Quantitative Analysis

Assignment: Graphical Analysis (Chapter 2)

Question 5

What types of graphs can be used to display nominal or ordinal observations? Discrete or continuous
observations?

Nominal and ordinal observations are types of qualitative data which can be visualized graphically using
bar and pie charts.

Bar charts represent data in the form of vertical or horizontal bars and describe the relationship
between two or more variables in terms of frequency or percentage.

Pie chart is a circular representation of the percentage of occurrence of a variable. The area of
the pie represents the frequency or percentage of the said variable.

Discrete and continuous observations which are types of quantitative data can be represented
graphically using scatter plots, histograms and box plots

Scatter plot represents data points on a two-dimensional plane using a cartesian system. It
shows the relationship between two quantitative variables measured for one value.

Histograms are similar to bar graphs in that they represent data using bars whose heights
correspond to frequency. The difference is that histograms plot numerical data and are
continuous (with no gaps) representing frequency distribution.

The boxplot graphically represents the distribution of a quantitative variable by visually

Listed are the per capita healthcare expenditures in 1989 of 23 out of 24 nations that make up the
Organization for Economic Cooperation and Development [10]. The per capita expenditure of Turkey
was not available.

a) Rank these countries according to per capita health care expenditure.

https://drive.google.com/file/d/1t6bnaqQEHv4hN8tCsluP3BOwjpkcLr-k/view?usp=sharing

b) Construct a histogram for the values of per capita expenditure.

c) Describe the shape of the histogram.

The histogram has a normal distribution as seen by the bell curve.

The table below categorizes 10,614,000 office visits to cardiovascular disease specialists in the United
States by the duration of each visit [15]. A duration of 0 minutes implies that the patient did not have
face-to-face contact with the specialist.

The statement is made that office visits to cardiovascular disease specialists are most often between
16 and 30 minutes long. Do you agree with this statement? Why or why not?

Yes, office visits made to the cardiovascular disease specialists are often between 16-30 minutes long.
This can be verified by the data in the table given above. The most amount of visits i.e. 4431 thousand
visits lasted between 16 to 30 minutes. This can also be seen in the graph below:
Question 10

The frequency distribution table below displays the number of cases of pediatric AIDS reported in the
United States between 1983 and 1989 [9].

Construct a bar chart showing the number of cases by year. What does the graph tell you about
pediatric AIDS in this time period?

The bar chart shows a steady increase in the number of AIDS cases over time.

Question 11
Listed below are the numbers of people who have been executed in the United States in each year
since the 1976 Supreme Court decision allowing the death penalty to be carried out [16].

Use this data to create a bar chart of executions by year. How has the number of executions varied
since 1976?

Question 17
The percentages of low birth weight infants in various countries around the world are contained in the
data set unicef [22] (Appendix B, Table B.2).The measurements themselves are saved under the
variable name lowbwt.

a) Construct a box plot for percentages of low birth weight infants.

Case Processing Summary

b) Do the data appear to be skewed? If so, are they skewed to the left or right?

The data is skewed right as seen by the shorter whisker on the bottom of the box. This means the lower
birth rates have more similar values in the least positive quartile.

c) Does the data contain any outlying observations?

The outliers include data from India and Bangladesh. These birth weights lie outside of the upper 75% of
the data.

Question 18
The number of nursing home residents at least 65 years old per 1000 population 65 years of age and
over for each state in the United States are contained in the data set nurshome [23] (Appendix B,
Table B.3). The state names are saved under the variable name state and the numbers of nursing
homes residents per 1000 population under the variable name resident.

a) Which state has the smallest number of nursing home residents per 1000 population 65 years of
age and over? Which state has the largest number? What factors might influence the substantial
amount of variability among different states?

Hawaii has the smallest population (13.6) whilst South Dakota has the largest population (74.9) of
nursing home residents per 1000 population 65 years of age and over. This may be due to the weather,
accessibility, the quality and variety of different nursing home services offered, the cultural and family
values each state has with regards to family structure, the cost of nursing homes and the general
affordability of public population.

b) Construct a box plot for the number of nursing home residents per 1000 population.

c) Are the observations symmetric or skewed? Are there any states that could be considered outliers?

The observations are slightly negatively skewed as seen by the median being closer to the top of the box
i.e. the third quartile. There are no outliers seen in the box plot graph.

d) Display the number of nursing home residents per 1000 population using a histogram. Do you find
this graph to be more or less informative than the box plot?
Since there is a single data set, the frequency distribution is much more detailed as represented by the
histogram. The box plot however tells the overall symmetry of the data. The shape of the symmetry is
determined by the histogram

Question 19
The declared concentrations of tar and nicotine for 35 brands of Canadian cigarettes are stored in the
data set called cigarette [24] ( Appendix B, Table B.4). The concentrations of tar per cigarette in
milligrams are saved under the variable name tar and the corresponding concentrations of nicotine
under the name nicotine

a) Produce a one way scatter plot of the declared concentration of tar per cigarette. Be sure to identify
instances in which two or more measurements have the same value and therefore overlap.

b) Describe the distribution of the values.

c) Construct a two-way scatter plot of the concentration of tar versus the concentration of nicotine.
Label the axes appropriately.

d) Does there appear to be a relationship between these two quantities?

The relationship between tar and nicotine is seen to be strong, positive and linear in relationship as seen
by the closeness of the dots to the fit line. As the tar increases so does the nicotine in the cigarette.