-buffet, Family, Fast-casual, Fast-food

Pricing arrangement A la carte, Inclusive Food group Steak, Seafood, Chicken, Vegetarian, Pasta, Other Ethnic food type American, Italian, French, Mexican, Asian, Mediterranean, Other Ambient noise level Loud teeming restaurant. Quiet intimate restaurant

TABLE 3. 7-Point scale of importance, Delphi panel item survey items

Scale items

How important is the restaurant's service? (e.g., have attentive service staff, serve food timely, get your order correctly)

How important is the restaurant's food quality? (e.g., tastes good, nutritionally balanced, wholesome)

How Important is the restaurant's atmosphere? (e.g., décor, lighting, temperature, color) How important is it that the restaurant provides value? (i.e., the price matched your expec-

tation for what you received) How important is it that the restaurant does not allow smoking in the dining room? How important is It that the restaurant use local ingredients? (e.g., local produce, shrimp,

oysters) How important is it that the restaurant provides live music? How important is it that the restaurant accepts reservations?

Frash, Antun, and Hodges 391

was completed over a 5-day period. No inducements (e.g., gifts) were offered to respondents for participation.

SPSS 11 was used to perform all statistical analyses. Nominal alpha for all analyses was p < 0.05. A series of descriptive analyses, including fre- quencies, means, and standard deviations, were conducted for all mea- sures of the dependent variables. Diagnostics were performed to assess assumptions of normality, homogeneity of variance, and independence. Two-way contingency table analysis and ANOVA was conducted to examine the relationships among categorical and scale survey items, respectively. Post hoc comparisons were conducted when the ANOVA F statistic proved significant. In order to gain a deeper understanding of the underlying constructs, FLC component variables (i.e., age, marital status, and children) were also analyzed.

RESULTS AND DISCUSSION

Sample Characteristics

When compared to U.S. Census Bureau statistics (2005), the average (n = 362) respondent's social demographics create a profile emblematic of a U.S. citizen. The sampled respondent's age was normally distributed with a median age of 35 years. Fifty-two percent were married. Thirty-seven percent reported having dependent children at home. Gender was represented nearly equally (52% male). The most frequently reported household income bracket was $40,000 to $59,999. About 81% were White/Caucasian, 11% were Black/African American, 2% were Hispanic/Latino, and 2% were Asian American. Roughly half the respondents were visitors, with 47% reporting living more than 50 miles away.

About 61% reported purchasing dinner away from home at least twice weekly. According to the Senior Director of Information Services & Library at the National Restaurant Association, this purchase behavior exceeds the average U.S. adult, who goes out to dinner at a sit-down res- taurant 1.4 times a week. The greater than average foodservice purchase behavior is attributed to the fact that about half of the respondents were visitors.

Almost 70% of the respondents reported eating these meals at restaurant categories within the sit-down/full-service sector. Seventy-five percent preferred inclusive pricing, in which a salad and a side dish is included with the entrée, versus á la carte pricing, where only the entrée is included.

392 JOURNAL OF FOODSERVICE BUSINESS RESEARCH

American food was the most selected (31.5%) ethnic food category. Steak, seafood, and chicken were all comparably chosen as the food group most ordered when dining out (= 29% each). There was no clear preference as to ambient noise levels, with about 49% preferring a loud, teeming atmosphere, versus the rest, who prefer a quieter, more intimate setting.

Means, standard deviations, and medians for the scale survey items of relative importance are given in Table 4. Scale item reliability was sound; the value for coefficient alpha was 0.87. Among the classically critical areas when dining out (food quality, service, and atmosphere; Lockyer, 2005), food quality was of greatest consequence. Considering the 7-point scale, the mean (m = 3.61) for the importance of meal value was surprisingly dispassionate. Additionally, respondents appear to be more concerned about a restaurant's smoking and use of local ingredients policies than they do about live music or reservations policies.

