Abstract
TV use could be detrimental to children’s development because it may displace enriching activities like shared reading and caregiver-child interaction or disrupt children’s interactions with the adults in their lives by distracting both parties. Some prior research has shown that demographic factors (maternal education, siblings) and maternal mental health (depression risk and parenting stress) may predict household TV, but findings are mixed. Household TV is defined here as whether or how much the TV is on in the home. In the current study, we examine the following research questions: (a) what is the prevalence of household TV (i.e., whether or how much the TV is on) in the homes of infants and toddlers in a sample of families from low-income homes? and (b) do demographic characteristics (i.e., maternal education level and presence of siblings in the home), and maternal mental health (i.e., depression risk and parenting stress) predict household TV (i.e., whether or how much the TV is on)? Mothers (N = 220) reported on their household TV, education level, siblings in the home, depression risk, and parenting stress when children were four to seven months of age and again at 15 to 19 months of age. Results showed a high level of household TV, especially during infancy. Furthermore, education level, but not siblings, depression risk, or parenting stress, was related to household TV during infancy and toddlerhood. These findings are discussed in relation to prior research and the potential for informing future interventions or education efforts.
Keywords: Household television, media, parenting stress, depression
1.0. Introduction
Despite the rise of mobile devices (e.g., Operto et al., 2020), TV use is still quite common in children’s homes, with a recent nationally representative report showing that TV sets are present in 97% of homes of children under eight years of age and that 60% of children watch television or videos at least once a day compared to 30% who use a mobile device (Rideout & Robb, 2020). Furthermore, 39% of children under eight live in homes that parents characterize as having a TV on most of the time or always, even if no one is watching (Rideout & Robb, 2020).
Indeed, compared to older children, infants and toddlers might be particularly likely to live in homes where TVs are often on in the background of other household activities. For example, in a nationally representative telephone survey, Lapierre et al. (2012) collected 24-hour time diaries from parents of children between 8 months and eight years of age. Parents reported that younger children in their sample were exposed to more background TV than older children, with children under 24 months being exposed to an average of 5.5 hours of background TV per day, compared to an average of 3.5 hours for older children. These researchers speculate that young children may have high background TV exposure because spending time alone with children in this age range can be isolating, and parents may turn to TV for extra stimulation (see also Tomopoulous et al., 2014).
Notably, children’s TV exposure may have meaningful impacts on development in young children, including in the domain of language (Clarke & Kurtz-Costes, 1997; Duch et al., 2013; Pagani et al., 2013), by displacing or disrupting activities like shared reading and caregiver-child interaction (Khan et al., 2017; Kirkorian et al., 2009; Nathanson & Rasmussen, 2011; Vandewater et al., 2006). Furthermore, although by age 3 or 4, children are able to learn from high-quality, educational TV (e.g., Mares & Pan, 2013), evidence suggests that infants’ and toddlers’ ability to learn from TV is more limited (Anderson & Pempek, 2005; DeLoache et al., 2010).
1.1. TV Exposure Among Infants and Toddlers from Low-Income Backgrounds
Evidence suggests that TV exposure may be higher amongst infants and toddlers from low socioeconomic status (SES) backgrounds compared to children from middle- and upper-SES backgrounds. Rideout and Robb (2020) found that children under eight from lower-income homes spent significantly more time using screen media than children from middle- or upper-income homes. Furthermore, a study by Masur et al. (2015) found that 69% of all parents reported that the TV was on for half or more than half of their child’s waking hours, whereas the parallel percentage for low-income families was 94% (cf. Vandewater et al., 2005). Similarly, parents reported that the TV was on in the room for half or more than half of the time during dyadic toy play between children and caregivers for 68% of children overall, but 90% of low-income children (Masur et al., 2015; see also Barr et al., 2010).
