Next Article in Journal
Anti-Inflammatory and Anti-Oxidative Synergistic Effect of Vitamin D and Nutritional Complex on Retinal Pigment Epithelial and Endothelial Cell Lines against Age-Related Macular Degeneration
Next Article in Special Issue
Involving Parents in Promoting Healthy Energy Balance-Related Behaviors in Preschoolers: A Mixed Methods Impact and Process Evaluation of SuperFIT
Previous Article in Journal
Creatine Supplementation for Patients with Inflammatory Bowel Diseases: A Scientific Rationale for a Clinical Trial
Previous Article in Special Issue
Obesogenic Programming Effects during Lactation: A Narrative Review and Conceptual Model Focusing on Underlying Mechanisms and Promising Future Research Avenues
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Family Social Capital: Links to Weight-Related and Parenting Behaviors of Mothers with Young Children

Nutritional Sciences Department, Rutgers University, New Brunswick, NJ 08901, USA
*
Author to whom correspondence should be addressed.
Nutrients 2021, 13(5), 1428; https://doi.org/10.3390/nu13051428
Submission received: 13 March 2021 / Revised: 19 April 2021 / Accepted: 19 April 2021 / Published: 23 April 2021
(This article belongs to the Special Issue Weight Outcomes of Kids: What Can Parents Do?)

Abstract

:
Family social capital includes the social relationships, values, and norms shared by a family and is positively linked with children’s mental and physical health status. This cross-sectional study addresses a gap in the literature related to family social capital vis-à-vis weight-related behaviors and home environments of 557 mothers and their young children (ages 2 to 9 years). Mothers completed an online survey comprised of valid, reliable questionnaires assessing family relationships and weight-related behavioral and home environment measures. The measures that determined family social capital (i.e., supportive, engaged parenting behaviors; family cohesion; family conflict; and family meal frequency) yielded distinct tertile groups that differed significantly (p < 0.001) on every family social capital measure with large effect sizes. Analysis of variance with Tukey post-hoc test revealed greater family social capital was linked to significantly better maternal health, dietary intake, physical activity, and sleep behavior. Additionally, maternal modeling of healthy eating and physical activity, child feeding practices, and home environments was higher in groups with greater family social capital. Child mental and physical health, physical activity, and sleep quality were better in families with greater family social capital. Findings suggest greater family social capital is linked to healthier weight-related behaviors and home environments. Future intervention studies should incorporate strategies to build family social capital and compare longitudinal outcomes to traditional interventions to determine the relative value of family social capital on health behaviors.

1. Introduction

The obesity epidemic is a serious worldwide concern that impairs the health and wellbeing of millions of adults and children. Obesity prevention is crucial, especially for children because children obesity is positively correlated with obesity in adulthood and an increased risk for comorbidities [1]. Parents play a vital role in childhood obesity prevention in that they have great influence on children’s weight-related behaviors, such as diet, physical activity, and sleep [2,3,4]. Parental influence may take many forms, such as modeling behaviors, establishing household routines, and creating home environments that support or dissuade behaviors. Family social capital is another type of influence that affects children [5,6].
Capital is typically thought of as economic (e.g., money, wealth), material (e.g., tools, possessions), and human (e.g., education, skills). Another type of capital, known as social capital, was conceptualized in Coleman’s social capital theory as the interpersonal relationships within social groups, such as families, neighborhoods, and communities, that facilitate access to resources (e.g., information, advice) needed to accrue resources and achieve certain goals [7]. The quality and amount of social capital available is positively correlated with the possibility of reaping benefits and reaching goals [8]. The effect of social capital is considered particularly crucial to providing children with access to opportunities that support their optimal development and positive outcomes [5,6,8,9,10]. Children’s social capital is transmitted primarily by the family [7].
Family social capital is built in the home and considered one of the most powerful, enduring types of social capital [11]. It includes the social relationships, values, and norms shared by a family and is created through emotionally warm home environments that promote parent:child engagement, family cohesion, and lasting affectionate attachments [6,12]. Family social capital is also developed by teaching children behavioral norms that facilitate their successful integration into other social structures, such as schools, workplaces, and the community [12]. For children, their social capital is clearly associated with the degree to which they can access, trust, and benefit from parents’ human capital (i.e., parent education, skills, values, social competence). Factors that can compromise family social capital include physical, emotional, and/or mental disengagement between parents and children (e.g., parental absence from the home, extensive parental employment commitments, parents’ preference for adult pursuits, parent illness, children’s extensive use of media or engagement with social groups outside the family), household conflict, and/or lack of clearly communicated and reinforced rules about acceptable, prosocial norms and behaviors [13,14].
Family social capital influences the health and well-being of children [5,6]. For example, a comprehensive systematic review and meta-analysis found 55 studies investigating relationships between family social capital and children’s mental health and problem behaviors [5]. The researchers concluded that family social capital was positively linked with children’s self-esteem and negatively associated with anxiety, depression, suicide ideation, and engaging in anti-social behaviors such as aggression and defiant actions [5]. The 21 studies investigating general health status that were analyzed in this same systematic review revealed that children and adolescents in families with higher levels of social capital had better overall health status, higher quality of life, and greater wellbeing [5].
Despite the obesity epidemic and the intense research effort to identify factors that protect against excess weight gain, little attention has focused on associations between family social capital and weight-related behaviors, such as dietary intake and physical activity [5,15]. In fact, McPherson et al.’s systematic review of family social capital’s influence on children located just two studies addressing nutrition, three studies examining physical activity, and two studies investigating weight status [5]. The systematic review conducted by Alvarez et al. identified a single study related to nutrition behaviors in children and family social capital [6]. The scant evidence available suggests that higher levels of family social capital are associated with better nutritional outcomes, greater physical activity, and better weight status [5,6].
It is surprising that so few studies have considered how family social capital is associated with weight-related behaviors. Equally surprising is the lack of attention given to family social capital and parental weight-related behaviors given parents are children’s role models, are family food gatekeepers, and create the structure/lifestyle environment within the home; thereby, strongly influencing weight-related behaviors of children that track into adulthood [2,3,4,16,17,18,19,20,21,22]. Thus, this exploratory, secondary analysis reported here was conducted in response to the dearth of research related to family social capital vis-à-vis weight-related behaviors and home environments of parents and children. Based on prior work [5,6,15], it was hypothesized that family social capital would be significantly associated with positive weight-related behaviors and healthy home environments of mothers with young children. A sound understanding of the associations between family social capital and health, weight-related behaviors, and home environments has the potential to inform the development of more effective health, nutrition, and obesity prevention interventions.

2. Materials and Methods

The Institutional Review Board at the authors’ university approved this investigation (Protocol #11-294). This secondary analysis used data collected at baseline (pre-randomization) for the HomeStyles randomized controlled trial [23,24].

