GROWING UP IN SCOTLAND: THE CIRCUMSTANCES AND EXPERIENCES OF 3 YEAR OLD CHILDREN LIVING IN SCOTLAND IN 2007/08 AND 2013

Chapter 5 Child Television Viewing

At the age 3 interview parents in BC1 and BC2 were asked a series of questions to measure the amount of television their child watched. These questions asked about the number of days the child watched television over the past week as well as how long the child spent watching television on a typical weekday and over a typical weekend. Parents were asked to include any time the child spent watching a film or DVD on a computer or laptop - as well as a television - but to exclude any other type of 'screen time' such as using a games console or computer to play games.

A range of international studies have shown that high exposure to television viewing in the early years has an adverse impact on cognitive development (see e.g. Zimmerman and Christakis, 2005; Christakis et al., 2009). Within the UK, analysis of the Avon Longitudinal Study of Parents and Children (ALSPAC) data found that greater usage of television in the home during a child's early years was associated with lower levels of language development at 24 months (even after controlling for socio-economic differences, including maternal occupation and parental educational status) (see Roulstone et al., 2011).

Data from longitudinal studies have also been used to explore the links between television viewing and childhood obesity. Earlier analysis from GUS has found that child's screen time between the ages of 3 and 6 was not associated with overweight and obesity at age 6 (Parkes et al., 2012). By comparison, analysis of ALSPAC data has found that a high number of hours of exposure to television at age 3 was associated with an increased risk of childhood obesity at age 7, again after controlling for socioeconomic differences (Reilly et al., 2005). The researchers concluded that this was due to television being associated both with a decrease in physical activity, and an increase in dietary intake. It is possible that the lack of a significant association between screen time and obesity in earlier GUS analysis is because screen time was a measurement not only of time spent watching television but also of playing video games and using a computer. A review of studies on the links between obesity and sedentary activities, suggested that time spent watching television and time spent playing video games should be analysed separately. The researchers found that while there was evidence that suggested parents should limit the time their younger children spent watching television, the evidence for any link between obesity and the time spent playing video games was not as strong. The researchers concluded that the link between time spent viewing television and obesity was likely to be related to a higher energy intake during times of inactivity rather than being a symptom of sedentary activity itself (Rey-Lopez et al., 2008).

5.1 Number of days in past week child watched television

The vast majority of children in BC1 (82%) and BC2 (86%) had watched television every day in the seven days prior to the interview (Table 6.1). A very small proportion of children in both cohorts had not watched television on any day during the past week (4% in BC1 and 2% in BC2). These differences across the cohorts are statistically significant.

Table 5.1 Days watched television in past week, by cohort

	BC1	BC2
	%	%
None	4	2
1-3 days	6	5
4-6 days	8	7
7 days	82	86
Unweighted bases	4191	5013

Tested on category '7 days', differences between cohort p <.001.

Table 5.2 shows the number of days the child watched television over the past week by equivalised household income and cohort. Focussing on the proportion of children who watched television every day during the past week, it is difficult to see a clear, linear pattern between television viewing and income. However, children from households in the lowest income quintiles were more likely to watch television on every day in the past week than children in the highest two income quintiles. This pattern is evident for both BC1 and BC2. As noted before, number of days of television viewing has increased overall between BC1 and BC2. This increase can be seen in almost all income groups (with the exception of the second income quintile) in a similar way. As such, there has been no statistically significant change in the nature of the relationship between household income and television viewing across cohorts.

Table 5.2 Numbers of days in past week child watched television by equivalised household income (quintile) and cohort

	Lowest quintile	2nd quintile	3rd quintile	4th quintile	Highest quintile
	%	%	%	%	%
BC1
None	4	4	3	2	3
1-3 days	7	5	6	7	6
4-6 days	6	6	8	9	12
7 days	83	85	83	81	79
BC2
None	2	3	2	1	1
1-3 days	5	6	4	5	5
4-6 days	5	6	7	9	9
7 days	88	85	87	85	85
Unweighted bases - BC1	783	802	761	857	721
Unweighted bases - BC2	980	777	813	775	1021

Tested on category '7 days': differences by income p<.05; differences by cohort p< .001; cohort* income p=NS.