1.2. Predictors of TV Exposure
The prevalence of TV exposure amongst infants and toddlers from low-SES and disadvantaged backgrounds is a vital topic to understand, given that these populations are already at risk for developmental delays and lower academic achievement later in life (Duncan & Brooks-Gunn, 2000). However, prior research examining demographic factors as predictors has resulted in mixed findings. Using a nationally representative sample of children under six years of age, Vandewater et al. (2005) found that children with no siblings had higher background TV exposure than children with siblings (Vandewater et al., 2005). Thompson and Tschann (2016) found that the total number of people living in the home was not predictive of TV among low-income Mexican-American 3- to 5-year-olds, although they did not report the association with the number of adults and the number of children separately in their analyses.
One set of predictors that have been consistently linked to young children’s TV exposure relates to maternal mental health. For example, two studies, one with a high-SES and one with a low-SES sample, report that infant TV viewing is related to maternal depressive symptoms (Bank et al., 2012; Anand et al., 2014). Similarly, McLearn et al. (2006) found that mothers with depressive symptoms in a mixed-SES sample were more likely to have toddlers who watched more than two hours of TV or videos per day. Parallel results have been found with low-SES 3- to 5-year-olds (Burdette, Whitaker, Kahn, & Harvey-Berino, 2003; Conners, Tripathi, Clubb, & Bradley, 2007). In addition to depression, some evidence from nationally representative samples suggests that TV exposure among infants and toddlers is also related to other maternal mental health measures, including anxiety or stress (e.g., Bank et al., 2012; Thompson & Christakis, 2007). For example, Walton et al. (2014) found that parents of 2- to 5-year-olds in a low-SES sample who were experiencing high stress were less likely to limit their children’s TV watching. These studies suggest that, in addition to maternal depression, parenting stress and related constructs like anxiety may contribute to children’s TV exposure. In the current study, we focus specifically on maternal stress.
Importantly, little research has attempted to disentangle how the related constructs of depression risk and maternal stress may uniquely predict children’s TV exposure. Disentangling these two variables is important because mental health variables are likely to be comorbid amongst postpartum mothers (e.g., Farr, Dietz, O’hara, Burley, & Ko, 2014; Leigh & Milgrom, 2008). Thus, researchers may draw inaccurate conclusions about TV exposure’s true antecedents by only measuring one construct. In the current study, we measure depression risk and parenting stress to tease apart their unique associations with children’s TV exposure. Furthermore, to our knowledge, existing studies focusing on maternal mental health have exclusively examined predictors of children’s TV viewing and not overall household exposure to TV, including both purposeful child viewing and TV that is left on in the background, which is essential given that both may impact child development. Here, we define household TV as whether or how much the TV is on in the home. Notably, using screen media in the background of other activities is related to overall child screen exposure (Elias & Sulkin, 2019; Rideout & Robb, 2020), highlighting the value of considering both types of use together.
Additionally, little prior research has examined the prevalence of household TV across development in a fine-grained manner or for very young infants (i.e., under nine months). This gap is important because household TV in this developmental stage may have different predictors compared to use among older infants or toddlers. Further, although prior research has examined predictors of household TV at different points across early childhood, to our knowledge, no studies have reported longitudinal data examining whether these predictors may differ for the same children at different stages of development. It is feasible that different factors could influence household TV across development. Indeed, mothers’ daily experiences with an infant are quite different from when that child is a toddler. For example, although infants lack expressive language abilities, by toddlerhood, children may be able to have more meaningful and in-depth back-and-forth verbal interactions with parents.
1.3. The Current Study
Overall, household TV (whether or how much the TV is on) seems to be quite prevalent in infants’ and toddlers’ homes, perhaps especially for those from low-SES backgrounds, despite potential adverse effects on children’s interactions and development. A growing body of research has explored potential antecedents to high levels of household TV in these homes, but findings have been mixed and have not examined developmental change across infants and toddlers. In the current study, we use a quantitative approach to examine the following research questions: (a) what is the prevalence of household TV (i.e., whether or how much the TV is on) in the homes of infants and toddlers in a sample of families from low-income homes? and (b) do demographic characteristics (i.e., maternal education level and presence of siblings in the home), and maternal mental health (i.e., depression risk and parenting stress) predict household TV (i.e., whether or how much the TV is on)? We expect to find that, replicating past research, infants and toddlers in low-income homes will experience a high level of household TV. Given the mixed findings and lack of research on household TV predictors, we consider this research question exploratory.