2.1. Sample

Participants were recruited with electronic announcements, flyers, and in-person invitations to participate in a program to “build even happier, healthier, safer families”. Recruitment announcements were distributed in locations frequented by parents, such workplace listservs, community settings (e.g., farm markets), and school-related activities. A study recruitment company also assisted with participant recruitment. All participants gave informed consent and were compensated $15 after completing the study baseline survey.
Eligibility criteria for this secondary analysis were aged 20–45 years, parent of one or more children between the ages of 2 to <9 years, ability to read English, made all or most decisions about family food choices, lived in study defined New Jersey or Arizona (study catchment area), had regular access to the Internet, and gave credible answers (e.g., did not answer all questions in a series similarly). Fathers were not included in this analysis a due to too few responses. Of the 5494 individuals who visited the study survey website, 557 met all eligibility criteria, gave informed consent, and completed the entire survey.

2.2. Instruments

Data were collected using the “Home Obesogenicity Measure of EnvironmentS” (HOMES) online survey. This survey was comprised of valid, reliable questionnaires described in detail elsewhere [23,24,25]. To summarize, participants reported sociodemographic characteristics (e.g., age, race/ethnicity, highest education completed children under age 18 years in the household, maternal employment, family affluence). Participants also completed scales assessing maternal and child health and weight-related behaviors, parenting behaviors, and family social capital.
Maternal and child health status was evaluated using the Health Quality of Life assessments for health status, physical health, and mental health from the Centers for Disease Control and Prevention [26,27]. The Patient Health Questionnaire (PHQ-2) assessed maternal depression severity [28]. Maternal stress level was assessed with Cohen’s brief Perceived Stress scale (PSS-2) [29].
Maternal and child weight-related behaviors included dietary intake, physical activity, and sleep. Dietary intake behaviors focused on an indicator of healthy behaviors (i.e., fruit/vegetable servings/day) and an indicator of unhealthy choices (i.e., sugar-sweetened beverage servings/day). Block’s Fruit/Vegetable Screener assessed daily fruit/vegetable intake for mothers and daily fruit/vegetable juice intake for children [30]. Daily servings of sugar-sweetened beverages (i.e., soft, fruit, tea, coffee, and energy drinks) were measured using the HOMES Drinks Intake Screener [25]. The HOMES Physical Activity Questionnaire assessed physical activity level with sedentary behavior evaluated using an indicator item of total sedentary screen time duration daily [25,31]. Pittsburgh Sleep Quality Index components evaluated overall sleep quality and total duration of sleep each night [32,33].
Parenting behaviors focused on modeling behaviors, child feeding practices, and home environment conditions [23,24,25]. Mothers reported their modeling of healthy eating and physical activity behaviors to children. Child feeding practices evaluated were restriction of children’s access to low nutrient density foods, pressure on children to eat healthy foods, and instrumental feeding (i.e., offering children palatable food as a reward for eating healthy foods). Home physical environment conditions included physical activity and food availability. The HOP-Up questionnaire was used to evaluate physical activity space and supports inside homes and in the outdoor areas immediately outside homes [34]. Household availability of 100% fruits/vegetable juice and sugar-sweetened beverages were assessed with the Household Food Supplies Questionnaire [35].
Family social capital is evaluated in various ways, with no standard method for determining it [6,36,37]. The measures used in this study were modeled on previous research [6,12,36,38,39,40,41,42,43] and included family social environment indicator scales assessing supportive, engaged parenting behaviors; family cohesion; family conflict; and frequency of family interactions. The mother:child verbal engagement scale contained 2 items that assessed the frequency mothers talked with children. The single item mother:child physical engagement scale evaluated physical warmth (i.e., I give my kids lots of hugs and kisses). Both engagement scales were modeled on the Earls et al.’s home and life interview survey [44]. Family cohesion and family conflict constructs from the brief Family Environment scale assessed feelings of family support and togetherness (e.g., there is a feeling of togetherness in our family) and family disapproval and disputes (e.g., family members often criticize each other) with 2 and 3 items, respectively [45,46,47]. The Family Meal Frequency scale was analogous to measures used by De Clercq et al. [39]. This scale assessed how many times each week families interacted by eating meals together (breakfast, lunch, and dinner). Family meals are considered a “fundamental aspect of family life that offers the opportunity of socialization” [39], with frequency of these meals considered a gauge of family closeness [48]. All family social capital scales, except Family Meal Frequency, had these 5 answer choices: Strongly disagree, disagree, neither agree nor disagree, agree, strongly agree, scored 0 to 4, respectively, with scoring reversed for negatively worded statements. The Family Meal Frequency questionnaire included all meals (breakfast, lunch, and dinner) eaten as a family each week with answers scored as 0 to 4, respectively, if families ate together <5 meals/week, ≥5 to <9 meals/week, ≥9 to <13, ≥13 to <17 meals/week, or ≥17 meals/week. For all family social capital scales, a higher scale score indicates greater expression of the characteristic. The total family social capital score was calculated by summing the score for each of the family social environment scales (possible score range from 0 to 20).

2.3. Data Analysis

Mothers were partitioned into tertiles based on their total family social capital score. Descriptive statistics for all variables by tertiles were conducted along with determining the internal consistency of scales. ANOVA and Tukey post-hoc tests were conducted to determine how mothers’ weight-related behaviors, parenting behaviors, and health characteristics significantly differed among and between tertiles. To reduce the Type I error risk caused by multiple comparisons, the Benjamini Hochberg procedure was applied at the 5% level for 2-tailed tests, generating a probability level for ANOVA main effects of p ≤ 0.03 to reduce the risk of type I errors [49]. Probability for post-hoc pairwise comparisons was set at p< 0.05. Effect size of ANOVA main effects was determined by calculating partial eta-squared values. Effect sizes of 0.01, 0.06, and 0.14 indicated small, medium, and large effects, respectively [50]. Analytic procedures were completed with SPSS software version 27.0 (IBM Corporation, Chicago, IL, USA).

