FSD Bulletin

Issue 15 (3/2004)

ISSN 1795-5262

Front page
Previous issues
Editorial staff

» latest issue

FSD Bulletin is the electronic newsletter of the Finnish Social Science Data Archive. The Bulletin provides information and news related to the data archive and social science research.


Finnish Social Science Data Archive
Tel: +358 40 190 1432
Fax: +358 3 343 9088
E-mail: fsd@uta.fi

How to Collect Data on the Web

Internet surveys require know-how

Tuomas J. Alaterä 8.2.2005

All Web users are familiar with Internet surveys - short surveys conducted by newspapers, surveys on customer satisfaction, or extensive scientific surveys. Responding to them, one soon learns that while Web techniques open up new opportunities, they also set new limitations. Responding to a badly designed Web survey can be both frustrating and meaningless.

Internet survey design does not differ very much from mail survey design - to a large extent the same rules apply. But there are some differences, and if one chooses to ignore them, it may lead to surveys that leave a lot to be desired, or even to biased responses. It is worthwhile to learn and benefit from the characteristics of the medium.

For three years, the University of Tampere has offered a course on Internet survey design and implementation. The course is part of a project funded by the Ministry of Education, and the idea for the course came from Sami Borg, Director of FSD. The course is meant for social science students, but every year there have been students from other fields of study, which has brought a wider range of views to discussions.

Main focus is on the principles and practicalities of Internet surveys with special emphasis on questionnaire design. Crucial features of online surveys are explored: when to use a Web survey, how to reach the target population, sampling design, functionality of the questionnaire, anonymity, IT issues etc.

Survey process resemblance

One goal is to improve skills needed in working life. Therefore, some sections of the course focus on computer and data collection skills needed at work.

After the introductory section, the course is structured to resemble research process. Students are divided into smaller groups, and each group will conduct their own survey. First, a research problem is chosen and a target population specified. A questionnaire is then planned. When the questionnaire design has been approved by the teachers, it is time to start thinking about the implementation. E-lomake online survey software was used to create the questionnaire Web forms.

In addition to the lecturer in charge of the entire course, there are assistant teachers taking care of HTML and SPSS teaching and group guidance. Teachers monitor the survey process of each group closely, and each stage is submitted for their approval.

Approved questionnaires are published on the Web. Each group must find a way to recruit a sufficient number of respondents for their questionnaire. Groups collect either a real dataset or an exercise dataset using other course participants as respondents. Most groups collect a real dataset with student organisations, municipal service units or enterprise employees as target populations. A few datasets contain more than 300 respondents, some have been used for a master's thesis.

Each group writes a short report on the data they have collected. This requires performing statistical analysis of the dataset. The report does not need to be very extensive, main focus is on technicalities like the how results are presented or how charts should be used.

Completing the course requires quite a lot of work both in and out of classroom, and students get 3 credits for it. Those who persevere with the course have been very motivated to do their assignment well. Because Web survey courses cover a wide variety of study topics and implementation techniques, teachers should have extensive knowledge of the subject and adequate technological competence.

Positive feedback

Student comments have been extremely positive. It seems that the course has succeeded in reaching its goal. Some students felt that there was perhaps not enough time to learn everything. Extra time would certainly be useful, but considering that the course aims to improve working life skills, strict deadlines and a lot of independent work are perhaps beneficial. Many expressed a wish for more courses on Internet surveys. This is something that the university should take into account when planning the curriculum.

The writer is in charge of the course on Internet surveys, and works as Information Network Specialist at FSD.