2.0. Method
2.1. Participants
The present study represents a subset of participants from a larger study on the timing and duration of resource use among mothers with young children living in poverty in Ohio (N = 326). Mothers who were pregnant or had an infant from birth to three months old were recruited for participation from Women, Infants, and Children (WIC) centers by trained research staff. Other inclusion criteria included that participants: (a) planned on being in the county for the entirety of the study (five years), (b) were at least 18 years old, (c) had conversational English skills in order to be able to respond to the standardized study measures and communicate with project staff, and (d) had a child that was not premature or diagnosed with any severe medical conditions. The data reported here were drawn from 220 mothers who completed the first parent meeting during infancy (see Procedures) and were collected between 2015 and 2017. At enrollment, mothers were an average of 26-years-old (SD = 5.22 years, Range = 18- to 43-years-old). The plurality of mothers had completed high school (38.9%); whereas 32.6% had completed some college, 14.0% had completed less than a high school diploma, 8.1% had completed a bachelor’s or master’s degree, and 3.2% had completed an associate’s degree (0.9% unreported).
The plurality of mothers reported an income of $10,000 or less (44.8%), whereas 16.3% reported an annual income of $10,001 to $20,000, 16.3% reported an annual income of $20,001 to $30,000, 9.0% reported an annual income of $30,001 to $40,000, and 6.9% reported an annual income of $40,001 or more (6.8% unreported). Additionally, 26.2% percent of mothers were married, 43.0% were single, 24.0% were single and living with a partner, and 6.3% were divorced, separated, or widowed (.5% unreported). Most mothers spoke only English at home (85.5%). A majority of children were female (56%), and a plurality were identified by mothers as African-American (45.5%). The remainder were primarily identified as Caucasian (38.6%) with 15.9% other, including multiple races, American Indian, Asian, and Pacific Islander. When their mothers enrolled in the study, children’s ages ranged from 0 to almost 4-months-old (59.8% of mothers were pregnant at enrollment).
2.2. Procedures
2.2.1. Enrollment.
Enrollment occurred when the mothers were pregnant or had a new infant (birth to three months old). Mothers completed a background questionnaire and a health survey and were provided with a set of diapers and wipes, a $5 gift card, and a children’s book to compensate them for their time.
2.2.2. Parent Meetings.
The first parent meeting was when the infant was four to seven months old. These meetings lasted about a half-hour and occurred in either the mother’s home or a neutral location (e.g., McDonald’s lobby, libraries). Mothers could attend these meetings with or without their child, and they were provided with a set of diapers and wipes, a $10 gift card, and a children’s book to compensate them for their time. The second parent meeting was when the infant was 15- to 19-months-old. The procedures for this meeting were the same as the first parent meeting. Incentives included two sets of diapers and wipes, a $10 gift card, and a children’s book.
2.3. Measures
2.3.1. Household TV.
Household TV was measured using one item from the short form of the Confusion, Hubbub, and Order Scale (Petrill, Pike, Price, & Plomin, 2004) collected during both parent meetings. The item reads: “There is usually a TV turned on somewhere in our home.” This item was scored on a five-point Likert scale from 0 (definitely untrue) to 4 (definitely true), with higher scores indicating more household TV. In past studies, the full CHAOS scale has shown adequate internal consistency (e.g., Petrill et al., 2004) and is negatively related to early language skills (e.g., Vernon-Feagans et al., 2012), as well as children’s screen media use (Emond et al., 2018).
2.3.2. Maternal Education Level.
For maternal education level, mothers indicated the highest level of education that they obtained at enrollment. Responses ranged from eighth grade or less to doctoral degree, and these were assigned to numeric values between 0 and 8 for analyses.
2.3.3. Siblings in the Home.
Siblings in the home was collected during the first parent meeting from a roster of everyone currently living in the home. Rosters were coded for whether or not the child had any siblings under 17 living with them in the home.
