![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
Abstract
This article reports on an exploratory study aimed to identify which ways of marking required and optional fields help older people fill in web forms correctly. Drawing on a pilot study and selective attention research in ageing, modified versions of widely used forms were created, in which standard asterisks were replaced with one of three alternatives: large asterisks (double standard size), textual labels and a simple binary classification of fields. Older novice computer users participated in this study (N = 88). The quantitative results were independent of the type of form and showed that the method of denoting required fields had a significant effect on the number of errors (measured as required fields not being filled (p < 0.05)). The participants made fewer errors in the binary classification than in the others (p < 0.05), while the differences between standard asterisks, large asterisks and labels were not significant (p > 0.05). The qualitative findings revealed that the participants felt overloaded with information when using every method except the binary classification. These results challenge widespread design solutions for older people (e.g. enlarging elements) and suggest that reducing selective attention demands can help us meet their accessibility needs.
Acknowledgements
The authors thank Àgora, the Casal dels Avis de l'Hospitalet and our participants for their support and collaboration; colleagues E. Arroyo and D. Hernández-Leo for their comments on earlier versions of this manuscript and A. Evans for improving our English and the reviewers for their stimulating comments.
Notes
1. Although the surname field tends to be required, we made it optional because older people showed concerns about privacy in the training session.
2. Using keywords such as web forms; online forms; older people; older adults; elderly, only did we find this reference (and others by the same author) in the main HCI journals and Conference Proceedings of the last few years.
3. Guidelines have helped designers to cope with diversity by spreading awareness and best practices (Nicolle and Abascal 2001). This is especially important if we consider that older people differ from the type of users that tend to be addressed in HCI (Dickinson et al. 2007).
5. Fifty owned their mobile phones. None of them were targeted at older people.
6. We have instructed almost 400 older people in a different computing courses in Àgora, see in Appendix 2.
7. The computer for the trainer was used by a participant.
8. The modified version of the Vueling form used an icon different from the standard asterisks, which was associated with the text labels at the top of the form, because the textual labels used in Yahoo! and Hotmail modified forms were too long to be included in the Vueling form.
9. Each participant was assigned with a number between 1 (Group A) and 12 (Group L). This number was generated by using MS Excel, function RANDBETWEEN(1;12). If one participant was assigned with a group that was full (10 or 12 participants), we generated another random number for that participant. This process was repeated until the participant was assigned to a free group.
10. Such as older people not willing to do the same test another day or difficulties because of conflicts with personal agendas (e.g. travelling, appointments with their GP, taking care of their grandchildren).
11. The content of this message is the same as in the error messages in the original forms.
12. The platform recorded the number of errors made by the participants until the form was filled in correctly. If more than one attempt was needed, the software added up the errors made in the iterations.
13. There was a break of 10 min when participants had to fill in two forms.
14. Another option would have been to conduct a MANOVA, but we only had one dependent variable. ANCOVA could have been used. However, the participants were distributed randomly.