I've spent the past year or so investigating social networking sites and older adults, and it seems to be a very tricky and open topic to come up with any firm answers. The idea, or question, behind my thesis is relatively simple; Is it possible to increase the number of older adults on social networking sites by categorizing and analyzing opinions and barriers discovered in research? While this doesn't sound like much of a challenge, difficulties are frequently becoming apparent, from convincing people that this research is even worthwhile, to making assumptions based on a relatively small pool of participants.
People commonly ask "Why?" when I explain a bit about my research. They ask why it matters if less people over 65 use social networking sites, as long as the people themselves are happy. It's a valid argument, one that I've had to deal with so often I've committed my answer to memory. I try and emphasize the following: the research is not about forcing this technology on people who do not want to use it, it is about asking why they don't want to use it and then looking into whether or not the mainstream sites could adapt to avoid some of the common issues that discourage people. It is about trying to improve these sites to make them inclusive to wider ranges of people, and in order to do that we need to investigate exactly what these issues are and how much of a problem each of them represent.
This leads to another problem, however. When dealing with this sort of topic, which focuses on opinions, it is very difficult to be fully confident in the data you receive. In a research project which relies on the evaluation of a system, or a comparison of two experiments, the data is relatively easy to defend. You can use existing System Usability Scales, or take data straight from the study, and unless you have done something wrong then that data is robust enough to analyze. Dealing with opinions is an entirely different matter. There are ways that you can gather qualitative feedback and then use that to obtain quantitative data (which I'll mention later) however it always relies on the small group of people you initially had for the focus group or interviews.
Recently we held two focus groups with small groups of adults over 60. These participants all lived around the Angus area of Scotland, and all of them had agreed to be part of the SiDE user pool, a project which I am part of which investigates social inclusion in a digital context. These two factors both indicate bias, firstly the location of all participants and secondly the fact that all of them at least had some interest or motivational factor to agree to participate in a social inclusion research project to do with technology. We can't assume that all older adults in the UK feel the same way, neither can we even assume that all older adults in Angus do. This seems to be one of the difficulties in this type of research; while focus groups can be useful to get some ideas as a starting point they do not tell you any more than what they actually are - six to ten people who may all agree on a specific point.
To try and move forward with this, we're planning to try and turn our focus group feedback into qualitative data, by trying to use the main issues discussed to construct a questionnaire that we can send to the user pool. This questionnaire is not going to be perfect, as the participants probably haven't covered every major barrier preventing people from using such sites, the feedback from the questionnaire will still be biased due to location and participants and other issues will almost certainly surface, but it's a starting point which will allow us to hopefully gather large amounts of data from the opinions of a few people in a focus group.
I'm hoping that this questionnaire will lead to something. Hopefully when the feedback is analyzed it will map out the key areas to follow up, and eventually something can be built which can be evaluated in a more traditional sense, but it is unfortunate that there has been so little research in this field to map out the foundations. The topic itself feels almost sociological, rather than anything computer based, but that seems to be another challenge for me when working in this opinion-based and relatively new field. I'm certain, though, that this is worth investigating, and it'll give fascinating insights into older adults and sites with social functionality, which are increasingly becoming the norm. As some researchers are commenting, the web is being rebuilt around people. It is, however, our responsibility to ensure that people don't get left behind.