Design Choices for Crowdsourcing Implicit Discourse Relations: Revealing the Biases Introduced by Task Design
This work highlights a source of bias in linguistic annotation that could impact model training and testing, though it is incremental in focusing on task design rather than new methods.
The study investigated how task design biases affect crowdsourced annotations of implicit discourse relations, comparing two annotation methods across four domains and showing that design choices can push annotators toward specific relations.
Disagreement in natural language annotation has mostly been studied from a perspective of biases introduced by the annotators and the annotation frameworks. Here, we propose to analyze another source of bias: task design bias, which has a particularly strong impact on crowdsourced linguistic annotations where natural language is used to elicit the interpretation of laymen annotators. For this purpose we look at implicit discourse relation annotation, a task that has repeatedly been shown to be difficult due to the relations' ambiguity. We compare the annotations of 1,200 discourse relations obtained using two distinct annotation tasks and quantify the biases of both methods across four different domains. Both methods are natural language annotation tasks designed for crowdsourcing. We show that the task design can push annotators towards certain relations and that some discourse relations senses can be better elicited with one or the other annotation approach. We also conclude that this type of bias should be taken into account when training and testing models.