Content uploaded by Shira Offer
Author content
All content in this area was uploaded by Shira Offer on Nov 08, 2017
Content may be subject to copyright.
This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/authorsrights
Author's personal copy
Assessing the relationship between family mealtime
communication and adolescent emotional well-being using
the experience sampling method
Shira Offer
*
Department of Sociology and Anthropology, Bar-Ilan University, Ramat Gan 52900, Israel
Keywords:
Adolescent well-being
Family meals
Family mealtime communication
Multilevel models
abstract
While most prior research has focused on the frequency of family meals the issue of which
elements of family mealtime are most salient for adolescents’well-being has remained
overlooked. The current study used the experience sampling method, a unique form of
time diary, and survey data drawn from the 500 Family Study (N¼237 adolescents with
8122 observations) to examine the association between family mealtime communication
and teens’emotional well-being. Results showed that in approximately half of the time
spent on family meals (3 h per week on average) adolescents reported talking to their
parents. Hierarchical linear model analyses revealed that controlling for the quality of
family relationships family mealtime communication was significantly associated with
higher positive affect and engagement and with lower negative affect and stress. Findings
suggest that family meals constitute an important site for communication between teens
and parents that is beneficial to adolescents’emotional well-being.
Ó2013 The Foundation for Professionals in Services for Adolescents. Published by Elsevier
Ltd. All rights reserved.
Family meals have been heralded in both scholarly research and the popular press as an important contributor to youth
development. Studies show that the frequency of family meals is positively associated with a myriad of health and well-being
outcomes. Adolescents who frequently eat meals with their family are healthier (Fulkerson, Kubik, Story, Lytle, & Arcan, 2009;
Hammons & Fiese, 2011), less likely to have eating disorders (Hammons & Fiese, 2011;Neumark-Sztainer, Eisenberg,
Fulkerson, Story, & Larson, 2008), engage in risky behaviors such as smoking, drinking alcohol, or using drugs (Eisenberg,
Olson, Neumark-Sztainer, Story, & Bearinger, 2004;Franko, Thompson, Affenito, Barton, & Striegel-Moore, 2008;Fulkerson,
Story, et al., 2006;Musick & Meier, 2012;National Center on Addiction and Substance Abuse, CASA, 2010;Sen, 2010), and
have higher school achievement (CASA, 2010;Eisenberg et al., 2004).
Research further shows that family meals are beneficial to adolescents’psychosocial adjustment. Using data from the
National Longitudinal Survey of Adolescent Health (Add Health) Musick and Meier (2012) reported that frequent family
dinners are related to a reduction in depressive symptoms. A similar pattern was observed among adolescents who partic-
ipated in the Eating Among Teens (EAT) Project (Eisenberg et al., 2004). The frequency of family meals was also found to be
related to higher social competence (Fulkerson, Story, et al., 2006) and lower perceived stress (Franko et al., 2008). This study
focuses on the association between family mealtime and teens’emotional well-being.
The burgeoning literature on family meals has proposed several mechanisms by which family meals promote youth health
and well-being. Family meals, it is argued, constitute an important ritual and a major site for socialization (Fiese & Schwartz,
*Tel.: þ972 3 5318651; fax: þ972 3 7384037.
E-mail address: shira.offer@biu.ac.il.
Contents lists available at SciVerse ScienceDirect
Journal of Adolescence
journal homepage: www.elsevier.com/locate/jado
0140-1971/$ –see front matter Ó2013 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.adolescence.2013.03.007
Journal of Adolescence 36 (2013) 577–585
Author's personal copy
2008;Fiese et al., 2002;Larson, Branscomb, & Wiley, 2006). Gathering at the table gives teens and their parents the op-
portunity to converse, express their feelings, and provide support to each other, which is important for reinforcing the social
bonds between them and building a sense of family togetherness (Fiese et al., 2002;Ochs & Shohet, 2006). Ochs and Shohet
(2006) further contended that family meals serve a moral function because “they are pervaded by talk oriented toward
reinforcing what is right and wrong about both the family and outsiders”(p. 42). Hence by eating meals together, parents can
convey key values to their children (Fiese, Foley, & Spagnola, 2006;Shaw, 2008). Family meals also facilitate parental
monitoring and supervision (Fiese et al., 2006;Neumark-Sztainer, Hannan, Story, Croll, & Perry, 2003;Ochs & Shohet, 2006;
Sen, 2010). Through conversations at the dinner table parents learn about their children’s daily activities and whereabouts.
According to findings from the National Center on Addiction and Substance Abuse study nearly 80% of the parents inter-
viewed mentioned that family meals gave them the opportunity to learn more about what was going on in their teens’lives
(CASA, 2010).
Overall, these studies suggest that one of the most important aspects of family mealsis communication between teens and
parents (Fiese et al., 2006;Fiese & Schwartz, 2008;Larson et al., 2006;Ochs & Shohet, 2006). Research shows that mealtime
communication is associated with lower emotional distress (Fiese et al., 2006) and a reduced risk of overweight among teens
(Jacobs & Fiese, 2007). Families, however, vary considerably in the extent to which they directly interact during mealtime
(Fiese & Schwartz, 2008). One major source of distraction during family meals is television, which is considered an important
inhibitor of social interactions (Feldman, Eisenberg, Neumark-Sztainer, & Story, 2007;Fiese & Schwartz, 2008;Jordan,
Hershey, McDivitt, & Heitzler, 2006).
