Formulae For Mhrm510: Descriptive Statistics
Formulae For Mhrm510: Descriptive Statistics
Formulae For Mhrm510: Descriptive Statistics
FOR MHRM510
Descriptive Statistics
Mean for Ungrouped Data
1
x =
n
∑x
Variance For Ungrouped Data
1 2
S 2= ∑ x −x
n−1
Standard Deviation for Ungrouped Data
S=
1
n−1
∑ x −x
2
Median for Ungrouped Data
th
The median is the
n+1
2
observation if the numbers are arranged in either ascending or
discending order.
Mean for Grouped Data
1
x = ∑ fx
n
where: f represents frequency of the respective class
Variance for Gropued Data
1 2
S 2= ∑ f x−
x
n−1
Standard Deviation for Grouped Data
S=
1
n−1
∑ f x− x
2
Median for Grouped Data
Median=X med
n
c −F m−1
2
f med
where: Xmed = lower limit of the median class
c = class width
fmed = frequency of the median class
Fm1 = cumulative frequency before the median class
n = sample size
Mode for Gruped Data
c f m − f m−1
Mode=X mod
2f m − f m−1− f m+ 1
Xmod is the lower limit of the modal class
c is the class width
fm is the frequency of the modal class
fm1 is the frequency of the class preceding modal class
fm+1 is the frequency of the class just after the modal class
Correlation
Pearsonian Product Moment Correlation Coefficient (rp)
r p=
∑ x −x y−y
∑ x− x 2 ∑ y− y 2
Spearman's Rank Correlation Coefficient (rs)
6 ∑ d 2
r s =1−
n n 2 −1
Regression Coefficients
The general linear regression equation taking the form Y = a + bx has coeffients calculated as:
Slope of the regression equation
b=
∑ x− x y− y
2
∑ x −x
Intercept of the regression equation
a= y −b x
Standard Normal Distribution
The ZScore transformation that can be used for the calculation of probabilities, assuming the
validity of the central limit theorem is as follows:
x− μ
z=
δ
where: δ is the standard deviation and
μ is the mean.
Hypothesis Test for The Mean
Small sample case
The small sample case applies the tstatistic calculated as:
x −μ
t c=
s / n
The large sample case applies the zscore statistic calculated as:
x −μ
z c=
s / n
ChiSqaured Test
Expected count (E) is determined as:
rowtotal∗columntotal
E=
sample
The chisquared test statistic is:
2
2 ∑ O−E
X c=
E
where: O is the observed count and
E is the expected count