T34_Group_643
A: I know technically it wasn’t asked for in the assignment brief but include a proper title to add
realism to the report
B: Solid introduction, but I would give a bit of 1-2 sentence background to the pandemic too to sort
of introduce the topic a bit more.
C: One thing that stands out to me about this report is that you’re missing any visuals and tables.
It’s really important to include these so the reader gets an idea of what the data looks like! I see
you put them in an excel file, but please include them in the main report for the final submission.
Example tables you can use for summary statistics (you NEED these for the final report):
World
Africa
GDP
Asia
Europe
North
America
…
Mean
Median
Standard
deviation
Min
Max
…
And then for correlations, something like this:
Correlations between GDP and Obesity
Continent
Africa
Asia
Europe
…
Correlation
And remember to include the graphics like histograms and scatter plots!!! These will help a lot
with showing the reader what the data looks like, and to be honest most of the lost marks for you
will be because these were missing.
D: Some decent commentary but would be a lot better if you had the table of statistics and graphics
to refer to. I would also try to interpret this in the context of the data; i.e what does positive
skewness mean for the wealth of the world? Etc.
E: I would probably add headings to better break up where in the document you’re talking about
CFR, GDP, Obesity, etc. Visuals like histograms and tables can also serve as good separators.
-
Additionally, talk about the differences by continent when discussing these variables too. E.g
what continent has the highest mean GDP? Etc etc.
F: I wouldn’t say it’s that weak. Additionally, explain whether this result make sense or not. Would
you expect obesity/GDP to have a high correlation?
-
For the final report, definitely looking for more discussions and interpretations of the
statistics you find.
G: I’m actually not sure this is true. The correlation between CFR and Obesity is quite low. Further,
why not just look at the correlation coefficient to determine this?
H: No need to describe how you generated the statistics
I trust you are very capable.
I: Don’t usually talk about skewness in scatter plots; it’s more of a histogram thing.
J: Also not sure this is true; I believe the relationship in this dataset is low GDP > low obesity
(because of the positive correlation).
K: In the conclusion section, try to sum up more of your report and provide some recommendations
Overall:
-
-
Good job! A solid effort
For the final report please include the scatter plots and histograms with the main report! I
was a bit confused as to which plots you were referring to during your report.
o If you don’t put the plots etc in the final report, you will be penalized. I didn’t do it
for the draft, but just make sure you’re including them on the final report.
Try to include some more discussion of your statistics that you find. Do they make sense to
you? What do they imply?
I look forward to reading the final report
Running head: QUANTITATIVE ANALYSIS
1
QUANTITATIVE ANALYSIS
Name
Institution
Tutor
QUANTITATIVE ANALYSIS
2
Introduction
Quantitative methods allow the analysis of collected data to interpret the trends and
relationships between them using certain statistical tests. This analysis report has data from five
categories i.e., country, continent, the GDP per capita in USD, the percentage population
classified as obese, and case fatality rate due to covid 19.
Country Level Distribution of GDP per capita
From the statistical summaries carried out for the GDP as shown in table 1 below, the
data includes central tendency, dispersion, distribution properties, sum, and count. The generated
statistic of the country-level GDP indicated that the mean and median of GDP are $14792.48 and
$6072.2. This data is skewed to the right because value of skewness being greater than 0 and as a
result a good measure of central tendency would be done using median. This is because, for
skewed data distribution, the median is minimally affected even with extremities of some values.
The minimum and maximum GDP are $381.30 from Malawi and $81734.5 from
Luxembourg.
For measures of dispersion the standard deviation, range and sample variance are
$19330.27, $81353.2, and 3.74 respectively.
QUANTITATIVE ANALYSIS
3
Summary statistics of GDP per capita
Mean
Standard Error
Median
Mode
Standard Deviation
Sample Variance
Kurtosis
Skewness
Range
Minimum
Maximum
Sum
Count
14792.4813
1742.95153
6072.2
#N/A
19330.26757
373659244.2
2.258169831
1.758094744
81353.2
381.3
81734.5
1819475.2
123
Table 1:Summary Statistics of GDP per capita
Country Level distribution of rates of Obesity
From table 2, obesity has a mean percentage of 18.78% with a median and mode of
21.2% and 20.2% respectively. The minimum and maximum percentages of obesity recorded
were 2.1 % and 37.9% from Vietnam and Kuwait respectively. In this case, the data is skewed to
the left and the best way to measure central tendency would be to use median.
