8/18/2018

NORMAL
DISTRIBUTION

SUMMARY
Descriptive statistics can be used to summarize and describe a single
variable (aka, UNIvariate)
• Frequencies (counts) & Percentages
– Use with discrete (nominal/ordinal) data
• Levels, types, groupings, yes/no, Drug A vs. Drug B
• Means & Standard Deviations
– Use with continuous (interval/ratio) data
• Height, weight, cholesterol, scores on a test

1
 8/18/2018

SUMMARY
Frequencies and percentages can be computed for
discrete data
– Examples: Likert Scales (Strongly Disagree to Strongly Agree);
High School/Some College/College Graduate/Graduate School

60
50
40
30
20
10
0
Strongly Agree Disagree Strongly
Agree Disagree

SUMMARY
We can compute frequencies and percentages for
continuous data
–Examples: Temperature, Height, Weight

2
 8/18/2018

SUMMARY
Measures of central tendency and measures of dispersion are often computed with
continuous data.

• Measures of Central Tendency (aka, the “Middle Point”)

– Mean, Median, Mode
– If your frequency distribution shows outliers, you might want to use the median instead of the
mean

• Measures of Dispersion (aka, How “spread out” the data are)

― Variance, standard deviation, standard error of the mean
― Describe how “spread out” a distribution of scores is
― High numbers for variance and standard deviation may mean that scores are “all
over the place” and do not necessarily fall close to the mean

In research, means are usually presented along with standard deviations or standard errors.

The Normal Distribution

The distribution of continuous data (interval-ratio) often forms a “bell
shaped” curve.
– Many phenomena in life are normally distributed (age, height,
weight, IQ). Empirical Rule

3
 8/18/2018

THE NORMAL DISTRIBUTION

Important things to note:

The normal distribution is fully defined by two parameters:

its standard deviation and mean

The normal distribution is bell shaped and

symmetrical about the mean

Normal distributions range from minus infinity to plus infinity

1.7

STANDARD NORMAL DISTRIBUTION

• A normal distribution whose mean is zero and standard deviation is one is called
the standard normal distribution.

0
1

1.8

4
 8/18/2018

ROLE OF NORMALITY
• Many statistical methods require that the numeric variables we are working
with have an approximate normal distribution.
• For example, t-tests, F-tests, and regression analyses all require in some sense that
the numeric variables are approximately normally distributed.

Standardized normal
distribution with
empirical rule
percentages.

USING THE NORMAL TABLE

• What is P(Z > 1.6) ?
P(0 < Z < 1.6)

0 1.6

P(Z > 1.6) = .5 – P(0 < Z < 1.6)

1.10

5
 8/18/2018

USING THE NORMAL TABLE

• What is P(Z < -2.23) ?
P(0 < Z < 2.23)

P(Z < -2.23) P(Z > 2.23)

-2.23 0 2.23

P(Z < -2.23) = P(Z > 2.23)

1.11

USING THE NORMAL TABLE

• What is P(Z < 1.52) ?

P(Z < 0) = .5 P(0 < Z < 1.52)

0 1.52

P(Z < 1.52) = .5 + P(0 < Z < 1.52)

1.12

6
 8/18/2018

USING THE NORMAL TABLE

• What is P(0.9 < Z < 1.9) ?
P(0 < Z < 0.9)

P(0.9 < Z < 1.9)

0 0.9 1.9

P(0.9 < Z < 1.9)

= P(0 < Z < 1.9) – P(0 < Z < 0.9)

1.13

APPLICATION OF NORMAL DISTRIBUTION

• We can use the following function to convert any normal random variable to a
standard normal random variable (z-score).

Some advice: always

draw a picture!

1.14

7
 8/18/2018

APPLICATION

Example: The time required to build a computer is

normally distributed with a mean of 50 minutes
and a standard deviation of 10 minutes:
What is the probability that a computer is
assembled in a time between 45 and 60 minutes?
Algebraically speaking, what is P(45 < X < 60) ?

1.15

APPLICATION
• P(45 < X < 60) ?

…mean of 50 minutes and a

standard deviation of 10 minutes…

1.16

8
 8/18/2018

LATIHAN
Ketinggian semua pemain bola jaring sekolah di Malaysia
bertaburan secara normal dengan nilai min dan sisihan
piawai masing-masing adalah 179 cm dan 5 cm.
Kirakan kebarangkalian seorang pemain terpilih mempunyai :
a) Ketinggian kurang daripada 181 cm
b) Ketinggian melebihi 189
c) Ketinggian sekurang-kurangnya 177 cm
d) Ketinggian di antara 176 cm dan 181 cm

TOOLS FOR ASSESSING NORMALITY

 Histogram

 Boxplot Visual inspection

 Normal Quantile Plot

(also called Normal Probability Plot)

 Skewness and Kurtosis

 Statistical Tests - Goodness of Fit Tests

Shapiro-Wilk Test (n<50)
Kolmogorov-Smirnov Test (K-S)
Lilliefors corrected K-S test
Anderson-Darling Test

9
 8/18/2018

HISTOGRAM

• Look for a “bell-shape”. Severe skewness and/or outliers are indications of non-
normality.
• Histograms are not useful for small sample sizes as it is difficult to get a clear
picture of the distribution.

RR1

BOXPLOT

It is hard to detect normality using a box-plot. But, at the very least, look for
symmetry. Severe skewness and/or outliers are indications of non-normality.

10
Slide 20

RR1 Roslinda Rosli, 3/1/2017

 8/18/2018

NORMAL QUANTILE PLOT

(NORMAL PROBABILITY PLOT)

SKEWNESS AND KURTOSIS

• For both measures, a perfectly normal
distribution should return a score of 0 (Rose
et al., 2015)
– A positive skewness value indicates positive (right)
skew; a negative value indicates negative (left) skew.
The higher the absolute value, the greater the skew.
– A positive kurtosis value indicates positive kurtosis;
a negative one indicates negative kurtosis. The
higher the absolute value, the greater the kurtosis.

– Rule of thumb
• divide either score by its standard error, if the result
is greater than 1.96, data are not normal.

11
 8/18/2018

STATISTICAL TESTS
• Statistical tests for normality are more precise since actual probabilities are calculated.
• Tests for normality calculate the probability that the sample was drawn from a normal
population.
• The hypotheses used are:
Ho: The sample data are not significantly different than a normal population.
Ha: The sample data are significantly different than a normal population.

• So when testing for normality:

Probabilities > 0.05 mean the data are normal.
Probabilities < 0.05 mean the data are NOT normal.

SPSS OUTPUT

Shapiro-Wilk Test (n < 50)

Kolmogorov-Smirnov Test (n > 50 )
p-value (Sig.) must > 0.05 for a normal distribution

12
 8/18/2018

DEAL WITH NON-NORMALITY

 Option 1 is to leave you data non-normal and conduct the parametric tests
(best for slightly normal data).

 Option 2 is to leave your data non-normal and conduct the non parametric
tests.

 Option 3 is to conduct “robust” tests.

 Option 4 is to transform the data. Transforming you data involving using

mathematical formulas to modify the data into normality.

13