Dining Preference Segmentation by FLC

Given the heretofore mentioned $10 billion in annual marketing outlays, the foodservice industry's need for fecund market segmentation strategy is salient. The intent of this research was to address these needs through an examination of the relative utility of FLC staging, as a segmentation tool, across a variety of consumers' dining preferences. It was theorized that because the FLC had been shown a useful correlate with leisure

TABLE 4. Means, standard deviation, and medians for scale survey items

Item

Service standards Food quality Atmosphere Meai value Smoking poiicy Local ingredient use Live music Reservation's poiicy

Mean

4.68 5.95 3.08 3.61 4.14 4.24 2.61 3.46

Std. dev.

1.92 1.83 1.89 2.08 2.46 2.07 1.68 2.07

Median

4.00 5.00 3.00 3.00 4.00 4.00 2.00 3.00

Wofe. Statistics are in reference to a 7-point Lii<ert scaie of Importance.

Frash, Antun, and Hodges 393

behavior, the concept might similarly hold for dinning behavior. Although important, plausible, and significant relationships were found between FLC stage and certain dining preference variables, the number of statistical associations was not as robust as anticipated.

Of the categorical variables, FLC staging was related to the restaurant's ambient noise level. Examination of the cell frequencies indicated that the young single stage preferred a louder and more boisterous dining atmo- sphere, whereas the empty nest stage favored quieter, more intimate settings. This finding seems reasonable because of the disparate ages and tradi- tional sociological patterns of the two FLC stages (Carter & McGoldrick, 1999; Schaefer & Lamm, 1995).

One-way analysis of variance was conducted to evaluate the relationship between FLC and the scale items. Significant relationships were found among consumers' preferences about the restaurant's smoking policies and reservations policies. Tukey post hoc comparisons revealed that going to a restaurant that does not allow smoking was more important to the empty nest stage than to the young single stage, with a mean difference of 1.56, on the 7-point scale. The follow-up tests also indicated that the accep- tance of reservations is more important to the full nest II stage than to the young single stage, with a mean difference of 1.64.

Dining Preference Segmentation by FLC Components

Considering the results of statistical analyses, it appeared that the young single stage had more pronounced dining preferences than other FLC stages. This caused the researchers to consider possible derivations. That is, might one of the components on which the multidimensional FLC con- cept is comprised (age, marital status, or children) be playing a particu- larly significant role? As such, each component variable was examined to see if one was more singularly indicative of consumer dining preference than the overarching FLC multidimensional construct.

The post analyses results were intriguing. Indeed, one of the three FLC component variables, when considered individually, was related to twice as many consumer dining preferences than the full FLC model. Age was significanUy related to 6 of the 13 dining preference dependent variables. To conduct the subsequent examination of the age variable, the respondent's years of age were categorized into 10-year groupings (18-29, 30-39, 40-49, 50-59, and 60-(-). Cross tabulation was again utilized for categorical vari- ables. As might be expected, and consistent with FLC stages inclusive of these age brackets, those in the 18 to 29 group preferred loud atmosphere

394 JOURNAL OF FOODSERVICE BUSINESS RESEARCH

TABLE 5. Relevant statistically significant associations between independent and criterion variables

Independent variable

FLC staging FLC staging FLC staging Age Age Age Age Age Age

Criterion variable

Ambient noise ievel Smoking poiioy Reservation policy Ambient noise levei Pricing arrangement Meai vaiue Smoking policy Reservation policy Use of iocai ingredients

Method

Contingency table ANOVA ANOVA Contingency tabie Contingency tabie ANOVA ANOVA ANOVA ANOVA

Test statistic

X'(5,N = 335) = 26.10** '=(5, 330) = 3.74* ^(5,330) = 4 . 2 7 * - X {3,/V= 335) = 2 9 . 9 1 * * X (3,/V=337) = 7.93**

'̂ (3, 337) = 3.67** F{5, 333) = 3.21 ** F(5, 334) = 6.72*^

' Ï 5 , 332) = 3.77**

*p < 0.05. **p< 0.01.

when dining, whereas those in the two 50-plus groups preferred quiet sur- roundings. Those in the 18 to 29 group also distinguished themselves with a clear preference for inclusive pricing.