In other words, it is not so much the family meal per se that is important for teens’well-being but what happens between
family members when they gather to share a meal. This important aspect of the family meal, however, has been relatively
overlooked in current research (Neumark-Sztainer, Larson, Fulkerson, Eisenberg, & Story, 2010). The major goal of the present
study is to fill in this gap in the literature by focusing on family mealtime communication. Musick and Meier (2012) noted that
“future work needs to go further in assessing which elements of mealtime are most salient, looking beyond how often families
eat together.”To respond to this call, this study examined the frequency of family mealtime communication and assessed its
relationship with adolescents’emotional well-being. Specifically, it evaluated (1) what proportion of family mealtime is spent
on direct communication between adolescents and their parents, and (2) how adolescents feel when they communicate with
their parents during mealtime. On the basis of the literature suggesting that family meals constitute an important site for
familial social interactions, I hypothesized that family mealtime communication would be associated with increased
emotional well-being among teens.
Measuring family meals and family mealtime communication
Most of the research conducted on family mealtime has examined the number of days per week that children eat
with their family, with at least five meals per week considered to be the optimal cut-off point for healthy outcomes
(CASA, 2010;Fiese & Schwartz, 2008). Studies, however, differ in the way they define a family meal in terms of the
people present and whether it refers to a specific meal, typically dinner, or not. For example, in the Eating Among Teens
(EAT) project, respondents were asked about the number of times a week all, or most, of the family in their house ate a
meal together without specifying who was present (Neumark-Sztainer et al., 2010). Musick and Meier’s(2012)study
used a more specific measure that asked whether respondents ate five or more evening meals a week with at least one
of their parents. They found that 60% of the adolescents in their sample fit this category. A similar number was reported
by Sen using the National Longitudinal Survey of Youth (NLSY) who indicated that 57% of adolescent girls and nearly
64% of adolescent boys ate dinner with their family at least 5 days a week, without specifying who was present during
the meal.
Other studies are based on time diaries and calculate the number of minutes a day (or hours per week) adolescents eat
with their family. Using the American Time Use Survey (ATUS), Wight, Price, Bianchi, and Hunt (2009) calculated that a third
of the teens ate a meal with their family for at least 20 min a day between 4 pm and 9 pm. Crouter, Head, McHale, and Tucker
(2004) employed a more inclusive measure of family meal and found that teens spent almost 2 h a week eating meals with
both of their parents and siblings present.
Most of these studies focus on the frequency of family meals but they do not account for what teens and their parents
do besides eating when they share a meal. The current study uses a different methodology, the experience sampling
method (ESM), to estimate not only the frequency of family meals but also the extent of family mealtime communication.
The ESM is a form of time diary that collects information in situ about respondents’activities, companionships, and
emotional states (Hektner, Schmidt, & Csikszentmihalyi, 2007). The ESM has several advantages. First, it makes it possible
to calculate accurate estimates of the overall amount of time adolescents spend eating meals with their parents under
different family constellations (i.e., with the mother only, with the father only, and with both parents present) and is not
limited to dinner time. As Musick and Meier (2012) pointed out, the benefits of shared meals may not be restricted to the
“evening meal.”Second, because the ESM collected data on both primary and secondary activities it can be used to learn
about what other activities adolescents engage in besides consuming food when they eat meals together with their family.
The ESM thus allows estimating how much time adolescents spend communicating with their parents during family
mealtime.
S. Offer / Journal of Adolescence 36 (2013) 577–585578
Author's personal copy
Family meals, family mealtime communication, and the quality of family relationships
This study estimated the association between family mealtime communication and adolescents’emotional well-being
while controlling for the quality of family relationships. As pointed out in previous research, the frequency of family meals
may be confounded with other variables, most notably the quality of family relations. Studies have found that teens spent
more time (Crouter et al., 2004) and ate meals (CASA, 2010;Franko et al., 2008) more frequently with their family when
relations with their parents were warmer and more supportive, and when they reported a positive atmosphere during
mealtimes (Fulkerson, Neumark-Sztainer, & Story, 2006;Fulkerson, Story, et al., 2006), suggesting that the association be-
tween family meals and well-being may be the result of selection bias. The same concern arises with respect to family
mealtime communication. Family members may be more likely to communicate during meals when their relationships are
good (Jordan et al., 2006).
Research shows, however, that for many outcomes the association between family meals and adolescent well-being re-
mains significant after adjusting for variables such as family connectedness (Eisenberg et al., 2004;Neumark-Sztainer et al.,
2003;Sen, 2010), cohesion and closeness (Franko et al., 2008;Musick & Meier, 2012), support (Fulkerson, Neumark-Sztainer,
et al., 2006;Fulkerson, Story, et al., 2006), and parental control (Musick & Meier, 2012). Additionally, longitudinal studies have
found that frequent family meals are correlated with lower rates of substance use (Eisenberg et al., 2004;Neumark-Sztainer
et al., 2008;Sen, 2010) and depressive symptoms (Musick & Meier, 2012) in subsequent years.
Altogether, these studies suggest that family meals are more than a mere proxy for some other family characteristics.
Although they do not make it possible to rule out the possibility that families with good (poor) quality relationships seek out
(avoid) eating meals together and communicating during family meals, they underscore the importance of controlling for the
quality of family relationships when estimating the association between family mealtime communication and adolescents’
well-being. The present study includes three variables to control for the quality of family relationships: parental monitoring,
teen self-disclosure, and parental support.
