Scribd
Upload
68 views

STAT - Measures of Shape

This document discusses various statistical tests that can be used to determine if a distribution is normal, including measures of skewness and kurtosis, z-scores, the Shapiro-Wilk test, Q-Q plots, and box plots. Skewness measures symmetry and whether the mean or median is higher/lower, while kurtosis measures peakedness. Z-scores of skewness and kurtosis below ±1.96 for samples under 50 indicate normality. The Shapiro-Wilk test compares data to a normal distribution, with p>0.05 indicating normality. Q-Q plots show if points fall in a straight line, and box plots examine median and quartiles. Passing

Uploaded by

Jessica Nathania Gagua
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
68 views

STAT - Measures of Shape

This document discusses various statistical tests that can be used to determine if a distribution is normal, including measures of skewness and kurtosis, z-scores, the Shapiro-Wilk test, Q-Q plots, and box plots. Skewness measures symmetry and whether the mean or median is higher/lower, while kurtosis measures peakedness. Z-scores of skewness and kurtosis below ±1.96 for samples under 50 indicate normality. The Shapiro-Wilk test compares data to a normal distribution, with p>0.05 indicating normality. Q-Q plots show if points fall in a straight line, and box plots examine median and quartiles. Passing

Uploaded by

Jessica Nathania Gagua
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 5

DESCRIPTIVE STATISTICS: MEASURES OF SHAPE AND OTHER TESTS OF NORMALITY

Prepared by: Ms. Maridette Joyce D. Maranan, M.A.

Skewness – Absence of symmetry


– Extreme values in one side of a distribution

Symmetric = mean and median are equal


Negatively Skewed = mean is lower than the median
Positively Skewed = mean is higher than the median

Coefficient of Skewness
Summary measure for skewness

3   Md 
S

If S < 0, the distribution is negatively skewed (skewed to the left).


If S = 0, the distribution is symmetric (not skewed).
If S > 0, the distribution is positively skewed (skewed to the right).
Kurtosis - Peakedness of a distribution

Leptokurtic: high and thin


Mesokurtic: normal in shape
Platykurtic: flat and spread out

When can we say Compute for the z-score: Divide each value by its standard error (Std. Error).
that a distribution is
normal using Rule of thumb:
Skewness and
Kurtosis? For small samples (n < 50): If both values are greater than ±1.96, then data is not normal.
The value should be somewhere in the span of -1.96 and +1.96 for data to be considered as
normally distributed.

For medium to large sample size (n = 50 to 300): ): If both values are greater than ±2.58 or
±3.29, then data is not normal.

Example:

Case Processing Summary

Cases
Valid Missing Total
N Percent N Percent N Percent

Motivation 15 100.0% 0 0.0% 15 100.0%


Descriptives

Statistic Std. Error


Motivation Mean 3.0333 .24624

95% Confidence Interval for Lower Bound 2.5052


Mean Upper Bound 3.5615
5% Trimmed Mean 3.0370
Median 3.0000
Variance .910

Std. Deviation .95369


Minimum 1.00
Maximum 5.00
Range 4.00
Interquartile Range 1.00

Skewness -.109 .580


Kurtosis 1.047 1.121

z-scoreSKEWNESS = -.109/.580 = -0.19


z-scoreKURTOSIS = 1.047/1.121= 0.94
Since n = 15, we will compare the z-score to ±1.96.
Both values are lower than ±1.96 so the data is normally distributed.

Shapiro-Wilk Test Rule of Thumb:

sig. value > 0.05 = Normal


sig. value < 0.05 = Not Normal

***Note: Shapiro-Wilk Test is not recommended for large sample size. If it will be used
with large sample size, make sure to also check graphical presentations (e.g. histogram)

Q-Q Plots

If points converge (or fall


approximately) in a straight
line, then data is normal.

The picture shows a normal


distribution despite the points
on either sides are gearing
away from the line (unless a lot
of points are plotted too far
from the line, it may be
considered as not normal).
Important to look at: Do most
of the points in the middle
converge in the line?
Box and Whisker
Plots (also known as
Box Plots)

If the answer to all three questions is yes, your data is normally distributed.

References: Gravetter, L. & Wallnau, L. (2009). Statistics for the behavioral sciences (8th ed.). Singapore:
Wadsworth/Cengage Learning; Pagano, R.R. (2010). Understanding Statistics in the Behavioral Sciences (9th ed.).
Belmont, CA:Wadsworth/Cengage Learning; Broto, A.S. (2006). Statistics Made Simple (2nd ed.). Mandaluyong,
Philippines: National Book Store; Tan, W. (2007). Lecture notes in Statistics; Lu, G. (2015). Lecture notes in Advaced
Educational Statistics. http://documents.routledge-interactive.s3.amazonaws.com/9780415628129/Chapter%2013%20-
%20Tests%20for%20the%20assumption%20that%20a%20variable%20is%20normally%20distributed
%20final_edited.pdf; http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3591587/; youtube.com

You might also like