OxIS is the only ongoing survey of Internet use in Britain; there is nothing with the scope and continuity of the OxIS data. It describes how Internet use has evolved from the first survey in 2003 to the present day. It includes information on use of the Internet, attitudes toward the Internet and technology, demographic information, and geographic information. It surveys the entire British population including users, non-users, and ex-users.
It can be used to track rising use of the Internet, to study reasons for non-use, and reasons to cease using the Internet, among many other questions. There is no other survey like it in Britain.
Why is OxIS a Face-to-Face Survey?
In a face-to-face survey, an interviewer talks to each respondent in their home. This generally improves the quality of the data because it increases the proportion of people who complete the survey (the ‘response rate’). Since it is a comprehensive questionnaire, OxIS is very long. A face-to-face interview improves the likelihood of getting high-quality answers to the very last question in the survey.
Surveys that use only the Internet are easy (and cheaper) because they can collect their data so readily: but easy and inexpensive data collection is balanced by several serious problems. The most important is that it is impossible to assemble a list of all the people who have Internet access in Britain. Initial contact would have to be by email and many people have multiple email addresses, email addresses are often private, and there is no authoritative list of email addresses: this makes it impossible to create a random sample. In technical survey terms the problem is that there is no way to construct a ‘sampling frame’; the list from which respondents are randomly chosen for the sample.
A further problem is that by restricting sampling to the Internet we could not include people who have stopped using the Internet or who have never used the Internet. Without these respondents we would not be able to compare users and non-users. In Internet surveys, we never know who actually answered the survey (or even if it was only one person). People vary widely in their computer skills but almost all adults can talk to an interviewer.
How Representative of Britain is OxIS Data?
The key to representative data is properly selecting the sample. Respondents for OxIS are selected randomly by a multi-stage process. Random selection ensures that every adult in Britain had an equal probability of being selected for an interview. This ‘equal probability’ is important because it ensures that the 1,818 people who responded to the 2019 survey are a representative sample of the entire 67 million adults in Britain. Since an identical sampling process has been used for each survey the results are comparable since 2003. For a more detailed description of the sample selection process see the Methodology section of the report of the 2019 survey.
How to Request OxIS Datasets
The data are checked and analyzed by researchers at the Oxford Internet Institute (University of Oxford). Sponsors of OxIS can use the data as soon as they are clean. Two years after they are collected the data are released to outside academic researchers and other government users. For-profit and not-for-profit users may be able to use the data for a fee. Contact us at: oxis@oii.ox.ac.uk for details.
Keep in Touch
To be notified of OxIS news and events, please sign up to the OxIS mailing list.
A panel at the OxIS 2019 launch: Vicki Nash, Katie O’Donovan, Ian Macrae, Liz Williams (right to left)
