Mass media plays a key role in the formation of individuals and social groups. Television (TV) can provide children with information that may conflict with the knowledge and guidance that a family or school gives them (Yalgin et al, 2002; Matziou et al, 2006).
TV is a source of learning and entertainment, but it can also shape behavioural models and alter a child's personality and character (Tandon et al, 2011; Rideout, 2013). TV is often used by children to remedy loneliness and fear (Christakis and Zimmerman, 2006; Taveras et al, 2006). There is evidence to suggest that there has been a tremendous increase in the time children and adolescents spend watching TV.
Children are considered passive users of TV, because of their inability to distinguish between programmes that are useful or harmful, real or fake (Christakis et al, 2004; Sigman, 2007). Studies have shown that the frequency and duration of TV viewing can cause physical, mental, and behavioural disorders in children and adolescents (Christakis et al, 2004; Sigman, 2007).
Many TV series or programmes include scenes of violence, which can affect children's character and make them vulnerable to immorality and underestimation of the value of human life (Bandura, 2002). Bandura (2002) concluded that TV can help to reinforce aggressive behaviour, as children tend to imitate what they see and that violence in TV programmes plays an important role in children's future behaviour. In addition, a study by Huesmann et al (2003) on 450 children aged 6‒10 years old who watched aggressive films, showed that children identified with aggressive heroes and did not differentiate reality from fictional TV violence. These children displayed aggressive behaviour 15 years later at follow up, regardless of gender, socioeconomic or mental status and factors related to their parents' status.
Additional research has shown a significant negative correlation between the time children spend watching TV and their school performance. Imitation of violent behaviour on TV, desensitisation because of frequent exposure leading to tolerance, and identification with violent heroes were among the most powerful mechanisms for violence in children and adolescents (Turow, 1996). Children who spend a lot of time watching TV may have poorer school performance and less interest in other social activities (Sharif et al, 2010).
Exposure to TV programmes may also have a negative effect on children's nutritional patterns and sleep habits. According to Nuutimen et al (2013), the time children and adolescents spend watching TV or playing computer/video games can influence sleep habits by decreasing the duration of rest time. Children who delay going to bed because of time spent watching TV often have a poor school performance and educational disabilities (Oka et al, 2008). Studies have shown that the main causes of sleep disorders among children of school age is watching TV for more than 2 hours per day and having a TV set in their room (Nuutimen et al, 2013). When children who are 4–10 years old spend more time watching TV, this reduces the time remaining for night rest and sleep (Li et al, 2008; Oka et al, 2008; Cain and Gradisan, 2010). Additionally, the time available for rest and sleep decreases as children grow older (Spruyt et al, 2005). Cain and Gradisan (2010) investigated the relationship between sleep and electronic media in school-aged children and adolescents, including TV viewing, and concluded that delayed bedtime and shorter total sleep time were most consistently related to media use.
Overexposure to TV programmes has been correlated with disorders in sleep-awakening pattern, excessive sleepiness and sleep disorders in 5‒6-year-old children (Li et al, 2008; Cain and Gradisan, 2010). Li et al (2008) found that fewer hours of rest and sleep are correlated with problems concentrating, poor school performance, more aggressiveness, behavioural disorders, and increased likelihood of having an accident. A study in Washington DC and Minnesota revealed that babies up to 2 years old are exposed to TV for between 1 and 1.5 hours per day (Zimmerman et al, 2007). Parents who participated in this survey reported that they were using TV for education, entertainment and mainly as a ‘babysitter’.
The American Academy of Pediatrics has expressed concerns about the time children spend watching TV (Jordan et al, 2006; Bleakley et al, 2013). They have produced guidelines focused on how parents can decrease time spent watching TV and control the quality of TV programmes (Garrison et al, 2011; Bleakley et al, 2013). Removing TV sets from children's rooms and motivating children and adolescents to take part in other activities, such as reading books or practicing sports have also been recommended (Patriarca et al, 2009). Patriarca et al (2009) studied TV and other electronic media use in children and adolescents in Italy and found that the majority of children and teenagers had a TV in the bedroom, and almost half of them always watched television there.