3. Results

Mothers in the study had a mean age of 32.65 ± 5.53 SD years and were mostly (60%) White. The majority (86%) had at least some post-secondary education. On average, family affluence was moderate (mean = 5.45 ± 1.74 SD on a 0 to 9 scale) and 36% of mothers did not have paid employment. Mothers had an average of 2 children and most (82%) lived in dual parent households.
Mothers were grouped into tertiles based on their total family social capital scores. As shown in Table 1, all tertiles differed significantly from each other on all family social capital scales with large effect sizes, thereby indicating distinct groups. As expected, scores on each scale were significantly higher in upper compared to lower tertiles indicating that family social capital was greater as tertile level incremented. Tertiles did not differ by maternal age, ethnicity/racial distribution, education level, number of children in the household, or number of parents in the household (Table 2). Mothers in Tertile 2 were significantly more likely to work full-time than Tertile 3, with a very small effect size. Tertile 1 had significantly lower family affluence than Tertile 3 (5.17 ± 1.76 SD vs. 5.71 ± 1.72 SD, an approximately 5% difference), but the effect size was small.
Data for maternal health status measures are displayed in Table 3. Higher tertile mothers tended to have significantly better health status, fewer days of “not good” physical and mental health, and lower depression severity and perceived stress than lower tertile comparators; effect sizes were small. The dietary behaviors revealed that Tertile 1 mothers tended to eat fewer fruits and vegetables and drink more sugar-sweetened beverages than other mothers. High tertile mothers tended to have significantly higher physical activity levels and, though not significant, less sedentary activity, than lower tertile mothers. Sleep quality and sleep duration were positively correlated with higher tertile assignment, with Tertile 3 scoring significantly higher than Tertile 1 mothers. However, effect sizes for dietary, physical activity, and sleep behaviors were small.
An examination of parenting behavior findings indicates that maternal modeling of healthy eating and modeling of physical activity was significantly different among and between tertiles with the highest scores in Tertile 3 and medium effect sizes (see Table 3). The child feeding practices scale scores indicated no significant differences in restriction across tertiles; however, the use of pressure was lower in higher tertile assignment, with significant differences between all pairwise comparisons and a medium effect size. Additionally, Tertile 1 was significantly more likely to use food rewards than Tertile 3, yielding a small effect size.
The home physical environment revealed that space and supports for physical activity inside the home as well as in the area outside the home was significantly higher in upper versus lower tertiles, with significant differences between all pairwise comparisons with small effect sizes (see Table 3). The home fruit/vegetable availability differed significantly among and between all pairwise comparisons with amounts greater in upper tertiles and a medium effect size. Sugar-sweetened beverage availability in households did not differ significantly among tertiles.
The child health findings displayed in Table 4 indicate that mothers’ rated children’s health as very good to excellent with the ratings being significantly greater in higher tertile assignments, with a medium effect size. Children in Tertile 3 tended to have fewer “not good” mental and physical health days than comparators. Children’s dietary intake did not differ significantly among tertiles; however, greater family social capital was associated with higher fruit/vegetable juice intake and lower sugar-sweetened beverage intake. Physical activity level was greater in higher tertiles, with Tertile 1 getting significantly less physical activity than those in Tertile 3; the effect size was small. Screen time and sleep duration did not differ significantly among tertiles. Sleep quality was significantly higher in families with greater family social capital, with Tertile 3 scoring significantly higher than other tertiles.

4. Discussion

This study aimed to expand the understanding of associations of family social capital with weight-related behaviors and home environments of families with young children and the potential value of incorporating strategies for building family social capital in health, nutrition, and obesity prevention interventions. The measures used to determine family social capital (i.e., supportive, engaged parenting behaviors; family cohesion; family conflict; and family meal frequency) yielded distinct tertile groups that differed significantly on every family social capital measure with large effect sizes. These differences were independent of sociodemographic characteristics. The findings indicate that greater family social capital is linked to better maternal health, dietary intake, physical activity, and sleep behaviors. In addition, maternal modeling of healthy eating and physical activity, child feeding practices, and home environments improved as family social capital increased. Child health status, mental health, physical activity, and sleep quality were also better in families with greater family social capital.
Family social capital is built through positive, supportive family relationships that teach values and societal norms [6,12]. Findings of this study suggest family social capital also confers health benefits to mothers and children which supports prior work among adolescents in Russia, Croatia, and Taiwan [51,52,53]. Mothers in families with greater social capital had better mental health, which likely offered protection to their children [54,55,56]. For instance, Reynolds and Crea reported that among U.S. families, parent depression and anxiety increased the vulnerability of adolescent children for engaging in anti-social behaviors [57]. Social capital, in the form of family cohesion, is also inversely related with disordered eating behaviors that may escalate into psychological disorders such as Anorexia Nervosa [58]. Further evidence of the effects of family social capital on children’s mental health is provided by Springer et al.’s report that close family relationships reduced the odds of teens in El Salvador from engaging in aggressive behaviors, risky behaviors such as binge drinking, drug use, and sexual relationships, and having suicidal thoughts [59].
The findings of the study reported here indicate that higher family social capital is associated with healthier maternal dietary intake, maternal role modeling of healthy eating, and household food supplies. The lack of differences in dietary intake in children contrasts with a prior report that found greater family capital in the form of family communication and role modeling was associated with better nutritional practices in Taiwan adolescents [53]. Contradictory findings may have been due to differences in measures used to assess dietary intake between studies. Similarly, a higher level of family belongingness was linked with greater fruit and vegetable intake in adolescents living in the United Kingdom [60]. The tandem increase in household availability of fruits/vegetables and family social capital supports previous work in the Philippines which found that primary food decision makers in more cohesive families had a greater propensity to make healthy food choices [6,61].
In the present study, maternal modeling of physical activity, as well as space and supports for physical activity increased as family social capital rose. In addition, physical activity level of both mothers and children tended to be higher in families with more social capital. These findings parallel reports from Taiwan where adolescents in families with better communication and adult role models had better physical activity outcomes [53]. Similarly, higher levels of family social capital were associated with regular overall physical activity in teenage Croatian boys, however this association was not observed in girls [62].
Mothers in the highest family social capital tertile group had higher quality sleep and longer sleep duration than those with the least family social capital. In fact, those in the highest family social capital group were the only mothers who met sleep recommendations of 7 to 9 h/night [63]. Children’s sleep duration did not differ among family social capital tertiles. However, children in the highest family social capital tertile had the best sleep quality, which aligns with a study of Canadian parents where greater social capital (as measured by occupation) had fewer sleep disturbances than comparators [64].
The supportiveness of the home environment for healthy eating and physical activity behaviors increased with family social capital in the study reported here. For example, mothers’ child feeding practices became more congruent with recommendations as family social capital increased. That is, restricting children’s food choices, pressuring them to eat, and using highly palatable food (e.g., sweets) to reward them for eating healthy foods (e.g., vegetables) are not recommended because they can promote or exacerbate feeding problems such as picky eating. Additionally, these feeding behaviors are not recommended because it may contribute to children’s development of an insensitivity to physiological signals of hunger and satiety that help regulate food intake amounts as well as development of preferences for highly palatable foods and reduced preference for healthy foods [65,66,67]. The home physical environment also became more supportive of health behaviors. That is, as family social capital rose so did space and supports for physical activity as well as availability of fruits/vegetables in the home. Although no comparable studies of home environments could be located, these findings are congruent with the overall premise of family social capital [6,12] in than parents in cohesive, engaged families would be inclined to create home environments supportive of optimal child development.
This study is, to the best of the authors’ knowledge, one of the first to investigate relationships of family social capital with weight-related behaviors and home environments of mothers and young children. Placing the study findings in the context of previous research is difficult for several reasons. First, despite its long history in other fields such as economics, sociology, and political science, the study of social capital in the health field began only about 20 years ago and thus there are few published studies [68]. Second, social capital is measured in a variety of ways which make direct comparisons of study results difficult. A “glaring gap in the conceptualization of social capital within the empirical literature has been the level of the family” [6]. Few studies have been conducted with children [42] or in the United States [5,6]. Additionally, many social capital studies, like the one reported here, are limited by their secondary analysis nature. That is, many published studies were not originally designed to assess social capital and, thus, are inherently constrained by available variables that are conceptually linked with social capital [36,69]. Finally, it is not possible to determine causation and/or direction of relationships due to this study being cross sectional.
Despite the study limitations, findings support the hypothesis and social capital theory [8] that family social capital is linked with positive weight-related behaviors and home environments of mothers and young children. Moreover, it suggests that strategies that teach parents how to build family social capital through amplified family verbal and physical engagement, more effective family conflict management, greater family cohesion, and more frequent family interactions such as those at mealtime could improve the effectiveness of health, nutrition, and obesity prevention interventions. Indeed, findings from the one nutrition intervention study located, which was conducted with Mexican American adults with type 2 diabetes mellitus, demonstrate it is possible to improve family social capital and, thereby, improve disease self-management [70]. Future intervention studies should consider incorporating strategies to build family social capital and compare longitudinal outcomes to traditional interventions to determine the relative value of family social capital on health behaviors [71].