Among the continuous variables, one-way analysis of variance revealed four significant relationships between age grouping and dining prefer- ence. Post hoc tests found that the 18 to 29 group was significantly less concerned with meal value than the two age groups falling in the 50-plus grouping. Moreover, similar difference between the youngest and two oldest groups were found with smoking policies, reservation acceptance, and the use of local ingredients. All these were more important to the older age groupings, with mean differences ranging from 1.65 to 2.45.

To facilitate the readers understanding of relevant results, significant asso- ciations between independent variables (i.e., FLC staging and age) and crite- rion variables (i.e., select dining preference variables) are given in Table 5.

With regard to the marriage and children FLC component variables, they did not yield much insight about consumers' dining preferences. Although ANOVA revealed significant relationships between select din- ing preference variables and both marital status and dependent children, mean differences were negligible at less than 0.50.

CONCLUSIONS

If one reviews the literature, it is clear that the family life cycle (FLC) concept has proven itself to be a prolific market segmentation strategy for

Frash, Antun, and Hodges 395

more than half a century. Tourism research also reveals FLC to be effec- tive in predicting vacation purchase behavior. However, the findings of this research did not support a similar robustness of the FLC concept when looking at the connectedness of consumers' dining preferences. Only three significant relationships were evidenced. That said, the researchers are not suggesting FLC should not have a role in subsequent foodservice market segmentation endeavors. It would be unwise for the foodservice industry to fully discount its use in the future. Sample size constraints could well have masked the ability to find more significant FLC correlations. There appeared, for example, to be a relationship between FLC and two other preference variables (i.e., restaurant category and food group) but expected cell counts fell below acceptable levels. That is, more than 20% of the cells had less than five (Green & Salkind, 2003).

Also, the small number of significant findings does not discount the utility of what was found herein, with regard to FLC segmentation. Given the results of this research, foodservice marketers would be wise to target advertising of restaurants with upbeat lively atmospheres to those in the young single FLC stage. Those restaurants with a more reserved atmosphere and those that do not allowing smoking should target their promotions to consumers in the empty nest stage. Further, restaurants that accept reservations would be wise to focus on those falling in the full nest II stage.

What is perhaps the most important outcome of this research, however, are the numerous correlations found between an individual's age and the consumers' dining inclinations. As noted previously in this article's review of the literature, market segmentation approaches should be as parsimoni- ous as possible (Kotier & Killien, 1983). It is indeed surprising that the well-acknowledged FLC concept was less informative of dining behavior patterns than age. However, age is a far simpler, more straightforward concept to measure. Effectively, though informal, a restaurateur could walk through his or her restaurant and gain a sense of his or her guests' ages simply by the color of their hair. Should future research support the strong association between age and dining preference found here, such segmentation efficiencies might make future foodservice marketing efforts more cost effective. Indeed, the results of this research warrant further investigation into market segmentation strategies for the foodservice industry. The better the foodservice industry can understand their individual guest's needs, the easier it will be to target their products and services, and most importantly, exceed those needs.

396 JOURNAL OF FOODSERVICE BUSINESS RESEARCH

REFERENCES

Andreasen, A. R. (1984). Life status changes and changes in consumer preferences and satisfaction. Journal of Consumer Research, 11(3), 784-795.

Blose, J. E., & Litvin, S. W. (2005). Social values and restaurant patronage. Tourism Review International, 8(4), 311-321.

Bojanic, D. C. (1992). A look at modernized family life cycle and overseas travel. Journal of Travel & Tourism Marketing, 7(1), 61-80.