Healthcare professionals should TV parents and teachers about the negative impact of TV on children's physical and mental health (Higuchi et al, 2005; Nuutimen et al, 2013). Snell et al (2007) found that there was a positive correlation between sleeping for less than 8 hours and body mass index increase. On the other hand, increased sleep time duration (10‒11 hours per night) was found to be related to decreased body mass index (Snell et al, 2007). Other negative results of time spent watching TV include nutritional disorders, such as obesity, that may cause heart attack, cancer and type 2 diabetes (Li et al, 2008; Zhang et al, 2010). Patriarca et al (2009) found that 54.1% and 61% of the children usually had lunch and supper respectively while watching TV. These correlations between preschool TV viewing habits and later outcomes for children may be an important topic for future longitudinal studies.
The relationship that young children develop with TV has troubled both parents and the scientific community. Children who are 1 years old watch 1‒2 hours each day, increasing to 2‒4 hours as children reach the age of 4 years old (Comstock and Scharrer, 2001). Parental mediation helps young children understand the conventions of TV programmes, but also acts as a regulator of children's use. The parental role is of great importance, especially when protecting children from violence on TV. Children can experience sleep disorders because of TV viewing; thus, parental guidance is essential (Nuutimen et al, 2013).
Aims
The main aims were to assess preschool children's TV viewing habits and sleep habits, as well as the way the time spent watching TV affects children's sleep patterns.
The secondary aims were to assess parent's perceptions of TV viewing habits and how their perceptions, attitudes and demographic characteristics (such as educational level and age) may correlate with children's sleep disorders.
Methods
This cross-sectional study was conducted in randomly selected kindergartens in Thessaloniki, Greece. A lucky numbers table was used to select kindergartens randomly. A lucky numbers table is a method for recruiting a random sample, where each participant is given a number (code) and a code is chosen randomly from a table by a blinded co-researcher. Every fifth code to the right of the one selected in the table is then selected for inclusion in the study. The same method was used to randomly choose parents to participate, provided they met the eligibility criteria, which were having a child who was 1‒5 years old, good knowledge of Greek, and the absence of chronic illness. Power analysis of 0.8 with P<0.05 was used to determine the sample size. The sample consisted of 100 children aged 1‒5 years old and their parents.
Data collection
The Children's Sleep Habits Questionnaire (CSHQ) was developed by Owens et al (2000) and is used to record children's sleep habits. The questionnaire consists of 22 closed questions concerning children's sleep habits, possible difficulties falling asleep and their sleep behaviour in the last 7 days. Questions are scored according to the following scale: 7=always, 6‒5=often, 4‒2=sometimes, 1=rarely and 0=never. The total score ranges from 22 to 110, with higher scores reflecting more sleep disorders. The scores for questions 1, 2, 3, 10 and 19 were reversed, according to developer instructions (Goodlin–Jones et al, 2008).
The Children's Television-Viewing Habits Questionnaire (CTVQ) developed by Owens et al (1999) was used to evaluate parents' perceptions of TV viewing habits at home. The first part consists of 18 closed questions, which include items regarding the influence of time spent watching TV on children's behaviour and parents' compliance with prevention standards for children's TV viewing. The second part asks about viewing habits at home and consists of 15 closed questions about time spent watching TV, programmes' content and children's participation in other activities. Cronbach's alpha showed that the internal consistency of the tool was 0.71.
The demographic characteristics questionnaire covered information such as parents' and children's gender, age, nationality, house structure, parents' educational level, occupation and number of siblings.
A 3-step process for translating the questionnaires into Greek was followed. Two independent translators translated the original English versions into Greek. Each Greek version was given to another two translators who worked separately to perform a backward translation. A panel of experts, consisting of two university professors, one paediatrician and two psychologists specialised in children, compared the original English version with the back-translation of the questionnaire. Minor changes regarding linguistic differences between the two texts and cultural adaptations were made. Permission was obtained from Owens to translate the questionnaires.
Time to complete the demographic sheet and the questionnaires did not exceed 15 minutes. A researcher was present during completion, in order to address any possible questions about the questionnaires.