Differences in television viewing by parental level of education are shown in Table 5.3. Whilst there was no linear relationship between parental level of education and television viewing, for both cohorts, children of parents with no qualifications were more likely to watch television every day than children of parents educated to degree level (for example, in BC2, 91% compared with 84%). The proportion of children watching television every day has increased similarly across almost all parental education sub-groups. As such, there has been no change in the relationship between the number of days the child watched television over the past week and parental education between the cohorts.

Table 5.3 Numbers of days in past week child watched television by parental level of education and cohort

	No qualifications	Lower Standard Grades or VQs or Other	Upper level SGs or Intermediate VQs	Higher grades and upper level VQs	Degree level academic and vocational qualifications
	%	%	%	%	%
BC1
None	5	3	5	3	3
1-3 days	7	6	5	7	7
4-6 days	3	6	6	8	11
7 days	85	86	84	83	79
BC2
None	4	5	3	2	1
1-3 days	3	3	5	5	6
4-6 days	1	5	4	5	9
7 days	91	87	87	88	84
Unweighted bases - BC1	200	206	804	1378	1596
Unweighted bases - BC2	163	217	710	1428	2317

Tested on category '7 days': differences by education p<.001; differences by cohort p< .001; cohort* education p=NS.

Table 5.4 shows that for BC1 children there was no association between the number of days a child watched television in the past week and maternal age at the birth of the child. For BC2 children it appears as though maternal age is associated with television viewing -children of mothers aged 40 or over were less likely to watch television than children of mothers aged under 20 (81% versus 88%). However, differences by age were not statistically significant. There has therefore been no change in the nature of the relationship between the number of days the child watched television over the past week and maternal age between the cohorts.

Table 5.4 Numbers of days in past week child watched television by maternal age at child's birth and cohort

	Under 20 years old	20 to 29 years old	30 to 39 years old	40 or older
	%	%	%	%
BC1
None	4	4	3	5
1-3 days	8	6	6	4
4-6 days	6	7	9	9
7 days	82	83	82	83
BC2
None	2	2	2	2
1-3 days	5	5	5	6
4-6 days	4	6	7	12
7 days	88	86	86	81
Unweighted bases - BC1	221	1563	2229	155
Unweighted bases - BC2	217	1976	2570	227

Tested on category '7 days': differences by age p=NS; differences by cohort p< .001; cohort* education p=NS.

Differences in the number of days a child watched television in the past week were also examined by area deprivation (Table 5.5). Between BC1 and BC2, television watching increased for children in all areas, regardless of the level of deprivation. These differences are statistically significant. Table 6.5 shows that there is a slight difference between number of days of television viewing by area deprivation in BC1; children in the three most deprived areas were more likely to watch television every day over the past week than children in the least deprived areas (around 83% in the three most deprived areas compared with 80% in the two least deprived areas). In BC2, the pattern is similar; children living in more deprived areas are still more likely to have watched television every day than those living in less deprived areas. However, the level of difference between the groups was smaller than in BC1. Nevertheless, there has been no statistically significant change in the nature of the relationship between number of days the child has watched television and area deprivation between cohorts.

Table 5.5 Numbers of days in past week child watched television by area deprivation (quintiles) and cohort

	1 Most deprived	2	3	4	5 Least deprived
	%	%	%	%	%
BC1
None	4	4	4	3	3
1-3 days	6	6	5	8	7
4-6 days	6	7	8	9	10
7 days	84	83	83	80	80
BC2
None	3	2	2	2	1
1-3 days	4	5	5	6	5
4-6 days	5	5	6	9	8
7 days	88	87	87	83	86
Unweighted bases - BC1	833	698	872	883	905
Unweighted bases - BC2	942	935	1029	1063	1010

Tested on category '7 days': differences by area deprivation p<0.5; differences by cohort p< .001; cohort* area deprivation p=NS.

5.2 Time spent watching television on a typical weekday

Table 5.6 presents the time the child spent watching television on a typical weekday by cohort. The most common length of time children watched television on a typical weekday was between 1 to 2 hours (35% of children in BC1 and 38% of children in BC2). Children aged 3 in 2013 were more likely to watch television for longer than children aged 3 in 2007/08. The proportion of children watching more than three hours of television on a typical weekday was 12% in BC1 increasing to 15% of children in BC2. These differences are statistically significant.