2.3.4. Maternal Depression.
Maternal depression risk was measured at both parent meetings. During the first parent meeting, maternal depression risk was measured using the Edinburgh Post-Natal Depression Scale (EPDS; Cox, Holden, & Sagovsky, 1987). The EPDS does not measure postpartum depression severity; instead, it assesses the probability of postpartum depression. Mothers responded to 10 items that gauged how they have felt during the past week. Items are scored on a zero- to three-point Likert scale; higher scores indicated a higher probability of depression (possible range = zero to 30). Prior research has shown that the EPDS is a valid screener of postpartum depression (Gaynes et al., 2005); Cronbach’s alpha for the current study was 0.86. During the second parent meeting, a postpartum measure was no longer appropriate, and thus depression risk was measured using the Center for Epidemiological Studies – Depression (CES-D; Radloff, 1977). The CES-D has 20 items scored on a Likert scale from zero (rarely or none of the time) to three (most or all of the time) for a possible range of zero to 60. The CES-D has been found to have high internal consistency and was moderately correlated with similar measures; Cronbach’s alpha for the current study was 0.78.
2.3.5. Parenting Stress.
Parenting stress was measured using the Parenting Stress Index – Short Form (PSI; Abidin, 2012) during both parent meetings. The PSI short form included 36-items that assessed parenting stress based on the mothers’ perception of mother-child interactions. Examples include “I feel trapped by my responsibilities as a parent” and “I expected to have closer and warmer feelings for my child than I do, and this bothers me.” Items were scored using a one- (strongly disagree) to five-point (strongly agree) Likert scale. The total stress score was used in the final analyses (range for the total stress score = 36 to 180). The total stress score of the PSI – short form has been found to have high internal consistency and is highly correlated with the full-length PSI (r = .98). Cronbach’s alpha for the current sample was .93.
2.4. Missing Data
The proportion of missing data ranged from 0.5% to 25%, with the greatest amount of missing data on time 2 maternal depression risk variable. Missing value analysis was conducted on all of the variables included in the model. Results from the Little’s MCAR test provided evidence the data were missing completely at random (χ2 = 26.55, df = 35, p = .85).
2.5. Power Analysis
A power analysis was conducted using Monte Carlo simulation, following the method described by Lebeau (2019), in R 4.0.2 (R Core Team, 2020), using simglm (LeBeau, 2019) and lme4 (Bates et al., 2007) packages. We simulated 1000 samples using standardized β coefficients for the effect sizes as determined by previous research (Dore et al., 2020 [βmom_ed = −0.11]; Vandewater et al., 2005, [βsiblings = 0.55]; and Conners et al., 2007 [βdepression = .19, βstress = .19]). The hierarchical nature of the design was accounted for by nesting time within mothers. Household TV (i.e., whether or how much the TV is on), measured at two time points, was the outcome; mother’s education and whether there were siblings in the home were included as time-invariant predictors; and depression risk and stress, measured at two time points, were included as time-varying predictors. The interactions between time and depression risk and time and stress were also included in the model to examine if the model had the power to detect if the effects of depression risk and stress varied by time point. Power was computed from 1000 replications of sample sizes of 150, 200, and 250. The results showed that for sample sizes at or above 200, the models had power of at least .80 at α = .05 to detect the expected effect sizes for mother’s education, siblings, depression risk, stress, and the interactions between time and depression risk and time and stress.
2.6. Analytic Strategy
We first used multiple imputation to impute 40 datasets (Graham, 2012) to use in the analyses. Then, to answer the first research question, what is the prevalence of household TV (i.e., whether or how much the TV is on) in this sample, we used the collected data to report descriptive statistics, including means, standard deviations, and ranges for the household TV variables at both time points. We compared this variable across time points using a Wilcoxon test for paired samples.
To answer the second research question, do demographic characteristics (i.e., maternal education level and presence of siblings in the home) and maternal mental health (i.e., depression risk and parenting stress) predict household TV (i.e., whether or how much the TV is on), we tested a multilevel regression analysis with an ordinal outcome, using the multiply imputed data sets with household TV use as the outcome and mother’s education level, the number of siblings in the home, maternal depression risk, parenting stress, and timepoint as predictors, including the interactions between timepoint and the other predictors to assess whether predictors change for infants vs. toddlers.
