HEALTH EDUCATION RESEARCH Vol.21 no.

6 2006
Theory & Practice Pages 862–871
Advance Access publication 10 October 2006

Is parenting style related to children’s healthy eating

and physical activity in Latino families?

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

Elva M. Arredondo*, John P. Elder, Guadalupe X. Ayala,
Nadia Campbell, Barbara Baquero and Susan Duerksen

Abstract associated with healthier behaviors. Moreover,

intervention researchers may want to encourage
Parenting styles influence a child’s risk for obesity. Latino parents to use less controlling styles with
The goals of this study are to evaluate the influence girls as this parenting style increased girls’ risk
of (i) parenting style on children’s health behaviors for unhealthy eating.
(physical activity and dietary intake), (ii) child-
ren’s sociodemographic characteristics on parent-
ing style and on children’s health behaviors and Introduction
(iii) parents’ sociodemographic characteristics on
their use of controlling styles to promote a healthy The obesity epidemic among children has been
home environment. Survey and anthropometric growing rapidly over the past 10 years, particularly
data were collected from a community sample in Latino children [1]. In 1999–2002, US national
of Latino parents (n 5 812) and their children data suggested that 26.3% of children ages 2–5
in kindergarten through second grade. Parental years and 38.9% of children ages 6–11 years were
use of positive reinforcement and monitoring was overweight or at risk for overweight [1]. Although
associated with children’s healthy eating and biological factors may influence children’s risk for
exercise. Also, parents’ use of appropriate disci- becoming overweight [2], compelling evidence
plining styles was associated with healthier eating, suggests that children’s contexts such as the home
while parental use of control styles was associated and school environments promoting unhealthy eat-
with unhealthy eating. The daughters of parents ing and exercise habits, precursors for the emer-
who used controlling styles ate more unhealthy gence of obesity [3]. Because children’s health
foods than did the sons. Older, employed and more habits develop in the home context and are highly
acculturated parents used less controlling styles dependent on parents’ actions and attitudes regard-
than their counterparts. Parenting interventions ing eating and exercise, the home environment
targeting children’s dietary intake and physical can have lasting effects on children’s weight
activity should encourage parents to use more trajectories [3].
positive reinforcement and monitor their child- A number of effective parenting styles can reduce
ren’s health behaviors as these parenting styles are the risk of childhood overweight. Parental monitor-
ing of children’s dietary intake and physical activity
has been associated with children’s health practices
[4, 5]. Also, parents who set appropriate limits (i.e.
Center for Behavioral and Community Health Studies, provide structure and boundaries) and consequate
Graduate School of Public Health, San Diego State
healthy eating with either tangible (e.g. stickers) or
University, 9245 Sky Park Court, Suite 221, San Diego,
CA 92123, USA
intangible (e.g. praise) reinforcers are more likely
*Correspondence to: E. M. Arredondo. to have children who eat healthy [6]. In contrast,
E-mail: earredondo@projects.sdsu.edu research shows that parents who are authoritarian

Ó The Author 2006. Published by Oxford University Press. All rights reserved. doi:10.1093/her/cyl110
For permissions, please email: journals.permissions@oxfordjournals.org
 Parenting style in Latino families

(i.e. highly directive, demanding and strict) regard- White/non-Hispanic, the parenting style of first
ing health behaviors increase their children’s risk for generation Mexican immigrants and US born Mex-
overweight. When considering children’s dietary ican American was more authoritarian than that

