Sociology

Workplace Flexibility Christensen, Kathleen, Schneider, Barbara

Published by Cornell University Press

Christensen, Kathleen & Schneider, Barbara. Workplace Flexibility: Realigning 20th-Century Jobs for a 21st-Century Workforce. Ithaca: Cornell University Press, 2010. Project MUSE., https://muse.jhu.edu/.

For additional information about this book

Access provided at 7 Aug 2019 22:22 GMT from University of California @ Berkeley

https://muse.jhu.edu/book/27549

57

3

COMING TOGETHER AT DINNER A Study of Working Families

Elinor Ochs, Merav Shohet, Belinda Campos, and Margaret Beck

Family mealtimes have received considerable attention in the popular media as a barometer of family well-being. Dinnertime, they report, is an endangered or de- funct family ritual that has given way to the demands of parents’ work and chil- dren’s extracurricular activities (RMC Research Corporation 2005). In the United States, as in other societies, the family dinner is viewed as an icon of the family and an ideal toward which contemporary families should strive. Cultural expectations that the family should eat a healthy home-cooked meal together present a challenge to working parents, whose finite time and energies are too often expended on work- place demands.

This study examines the extent to which, and how, working parents and their children manage to come together to share an evening meal. Eating together is a primordial form of sociality, often defining boundaries for a social group (Du- mont 1980; Lévi-Strauss 1963, 1969; Malinowski 1935). Across communities and historical time, sharing a meal has constituted an important locus for reifying and structuring social order and for social sense-making. Beyond fulfilling the body’s bi- ological need for nourishment, mealtimes are vital cultural sites that are “historically durable, yet transformable, socially organized and organizing, and tempospatially situated arenas, which are laden with symbolic meanings and mediated by material artifacts” (Ochs and Shohet 2006, 35). Distinctions in food types and manner of preparation, along with meal duration, locale, participation, communication, and demeanor are historically rooted, conventionalized, and morally evaluated. Indeed, as Norbert Elias (2000 [1939]) observes, table manners have been viewed as critical to good breeding and “the civilizing process” since at least the Middle Ages. For this reason, mealtime practices are deeply intertwined with social hierarchies (Bourdieu

58 OCHS, SHOHET, CAMPOS, AND BECK

1979), and mealtimes themselves are cultural sites of socialization into standards of comportment and taste (Ochs, Pontecorvo, and Fasulo 1996).

Yet the structure of mealtimes is also dynamic and subject to shifting social and economic exigencies and preferences. Eating together under the same roof was a singularly defining feature of peasant family communities in the late Middle Ages (Segalen 1986). Yet shared family mealtimes at home as a routine, bounded formal activity in Europe and the United States have been strongly associated with industri- alization, which separated the worlds of work and home, and deemed the home a private sphere for cultivating family (Cinotto 2006; Davidoff and Hall 2002; DeVault 1991; Draznin 2001; Gillis 1989; Jordan 1987). The construction of dining rooms, the centrality of the dining table, and the act of dining together on a regular basis became markers of status for the middle class and beyond. Judith Flanders (2003) notes the social importance of dinnertime in Victorian England:

Dinner had, earlier, been a meal eaten at midday. . . . By midcentury, when most middle-class men were no longer working at home, dinner moved to the later hour of five or six, after the office workers returned home. From this hour, those who did not have to get up for work the next morning pushed dinner ever later, as a sign of leisure. The upper middle classes copied them, in order to indicate their own gentility, and the middle classes, in turn, followed their lead, in order to separate themselves from those beneath them. (266)

In the United States, the Victorian standard of families eating dinner around the table during evenings at home has waxed and waned over time. For example, in their study of Middletown in the 1920s, Lynd and Lynd (1929) report that families bemoaned the decline of the family dinner ritual in their community. Decades later, the 1950s middle class family apparently insisted on dinnertime as a centerpiece of the “Leave It to Beaver” American way of life. However, it is difficult to gauge the extent to which the family dinner meal historically and currently constitutes an ideal toward which families strive rather than a quotidian practice in American households (Cinotto 2006).