Author Contributions

Conceptualization, all authors; methodology, V.Q. and C.B.-B.; formal analysis, C.D. and V.Q.; data curation, C.D. and V.Q.; writing—original draft preparation, C.D., K.E., and C.B.-B.; writing—review and editing, All authors; funding acquisition, C.B.-B. All authors have read and agreed to the published version of the manuscript.

Funding

Funding for this study was made possible by the United States Department of Agriculture, National Institute of Food and Agriculture, Grant Numbers 2011-68001-30170 and 2017-680001-26351.

Institutional Review Board Statement

The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Rutgers University (protocol code 11-294c and 12/26/2010).

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data sharing not applicable. No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Davis, M.M.; Gance-Cleveland, B.; Hassink, S.; Johnson, R.; Paradis, G.; Resnicow, K. Recommendations for prevention of childhood obesity. Pediatrics 2007, 120, S229–S253. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  2. Rosenkranz, R.; Dzewaltowski, D. Model of the home food environment pertaining to childhood obesity. Nutr. Rev. 2008, 66, 123–140. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Gruber, K.; Haldeman, L. Using the family to combat childhood and adult obesity. Prev. Chronic. Dis. 2009, 6, A106. [Google Scholar] [PubMed]
  4. Institute of Medicine. Progress in Preventing Childhood Obesity: How Do We Measure up? National Academy Press: Washington, DC, USA, 2007. [Google Scholar]
  5. McPherson, D.K.; Kerr, S.; McGee, E.; Cheater, F.; Morgan, A. The Role and Impact of Social Capital on the Health and Wellbeing of Children and Adolescents: A Systematic Review; Glasgow Centre for Population Health (GCPH): Glasgow, UK, 2013. [Google Scholar]
  6. Alvarez, E.C.; Kawachi, I.; Romani, J.R. Family social capital and health—A systematic review and redirection. Sociol. Health Illn. 2017, 39, 5–29. [Google Scholar] [CrossRef] [Green Version]
  7. Coleman, J.S. Social capital in the creation of human capital. Am. J. Sociol. 1988, 94, S95–S120. [Google Scholar] [CrossRef]
  8. Parcel, T.L.; Menaghan, E.G. Family social capital and children’s behavior problems. Social Psychol. Q. 1993, 56, 120–135. [Google Scholar] [CrossRef]
  9. Parcel, T.L.; Menaghan, E.G. Maternal working conditions and children’s verbal facility: Studying the intergenerational transmission of inequality from mothers to young children. Social Psychol. Q. 1990, 53, 132–147. [Google Scholar] [CrossRef]
  10. Rogers, S.J.; Parcel, T.L.; Menaghan, E.G. The Effects of Maternal Working Conditions and Mastery on Child Behavior Problems: Studying the Intergenerational Transmission of Social Control. J. Heal. Soc. Behav. 1991, 32, 145. [Google Scholar] [CrossRef]
  11. Arregle, J.-L.; Hitt, M.A.; Sirmon, D.G.; Very, P. The Development of Organizational Social Capital: Attributes of Family Firms. J. Manag. Stud. 2007, 44, 73–95. [Google Scholar] [CrossRef]
  12. Coleman, J. Foundations of Social Theory; Harvard University Press: Cambridge, MA, USA, 1990. [Google Scholar]
  13. Wright, J.P.; Cullen, F.T.; Miller, J.T. Family social capital and delinquent involvement. J. Crim. Justice 2001, 29, 1–9. [Google Scholar] [CrossRef]
  14. Parcel, T.L.; Dufur, M.J.; Zito, R.C. Capital at Home and at School: A Review and Synthesis. J. Marriage Fam. 2010, 72, 828–846. [Google Scholar] [CrossRef]
  15. Looman, W.S. Development and testing of the social capital scale for families of children with special health care needs. Res. Nurs. Health. 2006, 29, 325–336. [Google Scholar] [CrossRef] [PubMed]
  16. Ochoa, A.; Berge, J.M. Home Environmental Influences on Childhood Obesity in the Latino Population: A Decade Review of Literature. J. Immigr. Minor. Health. 2016, 19, 430–447. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  17. Golan, M.; Crow, S. Parents are key players in the prevention and treatment of weight-related problems. Nutr. Rev. 2004, 62, 39–50. [Google Scholar] [CrossRef] [PubMed]
  18. Lagström, H.; Hakanen, M.; Niinikoski, H.; Viikari, J.; Rönnemaa, T.; Saarinen, M.; Pahkala, K.; Simell, O. Growth Patterns and Obesity Development in Overweight or Normal-Weight 13-Year-Old Adolescents: The STRIP Study. Pediatrics 2008, 122, e876–e883. [Google Scholar] [CrossRef]
  19. Krahnstoever, D.; Francis, L.; Birch, L. Reexamining obesigenic families: Parents’ obesity-related behaviors predict girls’ change in BMI. Obes. Res. 2005, 13, 1980–1990. [Google Scholar]
  20. Mazarello Paes, V.; Ong, K.K.; Lakshman, R. Factors influencing obesogenic dietary intake in young children (0–6 years): Systematic review of qualitative evidence. BMJ Open 2015, 5, e007396. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  21. Gerards, S.M.; Kremers, S.P. The role of food parenting skills and the home food environment in children’s weight gain and obesity. Curr. Obes. Rep. 2015, 4, 30–36. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  22. Lindsay, A.C.; Greaney, M.L.; Wallington, S.F.; Mesa, T.; Salas, C.F. A review of early influences on physical activity and sedentary behaviors of preschool-age children in high-income countries. J. Spec. Pediatr. Nurs. 2017, 22, e12182. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  23. Byrd-Bredbenner, C.; Martin-Biggers, J.; Povis, G.A.; Worobey, J.; Hongu, N.; Quick, V. Promoting healthy home environments and lifestyles in families with preschool children: HomeStyles, a randomized controlled trial. Contemp. Clin. Trials. 2018, 64, 139–151. [Google Scholar] [CrossRef] [PubMed]