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

intake, observational, experimental and longitudinal of Mexican parents. Varela et al. [23] concluded
studies suggest that parents who use a controlling that Mexican descent parents living in the United
parenting style have children who are less likely to States may amplify their use of authoritarian styles
consume healthy foods, more likely to consume un- with their children as an adaptation to living in the
healthy foods and consume more food overall than United States. Several investigators have noted
children with parents who do not [7–18]. that the function of this parenting style is that it
The impact of parental styles on children’s health may help protect children from problem behavior
behaviors may depend, in part, on characteristics of [24]. However, additional research is needed to
the child. Studies suggest that overweight children shed light on how sociodemographic and cultural
and girls respond differently to a given parenting characteristics influence parents’ use of authoritar-
style than normal weight children or boys. Over- ian style.
weight children may respond more negatively to par- The overall goal of this study was to expand
ental control by eating more calories [19]. Similarly, current research evaluating the influence of effective
girls are likely to consume more energy when their and ineffective parenting styles associated with
mothers restrain or control their eating [10, 20]. obesity-related behaviors by including Latino fami-
Almost all studies examining the impact of parenting lies. The first aim of this study examined the influence
styles on children’s health practices have included of parental styles on children’s health behaviors,
primarily White/Anglo samples limiting the gener- controlling for parents’ sociodemographic factors.
alizability of these findings. Among children from Previous research involving an exclusively Latino
diverse ethnic backgrounds, there may be culturally sample and evaluating the role of parenting styles
mediated mechanisms that alter children’s inter- for promoting a healthy home did not control for
pretations and responses to their parenting styles. confounding variables such as parents’ sociodemo-
Understanding the health practices and parental graphic characteristics. We hypothesized that a
response of children from diverse cultural back- parenting style characterized by monitoring, use of
grounds will help researchers develop culturally reinforcement, appropriate limit setting and disci-
appropriate parenting intervention/prevention pro- pline (i.e. neither severe nor permissive) would be
grams targeting childhood obesity. positively associated with children’s healthy eating
Cultural values may provide a general or initial and physical activity. We also hypothesized that the
template for guiding parental decisions and social- use of a controlling style would be negatively
ization practices. Because parents’ attitudes toward associated with children’s healthy eating and physical
child rearing are influenced by cultural norms and activity. Moreover, as previous studies have found
sociocultural issues, parenting practices may differ that overweight children and girls respond differently
across ethnic groups. The Mexican culture has been to parenting styles, we examined the moderating
generally described as one emphasizing respect for role of children’s characteristics, namely body
authority where parents employ strict rules when mass index (BMI) and gender, on parenting styles
disciplining children [21, 22]. However, Mexican (monitoring, reinforcement, discipline, limit setting
American parents demonstrate a variety of styles, and control) and children’s health behaviors. Finally,
and socioeconomic factors associated with minority we examined the association between parents’
status (e.g. low educational and income level) rather sociodemographic variables, including accultura-
than affiliation to Mexican culture may instead ac- tion, and their use of a controlling parenting style
count for parents’ use of authoritarian styles [23]. In regarding children’s dietary intake and physical
a study comparing the parenting practices of two- activity, with the expectation that parents who are
parent Mexican, Mexican American families and unemployed, less educated and holding more

863
E. M. Arredondo et al.

traditional values will use a more controlling parent- Measures

ing style. Children’s dietary intake
A food frequency questionnaire was used to assess

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

children’s dietary intake. Food items included in the
Methods
survey were identified from previous studies with
the target population [25]. Parents were asked to rate
Participants
how often their child consumed each food item with
Parent–child dyads were recruited as part of a four- responses collapsed into the following categories:
arm-randomized community intervention trial called (i) never, (ii) at least once a month, (iii) at least once
‘Aventuras para Niños’ designed to implement and a week and (iv) everyday. The foods were catego-
evaluate an environmentally centered intervention rized by three independent reviewers as healthy or
study for Latino children grades K-2 and their unhealthy in terms of their contribution to risk for
families. The current study uses survey data collected obesity. Items in each category (e.g. dairy foods)
from the caretakers and anthropometric data on the were averaged to form a scale and each food group
children and caretakers. Thirteen schools in the (healthy versus unhealthy) was summed. ‘Healthy
southwest region of San Diego county, CA, USA, foods’ included (i) fruits and vegetables (e.g. 100%
averaging 6 miles from the US–Mexican border, were orange juice, green salad, fruit, vegetables—not
recruited to participate in the trial. To be eligible, potatoes), (ii) low-fat dairy foods (e.g. cheese, milk,
the schools had to have a minimum of 70% of Latino milk, yogurt), (iii) low-sugar cereals (e.g. oatmeal,
enrollment, no involvement in an obesity-related KixÒ, CheeriosÒ), (iv) wheat bread and (v) crack-
intervention program in the past 4 years and have ers. ‘Unhealthy foods’ included (i) regular soda,
a defined attendance boundary (i.e. not a charter or (ii) flavored drinks (e.g. Kool AidÒ), (iii) fats and
magnet school). Families of children enrolled in sweets (e.g. butter, salad dressing, potato chips) and
kindergarten to second grade, regardless of their (iv) sugar cereals (e.g. Lucky charmsÒ).
ethnicity, were recruited to participate if they met
the following criteria: (i) no major health problems,
(ii) residence within the school attendance bound- Children’s physical activity
aries and (iii) family did not intend to move away Parents rated their children’s physical activity level
from the area within a year. Families were recruited compared with other children. Answer choices were
via flyers, letters, telephone calls, face-to-face con- on a five-point Likert scale ranging from (1) ‘much
tact at the school and presentations at school events. less than others’ to (5) ‘much more than others’.
The single-item question was positively associated
with several other questions assessing children’s
Procedure
physical activity level, therefore, providing some
Parents completed a self-administered survey at evidence for construct validity: ‘During a typical
their children’s school; surveys were available week, how many days do you or another adult in the
in Spanish and English and took ;60 min to house participate in physical activities with your
complete. The children and their parents’ weight child?’ (r = 0.28, P < 0.001) and total number
and height were measured by trained research of sports and other organized activities engaged
assistants using portable scales and stadiometers. in by the child from a list of 16 activities (r = 0.17,
Following completion of the parent survey and P = 0.001).
anthropometric measurements, parent–child dyads
were given a $20 incentive. This protocol was
approved by the San Diego State University In- Parenting styles
stitutional Review Board to ensure the protection This scale assesses parents’ styles associated with
of human participants. children’s eating and activity. The 26-item scale

