Statistics Canada
www.statcan.gc.ca
Common menu bar links
An overview of Internet users and their time use patterns
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Past research has noted that when comparing Internet users with non-users, it is important to capture the extent of Internet use since decisions about how to allocate the remainder of one’s time are likely to be influenced by the amount of time spent online (Nie and Hillygus 2002). For the purpose of comparison, respondents were grouped according to time spent on the Internet in the following manner: those who did not report using the Internet for any one episode of at least five minutes during the diary day (‘non-users’); those who spent between five minutes and one hour on personal use of the Internet (i.e., email, browsing the Net for recreational purposes, chat groups, and other Internet communication) (‘moderate users’); and those who spent more than one hour on personal use of the Internet (‘heavy users’). This approach offers a preliminary attempt to gauge the extent to which duration of time online affects other activities. Future work might refine such a comparison by adding more detailed categories or perhaps using a continuous measure of time spent online1.
Table 1
Summary characteristics of Internet users and non-users during the diary day
In terms of their demographic characteristics, heavy Internet users differed considerably from moderate Internet users and non-users2. Heavy users tended to be younger than both moderate and non-users (Table 1). There were also greater proportions of students and unemployed individuals in the heavy user group. Further, while both non-users and moderate users were represented relatively equally among males and females, a much higher proportion of males could be found in the heavy user group.
These basic characteristics of the different groups are important since they help to explain differences in the general time use patterns observed for Internet users and non-users. Figure 1 shows that the principal difference between the three groups is that heavy Internet users spent considerably less time working during the diary day than both moderate and non-users3. They also spent less time on domestic work than the other groups, and somewhat less time on personal care and entertainment, compared with non-users. In fact, heavy users spent just over 10% of their total time (or on average about two-and-a-half hours) during the day on the Internet for personal use, a figure that was exceeded only by time spent sleeping (about 35% of total time), working or studying (15%), and consuming television and other media (11%). This figure is in addition to any time spent on the Internet for work or education-related purposes. Moderate Internet users spent the most time working and just under 3% of their total time (or about 40 minutes on average) using the Internet for personal use. Those who did not use the Internet during the diary day were older, and therefore a greater proportion of them were retired. Despite this fact, collectively they spent a greater amount of time working, both for paid employment and domestic work, than heavy Internet users. These findings reflect the positive association observed between amount of Internet use and the amount of discretionary or ‘free’ time held by respondents.
The magnitude of the gap in time spent working between heavy Internet users on the one hand, and moderate and non-users on the other potentially clouds the ability to discern less obvious relationships between Internet use and other uses of time. This is simply due to the fact that time spent at work may have a sizeable effect on other time use decisions. For this reason, the analysis includes an adjustment for time spent at work, so that comparisons can be made between users and non-users while assuming equal work time. Unadjusted results are also presented, to give readers an appreciation of the extent to which work time might influence other decisions (see Text Box 1 for more details).
Figure 1
Principal differences in time allocation during the diary day by type of Internet user
Text Box 1:
Analytical approach
Both simple bivariate and multivariate methods are used in the analyses for a number of reasons. Bivariate analyses offer effective means for exploring and identifying major patterns in how people spend their time in a manner that is relatively simple to convey. Past research has noted, however, that any examination of how people allocate their time must not ignore the array of social and economic factors which shape peoples’ lifestyles and form the context around which decisions are made (see, for example, Nie and Hillygus 2002, de Haan and Huysmans 2002, Pronovost 2002, Anderson and Tracey 2001, Nie 2001, Franzen 2000). Multivariate methods are therefore essential for examining differences in how people spend their time while also accounting for other factors that may influence overall behaviour and decisions, such as age, sex, labour force status, education, and the presence or absence of children in the household.
The specific methods used include multiple regression and multiple classification analysis (MCA) procedures, which generate estimates of the time that individuals in different groups devote to specific activities, both before and after adjustment for relevant control variables (specified at the bottom of each table). The adjusted and unadjusted figures are presented so that the effects of the adjustments are transparent to readers. Both of these methods can be used to obtain the desired results and the choice to include MCA is more a matter of the researcher’s preference regarding input, output and presentation (for more on MCA, see Andrews, Morgan, Sonquist and Klem 1973). Readers should note that the reported differences between the adjusted and unadjusted figures are subject to rounding.
It should also be noted that time use results presented in this study do not provide information about the causality of relationships between Internet use and time spent on other activities. Rather, the results reveal the existence of associations between time spent on the Internet and other activities, as well as the strength of these associations.
The next sections examine how Internet use fits into other daily activities, including both non-discretionary activities such as work, as well as those respondents choose to perform during their free time.
1. The classification of Internet users was also influenced by a few additional factors. A categorical classification was selected in order to assist with presentation and interpretation of findings. In order to preserve data quality, the ‘moderate’ and ‘heavy’ use categories were not broken into additional groups due to issues of sample size. Readers should note that although the ‘moderate’ use category spans a range of five minutes to one hour, the majority (79%) of moderate users reported spending at least 30 minutes online. The mean Internet time for moderate users was 40 minutes, while heavy users averaged over 150 minutes online (the median value was 120 minutes, but some extreme high values resulted in the inflated figure for the mean). Heavy users accounted for 43% of the Internet users in this study.
2. Readers should be cautioned that in this study, data on Internet users reflect only those individuals using the Internet during the 24-hour period captured by the time diary for an episode of at least five minutes. See the Note to readers for further information.
3. When reading figures, readers should note that only about two-thirds (65.5%) of respondents reported that their main activity in the week leading up to the survey was work or school. This, together with the fact that collection of the diary data included weekends, should be considered when interpreting values reported for time spent on work and educational activities.
