4. Checking consents and collecting data
After your IRB application is approved, the next step is to check the consents of your research participants. Having a system to check who has consented is important for ensuring ethical data collection.
First, in order to keep track of consents, creating a spreadsheet with consenting participants’ names can be helpful for organizing the process.
The screenshot below shows an example of a spreadsheet that is similar to what we use. In this spreadsheet, each sheet includes consented students from a particular instructor that are included in our corpus. Note that this data is not real and only used here for demonstration purposes.
Once you have collected all the information about consented participants and organized them in the spreadsheet, you can start collecting data. There are different ways to collect this data.
You can collect data directly from participants through a digital form such as Qualtrics or JotForm. However, you may not get the same level of participation this way, particularly if you do not have incentives for participation.
A more effective way of collecting data can be to work with instructors or other stakeholders who can provide access to student/participant texts. To develop the Crow corpus, we contact instructors at the end of the semester with a list of consented students from their class. The consented student spreadsheet can facilitate this process. To make sure that there is no conflict of interest for the instructors, you can wait to provide instructors with the information about the consented students after final grades have been submitted.
There are two options that we have found helpful to offer instructors when collecting student data:
In order to reduce instructors’ burden, they can share access to their learning management systems (LMS), such as D2L or BlackBoard, or another web platform where texts are stored. That way, instructors or other stakeholders do not have to spend time downloading texts. Once you have access to the LMS, you can use the consented participants spreadsheet that you created previously to select and download only the texts from the consented participants. Alternatively, you can batch download all student texts from a given instructors’ course for a given assignment or draft. Then you need to delete texts from non-consented students.
Tip: if you are directly collecting texts from a learning management system, you may want to contact your IT office to check if you could automatically scrape your LMS. The key here is whether an application programming interface (API), which allows the automatic download of data, is available. We unfortunately have not been able to use this method for our own data collection at University of Arizona, since D2L, our learning management system, does not have an API. However, other websites generally have an API, so if you are not collecting from an LMS this can be an effective way to gather data from the web.
Sometimes instructors or other collaborators might not want to share access to participant data/their LMS, or prefer to be more actively engaged in the process of data collection. In this case, you can invite them to submit the consented participants’ texts to a secure drive. We use Box for this purpose because it is approved by our IRB. You should check with your local IRB to make sure where they prefer you store identifiable student data.
A video version of this content is available on the Crow YouTube channel.
Previous: 3. Ethical issues in corpus building
Next: 5. Organizing your data