The concern with the status of the family meal and, more particularly, family dinnertime in the United States, is highly palpable in public health, social science, literature, film, television, and the popular press. Family dinners are potent cultural sites for generating wistful longings for an arguably imagined past or a possible fu- ture family way of life (Sutton 2001). In addition, family meals are charged with exceptional predictive powers for children’s well-being and, as a corollary, for sus- taining the family as a stable unit. Surveys of family food consumption in the United States link children’s reported nutritional habits to their reported participation in family dinners (Neumark-Sztainer et al. 2004; RMC Research Corporation 2005). Recent publications with titles such as The Surprising Power of Family Meals: How Eating Together Makes Us Smarter, Stronger, Healthier and Happier (Weinstein 2005) and a plethora of fact-sheet-type websites attest to a public yearning to see family

COMING TOGETHER AT DINNER 59

mealtimes as a resource for curing many ills of contemporary life such as family dissolution, childhood eating disorders, cigarette, alcohol and drug use, depression, and low educational achievement.

In addition to frequency of family mealtimes, the composition of family meals, especially the fate of the home-cooked family dinner, and the gendering of food preparation have been a source of considerable attention. Social and cultural stud- ies of food practices have emerged as a lively area of research, relating food pref- erences and taboos to social status, social relationships, and esthetic, moral, and religious frameworks (De Matos Viegas 2003; Grieshaber 1997; Murcott 1982, 1995; Ochs, Pontecorvo, and Fasulo 1996; Sobal, Bove, and Rauschenbach 2002; Thompson 1994). These studies document the importance of creating, solidify- ing, and structuring human relationships through daily meals across the world’s societies, and examine ideologies surrounding the intake of foods and folk models of what constitutes a meal.

In the United States, popular media describe a trend toward cooking less or not at all (e.g., Scattergood 2006; Scrivani 2005; Visser 1989), one that some individu- als make a self-conscious and deliberate attempt to reverse (Calta 2005; Jones et al. 2003; Pisano 2006; Severson and Moskin 2006). Despite their fascination with cook- ing shows, books, and equipment, modern Americans “eat breakfast in their cars, lunch at their desks and chicken from a bucket” (Scrivani 2005, 15). Some authors suggest that all of the information and resources for cooking in the modern United States may actually overwhelm potential cooks, producing “a terrible anxiety-rid- den limbo. There are the empty Calphalon pots. The waiting stove. The drawers full of enticing magazine recipes . . . but what to serve for dinner three hours from now?” (Scattergood 2006, F1). The failure to plan meals ahead of time may be increas- ingly common, complicating at-home meal preparation and increasing the expense involved (Chatzky 2006). The New York Times reports a “50-year slide away from home cooking,” and quotes one parent frustrated with the entire task: “We’re always hearing that eating dinner together is the cure for obesity, learning disorders, drugs, divorce, and every kind of problem we have in society . . . but what no one tells you is how to do all that cooking” (Severson and Moskin 2006).

Although in some contexts, cooking may be a form of religious expression (Sered 1988) and a source of domestic influence (Counihan 1988), scholarship on din- ner preparation in industrialized countries generally presents this task as a burden placed unequally on women. Women still report that food preparation is primarily their responsibility, although they spend less time cooking than women in earlier historical periods (Bittman 1995; DeVault 1991; Lupton 2000). It becomes increas- ingly difficult to find time for this and other household tasks with increased female participation in the workforce, producing significant identity conflicts for some women (Counihan 1988).

Several surveys indicate that children and parents in the United States frequently eat dinner together, but results vary with the age of the child and the ethnicity of the

60 OCHS, SHOHET, CAMPOS, AND BECK

family. The Child Trends Data Bank survey (2003) found that 56 percent of Ameri- can children ages six to eleven and 42 percent of adolescents ate a meal with their family six or seven days a week. The frequencies were higher for foreign-born than native-born, and higher for Hispanic than non-Hispanic adolescents and young children. The 2004 Healthy Youth Survey in Washington State (RMC Research Cor- poration 2005) also found that the frequency of eating together declines in American families as children grow older: approximately 78 percent of sixth-grade children and 60 percent of tenth-grade children, for example, reported that they ate dinner with their families always or most of the time. Culling their biennial assessments of the home environment of children from 1986 to 1994, the National Longitudinal Survey of Youth (Bradley et al. 2001) found that nonpoor children ate together with both parents more often than poor children, and that frequencies varied across eth- nicities. For example, 69 percent of nonpoor Euro-American and Hispanic children 6–9 years old reported eating a meal with both parents once a day or more, while 60 percent of nonpoor Asian American and 50 percent of nonpoor African Ameri- can children reported eating a meal with both parents once a day or more.

