Forms design

In 1988, Carel Jansen and I conducted a study with 200 participants who completed one or two forms each, while thinking aloud. These forms were issued by the city of Hengelo, the Dutch Income Tax authorities, the Institution for the national students grants, and the Ministry of Education in the Netherlands. The study was financially supported by the Government Information Service.
What problems do people have with forms? Several types of problems emerged from the study.
  • Orientation problems. Form-fillers have a lot of difficulties to keep the thread and to work systematically. Rather they work step-by-step and question-by-question.
  • Selection problems. It is often unclear which questions have to be answered and which can be skipped.
  • Problems with the explanations. Form-fillers do often not see that they should refer to the explanations. It is difficult to find the right information in the explanations, and therefore form-fillers read much information that is not relevant to their individual situation, and consequently they are confronted with all kind of mysteries.
  • Switching problems. People who fill-in forms have many problems to goi back and forth between the form, the explanations, and the other documents they have to use, such as their annual income statement or their rental agreement.
  • Problems with counting. It is often requested to make calculations on a form, and people who fill in forms make many mistakes in doing that. Even simple additions cause problems, not to mention calculations of percentages or averages.
  • Problems with verifying. After having completed a form, people seldom verify whether they gave the right answers. Calculations were not checked. When figures had to be taken over from other documents, most form-fillers refrained from making sure whether they did it correctly.
  • Problems with prior knowledge. Governments forms suppose erroneously that form-fillers are knowledgeable about the regulation, understand the legal terms, understand why particular information is requested. In practice, many people do not even understand the purpose of the form, nor are they aware of what is going to happen with the information they provide.
In sum, the behaviors of form-fillers can be characterized as a kick-and-rush strategy. They go straight to the point, they read as little text as possible and they give the answer that comes across first - and that is not always the right answer. Such behavior is perfectly understandable, since completing a form is a awkward job - a necessary evil, just like reading user instructions if you want to get down to work quickly with a new product. As a consequence of all these problems, the participants in the study made so many mistakes that approx. 87% of the completed forms would have resulted in an incorrect decision if they were processed by civil servants just like that. In reality, this figure is probably lower because many people get help from others, and because the experimental setting was a bit unusual. However, a follow-ups study by Jansen et al. (1991) showed that the types of mistakes found in the 1988 study, do occur also under more realistic conditions.
Better forms design Based upon our observations, we formlated three basic principes that should improve the usablility of government forms.
  • The action perspective principle. This means that the starting point in the design of a form is not the law or the regulation the forms refers to, but the personal and financial situation of the person who fills in the form.
  • The strict control principle. This means that the form strictly and in full detains prescribes what the user has to do.
  • The well-balanced offering of explanations and background information. All and only relevant information has to me added at precisely the place where the user needs it.
These principles lead to a large number of specific guidelines which we presented in an advisory textbook Formulierenwijzer (1989). The guidelines apply to design issues such as
  • The wording of questions on forms
  • The sequential order of the questions
  • Content and format of explanations and background information
  • Language and style
  • Procedures for usability testing of forms
The guidelines were applied to redesign nine of the forms that ware tested in our previous research. The result was promising. All-in-all the percentage of incorrect completed form decrease with 75%. However, that number was still alarming high. Apparently, a good design of government forms is not the complete solution for forms that refer to extremely complex laws and regulations.
Digital forms As early as 1981 we experimented with the idea that interactive software could be a useful tool to communicate complex regulations to citizens. Together with our research on paper forms, we experimented with electronic forms. This innovative approach got a strong impetus with the launching of the electronic Income Tax Declaration in the Netherlands, in 1995. Of course, the introduction of electronic forms is primarily motivated by the government organisations, to increase the efficiency of procedures. But electronic forms offer also great perspectives to overcome the usability problems we signalled in our earlier research.
    Selection problems – the software can use the answers on questions to select the follow-up question in a way that user only see relevant questions. The selection decisions move from the user to the software.
  • Problems with explanations and background information – software can offer context-sensitive help. Moreover, the help can be much more extensive as with paper forms. Indexes and hypertext help the user to find the relevant information.
  • Calculation problems are completely solved by the software, since the users do not have to calculate outcomes themselves.
  • Problems with verifying – Apart from the fact that checking calculation is no longer necessary, software can also signal missing, illogical and improbable answers (such as birth dates before 1900)
Of course, not all problems can be solved by the software. Users have still to insert their personal and financial data. Orientation problems can even be increased, since the user sees only a small part of the form in a time. Inexperienced computer users may ne confronted with many extra ‘computer problems’ including typing errors, navigation, etcetera. A Master Thesis project by Ivo d’Haens, sponsored by the Society for Technical Commuication (STC) was set up in 1999 to compare the problems of electronic and paper forms. Thirty citizens completed a Income Tax Declaration Form, 15 on paper and 15 electronic, while thinking aloud. The study showed that the alleged advantages of electronic forms were less obvious than we expected. The electronic form did nor reduce the number of errors, neither the time that the participants needed to complete it. The participant did not value the electronic form as easier or more motivating than the paper version. The most obvious advantage of the electronic form was the automatic calculation of totals and subtotals, as a result of which the number of calculation errors was reduced to zero. On the other hand, the program did not make it easier to select relevant questions, since the users had to indicate which categories apply to them and which not. This requires a more thorough understanding of the regulations than most participants in our study had. The electronic form did not lead to an increase in the use of explanations and background information. Context-specific help was not always useful since the participants had often questions beyond the specific questions they were answering. If they activated the online help, it covered the form, so that they could not easily switch between the form and the help information. And finally, the participants did often not use the scrollbar, so that they only read the information that was visible at a certain moment. Of course, our conclusions have to b taken with caution. The number of participants was low, the Income Tax Form is only one from a increasing collection of electronic forms, and the skills of the average computer user have been increased considerably since 1999. However, the results should make us cautious with too optimistic expectations of new developments in the design of electronic forms.
New developments in electronic services Electronic administrative services are growing rapidly. Travelers to the US, for instance, have to do with electronic bookings for their flights, Internet check-in, Electronic Visa applications and many other administrative procedures that run via the internet. This raises many issues that are at least interesting from a communication science perspective:
  • Issues of usability like the ones we investigated with electronic forms
  • Issues of acceptance: will citizens accept such advances in administrative communication? How to deal with privacy and individual control?
  • Issues of social (in)equality: there is an increasing gap between those who have access these new facilities and those who have not. Such questions are currently addressed in the research program on Electronic Services of the departments of Communication, Media and Organizations (CMO) and Technical and Professional Communication (TPC) of the University of Twente, in cooperation with meny other institutions such as the Duct Tax Service, the Telematics Institute and other.
    Publications in English about this research Steehouder, M. & Jansen, C. (1992). Optimizing the quality of forms In: H. Pander Maat & M.F. Steehouder (eds.). Studies in functional text quality (pp. 159-172). Amsterdam, Atlanta GA: Editions Rodopi. Jansen, C. & Steehouder, M. (1992). Forms as a source of communication problems. Journal of technical writing and communication 22, 179-194. Jansen, C. & Steehouder. M. (1999). How research can lead to better government forms. In D. Janssen & R. Neutelings (Eds). Reading and writing public documents (pp. 11-36). Amsterdam: Benjamins.