Because we measured maternal depression risk differently from time 1 to time 2 (i.e., the measure at time 1 is a measure of postpartum depression risk and the measure at time 2 is a general measure of depression risk), we accounted for this difference during both imputation and analysis. We treated the depression risk variables as two different variables and estimated a coefficient for postpartum depression risk at time 1 and another coefficient for general depression risk at time 2, by using interaction terms between dummy variable codes for time and the two depression measures.
3.0. Results
In relation to our first research question, we found that at the first parent meeting when children were infants (about six months of age), 72% of mothers reported that it was “somewhat true” or “definitely true” that there is usually a TV turned on somewhere in their home (M = 2.93 out of 4, SD = 1.17). The parallel percentage at the second parent meeting, when children were toddlers (about 18 months of age), was 65% (M = 2.66 out of 4, SD = 1.23). See Table 1 for frequencies of mothers that gave each response. Household TV for infants (first parent meeting) was related to household TV for toddlers (second parent meeting; r = .43). See Table 2 for all descriptive statistics and Table 3 for all bivariate correlation results.
Table 1.
Frequencies of Household TV item
| n | % | |
|---|---|---|
| Parent Meeting 1 (n = 219) | ||
| Definitely true (4) | 85 | 38.5 |
| Somewhat true (3) | 78 | 35.3 |
| Not really true or untrue (2) | 23 | 10.4 |
| Somewhat untrue (1) | 21 | 9.5 |
| Definitely untrue (0) | 12 | 5.4 |
| Parent Meeting 2 (n = 164) | ||
| Definitely true (4) | 49 | 29.9 |
| Somewhat true (3) | 56 | 34.1 |
| Not really true or untrue (2) | 27 | 16.5 |
| Somewhat untrue (1) | 18 | 11.0 |
| Definitely untrue (0) | 14 | 8.5 |
Table 2.
Descriptives for Variables of Interest
| Measure | n | M | SD | Min | Max |
|---|---|---|---|---|---|
| Parent Meeting 1 | |||||
| Household TV | 220 | 2.93 | 1.17 | 0 | 4 |
| Maternal Depression | 218 | 5.85 | 4.92 | 0 | 19 |
| Parenting Stress | 220 | 60.51 | 18.43 | 33 | 125 |
| Parent Meeting 2 | |||||
| Household TV | 173 | 2.66 | 1.23 | 0 | 4 |
| Maternal Depression | 174 | 14.07 | 7.87 | 1 | 43 |
| Parenting Stress | 171 | 62.18 | 18.70 | 35 | 131 |
Note. Household TV = “There is usually a TV turned on somewhere in our home” from the Confusion, Hubbub, and Order Scale (Petrill et al., 2004). Maternal Depression at Parent Meeting 1 = Edinburgh Post-Natal Depression Scale (EPDS; Cox et al., 1987). Maternal Depression at Parent Meeting 2 = Center for Epidemiological Studies – Depression (CES-D; Radloff, 1977). Parenting Stress = Parenting Stress Index – Short Form (PSI; Abidin, 2012)
Table 3.
Bivariate Correlations for Variables of Interest
| Measures | 2. | 3. | 4. | 5. | 6. |
|---|---|---|---|---|---|
| 1. Household TV – PM1 | .43** | .02 | .05 | .06 | .16* |
| 2. Household TV – PM2 | -- | .20** | .15 | .18* | .25** |
| 3. Maternal Depression – PM1 | -- | .55** | .41** | .51** | |
| 4. Parenting Stress – PM1 | -- | .31** | .61** | ||
| 5. Maternal Depression – PM2 | -- | .46** | |||
| 6. Parenting Stress – PM2 | -- |
Note:
p<.05,
p<.01
To examine the predictors of household TV for infants and toddlers, a multilevel regression analysis was completed in Mplus (Muthén & Muthén, 2006). We used multiple imputation techniques to impute values for the mental health variables and the outcome (TV) variable to account for missing data. Specifically, we used Mplus to impute 40 datasets.
