Sensitive questions must be asked, it’s the only way to peel away more layers and reveal a deeper and more detailed truth. That is of course if they are asked in the right way, otherwise respondents can be harmed, are alienated, dropout rates increase, and the detail and honesty of answers decreases.
So, what is the ‘right way’ to ask sensitive questions? Well, that depends on the respondents and subject involved but there are some universal rules that will always help:
1) Understand the Ethical Landscape
Respect for respondents' privacy, cultural differences, and emotional well-being should always take precedence. This must be considered in the processes, methodology and questions so that all respondents are clear on the research objectives, give informed consent, are allowed to decline to answer specific questions without consequence and are kept anonymous. Beyond this, questions should be phrased in neutral and non-discriminatory ways that avoid bias, stereotypes, or prejudiced language. And then they need to be asked in a manner that allows all equal ability to answer too.
2) Methodology
Use a research methodology that allows respondents as much privacy as possible. This can mean using platforms like chatrooms, WhatsApp, SMS, computer assisted self-interviewing (CASI) or even voice only Zoom interviews. Being able to answer without showing face gets more honest and more detailed answers, multiple studies confirm this. Duffy and colleagues show that respondents in face-to-face interviews are more susceptible to social desirability bias because of the interviewer's presence, Newman showed that sensitive question response rates increase in the absence of an interviewer, and Kreuter and team found lower misreporting of sensitive activities when respondents could do this on their own.
3) Sensitivity Training
Sensitivity training ensures that not only is no harm done but that more trust is built, and deeper insight is gained. All fieldworkers need to participate, and the training needs to very specifically deal with the issue being researched. Beyond this all fieldworkers should be provided with a list of resources and organisations to refer respondents to should it become clear further assistance is required.
4) Questions
Questions need to be asked in a way that is not only neutral but also allows respondents to answer with dignity. It is difficult to admit you earn $0 but less so to say that you earn under $1000. So, when asking income instead of having a scale that starts with ‘$0 to $1000’ change that to ‘Under $1000’, especially if respondents are reading the answer options themselves. Also, use inclusive language, so that respondents don’t feel like they are singling themselves out by answering honestly, for example ‘Which of the following do you identify with…?’ as opposed to ‘Are you…?’ or ‘What is your relationship status?’ as opposed to ‘What is your marital status?’
Of course all this is just the start, but a great start and strong foundation on which to build the rest of any study. Need help starting with the rest? Get in touch and we’ll have you on the way in no time!
