Virtual Assistants & Stereotypes

Description

This activity is a re-creation of a well-known study conducted in the '90s by Elizabeth Loftus and John Palmer on the fallibility of eyewitness testimony.

They were studying how the verbal description of an automobile accident affects what eyewitnesses recall about the accident - specifically how fast eyewitnesses thought the cars were going and whether or not they saw broken glass on the road.

In this recreation, participants watch a video of a car accident and then are asked to estimate the speed of one of the vehicles. One group was told the vehicle "zoomed" past while the other group was told that it simply "passed" the other vehicle. Participants are then asked to estimate the speed of the vehicle in miles per hour.

Instructional objective: participants learn by comparing the data from the two groups that the wording used in the question ("passed" s. "zoomed passed") caused participants to estimate the car's speed differently.

Tools Employed:

  • A Google Site was used to contain the 7 steps in the study.

  • Google Forms were used to gather participants' estimates of the speed of the cars in mph.

  • Google Spreadsheets collected the data from the Forms and this data was embedded into the Google Site.

  • Youtube: a video of a car crash was embedded from Youtube.

  • Random Assignment: participants were placed into group 1 or 2 based on the results of a random number picker emedded into the page (from Random.org).

Flow of the Activity

Click image to enlarge