Results from the multilevel ordinal regression analysis showed that at the alpha level of .05, there was a significant main effect of maternal education on household TV (whether or how much the TV is on; B = −0.25, p = .04), such that lower maternal education was associated with higher household TV. The interaction between maternal education and time was not significant (p = .28), indicating that the association between maternal education and household TV was similar for infants and toddlers. There were no significant main effects for time (p = .73), stress (p = .16), time 1 maternal depression risk (p = 0.27), time 2 maternal depression risk (p = .28), race (p = .07), or siblings (p = .72), suggesting that neither these demographic characteristics nor maternal mental health were associated with household TV (whether or how much the TV is on). In addition, the interactions of these variables with time were not significant (p > .15), indicating that associations did not vary for infants and toddlers. See Table 4 for full regression results.
Table 4.
Multilevel regression results
| β | SE | p | |
|---|---|---|---|
| Time | 0.63 | 1.8 | .18 |
| Parenting Stress | 0.02 | 0.01 | .16 |
| T1 Maternal Depression | −0.05 | 0.04 | .27 |
| T2 Maternal Depression | 0.03 | 0.03 | .28 |
| Race | 0.66 | 0.37 | .07 |
| Education | −0.25* | 0.12 | .04 |
| Siblings | −0.14 | 0.40 | .72 |
| Parenting Stress × Time | 0.003 | 0.02 | .85 |
| Race × Time | −0.64 | 0.45 | .15 |
| Education × Time | 0.15 | 0.14 | .28 |
| Siblings × Time | 0.29 | 0.47 | .54 |
Note.
p < .05
4.0. Discussion
In this study, we set out to investigate the prevalence and predictors of household TV (whether or how much the TV is on) in the homes of a sample of infants and toddlers from low-income homes. Prior research has shown that household TV can disrupt both children’s play behaviors and parent-child interaction, which, in turn, are likely to influence developmental outcomes. Our results yielded three main findings: (a) household TV was highly prevalent in the homes of infants and toddlers from low-income homes in our sample; (b) lower maternal education was related to higher household TV for both infants and toddlers; (c) maternal mental health was not predictive of household TV for infants or toddlers. We will discuss each of these findings in turn.
4.1. Prevalence of Household TV
Replicating prior research, we found that mothers in our low-income sample reported high levels of household TV (whether or how much the TV is on) in their homes. More than half of mothers reported at least moderate levels of agreement at both parent meetings that there was usually a TV turned on in their home, whereas 72% of mothers did so when children were infants. Although it is difficult to compare directly with prior research given disparate methods for assessing household TV, this finding is in line with prior research showing that household TV is prevalent in the homes of infants and toddlers, especially those from low-income backgrounds (Lapierre et al., 2012; Masur et al., 2015).
4.2. Demographic Factors and Household TV
The finding that maternal education predicts household TV (whether or how much the TV is on) is in line with some prior research, such as Masur, Flynn, and Olson (2015), who found that household TV exposure was higher among toddlers from low-SES backgrounds than those from middle-SES households. However, other research has found that neither household income nor parent education was predictive of household TV among children under two (Vandewater et al., 2005). Notably, these prior studies examined both low- and middle-SES families, whereas the families in the current study were all part of a low-SES sample identified through their participation in the WIC program, so the finding that maternal education was predictive even within this population is a novel contribution to this literature.
Counter to some prior research (Vandewater et al., 2005), our results did not show an association between the presence of siblings in the home and household TV (whether or how much the TV is on). Vandewater et al. (2005) reported that children with no siblings had more background television than children with siblings, perhaps because older siblings are more likely to pay overt attention to television, leading parents to avoid keeping it on during other activities. The fact that we did not find such a relation in the current study may reflect our focus on families from low-SES backgrounds or our focus on overall household TV (whether or how much the TV is on) rather than solely background TV.
4.3. Maternal Mental Health and Household TV
Counter to our hypotheses, we found no significant associations between maternal mental health (depression, stress) and household TV (whether or how much the TV is on). This null effect was surprising as maternal mental health has been consistently linked to children’s TV viewing in infancy and toddlerhood (e.g., Anand et al., 2014). For example, our finding that parenting stress is not related to household TV differs from prior studies, such as Thompson and Christakis (2007), who found that maternal mental distress predicted overall TV viewing in children under three years of age.