864
 Parenting style in Latino families

(14 for eating and 12 for limiting activity) was Table I. Items describing each of the parenting scales
developed for this project using qualitative and a
quantitative methods [26]. Focus group discussions

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

with 30 Latina mothers and a review of existing scales Monitoring
Diet How much do you keep track of the ...?a 0.76
used with Hispanic mothers (Child Feeding Ques-
Sweet snacks (candy, ice cream,
tionnaire [27, 28]) yielded a scale representing the cake) that your child eats?
following constructs: parental concern, limit setting, Salty snack foods (potato chips,
monitoring, reinforcement, discipline and control tortilla chips) that your child eats?
regarding children’s eating and activity. An explor- High-fat foods that your child eats?
Servings of fruits and vegetables your
atory factor analysis (EFA) was conducted with
child is eating
a sample of 240 Latina mothers yielding a scale How often must your child ask
with 33 items. Following the EFA, a confirmatory permission before ...?a
factor analysis (CFA) was conducted to assess Getting a snack
construct validity (n = 387). Because parental Activity How much do you keep track of N/A
the ...?a
concern was not related to any of the four constructs,
Amount of TV or videos your child is
a second CFA without parental concern was tested. watching?
All items loaded significantly to their respective Exercise your child is getting?
factors. The full parenting styles scale consisted of Discipline
five subscales (monitoring, limit setting, reinforce- Diet How often do you discipline 0.73
your child for doing the
ment, discipline and control) yielding good fit [Root
following without your
Mean Square Error of Approximation = 0.06, v2 permission?a
(279) = 2.79]. In the original scale, each factor Getting a snack
includes parenting styles associated with both child- Drinking a soda
ren’s activity and diet. For the purposes of the study How often must your child ask
permission before ...?
goals, however, questions regarding children’s eat-
Drinking a soda
ing (e.g. monitoring children’s eating) and activity Activity How often do you discipline N/A
were analyzed separately. For a description of the your child for doing the
items of each scale, please refer to Table I. following without your
Monitoring: Seven items (five for eating and two permission?a
Watching TV or videos
for activity) measured the frequency with which
Playing video games or the computer
parents’ monitored their children’s health behav- Control
iors. Discipline: Five items (three for eating and Diet How much do you agree or disagree 0.72
two for activity) measured the frequency with with each statement?b
which parents disciplined their children for un- Offer sweets (candy, ice cream, cake)
to my child as a reward for good
healthy eating (e.g. drinking soda) and engaging in
behavior
sedentary behaviors (e.g. watching TV) without My child should always eat all the
their permission. Control: Six items (five questions food on his/her plate
for eating and one question for activity) measured I have to be especially careful to
parents’ use of control styles. Limit setting: Six make sure my child eats enough
If my child says ‘I’m not hungry’,
items (two for eating and four for activity) assessed
I try to get him/her to eat anyway
parents’ use of appropriate boundaries with un- If I don’t regulate or guide my child’s
healthy eating and sedentary behavior. Reinforce- eating, he/she would eat much less
ment: Two items (one for eating and one for than he/she should
activity) measured parents’ use of praise when Activity I offer TV, videos, or video games to N/A
my child as a reward for good
their children ate healthy snacks or engaged in
behavior
activity.