QUANTITATIVE ANALYSIS
4
Summary statistics of Obesity
Mean
Standard Error
Median
Mode
Standard Deviation
Sample Variance
Kurtosis
Skewness
Range
Minimum
Maximum
Sum
Count
18.78211
0.823354
21.2
20.2
9.131436
83.38312
-1.07573
-0.11193
35.8
2.1
37.9
2310.2
123
Table 2:Summary statistics of rates of obesity in different countries.
Relationship between GDP per capita and obesity rate.
To figure out the relationship between these two data groups, the data was represented
graphically using a scatter plot. From the scatter plot in fig 1 below, it is noticed that for very low
GDP, there was a high cluster of data collected for obesity but they were distributed between the
greater percentages of obesity and the lesser percentages. However, as the GDP per capita
increased, there was a high percentage of people with obesity rather than a low percentage. This
means that obesity rate is increased in high prosperity countries.
QUANTITATIVE ANALYSIS
5
Scatter plot of country level GDP against Obesity
40
35
Obesity
30
25
20
15
10
5
0
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
GDP per Capita
Figure 1:Scatter plot showing the relationship between GDP per Capita and Obesity rate
Relationship between GDP per capita and obesity rates
A correlation coefficient was calculated to find out how strongly the two variables relate
to each other; the correlation coefficient was found to be 0.4028. This shows a weak positive
correlation. Meaning that as the GDP increases, the percentage of people living with obesity
increases but the correlation is weak meaning this relationship has other variables. To investigate
the average rates of distribution of obesity and GDP per capita in each continent, the below
histograms were constructed.
QUANTITATIVE ANALYSIS
6
Average of GDP per capita by Continent
Average GDP per capita
40000
35000
30000
25000
20000
15000
10000
5000
0
Africa
Asia
Europe
North
America
Oceania
South
America
Continents
Figure 2: Average distribution of GDP in different continents
Average rate of obesity per continent
Average of obesity
30
25
20
15
10
Total
5
0
Africa
Asia
Europe
North
America
Oceania
South
America
Continents
Figure 3:Average distribution of Obesity in different continents
Variation of the relationship among continents
To investigate how the relationship between GDP and obesity rate varies among
continents, the correlations of the different continents were calculated and graphically
represented in figure 2. Oceania, north America and south America had the strongest positive
QUANTITATIVE ANALYSIS
7
correlation between the variables. Asia and Africa had weak positive correlations of 0.42 and
0.28 respectively. Lastly Europe had a weak negative correlation coefficient of -0.06 meaning
that as GDP increases, obesity rates increase.
Relationship between GDP per capita and
Obesity rates for different continents
1
0.8
0.6
0.4
0.2
0
Africa
-0.2
Asia
Europe
North
America
Ocenia
South
America
Figure 4: Histogram comparing relationship between GDP per capita and obesity rate in different
cotinents
Country Level Variations of COVID 19 Fatality Rates.
The mean, median, and mode for the COVID 19 mortality rates are 2.022, 1.7, and 1.3
respectively. The data is skewed to the right meaning more data falls on the right of the
distribution curve.