Contextual and demographic controls
The present study also controlled for several contextual and demographic characteristics which have been shown to affect
teens’likelihood of spending time and eating meals with their family. Previous research indicates that adolescents spend
more time with their family over the weekend (Larson & Richards, 1998;Larson, Richards, Moneta, Holmbeck, & Duckett,
1996) and when they are at home (Larson & Richards, 1994). It has also been shown that adolescents eat meals with their
family less frequently as they grow older and that adolescent girls are less likely to have family meals than adolescent boys
(Fulkerson et al., 2009,Fulkerson, Neumark-Sztainer, et al., 2006;Fulkerson, Story, et al., 2006;Neumark-Sztainer et al., 2003;
Sen, 2010).
Method
Participants
The data are drawn from the 500 Family Study, a research project on the work, school, and home experiences of parents in
dual-earner families and their children. Participants were recruited in 1999 and 2000 through posts at local schools and
newspapers in eight urban and suburban communities across the United States. The study includes predominantly non-
Hispanic white families with highly educated parents whose average earnings were above the national average for mar-
ried parents in the United States. For more information about the study and the participants’demographic characteristics, see
Hoogstra (2005).
Procedure
The 500 Family Study is a mixed-method study. Participants completed a survey and filled in the experience sampling
method (ESM). The ESM is a unique form of time diary that collects information about respondents’activities, surroundings,
and emotional states as they occur in their natural setting over the course of a typical week (Csikszentmihalyi & Larson, 1987;
Hektner et al., 2007). In this study, participants carried alarm watches that were pre-programmed to randomly emit eight
signals during their waking hours for seven consecutive days. When signaled, participants were asked to indicate what they
were doing, where they were, who was with them, and how they were feeling in a self-report questionnaire. The ESM is
considered a highly valid and reliable instrument that provides accurate information about individuals’time uses and sub-
jective experiences (Hektner et al., 2007). A major advantage of the ESM is that it provides detailed information about the
multiple activities respondents are engaged in simultaneously (Hektner et al., 2007;Schneider, 2006) and therefore makes it
possible to examine what activities respondents are engaged in besides consuming food when they eat with their family.
The analyses in the current study are based on data collected from the adolescents who filled in both the survey and the
ESM. Preliminary analyses showed that the excluded teens (i.e., teens who did not fill in the ESM) were more likely than those
included to be boys and report lower family support. The adolescents responded, on average, to 30 out of 56 signals and,
S. Offer / Journal of Adolescence 36 (2013) 577–585 579
Author's personal copy
consistent with previous research, those respondents who missed more than one fourth of the signals (n¼65) were excluded
from the sample (Schneider, 2006). The final sample consisted of 237 respondents with 8122 signals.
Missing data in the four emotional outcomes and survey measures were imputed using a multiple imputation technique
with AMELIA II software (Honaker, King, & Blackwell, 2009). Seventy-four signals could not be imputed because the re-
spondents did not answer any of the emotional items prompted by these signals. This is not problematic, however, because in
multilevel modeling respondents can miss some signals, or items in a signal, and still be included in the model. Imputing
missing data did not yield significantly different results.
Measures
Family meals
In the ESM respondents were asked to report the main activity they were engaged in when signaled. Responses to this
question were originally coded by trained coders into more than 400 activity codes (all items were double coded with
interrater reliabilities ranging from
k
¼0.79–0.95). One of these codes was “eating a meal”. Respondents were also asked to
indicate who they were with when signaled. I used these two questions to construct three dummy family meal measures:
meals with the mother only, meals with the father only, and meals with both parents. Due to low frequencies, I later collapsed
these three measures into one measure, referred to as family meals, which indicates whether respondents were having a meal
with either their mother only, fatheronly, or both parents when signaled (yes ¼1, no ¼0). This global measure of family meals
was used in all the multivariate analyses.
Family mealtime communication
A major advantage of the ESM is that it asked respondents to report not only on their primary activity but also on their
secondary activity, if they had any. I used answers to this question to identify signals in which respondents indicated that they
were directly interacting with their parents during mealtime. This dummy measure, referred to as family mealtime
communication, indicates whether respondents were talking to their parents while eating with them (yes ¼1, no ¼0).
Other family mealtime activities
Using responses to the second activity item, I identified two other categories of family mealtime activity. Family mealtime
leisure refers to activities such as watching television, reading, listening to music, and playing computer games during family
meals. Because the frequency of watching television during family meals in this sample was small (3 signals for meals with
the mother only, 8 signals for meals with the father only, and 8 signals for meals with both parents) I did not treat it as a
separate category. The measure other family mealtime activities refers to activities such as helping with the dishes, talking on
the phone, cooking, and playing with a pet. Finally, signals inwhich respondents did not report any additional activity besides
eating with their parents were coded as nothing else.
Emotional well-being
In the ESM respondents were asked to report how they were feeling when signaled. I used these items to construct four
composite measures of emotional well-being. These are the means of (a) positive affect dfeeling cheerful, happy, and good
about oneself when signaled (
a
¼0.69); (b) engagement denjoying the activity, finding it interesting, feeling involved in the
activity, and wishing to do something else (reverse coded) (
a
¼0.65); (c) negative affect dfeeling angry, irritated, and
frustrated (
a
¼0.90); and (d) stress dfeeling stressed and strained (
a
¼0.80). The response categories for all these items
ranged from 0 (not at all)to3(very much). As Table 1 shows, adolescents in this study reported, on average, relatively high
levels of positive affect and engagement and low levels of negative affect and stress.