However, much of the prior research has examined maternal mental health in relation to children’s TV viewing (e.g., Bank et al., 2012; Anand et al., 2014), rather than overall household TV as measured here (whether or how much the TV is on). One possible explanation for our divergent findings is that there are different predictors of these two TV variables: variable characteristics like mothers’ stress and depression risk may play an important role in how much TV they purposely expose their children to, but the tendency to leave the TV on throughout the day, including in the background of other household activities, may be linked to demographic factors like education. Household TV may reflect habitual patterns of behavior that carry over from mothers’ lives before their children’s arrival, whereas children’s purposeful TV watching is likely to be influenced by specific circumstances during parenthood, including mothers’ mental health. Relatedly, there was no difference between predictors for infants versus toddlers, potentially in line with the idea that household TV is less related to characteristics that vary over time than other types of TV exposure.
Another possible explanation is we examined parenting stress and depression risk rather than clinical levels of anxiety or depression, and although there was variability on these measures, overall, mothers in our sample did not report experiencing these mental health difficulties at high levels. However, levels of depression were comparable to other studies examining mothers in low-income samples that have found such associations (Burdette et al, 2003; Conners et al., 2007). Studies examining mothers with clinical levels of these challenges may find differing associations with household TV (whether or how much the TV is on), which would have additional implications for targeting families for intervention.
4.4. Implications
Our finding that household TV (whether or how much the TV is on) was high among this sample of mothers and their young children from low-SES homes highlights the need for intervention around this topic among at-risk families. Given the low overt attention that infants and toddlers pay to adult-directed TV, parents may assume that the presence of household TV does not have any impact on children. However, research suggests that household TV can affect both children’s solo play and dyadic interactions between children and their caregivers (Kirkorian et al., 2009; Schmidt et al., 2008). Future research should explore educational interventions to reduce household TV for young children from disadvantaged backgrounds. The predictors of household TV in this population may also help identify families that could benefit from intervention. Even within this low-SES sample, the mothers with the lowest levels of education reported the highest levels of household TV exposure, suggesting that maternal education may be particularly important to consider in targeting any interventions or educational efforts to reduce TV in this population.
4.5. Limitations and Future Research
Although the current study’s findings are novel contributions to the extant literature on household TV for low-income infants and toddlers, we note several important limitations. One limitation of this study is that our measure of household TV (whether or how much the TV is on) was based on a single item. Future research should use more extensive measurements of household TV, including observational assessments. Furthermore, more nuanced measures of household TV could capture both the extent of household TV in the home and the amount of household TV that children are exposed to (i.e., that occurs in the same room, Masur, Flynn, & Olson, 2015).
It would also be useful for future research to follow women for a longer period of time, including prior to the birth of children, to examine the extent to which household TV (whether or how much the TV is on) is consistent across time and whether it is associated with similar demographic factors both before and after children. Understanding the role of household TV in different stages of individuals’ lives and the motivation for higher levels of use may be valuable in determining intervention and education strategies.
4.6. Conclusion
In sum, the current study shows that household TV (whether or how much the TV is on) was prevalent in the homes of low-income infants and toddlers in our sample. High levels of household TV could negatively impact children from low-income homes who are already at-risk for poorer developmental outcomes than their middle-SES peers. Counter to prior research, we found that maternal mental health was not linked to household TV, whereas maternal education was a consistent predictor for infants and toddlers. These findings can be used to inform future research and intervention efforts.
Highlights.
Household TV may be detrimental to children’s development.
What is the prevalence of household TV among infants/toddlers from low-income homes?
What are the predictors of household TV among infants/toddlers from low-income homes?
Mothers reported a high level of household TV, especially during infancy.
Education, but not siblings, depression, or stress, was related to household TV.