865
E. M. Arredondo et al.

Table I. Continued Child and parent BMI

a BMI scores were calculated for each child and their
caregiver/parent. Weight was measured three times

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

Limit setting
to the nearest pound using a Health-o-Meter stan-
Diet How much do you agree or disagree N/A
with each statement?b dard scale with the participant standing without
I limit the amount of soda my shoes. Height was measured three times to the
child drinks nearest one-quarter of an inch, using a standard
I limit the number of snacks my portable stadiometer with shoes removed. Indices
child eats
of BMI were calculated using the Quetelet index
Activity I limit the amount of time my child 0.87
watches TV or videos during the (w hÿ2, [30]). Overweight status was determined by
week (Mon-Fri) using children’s gender-specific BMI—for age charts
I limit the amount of time my child from Centre for Disease Control (CDC’s) National
watches TV or videos during the Center for Health Statistics and adult BMI cutoff
weekend (Sat/Sun)
points for overweight and obesity from the CDC.
I limit the amount of time my child
plays video games (like Game boy,
Sega, Play station) or is on the Data analysis
computer during the week (Mon-Fri)
I limit the amount of time my child The linearity of the dependent variables modeled
plays video games (like Game boy, as continuous measures was assessed by visual
Sega, Play station) or is on the inspection of scatter plots of univariate models.
computer during the weekend
Multiple linear regression was used to examine the
(Sat/Sun)
Reinforcement influence of parents’ sociodemographic variables on
Diet How often do you ...?a N/A parenting styles. To perform these analyses, parents’
Praise your child for eating a healthy marital status was dichotomized as married (refer-
snack ence) versus not, employed (reference) versus not
Activity How often do you ...?a N/A
and attained a high school education or less (refer-
Praise your child for being physically
active ence) versus those who attained more education.
Children’s weight was categorized as underweight/
a
Five-point Likert scale from ‘never’ to ‘always’. bFive-point normal versus ‘at risk for overweight/overweight’.
Likert scale from ‘disagree’ to ‘agree’.
Parental reports of children’s physical activity,
healthy eating and unhealthy eating were regressed
Parents’ acculturation on each of the parenting styles controlling for
The Acculturation Rating Scale for Mexican potential confounding variables (age, marital status,
Americans (ARSMA-II) developed by Cuéllar education and employment). Income was not in-
et al. [29] consists of 30 items on a five-point cluded as a covariate due to significant collinearity
Likert-type response format, with 1 = ‘not at all’ to with employment (r = 0.43, P < 0.001) and age (r =
5 = ‘often’. Respondents receive a composite score 0.26, P < 0.001). To examine the moderating role
reflecting greater assimilation to the Anglo culture of children’s characteristics on parenting styles
with higher scores (a = 0.72). and children’s health behaviors, a series of main
effects with their corresponding interaction terms
Parents’ demographic variables were entered predicting children’s health behaviors
Participants were asked to provide information (healthy eating, unhealthy eating and physical activ-
about their age, gender, marital status, educational ity) separately. To examine the moderating role of
attainment, employment status, household income, children’s characteristics, namely BMI and gender,
years in the United States and country of origin if on parenting styles for activity and eating and on
born outside the United States. children’s health behaviors (eating and physical

866
 Parenting style in Latino families

activity), a series of main effects with their correspond- Table II. Parent and child demographic characteristics (n = 812)
ing interactions between the moderating variables Parent characteristics
and the predictors were estimated. Each model in-

