Unit 10 - Assignment Without Answers PM
Unit 10 - Assignment Without Answers PM
Comprehending the meaning and implications of basic summary measures for ratio
variables
Read this assignment carefully and answer all questions.
Hand in the assignment via Canvas.
A researcher is interested in the use of mobile devices by three large groups of students (A= languages,
B= beta sciences and C= social sciences). He draws three random samples from the three populations
and convinces all participants to install an app measuring cell phone use during lecture hours (lecture
hours are identical for all groups of students). The data are presented in Table 1.
Table 1: minutes of cell phone use during lecture hours by three groups of students
Group Observations
A 67 93 32 80 83 117 70 5 48 100 69
B 103 165 5 25 75 15 129 20 128 30
C 121 100 264 147 12 180 411 120 215
Meaning: the third student in group B only uses his/her cellphone 5 minutes during lecture hours.
In order to simplify the analysis, the data are rearranged in ascending order in Table 2.
Group Observations
A 5 32 48 67 69 70 80 83 93 100 117
B 5 15 20 25 30 75 103 128 129 165
C 12 100 120 121 147 180 215 264 411
1. Find or calculate the medians, the Q1, Q3 and the interquartile range (IQR) for each of the three
groups. Do this by hand, without using statistical software. Note: we would like you to include the
median when computing Q1 and Q3 for an odd number of observations (see reading material and
micro lecture).
1
2. Use the grid in Figure 4 to draw the boxes of the box plot. In this grid, Group 1 = Group A, Group 2 =
Group B and Group 3 = Group C. Assume that the line underneath the number belongs to the
number on the Y-axis.
3. The upper parts of the boxes of groups 2 and 3 are bigger than the lower parts of these boxes. A
colleague wonders whether this means there are more cases with higher values (bigger than the
median) than cases with lower values within these boxes. Is this correct?
4. The box of group 2 is bigger than the box in group 1. Why is that?
5. What does the median tell you in terms of the distribution of observations?
6. We will now identify the whiskers of the box plot. Please fill-in the following table:
Group
A B C
Q1 (see table 3)
IQR (see table 3)
Lower whisker Step 1: 1st quartile minus 1.5 * IQR
Step 2: Smallest value in the dataset, which is
within the 1st quartile minus 1.5 * IQR
Q3 (see table 3)
IQR (see table 3)
2
Upper whisker Step 1: 3rd quartile plus 1.5*IQR
Step 2: Highest value in the dataset within 3rd
quartile plus 1.5*IQR
8. Are there any data points NOT included in the range of the whiskers? Use stars to indicate them in
Figure 4. What are these called?
9. Copy the medians from table 3 into table 5. Calculate the means for the groups.
Median Mean
A
B
C
10. Why do you think the medians for group A and B are less different from the mean than the median
and mean in group C?
11. Since the concepts ‘variance’ and ‘standard deviation’ are central in the context of statistics, we
want you to calculate the sample standard deviations of the observations in the groups. After this
course you will always use statistical software to do that, but it helps if you understand which steps
you need to take to compute the standard deviation as well. You can use excel, but make sure you
understand all steps (so do not ask the program to compute the standard deviation for you).
12. What information does the standard deviation give us? What does the calculated SD ‘mean?