IRB Advice and best practices
About this document
These suggestions have been compiled in an effort to streamline the IRB process by answering frequently asked questions and highlighting common mistakes. Doing everything here is not a guarantee that your application will be approved, but cumulative experience of faculty in the School suggests that these strategies can help. If you have a suggestion or correction, please send an email to the chair of the Research Committee.
For now (February 2014), the IRB is short staffed. There are efforts underway to remedy this, but the problem will not be fixed instantly. As a consequence all reviews are taking longer than they ideally would. This applies to all types of reviews—exempt, expedited, amendments, and full reviews. The exempt queue in particular is backed up, often because investigators are submitting protocols that are not suitable for exempt review (see below for key differences between exempt and expedited protocols.) There are a few things you can do to help minimize the delay:
- Use amendments for related projects, rather than submitting new applications each time. The follow-up projects must be on the same general research topic, but they can address different questions, use different methods, etc. This is can save you multiple weeks on approval for a new study. You do not have to indicate in your initial application that you plan to do this, though choose a generic name for the study is a good idea.
- Make sure that all members of the research team have completed the eCOI and CITI before submitting. You might even ask your collaborator(s) or research assistant(s) to provide you with a screenshot or PDF documenting CITI completion so that you can respond promptly if the IRB should ask. Unfortunately, the eCOI does not have a receipt.
- You do not have to build your survey online (e.g., using Qualtrics) in advance of approval. A Word document listing the items is sufficient. Including supporting material for the reviewer, such as scale names or cites, may also help. If you receive an exempt review that demands that you build a survey in advance, talk with one of our IRB representatives.
- Be clear about online panels. If you use a panel company to recruit participants (e.g., Qualtrics), the description of your recruitment should explain that panel company is making individuals who have expressed an interest in completing surveys aware of your project, and that they are managing the eligibility parameters. They are not involved in the design or execution of the study itself. Include this information both in your initial application and your research protocol.
- You do not have to buy/submit copies of all media materials (e.g., movies) used in your study. If you can reference an online link or provide a clip, this is helpful. Mention movie ratings. If clips are from G, PG, PG-13 movies and being shown to college students, for example, this is good to mention. There may be other helpful links (e.g., a parents’ website review of objectionable content in the movie) you can provide as well. If a reviewer indicates buying a copy is necessary, talk to one of our IRB reps for help.
- You do not have to submit every possible stimulus you may use. Instead, you should provide typical stimuli as well as “boundary condition” or extreme examples of stimuli. Explain how the samples are representative of the stimuli. You may also want to consider what proportion of your stimuli you include (e.g., 2 pictures may not be sufficient if you are using 200 photos). If stimuli have been approved in another approved protocol, you may reference it (although this does not guarantee instant approval, it does give the reviewer more support in decision.)
- Deception is not permissible in exempt protocols.
- Deception can be a somewhat gray area. Whether or not omission constitutes deception is particularly unclear and responses will depend on the reviewer. Some reviewers do not consider obscuring the purpose of the study beforehand to be deception; instead, they interpret it as providing less detail up front. Others consider this deception, particularly if it may cause discomfort in participants (e.g., if you describe a study as a study on television characters, and then reveal in the debriefing you were more specifically looking at how television characters may evoke racist attitudes, participants may feel “tricked” by the nature of the research.)
- For an expedited protocol, omission (i.e., a lack of detail) in consent can be compensated with a clear debriefing and the clearly elaborated opportunity for participants to withdraw their data for a reasonable time afterwards (e.g., 2 weeks) in the debriefing. This is a determination made by individual reviewers, but in general this should not require deception paperwork. In the example above, if the nature of the study was explained in the debriefing and participants were given the opportunity to withdraw their data at that time or up to two weeks after participation, this should satisfy the reviewers.
Exempt vs. expedited protocols
Although they require slightly more paperwork and an extra signature, expedited protocols have one big advantage: you can make changes to the protocol.
- Changes are not permitted to exempt protocols. This includes changes in personnel, participants, measures, etc.
- Sensitive material cannot be included in exempt protocols. From the IRB website on exempt clarification: “In some studies, even when the data collected are anonymous the potential for social stigma of an identifiable group can make IRB review a requirement. Surveys that contain invasive questions that may cause participants to experience emotional distress or discomfort while answering are also not exempt from IRB review, even if the data are collected from participants who remain anonymous.”
- Deception cannot be used in an exempt protocol. Exempt protocols cannot include “research that is greater than minimal risk, involves coercion, undue influence, deception, or any practice that does not follow the ethical principles described in the Belmont Report.” As noted above, there is some ambiguity as to what constitutes deception.
- Experimental manipulations are not (consistently) allowed in an exempt protocol. The general consensus among IRB reviewers is that any “manipulation” or “intervention” is not acceptable for exempt review. The School is currently asking for justification for this from the IRB. If you’ve gotten experiments approved in the past, you can try going through exempt review. Fair warning: if you take this risk, you may be prolonging your own process, however, as it may get bounced back for expedited review instead.
Suggestions initially compiled by Jesse Fox and Kelly Garrett in February 2014
This document offers suggestions that may help researchers when preparing IRB applications for TESoC studies, which use Qualtrics Panelists. Following these suggestions does not guarantee success, but it may allow researchers to avoid some common pitfalls. If you have a suggestion that you think belongs here, please describe it in an email to TESoC@osu.edu.
- In your description of how you recruit participants you should explain that Qualtrics is making individuals who have expressed an interest in completing surveys aware of your project, and that they are managing the eligibility parameters. They are not involved in the design or execution of the study itself. Include this information both in your initial application and your research protocol.
- If you are using a general population sample, African American and Caucasian, age 18+, under incentives you can say something like, "Participants in this survey receive cash incentives paid in USD in the amount of $1.40." (Note: This specific language is untested, but similar language was used for studies using an SSI sample in the past.)
- You cannot guarantee confidentiality with an online study. You can, however, include a statement like the one below in your consent form. Although this has been used with some success in prior applications, we do not have official approval from the IRB at this point.
…This study takes around 20 minutes to complete and your responses are confidential*…
* While we strive to use industry-standard security technology to help protect your personal or demographic information, we cannot guarantee the security of the information you transmit.
Suggestions initially compiled by Kelly Garrett in February 2014