This document discusses variables related to human development across U.S. states. It analyzes the percentage of population with bachelor's degrees (%BA), percentage with master's degrees (%MA), urban population (%Urban), and percentage with long commutes (%LongCommute) in each state. For each variable, it provides summary statistics and finds that %BA and %MA are right-skewed, %Urban follows a normal distribution, and %LongCommute is right-skewed. It hypothesizes relationships between the variables and plans univariate and bivariate analyses.
4. Stage One: Univariate Analysis
Percentage population with a bachelor's degree (%BA)
Summary statistics
Column n Mean Std. dev. Median Range Min Max Q1 Q3
%BA 51 27.93333 5.723309 27.1 32.6 17.5 50.1 24.5 30.8
According to the data, the median
percentage of population with a bachelor’s
degree is 27.1% and the mean is about 28%.
41.1% of U.S. states contain populations
where 25% to 30% of people have at least a
bachelor’s level degree.
The distribution of the curve is positively
(right) skewed with most states having a
%BA population below the mean. About 70%
of the U.S. states have %BA populations
between 17.5% to 30%. Only about 30% of
U.S. states have %BA populations above
30%.
1 state, in particular, The District of
Columbia is an outlier in the data with 50.1%
of the population being bachelor degree level
holders and above. Excluding the outlier, the
data follows a normal distribution very
closely.
One standard deviation from the mean
is 22.2% and 33.7%. 84.3% of the data falls
within one standard deviation of the mean,
suggesting that the percentage of commute
times over 1 hour do not vary largely between
regions or states.
5. Percentage population with a master’s degree (%MA)
Summary statistics
Column n Mean Std. dev. Median Range Min Max Q1 Q3
%MA 51 10.323529 3.4059705 9.4 20.6 6.3 26.9 8.1 11.3
According to the data, the median
percentage of population with a master’s
degree is 9.4% and the mean is 10.3%. 51%
of U.S. states contain populations where
7.5% to 10% of people have at least a
master’s level degree.
The distribution of the curve is positively
(right) skewed with about 61% of states
having a %MA population below the mean.
More than half of U.S. states have %MA
populations between 7.5% to 10%. Only
about 38.5% of U.S. states have %MA
populations above 10%.
Again, 1 state, in particular, The District
of Columbia is an outlier in the data with
26.9% of the population being master degree
level holders and above. The data without the
outlier still shows a positive (right) skew.
One standard deviation from the mean
is 6.0% and 12.8%. 84.3% of the data falls
within one standard deviation of the mean,
suggesting that the percentage of commute
times over 1 hour do not vary largely between
regions or states.
7. Urban population (%Urban)
Column n Mean Std.
dev.
Median Range Min Max Q1 Q3
%Urban 51 74.104 14.887 74.2 61.3 38.7 100 64.8 87.9
According to the distribution of the data,
the median is an urban population of 74.2%,
while the mean is not far off, at 74.1%.
The data closely follows a normal
distribution. The only outlier for the data, (by
less than 1%) is The District of Columbia
(Washington, D.C.) with an urban population
of 100%. It does not make much of a
difference to the appearance of the
distribution when it has been removed, but
the mean and standard deviation decrease to
73.586% and 14.567%. If anything, the data
has a slight negative skew to the left.
One standard deviation from the mean
is 59.3% and 89.1%. Exactly 66.7% of the
data falls within one standard deviation of the
mean (34 out of 50 states and 1 district).
The fact that the data has a near normal distribution seems logical, given that it
accounts for very different regions of one country. The East Coast has largely urban
populations, as there is a higher population density in these states than in many
Western states. The distribution reflects this trend.
8. Percentage population with long commutes (%LongCommute)
Column n Mean Std.
dev.
Median Range Min Max Q1 Q3
%Long
Commute
51 6.806 2.851 5.8 14 2.6 16.6 4.9 8.3
The mean percentage of a state
population with commutes over 1 hour in
duration is 6.81%, while the median is 5.8%
of a state’s population.
The distribution is skewed positively to
the right. With the removal of the outliers
(Maryland, 13.9%, New Jersey, 13.9%, and
New York, 16.6%), the skew is far less
drastic. The mean decreases to 6.3%, while
the median decreases to 5.75%.
For this data set, one standard deviation
from the mean is 4.0% and 9.7%. 76.5% of
the data falls within one standard deviation of
the mean, suggesting that the percentage of
commute times over 1 hour do not vary
largely between regions or states. To further
research this data set, it may be useful to
separate states by the types of transportation
used for commutes, or the average distance
traveled to work.