4 - The Shape of The Distribution
4 - The Shape of The Distribution
Skewness
Kurtosis
THE SHAPE
OF THE
DISTRIBUTION
Prepared by:
Danah A. Mama, RPm
SHAPE OF THE DISTRIBUTION
Skewness Magnitude
Approximately -0.5 to 0.5
Symmetrical
(Slight)
Moderately -1 to -0.5 (left)
Skewed 0.5 to 1 (right)
Significantly Less than -1 (left)
Asymmetrical Greater than 1 (right)
(Strong)
Descriptive Statistics
Sleep Hours
Valid 93
Missing 0
Mean 6.452
Std. Deviation 1.212
Skewness 0.125
Std. Error of Skewness 0.250
Kurtosis -0.444
Std. Error of Kurtosis 0.495
Shapiro-Wilk 0.961
P-value of Shapiro-Wilk 0.007
Minimum 4.000
Maximum 9.500
KURTOSIS
“Tailedness” or “Peakness”
Note:
Outliers among the low scores are defined as any score
which is smaller than the smallest score in the
interquartile range
Outliers among the high scores are defined as any score
which is bigger than the largest score in the interquartile
range
Calculate Interquartile Range
1. 120, 115, 65, 140, 122, 142, 125, 135, 122,
138, 144, 118
2. 65, 115, 118, 120, 122, 122, 125, 135, 138,
140, 142, 144
3. 120, 122, 122, 125, 135, 138
4. IQR = 138 – 120 = 16 x 1.5 = 24
5. 120 – 24 = 96
6. 138 + 24 = 162
7. Scores NOT between 96 and 162 are outliers!
8. Thus, 65 is an outlier.
Calculate Interquartile Range
1. Extreme outliers are identified in much
the same way but the interquartile range
is multiplied by 3 (rather than 1.5)
2. It would be usual practice to delete
outliers from your data.
3. You might also wish to compare the
outcome of the analysis with the
complete data and with outliers
excluded.
COMPARING MEASURES OF
CENTRAL TENDENCY
Differences among the measures occur with skewed
distributions
Pattern:
Mode will remain at the highest point in the
distribution
Median will be pulled slightly out into the skewed tail
Mean will be pulled the farthest out
So, mean is more sensitive to skew than the median or
mode, and in cases of extreme skew, the mean may no
longer be appropriate to use.
In media, median is usually reported to summarize the
center of skewed distributions
When to Use Each Measure
Mean: Best for normally distributed data
without outliers.
Median: Best for skewed distributions or
when outliers are present.
Mode: Useful for categorical data or to
identify the most common item.
Practical Applications in
Psychology
Personality Assessment
Mean: Used to determine the average score on
personality tests to assess where an individual falls
compared to the norm.
Median: Helpful in analyzing skewed distributions of
personality traits, as the median is less affected by
outliers.
Clinical Psychology
Mode: Identifies the most common symptoms or disorders
in a patient population, aiding in diagnosis and treatment
planning.
Mean: Used to track changes in symptom severity over
time, such as the average depression score before and
after therapy.
Practical Applications in
Psychology
Cognitive Psychology
Median: Helpful in analyzing reaction time
data, which is often skewed, to determine
the typical response speed.
Mode: Identifies the most frequent types
of errors or strategies used by
participants in cognitive tasks.
Practical Applications in
Psychology
Developmental Psychology
Median: Useful for determining milestones like the
median age of first steps or first words, as the
median is less influenced by early or late
developers.
Mode: Identifies the most frequent developmental
stages or behaviors observed in a sample of
children.
Social Psychology
Mean: Used to calculate the average number of
friends or social interactions for individuals in a
study.
Mode: Identifies the most common types of social
relationships or interactions observed.
Conclusion
Measures of central tendency provide
valuable insights into datasets.
Understanding when and how to use
mean, median, and mode is essential
for effective data analysis.