Using Peer Tutoring in Evaluating the Usability of a Physically Interactive Computer Game with Child

1
Using Peer Tutoring in Evaluating the Usability
of a Physically Interactive Computer Game with
Children

• Johanna Höysniemi
• 29 August 2002
• IDC2002, Eindhoven

2
Contents

• Introduction (1)
• The background of the method (1)
• The application of the method (2)
• The benefits of peer tutoring (4)
• Drawbacks and challenges (1)
• Discussion (1)
• Today, we are not going to talk about
• Usability and computer games

3

• Definition Peer tutoring is a usability
  evaluation method where children teach other
  children how to use the product that is being
  evaluated in a familiar social setting.
• Usability flaws can be found by observing and
  analyzing childrens communication and behavior
  in a peer teaching situation.
• Peers are children of similar age and status who
  spend time together on a daily basis.
• Tutor teacher, expert
• Tutee student, novice user

4

• Based on a well-known approach in childrens
  education.
• Development psychology (Corsaro, Damon Phelps)
• Peer interaction and learning (Topping, Garvey,
  Goodblad, Cooper)
• Social psychology and development (Rogoff,
  Resnick, Doise)
• Cognitive psychology (Piaget, Vygotsky, Wood)

• Peer tutoring vs. Co-discovery learning,
  Co-participation, Thinking aloud and Paired
  usability testing
• Druin, Nielsen, Hanna the learnability of the
  product relates to childrens enjoyment of it.

5

• Developed for the usability evaluation purposes
  of QuiQuis Giant Bounce - a physically and
  vocally interactive computer game for 4 to 9
  year old children.
• Existing usability evaluation methods did not
  suit our needs well enough.
• What were we interested in?
• Physical actions and playability
• Learnability and teachability
• Communication and instructions
• Duration of play

6

• Two test sessions with 28 children aged 5 to 9.
• Setup a webcam, a laptop, and a video camera.

• Roles interactor, tutee, tutor and cameraman.
• Two variations of the method were used.

Each-one-teach-one
Two-on-one
Images by www.lego.com
7

• Testing can be carried out in an environment
  familiar to the children.
• Peers are usually so engaged with the game and
  with each other that they do not even notice the
  testing is going on.
• When a child has a peer in the test space, the
  children are more relaxed to communicate.

8

• Adult-child communication and power structures
  could sidetrack the test situation.
• Tutor and tutee are cognitively closer and the
  cognitive load can be split.
• Children need not be asked to think aloud.
• Tutor helps adults to see the situation through
  the eyes of the other child.

9

• Children can manage their own interactions.
• One child teaches the other and thus takes an
  active role as an instructor in a testing
  situation.
• Rogoff Children are likely to treat the
  situation differently if they are in charge of it
  rather than being given a task by adults.

10

• Crucial in evaluating the physical
  appropriateness and playability of the game
  tasks.
• Physical actions and communication can be
  further analyzed.
• Children were not intimidated by the video
  camera.

11
Drawbacks and challenges

• Requires a lot of work and a multidisciplinary
  research team.
• Should be used together with a collaborative
  design method.
• If applied in a different context, the challenge
  is in designing the test setup and tasks and how
  the tutor-tutee roles are explained to the
  children.

12

• Further development and analysis of the method
  still needed

• Variations of the method
• Role issues (interactor, tutor-tutee)
• Age appropriateness
• Childrens preferences of being tutors and
  tutees
• Teaching experiences and skills needed

13
Thank you! Questions? johanna_at_cs.uta.fi P.S
More information about the QuiQuis Giant Bounce
project can be found at www.kukakumma.net