Table 1
Descriptive statistics for well-being measures and controls (N¼237 adolescents with 8122 signals).
Mean SD Range
a
Well-being outcomes (ESM)
a
Positive affect 1.98 0.38 0–3 0.69
Engagement 1.62 0.32 0–3 0.65
Negative affect 0.48 0.35 0–3 0.90
Stress 0.51 0.38 0–3 0.80
Contextual controls (ESM)
a
Weekend 0.25 0.12 0–1
Home 0.45 0.14 0–1
Demographic controls (survey)
Adolescent boy 40.5
Age 15.38 1.60 11–18
Quality of family relationships (survey)
Parental monitoring 1.77 0.64 0–3 0.66
Teen self-disclosure 1.71 0.60 0–4 0.80
Family support 2.24 0.46 0–3 0.87
a
Estimates were calculated at the aggregated person-level.
S. Offer / Journal of Adolescence 36 (2013) 577–585580
Author's personal copy
Controls
I included two contextual controls drawn from the ESM. Weekend refers to whether the signal occurred during the weekend
and home to whether it occurred while the respondent was at home (yes ¼1, no ¼0). The demographic controls were drawn
from the survey. Gender was coded as a dummyvariable (boy ¼1, girl ¼0) and age was measured in years. As shown in Table 1,
the sample included fewer adolescent boys (40.5%) than adolescent girls. The average age was 15.4 (SD ¼1.6).
Using the survey data, the analyses also adjusted for the quality of family relationships. Parental monitoring was the mean of
five items that asked respondents to indicate, on a scale ranging from 0 (never)to3(often), how often their parents check on
whether they have done their homework, limit the amount of time they spend watching TV or playing video or computer games,
limit the amount of time they go out with friends on school nights, call to check when they are out with friends, and ask them to
call home when they are out with friends (
a
¼0.66). Teen self-disclosure was the mean of 16 questions that asked respondents
how often they discuss issues with their parents, such as things they have studied in class, going to college, who theirfriends are,
drugs, and dating (
a
¼0.80). The response categories forall these itemsranged from0 (never)to4(2–3 times a week). These two
measures were adapted from the Sloan Study of Youth and Social Development (see Hoogstra, 2005). Family support, based on
Rathunde’s (1996) index, was the mean of 16 items that asked how often in the respondent’s family, for example, “Ifeel
appreciated for who I am”;“I dothings I like to do without feeling embarrassed”; and “If I have a problem, I get special attention
and help”(
a
¼0.87). The response categories ranged from 0 (never)to3(often). The adolescents in this study reported, on
average, relatively high levels of parental monitoring and family support and medium levels of self-disclosure (see Table 1).
Analysis strategy
I began by providing descriptive information about the frequency of family meals and family mealtime communication. In the
next stage, I used hierarchical linear modeling to examine the association between family mealtime communication and teens’
emotional well-being. The major advantage of hierarchical linear modeling is that it accounts for the nonindependence of ob-
servations within individuals and can be used to estimate the signal-level (within-respondent level) and person-level (between-
respondent level) simultaneously (Raudenbush & Bryk, 2002). First, the global measure of family meals was used asa predictor of
the four emotional well-being outcomes then the measure of family mealtime communication was used as a predictor. These
models can be expressed with two sets of equations. The first is awithin-person (i.e. signal) levelequation that models emotional
well-being as a function of whether the respondent was eating a meal with his or her parents and level-1 controls:
Positive affectij ¼
b
0jþ
b
1jðfamily mealtime communicationÞij þ
b
2jðweekendÞij þ
b
3jðhomeÞij þ
3
ij (1)
The dependent variable is the score on emotional outcome (positive affect in this illustration) on signal ifor person j. In this
equation
b
0j
is the intercept,
b
1j
denotes the association between having a family mealtime communication signal and positive
affect,
b
2j
and
b
3j
respectively denote the association between having weekend and home signals and positive affect, and
3
ij
is
the within-person residual. In this equation the intercept is allowed to vary randomly across individuals. This variation is then
explained using the second equation, which estimates variation across respondents in
b
0j
as a function of covariates Z:
b
0j¼
g
00 þ
g
01Z1jþ
g
02Z2jþ.þ
g
0kZkj þ
n
0j(2)
In the person-level equation
g
00
is the average score on positive affect,
g
01
to
g
0k
are the regression coefficients of the
estimated effects of the covariates Z(i.e. age, gender, parental monitoring, teen self-disclosure, and parental support) on the
adjusted positive affect score, and
n
0j
is the person-level error term assumed to be normally distributed with mean zero and
unknown variance.
The hierarchical linear models also control for the aggregated person-mean levels of time spent on family meals in order to
differentiate between the within-person and between-person effects of family meals on well-being (Hoffman & Stawski,
2009). This control accounts for potential variation between respondents in the overall amount of time they spend eating
meals with their parents and consequently provides more accurate estimates of the association between family meals, family
mealtime communication, and well-being.