Acknowledgements
This work was supported in part by the Crane Center for Early Childhood Research and Policy of The Ohio State University and the Institute for Population Research at The Ohio State University, which was supported by NICHD (P2CHD058484). The content is solely the authors’ responsibility and does not necessarily represent the official views of the funders. We want to thank the research team and our partner, Columbus Public Health, without whom this study would not have been possible, and Women, Infant, Children centers in Franklin County, OH. Special thanks to James Uanhoro, Susie Mauck, and Adam Scharfenberger for assistance with data analysis.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Declaration of interest: none
References
- Bank AM, Barr R, Calvert SL, Parrott WG, Mcdonough SC, & Rosenblum K (2012). Maternal depression and family media use: A questionnaire and diary analysis. Journal of Child and Family Studies, 21(2), 208–216. doi: 10.1007/s10826-011-9464-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Burdette HL, Whitaker RC, Kahn RS, & Harvey-Berino J (2003). Association of maternal obesity and depressive symptoms with television-viewing time in low-income preschool children. JAMA Pediatrics, 157(9), 894–899. doi: 10.1001/archpedi.157.9.894 [DOI] [PubMed] [Google Scholar]
- Conners NA, Tripathi SP, Clubb R, & Bradley RH (2007). Maternal characteristics associated with television viewing habits of low-income preschool children. Journal of Child and Family Studies, 16(3), 415–425. doi: 10.1007/s10826-006-9095-0 [DOI] [Google Scholar]
- Cox JL, Holden JM, & Sagovsky R (1987). Detection of postnatal depression. Development of the 10-item edinburgh postnatal depression scale. The British journal of psychiatry, 150(6), 782–786. [DOI] [PubMed] [Google Scholar]
- Farr SL, Dietz PM, O’hara MW, Burley K, & Ko JY (2014). Postpartum anxiety and comorbid depression in a population-based sample of women. Journal of Women’s Health, 23(2), 120–128. doi: 10.1089/jwh.2013.4438 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gaynes BN, Gavin N, Meltzer-Brody S, Lohr KN, Swinson T, Gartlehner G, … Miller WC (2005). Perinatal depression: Prevalence, screening accuracy, and screening outcomes: Summary. Rockville, MD: Agency for Healthcare Research and Quality. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lapierre MA, Piotrowski JT, & Linebarger DL (2012). Background television in the homes of US children. Pediatrics, 130(5), 839–846. doi: 10.1542/peds.2011-2581 [DOI] [PubMed] [Google Scholar]
- Leigh B, & Milgrom J (2008). Risk factors for antenatal depression, postnatal depression and parenting stress. BMC Psychiatry, 8(1), 24. doi: 10.1186/1471-244X-8-24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Masur EF, Flynn V, & Olson J (2015). The presence of background television during young children’s play in American homes. Journal of Children and Media, 9(3), 349–367. doi: 10.1080/17482798.2015.1056818 [DOI] [Google Scholar]
- Mclearn KT, Minkovitz CS, Strobino DM, Marks E, & Hou W (2006). The timing of maternal depressive symptoms and mothers’ parenting practices with young children: Implications for pediatric practice. Pediatrics, 118(1), e174–e182. doi: 10.1542/peds.2005-1551 [DOI] [PubMed] [Google Scholar]
- Petrill SA, Pike A, Price T, & Plomin R (2004). Chaos in the home and socioeconomic status are associated with cognitive development in early childhood: Environmental mediators identified in a genetic design. Intelligence, 32(5), 445–460. doi: 10.1016/j.intell.2004.06.010 [DOI] [Google Scholar]
- Radloff LS (1977). The ces-d scale:A self-report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385–401. doi: 10.1177/014662167700100306 [DOI] [Google Scholar]
- Thompson DA, & Christakis DA (2007). The association of maternal mental distress with television viewing in children under 3 years old. Ambulatory Pediatrics, 7(1), 32–37. doi: 10.1016/j.ambp.2006.09.007 [DOI] [PubMed] [Google Scholar]
- Thompson DA, & Tschann JM (2016). Factors contributing to background television exposure in low-income mexican–American preschoolers. Maternal and child health journal, 20(9), 1835–1841. doi: 10.1007/s10995-016-1986-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Vandewater EA, Bickham DS, Lee JH, Cummings HM, Wartella EA, & Rideout VJ (2005). When the television is always on: Heavy television exposure and young children’s development. American Behavioral Scientist, 48(5), 562–577. doi: 10.1177/0002764204271496 [DOI] [Google Scholar]