Managing bots, inactive players and dropouts
SMARTRIQS offers flexibility in handling inactive participants and dropouts. You can choose from three options:
1. Bots and default responses
This method uses bots and default responses to fill missing roles and missing responses, whenever participants exceed the maximum waiting time (specified by the maxWaitTime parameter), ensuring that every participant can complete the study, even if there are not enough other participants available within the maximum waiting time. You can specify default responses for bots in the Survey Flow (see the examples below)—these can be stochastic as well, or depend on previous decisions.
EXAMPLE 1: fixed default
Generate a fixed response of 50:
EXAMPLE 2: random default response
Generate a random response between 0 and 100:
EXAMPLE 3: dynamic default response
Generate a response based on other variables:
EXAMPLE 4: multiple default responses
Generate random responses between 0 and 100 for 4 players:
Note that responses are separated by commas
To avoid any deception, participants are always notified whether they have been matched with other participants or bots (or some combination thereof), and they also receive notifications whenever a response is a default response (as opposed to a `real’ response). For example:
To set up a study that allows for bots and default responses, set botMatch = yes before the MATCH block, and define defaultData before every GET block.
2. Default responses only
The second option does not allow participants to be matched with bots but uses default responses whenever someone becomes inactive or drops out in a later stage. If there are not enough other participants available during matching, the survey is terminated. However, if a participant becomes inactive in a later stage, the study continues and the timed out participant’s response is replaced by the default response. In this case, both the timed out participant and the other group members are notified about the timeout and that the default response has been applied. If the timed out participant becomes active again, they can still submit their responses in subsequent stages and complete the study. SMARTRIQS automatically creates a log file (timeOutLog) detailing which participant has timed out in which stage, allowing researchers to decide later whether to keep or exclude their data.
To set up a study that does not allows for bots but allows for default responses, leave the botMatch parameter blank before the MATCH block, but define defaultData before every GET block.
If a study is terminated, the participant receives the following message by default: “The survey has been terminated. Please contact the researcher to receive partial compensation for your participation.” You can use any custom text by defining the terminateText parameter before the MATCH block. You can also include dynamic variables, such as completion codes in these messages.
3. No bots or default responses
The third option is the most conservative as it does not allow for using either bots or default responses. If there are not enough participants available during matching, or if any of the participants time out, the survey is immediately terminated for everyone in the group. Termination is logged in Qualtrics, and the researcher can specify a custom survey termination message with a partial survey completion code, should the participants receive partial compensation for their participation.
To set up a study without bots or default responses, leave the botMatch parameter blank before the MATCH block, and omit defaultData before every GET block (you can either delete or leave blank the defaultData embedded data in the Survey Flow).