lOMoARcPSD|53991452
Week1 - Assignment on Word Document
Intro to Biomedical Statistics (National University)
Scan to open on Studocu
Studocu is not sponsored or endorsed by any college or university
Downloaded by kuchy kuchy ()
, lOMoARcPSD|53991452
**I have carefully studied my instructor’s written homework suggestions for this week’s
homework**
Week 1 Assignment MUST SUBMIT IN WORD
Version (revision) date: August 4, 2021
Write your answer after the question (in this document)
Required Questions A and B added for this month
Your Name : Whitney Macaulay
PLEASE READ THE HOMEWORK SUGGESTIONS BEFORE STARTING THIS ASSIGNMENT
Confirm that you have used the Homework Suggestions provided by your instructor: True
A. Page 43: Give an example of skewed data. Is it positive or negative skew? What do these
terms mean?
In distributions that are skewed it means that the data tends to have a tail of data that is has more
higher values and the tails goes off to the right means that it is positively skewed. Opposingly,
when data has a tendency of higher values and the tail of data goes off to the left that means the
data set is negatively skewed. If the relationship is positively skewed , the mean is greater than
the mode and median. When a relationship is negatively skewed, the mean is less than the
median and mode.
Contrast a “skewed distribution” with an “approximately normal distribution.” A short written
answer in a short paragraph will be sufficient.
A “normal distribution” the mean and the median are the same number with the arch peaking in
the middle. In a skewed distribution, there tends to be outliers in the data that make the data
either skewed with a tail to either the right or left and making the mean different than the median
and mode. An example of a skewed distribution would be the age of deaths in a population. The
data is negatively skewed because the majority of people die later in life however there are still a
portion of the population that die at an early age.
Note: many of the biological / clinical variables in our course have “approximately normal”
distributions. Economic variables (income, wealth, etc.) are often skewed.
B. For the following two important healthcare variables, explain why you think they may be
approximately normal distributions, or skewed distributions:
a. Heart rate of patient upon admission to hospital
Normal – because there is an average range of heartbeats typically.
b. Length of stay for patients in the hospital
Downloaded by kuchy kuchy ()
, lOMoARcPSD|53991452
Positively skewed – because most patients tend to stay in the hospital for only a couple of
days or so but there are patients with more critical conditions that have to stay for months to
years at a time which is uncommon in relation to the entire population.
1. REAL WORLD: This published table summarizes the actual clinical trial results for a colon
cancer drug, panitumumab (Vectibix™). The experimental group is summarized in the middle
column, described as P + BSC, i.e., experimental drug (P = panitumab) plus best supportive care
(BSC). The far right column is the control group, who received the best supportive care (BSC)
without the experimental drug.
a. The endpoint used in many cancer drug trials is survival. The median is usually used, rather than
the mean, for the comparison in these survival trials. Why would the median be used, when we
“generally” use the mean to describe central tendency? Read chapter three (p. 43) in our
textbook. Look at the data table below: does the survival data appear to be skewed? How do
we know?
The reason that the median is usually used in trials like this is because if the data is skewed it
will cause a drastic shift in the mean due to one or a few outliers, making the mean less
meaningful of a statistic than the median. Yes, in the data set that is P+BSC the data appears to
be skewed because the mean of that column is much higher than the median showing us that
there could’ve been an outlier(s) that were accounted for in the mean, creating “skewed
distribution” and not a “normal distribution”.
b. The survival measurement below is PFS, progression-free survival. The table shows the mean
and median PFS for the experimental and control groups. Do you think that the sponsor of the
drug (the drug company) wanted the FDA to focus on comparing the means or the medians?
Given your answer to (a) above, which do you think better supported the drug’s approval, the
comparison of the means or the medians? Hints: look at the difference in means, compared with
the difference in medians. Notice that the means are also shown with the SD, standard
deviations. Why is this SD information relevant?
I think that the drug company probably would want to disclose the statistics with emphasis on the
median of the data because it has a better representation of the outcome of the use of the drug. The
reason that they would put the SD next to the mean is probably to show that even though in the first
column it looks like the mean was much higher the standard deviation of the data has more relevance to
the overall outcome of the use of the drug.