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

cluded main effects and interaction terms (separated Mean age (SD) 34 (8)
Gender (%)
by parenting style for eating and activity by child
Female 96
characteristic), adjusting for confounding variables Marital status (%)
(parents’ age, marital status, employment and educa- Married 70
tion). Standardized betas were reported and an alpha Education (%)
of 0.05 was used for all statistical tests. All analyses High school or less 67
Employment (%)
were conducted using the Statistics Analysis Software
Employed 38
8.0 (1999; SAS Institute Inc., Cary, NC, USA). Mean income (per month) (%)
Participant characteristics are shown in Table II. <$1500 40
Mean years in the United States (SD) 16 (12)
Country of origin (%)
Mexico 71
Results Other 28
Weight status (%)a
The influence of parenting styles on Underweight 1
Normal weight 22
children’s health behaviors Overweight 33
Table III shows bivariate associations between Obese 40
parenting styles and children’s health behaviors,
Child characteristics
controlling for potential confounding variables (pa-
rental age, marital status, employment and educa- Mean age (SD) 6 (0.94)
tion). Parental monitoring for healthy eating was Gender (%)
positively related to children’s healthy eating and Girls 52
Weight status (%)b
negatively related to children’s unhealthy eating. Underweight 12
Parental monitoring for activity was positively re- Normal 55
lated to children’s activity. Parental use of reinforce- At risk of overweight 6
ment styles for healthy eating was positively related Overweight 26
to children’s healthy eating and negatively related to a
Categories recommended by CDC: <18.5 (underweight), 18.5–
children’s unhealthy eating. Parental use of rein- 24.9 (normal weight), 25.0–29.9 (overweight), >30 (obese).
b
forcement styles for activity was positively related to BMI adjusted for age and gender: <5th percentile
(underweight), >5th–85th (normal), >85th and <95th (at risk for
children’s physical activity. Parents’ use of appro- overweight), >95th (overweight).
priate disciplinary styles was positively related to
children’s healthy eating. Parental discipline was not significant main effects for the child’s gender (b =
significantly related to children’s unhealthy eating or ÿ0.50, P < 0.05) and parental control (b = ÿ0.17,
physical activity. However, parental control was P < 0.01) predicting unhealthy eating above and
positively related to children’s unhealthy eating. beyond the effects of the confounding variables.
Parental control was not significantly associated There was a significant two-way interaction be-
with children’s healthy eating or physical activity. tween children’s gender and parents’ use of con-
trolling styles predicting unhealthy eating (b =
Moderating role of child characteristics ÿ0.16, P < 0.05). The simple slope of each line
on parenting styles and child health was tested and found that the regression parameter
behaviors of girls was positively related to unhealthy diet (b =
Main effects and their corresponding interactions 0.17, P < 0.01; see Fig. 1). In other words, girls
were entered in each model. The data revealed were significantly more likely to eat unhealthy than

867
E. M. Arredondo et al.

Table III. Bivariate associations between parenting styles Child Gender by Parental Control Predicting Child
Unhealthy Eating
subscales and children’s health behaviors 10

Unhealthy Eating
Healthy Unhealthy Physical 8

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

eatinga (b) eatingb (b) activity (b)
6
girls
Monitoring 0.45*** ÿ0.17** 0.19*** boys
4
Reinforcement 0.32*** ÿ0.08** 0.13***
Discipline 0.20* ÿ0.004 0.04 2
Limit setting 0.08 ÿ0.05 0.03 0
Control ÿ0.07 0.10** ÿ0.02 low
Control
high

Models adjusted for parents’ age, marital status, employment Fig. 1. The moderating role of children’s gender on controlling
and education. parenting styles and children’s unhealthy eating. Model adjusted
a
Fruits and vegetables, low-fat dairy foods, low-sugar cereals, for parents’ age, marital status, employment and education.
wheat bread and crackers. bRegular soda, flavored drinks,
fats and sweets and sugar cereals. *P < 0.05, **P < 0.01,
***P < 0.001. Child Gender by Parental Limit Setting
Predicting Unhealthy Eating
8

Unhealthy Eating 6
boys when parents’ used more control strategies
girls
for eating. Moreover, there was a significant two- 4
boys
way interaction between children’s gender and 2
parents’ use of limit setting in predicting unhealthy
0
eating (b = ÿ0.18, P < 0.02; see Fig. 2). The simple low high
Limit Setting
slope of each line was tested and found that the
regression parameter of boys was negatively related Fig. 2. The moderating role of children’s gender on limit setting
to unhealthy eating (b = ÿ0.15, P < 0.01). In other and children’s unhealthy eating. Model adjusted for parents’
words, boys ate significantly less unhealthy than age, marital status, employment and education.
girls when their parents engaged in limit setting
regarding food. No other interactions were evident
Table IV. Associations between parents’ demographic
between children’s characteristics and parenting characteristics and parental control strategies
styles (for eating and activity) predicting children’s
health behaviors. Parent Control for Control for
characteristics eating activity
Demographic influences on parenting Age ÿ0.01** 0.006
controlling style subscale Marital statusa 0.14 0.001
To investigate the independent association of pa- Educationb ÿ0.09 ÿ0.01
Employmentc 0.25** 0.24*
rents’ sociodemographic variables and parental
Acculturation ÿ0.09** ÿ0.17***
controlling style subscale, a multiple linear regres-
sion was conducted. Table IV shows the associations Predictors for each DV were entered simultaneously in one
between parents sociodemographic characteristics block.
a
‘0’ = married/living as married, ‘1’ = divorced/widowed/
and controlling parenting style. The omnibus test separated/never been married. b‘0’ = high school or less,
for the association between parental control and ‘1’ = greater than high school. c‘0’ = employed, ‘1’ = not
child eating was rejected [F (5,696) = 6.46, P < employed.
0.001, R2 = 0.04]. Parents’ who were younger,
unemployed and less acculturated were significantly [F (5,701) = 4.37, P < 0.001, R2 = 0.03]. Parents
more likely to use a controlling style regarding their who were not employed and were less acculturated
children’s eating. In terms of parental control for were significantly more likely to report using a con-
children’s activity, the omnibus test was also rejected trolling style for their children’s activity.

