NORMALITY TEST
LINDA AYU RIZKA PUTRI,S.K.M.,M.SC.
DATA SCALE REFRESHMENT
Normality test
• A test used to determine whether
sample data has been drawn from a
normally distributed population
• A number of statistical tests, such as
the Student's t-test and the one-way
and two-way ANOVA, require a
normally distributed sample population.
Selecting normality test
• Analytically
• Graphically
Kolmogorov Smirnov
• The Kolmogorov-Smirnov test is used to test the null hypothesis
that a set of data comes from a Normal distribution. The
Kolmogorov Smirnov test produces test statistics that are
used (along with a degrees of freedom parameter) to test
for normality.
• Kolmogorov–Smirnov test is used for n ≥50.
• large deviation has a low p-value. As a rule of thumb, we
reject the null hypothesis if p < 0.05. So if p < 0.05,
we don't believe that our variable follows a normal
distribution in our population.
Decision-making process in the normality test with
Kolmogorov-Smirnov
1.If the value Asymp.Sig. > 0.05, then the data is normally distributed research.
2.If the value Asymp.Sig. <0.05, then the research data is not normally distributed.
Performing KS in SPSS
Saphiro- Wilk
• The Shapiro–Wilk test is more appropriate method for small
sample sizes (<50 samples) although it can also be handling
on larger sample size while Kolmogorov–Smirnov test is used
for n ≥50.
• A Shapiro-Wilk test is the test to check the normality of the data.
The null hypothesis for Shapiro-Wilk test is that your data is
normal, and if the p-value of the test if less than 0.05, then
you reject the null hypothesis at 5% significance and
conclude that your data is non-normal.
Saphiro wilk in spss
Interpreting result
1.If the value Asymp.Sig. > 0.05, then the data is normally distributed research.
2.If the value Asymp.Sig. ≤ 0.05, then the research data is not normally distributed.
Normal probability plot
Normal probability plot
• The normal probability plot (Chambers et al., 1983) is a
graphical technique for assessing whether or not a data set
is approximately normally distributed
Histogram
Decision
In univariate
• Data follow normal distribution (mean ± SD)
• Data violated normal distribution (median (IQR)
Bivariate/multivariate
• Data follow normal distribution = parametric test
• Data violated normal distribution = non-parametric data
How to select appropriate test
Distribusi Statistik parametrik
normal
Uji asumsi Statistik non-
(normalitas, parametrik
independensi, dll)
Distribusi-log
Distribusi tidak tidak normal
normal
Log-
transformasi
Distribusi-log
normal
Statistik parametrik lalu
dieksponensial/anti-log
Question?
Skewness and kurtosis (next week, normal
plot in excel)
skewness
• Kemencengan atau kecondongan (skewness) adalah tingkat
ketidaksimetrisan atau kejauhan simetri dari sebuah distribusi.
• If skewness is less than -1 or greater than 1, the distribution is highly
skewed.
• If skewness is between -1 and -0.5 or between 0.5 and 1, the
distribution is moderately skewed.
• If skewness is between -0.5 and 0.5, the distribution is approximately
symmetric.
Kurtosis
• Kurtosis, keruncingan distribusi
data adalah derajat atau ukuran
tinggi rendahnya puncak suatu
distribusi data terhadap distribusi
normal data.
• When kurtosis is equal to
3, the distribution is
mesokurtic. This means the
kurtosis is the same as the
normal distribution, it is
mesokurtic (medium peak).
Confidence interval
A confidence interval is the mean of your estimate plus and
minus the variation in that estimate. This is the range of values
you expect your estimate to fall between if you redo your test,
within a certain level of confidence. Confidence, in statistics, is
another way to describe probability
• Jika confidence interval melewati nol maka tidak ada perbedaan
signifikan antara kelompok yang dibandingkan.
Misal CI = -2 – 0,5
Sebaliknya jika CI tidak mengandung nol, maka terdapat perbedaan
signifikan antara kelompok
Misal CI = 0,5 – 2,3