Table 5.6 Time spent watching television on a typical weekday by cohort

	BC1	BC2
None	1	4
Up to 1 hour	28	20
Between 1 and 2 hours	35	38
Between 2 and 3 hours	24	24
Over 3 hours	12	15
Unweighted bases	4024	4958

Tested on category 'Over 3 hours': differences by cohort p <.05

Table 5.7 presents the number of hours the child watched television by equivalised household income and cohort. Looking specifically at children who watched more than three hours of television on a typical weekday, for both cohorts there was a linear relationship between household income and television viewing: the lower the income quintile, the more likely the child was to watch television for over three hours. For example, in BC2, 19% of children from households in the lowest income quintile watched more than three hours of television every day compared with 14% of children from households in the third income quintile and 5% of children from households in the highest income quintile. Differences by income were statistically significant. Each income quintile has seen an increase (between BC1 and BC2) in the proportion of children watching three hours or more of television on a weekday so there has been no statistically significant change in the strength of this relationship between the cohorts.

Table 5.7 Time child spends watching television on an average weekday by equivalised household income (quintile) and cohort

	Lowest quintile	2nd quintile	3rd quintile	4th quintile	Highest quintile
	%	%	%	%	%
BC1
None	0	0	1	1	1
Up to 1 hour	25	27	25	29	32
Between 1 and 2 hours	31	32	35	40	40
Between 2 and 3 hours	24	27	25	23	21
Over 3 hours	19	15	14	7	5
BC2
None	3	5	4	4	4
Up to 1 hour	16	20	18	22	27
Between 1 and 2 hours	32	33	40	41	44
Between 2 and 3 hours	27	25	24	23	18
Over 3 hours	21	16	15	9	6
Unweighted bases - BC1	741	768	732	834	698
Unweighted bases - BC2	963	769	808	769	1013

Tested on category 'Over 3 hours': differences by equivalised household income p< .001; differences by cohort p< .05; cohort* equivalised household income p=NS.

The amount time a child watched television on an average weekday also varied by level of parental education (Table 5.8). Children of parents educated to degree level were less likely to watch television for more than three hours than children of parents with no qualifications (in BC1, 7% compared with 20%). As noted earlier, BC2 children were more likely to watch more television overall and this increase is apparent amongst children in all education sub-groups. Thus there has been no notable change in the nature of the relationship between time spent watching television on a weekday and parental level of education between the cohorts.

Table 5.8 Time child spends watching television on an average weekday by parental level of education and cohort

	No qualifications	Lower Standard Grades or VQs or Other	Upper level SGs or Intermediate VQs	Higher grades and upper level VQs	Degree level academic and vocational qualifications
	%	%	%	%	%
BC1
None	<1	<1	<1	<1	1
Up to 1 hour	20	24	24	28	31
Between 1 and 2 hours	33	29	32	36	38
Between 2 and 3 hours	26	27	27	24	22
Over 3 hours	20	20	17	12	7
BC2
None	4	3	4	4	5
Up to 1 hour	12	13	14	18	25
Between 1 and 2 hours	32	31	30	38	42
Between 2 and 3 hours	26	32	29	25	19
Over 3 hours	26	21	23	15	9
Unweighted bases - BC1	186	200	756	1338	1537
Unweighted bases - BC2	160	208	698	1412	2304

Tested on category 'Over 3 hours': differences by education p< .001; differences by cohort p< .01; cohort* education p=NS.

Table 5.9 shows the amount of television a child watches on a typical weekday by the mother's age at the time of the child's birth and cohort. We see that there was an association between mother's age and the child's television viewing habits. Likelihood of watching three or more hours of television on a typical day decreases as maternal age increases. In BC1, 17% of children whose mother was aged under 20 watched television for more than three hours compared with 11% of children whose mother was aged 40 or older. The corresponding figures in BC2 were 23% and 11%. There has been no statistically significant change in the nature of this relationship between the cohorts.

Table 5.9 Time child spends watching television on an average weekday by maternal age at child's birth and cohort

	Under 20 years old	20 to 29 years old	30 to 39 years old	40 or older
	%	%	%	%
BC1
None	<1	<1	1	<1
Up to 1 hour	23	27	30	26
Between 1 and 2 hours	37	34	35	41
Between 2 and 3 hours	23	24	24	22
Over 3 hours	17	14	10	11
BC2
None	4	4	4	5
Up to 1 hour	16	18	22	27
Between 1 and 2 hours	26	36	41	36
Between 2 and 3 hours	31	24	22	21
Over 3 hours	23	17	11	11
Unweighted bases - BC1	209	1500	2145	147
Unweighted bases - BC2	211	1953	2547	224

Tested on category 'Over 3 hours': differences by maternal age p< .001; differences by cohort p=NS; cohort* maternal age p=NS.

