Combining telephone and web interviewing

In conjoint analysis, market research, market research fieldwork on January 30, 2013 by sdobney

We’ve become so used to web-based interviewing that it’s easy to forget that just a few years ago telephone interviewing was the major means of collecting data for market research surveys. The switch to web has been so rapid that many telephone fieldwork companies are now struggling. However, we’ve also seen situations where response rates to, say, email invites from a companies own lists have dropped dramatically. For some business-to-business research telephone has come back into fashion, but this time we can use combined phone and web (web-assisted telephone interviewing).

The great advantage of telephone interviewing over email or web-recruitment is that you have an interviewer working to make contact and pursuade the respondent to take part in the research, adding a legitimacy but also a friendly voice. Web on the other hand is quick and cheap and since we are showing people items on screen it is much more useful where we need to show material (eg testing advertising), or ask people to make choices between different options as is done in conjoint or discrete choice research. Unfortunately the speed and low-cost has led to situations where email lists have been overused and now respondents no longer bother to take part and response rates are as low as 1-2% or less.

What has happened is that companies have found they need to go back to telephone interviewing, but that they like the prompting and showing elements of the web-based surveys. Consequently we’ve seen a number of companies using our web-assisted telephone interviewing. In WATI, the survey can be set up so that the interviewer can be responsible for completing the questionnaire and the respondent tracks them on screen, or the respondent completes the questionnaire and the interviewer tracks on screen. More commonly it’s the interviewer as it enables the interviewer to see questions and prompts that can be hidden from the respondent allowing for unbiased or unprompted measurement.

Some simple systems use single HTML pages and perhaps involve sending an email with a link. Our system allows the interviewer and respondent to see the same questionnaire at the same time, albeit with different views. So for choice-based tasks where each individual has a different computer-generated questionnaire, the survey on screen is completely specific to the respondent in question. The process of access is via a simple URL that can be read out and a passcode to gain access.

The advantage for the respondent is the ability to read and to see the materials that the interviewer wants to show at the time the interviewer wants to show them (the survey tracks where the interviewer is). So detailed materials later in the questionnaire can only be seen when the interviewer reaches those parts of the survey. It also allows images and more complex items such as video or audio to be shown – for instance to test for brand and advertising awareness, where showing the advert under test must come after checking for brand measurements.

For pan-European projects it even allows us to have the pages in multiple languages so the respondent can choose to view the questionnaire in say French, while conducting the interview in English.

The hidden extra benefit is that like standard web-surveys, the data captured can be monitored in realtime, rather than running on a separate in-house CATI system. Surveys can also be run using interviewers working in a number of different locations – for instance if you have native language speakers in different countries.

We’re also able to use the same principles of one survey with two (or more views) for telephone-based group discussions or depth interviews. Again, making the telephone interview much more like a traditional face-to-face experience in terms of prompt material without the disadvantages of typing for online qual.

It is also possible to use the system to allow a telephone interviewer to follow up responses collected via the web, or to allow different respondents to see each others answers and so enable consensus testing, or delphi-type interviewing.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: