TAO/TAO Survey Among Elderly - Wave 1

The purpose of this survey was to get insights into the motivations of not using the Internet and to assess how non-usage in this age group affects social inclusion and well-being. This first wave survey provides us baseline data for the assessment of mobilizing opportunities and barriers, e.g. in terms of differences between groups regarding Internet skills, preferences and usage patterns. The TAO onliner survey targeted Internet users at an age of 50 or more years. The geographical scope has covered the countries of the TAO consortium members, i.e. the Netherlands, Germany and Switzerland. The purpose of the survey was to get insights in Internet usage patterns and motivations and to assess how Internet usage in this age group affects social inclusion and well-being. With the TAO offliner survey non-Internet users at an age of 50 or more years were targeted in the same geographical scope.

Methods & data

The online survey was set up via the Web-Application ‘Survey Monkey’. Survey languages were Dutch and German. Due to subtle differences in the wording and differences in the educational systems two German versions were developed, one for Germany and one for the German-speaking part of Switzerland. Several Dutch, German and Swiss Internet- and gerontology-related organisations were asked to distribute the surveys’ URLs to their members and to their wider networks. The offliners paper and pencil survey was distributed via personal relations of the research team. The BUAS team in Switzerland, for example, asked faculty and staff members to distribute the questionnaire to relevant persons. Additionally, elderly residents of care facilities were approached. However, it became obvious that recruitment of elderly offliners becomes a more and more difficult issue due to the widespread Internet use in this age group. The questionnaire was developed by the use of different sources. It was intended to rely on well-known and tested questionnaires and instruments as far as possible (e.g. Mental Health Index 5, the Psychosocial Consequences Scale (PCS), the GVU WWW User Survey).

Limitations

The results of the TAO-survey need to be interpreted keeping several limitations in mind. First, and most important, the sample is a convenience sample which is certainly biased by the channels through which the URLs were distributed. However, the respondents were also recruited without any organisational link. Second, a lot of difficulties were faced while recruiting offliners. The offliners’ sample size is surely not optimal in comparison to the large onliners sample.

Socio-demographics

Onliners

Overall, 2912 persons participated in the online survey by the beginning of March 2012. The share of Swiss and Dutch respondents was approximately 20% each, respectively 561 and 426 respondents, while the share of German respondents makes up 60% with a total of 1925 respondents. Overall, the respondents are on average 68.57 years old and 60% are male. As expected in this age group, the overwhelming majority of the respondents are married (70%). 12% are divorced and 12% are widowed. Over two thirds of the respondents live in a two-person household, roughly one quarter in a single household, and 7% live in a household with more than two persons. The largest share of respondents (40%) has a university degree (including university of applied sciences). Compared to the results of the European Social Survey we see overlap with regard to our results of the questions on gender, marital status and living situation. With regard to education we see that our sample has a higher education.

Offliners

The offliners’ sample size is surely not optimal in comparison to the large onliners sample. The offline respondents tend to include more female respondents, that are older, have a lower educational level, that live more frequently outside big cities and that are more often widowed or divorced than the online respondents of our survey. On the other hand, the offliners show no differences from onliners with regard to the frequency of meeting friends. The offliners even indicate more often than onliners that they meet friends and that they have a person to talk to about personal issues. The offline respondents seem to be less involved in clubs, associations or non-profit organizations (charity activities) compared to online respondents. Something that does stand out is the mental health index, which is slightly lower.

Beginners

Beginning Internet users (less than 5 years of Internet use) seem to have specific characteristics that differentiate them from longer-term Internet users. They differ on many sociodemographic characteristics, social inclusion, mental well-being and Internet use patterns from those who have used the Internet for more than five years. In several sociodemographic regards, beginners are more similar to offliners than to other groups of Internet users.

Online Survey: Well-being and Social Inclusion

The mental health index consists of five items which mainly cover emotional states (e.g., feeling happy, down or nervous). In general, the sum score is positively skewed. Across the countries the mean score seems to be very positive, too. There is only little variation between the three countries. Our analysis reveals that the survey’s respondents are socially well-included and have on average a good mental health.

