JMP platforms and statistical methods

Distribution:
Histogram, box plot, summary statistics, one-sample t-test

Fit Y by X:
Two-sample t-test

Specialized Modeling:
Paired t-test

Formula Editor:
Creating a new variable

Objective

Use statistical inference for means, including one-sample t-test, two-sample t-test, and paired t-test, to determine if an experiment demonstrates statistical evidence that dogs exhibit jealous behavior when their owners interact with other dogs.

Problem statement

Psychology researchers from University of Auckland designed an experiment to study the behavior of dogs in potential jealousy-inducing social interactions.

The study selected 18 dogs (11 female, seven male). Each of the dogs (the subjects) experienced two test conditions. The Dog Condition was such that the subject first saw its owner next to a realistic looking fake dog. The fake dog was then concealed from view from the subject using a barrier. The owner interacted with the fake dog (e.g., petting), though the subject dog could not directly see the fake dog that the owner was interacting with. This condition was to test if the subject dogs were potentially envisioning the interaction.

The other experimental condition (Cylinder) allowed the subject dog to see its owner interacting with a fleece cylinder as if it were a dog, although the subject can clearly see it is simply a non-dog object. In this condition, the fake dog was also placed so that the subject dog could see it was in the same room, but its owner was not interacting with it.

Following the two experimental conditions, a third trial was conducted (Confirmation). In this trial, the owners interacted with a fake dog that was visible to the subject dog for the entire time. This trial ensured that all the subject dogs would react enough to a visible jealousy-inducing condition and also create a baseline of the amount of force each dog would apply in such a condition.