Most of what is known about the patterning of family mealtimes in the United States is based on reported behavior, rather than direct observation of how, when, where, and how often family members eat dinner. When children report that they eat dinner with their families, it is not clear whether all family members are pres- ent, and whether all the family members are eating together at the same time and in the same place. Family members may eat in different rooms, for example, with one child eating in front of the television in a bedroom while other family members eat elsewhere. Or some family members may begin eating long before another fam- ily member comes to eat, whereupon the earlier diners may leave the latecomer to finish eating alone. Moreover, reports of which family member is preparing dinner, what he or she is preparing for dinner, and how long it takes to prepare a family dinner generally do not discern the different kinds of involvement in food prepara- tion assumed by family members, and the extent to which a “home-cooked” meal consists of fresh ingredients, convenience foods (e.g., frozen pizza), or a combina- tion of both.

The present anthropological study addresses these lacunae by presenting results of ethnographic observations of family dinnertime preparation and eating patterns across thirty U.S. dual-earner families, whose activities were videotaped during the workweek and on the weekend. This analysis articulates (a) a typology of dinner- time arrangements, (b) the frequencies with which these arrangements characterize the families in this study across the days they were observed, and (c) the status of particular family dinnertime arrangements as sociocultural practices, that is, as the outcome of a familial disposition or habitus that structures how they routinely par- ticipate in their everyday lifeworlds (Bourdieu 1977, 1990). Observations are then related to two popular idioms regarding the fate of family dinnertime in the Ameri- can cultural imagination, namely the optimistic “Apple Pie” view that regular family

COMING TOGETHER AT DINNER 61

dinners continue to characterize American family life, and the pessimistic “Gloom and Doom” view that regular family dinners are no longer viable in the busy lifestyle of contemporary American families.

Method Participants

Families were recruited from the greater Los Angeles area as part of an Alfred P. Sloan Foundation–funded interdisciplinary investigation of the everyday lives of middle-class, dual-income families with young children. In the present study, thirty families from the UCLA Sloan Center on Everyday Lives of Families (CELF) data set were assessed. These thirty families were heterosexual, dual-income, middle-class households consisting of a married couple with two to three children, one of whom was required to be between the ages of eight and ten. As a marker of middle-class socioeconomic status, families were required to be homeowners with a monthly mortgage. Fliers and newspaper advertisements were used to recruit participants, and emphasis was placed on recruiting an ethnically and occupationally diverse sample. Families were compensated $1,000 in exchange for their participation in the study.

Procedure

CELF families participated in a study that used a combination of naturalistic ob- servation and self-report methods, including ethnographic video-recordings and timed observations of daily activities, semistructured interviews, questionnaires, and salivary cortisol, to document daily family life. As part of this research project, each family was recorded by two videographers during the course of a week for two weekday mornings, late afternoons, and evenings; one Saturday morning; and one Sunday morning and evening. These video recordings yielded approximately fifty hours of video footage per family. To capture family and home life, video recording began as the family members rose to begin their morning routine, stopped dur- ing the school and workday, resumed once the children left school and parents left work, and continued throughout the evening until the children retired to bed for the night. Thus, family dinner practices were captured in the course of the standard videotaping procedure across three evenings. Researchers then used these video data to examine families’ dinner-related behaviors.1

D E M O G R A P H I C S

Participants self-reported their ethnic backgrounds, income, education levels, jobs, and number of hours worked per week. The sample was ethnically diverse, but the majority of parents were of white and non-Hispanic descent born outside the

62 OCHS, SHOHET, CAMPOS, AND BECK

greater Los Angeles area, who lived in households with double the median Los An- geles County household income (median = $46,000; U.S. Census Bureau 2006b), and had at least a college degree. In terms of working hours, 63 percent of fathers and 50 percent of mothers reported working between forty and forty-nine hours per week, and 30 percent of fathers and 13 percent of mothers reported working over fifty hours per week.

FA M I LY D I N N E R PA RT I C I PAT I O N

A family dinner was defined as the main household meal of the evening, and the participation of each family member was documented. To understand the extent to which dinners were characterized by the full participation of the family mem- bers, dinners were organized as either having no family member missing or fam- ily member(s) missing. Dinners were then subcategorized into one of the following four patterns: (a) in unison, where all family members ate in the same location and at the same time; (b) in partial unison, where at least one member was missing but the remaining family members ate in the same location and at the same time; (c) fragmented, where family members ate dinner in different locations and/or at nonoverlapping times (i.e., with at least a 10-minute interval between start times); or (d) part-fragmented, where at least one member was missing and the remaining family members ate in different locations and/or at nonoverlapping times.

