Policy on Archival Data

Background

The analysis of already existing data is an invaluable research practice for some situations such as meta-analysis, re-working of particular analyses guided by new conceptual ideas, content or discourse analyses. In such cases there can be ethical concerns. The following has been prepared to provide some guidance to researchers employing this investigative practice. It presents an ethics protocol that is strongly recommended for researchers using archival data.

What is Archival Data?

Any observations, texts or analyses that pre-date a planned research project. Typically these have been generated under the auspices of some agency other than the Department of Psychology. Archival data can include:

• published results of investigations (e.g., for meta-analysis),
• data gathered under research projects that already have ethical clearance (e.g., for re-analysis),
• information both gathered and owned by agencies other than the Department (e.g., interview transcripts collected by a service agency, client files, photos in a museum archive);

A Protocol for Archival Data

In studies involving the analysis of archival data appropriate permission of the original data gatherers/owners is required in order to gain ethics clearance from the Department of Psychology. The requirements vary with specific types of data.

1. In cases of published data in the public record, no formal permission is required. However, responsible reportage of the sources of the data is required;

    

2. In cases of data collected under other ethically-cleared U of C, projects a letter from the owner of the data indicating permission to use them, procedures expected to insure anonymity of response protocols, any conditions pertaining to use of the data, and evidence of original ethical clearance are required;

    

3. In cases of data gathered by other agencies a letter from the agency is required. This letter should contain:
   • written permission to use the data,
   • indication of the procedures expected to insure anonymity of response protocols,
   • description of any conditions pertaining to use of the data,
   • description of the agency's ethical expectations and practices, and
   • a clear indication that the data were gathered in keeping with these ethical procedures.

Procedures

In cases (2) and (3) in the previous section, researchers using archival data should submit a covering letter explaining their project as well as copies of the relevant documentation to the Chair of the Department of Psychology Research Ethics Board. For students, this letter and enclosures should be accompanied by a memo from their supervisor indicating knowledge of and agreement with the particular arrangements being described.

The Board Chair will examine this documentation and either clear the project or, if there appear to be ethical concerns, refer it for review by the Committee. In such cases the Chair will meet the the researcher to discuss the problem and try to find a workable solution.

Potential Problems

One thorny issue is the situation where the ethical procedures of the agency owning the data are not up to the same standards as those in the Department. For example, concerns about informed consent and anonymity are, in some "downtown" agencies, not as closely regulated as ours. What should we do here? One approach is to insure that there is no possibility of participant anonymity being compromised upon analysis of the archival data. This can be addressed by adopting a procedure whereby the researcher receives non-named data protocols. If this is not reasonably possible (e.g., client files for re-analysis that are too lengthy to reproduce and remove identifying information), then signed pledges of confidentiality on the part of the researcher will be required and especial attention will be paid to the procedures required by the agency owning the data. Any possibility of anonymity breaches will result in the project not being cleared.

Electronic Texts

In some situations already-existing electronic texts (e.g., email interchanges, list serve debates etc.) can provide invaluable research data (e.g., opinion polling, attitude research, discourse analyses of various psychological constructions). In such cases, the researcher should approach the individuals concerned asking for their permission to include their texts in the analysis. Anonymity should be guaranteed. This can usually be accomplished by sending a request to the list serve or the principals involved asking anyone with an objection to reply. This request should fully explain the intent of the analysis and the safeguards in place. Researchers must remove the texts of any objectors from the data set. Copies of the messages both sent and received regarding these procedures must be presented with the covering letter and supervisor's memo. (Note: If you intend to use electronic texts that you are soliciting, rather than ones that already exist, you should follow the regular ethics protocols regarding informed consent and confidentiality)