In order to investigate further what intentions are behind the respondents’ Internet usage they were asked to assess the importance of the Internet with respect to aspects relevant for the well-being and social inclusion. The sum of the scores on a 5 point Likert scale across the 18 items ranges between 18 and 90. Higher scores indicate that more positive psychosocial consequences are experienced. Overall, the ICONS score for all respondents is generally high, and there are only little differences between this overall value and the country-specific ICONS scores (Clark & Frith, 2005), as the score for the Swiss sample is 63.03, the score for the German sample is 62.63, and the score for the Dutch sample is 61.17. In general, our respondents have on average made very positive experiences with using the Internet.

Conclusions

The TAO onliner and offliner survey has revealed a number of interesting insights into Internet usage and its impact on social inclusion and well-being. This first wave survey provides us baseline data for the assessment of mobilizing opportunities and barriers, e.g. in terms of differences between groups regarding Internet skills, preferences and usage patterns. In stylizing words, onliners are about 69 years old, rather male than female (ratio: 60:40), usually live in a two-person-household and have a relatively high educational degree. Offliners are more likely to be female, older, and show lower levels of educational degree. Offliners are also more often divorced or widowed than onliners, but they are not at all isolated, as indicated by their contacts to friends and other social aquaintances.

Another distinction that turned out to be meaningful in our survey is between Internet beginners and long-term Internet users. Regarding many sociodemographic features, social inclusion and mental well-being Internet beginners seem to have more in common with offliners than with experienced onliners. Apparently, their characteristics seem to shift from offliner to onliner attributes.

We could observe different Internet usage patterns between the respondents, though they are difficult to explain by differences with regard to the respondents’ social inclusion or state of well-being. These causal relationships have to be examined and answered in the second wave of the survey.

Overall, our findings suggest that reluctance towards active Web 2.0 usage is widespread among individuals at the age of 50+. However, we also found that those persons within the target group that are active Web 2.0 users seem to form what we call elite users that have a lot of expertise to share and are willing and able to spend above average shares of their time on these activities.

With regard to predictors for Internet and social media use, we identified age, gender and educational degree as the factors having the highest impact on these activities, whereas the impact of social inclusion and well-being factors is inconclusive, most likely depending on overall context conditions of the respondent’s living situation.

Taking these results together, the TAO survey has found a number of aspects that deserve to be considered when the “silver market” is targeted by – commercial or not-for-profit – activities that aim to tap the potential of this group. These aspects provide a number of suggestions for segregating this target group into meaningful subgroups that obviously differ significantly with regard to their expectations and needs and their capacities and preferences to interact and collaborate with others. This is of particular importance because interaction and collaboration, namely in form of ‘co-creation’, is considered to be the most promising way to adapt society and economy to the fundamental changes induced by the demographic shift towards ageing populations.

References

• D5.1 – Survey Among Elderly Beginners: Questionnaire and administration completed and the intermediate report
• Berwick, D.M., Murphy, J.M., Goldman, P.A., Ware, J.E., Barsky, A.J., Weinstein, M.C. (1991). Performance of a five –item mental health screening test. Medical Care, Vol. 29, 169-176.
• Clark, D.J., Frith, K.H. (2005) The development and initial testing of the Internet Consequences Scales (ICONS). Comput Inform Nurs. 2005 Sep-Oct; 23(5):285-91.
• Friedman, B., Heisel, M., Delavan, R. (2005). Validity of the SF-36 five-item mental health index for rmajor depression in functionally impaired, community-dwelling elderly patients.Journal of the American Geriatric Society, Vol. 53, 1978-1985.
• Schlosser, A., Pirolli, P., GVU’s WWW User Survey (http://www.cc.gatech.edu/gvu/user_surveys/survey-1998-10/questions/use.html; accessed 12 July 2012).

Links to other Handbook chapters

Target Groups

(N)Onliner & Offliner

Older Adults and Online Communities

Usability