Question 2.1: The first experiment examined subjects' perception of height by adjusting the screen position camera position camera focus camera exposure Assign the letter corresponding to your answer to q2_1 below. [41] : 92_1 = . . . [ ] : grader . check("q2_1") Now we'll read in the experimental data from the paper. The table below gives the condition, height estimate, and true height for each subject. Recall that individuals can be in one of two conditions: "down", where they're looking up at the pictured individual, and "up", where they're looking down at the pictured individual. [44] : expi_raw = Table. read_table("frontiersi.csv") expi_raw In this section, we will be conducting an A/B test, wherein we split some data into two groups randomly and calculate some test statistic on each group. The first component of an A/B test is a test statistic, a value that we calculate that compares two groups. A good test statistic is one for which large or small values but not both indicate in the direction of the alternative hypothesis. As a precursor to our test statistic (i.e. not as the test statistic itself), we will calculate the error ratio between the estimate and the true height: estimate _ true height true height This will give us the magnitude of the difference relative to the true height, to correct for any scale differences based on the actual height of the figure, and whether the estimate was above or below the actual height. Question 2.2: How can we turn the error ratio column into a good test statistic for comparing the two conditions