868
 Parenting style in Latino families

Discussion [10, 20], girls ate more unhealthy food when their
parents engage in a controlling parenting style
One of the primary study goals was to examine the regarding their eating. Several explanations may

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

influence of parenting styles on children’s health account for these findings. One interpretation may
behaviors. Our findings suggest that the children of be that boys and girls differ in their awareness of
parents who monitored and reinforced healthy be- and responsiveness to maternal restriction. Alterna-
haviors ate more healthy foods and less unhealthy tively, it may be that parents treat girls differently, but
foods, and were more physically active compared are not aware of it. In an observational study, Olvera-
with the children of parents who did not use these Ezzell et al. [32] found that Latino mothers engaged
parenting styles. These findings are consistent with in more restraining behaviors with their daughters
previous research [6, 31]. In a longitudinal study, than with their sons. If indeed, mothers treat girls
Faith et al. [31] found that parents who monitored differently than boys, their behaviors may be influ-
their children’s fat intake were less likely to have enced by several factors such as the concern that their
children who were overweight 2 years later than daughters will become overweight and likely face
parents who did not monitor their children’s diet. health or self-esteem issues and the possibility of
Parental monitoring of children’s sedentary activi- teasing by their peer group; these concerns may
ties such as television viewing may help protect reflect society’s pressures toward thinness, particu-
larly among women and young adolescent females
children from becoming overweight given the links
[34, 10]. Regardless of parents’ motives, it is
among television watching, inactivity and increased
apparent from our findings that, at least with girls,
caloric consumption [4]. The use of reinforcement
control styles are counterproductive and may be
techniques by parents has been shown to cause an
placing them at risk for overweight.
increase in children’s healthy snack intake [6].
In comparison, boys were less likely to eat
Moreover, study finding suggests that parenting
unhealthy when their parents set appropriate limits
styles characterized by the use of appropriate dis-
regarding eating. These findings suggest that overly
cipline was associated with eating healthier foods. restrictive parenting has deleterious effects on their
These results held even after controlling for poten- children’s eating, but parenting characterized as lax
tially confounding variables such as parents’ age, and little limit setting is also ineffective, particularly
marital status, employment status and education. with boys. When examining the moderating in-
Consistent with previous studies involving primar- fluence of children’s characteristics on the relation-
ily White middle- and upper-class families, these ship between each of the parenting styles and
parental styles appear to reduce the risk of over- children’s health behaviors, children’s BMI did not
weight among Latino children [4, 5]. moderate this relation. These findings are in contrast
Parents who engage in a controlling parenting to those found by Birch and Fisher [19]. Although
style regarding children’s eating had children who mothers of children who were overweight were more
ate more unhealthy foods. These findings are con- likely to accurately report that they were overweight
sistent with those reported by Olvera-Ezzell et al. (data not shown), it is possible that given the per-
[32] who found that parents who used threats and ception that better health is ascribed to heavier weight
bribes had children who consumed less healthy [35], the children’s weight status may not impact
foods. Also, Patrick et al. [33] found that, after ad- parenting style related to diet and physical activity.
justing for caregiver education, African–American Another study aim was to evaluate the influence
and Hispanic parents who exhibited an authoritar- of parents’ sociodemographic variables on their use
ian feeding style had children who consumed less of control or authoritarian styles for children’s
fruits and vegetables. Parental control regarding eating and activity. Our study findings provide a
children’s eating impacted daughters’ eating com- mixed picture as to whether sociodemographic
pared with sons. Consistent with previous research correlates of minority status (e.g. low education

869
E. M. Arredondo et al.

and income) or aspects of Latino culture influence behaviors associated with obesity. If replicated and
parental styles characterized as authoritarian. When confirmed, findings from the current study may help
considering children’s eating, parents who were inform the development of culturally appropriate

