Lab

Name1*

First NameLast Name
Name2*

First NameLast Name
E-mail*
example@example.com

Investigation 1: Helper or Hinderer?

An investigation reported in the November 2007 issue of Nature (Hamlin, Wynn, and Bloom) aimed at assessing whether pre-verbal infants take into account an individual's actions towards others in evaluating that individual as appealing or aversive, perhaps laying for the foundation for social interaction. In one component of the study, 10-month-old infants were shown a "climber" character (a piece of wood with "google" eyes glued onto it) that could not make it up a hill in two tries. Then they were shown two scenarios for the climber's next try: one where the climber was pushed to the top of the hill by another character ("helper") and one where the climber was pushed back down the hill by another character ("hinderer"). Each infant was alternately shown these two scenarios several times. Then the child was presented with both characters and asked to pick one to play with.

Video 1

Video 2

Video 3

Step 1: Ask a Research question
(a) State a research question for this study.
Step 2: Design a study and collect data
(b)
Identify the observational units in this study blank
Define the variable of interest blank
Is the variable quantitative or categorical? blank
(c) The Methodology section of the paper also states that for the 10-month-olds, the climber was a yellow triangle; helper and hinderer were a red square and a blue circle (counterbalanced). Also counterbalanced were which event (helping or hindering) the infants observed first and the positions of helper and hinderer when presented to the infants (on left or right). Why are these important considerations?

Step 3: Explore the data
Open this data file. Because the variable is categorical, to summarize the results numerically, instead of looking at the mean, standard deviation, and skewness, we will simply count the number of outcomes in each category. With a binary categorical variable, we often denote one outcome as "success" and one as "failure." Which is which doesn't really matter. For this study, let's consider choosing the "helper toy" as success.
(d) Report the sample size blank , the number of successes blank , and the proportion of successes blank .

With categorical data, instead of a dotplot we will start with a bar graph.

Using the applet, enter the data by pressing Clear and then typing InfantData.txt in the Enter data box.
Press Use Data twice. (Make sure the sample size n matches.)
Now drag the “choice” variable from the Variables list to the Response box.
Check the Show descriptive box (scroll down?).
On the far left, check the Edit title/axes box.
Scroll right and change the category names to not be truncated after 4 letters. Add a descriptive title to the graph (briefly summarize what the results are about).
- Titles should provide enough information so we can understand the variable (e.g., body mass index vs. body weight), who, when, and where (e.g., 100 Stat 217 students). The title should help the graph stand on its own and allow the reader to comprehend key takeaway messages with a quick skim of the graph. Rather than a generic phrase, use a description sentence that encapsulates the graph’s finding or “so what.”
Make a screen capture of your graph (only).

(e) Upload your screen capture here.

Cancelof
(f) Did a majority of the infants pick the helper toy as the researchers suspected?

Step 4: Draw Inferences Beyond the Data
We could stop here, but that would only tell us about these 16 infants. Instead, let's learn a little more to see whether we can make reasonable claims about 10-month-old infants in general...
YesNo
So let's figure out how plausible or believable it is that the probability a 10-month infant picks the helper toy is 0.50. In other words, I want to evaluate the following model:
- All 10-month infants have the same probability of picking the helper toy
- That probability is equal to 0.50
- What color or shape the object is or which video they watched first do not impact this probability (these effects have been 'averaged out' across the infants)
In other words, the graph of the proportion choosing the helper toy could still be bouncing around, giving us a proportion of 0.875, just by random chance variation alone, and if I was to keep testing infants forever, the graph would converge to 0.50?

To compare our data to this model, we are going to simulate sets of 16 infants where we know each infant is equally likely to pick the helper toy and the hinderer toy.
(h) Toss a coin 16 times to represent the 16 infants (assuming they are identical to each other). Record the number of heads in your 16 coin tosses.
(i) Pool your results with the rest of the class and include a screen capture of the class results.

Cancelof

Computer Simulation

Let's turn to the computer to repeat this random process many many times.

Keep the probability of heads set to 0.5
Set the Number of tosses to 16 (Why?)
Press Draw Samples.

The computer replicates what you did with the coin and keeps track of the number of heads in a dotplot.

Uncheck Show animation and press Draw Samples again.

TAIQ: Did you get the same value for the number of heads both times? Should you? Why or why not?

Set the Number of repetitions to 98 and press Draw Samples.

(j) Based on your simulation, results does 15 appear to be an unusual outcome for 16 attempts with the "coin tossing model"? Explain how you are deciding.
In the long-run, 10-month-old infants have a genuine tendency to pick the helper toy (the probability is larger than 0.50)In the long run, 10-month-old infants do not have a genuine tendency and we found a majority in this sample of 16 infants just by “random chance”
(l) Explain your reasoning for your answer to (k) as if to someone not in a statistics course. (Hint: Refer to whether the observed data (15 out of 16) were consistent or inconsistent with the proposed model?)
Should be Empty:

Investigation 1: Helper or Hinderer?

Step 1: Ask a Research question

Step 2: Design a study and collect data

Step 3: Explore the data

Step 4: Draw Inferences Beyond the Data

Computer Simulation