When analyzing second hand information in a data set (participant is recounting someone else's experience), if you cannot access that person to confirm their experience, what are some cautions to take in interpretation? Can you use the data at all?

Also, if you do use the data, wouldn't there be potential ethics issues considering the other person has not signed a consent form?

  • 1
    Could you kindly expand on what type of interview you mean? – Ébe Isaac Jan 12 '17 at 8:46
  • I don't think there would be an ethics problem, if the third party is not identified. I think the main consideration would be that you would not be able to assign the same a priori level of reliability to the second-hand reported data. But I think you can still collect it. Example: parents are the primary source of information used as a basis for diagnosing Tourette Syndrome in children. – aparente001 Jan 12 '17 at 8:57
  • @ÉbeIsaac A semi-structured interview for multiple case studies in social sciences – curious Jan 16 '17 at 2:26

...what are some cautions to take in interpretation? Can you use the data at all?

The most obvious caution to take here is simply to be upfront in your research about the fact that certain data is unconfirmed 'hearsay' data, consisting of a recounting of another person's experience, without confirmation from the primary person. Whether it can be used legitimately depends on the purpose. There is no particular reason that data could not be used at all, but that really depends on whether hearsay evidence constitutes valuable evidence on your research question. Regardless, in your description of your data, and subsequent interpretation, you should be clear about the fact that this data is hearsay, and your interpretation should take account of this (e.g., one possible hypothesis would be that the hearsay account is incorrect).

As in courts of law, hearsay should be considered to be weak evidence, due to the possibility that the account of another person's experience may not accord with that person's own description. Hearsay evidence has an additional degree of removal from reality relative to direct evidence from a person about their own experience and actions. As such, it has a higher chance of error, and this should be conceded clearly in your analysis and interpretation of your data.

...wouldn't there be potential ethics issues considering the other person has not signed a consent form?

As with any academic research project, if you are using data that identifies the characteristics or experiences of individuals, it is generally the case that you will need to present your results in a way that anonymises the data, and does not allow the reader to identify the characteristics of any specific individual. The same should apply whether the information comes directly from that individual, or as hearsay from another person.

There may be cases where it is legitimate to report hearsay information about individuals as part of research. For example, a research project might ask respondents to give feedback and opinions on the actions of local politicians, in which case it would be legitimate to report this data without the consent of those politicians. (In this case the respondents would be anonymised, but the politicians probably would not be.) In other cases, when you are dealing with people that are not public figures, it might not be legitimate to report outcomes that allow identifiability of those people.

If you are unsure about the ethical requirements for your study, you should contact the relevant ethics boards or ethics officers at your university. Academic studies involving people generally require ethics clearance anyway, but if this matter has arisen as an additional issue, you should contact the ethics board and seek further guidance from them.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.