Scribd
Upload
964 views10 pages

Normality Test

1) The document outlines a normality test conducted on rainfall and temperature data from Brisbane, Australia over 30 years to determine if the data is normally distributed. 2) Statistical analysis was performed using SPSS, and the Shapiro-Wilk test results showed the rainfall data was not normally distributed while the temperature data was normally distributed. 3) While the analysis provided insight, the author recommends collecting over 50 years of data to allow use of the Kolmogorov-Smirnov test for a more comprehensive normality test analysis.

Uploaded by

dipen sapkota
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
964 views10 pages

Normality Test

1) The document outlines a normality test conducted on rainfall and temperature data from Brisbane, Australia over 30 years to determine if the data is normally distributed. 2) Statistical analysis was performed using SPSS, and the Shapiro-Wilk test results showed the rainfall data was not normally distributed while the temperature data was normally distributed. 3) While the analysis provided insight, the author recommends collecting over 50 years of data to allow use of the Kolmogorov-Smirnov test for a more comprehensive normality test analysis.

Uploaded by

dipen sapkota
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 10

Normality Test

SUBMITTED BY: DIPESH SAPKOTA


DUE DATE: 15 MAY 2021
Outline of Presentation
Introduction
Normality Test Method and Understanding of Normality Test
Test Procedure
Test Process in SPSS
Previous Assumption on Normality of Data
SPSS Output and Interpretation
Conclusion and Recommendations
Introduction
Normality Test is used to determine whether the sample data that has been drawn is from
normally distributed population.
This test can be performed mathematically or graphically.
The hypothesis used during the normality test are:
1) Null Hypothesis: Data follows the normal distribution
2) Alternative Hypothesis: Data doesn’t follow the normal distribution.

Objective of study:
•To determine whether the rainfall and temperature data are drawn from normally distributed
population.
Normality Test Method and
Understanding of Normality Test
Normality Test Method
For assessing whether data are normally distributed or not, we can carry out in two ways:
Using statistical test (Shapiro’s-Wilks test and Kolmogorov-Smirnov test)
Observing graphs(Q-Q plots).
Kolmogorov-Smirnov (KS) test is used for the larger amount of data whereas Shapiro-Wilk test is used for small
number of samples (n≤50) .
Understanding of Normality Test
If significant value>0.05, null hypothesis accepted which signifies data sets are normally distributed.
If significant value<0.05, alternative hypothesis accepted which signifies data sets are not normally distributed.
If the data points are near to the diagonal line in Q-Q plots, data are said to be normally distributed ,otherwise
not normally distributed.
Test Procedure
Rainfall and temperature data was collected from Australian bureau of meteorology by
selecting Peterson field of Brisbane.
An excel sheet was prepared for 30 years of data
Statistical analysis of collected data was performed using SPSS software
Test Process of Normality in SPSS
Normality Plots
Analyze Plots
with tests

Statistics
Descriptive (Confidence Continue
Statistics interval of mean
95% or other)

Explore Dependent List Okay


Previous Assumptions on Normality
of Data

Skewness =-0.434
Skewness =1.323
Kurtosis=-0.380, which shows in range of=+1 and -1 , so data
Kurtosis=1.285 ,which shows not in range of=+1 and -1,so are approximate normally distributed
data aren’t normally distributed
The distribution curve of ADMT shows nearly symmetric
The distribution curve of ADMR is skewed to the right distribution of data which depict that data are approximate
which indicates data are not normally distributed. normally distributed.
SPSS Output and Interpretation

In our case, total no of samples is 30. So, SW test will be applicable.
Shapiro-Wilk significant value (0.001) ≤0.05. So, alternative hypothesis accepted which signifies
data are not normally distributed for ADMR.
Similarly, SW value for ADMT is 0.529>0.05. So, null hypothesis accepted which signifies data are
normally distributed.
Q-Q Plot of ADMR and ADMT

The preceding ADMR diagram shows that many The preceding ADMT diagram shows that only a
points deviate from the diagonal line. As a result, few of the points deviate from the diagonal line. As
the data are not normally distributed. a result, the data are normally distributed.
So, normality test of ADMR matches with So, normality test of ADMT matches with
distribution of data as both shows data aren't distribution of data as both shows data are
normally distributed normally distributed.
Conclusion and Recommendation
Conclusion
Statistical data analysis of annual rainfall and temperature of Brisbane was carried out by using
SPSS program.
While analyzing the SPSS output, ADMR data are normally distributed whereas ADMT data are
not normally distributed.
Recommendations:
We can calculate more than 50 years of data in-order to have comprehensive result of the
normal distribution.
It is because if we employ more than 50 years of data, KS method will be applicable which will
provide more information about distribution of data .So, normality test will provide
comprehensive result.

You might also like