  24. Quick, V.; Martin-Biggers, J.; Povis, G.A.; Worobey, J.; Hongu, N.; Byrd-Bredbenner, C. A Socio-Ecological Examination of Weight-Related Characteristics of the Home Environment and Lifestyles of Households with Young Children. Nutrients 2017, 9, 604. [Google Scholar] [CrossRef] [PubMed]
  25. Byrd-Bredbenner, C.; Martin-Biggers, J.; Koenings, M.; Quick, V.; Hongu, N.; Worobey, J.; Laws, R. HomeStyles, A Web-Based Childhood Obesity Prevention Program for Families with Preschool Children: Protocol for a Randomized Controlled Trial. JMIR Res. Protoc. 2017, 6, e73. [Google Scholar] [CrossRef] [Green Version]
  26. Centers for Disease Control and Prevention. HRQOL Concepts. Why is Quality of Life Important? 2011. Available online: http://www.cdc.gov/hrqol/concept.htm (accessed on 9 May 2016).
  27. Centers for Disease Control and Prevention. CDC HRQOL-14 Healthy Days Measure. 2011. Available online: http://www.cdc.gov/hrqol/hrqol14_measure.htm (accessed on 9 May 2016).
  28. Kroenke, K.; Spitzer, R.L.; Williams, J.B.W. The Patient Health Questionnaire-2: Validity of a two-item depression screener. Med. Care. 2003, 41, 1284–1292. [Google Scholar] [CrossRef] [PubMed]
  29. Cohen, S.; Kamarck, T.; Mermelstein, R. A global measure of perceived stress. J. Health Soc. Behav. 1983, 24, 385–396. [Google Scholar] [CrossRef] [PubMed]
  30. Block, G.; Gillespie, C.; Rosenbaum, E.H.; Jenson, C. A rapid food screener to assess fat and fruit and vegetable intake. Am. J. Prev. Med. 2000, 18, 284–288. [Google Scholar] [CrossRef]
  31. Quick, V.; Byrd-Bredbenner, C.; Shoff, S.; White, A.A.; Lohse, B.; Horacek, T.; Kattelmann, K.; Phillips, B.; Hoerr, S.; Greene, G. A streamlined, enhanced self-report physical activity measure for young adults. Int. J. Heal. Promot. Educ. 2016, 54, 245–254. [Google Scholar] [CrossRef]
  32. Buysse, D.J.; Reynolds, C.F.; Monk, T.H.; Berman, S.R.; Kupfer, D.J. The Pittsburgh sleep quality index: A new instrument for psychiatric practice and research. Psychiatry Res. 1989, 28, 193–213. [Google Scholar] [CrossRef]
  33. Carpenter, J.S.; Andrykowski, M.A. Psychometric evaluation of the pittsburgh sleep quality index. J. Psychosom. Res. 1998, 45, 5–13. [Google Scholar] [CrossRef]
  34. Cheng, C.; Martin-Biggers, J.; Quick, V.; Spaccarotella, K.; Byrd-Bredbenner, C. Validity and reliability of HOP-Up: A questionnaire to evaluate physical activity environments in homes with preschool-aged children. Int. J. Behav. Nutr. Phys. Act. 2016, 13, 91. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  35. Martin-Biggers, J.; Koenings, M.; Quick, V.; Abbot, J.M.; Byrdbredbenner, C. Appraising nutrient availability of household food supplies using Block Dietary Screeners for individuals. Eur. J. Clin. Nutr. 2015, 69, 1028–1034. [Google Scholar] [CrossRef] [PubMed]
  36. Reynolds, J.C.; Damiano, P.C.; Glanville, J.L.; Oleson, J.; McQuistan, M.R. Neighborhood and family social capital and parent-reported oral health of children in Iowa. Community Dent. Oral. Epidemiol 2015, 43, 569–577. [Google Scholar] [CrossRef]
  37. Zolotor, A.J.; Runyan, D.K. Social capital, family violence, and neglect. Pediatrics 2006, 117, e1124–e1131. [Google Scholar] [CrossRef] [PubMed]
  38. Vandewater, E.A.; Lansford, J.E. A family process model of problem behaviors in adolescents. J. Marriage Fam. 2005, 67, 100–109. [Google Scholar] [CrossRef]
  39. De Clercq, B.; Pfoertner, T.-K.; Elgar, F.J.; Hublet, A.; Maes, L. Social capital and adolescent smoking in schools and communities: A cross-classified multilevel analysis. Soc. Sci. Med. 2014, 119, 81–87. [Google Scholar] [CrossRef] [PubMed]
  40. Runyan, D.K.; Hunter, W.M.; Socolar, R.R.S.; Amaya-Jackson, L.; English, D.; Landsverk, J.; Dubowitz, H.; Browne, D.H.; Bangdiwala, S.I.; Mathew, R.M. Children who prosper in unfavorable environments: The relationship to social capital. Pediatrics 1998, 101, 12–18. [Google Scholar] [CrossRef]
  41. Fram, M.S. Managing to Parent: Social Support, Social Capital, and Parenting among Welfare-Participating Mothers with Young Children in School of Social Work-Institute for Research on Poverty; Discussion Paper #1263-03; University of Washington: Seattle, WA, USA, 2003. [Google Scholar]
  42. Harpham, T. Measuring the Social Capital of Children, Young Lives Working Paper; South Bank University London: London, UK, 2002. [Google Scholar]
  43. Bala-Brusilow, C. A Study of the Associations between Childhood Obesity and Three Forms of Social Capital. Ph.D. Thesis, Wayne State University, Detroit, MI, USA, 2010. [Google Scholar]
  44. Earls, F.; Brooks-Gunn, J.; Raudenbush, S.W.; Sampson, R.J. Project on Human Development in Chicago Neighborhoods (PHDCN): Home and Life Interview, Wave 2, 1997–2000. Instruments for ICPSR 13630. Inter-University Consortium for Political and Social Research: Ann Arbor, MI, USA, 2005. [Google Scholar]
  45. Moos, R.; Moos, B. Family Environment Scale Manual: Development, Applications, Research; Mind Garden, Inc.: Palo Alto, CA, USA, 2002. [Google Scholar]
  46. Charalampous, K.; Kokkinos, C.; Panyiotou, G. The family environment scale: Resolving psychometric problems through an examination of a Greek translation. Int. J. Educ. Psychol. Assess 2013, 13, 81–99. [Google Scholar]
  47. Greene, R.C.; Plank, R.E. The Short-Form Family Environment Scale: Testing a Different Response Format. Psychol. Rep. 1994, 74, 451–464. [Google Scholar] [CrossRef]
  48. Rhee, K. Childhood Overweight and the Relationship between Parent Behaviors, Parenting Style, and Family Functioning. Ann. Am. Acad. Politi- Soc. Sci. 2008, 615, 11–37. [Google Scholar] [CrossRef]