Data analysis
Data analysis was conducted using the Statistical Package for Social Sciences 19.0 (SPSS Inc., Chicago, IL, USA). Continuous data were expressed as means (±standard deviation) and medians (range), while categorical and dichotomous variables were expressed as absolute values (n) and percentages (%) of the groups. The Kolmogorov-Smirnov test for normality and normality plots were used to check the normal distribution of continuous variables. The Student's t-test was used to examine whether a quantitative normally distributed variable presented differences between groups. Pearson's R was used to assess the correlation between two quantitative normally distributed variables, while Spearman's rho was used for variables that were not normally distributed. Pearson's chi-squared test and Fischer's exact test were used to determine if there were differences between groups in terms of the percentages of some characteristics. When two or more independent variables were found to be at an alpha level of 0.2 (P<0.2), multivariate linear regression was performed. A backward stepwise linear regression was conducted and coefficients beta (B) and 95% confidence intervals were calculated. A two-tailed P<0.05 was considered to demonstrate statistical significance.
Ethical considerations
Approval was obtained from the Pedagogic Institute of the Educational Ministry of Greece. Parents were approached at the entrance to the kindergartens, at the beginning or end of the school programme. Meetings with parents were scheduled by the kindergarten directors. At the appointment, participants were led to a separate room where the aims of the study were explained and parents were informed that their participation was voluntary and anonymity would be maintained. Every parent was given a copy of the permission from the Pedagogic Institute of the Educational Ministry of Greece and a text with information about the aims and methodology of the study. Written informed consent was obtained by parents who met the eligibility criteria and they were notified about their ability to withdraw from the study at any time. After giving their consent, parents completed the demographic sheet and the questionnaires.
Results
Demographics
Demographic characteristics of the sample are shown in Table 1. The majority (52%) of the children were boys. The children's mean age was 3.5±0.9 years. The parents' mean age was 37.4±4.4 years (fathers), 32.9±4.0 years (mothers).
Table 1. Demographic characteristics of sample
| Characteristic | n=100 |
|---|---|
| Gender (children) | |
| Boys | 52 |
| Girls | 48 |
| Nationality | |
| Greek | 9 |
| Albanian | 44 |
| Armenian | 8 |
| English | 6 |
| Bulgarian | 14 |
| Russian/Georgian | 3 |
| Italian | 2 |
| Ukranian | 12 |
| Czech | 2 |
| Occupation (mother) | |
| Unemployed/retired | 51 |
| Public servant | 0 |
| Private clerk | 49 |
| Occupation (father) | |
| Unemployed/retired | 10 |
| Public servant | 0 |
| Private clerk | 89 |
| Education (mother) | |
| <2 years at school | 0 |
| Elementary | 5 |
| High school | 41 |
| Senior high school (lyceum) | 21 |
| College | 14 |
| Technological educational institute | 6 |
| University | 8 |
| Master/doctoral degree | 5 |
| Education (father) | |
| <2 years at school | 0 |
| Elementary school | 4 |
| High school | 40 |
| Senior high school (lyceum) | 33 |
| College | 6 |
| Technological educational institute | 5 |
| University | 10 |
| Master/doctoral degree | 2 |
| Place of living | |
| Apartment/block of flats | 92 |
| House (cottage, detached etc) | 8 |
| Characteristic Mean (SD) | |
| Number of rooms | 3.0 (1.3) |
| Children's age (years) | 3.5 (0.9) |
| Father's age (years) | 37.4 (4.4) |
| Mother's age (years) | 32.9 (4.0) |
| Number of siblings | 2.0 (0.9) |
Parents' and children's television viewing habits
Children spent 2–3 hours watching TV daily, and 3 hours during weekends. Overall, 13% of children usually watched TV with their parents and 74% of parents were responsible for the quality and quantity of TV programmes. The majority (84%) of children did not have a TV set in their room. The most preferred TV programmes were cartoons (60%) and TV shows for children (27%). The majority of children (63%) preferred to play with other peers or spend time doing outdoor activities, rather than watch TV.