Results
The frequency of family meals and family mealtime communication
How much time do adolescents spend eating meals with their family and what proportion of that time is spent on direct
communication with parents? Table 2 presents the frequency distribution of family meals. The adolescents in this study spent
overall a small amount of their waking time eating meals with their parents. Only about 3% of the signals corresponded to
family meals with either the mother only, the father only, or both parents. ESM proportions can be multiplied by 112 (16
average waking hours per day 7 days a week) to derive a weekly estimate of the number of hours spent in family meals (see
Schneider, 2006). Hence this percentage corresponds to 3.36 h a week of family meals. Slightly more than one percent of the
signals were spent in family meals with both parents (1.2%) and with the mother only (1.1%), the equivalent of 1.34 and 1.23 h
a week, respectively. The adolescents ate less frequently with their father only (0.67 h a week).
S. Offer / Journal of Adolescence 36 (2013) 577–585 581
Author's personal copy
To estimate the proportion of family mealtime spent on direct communication with parents I categorized each family meal
measure by type of secondary activity. Table 3 shows the percentage of signals spent on direct communication, leisure, other
activities, and nothing else, out of all the signals corresponding to family meals for each family meal measure. The results
indicate that most family meal signals were spent in direct communication with parents (42.7%, 37.0%, and 54.7% for meals
with the mother only, the father only, and both parents, respectively). Overall, adolescents talked with their parents in almost
half of their shared meals (46.5%). Note that the highest percentage of family mealtime communication was found for meals
with both parents present. These teens were also less likely to engage in a leisure activity, such as watching television or
listening to music, during family meals when both of their parents were present. Teens engaged in a leisure activity when
both parents were present in only 13.7% of the family meal signals, compared to 22.5% when they ate meals with their mother
only and 34.8% when they ate meals with their father only. In 20% of all family meal signals teens reported doing nothing else
aside from eating when they shared a meal with both parents, compared to 14.6% and 17.4% when they ate meals with their
mother only and father only, respectively.
The association between family mealtime communication and adolescent’s emotional well-being
How do adolescents feel when they eat meals with their parents and, more specifically, how do they feel when they
communicate with their parents during mealtime? Table 4 presents the results of a series of hierarchical linear analyses that
estimated the association between family meals and the four ESM well-being measures (Model 1) and between family
mealtime communication and these outcomes (Model 2), controlling for location and time of the week at level-1 and age,
gender, and the quality of family relations at level-2. Because the number of signals in each family meal measure was small, I
did not examine meals with the mother only, meals with the father only, and meals with both parents separately but rather
collapsed them into one global measure (i.e., family meals).
As Table 4 shows, respondents reported higher positive affect (b¼0.16, p<0.001) and engagement (b¼0.22, p<0.001)
and lower negative affect (b¼0.13, p<0.001) and stress (b¼0.09, p<0.001) when they ate meals with their parents
compared to when they did not. As hypothesized, family mealtime communication was particularly beneficial to adolescents’
emotional well-being. Family mealtime communication was associated with higher positive affect (b¼0.22, p<0.001) and
engagement (b¼0.33, p<0.001) and lower negative affect (b¼0.15, p<0.001) and stress (b¼0.15, p<0.001). Note that
the associations between each of these outcomes and family mealtime communication were stronger than their associations
with the global measure of family meals.
Discussion
The literature highlights the beneficial effects of family meals for adolescents’health and well-being. Yet the question of
what particular aspect of the family meal contributes to adolescents’well-being has not been systematically examined in
quantitative research (Fiese & Schwartz, 2008;Musick & Meier, 2012). This study built upon and extended previous research
on family mealtimes by specifically focusing on family mealtime communication and its associations with adolescents’
emotional well-being.
The first research goal was to determine how often adolescents eat meals with their family and what proportion of that
shared time is spent on direct communication with parents. This study found thatoverall, teens spent slightly more than 3 h a
week eating meals with their parents, either with their mother only, father only, or both parents present. If, as Fiese and
Table 2
Frequency of family meals, by family constellation and overall (N¼237 adolescents with 8122 signals).
Mean
a
SD
a
Hours/week
Mother only 1.1 2.26 1.23
Father only 0.6 1.37 0.67
Both parents 1.2 2.29 1.34
All family meals 2.9 3.55 3.25
a
Means and standard deviations were calculated at the aggregated person-level out of the respondent’s total number of
signals.
Table 3
Family meals by type of secondary activity: percentages (N¼237 adolescents with 8122 signals).
Communication Leisure Other activities Nothing else Total
Mother only 42.7 22.5 20.2 14.6 100
Father only 37.0 34.8 10.9 17.4 100
Both parents 54.7 13.7 11.6 20.0 100
All family meals 46.5 21.3 14.7 17.4 100
Note: Estimates were calculated at the signal-level out of all family meal signals.
S. Offer / Journal of Adolescence 36 (2013) 577–585582
Author's personal copy
Table 4
Hierarchical linear models results predicting adolescents’emotional well-being with family meals (Model 1) and family mealtime communication (Model 2), and controls (N¼237 respondents with 8122signals).