  49. Benjamini, Y.; Hochberg, Y. Controlling the false discover rate: A practical and powerful approach to multiple testing. J. Royal. Statis. Soc. Series B. (Methodol) 1995, 57, 289–300. [Google Scholar]
  50. Watson, P. Rules of Thumb on Magnitudes of Effect Sizes. 2019. Available online: http://www.imaging.mrc-cbu.cam.ac.uk/statswiki/FAQ/effectSize (accessed on 12 December 2020).
  51. Dubrov, D. Family social capital as a predictor of parental and adolescent subjective well-being in Russia. SSRN Electron. J. 2017. [Google Scholar] [CrossRef] [Green Version]
  52. Novak, D.; Suzki, E.; Kawachi, I. Are family, neighborhood and school social capital associated with higher self-rated health among Croation high school students? A population-based study. BMJ Open 2015, 5, e007184. [Google Scholar] [CrossRef] [Green Version]
  53. Wang, R.; Chen, S.W.; Tang, S.M.; Lee, S.L.; Jian, S.Y. The relationship between selected developmental assets and health promoting behaviours of adolescents in Southern Taiwan. J. Clin. Nurs. 2011, 20, 359–368. [Google Scholar] [CrossRef]
  54. Dufur, M.; Parcel, T.; McKune, B. Capital and context: Using social capital at home and at school to predict child social adjustment. J. Health Social. Behav. 2008, 49, 146–161. [Google Scholar] [CrossRef] [PubMed]
  55. Maynard, M.; Harding, S. Perceived parenting and psychological well-being in UK ethnic minority adolescents. Child Care Health Dev. 2010, 36, 630–638. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  56. Maynard, M.; Harding, S. Ethnic differences in psychological well-being in adolescence in the context of time spent in family activities. Soc. Psychiatry Psychiatr. Epidemiol. 2010, 45, 115–123. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  57. Reynolds, A.D.; Crea, T.M. Household stress and adolescent behaviours in urban families: The mediating roles of parent mental health and social supports. Child Fam. Soc. Work. 2014, 21, 568–580. [Google Scholar] [CrossRef]
  58. Fulkerson, J.; Story, M.; Mellin, A.; Leffert, N.; Neumark-Sztainer, D.; French, S.A. Family dinner meal frequency and adolescent development: Relationships with developmental assets and high-risk behaviors. J. Adol. Health 2006, 39, 337–345. [Google Scholar] [CrossRef] [PubMed]
  59. Springer, A.; Parcel, G.; Baumler, E.; Ross, M. Supportive social relationships and adolescent health risk behavior among secondary school students in El Salvador. Soc. Sci. Med. 2006, 62, 1628–1640. [Google Scholar] [CrossRef]
  60. Morgan, A.; Haglund, B.J.A. Social capital does matter for adolescent health: Evidence from the English HBSC study. Health Promot. Int. 2009, 24, 363–372. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  61. Moxley, R.L.; Jicha, K.A.; Thompson, G.H. Testing the Importance of Family Solidarity, Community Structure, Information Access, and Social Capital in Predicting Nutrition Health Knowledge and Food Choices in the Philippines. Ecol. Food Nutr. 2011, 50, 215–239. [Google Scholar] [CrossRef] [PubMed]
  62. Novak, D.; Doubova, S.; Kawachi, I. Social capital and physical activity among Croatian high school students. Public Health 2016, 135, 48–55. [Google Scholar] [CrossRef] [PubMed]
  63. Hirshkowitz, M.; Whiton, K.; Albert, S.M.; Alessi, C.; Bruni, O.; DonCarlos, L.; Hazen, N.; Herman, J.; Katz, E.S.; Kheirandish-Gozal, L.; et al. National Sleep Foundation’s sleep time duration recommendations: Methodology and results summary. Sleep Health 2015, 1, 40–43. [Google Scholar] [CrossRef]
  64. Nagy, E.; Moore, S.; Gruber, R.; Paquet, C.; Arora, N.; Dube, L. Parental social capital and children’s sleep disturbances. Sleep Health 2016, 2, 330–334. [Google Scholar] [CrossRef] [PubMed]
  65. Yee, A.Z.; Lwin, M.O.; Ho, S.S. The influence of parental practices on child promotive and preventive food consumption behaviors: A systematic review and meta-analysis. Int. J. Behav. Nutr. Phys. Act. 2017, 14, 47. [Google Scholar] [CrossRef]
  66. Blaine, R.E.; Kachurak, A.; Davison, K.K.; Klabunde, R.; Fisher, J.O. Food parenting and child snacking: A systematic review. Int. J. Behav. Nutr. Phys. Act. 2017, 14, 146. [Google Scholar] [CrossRef] [Green Version]
  67. Vollmer, R.L.; Baietto, J. Practices and preferences: Exploring the relationships between food-related parenting practices and child food preferences for high fat and/or sugar foods, fruits, and vegetables. Appetite 2017, 113, 134–140. [Google Scholar] [CrossRef] [PubMed]
  68. Reynolds, J.C. Neighborhood and Family Social Capital and Oral Health Status of Children in Iowa. Master’s Thesis, University of Iowa, Iowa, IA, USA, 2013. [Google Scholar]
  69. Carrillo-Alvarez, E.C.; Villalonga-Olives, E.; Rier-Romani, J.; Kawachi, I. Development and Validation of a Questionnaire to Measure Family Social Capital. SSM-Pop. Health 2019, 8, 100453. [Google Scholar] [CrossRef]
  70. McEwen, M.M.; Pasvogel, A.; Murdaugh, C. Effects of a Family-Based Diabetes Intervention on Family Social Capital Outcomes for Mexican American Adults. Diabetes Educ. 2019, 45, 272–286. [Google Scholar] [CrossRef] [PubMed]
  71. Pinkerton, J.; Dolan, P. Family support, social capital, resilience and adolescent coping. Child Fam. Soc. Work 2007, 12, 219–228. [Google Scholar] [CrossRef]
Table 1. Family social capital scale comparison by tertile (n = 557).
Table 1. Family social capital scale comparison by tertile (n = 557).
Family Social
Capital Scales (Cronbach Alpha) #
Tertile 1 (n = 187)
Mean ± SD
(95% CI *)
Tertile 2 (n = 186) Mean ± SD
(95% CI)
Tertile 3 (n = 184) Mean ± SD
(95% CI)
F
df = 2554 ‡
ANOVA
p
