FORMULAS SHEET FOR
BASIC STATISTICS
INTRODUCTION TO DATA
(1) Total Number of Classes (groups) = 1 + 3.322 Log (n)
where n is the total number of the observed points
𝐑𝐚𝐧𝐠𝐞
(2) Class Size (width/Interval)= h = 𝐍𝐮𝐦𝐛𝐞𝐫 𝐨𝐟 𝐂𝐥𝐚𝐬𝐬𝐞𝐬
where Range = (Maximum value) - (Minimum Value)
𝐋𝐨𝐰𝐞𝐫 𝐋𝐢𝐦𝐢𝐭+𝐔𝐩𝐩𝐞𝐫 𝐋𝐢𝐦𝐢𝐭
(3) Mid Point (Class Mark) = X = 𝟐
𝐂𝐥𝐚𝐬𝐬 𝐅𝐫𝐞𝐪𝐮𝐞𝐧𝐜𝐲
(4) Relative Frequency = 𝐓𝐨𝐭𝐚𝐥 𝐅𝐫𝐞𝐪𝐮𝐞𝐧𝐜𝐲 x 100
CENTRAL TENDENCY
For Un-Grouped Data
̅ = ∑x/n
(1) Arithmetic Mean = 𝐗
(2) Geometric Mean = 𝐗 ̅ = (πx)1/n
̅ = Anti Log [ ∑Log(x)/n ]
(Or) 𝐗
̅ = n/∑(1/x)
(3) Harmonic Mean = 𝐗
̃ = [(n + 1)/2]th value
(4) Median = 𝐗
̂ = The most repeated value in data
(5) Mode = 𝐗
QUARTILES
For Un-Grouped Data
First Quartile = Lower Quartile = Q 1 = [(n + 1)/4]th value
Second Quartile = Median Quartile = Q2 = 2[(n + 1)/4]th value
Third Quartile = Upper Quartile = Q3 = 3[(n + 1)/4]th value
General Formula = QN = N[(n + 1)/4]th value; where N = 1, 2, 3
DECILES
For Un-Grouped Data
General Formula = DN = N[(n + 1)/10]th value; where N = 1, 2, 3, ......., 9
PERCENTILES
For Un-Grouped Data
General Formula = PN = N[(n + 1)/100]th value; where N = 1, 2, 3, ......., 99
CENTRAL TENDENCY
For Grouped Data
̅ = ∑fx/∑f
(1) Arithmetic Mean = 𝐗
̅ = Anti Log [ ∑f Log(x)/∑f]
(2) Geometric Mean = 𝐗
̅ = ∑f/∑(f/x)
(3) Harmonic Mean = 𝐗
̃ = l + (h/f) [(∑f)/2 - C]
(4) Median = 𝐗
̂ = l + h[(fm - f1)/(2fm - f1 - f2)]
(5) Mode = 𝐗
QUARTILES
For Grouped Data
General Formula = QN = l + (h/f) [N(∑f)/4 - C]; where N = 1, 2, 3
DECILES
For Grouped Data
General Formula = DN = l + (h/f) [N(∑f)/10 - C]; where N = 1, 2, 3, ........, 9
PERCENTILES
For Grouped Data
General Formula = PN = l + (h/f) [N(∑f)/100 - C]; where N = 1, 2, 3,....., 99
Relationship between AM, GM and HM
AM ≥ GM ≥ HM
Empirical Relationship Between Mean, Median and Mode
Mean - Mode = 3(Mean - Median); when data is not symmetrical
Mean = Median = Mode ; when data is symmetrical
DIPERSION
Un-Grouped Data (Absolute Measurements)
(1) Range = (Maximum value) - (Minimum Value)
(2) Quartile Deviation = Q.D = [(Q3 - Q1)/ 2]
where Q1 = [(n + 1)/4]th value and Q3 = 3[(n + 1)/4]th value
(3) Mean Deviation (M.D)
(i) Mean Deviation About Mean
̅∣ / n] ; where X
M.D𝐗̅ = [ ∑ ∣ X - 𝐗 ̅ = ∑x/n
(ii) Mean Deviation About Median
̃∣ / n] ; where X
M.D𝐗̃ = [ ∑ ∣ X - 𝐗 ̃ = [(n + 1)/2]th value
(iii) Mean Deviation About Mode
̂∣ / n] ; where X
M.D𝐗̂ = [ ∑ ∣ X - 𝐗 ̂ = The most repeated value in data