Positive affect Engagement Negative affect Stress
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
b(SE) b(SE) b(SE) b(SE) b(SE) b(SE) b(SE) b(SE)
Intercept 2.03*** (0.03) 2.03*** (0.03) 1.56*** (0.03) 1.56*** (0.03) 0.49*** (0.02) 0.49*** (0.02) 0.56*** (0.03) 0.56*** (0.03)
Signal-level
Family meals 0.16*** (0.04) 0.22*** (0.05) 0.13*** (0.03) 0.09** (0.03)
Family mealtime
communication
0.22*** (0.05) 0.33*** (0.07) 0.15*** (0.04) 0.15*** (0.04)
Weekend 0.08*** (0.02) 0.08*** (0.02) 0.17*** (0.02) 0.17*** (0.02) 0.05** (0.02) 0.05** (0.02) 0.10*** (0.02) 0.10*** (0.02)
Home 0.13* (0.02) 0.13* (0.02) 0.01 (0.02) 0.01 (0.02) 0.01 (0.02) 0.00 (0.02) 0.04** (0.02) 0.04** (0.02)
Person-level
Boy 0.03 (0.04) 0.03 (0.04) 0.01 (0.04) 0.01 (0.04) 0.00 (0.04) 0.00 (0.04) 0.01 (0.04) 0.01 (0.04)
Age 0.05** (0.01) 0.05** (0.01) 0.00 (0.01) 0.00 (0.01) 0.00 (0.01) 0.00 (0.01) 0.02 (0.01) 0.02 (0.01)
Parental monitoring 0.01 (0.04) 0.01 (0.04) 0.02 (0.04) 0.02 (0.04) 0.08* (0.04) 0.08* (0.04) 0.07 (0.05) 0.07 (0.04)
Teen self-disclosure 0.04 (0.05) 0.04 (0.05) 0.04 (0.04) 0.04 (0.04) 0.11* (0.05) 0.11* (0.05) 0.13** (0.05) 0.13** (0.05)
Family support 0.37*** (0.06) 0.37*** (0.06) 0.17** (0.05) 0.18*** (0.05) 0.24*** (0.07) 0.24*** (0.07) 0.11 (0.06) 0.11 (0.06)
Person mean
family meals
0.31 (0.55) 0.03 (0.06) 0.14 (0.59) 0.04 (0.05) 0.05 (0.61) 0.05 (0.47) 0.16 (0.61) 0.02 (0.06)
Within-person
variance
s
2
0.34 0.34 0.57 0.57 0.38 0.38 0.28 0.28
Between-person
variance
s
00
0.11 0.11 0.08 0.08 0.09 0.09 0.12 0.12
*p<.05. ** p<.01. *** p<.001.
S. Offer / Journal of Adolescence 36 (2013) 577–585 583
Author's personal copy
Schwartz (2008) suggest, family meals typically last 20 min, this corresponds to almost 10 shared meals a week and implies
that teens and their parents eat not only dinners together but other meals as well. More importantly, the findings indicate that
in nearly half of the time spent in family meals, adolescents reported talking to their parents. Family mealtime communi-
cation was especially high during family meals when both parents were present. These findings corroborate previous research
and provide further evidence that shared meals give family members the opportunity to interact and communicate (Fiese
et al., 2006;Fiese & Schwartz, 2008;Larson et al., 2006;Ochs & Shohet, 2006).
The second research goal was aimed at testing how adolescents feel when they eat meals with their parents. As hy-
pothesized, results from the hierarchical linear model analyses revealed that family mealtime communication was beneficial
to adolescents’emotional well-being. Family mealtime communication was found to be significantly associated with higher
positive affect and engagement and lower negative affect and stress among these adolescents. Consistent with previous
research (CASA, 2010;Fiese et al., 2006;Fulkerson, Neumark-Sztainer, et al., 2006;Fulkerson, Story, et al., 2006), these
findings suggest that family meals are an enjoyable activity for teens and constitute an important site for positive connections
and interactions between teens and parents, which contributes to adolescents’well-being.
Limitations, strengths, and directions for future research
This study has several limitations. First, because the sample consisted of dual-earner middle-class families it did not make
it possible to examine variations in family meals and family mealtime communication by socioeconomic status or parental
composition. Previous studies have shown that families of higher socioeconomic status eat meals together more frequently
(Fiese & Schwartz, 2008;Musick & Meier, 2012;Wight et al., 2009) and are less likely to watch television during shared meals
than families of lower status (Feldman et al., 2007). This important family routine may also be more difficult to maintain in
families headed by one parent compared to families headed by a couple because a single parent has no spouse or partner with
whom to share the work involved in preparing family meals (Tubbs, Roy, & Burton, 2005;Wight et al., 2009).
Second, the homogenous nature of the sample did not allow for testing differences by race and ethnicity. Previous research
has provided mixed results regarding the effect of race and ethnicity on family meal patterns. A recent study has shown that
minority teens in the United Kingdom are more likely than white teens to eat meals with their family (Maynard & Harding,
2010). Similar patterns have been observed among African American teenagers in the United States (Larson, Richards, Sims, &
Dworkin, 2001). By contrast, Wight et al. (2009) found that African American teens are less likely to eat meals with their
family than white teens. More research is needed to clarify the effects of race and ethnicity on the frequency and experience of
family meals.
Other limitations pertain to the cross-sectional and non-experimental design of the study, which is not suited for making
causal claims. This study cannot rule out the possibility that the beneficial association between family meals, family mealtime
communication, and adolescents’well-being may be due to reverse causation; that is, that well-adjusted teens are more
willing to eat with their family and communicate with them during shared meals than their less well-adjusted counterparts.
Nor can the analyses account for the possibility that these associations are due to some unobserved characteristic of the
adolescent, parent, or both. Interestingly, Jordan et al. (2006) indicated that some of the parents in their study mentioned
watching television during family meals as a way to avoid conflictual interactions with their children. Hence, although the
analyses controlled for the quality of family relationships, adolescents could have sought out or avoided communicating with
their parents during family meals as a function of some prior characteristic of the home environment. This is of particular
concern because respondents who did not fill in the ESM and were consequently excluded from the study were more likely
than those included to have reported lower family support. Therefore, adolescents in families with lower relational quality
may have been selected out, leading to inflated estimates. Another potentially confounding factor that was not accounted for
in this study is parenting style. Following Berge, Wall, Neumark-Sztainer, Larson, and Story's (2010) finding that an author-
itative parenting style was associated with more frequent family meals, one can expect parenting style to also affect patterns
of family mealtime communication.