Many parents (48%) did not agree that TV affects their children, and 25% reported that TV may affect children negatively. The majority (73%) of parents had a neutral opinion on the content of TV programmes, while only 3% of them reported a positive opinion. The majority of parents (85%) were responsible for time spent watching TV. The average time parents spent watching TV with their children was 1.5 hours daily and 2 hours during weekends. Parents reported that most of their children (76%) were influenced by the content of TV advertisements. The majority (73%) of parents rated the quality of TV programmes as average. Almost all of them (96%) encouraged their children to participate in other activities, although only 26% of them spent time in sports activities themselves.
The correlation between demographic characteristics and time spent watching TV daily is shown in Table 2. Variables that were statistically significant at P≤0.2 were inserted into the multivariable linear regression model. This regression analysis found that children's age (P=0.01), mothers' educational level (P=0.001) and number of siblings (P=0.001) were related to time spent watching TV daily. Older children spent more time watching TV daily, while children whose mother had a higher educational level tended to watch TV less, as did children with siblings. These variables explained 6% of the variance for time spent watching TV daily.
Table 2. Correlation between demographic characteristics and time children spent watching television daily
| Variable | Statistical test value | P |
|---|---|---|
| Child's gender | 0.38* | |
| Boys | 1.75 (1.17)† | |
| Girls | 1.66 (1.26)† | |
| Child's age | 0.14‡ | 0.001 |
| Father's age | -0.08‡ | 0.09 |
| Mother's age | -0.07‡ | 0.13 |
| Mother's education | -0.20‡ | <0.001 |
| Father's education | -0.17‡ | <0.001 |
| Number of siblings | -0.16‡ | <0.001 |
Note:
*Student's t-test,
†Mean (standard deviation),
‡Spearman's rho
Children's age (P=0.001) and mothers' educational level (P=0.01) were statistically significant influences on time spent watching TV during weekends. These two variables explained 5% of the variance for time spent watching TV during weekends (Table 3).
Table 3. Correlation between demographic characteristics and time children spent watching television at the weekend
| Variable | Statistical test value | P |
|---|---|---|
| Child's gender | 0.43* | |
| Boys | 2.88 (2.42)† | |
| Girls | 2.74 (1.67)† | |
| Child's age | 0.23‡ | <0.001 |
| Father's age | -0.01‡ | 0.77 |
| Mother's age | 0.04‡ | 0.37 |
| Mother's education | -0.17‡ | <0.001 |
| Father's education | -0.12‡ | <0.001 |
| Number of siblings | -0.07‡ | 0.1 |
Note:
*Student's t-test,
†Mean (standard deviation),
‡Spearman's rho
Parents' educational level (P=0.01 for fathers; P=0.04 for mothers) and number of siblings (P=0.01) were statistically significant factors in time spent watching TV with parents. Children whose parents were of higher educational level tended to spend less time watching TV with their parents, as did children who had siblings. These variables explained 5% of the variance for time spent by children and parents watching TV together (Table 4).
Table 4. Correlation between demographic characteristics and time children spent watching television with parents
| Variable | Statistical test value | P |
|---|---|---|
| Child's gender | 0.98* | |
| Boys | 0.94 (1.09)† | |
| Girls | 0.94 (1.01)† | |
| Child's age | 0.12‡ | 0.006 |
| Father's age | -0.04‡ | 0.33 |
| Mother's age | -0.02‡ | 0.66 |
| Mother's education | -0.13‡ | 0.003 |
| Father's education | 0.15‡ | 0.001 |
| Number of siblings | -0.09‡ | 0.03 |
Note:
*Student's t-test,
†Mean (standard deviation),
‡Spearman's rho
Children whose fathers were of higher educational level reported increased levels of parental control on the time spent watching TV than those of lower educational level (P=0.009). No statistically significant relationship was found between demographic characteristics and parents' perception regarding the impact of TV on their children, or between demographic characteristics and parents' encouragement to spend time on outdoor activities.