Cai, L. A., Hong, G.-S., & Morrison, A. M. (1995). Household expenditure patterns for tourism products and services. Journal of Travel & Tourism Marketing, 4(4), 15-41.

Calder, B. J., Phillips, L. W., & Tybout, A. M. (1981). Designing research applications. Journal of Consumer Research, 8(3), 197-207.

Carter, B., & McGoldrick, M. (1999). Overview: The expanded family lifecycle: Individual, family, and social perspectives (3rd ed.). Needham Heights, MA: Allyn & Bacon.

Derrick, F. W., & Lehfeld, A. K. (1980). The family life cycle: An alternative approach. Journal of Consumer Research, 7(2), 214—218.

Dominguez, L. V., & Page, A. L. (1981). Use and misuse of social stratification in consumer behavior. Journal of Business Research, 9(2), 151-173.

Fodness, D. (1992). The impact of family life cycle on the vacation decision-making process. Journal of Travel Research, 31(2), 8-14.

Gaines, J. (2007). Travelocity highlights 10 delicious destinations for foodies in 2007. Retrieved June 29, 2007, from http://news.travelocity.com/phoenix.zhtml?c=75787&p= irol-newsArticle&ID=974387&highlight

Gilly, M. C , & Enis, B. M. (1982). Recycling the family life cycle: A proposal for redefi- nition. In A. A. Mitchell & A. Arbor (Eds.), Consumer Research (Vol. 9, pp. 271-276). Ann Arbor, MI: Association of Consumer Research.

Green, S. B., & Salkind, N. J. (2003). Using SPSS for Windows and Macintosh: Analyzing and understanding data (3rd ed.). Upper Saddle River, NJ: Prentice Hall.

Hong, G.-S., Fan, J. X., Palmer, L., & Bhargava, V. (2005). Leisure travel expenditures pat- terns by family life cycle stages. Journal of Travel & Tourism Marketing, 18(2), 15-30.

Kotier, P., & Killien, G. L. (1983). Marketing decision making: A model building approach. New York: McGraw-Hill.

Lansing, J. B., & Morgan, J. M. (1955). Consumer finances over the life cycle. Consumer ße/iovior, 2(1), 36-50.

Lawson, R. (1988). The family life cycle: A demographic study. Journal of Marketing Management, 4(\), 13-32.

Lawson, R. (1991). Patterns of tourist expenditures and types of vacation across the family life cycle. Journal of Travel Research, 29(4), 12-19.

Lin, Y.-C, & Lehto, X. Y. (2006). A study of female travelers' needs trajectory and fam- ily life cycle. Journal of Hospitality & Leisure Marketing, I5(\), 65-87.

Lockyer, T. (2005). The dining experience: Critical areas of guest satisfaction. Journal of Hospitality and Tourism Management, 12(\), 50-63.

McCracken, V. A., & Brandt, J. A. (1987). Household consumption of food-away-from-home: Total expenditure and by type of food facility. American Journal of Agricultural Economics, 69(2), 274-284.

Frash, Antun, and Hodges 397

Murphy, P. E., & Staples, W. A. (1979). A modernized family life cycle. Joumal of Consumer Research, 6{\), 12-22.

National Restaurant Association. (2006). 2007 Restaurant industry forecast: What's next. Washington, DC: Author.

Rowntree, B. S. (1903). Poverty: A study of town life. London: Macmillan. Schaefer, R. T., & Lamm, R. P. (1995). Sociology (5th ed.). New York: McGraw-Hill. Schaninger, C. M., & Danko, W. D. (1993). A conceptual and empirical comparison of alter-

native household life cycle models. Joumal of Consumer Research, 79(1), 580-594. U.S. Census Bureau. (2005). Families and living arrangements. Retrieved July 5, 2007,

from http://www.census.gov/population/www/socdemo/hh-fam.html Wells, W. D., & Gubar, G. (1966). Life cycle concept in marketing research. Journal of

Marketing Research, 3{4), 355-363.