Wednesday, July 20, 2016

Ethical considerations in research data collection

There is a saying that "it doesn't hurt to ask." While that may often be true in casual conversation, asking questions as part of data collection is sometimes costly and has ethical considerations.

Is the data already available?

Census data is free. Okay, it's not exactly free, because your taxes pay for it – but anyone can get access to the statistical data. For example, if your business wants to know which neighbourhoods have dual-income households with no children, or the number of Indian-speaking households with teenagers in a specific suburb, that information is available from The Australian Bureau of Statistics (ABS), U.S. Census Bureau, UK Office for National Statistics, or other national Census office.  Sometimes you may need assistance in extracting the relevant data from Census data, but it is easier and less expensive than collecting data.

You might also be able to find relevant statistical data published by other sources, such as universities, chambers of commerce, and local governments.

What is the most efficient way to collect the data?

When you create a research plan, you need to determine the target sample size that is necessary. For example, if you want to research Martian immigrants in the USA, you would check the Census data to find that there are 950,000 Martians in the USA. You would aim for a sample of at least 600 participants. The sample size would produce a confidence level of 95% and a confidence interval of four.

A confidence interval of four means that if – for example – 57% of the survey participants pick an answer, you can be confident that if you had asked the question of the entire Martian population in the USA, between 53% (57- 4) and 61% (57 + 4) would have picked that answer.

When you put the confidence level and the confidence interval together, you can say that you are 95% sure that when 57% of our sample picks an answer, the true percentage of the entire Martian population in the USA who would have picked that answer is between 53% and 61%.

Depending on the type of data that you want, you might be able to collect data through online web surveys. Once you program the survey questions, your computers can tabulate the data automatically – it is just as easy to process 100 surveys as 9,000 surveys. Obviously, collecting data through one-on-one interviews and small focus groups is more costly and more intrusive, but those results can be combined with online survey results; for example, three focus groups, ten interviews, and 300 online surveys. This gives you valid data that is cost-effective. A mixed-methods approach also provides an advantage when exploring complex research questions. The qualitative data provides a deep understanding of survey responses, and statistical analysis offers a detailed assessment of patterns in responses.

Is it safe and ethical to collect the data?

There are types of research projects that have no ethical questions – for example, how many customers heard about the product from a specific advertising campaign.

Then there are projects have many ethical questions, such as:

  • If you thank research participants by giving them a small amount of cash or grocery store voucher, or free products, will that unduly influence the kind of answers they give?
  • Will our questions endanger or harm the participants, such as victims of people-smugglers, or survivors of institutional sexual abuse or spousal abuse? 
  • Do our questions give the participants false hope, such as expectations that completing an interview will get them a refugee visa?

There are some ethical considerations for which you have no doubts. For example, you may wish to observe an interview by using a one-way mirror from another room. If the participant has been guaranteed confidentiality, that would not be the right way to go about it. You could ask for a transcript of the interview, with identifying details redacted.


You can consider joining a professional research organisation -  such as Australian Market and Social Research Society (AMSRS) or U.S. Marketing Research Association  (MRA) and comply with their codes of professional standards, and with other relevant ethical standards for a project. Review all questions, participant criteria, and confidentiality procedures at the beginning of a project.

While you want to collect the data, you should never lose sight of the ethics involved.

No comments:

Post a Comment

Related Posts Plugin for WordPress, Blogger...