M E A L L E N G T H

Meal length was calculated at individual and family levels. At the individual level, each family member’s approximate start and stop time for eating was recorded. At the family level, family meal length was calculated as the time elapsed from the start time of the first family member who began eating, until the stop time of the last family member who finished eating. Individual and family meal lengths are both reported in minutes.

It was speculated that that concurrent engagement with noneating activities (e.g., watching television) during dinner may increase meal length. Therefore, fam- ily members’ engagement in any concurrent activities during dinner was docu- mented. Concurrent activities were defined as those occurring when one or more family members were observed engaging in a noneating activity during dinner (e.g., phone, television, homework).

M E A L P R E PA R AT I O N

To assess labor distribution during meal preparation, the extent to which family members participated in making dinners cooked at home was documented. Din- ners were characterized as being prepared by (a) Mom exclusively, (b) Dad exclu- sively, (c) Mom and Dad equally, (d) Mom and Dad, with Mom leading, (e) Mom and Dad, with Dad leading, (f ) Mom and kids, or (g) other relatives/nanny.

To assess the labor involved in meal preparation, the extent to which meals were prepared from scratch (using basic/raw ingredients) or assembled from commercial

COMING TOGETHER AT DINNER 63

foods was documented. Basic/raw ingredients include fresh vegetables, raw meat or fish, as well as dairy products, oils, dried pasta, and grains, among other items. A com- mercial dish was either purchased in finished form or finished by the home cook en- tirely according to package directions. Examples include pre-prepared frozen dishes such as pizza or pot pies. A modified commercial dish incorporates commercial items but is finished independently by the cook in a way not dictated by package directions. Examples include fish tacos made with frozen fish sticks, a baked pasta dish including commercial tomato sauce and pepperoni, or pigs-in-a-blanket (hot dogs wrapped in biscuit dough and baked). Meals cooked “from scratch” contained less than 20 per- cent commercial dishes and up to 50 percent modified commercial dishes.

Results To assess whether the contemporary family dinner practices of this sample fit the idealized or pessimistic models of family dinnertime, it was important to first ex- amine how often families were coming together for dinner. At the family level, the analysis revealed a mixed pattern. Although 77 percent of families had at least one dinner in unison during the three days, 63 percent of families had one or more frag- mented dinners during the three days, and 50 percent of families had one or more family members not home at least one day during the three days filmed. Across the sample, only five families (17%) consistently ate dinner in unison across all three days, and seven families (23%) never ate dinner in unison across the three days. To better understand these initial patterns, the individual participation and preparation practices of family dinners were then examined.

Participating in Dinner

In terms of individual participation in family dinnertime, all family members were present for dinner at least 83 percent of the time. As table 3.1 shows, however, Dads were more likely to be missing at weekday dinners than Moms, but Moms and Dads showed more similar levels of dinner participation on the weekends. Since Dads were more likely to be missing on weekdays, it raised the question of whether these men reported working more hours per week than Dads who never missed a dinner. This did not turn out to be the case. Of the eight Dads who missed a weekday dinner, 50 percent reported working forty to forty-nine hours per week and the other 50 per- cent reported working more than fifty hours per week. Interestingly, of the five fami- lies who ate dinner in unison all three days, two fathers reported working more than fifty hours per week. Notably, both were self-employed and more likely to have a flex- ible schedule than their counterparts who were not self-employed.

Although Dads were more likely to be missing at weekday dinners than Moms, the majority of weekday and weekend dinners were still largely consistent with the idealized perception of families eating together in the same space and location. As

64 OCHS, SHOHET, CAMPOS, AND BECK

table 3.2 shows, whether all members were present or not, families ate the majority of dinners in unison or partial unison. Of the forty-six weekday dinners when no family member was missing, 59 percent were eaten in unison. Similarly, of the twelve weekday dinners where one or more family members were missing, 67 percent were eaten in partial unison. A similar pattern was observed on the weekends. As table 3.2 shows, fewer than half of family dinners on weekdays and weekends were character- ized as fragmented or part-fragmented, whether all members were present or not.

