Research Projects
  • Labor Allocation in Paid Crowdsourcing: Experimental Evidence on Positioning, Nudges and Prices
    (w. John Horton)
    This paper reports the results of a natural fi eld experiment where workers from a paid crowdsourcing environment self-select into tasks and are presumed to have limited attention. In our experiment, workers labeled any of six pictures from a 2 x 3 grid of thumbnail images. In the absence of any incentives, workers exhibit a strong default bias and tend to select images from the top-left ("focal") position; the bottom-right ("non-focal") position, was the least preferred. We attempted to overcome this bias and increase the rate at which workers selected the least preferred task, by using a combination of monetary and non-monetary incentives. We also varied the saliency of these incentives by placing them in either the focal or non-focal position. Although both incentive types caused workers to re-allocate their labor, monetary incentives were more effective. Most interestingly, both incentive types worked better when they were placed in the focal position and made more salient. In fact, salient non-monetary incentives worked about as well as non-salient monetary ones. Our evidence suggests that user interface and cognitive biases play an important role in online labor markets and that salience can be used by employers as a kind of "incentive multiplier".

  • Predicting and Preventing Shootings among At-risk Youth
    (w. Steven Levitt and John List)
    Each year, more than 250 students in the Chicago Public Schools (CPS) are shot. The authors of this paper worked with the leadership of CPS to build a predictive model of shootings that helped determine which students would be included in a highly targeted and resource intensive mentorship program. This paper describes our predictive model and offers a preliminary evaluation of the mentoring intervention performed by Youth Advocate Programs, Inc. (YAP). We find little evidence that the intervention reduces school misconducts or improves educational outcomes. The scale of intervention was too small to generate meaningful findings on shootings.
    • Published in American Economic Review: Papers and Proceedings (Dec 2010)
    • Download the paper

  • Preventing Satisficing in Online Surveys
    (w. Adam Kapelner)
    Researchers are increasingly using online labor markets such as Amazon’s Mechanical Turk (MTurk) as a source of inexpensive data. One of the most popular tasks is answering surveys. However, without adequate controls, researchers should be concerned that respondents may fill out surveys haphazardly in the unsupervised environment of the Internet. Social scientists refer to mental shortcuts that people take as “satisficing” and this concept has been applied to how respondents take surveys. We examine the prevalence of survey satisficing on MTurk and find that MTurk respondents pay as much or more attention to instructions as do traditional undergraduate populations. We present an alternative way of presenting survey questions, called Kapcha, which reduces satisficing by slowing respondents down and by "fading-in" the text on survey questions. This method is found to substantially improve the quality of survey data. Finally, we present an open-source platform that can be used for further survey experimentation.

  • Breaking Monotony with Meaning: Motivation in Crowdsourcing Markets
    (w. Adam Kapelner)
    We conduct a natural field experiment that explores the relationship between the “meaningfulness” of a task and people’s willingness to work. Our study uses workers from Amazon’s Mechanical Turk (MTurk), an online marketplace for task-based work. All participants are given an identical task of labeling medical images. However, the task is presented differently depending on treatment. Subjects assigned to the meaningful treatment are told they would be helping researchers label tumor cells, whereas subjects in the zero-context treatment are not told the purpose of their task and only told that they would be labeling “objects of interest”. Our experimental design specifically hires US and Indian workers in order to test for heterogeneous effects. We find that US, but not Indian, workers are induced to work at a higher proportion when given cues that their task was meaningful. However, conditional on working, whether a task was framed as meaningful does not induce greater or higher quality output in either the US or in India.

   




Ph.D. Student in Economics
Massachusetts Institute of Technology
Contact me:  
dchandler (at) mit.edu

Keep in Touch: