Elementary Statistics Assignment PDF

Which of the following is not a measure of variability?
Mode
Standard deviation
All of the answers on this list are measures of
variability!
Range
A scatter plot or scatter chart can be used to identify the association between variables.
According to Dr. Coy in the lecture video, if we can approximate the relationship with a line,
then there is probably a linear relationship between the two variables.
True
False
Unfortunately, if we wish to compute a table of descriptive statistics, we are forced to
calculate all of our statistics using individual Excel formulas. There are no shortcuts!
True
False
The confidence level in the descriptive statistics output tells us how much to add and
subtract from the mean to obtain the 95% confidence interval of the mean.
True
False
Match the clues with the answers at right.
A single
value of
each
variable
at a
given
time,
which
often
appears
as one
data row
in your
A
.
Bins
B
.
Normal
C
.
Time series
D
.
Sample
mean
E. Observation
F. Hypothesis
E.

spreadsh
eet.
Grouping
s used to
determin
e the
intervals
in a
frequenc
y
distributi
on.
An
example
is the
record of
closing
stock
prices for
Exxon for
the last
three
years.
A point
estimate
Comparin
g P-value
to the
research
alpha
This
distributi
on has no
skew
Test
Skew has to do with the direction of the longer tail of the distribution. If the skew is positive,
then the longer tail points to the right. For symmetrical distributions like the normal
distribution, the tails are identically sized and have, therefore, no skew.
True
False
It is possible that two variables could have a strong non-linear correlation but have a linear
correlation close to zero.
True
A.
C.
D.
F.
B.

False
if the data exhibits a bell-shaped curve, roughly 68% of the observations will be within
_____________.
3 standard deviations of the mean.
2 standard deviations of the mean.
1 standard deviation of the mean.
1.5 standard deviation of the mean.
I know that this question is about the empirical rule, but none of the answers
on this list are correct!
Histograms are used to learn about the properties of a distribution. These properties can
include shape, skew and multiple modes (e.g., bimodal distribution).
True
False
Variance and standard deviation measure the same attribute of a distribution.
True
False
although correlations close to 1 or -1 are obviously strong, determining whether or not the
correlation between two variables is strong is context specific and somewhat subjectively
determined.
True
False
Which of the following correlation statistics is the strongest?
R
= -
12
%
R
=
41
%
R
= -
91

%
R
=
29
%
In a study of countries where data are available, the homicides by firearm rates were
compared with the number of guns per capita. Of 186 countries in the data set, 107
countries had complete data. So, only these countries were available for analysis.
When using the entire list of 107 countries, the correlation between homicide by
firearm rate and guns per capita was R=10%.
Researchers argued that this was probably not the best approach since law
enforcement in developing and developed countries is very different. So, they
restricted the dataset to developing countries. The resulting dataset had 27
countries.After running their analysis, they found that the correlation between
homicide by firearm rate and guns per capita was R =2%. See the graph below.

Neither correlation is particularly interesting. But the researchers took a look at the
graph and argued that South Africa is a distinct outlier that should be removed
because it has unique characteristics related to its history and current socio-political
environment, which makes it distinctively different than the rest of the countries in
the dataset. When the correlation is run again without South Africa, they found that it
is increased to slightly above 71%. See the graph below. Based on this analysis,
they conclude that an increased number of guns per capita leads to an increased
number of firearm-related homicides, which is evidence of the benefit of more
restrictive gun control in countries similar to the U.S.

In 200 words or less, discuss what this research tells us about the relationship
between the homicide by firearm rate and guns per capita. Hint: This is NOT a
political question nor is this a question about your opinion regarding the possible
linkage between the number of firearms in the US and firearm-related homicides.
Instead, focus on the correlation statistics, correlation theory, the approach they
used to achieve their results, and your understanding of the problem context.
Answer: The above graph clearly highlights that there is a positive relationship between firearm rate
and guns per capita for the developed countries which is apparent from the positive slope of the best
fit line which is apparent when all the developed nations barring South Africa is taken. Clearly, South
Africa is an outlier and anomaly and should not be considered for the purposes of this study as it
tends to have a distortion on the overall results. The correlation coefficient for the above result tends
to exceed 0.7. This is indicative of the fact that strong correlation tends to exist between the guns per
capita and the homicide by firearm rate. The high correlation is also apparent from the scatter plot
where the deviations of the individual observations from the best fit line seem limited only. As a
result, it makes sense to control the availability of guns which can potentially lead to lower homicides
caused by firearm in nations such as USA.