Table 5.10 presents the number of hours the child watched television on an average weekday by area deprivation and cohort. If we consider those children that watch television for over three hours on a typical day, there is a statistically significant association between area deprivation and television viewing: the more deprived an area a child lived in, the more likely they were to watch more hours of television. For example, in BC1 nearly two in ten (18%) of children from the most deprived areas watched television for over three hours compared with nearly one in ten (8%) of children from the least deprived areas. The corresponding figures for BC2 are 19% compared with 8%. There was no statistically significant difference in results between the cohorts.

Table 5.10 Time child spends watching television on an average week day by area deprivation (quintiles) and cohort

	1 Most deprived	2	3	4	5 Least deprived
	%	%	%	%	%
BC1
None	0	1	0	1	1
Up to 1 hour	24	28	29	30	28
Between 1 and 2 hours	33	33	35	34	42
Between 2 and 3 hours	25	24	24	25	22
Over 3 hours	18	13	12	10	8
BC2
None	2	2	2	3	4
Up to 1 hour	18	23	24	27	27
Between 1 and 2 hours	33	34	37	39	42
Between 2 and 3 hours	27	25	24	22	20
Over 3 hours	19	16	14	10	8
Unweighted bases - BC1	793	669	832	848	882
Unweighted bases - BC2	926	926	1015	1052	1008

Tested on category 'Over 3 hours': differences by area deprivation p< .001; differences by cohort p=NS; cohort* equivalised household income p=NS.

5.3 Time child spent watching television during a typical weekend

Table 5.11 shows the amount of television that children aged three in Scotland watched during an average weekend. Children in BC2 spent significantly more time watching television during an average weekend than children in BC1. Looking at the proportion of children watching more than three hours during a typical weekend, 26% of BC1 children did this compared with 42% of children in BC2. This represents a notable and statistically significant increase in the amount of television viewing. This increase is evident amongst children from all socio-economic backgrounds. It is hard to draw any firm conclusions about what this increase in screen time might mean for later outcomes. Previous research on activity levels at age 6 using GUS data found that those children who had had higher levels of screen use when they were younger continued to have high levels of screen time at age 6. However, the impact on high levels of screen time on physical activity is less clear. Although children with high levels of screen time at age 6 were less likely to meet the recommended targets for 60 minutes a day of moderate to vigorous physical activity age 6, the majority of these children (83%) still met the target (Parkes et al., 2012).

Table 5.11 Time child spent watching television during a typical weekend by cohort

	BC1	BC2
None	4	6
Up to 1 hour	20	6
Between 1 and 2 hours	26	20
Between 2 and 3 hours	24	26
Over 3 hours	26	42
Unweighted bases	4021	4732

Tested on category 'Over 3 hours': differences by cohort p <.001

There is an association between the time a child spends watching television and household income (Table 5.12). Generally, children from lower income households were more likely to watch television for more than three hours over the weekend. For example, in BC1 around 30% of children in the lowest two income quintiles watched television for more than three hours compared with 21% of children from households in the highest income quintile. The pattern is slightly different in BC2 where household income was less related to television viewing. Whilst children in the lowest income group were more likely, and those in the highest income group were less likely, than those in any other group to have watched television for more than three hours, the children in the middle three groups were broadly similar. Whilst the pattern appears different, the social gradient remains and there was no statistically significant change in the nature of the relationship between income and television viewing between the cohorts.

Table 5.12 Time child spent watching television during a typical weekend by equivalised household income (quintile) and cohort

	Lowest quintile	2nd quintile	3rd quintile	4th quintile	Highest quintile
	%	%	%	%	%
BC1
None	6	4	4	3	3
Up to 1 hour	18	20	20	19	21
Between 1 and 2 hours	23	25	26	29	29
Between 2 and 3 hours	24	21	23	25	25
Over 3 hours	29	30	28	23	21
BC2
None	7	7	5	5	5
Up to 1 hour	6	7	6	7	6
Between 1 and 2 hours	16	21	21	21	23
Between 2 and 3 hours	23	26	25	26	29
Over 3 hours	48	39	42	40	37
Unweighted bases - BC1	738	769	731	835	698
Unweighted bases - BC2	915	726	776	730	975

Tested on category 'Over 3 hours': differences by equivalised household income p< .001; differences by cohort p< .001; cohort* equivalised household income p=NS.