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

younger, unemployed and less acculturated reported interventions. Intervention researchers may want to
using controlling or authoritarian styles. These focus on teaching parents how to monitor and
findings suggest that, in addition to acculturation, positively reinforce their children’s health behav-
sociodemographic factors associated with minority iors. For example, parents could be encouraged to
status in the United States were also related with reward their children’s healthy eating and physical
having an authoritarian parenting style. In compar- activity by providing them with more tangible and
ison, parents who were unemployed and less ac- intangible reinforcers. Moreover, parents could be
culturated were more likely to use parenting styles to encouraged to be mindful of the costs associated
promote physical activity characterized as authori- with the using a controlling parenting style, partic-
tarian. In both health behaviors, parents who were ularly the association with unhealthy eating practi-
less acculturated engaged in more authoritarian ces among daughters. For instance, parents may
styles, independent of other sociodemographic char- want to avoid being rigid and forcing their daughters
acteristics. These results are consistent with those to eat healthy foods and avoid unhealthy foods.
showing evidence that Latino parenting styles may It is crucial that interventions focus on the home
have aspects of authoritarian parenting [22], but environment and the types of parenting styles asso-
more studies are needed to confirm this. ciated with the risk of obesity because children’s
There are several limitations that merit noting. physical activity and dietary habits form in this
Due to the correlational nature of the study, it is not context.
possible to infer causation. For instance, we cannot
state whether each of the parenting styles subscales
cause healthy or unhealthy eating by their children Acknowledgements
or vice versa. Additionally, our measure of parent-
ing styles regarding children’s dietary intake and We would like to thank Kathleen Merchant and
activity is not identical to those published; there- Michelle Zive for their contribution to the de-
fore, our findings may not be comparable to other velopment of the food frequency questionnaire.
studies. Moreover, our measure of children’s phys- This effort was funded by the National Heart Lung
ical activity is limited to parents’ report of child- and Blood Institute (5R01HL073776) and the
ren’s activity compared with other children rather American Cancer Society (PFT-04-156-01-CPPB).
than an estimate of actual physical activity per
day. Despite these limitations, there were several
strengths. The large sample size in the present study Conflict of interest statement
augments the limited amount of research examining
the influence of parental styles on children’s health None declared.
practices involving Latino families. Our measure of
parenting styles captures both obesity-related behav-
iors diet and activity—thus providing additional References
evidence for the importance of parenting styles on
risk for obesity. Moreover, we augment previous 1. Ogden CL, Carroll MD, Johnson CL. Prevalence and trends
in overweight among U.S. children and adolescents, 1999–
studies by examining the parenting practices of fam- 2000. J Am Med Assoc 2002; 288: 1728–32.
ilies with a younger age group of children. Previous 2. Farooqi SI, O’Rahilly S. Recent advances in the genetics of
studies have largely included older children. severe childhood obesity. Arch Dis Child 2000; 83: 31–4.
3. Davison KK, Birch LL. Childhood overweight: a contextual
In conclusion, it is clear that parenting styles play model and recommendations for future research. Obes Rev
an important role in shaping children’s health 2001; 2: 159–71.

870
 Parenting style in Latino families

4. Gortmaker SL, Must A, Sobol AM et al. Television viewing 22. Cardona PG, Nicholson BC, Fox RA. Parenting among
as a cause of increasing obesity among children in the Hispanic and Anglo-American mothers with young children.
United States, 1986–1990. Arch Pediatr Adolesc 1996; 150: J Soc Psychol 2000; 140: 357–65.
356–62. 23. Varela R, Vernberg EM, Sanchez-Sosa JJ et al. Parenting

Downloaded from https://academic.oup.com/her/article-abstract/21/6/862/609279 by CSU San Marcos user on 05 December 2018