Children's sleep habits
Almost half of the children did not use a toy (such as a teddy-bear or dolls) to help them fall asleep. The majority (59%) of them had their own bed and did not need their parents' presence in order to fall asleep. Most children were reported to fall asleep at the same time and rest for the same time period every night. Only 35% of the children were nervous and moved on the bed during sleeping hours. The majority (85%) did not gnash, 80% did not snore and 86% did not have nightmares. The majority (67%) of the children did not take a nap during the day and 92% did not fall asleep during other activities.
The mean CSHQ score was 37.6±9.2. Mean sleep time at night was 10.4±1.4 hours, and mean nap time (usually in the afternoon) was 46±53 minutes.
Fathers' age (P=0.04), number of siblings (P=0.02) and time spent watching TV daily (P=0.001) affected CSHQ score. Older fathers' children tended to score lower than those of younger fathers. An increase in time spent watching TV daily was correlated with an increase in CSHQ score. Having more siblings was also associated with a lower CSHQ score. These variables explained 6% of the variance for the CSHQ score (Table 5).
Table 5. Correlation between demographic characteristics and Children's Sleep Habits Questionnaire score
| Variable | Statistical test value | P |
|---|---|---|
| Child's gender | 0.41* | |
| Boys | 43.2 (9.3)† | |
| Girls | 43.8 (8.5)† | |
| Child's age | 0.01‡ | 0.97 |
| Father's age | -0.13‡ | 0.004 |
| Mother's age | -0.10‡ | 0.02 |
| Mother's education | -0.06¶ | 0.21 |
| Father's education | -0.09¶ | 0.048 |
| Number of siblings | -.014‡ | 0.001 |
| Time (hours) children spent watching TV daily | 0.12‡ | 0.007 |
| TIme (hours) children spent watching TV on weekends | 0.07‡ | 0.09 |
| Time (hours) children spent watching TV with their parents (daily) | 0.10‡ | 0.02 |
| Time (hours) children spent watching TV with their parents during weekends | 0.06‡ | 0.21 |
Note:
*Student's t-test,
†Mean (standard deviation),
‡Pearson's r
¶Spearman's rho
Discussion
In this study, preschool children's sleep habits in relation to time spent watching TV was investigated. Time spent watching TV was negatively correlated with CSHQ score. Fathers' age and number of siblings were found to be modifying factors of children's sleep habits.
Sleep disorders related to TV viewing habits usually include bedtime refusal, delayed sleep onset or difficulty settling, shortened sleep duration and frequent night arousals because of nightmares (Owens et al, 1999). In the present study, sleep disorders included difficulties falling asleep, waking in the night more than once because of a nightmare and tiredness or sleepiness during the day.
A statistically significant correlation between children's age and time spent watching TV was found both during weekdays and weekends. Spruyt et al (2005) showed that as children grow older, time for rest and sleep decreases. Cain and Gradisan (2010) studied children 4‒10 years old and concluded that extended time spent watching TV is linked to having less time available for rest and sleep. In their study, time spent watching TV ranged from 2.5 hours daily to 3 hours on weekends. By comparing children who spend less than 2 hours watching TV to their peers who watch it for more than 2 hours per day, the researchers found that the latter group suffered from increased stress and sleep disorders, waking up during night hours and difficulty falling asleep (Cain and Gradisan, 2010).
In the present study, the majority of children (79%) usually watched TV between 5pm and 9pm every evening. Children who watch TV after 9pm more frequently experience sleep disorders than those who do not. Watching TV before going to bed is significantly related to negative sleep impact, sleep disorders, delayed bedtime and reduced total sleep time, especially for 5‒6-year-old children (Cain and Gradisan, 2010). Decreased sleep time may result in lack of concentration and poor school performance (Li et al, 2008). Children with less sleep time were found to be more aggressive and more vulnerable to injuries because of decreased reaction time and manifested behavioural problems. Higuchi et al (2005) found that night exposure to the strong light produced by TVs may reduce melatonin levels and have a negative impact on children's circadian rhythms. The findings of the present study showed that children had an average of 10.4 hours of sleep time daily, while their average nap time was 46 minutes per day.