Tukey
Post-Hoc Pairwise
Comparisons
Partial Eta-Squared
Verbal engagement A [44] (0.69)1.71 ± 1.00
(1.57, 1.86)
2.19 ± 0.83
(2.07, 2.31)
2.73 ± 0.90
(2.60, 2.86)
57.181<0.001ABC0.171
Physical engagement A [44] (n/a)3.07 ± 1.03
(2.93, 3.22)
3.62 ± 0.57
(3.54, 3.70)
3.85 ± 0.42
(3.79, 3.91)
56.233<0.001ABC0.169
Family conflict A [45,46,47] (0.86)2.27 ± 0.98
(2.13, 2.41)
2.95 ± 0.85
(2.83, 3.07)
3.62 ± 0.66
(3.53, 3.72)
120.262<0.001ABC0.303
Family cohesion A [45,46,47] (0.87)2.54 ± 0.78
(2.43, 2.65)
3.12 ± 0.52
(3.05, 3.20)
3.63 ± 0.45
(3.57, 3.70)
153.128<0.001ABC0.356
Family meal frequency B (n/a)1.80 ± 1.02
(1.65, 1.95)
2.52 ± 0.97
(2.38, 2.66)
3.12 ± 0.83
(3.00, 3.24)
90.294<0.001ABC0.246
# Cronbach alpha. * CI = Confidence Interval ‡ df = Degrees of Freedom Pairwise comparisons A = Tertile 1 and 2 differed significantly; B = Tertile 1 and 3 differed significantly; C = Tertile 2 and 3 differed significantly (p < 0.05). A 5-point agreement rating: Strongly disagree, disagree, neither agree nor disagree, agree, strongly agree; scored 0 to 4, respectively, with scoring reversed for negatively worded statements; scale score equals average of item scores; higher scale score indicates greater expression of the characteristic. B Days/week of having family meals at breakfast, lunch, and dinner. Total possible range = 0 to 4; scored as 0 if <5 meals/week, 1 if ≥5 to <9 meals/week, 2 if ≥9 to <13, 3 if ≥13 to <17 meals/week, 4 if ≥17 meals/week.
Table 2. Maternal sociodemographic characteristic comparisons by family social capital tertiles (n = 557).
Table 2. Maternal sociodemographic characteristic comparisons by family social capital tertiles (n = 557).
Family Social Capital ScalesTertile 1 (n = 187)
Mean ± SD
(95% CI *)
Tertile 2 (n = 186) Mean ± SD
(95% CI)
Tertile 3 (n = 184) Mean ± SD
(95% CI)
F
df = 2554
ANOVA
p
Tukey
Post-Hoc Pairwise
Comparisons
Partial Eta-Squared
Age32.51 ± 5.78
(31.68, 33.34)
32.82 ± 5.42
(32.03, 33.60)
32.61 ± 5.43
(31.82, 33.40)
0.1470.864 0.001
Race/ethnicity (white vs. non-white) A0.42 ± 0.50
(0.35, 0.49)
0.38 ± 0.49
(0.31, 0.45)
0.39 ± 0.49
(0.31, 0.46)
0.4610.631 0.002
Highest level of education B2.30 ± 0.71
(2.20, 2.41)
2.41 ± 0.69
(2.31, 2.51)
2.33 ± 0.73
(2.22, 2.44)
1.0650.345 0.004
Number of children Under 18 in the household2.35 ± 1.16
(2.18, 2.51)
2.23 ± 1.11
(2.07, 2.39)
2.16 ± 1.06
(2.00, 2.31)
1.3990.248 0.005
Marital status (single or dual parent household) C1.78 ± 0.41
(1.72, 1.84)
1.86 ± 0.35
(1.81, 1.91)
1.81 ± 0.39
(1.75, 1.87)
2.0190.134 0.007
Maternal employment D1.17 ± 0.92
(1.03, 1.30)
1.21 ± 0.86
(1.09, 1.33)
0.97 ± 0.93
(0.83, 1.10)
3.7700.024C0.013
Family affluence scale E5.17 ± 1.76
(4.92, 5.43)
5.49 ± 1.71
(5.24, 5.74)
5.71 ± 1.72
(5.46, 5.96)
4.5010.012B0.016
* CI = Confidence Interval ‡ df = Degrees of Freedom Pairwise comparisons A = Tertile 1 and 2 differed significantly; B = Tertile 1 and 3 differed significantly; C = Tertile 2 and 3 differed significantly (p < 0.05). A Coded 0 = white; 1 = non-white; B Coded 1 = high school or less; 2 = some post-secondary education; 3 = college degree or higher; C Coded 1 = single parent; 2 = dual parent; D Coded 0 = no paid employment; 1 = part-time paid employment; 2 = full-time paid employment; E scale scored 0 to 9 points; greater points indicate greater family affluence.
Table 3. Maternal health, weight-related behaviors, and home environment comparisons by family social capital tertiles (n = 557).
Table 3. Maternal health, weight-related behaviors, and home environment comparisons by family social capital tertiles (n = 557).
MeasuresTertile 1 (n = 187)
Mean ± SD
(95% CI *)
Tertile 2 (n = 186) Mean ± SD
(95% CI)
Tertile 3 (n = 184) Mean ± SD
(95% CI)
F
df = 2554
ANOVA
p
Tukey
Post-Hoc Pairwise
Comparisons
Partial Eta-Squared
Health
Health status A [26,27]3.28 ± 0.95
(3.15, 3.42)
3.37 ± 0.86
(3.25, 3.42)
3.72 ± 1.02
(3.58, 3.87)
11.216<0.001BC0.039
Physical health quality of life B [26,27]3.87 ± 6.79
(2.89, 4.85)
3.20 ± 5.07
(2.47, 3.93)
2.29 ± 5.23
(1.53, 3.05)
3.5380.030B0.013
Mental health quality of life B [26,27]6.70 ± 8.64
(5.45, 7.95)
4.83 ± 7.13
(3.80, 5.86)
3.46 ± 6.34
(2.54, 4.38)
8.871<0.001AB0.031
Depression severity C [28]1.74 ± 0.77
(1.63, 1.85)
1.57 ± 0.71
(1.47, 1.67)
1.52 ± 0.73
(1.41, 1.62)
4.5410.011B0.016
Perceived stress C [29]1.76 ± 0.79
(1.64, 1.87)
1.59 ± 0.78
(1.48, 1.70)
1.43 ± 0.67
(1.33, 1.53)
8.970<0.001AB0.042
Dietary Intake
Fruit/vegetable (serv/day) [30]4.01 ± 1.93
(3.73, 4.29)
4.47 ± 1.68
(4.22, 4.71)
5.05 ± 2.00
(4.60, 5.17)
10.377<0.001AB0.036
Sugar-sweetened beverages (serv/day) [25]0.85 ± 0.91
(0.72, 0.98)
0.74 ± 0.81
(0.62, 0.86)
0.61 ± 0.77
(0.49, 0.72)
4.0280.018B0.014
Physical Activity [31]
Physical activity level D [25,31]12.15 ± 9.21
(10.82, 13.48)
13.98 ± 10.07
(12.52, 15.43)
16.59 ± 9.66
(15.18, 17.99)
9.891<0.001BC0.034
Screen time (min/day)383.82 ± 289.32
(342.08, 425.56)
351.45 ± 287.70
(309.83, 393.07)
315.57 ± 248.85
(279.38, 351.77)
2.8370.059 0.010
Sleep [32,33]
Sleep quality E3.04 ± 0.88
(2.92, 3.17)
3.19 ± 0.95
(3.05, 3.33)
3.32 ± 0.87
(3.19, 3.45)
4.3910.013B0.016
Sleep duration (hours/day)6.93 ± 1.27
(6.75, 7.12)
6.99 ± 1.28
(6.81, 7.18)
7.28 ± 1.16
(7.11, 7.18)
4.0270.018B0.015
Parenting Behaviors [23,24,25]
Modeling of healthy eating F3.33 ± 0.82
(3.22, 3.45)
3.62 ± 0.71
(3.52, 3.73)
3.93 ± 0.76
(3.82, 4.04)
28.232<0.001ABC0.091
Modeling of physical activity (days/week)2.73 ± 1.22
(2.56, 2.91)
3.26 ± 1.19
(3.09, 3.43)
3.59 ± 1.29
(3.40, 3.78)
22.435<0.001ABC0.075
Child Feeding Practices F
Parent feeding: restriction3.77 ± 0.82
(3.65, 3.89)
3.77 ± 0.82
(3.65, 3.89)