5. Dennison BA, Erb TA, Jenkins PL. Television viewing and style of Mexican, Mexican American, and Caucasian-Non-
television in bedroom associated with overweight risk Hispanic families: social context and cultural influences.
among low-income preschool children. Pediatrics 2002; J Fam Psychol 2004; 18: 651–7.
109: 1028–35. 24. Mason CA, Cauce AM, Gonzales N et al. Neither too sweet
6. Stark LJ, Collins FL Jr, Osnes PG et al. Using reinforcement nor too sour: problem peers, maternal control, and problem
and cueing to increase healthy snack food choices in behavior in African American adolescents. Child Dev 1996;
preschoolers. J Appl Behav Anal 1986; 19: 367–79. 67: 2115–30.
7. Satter EM. Internal regulation and the evolution of normal 25. Zive MM, Frank-Spohrer GC, Sallis JF et al. Determinants
growth as the basis for prevention of obesity in children. of dietary intake in a sample of white and Mexican-
J Am Diet Assoc 1996; 96: 860–4. American children. J Am Diet Assoc 1998; 98: 1282–9.
8. Birch LL, Fisher JO. Development of eating behaviors among 26. Larios SE, Ayala GX, Baquero B et al. Development of
children and adolescents. Pediatrics 1998; 101: 539–49. Survey that Measures Home Environmental Factors Con-
9. Francis LA, Hofer SM, Birch LL. Predictors of maternal tributing to Obesity. Washington, DC: American Public
child-feeding style: maternal and child characteristics. Health Association, 2004.
Appetite 2001; 37: 231–43. 27. Birch LL, Fisher JO, Grimm-Thomas K et al. Confirmatory
10. Fisher JO, Birch LL. Restricting access to foods and factor analysis of the Child Feeding Questionnaire: a measure
children’s eating. Appetite 1999; 32: 405–19. of parental attitudes, beliefs and practices about child
11. Costanzo PR, Woody EZ. Domain-specific parenting styles feeding and obesity proneness. Appetite 2001; 36: 201–10.
and their impact on the child’s development of particular 28. Anderson CB, Hughes SO, Fisher JO et al. Cross-cultural
deviance: the example of obesity proneness. J Soc Clin equivalence of feeding beliefs and practices: the psycho-
Psychol 1985; 3: 425–45. metric properties of the child feeding questionnaire among
12. Kremers SP, Brug J, de Vries H et al. Parenting style and Blacks and Hispanics. Prev Med 2005; 41: 521–31.
adolescent fruit consumption. Appetite 2003; 41: 43–50. 29. Cuellar I, Arnold B, Maldonado R. Acculturation Rating
13. Birch LL. Development of food acceptance patterns in the Scale for Mexican Americans-II: a revision of the original
first years of life. Proc Nutr Soc 1998; 57: 617–24. ARSMA scale. Hisp J Behav Sci 1995; 17: 275–304.
14. Birch LL, Fisher JO. Mothers’ child-feeding practices 30. Garrow JS, Webster J. Quetelet’s index (W/H2) as a measure
influence daughters’ eating and weight. Am J Clin Nutr of fatness. Int J Obes 1985; 9: 147–53.
2000; 71: 1054–61. 31. Faith MS, Berkowitz RI, Stallings VA et al. Parental feeding
15. Carper JL, Orlet Fisher J, Birch LL. Young girls’ emerging attitudes and styles and child body mass index: prospective
dietary restraint and disinhibition are related to parental analysis of a gene-environment interaction. Pediatrics 2004;
control in child feeding. Appetite 2000; 35: 121–9. 114: e429–36.
16. Fisher JO, Birch LL. Restricting access to palatable foods 32. Olvera-Ezzell N, Power TG, Cousins JH. Maternal so-
affects children’s behavioral response, food selection, and cialization of children’s eating habits: strategies used by
intake. Am J Clin Nutr 1999; 69: 1264–72. obese Mexican-American mothers. Child Dev 1990; 61:
17. Birch LL, McPhee L, Shoba B et al. ‘‘Clean up your plate’’: 395–400.
effects of child feeding practices on the conditioning of meal 33. Patrick H, Nicklas TA, Hughes SO et al. The benefits of
size. Learn Motiv 1987; 18: 301–17. authoritative feeding style: caregiver feeding styles and
18. Fisher JO, Birch LL. Parents’ restrictive feeding practices children’s food consumption patterns. Appetite 2005; 44:
are associated with young girls’ negative self-evaluation of 243–9.
eating. J Am Diet Assoc 2000; 100: 1341–6. 34. Pike KM, Rodin J. Mothers, daughters, and disordered
19. Birch LL, Fisher JO. Mothers’ child-feeding practices eating. J Abnorm Psychol 1991; 100: 198–204.
influence daughters’ eating and weight. Am J Clin Nutr 35. Crawford PB, Gosliner W, Anderson C et al. Counseling
2000; 71: 1054–61. Latina mothers of preschool children about weight issues:
20. Johnson SL, Birch LL. Parents’ and children’s adiposity and suggestions for a new framework. J Am Diet Assoc 2004;
eating style. Pediatrics 1994; 94: 653–61. 104: 387–94.
21. Fromm EM, Maccoby M. Social Character in a Mexican
Village. Englewood Cliffs, NJ: Prentice-Hall, 1970. Received on February 1, 2006; accepted on August 24, 2006

871