̅)2/ n] (OR) Variance = ∑x2/n - (∑x/n)2
(4) Variance = [ ∑ (X - 𝐗
(5) Standard Deviation = S.D = √𝑽𝒂𝒓𝒊𝒂𝒏𝒄𝒆
Un-Grouped Data (Relative Measurements)
(1) Coefficient of Range = [(Max) - (Min) / (Max) + (Min) ] x 100
(2) Coefficient of Quartile Deviation = [(Q3 - Q1)/ (Q3 + Q1)] x 100
where Q1 = [(n + 1)/4]th value and Q3 = 3[(n + 1)/4]th value
(3) Coefficient of Mean Deviation
(i) Mean Deviation About Mean
̅ ] x 100
Coefficient of Mean Deviation About Mean = [ M.D𝐗̅ / 𝐗
where ̅X = ∑x/n
(ii) Mean Deviation About Median
̃ ] x 100
Coefficient of Mean Deviation About Median = [ M.D𝐗̃ /𝐗
where ̃X = [(n + 1)/2]th value
(iii) Mean Deviation About Mode
̂ ] x 100
Coefficient of Mean Deviation About Mode = [ M.D𝐗̂ / 𝐗
where X̂ = The most repeated value in data
̅] x 100
(4) Coefficient of Variation = CV = [S.D / 𝐗
where ̅
X = ∑x/n and S.D = √𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒
Grouped Data (Absolute Measurements)
(1) Range = (Upper limit of last class) - (Lower limit of first class)
(2) Quartile Deviation = Q.D = [(Q3 - Q1)/ 2]
where Q1 = l + (h/f) [ ∑f/4 - C ]and Q3 = l + (h/f) [3(∑f)/4 - C]
(3) Mean Deviation (M.D)
(i) Mean Deviation About Mean
̅∣ / ∑f ] ; where ̅
M.D𝐗̅ = [ ∑ f ∣ X - 𝐗 X = ∑fx/∑f
(ii) Mean Deviation About Median
̃∣ / ∑f ] ; where ̃
M.D𝐗̃ = [ ∑ f ∣ X - 𝐗 X = l + (h/f) [ ∑f/2 - C ]
(iii) Mean Deviation About Mode
̂∣ / ∑f ] ; where X
M.D𝐗̂ = [ ∑ f ∣ X - 𝐗 ̂ = l + h[(fm - f1)/(2fm - f1 - f2)]
̅)2/ ∑f ] (OR) Variance = ∑fx2/∑f - (∑fx/∑f)2
(4) Variance = [ ∑ f (X - 𝐗
(5) Standard Deviation = S.D = √𝑽𝒂𝒓𝒊𝒂𝒏𝒄𝒆
Grouped Data (Relative Measurements)
[(𝐔𝐩𝐩𝐞𝐫 𝐥𝐢𝐦𝐢𝐭 𝐨𝐟 𝐥𝐚𝐬𝐭 𝐜𝐥𝐚𝐬𝐬) − (𝐋𝐨𝐰𝐞𝐫 𝐥𝐢𝐦𝐢𝐭 𝐨𝐟 𝐟𝐢𝐫𝐬𝐭 𝐜𝐥𝐚𝐬𝐬)]
(1) Coefficient of Range = x 100
[(𝐔𝐩𝐩𝐞𝐫 𝐥𝐢𝐦𝐢𝐭 𝐨𝐟 𝐥𝐚𝐬𝐭 𝐜𝐥𝐚𝐬𝐬)+ (𝐋𝐨𝐰𝐞𝐫 𝐥𝐢𝐦𝐢𝐭 𝐨𝐟 𝐟𝐢𝐫𝐬𝐭 𝐜𝐥𝐚𝐬𝐬)]
(2) Coefficient of Quartile Deviation = [(Q3 - Q1)/ (Q3 + Q1)] x 100
where Q1 = l + (h/f) [ ∑f/4 - C ]and Q3 = l + (h/f) [3(∑f)/4 - C]
(3) Coefficient of Mean Deviation
(i) Mean Deviation About Mean
̅ ] x 100
Coefficient of Mean Deviation About Mean = [ M.D 𝐗̅ / 𝐗
where ̅X = ∑fx/∑f
(ii) Mean Deviation About Median
̃ ] x 100
Coefficient of Mean Deviation About Median = [ M.D 𝐗̃ /𝐗
where ̃X = l + (h/f) [ ∑f/2 - C ]
(iii) Mean Deviation About Mode
̂ ] x 100
Coefficient of Mean Deviation About Mode = [ M.D 𝐗̂ / 𝐗
where X̂ = l + h[(fm - f1)/(2fm - f1 - f2)]
̅] x 100
(4) Coefficient of Variation = CV = [S.D / 𝐗
where ̅
X = ∑fx/∑f and S.D = √𝑉𝑎𝑟𝑖𝑎𝑛𝑐𝑒