There was also an association between parental level of education and a child's television viewing over a typical weekend (Table 5.13). Children from households with more highly educated parents were less likely to watch television for over three hours than children whose parents had lower qualifications. In BC1, this relationship was linear: as the level of education increased the proportion of children watching television for longer hours decreased. 34% of children whose parents had no qualifications watched television for more than three hours compared with 23% of children whose parents were educated to degree level. Whilst in BC2 children with degree educated parents were still the group least likely to have watched television for three or more hours during a typical weekend, the broader linear pattern by level of education seen in BC1 is not as apparent. Nevertheless, there has been no statistically significant change in the nature of the relationship between parental level of education and television viewing over a typical weekend by cohort.

Table 5.13 Time child spent watching television during a typical weekend by parental level of education and cohort

	No qualifications	Lower Standard Grades or VQs or Other	Upper level SGs or Intermediate VQs	Higher grades and upper level VQs	Degree level academic and vocational qualifications
	%	%	%	%	%
BC1
None	4	4	5	4	4
Up to 1 hour	17	24	18	20	20
Between 1 and 2 hours	21	14	26	27	29
Between 2 and 3 hours	24	28	24	23	24
Over 3 hours	34	30	28	26	23
BC2
None	7	7	8	8	5
Up to 1 hour	6	5	5	7	6
Between 1 and 2 hours	16	15	16	19	23
Between 2 and 3 hours	24	26	21	25	28
Over 3 hours	46	47	50	41	38
Unweighted bases - BC1	185	200	757	1337	1536
Unweighted bases - BC2	155	200	666	1358	2185

Tested on category 'Over 3 hours': differences by education p< .001; differences by cohort p< .001; cohort* education p=NS.

Table 5.14 shows differences in duration of weekend television watching by maternal age. There was no statistically significant association between the mother's age at the birth of the child and the amount of television a child watched over the weekend in either cohort.

Table 5.14 Time child spent watching television during a typical weekend by maternal age at child's birth and cohort

	Under 20 years old	20 to 29 years old	30 to 39 years old	40 or older
	%	%	%	%
BC1
None	6	5	3	6
Up to 1 hour	15	20	20	20
Between 1 and 2 hours	24	24	28	22
Between 2 and 3 hours	31	22	24	22
Over 3 hours	25	28	25	30
BC2
None	5	8	5	4
Up to 1 hour	6	6	7	3
Between 1 and 2 hours	16	19	20	25
Between 2 and 3 hours	23	26	25	27
Over 3 hours	50	40	43	41
Unweighted bases - BC1	209	1495	2148	147
Unweighted bases - BC2	199	1849	2450	211

Tested on category 'Over 3 hours': differences by age NS; differences by cohort p< .001; cohort* education p=<01

Table 5.15 shows there was some association between area deprivation and the amount of time a child watched television. Children from the most deprived areas tended to watch more television than children from the least deprived areas. For example, in BC2, 43-48% of children living in areas in the three most deprived quintiles watched three or more hours of television in a typical weekend compared with around 35% of children living in the two least deprived quintiles. These differences across areas were statistically significant. The general increase between cohorts in the proportion of children watching three or more hours of television is evident amongst children from all areas, regardless of deprivation. As such, there was no statistically significant change in the nature or strength of the relationship between area deprivation and television viewing between the cohorts.

Table 5.15 Time child spent watching television during a typical weekend by area deprivation (quintiles) and cohort

	1 Most deprived	2	3	4	5 Least deprived
	%	%	%	%	%
BC1
None	5	5	4	4	4
Up to 1 hour	20	21	19	19	20
Between 1 and 2 hours	22	24	25	28	31
Between 2 and 3 hours	23	23	22	25	26
Over 3 hours	30	26	30	24	20
BC2
None	7	7	5	6	7
Up to 1 hour	5	6	7	8	7
Between 1 and 2 hours	17	17	19	22	24
Between 2 and 3 hours	22	26	26	28	26
Over 3 hours	48	44	43	35	36
Unweighted bases - BC1	793	667	831	848	882
Unweighted bases - BC2	886	881	973	993	969

Tested on category 'Over 3 hours': differences by area deprivation p< .001; differences by cohort p< .001; cohort* area deprivation p=NS.