Downloaded by kuchy kuchy ()
Week1 - Assignment on Word Document
Intro to Biomedical Statistics (National University)
Scan to open on Studocu
Studocu is not sponsored or endorsed by any college or university
Downloaded by kuchy kuchy ()
, lOMoARcPSD|53991452
**I have carefully studied my instructor’s written homework suggestions for this week’s
homework**
Week 1 Assignment MUST SUBMIT IN WORD
Version (revision) date: August 4, 2021
Write your answer after the question (in this document)
Required Questions A and B added for this month
Your Name : Whitney Macaulay
PLEASE READ THE HOMEWORK SUGGESTIONS BEFORE STARTING THIS ASSIGNMENT
Confirm that you have used the Homework Suggestions provided by your instructor: True
A. Page 43: Give an example of skewed data. Is it positive or negative skew? What do these
terms mean?
In distributions that are skewed it means that the data tends to have a tail of data that is has more
higher values and the tails goes off to the right means that it is positively skewed. Opposingly,
when data has a tendency of higher values and the tail of data goes off to the left that means the
data set is negatively skewed. If the relationship is positively skewed , the mean is greater than
the mode and median. When a relationship is negatively skewed, the mean is less than the
median and mode.
Contrast a “skewed distribution” with an “approximately normal distribution.” A short written
answer in a short paragraph will be sufficient.
A “normal distribution” the mean and the median are the same number with the arch peaking in
the middle. In a skewed distribution, there tends to be outliers in the data that make the data
either skewed with a tail to either the right or left and making the mean different than the median
and mode. An example of a skewed distribution would be the age of deaths in a population. The
data is negatively skewed because the majority of people die later in life however there are still a
portion of the population that die at an early age.
Note: many of the biological / clinical variables in our course have “approximately normal”
distributions. Economic variables (income, wealth, etc.) are often skewed.
B. For the following two important healthcare variables, explain why you think they may be
approximately normal distributions, or skewed distributions:
a. Heart rate of patient upon admission to hospital
Normal – because there is an average range of heartbeats typically.
b. Length of stay for patients in the hospital
Downloaded by kuchy kuchy ()
, lOMoARcPSD|53991452
Positively skewed – because most patients tend to stay in the hospital for only a couple of
days or so but there are patients with more critical conditions that have to stay for months to
years at a time which is uncommon in relation to the entire population.
1. REAL WORLD: This published table summarizes the actual clinical trial results for a colon
cancer drug, panitumumab (Vectibix™). The experimental group is summarized in the middle
column, described as P + BSC, i.e., experimental drug (P = panitumab) plus best supportive care
(BSC). The far right column is the control group, who received the best supportive care (BSC)
without the experimental drug.
a. The endpoint used in many cancer drug trials is survival. The median is usually used, rather than
the mean, for the comparison in these survival trials. Why would the median be used, when we
“generally” use the mean to describe central tendency? Read chapter three (p. 43) in our
textbook. Look at the data table below: does the survival data appear to be skewed? How do
we know?
The reason that the median is usually used in trials like this is because if the data is skewed it
will cause a drastic shift in the mean due to one or a few outliers, making the mean less
meaningful of a statistic than the median. Yes, in the data set that is P+BSC the data appears to
be skewed because the mean of that column is much higher than the median showing us that
there could’ve been an outlier(s) that were accounted for in the mean, creating “skewed
distribution” and not a “normal distribution”.
b. The survival measurement below is PFS, progression-free survival. The table shows the mean
and median PFS for the experimental and control groups. Do you think that the sponsor of the
drug (the drug company) wanted the FDA to focus on comparing the means or the medians?
Given your answer to (a) above, which do you think better supported the drug’s approval, the
comparison of the means or the medians? Hints: look at the difference in means, compared with
the difference in medians. Notice that the means are also shown with the SD, standard
deviations. Why is this SD information relevant?
I think that the drug company probably would want to disclose the statistics with emphasis on the
median of the data because it has a better representation of the outcome of the use of the drug. The
reason that they would put the SD next to the mean is probably to show that even though in the first
column it looks like the mean was much higher the standard deviation of the data has more relevance to
the overall outcome of the use of the drug.
Downloaded by kuchy kuchy ()