QUANTITATIVE ANALYSIS
8
Descriptive summary statistics of CFR
Mean
Standard Error
Median
Mode
Standard Deviation
Sample Variance
Kurtosis
Skewness
Range
Minimum
Maximum
Sum
Count
2.022764228
0.120628618
1.7
1.3
1.337836093
1.789805411
5.057993562
1.816339689
8.6
0
8.6
248.8
123
Table 3: Overall descriptive summary statistics of CFR for all GDP per capita
Because of the skewness of the data, it is best analyzed using the median. So, a summary
statistic of the data that falls above and below the median of the GDP were found as shown in
table 3 below:
CFR summary statistics of data below GDP median
Mean
Standard Error
Median
Mode
Standard Deviation
Sample Variance
Kurtosis
Skewness
Range
Minimum
Maximum
Sum
Count
2.2
0.159541693
1.9
1.5
1.246060459
1.552666667
1.162319144
1.16315905
5.6
0.5
6.1
134.2
61
CFR summary statistics of data above
GDP median
Mean
Standard Error
Median
Mode
Standard Deviation
Sample Variance
Kurtosis
Skewness
Range
Minimum
Maximum
Sum
Count
Table 4: CFR summary statistics of data below and above GDP median
1.848387097
0.179159297
1.6
2.5
1.410701713
1.990079323
8.772546224
2.44579067
8.6
0
8.6
114.6
62
QUANTITATIVE ANALYSIS
9
Both tables show that all the data on COVID 19 fatality rates both below and above the median
of the GDP per capita in a country level is skewed to the right and the range between the smallest
value and the largest is greater below the GDP.
CFR
Scatter plot of CFR against GDP per Capita
10
9
8
7
6
5
4
3
2
1
0
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
Gdp Per Capita
Figure 5: Scatter plot of CFR against GDP per capita on a country level
From the data, a relationship between COVID 19 fatality rates and obesity in each
country was observed and there were more fatalities for higher percentages of obesity in a
particular country. A histogram of the same data further highlighted that the data was bimodal.
QUANTITATIVE ANALYSIS
10
CFR
Scatter plot of CFR against percentage of obesity
in a country
10
9
8
7
6
5
4
3
2
1
0
0
5
10
15
20
25
30
35
40
Obesity
Figure 6: Scatter plot of CFR against obesity rates on a country level
Figure 7: A histogram of CFR against obesity
Variations per continent
From the histogram, South America, North America and Europe had medians above the
CFR of all the 123 countries combined while Africa, Oceania and Asia were below that value.
South America had the highest median with 2.5, followed by north America with 2.3 and Europe
with 1.8. Africa, Asia and Oceania had 1.6, 1.3 and 1.1 respectively. The data is shown below:
QUANTITATIVE ANALYSIS
11
CFR variations per continent
3
Median of CFR
2.5
2
1.5
1
0.5
0
Africa
Asia
Europe
North
America
Oceania
South
America
Continents
Figure 8: Histogram of CFR variations per continent
To show the relationship between the pre-existing conditions and Covid 19 fatality rates a
histogram of correlation coefficients was constructed for each continent. In the figure below it
can be seen that only Oceania has a positive correlation of 0.7 between GDP and CFR while the
rest have a negative correlation.
Correlation between GDP and CFR
0.8
0.6
Correlation
0.4
0.2
0
Africa
-0.2
Asia
Europe
North
America
Ocenia
-0.4
-0.6
Continents
Figure 9: Correlation between prosperity and CFR across different continents
South
America
QUANTITATIVE ANALYSIS
12
From the correlation plot below, Africa, north America and Oceania had positive
correlation coefficients of 0.15, 0.06 and 0.34 respectively. These positive correlations are weak.
Asia, Europe, and South America has negative correlations of -0.252873016, -0.172106188 and 0.489511111 respectively.
Correlation between Obesity and CFR
0.4
Correlation coefficient
0.3
0.2
0.1
0
-0.1
-0.2
Africa
Asia
Europe
North
America
Ocenia
South
America
-0.3
-0.4
-0.5
-0.6
Continents
Figure 10:Correlation between obesity and CFR across different continents
Conclusion
With all the analysis done above, it is recommended that more comprehensive tools are
used to look determine the relationships between each data category and the variables that can
affect these relationships which are not accounted for.
Tips for the final submission:
All described in the draft submission plus,
Include a section with inferential statistical analyse (hypothesis tests, confidence
intervals) to guide your investigation.
Note that you are not allowed to make significant changes to the topic of your report.
Slight changes to the topic are acceptable however with approval from your tutor.
The guidelines below integrate many writing instructions to shows how they all fit together.
Note that these guidelines only explain you the expectations. You should pay attention to the
structure of a written report and decide how you would like your report divided into sections.
Each section should serve a distinct purpose.
Purchase answer to see full
attachment