Guidelines for Secondary Data Analysis

Secondary data analysis occurs when a researcher wants to analyze data that differs from the purpose for which the participants originally gave consent. Data may include, for example, a data set saved in a spreadsheet format or the re-analysis of raw data such as the recoding of videotapes or transcripts. Data used in a secondary data analysis may have been collected in a previous study, but may also include data collected outside of research (e.g., medical or administrative records, student grades). Students who use data collected by their supervisor for their thesis are considered to be using the data for a purpose not usually intended in the original ethics application or informed consent and therefore should submit an application. Data that are within the public domain does NOT require ethical review. Data in the public domain involve information to which anyone can access, such as analyzing magazine articles, surveying a certain type of blog on the internet, or examining newspaper personal ads.

Research involving secondary data analysis does not necessarily have to involve a review by the UREB. What is of importance in deciding whether review is necessary is the nature of the data the researcher wishes to analyze. Specifically, the data can vary in terms of the amount of identifying information. The data can:
1. Be anonymized either by being stripped of identifying information or by never having contained identifying information.
2. Have had identifying information removed and replaced with a code. The applicant does not have access to the code.
3. Contain information that could potentially be linked together to identify participants
4. Contain identifying information on participants

Research that involves data that have been anonymized or for which the researcher does not have access to identifying information does not require review by the UREB (i.e., # 1& #2). The Chair of the UREB needs to merely confirm that the data to be used are indeed anonymous. To verify that the data are anonymous, the researcher needs to complete Section A of the secondary data analysis form.
For research that involves identifying data (i.e., #3 & #4), a thorough review of the project will be made by the Chair of the UREB and both sections A and B will need to be completed. Under these circumstances the researcher must: 
• make the case for why it is necessary or unavoidable for the researcher to have identifying information concerning the participants.
• indicate how participants’ privacy and the confidentiality of the data will be protected
• explain why participants would not object in principle to the secondary analysis of data (e.g., the consent form refers to future data analysis)

If the researcher can adequately address the above concerns then the researcher may not be required to obtain additional consent from the participants. If not, the researcher must explain how further consent will be achieved.

In the case where the researcher feels it is impossible or impractical to obtain consent again, the researcher must explain why. Furthermore, to obtain a waiver of consent, the researcher should identify how:
• the analysis will involve no more than minimal risk for the participants
• the waiver is unlikely to affect the rights and welfare of the participants

Additionally, the researcher should consider strategies to allow participants to object and/or communicate to relevant groups about the research (e.g., posting notices where participants may frequent) especially if the nature of the data collected is of a sensitive nature. In the case of research involving a particularly sensitive topic, the researcher can also consider obtaining proxy consent whereby a group of individuals from the same population as the study’s original participants reviews the project and provides feedback on whether they would have agreed to participate.