Despite these limitations, the current study makes an important contribution to the literature on family meals, and family
time more generally. The methodology it employed, the ESM, provided accurate measures of the overall amountof time teens
spent in family meals and made it possible to estimate how frequently they communicated during shared meals. The focus on
all family meals, rather than family dinners only, constitutes a major advantage of the present study.
Finally, by showing that the communication that goes on between teens and their parents during family meals was highly
beneficial to adolescents’well-being, this study helped answer the important question of “what is it about family meals that
explains better outcomes in youth”(Neumark-Sztainer et al., 2010, p. 1119). Hence as an important routine and site for social
interactions between family members, family meals should be encouraged and facilitated. Nevertheless it should be
remembered that family meals are not the only arena for family communication (Kremer-Sadlik & Paugh, 2007;Musick &
Meier, 2012). Future research should also pay attention to other activities and routines that give families the opportunity
to communicate and build positive and enjoyable relationships.
References
Berge, J. M., Wall, M., Neumark-Sztainer, D., Larson, N., & Story, M. (2010). Parenting style and family meals: cross-sectional and 5-year longitudinal as-
sociations. Journal of American Dietetic Association, 110(7), 1036–1042. http://dx.doi.org/10.1016/j.jada.2010.04.011.
S. Offer / Journal of Adolescence 36 (2013) 577–585584
Author's personal copy
Crouter, A. C., Head, M. R., McHale, S. M., & Tucker, C. J. (2004). Family time and the psychosocial adjustment of adolescent siblings and their parents. Journal
of Marriage and Family, 66,147–162. http://dx.doi.org/10.1111/j.0022-2445.2004.00010.x-i1.
Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the experience sampling method. The Journal of Nervous and Mental Disease, 175,
526–536.
Eisenberg, M. E., Olson, R. E., Neumark-Sztainer, D., Story, M., & Bearinger, L. H. (2004). Correlations between family meals and psychosocial well-being
among adolescents. Archives of Pediatrics and Adolescent Medicine, 158, 792–796. http://dx.doi.org/10.1001/archpedi.158.8.792.
Feldman, S., Eisenberg, M. E., Neumark-Sztainer, D., & Story, M. (2007). Associations between watching TV during family meals and dietary intake among
adolescents. Journal of Nutrition Education and Behavior, 39,257–263. http://dx.doi.org/10.1016/j.jneb.2007.04.181.
Fiese, B. H., Foley, K. P., & Spagnola, M. (2006). Routine and ritual elements in family mealtime: contexts for child well-being and family identity. New
Directions for Child and Adolescent Development, 111,67–89. http://dx.doi.org/10.1002/cd.156.
Fiese, B. H., & Schwartz, M. (2008). Reclaiming the family table: mealtimes and child health and well-being. Social Policy Report, 22(4), 3–19.
Fiese, B. H., Tomcho, T. J., Douglas, M., Joseph, K., Poltrock, S., & Backer, T. (2002). A review of 50 years of research on naturally occurring family routines and
rituals: cause for celebration? Journal of Family Psychology, 16, 381–390 http://dx.doi.org/10.1037//0893-3200.16.4.381.
Franko, D. L., Thompson, D., Affenito, S. G., Barton, B. A., & Striegel-Moore, R. H. (2008). What mediates the relationship between family meals and
adolescent health issues? Health Psychology, 27(Suppl.), S109–S117 http://dx.doi.org/10.1037/0278-6133.27.2.
Fulkerson, J. A., Kubik, M. Y., Story, M., Lytle, L., & Arcan, C. (2009). Are there nutritional and other benefits associated with family meals among at-risk
youth? Journal of Adolescent Health, 45, 389–395 http://dx.doi.org/10.1016/j.jadohealth.2009.02.011.
Fulkerson, J. A., Neumark-Sztainer, D., & Story, M. (2006). Adolescent and parent views of family meals. Journal of the American Dietetic Association, 106, 526–
532. http://dx.doi.org/10.1016/j.jada.2006.01.006.
Fulkerson, J. A., Story, M., Mellin, A., Leffert, N., Neumark-Sztainer, D., & French, S. A. (2006). Family dinner meal frequency and adolescent development:
relationships with developmental assets and high-risk behaviors. Journal of Adolescent Health, 39,337–345. http://dx.doi.org/10.1016/j.bbr.2011.03.031.
Hammons, A. J., & Fiese, B. H. (2011). Is frequency of shared family meals related to the nutritional health of children and adolescents? Pediatrics, 127, e1565–
e1574 http://dx.doi.org/10.1542/peds.2010-1440.
Hektner, J. M., Schmidt, J. A., & Csikszentmihalyi, M. (2007). Experience sampling method: Measuring the quality of everyday life. Thousand Oaks, CA: Sage.
Hoffman, L., & Stawski, R. S. (2009). Persons as contexts: evaluating between-person and within-person effects in longitudinal analysis. Research in Human
Development, 6,97–120. http://dx.doi.org/10.1080/15427600902911189.
