Paper - Lookit (Part 1): A New Online Platform for Developmental Research

  • Metadata:
  • Also see: scottLookitPartAssessing2017scottLookitPartAssessing2017

    Metadata:

    author: Kimberly Scott, Junyi Chu, Laura Schulz
    title: Lookit (Part 2): Assessing the Viability of Online Developmental Research, Results From Three Case Studies
    ...
  • Lookit aims to be a platform to conduct online research experiments for young children in developmental research.
  • According to the paper, there are three practical challenges for developmental sciences: recruiting an adequate number of children, reaching special populations and scheduling repeat sessions.
  • For adult testing: Amazon Mechanical Turk does a great job (original citations needed)
  • ethical issues: when looking at ethics, special care needs to be taken in the online environment.
    • fair and noncoercive reimbursment
    • ensuring the validity of informed consent
    • protecting parents privacy
    • -> The paper describes how they handled those
  • Lookit experienced a few technical issues
    • Video quality was dependent on upload speed (?!), videos with low effective framerate were excluded
    • Video usability. 35% of videos were unusable, as they were missing, incomplete or due to framerate issues
  • Lookit also experienced a few methodological issues
    • typical fussiness
    • Parents were not compliant (they should have their eyes closed during the trials)
  • It cites a paper by slaughter & suddendorf 2007 that claims that 14% of infant children are, on average, excluded from studies due to fussiness while only 2% of studies give an operational definition of fussiness.
  • Families participating in Lookit were more representative of the US population than in typical samples.
    • However, parents that did not finish highschool were still hard to reach
  • Recommendations:
    • Online studies can improve the outreach (more participants and more varied participants) while still providing usable data in devleopmental research
    • Parent blinding was problematic, so providing clear, simple explanation of why this is necessary combined with previewing of the shown material will help here
    • Clear exclusion criteria are extra crucial in online studies
  • Only 26% of the entries were usable in the study, the rest needed to be excluded. This was mainly caused by the online environment
    • The authors argue that the time saved by issuing studies online can still be a net positive
  • A weakness is that the experimenters cannot directly engage in the testing. This is only partially fixable for online studies.
  • Lookit still needs programming expertise to fully use.