In these limited data, there appears to be no clear link between the number of hours worked and family participation in dinner. For example, the percentage of families that had dinner in unison every night during the study is actually slightly higher among families with one parent working 50 + hours a week (18%) than those with both parents working less than fifty hours a week (16%). However, there was some indication that the scheduling of work hours may be a greater impediment to family dinners than the number of hours worked. Thus, the discussion shifts to fa- thers, as they were significantly more likely than mothers to miss family meals. Eight fathers in eight different families self-reported weekly work schedules that regularly involve working until 7 p.m. or later. In two additional families, the fathers reported working long shifts ranging from twenty-four hours to three days. None of these ten families, which make up 33 percent of the sample, had dinner in unison every night during the study. In the eight families in which the father completely missed dinner at least once, four fathers reported that their workdays ended between 7 p.m. and 10:30 p.m. A fifth father, an airline pilot, works away from home in three-day shifts. A sixth father did not report his usual working hours but held multiple jobs, including an evening job as a sports instructor.

In contrast, of the five families that ate dinner in unison every night, only two families have one parent working as late as 6 p.m. or 6:15 p.m. and the other three

Table 3.1. Family Dinner by Meal Level: Participation on Weekdays and Weekends

WEEKDAYS (N = 58) WEEKENDS (N = 26)

PRESENCE AT DINNERS FREQUENCY % FREQUENCY %

Mom 54 93 25 96

Dad 48 83 22 88

Child 1 56 97 25 96

Child 2 55 95 22 85

Child 3 16 89 6 100

Guests 7 12 4 13

Note: N = 60 possible meals on two weekdays for 30 families (2 meals were missing); 30 possible meals on one weekend day for 30 families (4 meals were missing).

COMING TOGETHER AT DINNER 65

families reported one parent whose return times ranged from 3 p.m. to 5:15 p.m. Three of the five fathers reported being self-employed or small-business owners, with presumably more flexible schedules than others, and one of them worked from an office at home. A fourth father, a lawyer who ends work at 6 p.m., works forty- eight hours per week but comments that his government position provides “more time at home” (perhaps relative to other employers he considered) while his chil- dren are young.

Meal Length and Concurrent Activities

On average, family dinners lasted for an average of 29.50 minutes on the weekdays and 33.23 minutes on the weekends. The average eating times of individual fam- ily members and guests ranged from 16.25 to 24.67 minutes on weekdays to 20 to 39.50 minutes on the weekends, with young children showing the shortest dinner times and parents and weekend guests showing the longest dinner times. Previous research has suggested that engaging in nondinner activities during dinner is an in- creasingly common practice. This analysis, however, suggests that dinners without concurrent activities are still the norm. The majority of weekday dinners (n = 39, 69%) occurred without concurrent activities, and only 33 percent (n = 19) of week- day dinners had concurrent activities where one or more family members was doing

Table 3.2. Family Dinners by Meal Level: How Often Family Members Are Present on Weekdays and Weekends

WEEKDAYS (N = 58) WEEKENDS (N = 26)

FREQUENCY % FREQUENCY %

NO FAMILY MEMBER MISSING

Family in unison 27 59 15 75

Family fragmented 19 41 5 25

FAMILY MEMBER MISSING

Family in part-unison 8 67 6 100

Family part-fragmented 4 33 0 0

IF NOT ALL AT HOME, WHO MISSING?

Mom 1 1

Dad 8 3

Child 1 1 1

Child 2 1 3

Child 3 1 0

Note: N = 60 possible meals on two weekdays for 30 families (2 meals were missing); 30 possible meals on one weekend day for 30 families (4 meals were missing). Percentages presented were calculated from total dinners characterized as no family member missing (46 weekday; 20 weekend) and family member missing (12 weekday; 6 weekend).

66 OCHS, SHOHET, CAMPOS, AND BECK

homework, watching TV, or talking on the phone. This pattern also held true for the weekends. The majority of weekend dinners did not contain concurrent activities (n = 15, 58%) although 42 percent (n = 11) of weekend dinners did contain concur- rent activities. When concurrent activities occurred during dinner, however, dinners were likely to be longer. Dinners with concurrent activities lasted forty minutes on average, whereas dinners without activities lasted twenty-five minutes on average.

