Monday, July 18, 2011

Recall Bias in Survey Design

I'm working on a survey which intends to measure whether a person's participation in a fellowship increased their research productivity (i.e. number of publications, new technologies developed and patented). At baseline the person is asked to report about their publications in the two years preceding the measurement. After two years of participation in the programme, the person is asked to reflect on their publications record since the start of the programme.

Besides the fact that publications usually have a long lead time, a recall bias may also be at play. The European Health Risk Monitoring Project explains response shift bias as follow:

Recent happenings are easier to remember but when a person is asked to recall events from the past, accuracy of the recall gets worse while time span expands. Long recall periods may have a telescoping effect on the responses. This means that events further in past are telescoped into the time frame of the question.

In my example, if the question asks if a person published a journal article in the past 2 years, the respondent might place the journal article which was published 2.5 years ago into the time frame of 2 years. Those people who do not publish regularly might be better able to provide accurate information. Those who publish frequently could potentially check their facts, but they are unlikely to do so if the survey is not seen as sufficiently important.

The EHRM recommends the following strategies for trying to address this type of bias: 
The process of recall of events from the past can be helped by questionnaire design and process of interview. The order of questions in the questionnaire can help respondents to recall events from the past. Also giving some landmarks (holidays, known festivals etc.) can help to remember when some events happened.  Also, use of a calendar may help a respondent to set events into the correct time frame.

2 comments:

Paul Sparks said...

I'm the field officer in charge of evaluation for little an NGO working in Kenya and Uganda.

I feel your pain about recall bias. I've been dragged into using a food security tool that asks a series of 9 questions all with a 1 year recall period. It's a standardized questionnaire so I can't even mess with the design!

Anyway, great blog! I'll return.

Best Regards

Benita Williams said...

Hi Paul. Thanks for dropping in. I hope your tool doesn't ask people to recall what they had for breakfast one year ago! :) Keep well!