Unit VIII Assessment: Nonparametric Tests and Data Visualization

This assignment delves into the realm of biostatistics and data visualization within public health research. It begins by comparing and contrasting two nonparametric tests, the Mann-Whitney U test and the Kruskal-Wallis H test, highlighting their applications and the scenarios where public health researchers employ them. The response also elucidates how these nonparametric tests differ from their parametric counterparts. Furthermore, the assignment underscores the significance of data visualization in public health, emphasizing its role in summarizing and interpreting large datasets. It then identifies and describes three key data visualization methods: line graphs, bar graphs, and pie charts, detailing their characteristics and suitability for representing different types of data. The assignment is supported by relevant scholarly references in APA format.

Unit VIII Assessment
1. Compare and contrast two nonparametric tests performed in biostatistical studies.
When do public health researchers use nonparametric tests? How do they compare
with parametric tests? Your response must be at least 350 words in length.
The distribution free statistical tests or non parametric statistical test are analysis based on
the models which do not follow any specific parameter or prior set assumptions (Edgington,
2015). Examples of non parametric tests include the Mann–Whitney U test, the Kruskal-Wallis H
tests and so on. Both the Mann-Whitney U and the Kruskal-Wallis H tests independent
parameters however the major difference between the two lie in the fact that Mann-Whitney U
can analyze data based on two independent data whereas the Kruskal-Wallis H can
accommodate multiple groups for statistical analysis (Vermeulen, Thas & Vansteelandt, 2015).
The Kruskal-Wallis H test can also be termed as an extension of the Mann-Whitney U test which
is applied for evaluating two or more than two independent samples of equivalent or diverse
sizes of samples (Cleophas & Zwinderman, 2016). As the Mann–Whitney U test efficient in
reporting the effect size of inferential tests it is extensively used by researchers for statistical
analysis. More parameters can be included in the Kruskal-Wallis H non parametric test thereby
increasing the required sample size for statistically accurate analysis of data.
The public health researchers use nonparametric tests as non parametric models require no
specific parameters and predictions of data being distributed normally as per prior postulation
(Edgington, 2015). As fewer assumptions are involved in tests with non parametric methods,
the applications of the non parametric methods are much more compared to the
corresponding parametric tests. The non parametric tests are also very sturdy and robust as it
does not rely much on pre-determined conjectures along with increased simplicity and ease of
usage compared to parametric tests.

The parametric tests presumes the primary statistical allotments present in the collected
data whereas non parametric method of testing does not depend on any such presumed
distribution of data and therefore can be utilized when parametric validity is not present in the
data. Parametric tests frequently have corresponding non parametric equivalent (Pohlert,
2016). In cases when both parametric and non parametric tests are both available, parametric
tests have more power to assert the relevance of the hypothesis whereas to achieve the same
confidence degree, a much bigger sample size would be required by the non parametric
method for statistically analyze the same study.
2. What is the importance of data visualization in public health research? List three
methods of data visualization, and identify their characteristics. Your response must be
at least 350 words in length.
The importance of data visualization in public health research is that it has the potential
of summarizing the entire data statistically, and representing the analyzed data visually. Public
health research generally involves data collection from large defined populations, however
recognizing relevant trends and patterns from such huge data set might be inconvenient (Ward,
Grinstein & Keim, 2015). Data visualization aims at simplifying the process and helps the
observer derive an analysis or conclusion from the visually represented information.
Line graphs, bar graphs and pie charts can be listed as different methods of data
visualization (Bree & Gallagher, 2016).
The line graph comprises of two axes, a horizontal axis and a vertical axis and each
variable is plotted along each axis. The most important characteristic of line graph is that it can
visually represent data of any given two variables. Line graphs are widely used to plot changes
in multiple groups over a short or long period of time. Even small changes can be highlighted
well in line graphs making it an essential tool for data representation.

Figure 1 shows data representation in line graph format. Here it can be seen the
changing number of lions at a national park over the recent years (arbitrary data used as
example).
2000 2005 2010 2015
0
100
200
300
400
500
600
700 Number of Lions
Axis Title
The bar graph is generally used to represent complex groups of data. The key attribute
of bar graphs compared to line graph that it can visually represent large differences or huge
amounts of data over a long period of time in a structured manner (Bree & Gallagher, 2016).
Bar graphs are also easy and convenient to compare different sets of collected data amongst
different groupings at one glance.
Figure 2. This data represented via bar chart reveals the time and the hours distributed
on laptop/desktop, smart phones and other smart gadgets (arbitrary data used as example).

2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
0.2
0.3
0.4
0.3
0.3
0.3
0.3
0.4
0.4
0.600000000000001
2.2
2.3
2.4
2.6
2.5
2.3
2.2
2.2
2.2
2.1
0.3
0.3
0.4
0.8
1.6
2.3
2.6
2.8
3.1
3.5
Time spent on smart devices
(hours/day)
Hours spent on other smart devices
Hours per day spent on laptop/desktop
Hours per day spent on smart phones
The pie chart is visual representation of statistically analyzed data in a circular chart
divided into portions to elucidate the proportion in numeric fractions in comparison to the
whole circle. The central angle along with the area and length of the arc of the divided
portion is directly proportional to the percentage of the data or the quantity it represents.
This allows the reader to gain a quick perceptive and understanding of the information
about the variable parts in comparison to the whole information (Bree & Gallagher, 2016).
Figure 3 shows data representation in pie chart format. Here it can be seen the varying
percentage of sales in a company over a period of one year (arbitrary data used as example)
59%23%
10% 9%
Sales
1st Qtr 2nd Qtr 3rd Qtr 4th Qtr

References
Bree, R. T., & Gallagher, G. (2016). Using Microsoft Excel to code and thematically analyse
qualitative data: a simple, cost-effective approach. AISHE-J: The All Ireland Journal of
Teaching and Learning in Higher Education, 8(2).
Cleophas, T. J., & Zwinderman, A. H. (2016). Non-parametric Tests for Three or More Samples
(Friedman and Kruskal-Wallis). In Clinical Data Analysis on a Pocket Calculator (pp. 193-
197). Springer, Cham.
Edgington, E. S. (2015). Nonparametric tests for single-case experiments. In Single-Case
Research Design and Analysis (Psychology Revivals) (pp. 145-170). Routledge.
Pohlert, T. (2016). Non-parametric trend tests and change-point detection. CC BY-ND, 4.
Vermeulen, K., Thas, O., & Vansteelandt, S. (2015). Increasing the power of the Mann‐Whitney
test in randomized experiments through flexible covariate adjustment. Statistics in
medicine, 34(6), 1012-1030.
Ward, M. O., Grinstein, G., & Keim, D. (2015). Interactive data visualization: foundations,
techniques, and applications. AK Peters/CRC Press.