Join Our Newsletter





Events Calendar

« < October 2026 > »
S M T W T F S
27 28 29 30 1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31
Home arrow Library of Research Articles arrow Online Research arrow The Trouble With Web Surveys
The Trouble With Web Surveys PDF Print E-mail
Written by Brian Cooper   
20 Oct 2005


Over the last few years there has been an explosion in “build it yourself” on-line web surveys. Whilst in some cases, these online solutions can be good and provide excellent value for money, they have been designed and built by software and web developers. The beauty of most key market research software, be it CATI, online or tabulation packages, is they have been designed for researchers by researchers. How can a web survey portal advise you on the positioning of a question or even the legal validity of your ethnic origin question? Trust your online survey to agencies with whom you would trust your face to face or telephone project.

A particular problem in these days of cost cutting and lower research budgets is to presume, often incorrectly, that an online survey is vastly cheaper than any other methodologies. Of course, this is often the case, but there are some variables to consider.

??Sample list – if you do not have a contact list and are attempting to fill certain quotas or interview quite specific people, then costs for sample can be substantially higher than telephone costs.

??Length of survey – it is easier to keep respondents on the phone when conducting a long survey than an online version- keep your online surveys to less than 10 minutes if possible.

??Development time – a survey that is highly designed and fairly complex can easily take twice as long to program as a CATI script. CATI software has no need to look good as it is the interviewers who see the screen, not the respondent and the tendency, as mentioned above, to “over design” can cause delays in putting a survey live and increase costs.

A point to always remember when conducting research – cheaper does not always mean better. Use the right methodology first and foremost.

Conclusion

As online survey providers and web survey software developers, we have a keen interest in seeing online surveying grow but we also have a deeper understanding of data capture and research methodology in general. We believe that a “one size fits all” scenario can not work in today’s mature research market. Different methods will always be required in order to get the best information for our clients, and even though online surveys can offer fantastic value for money, the proof will always be in the accuracy and quality of the results.

Beware of over complicated, over designed surveys – keep it simple and let the questions speak for themselves.

Avoid using non-research based sample companies – build your own if budgets permit or use one of the well known large agency panels. The costs may be higher but your data will be more robust.

Software solutions – DIY online surveys may be cheaper but there is still no substitute for a properly designed survey from a MRS agency.

Brian Cooper – Managing Director
Omnisis Ltd

172 Kerscott Road
Manchester
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it
0161 969 1445
Developers of Hyperweb, an online survey software package, and providers of online fieldwork


Last Updated ( 20 Oct 2005 )
 
< Prev   Next >

Polls

How important is market research to start-ups in the current economic climate?
 

RSS Feeds

Subscribe Now