3.86 ± 1.00
(3.72, 4.01)
0.7240.485 0.003
Parent feeding: pressure2.58 ± 0.95
(2.44, 2.71)
2.27 ± 0.90
(2.14, 2.40)
2.01 ± 0.96
(1.87, 2.15)
17.232<0.001ABC0.059
Parent feeding: instrumental (food reward)2.50 ± 0.72
(2.39, 2.60)
2.38 ± 0.72
(2.27, 2.48)
2.28 ± 0.80
(2.16, 2.40)
3.9030.021B0.014
Home Physical Activity Environment [34]
Indoor home physical activity space and supports F,G3.14 ± 0.84
(3.02, 3.26)
3.35 ± 0.84
(3.22, 3.47)
3.56 ± 0.77
(3.45, 3.67)
12.457<0.001ABC0.043
Outdoor/yard physical activity space and supports G4.22 ± 0.72
(4.11, 4.34)
4.39 ± 0.64
(4.30, 4.49)
4.56 ± 0.56
(4.47, 4.64)
11.276<0.001ABC0.043
Home Food Environment [35]
Fruit/vegetables (serv available/household member/week)5.41 ± 1.99
(5.12, 5.69)
5.97 ± 1.86
(5.70, 6.24)
6.58 ± 2.14
(6.27, 6.89)
16.051<0.001ABC0.055
Sugar-sweetened beverages (serv available /household member/week)1.79 ± 1.84
(1.52, 2.05)
1.74 ± 1.82
(1.48, 2.01)
1.49 ± 1.82
(1.22, 1.75)
1.4650.232 0.005
* CI = Confidence Interval ‡ df = Degrees of Freedom Pairwise comparisons A = Tertile 1 and 2 differed significantly; B = Tertile 1 and 3 differed significantly; C = Tertile 2 and 3 differed significantly. A 5-point excellence rating: poor, fair, good, very good, excellent; scored 1 to 5, respectively. B Days/month of “not good” health C 4-point frequency rating: not at all, several days, more than half the days, nearly every day; scored 1 to 4. Items averaged to create scale score. Cronbach alpha for Depression Severity and Perceived Stress are 0.74 and 0.78, respectively. D Days/week engaging in walking, moderate, or vigorous activity, for at least 10 min; days weighted by intensity of 1, 2, and 3, respectively, and summed; score range 0 to 42. E 5-point rating scale: very bad, bad, okay, good, very good; scored 1 to 5, respectively. F 5-point agreement rating: strongly disagree, disagree, neither agree nor disagree, agree, strongly agree; scored 1 to 5, respectively; scale score equals average of item scores. Cronbach alpha for modeling of healthy eating, parent feeding: pressure, parent feeding: instrumental, indoor home physical activity space and supports, and outdoor/yard physical activity space and supports are 0.74, 0.69, 0.74, 0.71, and 0.74, respectively. Cronbach alpha for other scales sharing this superscript could not be calculated due to scale length (<2 items) and/or scale type. G 5-point occurrence rating: almost never, 1–2 times/week; 3 to 4 times/week, 5 to 6 times/week, every day; scored 1 to 5, respectively; scale score equals average of item scores.
Table 4. Child health and weight-related behaviors comparisons by family social capital tertiles (n = 557).
Table 4. Child health and weight-related behaviors comparisons by family social capital tertiles (n = 557).
MeasuresTertile 1 (n = 187)
Mean ± SD
(95% CI *)
Tertile 2 (n = 186) Mean ± SD
(95% CI)
Tertile 3 (n = 184) Mean ± SD
(95% CI)
F
df = 2554
ANOVA
p
Tukey
Post-Hoc Pairwise
Comparisons
Partial Eta-Squared
Child Health Status A [26,27]4.16 ± 0.77
(4.04, 4.27)
4.37 ± 0.86
(4.25, 4.49)
4.64 ± 0.64
(4.55, 4.73)
19.034<0.001ABC0.055
Physical health quality of life B1.33 ± 2.57
(0.96, 1.70)
1.96 ± 4.77
(1.27, 2.65)
1.17 ± 2.85
(1.19, 1.78)
2.5590.078 0.009
Mental health quality of life B2.66 ± 4.47
(2.01, 3.30)
2.83 ± 5.43
(2.04, 3.61)
1.68 ± 1.97
(1.40, 1.97)
3.9410.020C0.014
Child Dietary Intake
Fruit/vegetable juice (serv/day) [30]0.68 ± 0.50
(0.61, 0.75)
0.65 ± 0.50
(0.57, 0.72)
0.75 ± 0.56
(0.67, 0.83)
1.7770.170 0.007
Sugar-sweetened beverages (serv/day) [25]0.38 ± 0.50
(0.31, 0.45)
0.33 ± 0.46
(0.26, 0.39)
0.27 ± 0.41
(0.21, 0.33)
2.7760.063 0.006
Physical Activity [31]
Physical activity level C23.82 ± 12.35
(22.04, 25.60)
26.56 ± 10.91
(24.99, 28.14)
28.2 ± 10.82
(26.63, 29.77)
7.0270.001B0.021
Screen time (min/day)317.57 ± 293.38
(275.24, 359.89)
292.82 ± 260.59
(255.13, 330.52)
274.89 ± 261.55
(236.85, 312.93)
1.1490.318 0.003
Child Sleep D [32,33]
Sleep quality4.09 ± 0.78
(3.97, 4.20)
4.18 ± 0.80
(4.07, 4.30)
4.48 ± 0.652
(4.38, 4.57)
7.740<0.001BC0.033
Sleep duration (hours/day)10.31 ± 1.34
(10.10, 10.52)
10.60 ± 1.50
(10.37, 10.83)
10.60 ± 1.38
(10.39, 10.81)
2.3020.101 0.009
* CI = Confidence Interval ‡ df = Degrees of Freedom Pairwise comparisons A = Tertile 1 and 2 differed significantly; B = Tertile 1 and 3 differed significantly; C = Tertile 2 and 3 differed significantly A 5-point excellence rating: poor, fair, good, very good, excellent; scored 1 to 5, respectively. B Days/month of “not good” health. C Days/week engaging in walking, moderate, or vigorous activity, for at least 10 min; days weighted by intensity of 1, 2, and 3, respectively, and summed; score range 0 to 42. D 5-point rating scale: very bad, bad, okay, good, very good; scored 1 to 5, respectively.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Quick, V.; Delaney, C.; Eck, K.; Byrd-Bredbenner, C. Family Social Capital: Links to Weight-Related and Parenting Behaviors of Mothers with Young Children. Nutrients 2021, 13, 1428. https://doi.org/10.3390/nu13051428

AMA Style

Quick V, Delaney C, Eck K, Byrd-Bredbenner C. Family Social Capital: Links to Weight-Related and Parenting Behaviors of Mothers with Young Children. Nutrients. 2021; 13(5):1428. https://doi.org/10.3390/nu13051428

Chicago/Turabian Style

Quick, Virginia, Colleen Delaney, Kaitlyn Eck, and Carol Byrd-Bredbenner. 2021. "Family Social Capital: Links to Weight-Related and Parenting Behaviors of Mothers with Young Children" Nutrients 13, no. 5: 1428. https://doi.org/10.3390/nu13051428

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop