Background

In a study reported in the November 2007 issue of Nature, researchers investigated whether pre-verbal infants (e.g., 10-month olds) take into account an individual’s actions towards others in evaluating that individual as appealing or aversive (Hamlin, Wynn, and Bloom, 2007). In one component of the studies, infants were shown the following two videos. Please watch each video at least once.

Video 1
Video 2

The infants were then presented with the two objects (the "helper" and the "hinderder") to see whether they would chose one to play with.

Video 3

Important note: The shapes and colors of the objects were randomized across the infants. It was also randomized whether the helper object was put next to the infant's left hand or right hand. 

Research Question: Do pre-verbal infants show a tendency to prefer the "helper" toy over the "hinderer" toy in the long-run?

Goals: In this lab, you will

• Investigate a research question through:
  • Simple numerical and graphical summaries
  • Use of the One Proportion Inference applet
• Explore the concept of “statistical significance” and p-values to assess a research conjecture
• Use sample data to draw conclusions about an overall process

Summarizing the Observed Data

After collecting the data, the next step is to produce some numerical and graphical summaries of the sample data. The goal is to discuss the overall patterns in the data. Here we don’t want to just present a bunch of "helper" and "hinderer" outcomes, but a single number and a picture or two that best represents what we learn from the data. First some terminology:

• The observational units are the objects we are measuring. Here, the observational units are the infants who made a clear choice.
• The sample size is the number of observational units for which we have data. Here, n = 16.
• The variable is the "question" posed to each observational unit. Here, "which object did the infant select."
• This is a categorical rather than a quantitative variable. A categorical variable with only two categories is often referred to as a binary variable. With a binary variable, we often label the two outcomes as "successes" and "failures."

Numerical Summary

To summarize a binary variable numerically, we just want to know the count and proportion for each category (Helper or Hinderer). Here we could choose to consider selecting the Helper toy to be the "success."

(a) Examine the results in the file, and then enter the count, proportion, and percentage of infants that chose the helper toy. Hint: Proportions are always numbers between 0 and 1 (inclusive). Please report proportions to at least 3 decimal places. 

Graphical Summary

(b) Choose between a free online applet or Excel and follow the appropriate link below for instructions on making a simple bar graph of these results.

Excel Enter your category proportions: In A1, describe the variable In B1, type: Proportion In A2 and A3, enter the category names. In B2 and B3, enter the observed proportions (or a formula, e.g., "=14/16") for each choice.
Drag your mouse over the six cells you have created to highlight them.
Choose the Insert menu and then select the Column or Bar Chart chart (shown to the right).
Then from the Column menu/Chart Wizard, select the first option: "Clustered Column."
Note you can change the colors and labeling. At least: Right click on a bar and select Add Data Labels Under Chart Tools > Design > Add Chart Element, select Axis Labels > Primary Horizontal to label the horizontal axis. Also, click on the title and type in a real title.