Parents' perceptions of viewing habits and their characteristics significantly affect CSHQ score and time spent watching TV (Zhang et al, 2010). The present study's findings showed that 25% of parents believed that TV could have a negative impact on their kids, while 48% had a neutral opinion of the effect of TV. This agrees with other studies where higher educational level of parents was related to decreased timer spent watching TV daily and during weekends. In the present study's sample, 13% of mothers and 12% of fathers had academic qualifications. Other studies have shown that parents of higher educational level were more informed about the right amount of time their children can watch TV and were stricter with time spent watching TV. Patriarca et al (2009) found that the children of families of lower socioeconomical status tended to spend more time watching TV. There was a statistically significant correlation between parent educational level and time children spent watching TV both daily and at weekends. Maternal educational level was a significant factor in time spent watching TV on a daily basis and at weekends. Only the father's educational level was found to correlate with children's sleep habits. However, there was no statistically significant difference or correlation between parental occupation and time spent watching TV or children's sleep habits.
The present study's results showed that only 13% of children usually watched TV with their parents, who can control the quantity and quality of TV programmes. Patriarca et al (2009) found that 49% of parents who accompanied their children watching TV had control of the amount and content of TV programmes. On the contrary, the presence of a TV set in a child's room is associated with a higher incidence of sleep disorders (including delayed sleep time and nightmares), because of children's exposure to violent TV content. In the present study, 84% of children did not have a TV set in their room. This is in contrast to Patriarca et al (2009), who found that the majority of children (89.5%) had a TV set in their room and 52.5% always watched TV there. A large proportion of parents who participated in the present study had a neutral opinion of the content of TV programmes, while few of them rated the TV content as good or very good. This conflicts with Zimmerman et al (2007), where parents reported that TV viewing is amusing and entertaining for children. Zimmerman et al (2007) also reported that parents encourage children to spend time watching educational TV programs. The present study found that the majority (96%) of parents encouraged their children to spend time in other activities (such as reading a book, listening to music, or playing a sport) instead of watching TV. The preferred activities included reading a book (30%) and sport activities (26%).
Conclusions
The results of this study are based on parental reports rather than preschool children's self-evaluations, since instruments determining parents' opinions were used. As preschool children are under parental control in Greece, obtaining the most accurate information regarding children's TV habits and sleep patterns is of great importance and is better achieved by assessing parental reports than children's self-estimations. The results highlight that parents were responsible for the quality and quantity of TV programmes that their children watched. Few parents reported a positive opinion of the content of TV programmes, while the majority encouraged their children to participate in other activities rather than watching TV shows. Parental educational level and the presence of siblings was associated with shorter time spent watching TV and better control of TV programmes. Prolonged time spent watching TV daily was linked with more frequent sleep disorders.
The impact of parents' education level on control of TV watching time has to be acknowledged. It is encouraging that many parents reported that there was no TV set in their children's rooms, and that they guided their children to participate in other activities, instead of watching TV.
Recommendations
As the time children spend watching TV increases, there is an urgent need for studies that assess the impact of TV. Studies evaluating factors that affect preschool children and their families in relation to the use of mass media in general should be planned. Parents should understand the strong link between the time children spend watching TV and sleep disorders. Parents, teachers and healthcare professionals should encourage children and adolescents to participate in activities other than watching TV.
Finally, early recognition of sleep disorders symptoms (increased aggressiveness, loss of concentration, low school performance and hyperactivity) is of tremendous importance for children, because of the increased time they spend watching TV. Proper information about the negative impact of TV on children's sleep pattern and in-target interventions should be provided.
As the link between increased time spent watching TV and sleep disorders is well established, there are many topics for future research. The in-depth evaluation of young children's sleep patterns, documentation of children's and adolescents' TV viewing habits, parental attitudes towards the proper use of TV and the negative impact of the increased time watching TV on the children's and adolescents' nutritional status should be investigated in future studies. The possible correlation between preschool TV viewing habits and long-term outcomes for children should be flagged up as another important topic for future research, possibly a longitudinal study.
Limitations
As the study was conducted in kindergartens of the prefecture of Thessaloniki, because of ease of access, the results may not be generalisable.