Honaker, J., King, G., & Blackwell, M. (2009). Amelia II: A program for missing data (Version 1.2–6) [Computer software]. Cambridge, MA: Harvard University.
Hoogstra, L. (2005). Design and sample characteristics of the 500 Family Study. In B. Schneider, & L. J. Waite (Eds.), Being together, working apart: Dual-career
families and their work –life balance (pp. 23–48). Cambridge, UK: Cambridge University Press.
Jacobs, M. P., & Fiese, B. H. (2007). Family mealtime interactions and overweight children with asthma: potential for compounded risks? Journal of Pediatric
Psychology, 32,64–68 http://dx.doi.org/10.1093/jpepsy/jsl026.
Jordan, A. B., Hershey, J. C., McDivitt, J. A., & Heitzler, C. D. (2006). Reducing children’s television-viewing time: a Qualitative study of parents and their
children. Pediatrics, 118, e1303–e1310. http://dx.doi.org/10.1542/peds.2006-0732.
Kremer-Sadlik, T., & Paugh, A. L. (2007). Everyday moments: finding “quality time”in American working families. Time & Society, 16, 287–308. http://dx.doi.
org/10.1177/0961463X07080276.
Larson, R., Branscomb, K., & Wiley, A. (2006). Forms and functions of family mealtimes: multidisciplinary perspectives. New Directions for Child and
Adolescent Development, 111,1–15.
Larson, R., & Richards, M. H. (1994). Divergent realities: The emotional lives of mothers, fathers, and adolescents. New York: BasicBooks.
Larson, R., & Richards, M. H. (1998). Waiting for the weekend: Friday and Saturday night as the emotional climax of the week. New Directions for Child and
Adolescent Development, 82,37–51.
Larson, R., Richards, M. H., Moneta, G., Holmbeck, G., & Duckett, E. (1996). Changes in adolescents’daily interactions with their families from ages 10 to 18:
disengagement and transformation. Developmental Psychology, 32,744–754. http://dx.doi.org/10.1037/0012-1649.32.4.744.
Larson, R., Richards, M. H., Sims, B., & Dworkin, J. (2001). How urban African American young adolescents spend their time: time budgets for locations,
activities, and companionship. American Journal of Community Psychology, 29, 565–597. http://dx.doi.org/10.1023/A:1010422017731.
Maynard, M. J., & Harding, S. (2010). Ethnic differences in psychological well-being in adolescence in the context of time spent in family activities. Social
Psychiatry þPsychiatric Epidemiology, 45,115–123. http://dx.doi.org/10.1007/s00127-009-0047-z.
Musick, K., & Meier, A. (2012). Assessing causalityand persistence in associations between family dinners and adolescent well-being. Journal of Marriage and
Family, 74,476–493. http://dx.doi.org/10.1111/j.1741-3737.2012.00973.x.
National Center on Addiction and Substance Abuse (CASA). (2010) The importance of family dinners VI. Retrieved from: http://www.casacolumbia.org.
Neumark-Sztainer, D., Eisenberg, M. E., Fulkerson, J. A., Story, M., & Larson, N. I. (2008). Family meals and disordered eating in adolescents: longitudinal
findings from Project EAT. Archives of Pediatric and Adolescent Medicine, 162(1), 17–22.
Neumark-Sztainer, D., Hannan, P. J., Story, M., Croll, J., & Perry, C. (2003). Family meal patterns: associations with sociodemographic characteristics and
improved dietary intake among adolescents. Journal of the American Dietetic Association, 103,317–322. http://dx.doi.org/10.1053/jada.2003.50048.
Neumark-Sztainer, D., Larson, N. I., Fulkerson, J. A., Eisenberg, M. E., & Story, M. (2010). Family meals and adolescents: what have we learned from Project
EAT (Eating Among Teens)? Public Health Nutrition, 13(7), 1113–1121 http://dx.doi.org/10.1017/s1368980010000169.
Ochs, E., & Shohet, M. (2006). The cultural structuring of mealtime socialization. New Directions for Child and Adolescent Development, 111,35–49. http://dx.
doi.org/10.1002/cad.153.
Rathunde, K. (1996). Family context and talented adolescents’optimal experience in school-related activities. Journal of Research on Adolescence, 16,
605–626.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models. Thousand Oaks, CA: Sage.
Schneider, B. (2006). In the moment: the benefits of the experience sampling method. In M. Pitt-Catsouphes, E. E. Kossek, & S. Sweet (Eds.), The work and
family handbook (pp. 469–488). Mahwah, NJ: Lawrence Erlbaum.
Sen, B. (2010). The relationship between the frequency of family dinner and adolescent problem behaviors after adjusting for other family characteristics.
Journal of Adolescence, 33,187–196. http://dx.doi.org/10.1016/j.adolescence.2009.03.011.
Shaw, S. M. (2008). Family leisure and changing ideologies of parenthood. Sociology Compass, 2, 688–703. http://dx.doi.org/10.1111/j.1751-9020.2007.
00076.x.
Tubbs, C. Y., Roy, K. M., & Burton, L. M. (2005). Family ties: constructing family time in low-income families. Family Process, 44,77–91. http://dx.doi.org/10.
1111/j.1545-5300.2005.00043.x.
Wight, V. R., Price, J., Bianchi, S. M., & Hunt, B. R. (2009). The time use of teenagers. Social Science Research, 38, 792–809.
S. Offer / Journal of Adolescence 36 (2013) 577–585 585