Non-interactive dyadic studies would include pairs of people such as coworkers, parents and their children, romantic partners, or friends. Connect allows you to target people who are in these types of relationships (and others). Below, we offer tips and best practices for conducting dyadic studies. 

Targeting Participants in Dyadic Relationships 

To begin, you’ll want to target participants who have previously indicated that they are in a relationship of the kind you are targeting. This can easily be achieved using Connect’s demographic targeting tool. There is no additional fee for using demographic targeting, and there are several options for potential dyadic relationships. 

For example, if you’d like to select participants in romantic relationships, you could use the relationship/marital status question, selecting in a relationship, married, or in a civil union/partnership: 

If there is a particular dyadic relationship that you are interested in studying, but no option currently on the Connect demographic targeting options, CloudResearch can add the question for you so that participants in your specified dyadic relationship are verified and answer the question in a way that is dissociated from your study.

Once you have selected the demographic target, your study will only show up to qualified participants who meet the given demographic criteria. 

Setting up Your Study 

Payment

Participants should be offered compensation that is higher than average to participate in dyadic studies. We recommend paying each participant (both the Connect participant and their partner) at least $10 per hour (more if you can afford it) because dyadic studies require more than a typical task on Connect. In addition to the participant’s payment, you will need to also account for their partner’s compensation in your pay rate. To compensate the partner who isn’t on Connect, you can offer to bonus participants for their partner’s participation or make the original payment twice as much and add in the instructions that participants should plan on sharing the pay with their partners. 

Title and Study Information

The title of your study should clearly indicate that the study will require 2 people (e.g., Relationship Study for Couples) in the Basic Information form. Additionally, in the description of your task under Basic Information, you’ll want to describe the task as requiring two people (e.g., “This is a romantic relationship study. Both you and your partner will need to complete questionnaires about your relationship, independently and confidentially, in order to be paid and approved for this task). Finally, you’ll want to instruct participants to send the study link to their partner (their spouse, mentor, child, for example) when you provide instructions to your participants in the Basic Information form. You can write something such as “You will need to send the survey link to your partner in order for them to take the survey. The link is: <insert hyperlink address here>. Please make sure that your partner also has your Connect ID to enter in their survey. Please note that your task will not be approved unless you and your partner both complete the survey and enter your Connect ID.” In the instructions, it can also be helpful to give participants a stern warning against filling the survey out twice. For example, “please do not attempt to fill out the survey more than once or have your partner complete the survey on your behalf. This is important for the integrity of the study. Participants found to be in violation of this will have their submissions disqualified and will not receive payment.” 

Make sure that there is a space in your survey for participants to enter the original participant’s Connect ID at the beginning of the study (e.g., Please enter the Connect ID that was given to you by your romantic partner. If you are the original person in the dyad to select this task, please enter your Connect ID. Remember that the Connect ID entered should be the same for you and your partner). In order to link the data of both people in each data, it is essential for each person to enter the original person’s Connect ID. 

Best Practices and a Word of Caution

Before launching your study, we recommend launching a pilot study for around 5-10% of the total complete number of participants you’re aiming for. The pilot data will help ensure that you can match up partners’ responses and that participants aren’t getting confused by the instructions you’re providing. If pilot participants provide you with feedback for how to make the study less confusing, you can then incorporate their ideas in your full study. 

As a word of caution, it is important to note that this recruitment strategy does require participants to be honest about their partnerships. There is no way to 100% verify that one person isn’t filling out both surveys or that the dyadic relationship is real. However, because demographic information questions are dissociated from the dyadic study, there would be no way for participants to purposefully lie in order to qualify for the study. So, make sure to use the demographic targeting tool discussed earlier so that these dyads are previously verified. As an added layer of protection, we strongly recommend researchers include questions to verify that two different people are completing the survey. You may need to get creative here. One great example comes from recent research by Nikolova and colleagues (2023) where the researchers have romantic couples living in the same household upload pictures of their hands individually and then upload a picture of their hands together. This procedure helps to show that they were completing part of the survey alone and another part of the survey together, while maintaining anonymity with the photos.

We also recommend only targeting relationships that are unlikely to pose friction. This would exclude relationships such as employee/manager dyads where there might be some level of discomfort involved with the request. We would recommend sticking to partner/friend/family/coworker dyads. And lastly, as previously mentioned, try to incentivize the study well, making it clear that fraudulent data will lead to a rejection/ban.

For dyadic studies that are more complex, we’d be happy to provide more advice or even help manage the study using our Managed Research service. Please feel free to contact us here.