The following report details the results of the profiling of the Science Museum’s virtual audience. The work forms part of a larger project to gain a detailed understanding of the demographic and psycho-graphic profile of our total audience – both physical and virtual. It is therefore run in conjunction with a survey of actual visitors to the Museum. Both Web and physical audience surveys will be run as longitudinal projects, so that we can build up a picture of fluctuations and changes in the profile over the coming years.
METHODOLOGY
The following report is based on responses to a short pop-up survey, which was posted on the Science Museum homepage on Monday 6 August 2001. Log data were also used to supply additional information on virtual visiting patterns. A cookie was employed to discourage visitors from completing the survey more than once, and to avoid irritating regular visitors. However, a link was provided further down the site so that anyone who had initially closed the survey could go back and fill it in later.
Survey responses were collected in the course of one 24-hour day. The survey was available for the full 24 hours to avoid excluding users accessing the site at night or in different time zones. A total of 252 responses were collected, representing 5.5% of the 4600 visitors to the site during the 24 hours of the survey period. While the pop-up survey dealt specifically with the Science Museum Website, the questions were closely linked to those used to profile the Museum’s physical visitors. This allows us to compare the data and look at both physical and virtual visiting patterns.
The Web survey was developed in close collaboration with the Science Museum’s Web Management Group. The demographic questions within the survey had been extensively piloted through prior use in the physical survey. Note of caution It should be noted that the survey was up on the Website for one day during the summer holidays. It therefore represents a snapshot of visitors at a particular time of year, rather than a conclusive summary of annual visiting patterns. The survey will be repeated during the autumn term, and at intervals throughout the year to explore seasonal fluctuations in visiting patterns.
The return rate of 5.5% represents a relatively high response rate for web surveys where return rates of 2% – 4% are standard. However, it should be noted that the responses of the sample taken may not be representative of the total population of Web visitors. Please note that the physical audience profiling data has only been collected for 6 months. Therefore figures quoted for the physical visitor profile below may alter once a full year’s data has been included. The data given for physical visitors excludes school groups to allow the best comparison for the Web survey, which was taken during school holidays.
The following graph shows fluctuations in patterns of Website visiting over the past two years. It is difficult to interpret the individual peaks and troughs in the graph due to changes in collection methods. However, the overall trend is an increase in visitor numbers to the site. Note on search phrases The log data also provides data on search phrases used to access the site from external search engines.
This data provides interesting information on the sorts of topics users are interested in when they enter the site. An initial review of the data showed that a very wide range of search phrases are being used. A high proportion of these comprised combinations of the words ‘science’ and ‘museum’. A detailed analysis of the search phrases would give an idea of the most common topics under which visitors enter the site. However, because of the amount and range of data, this analysis would need to be done separately to this report. Acknowledgements Thank you to Mike Ellis and Daniel Evans for their help with this project.