M E A L P R E PA R AT I O N

Idealized family dinners are perceived to be characterized by home-cooked meals, whereas pessimistic perceptions suggest that contemporary family dinners are char- acterized by convenience or take-out food. This study found that CELF families ate home-cooked meals 75 percent of the time on weeknights and 85 percent of the time on weekends. Many of these meals rely on commercial food, however, even though they are prepared (or at least heated up) at home. On weeknights, for ex- ample, only one-third of the forty-five home-cooked meals are prepared by a parent from scratch. Of all fifty-eight weeknight dinners, only 25 percent of the weeknight dinners were cooked by a parent from scratch (figure 3.1). A recent nationwide study reports that 32 percent of all dinners were prepared at home from scratch (Sloan 2006). This slightly lower figure may result from the use of direct observa- tions rather than self-reported data, a stricter definition of “from scratch,” or a focus on working parents or on weeknights, among other factors.

Home-cooked from scratch

25%

Home-cooked/take- out mixture

5%

Take-out 15%

Restaurant 5%

Home-cooked, some commercial food

25%

Home-cooked, mostly of entirely commercial food

17%

Guest at grandparents’ 3%

FIGURE 3.1. Sources of 58 weeknight dinners in the CELF Study

COMING TOGETHER AT DINNER 67

Consistent with previous research on the double shifts of working mothers, 60 percent of home-cooked meals were prepared exclusively by Mom, while only 7 percent of home-cooked meals were prepared exclusively by Dad. The study ex- amined dinners in which Mom participated in the preparation, either by herself or with other members of her family, and found that Mom was involved in 91 percent of weekday dinners and 81 percent of weekend dinners, whereas Dad was involved in dinner preparation only 31 to 33 percent of the time. Thus, the labor of meal preparation appears to be disproportionately carried out by women.

Discussion: Reasons for Tempered Optimism What do these observations indicate about the viability of dinnertime as a cultural site of family unity and healthy meals in the United States? Do the observations support the previously cited, relatively optimistic surveys that report roughly 50 to 70 percent of elementary school-age children eating a meal at home with their par- ents almost every day? Is the media avalanche bemoaning the demise of the family dinner warranted? Is the home-cooked meal still a family culinary tradition, or has it given way to take-out, restaurant, or other pre-prepared food alternatives? Have gender roles regarding meal preparation and participation remained unchanged since prefeminist times, or has there been a radical shift in response to the exigen- cies of two parents working outside the home?

The response to all but the last of these questions is a hedged positive, “yes, but . . .” Regarding children and their parents coming together for family dinner, the finding that most of the families ate at least one dinner all together across the week suggests that dinnertime continues to be a culturally important gathering for the families in this study, and supports an optimistic outlook for the robustness of this ritual. But, few families ate dinner all together consistently across the week, suggesting that bringing the whole family to the dinner table is not a quotidian practice among the families who were video-recorded.

On the optimistic side, when some or all family members come home for din- ner during the week, they tend to eat together at one time and in one location, such as around a table. But, it is important to note that between 33 and 40 percent of the weekday dinnertimes involve some or all of the family members eating at different times or in different rooms of the house. During these fragmented meals, one or more family members begin eating ten minutes or more before another family member comes to eat. Upon his or her arrival, one or more of the other family members may already be leaving the dinner area. Another frag- mented meal scenario involves one or more family members eating in one room, while another eats somewhere else in the house, such as in the bedroom watching television. Although it is heartening to observe that family members generally congregate for dinner, the fragmented family dinners observed here temper and

68 OCHS, SHOHET, CAMPOS, AND BECK

nuance the more optimistic survey reports of kids and parents frequently eating dinner at home.

Can the busy work lives of parents account for the centripetal and centrifugal pat- terns of participation in family dinners? The finding that working Dads are missing from the dinner table significantly more often than working Moms might seem to make sense in light of the fact that Dads in this study worked longer hours than did Moms. But, a deeper look reveals that a simplistic correlation between work hours and family life is misleading: the Dads who missed family dinners did not work lon- ger than those who came home to eat together with their families. Indeed, as noted, eating all together consistently across the week was more characteristic of families with a parent who worked more than fifty hours a week than of those with parents who worked fewer hours. Moreover, while relative flexibility in workplace sched- ules freed a few of the working parents to have dinner together with their families, flexible work hours did not account for which families ate in unison and which did not. Of five self-employed fathers, only two always had dinner together with their families, and two missed dinner at least once. Of the three fathers who own small businesses, only one always had dinner in unison with their families. Parents in this study who made it home for dinner left the workplace by about 6 p.m. Some of the work after 6 p.m. may have been voluntary, because it was undertaken by the self- employed or small business owners.

