High levels of electronic media use during childhood have detrimental effects on child and adult  health, including overweight [2, 3]. Despite this, children's use of such media appears to have increased over recent decades . Therefore, several countries now recommend a maximum of two hours per day electronic media use in their children's physical activity guidelines [5, 6]. Television is the form of electronic media used most by children .
Children's television viewing patterns have been well described in cross-sectional studies. Girls and boys appear to watch similar amounts of television , and older children spend more time in front of the television than their younger peers [8–12]. A weak positive association between television viewing and adiposity in children is commonly reported [4, 13].
Patterns of other electronic media use in children have received less attention, perhaps because children spend less time on these activities [9, 14, 15]. Boys tend to spend more time than girls using electronic games and computers [9, 16, 17], and usage appears to increase with age in both genders [9, 17]. Electronic game use [14, 16] and total sedentary time [15, 17] have also been reported to be associated with adiposity in children.
To date, most investigations of electronic media use patterns in elementary school children have come from cross-sectional studies. Longitudinal descriptions of electronic media use and changes in usage over time in individual children have been scarce. The available data suggests that television viewing tracks quite strongly within the middle childhood period (correlations range from r = 0.46 to r = 0.73) [18–20], and higher television viewing times at one point in childhood appear to predict higher viewing later in childhood [8, 21]. Electronic game use appears to track less strongly than television viewing time .
Longitudinal associations between electronic media use and adiposity in childhood have been investigated in different ways in different studies, making comparisons difficult. Unlike the consistent weak positive associations reported in cross-sectional studies for all forms of electronic media use , findings are more equivocal as to whether all forms of electronic media use predict increasing adiposity. Positive associations have been reported between television viewing and later adiposity [18, 21–26] or increased adiposity [18, 22, 23] in most, but not all  studies. Conversely, electronic games have generally been found not to be associated with adiposity longitudinally [18, 27], though some studies have reported an association for girls  or boys only . Total screen time may have a positive [20, 28] or no association with adiposity in childhood . Despite these conflicting findings, the larger-scale and more recent studies generally suggest a positive association between electronic media use and adiposity .
This paper aims to (1) describe longitudinal patterns of television viewing and electronic game/computer use in a large population-ascertained sample of Australian children as they move from the early to late elementary school years and (2) investigate relationships between television and electronic game/computer use and BMI trajectories.