Three conclusions can be drawn from these findings: first, being self-employed or a small business owner with flexible hours does not ensure that a parent will always eat together with his or her family. Second, eating all together at dinner may be pri- marily a lifestyle decision to return home to engage in this family practice, at least among middle-class families. Third, leaving the workplace before dinnertime (in this sample, by roughly 6 p.m.) may be linked to families having dinner together, and so pursuing this option may help working parents who want to eat with their families. These results are consistent with other studies that found that work schedule, not number of work hours, has a significant effect on quality of life (Barnett 2006).

What about the fate of the home-cooked dinner? Defying the pessimistic posi- tion that spending time in the kitchen preparing a meal is a relic of the past, this study found that the vast majority of family dinners were cooked at home. But “home-cooked” requires greater scrutiny, in that a home-cooked dinner usually entailed heating up or otherwise preparing packaged foods rather than raw ingre- dients. The American home cook’s reliance on convenience foods, however, may go back decades, as indicated by tuna casserole and other comfort food recipes in The Joy of Cooking (Rombauer 1931), which rely on canned or frozen items. Indeed, the editor of newer editions of the cookbook recently confessed that “In ‘97 we kind of lost our way . . . you had to make your own béchamel sauce to get the perfect tuna- fish casserole. But that is not what tuna casserole is about. It is about getting home and realizing that there’s nothing in the fridge for dinner and that there is nothing wrong with tuna casserole from a can” (Steinhauer 2006, 82).

COMING TOGETHER AT DINNER 69

Preparing tuna casserole from a can is not the only dinner practice that has withstood the test of time and social upheavals in the United States. The gender asymmetry observed in working Dads’ and Moms’ presence at family dinners is matched in their asymmetrical roles in dinner preparation. As in other stud- ies, the video-recordings, overwhelmingly and without caveat, affirm that Moms are still in the kitchen making dinner for the family most of the time. Even on the weekends, when Dads are present for 88 percent of the dinners, Dads rarely are the main chefs of the family dinner meal. Sociologist Luce Giard (1994) wrote “Man- ger sert . . . à concrétiser un des modes de relation entre la personne et le monde” (Eating serves . . . to concretize one of the modalities of relating the person and the world; 259). This study indicates that the practice of family dinnertime has a certain amount of resonance in the United States for concretizing the family as a social unit. After all, Moms in the study made an effort to prepare home-cooked meals and most families shared at least one dinner together. Yet, the inability of most families in the study to regularly eat together, the many families who ate dinner at home apart, and the quarter of the families who never ate together reveal the diminished status of dinnertime as an occasion for gathering the family in the sociocultural ecology of events that compose a family member’s day.

As attested in health and well-being interviews conducted with parents in the study, and in side conversations during filming, the vast majority considered “eating well” to be an integral part of attaining and maintaining health, and most couples further related having home-cooked meals to “eating well” and “being healthy.” Many parents likewise discussed their efforts to bring the family together for at least one meal a week, a practice borne out in our findings. Yet, parents also bemoaned their and their children’s busy schedules, suggesting that today’s “fast pace of life” often prevents them from all eating together at the same time. For some families, going out to eat as a family was a viable or preferred solution, while others relied on a combination of convenience foods and/or take-out meals at home, and enjoyed dinners that involve some, if not all, family members. In the opposition between centripetal forces bringing family members to the dinner table and centrifugal forces pulling them apart, the centrifugal forces appear to dominate during the workweek, while centripetal forces gain force only during the weekend. For many U.S. families, dinnertime is what might be called a “sometimes” practice, and as such it is only an intermittent cultural site for the primary socialization of children and the collective consideration of events that affect the lives of family members.

What are the practical implications of this study to policymakers and working families? Although the small sample of families that we studied limited generaliz- ability and our ability to conduct multivariate statistical analyses, our observations resonate with the existing literature and public perceptions of family dinners. Many families want to have dinner together because of the perceived and documented links between family dinners and child well-being (Fulkerson et al. 2006). The “yes, but . . .” conclusion of the family dinner study suggests that policymakers can help

70 OCHS, SHOHET, CAMPOS, AND BECK

families achieve schedules that enable them to fulfill their “yes” intentions of having dinner together. Naturalistic observations suggest that the problem is not the num- ber of hours that parents work, but the time of day when the hours are worked. Both employers and employees should encourage and use flexible schedules to improve work-life balance and preserve this important ritual that promotes family cohesion and child well-being.