Professional Documents
Culture Documents
DESCRIPTIVE STATISTICS
MEASURES OF CENTRAL TENDENCY
AND OTHER POSITION
COMPUTATIONAL VARIATIONS
► ungrouped data
► grouped data
MEAN (x)
► the most frequently used measure of central tendency
► defined as the arithmetic average of all scores or group
of scores in a distribution
where: MD = median
x = score
n = sample size
Solution:
A. MEAN
1.8 + 2.1 + 1.7 + 1.6 + 0.9 + 2.7 + 1.8
x=
7
𝐱 = 𝟏. 𝟖
COMPUTATION (UNGROUPED DATA)
Example: A food inspector examined a random sample
of 7 cans of certain brands of tuna to determine the
percent of foreign impurities. The following
percentages were recorded: 1.8, 2.1, 1.7, 1.6, 0.9,
2.7 and 1.8. Compute for the sample mean, median
and mode.
Solution:
B. MEDIAN
ARRAY: 0.9 1.6 1.7 1.8 1.8 2.1 2.7
MD = 1.8
COMPUTATION (UNGROUPED DATA)
Example: A food inspector examined a random sample
of 7 cans of certain brands of tuna to determine the
percent of foreign impurities. The following
percentages were recorded: 1.8, 2.1, 1.7, 1.6, 0.9,
2.7 and 1.8. Compute for the sample mean, median
and mode.
Solution:
C. MODE
Since 1.8 has 2 frequencies, therefore, MO = 1.8
COMPUTATION (UNGROUPED DATA)
Exercise: An article in the Journal of Structural
Engineering (Vol. 115, 1989) describes an experiment
to test the yield strength of circular tubes with caps
welded to the ends. The first yields (in kN) are as
follows: 140, 96, 164, 170, 102, 160, 128, 156,
102, 128, 126, 108, 128, 102, 160, 104 and 96.
Calculate the sample mean, median and mode.
Answer: 𝐱 = 127.65
MD = 128
MO = 102, 128
COMPUTATION (GROUPED DATA)
MEAN
a. LONG METHOD
𝐤
𝐢=𝟏 𝐗𝐢 𝐟𝐢
𝐱=
𝐧
where: x = arithmetic mean
Xi = class mark (class midpoint)
fi = frequency of each class
k = no. of class intervals
n = sample size
COMPUTATION (GROUPED DATA)
MEAN
b. DEVIATION METHOD
𝐤
𝐢=𝟏 𝐟𝐢 𝐃𝐢
𝐱=𝐀+
𝐧
where: x = arithmetic mean
A = assumed mean or class mark of the mean
class
fi = frequency of each class
Di = deviation from the assumed mean
n = sample size
COMPUTATION (GROUPED DATA)
MEAN
c. CODING METHOD
𝐤
𝐢=𝟏 𝐟𝐢𝐮𝐢
𝐱 = 𝐀+𝐜
𝐧
where: x = arithmetic mean
A = assumed mean or class mark of the mean
class
fi = frequency of each class
ui = code
c = class size
n = sample size
COMPUTATION (GROUPED DATA)
Example: Find the arithmetic mean of the distribution
using the three methods.
CLASS INTERVAL f
25-29 5
30-34 4
35-39 6
40-44 9
45-49 7
50-54 6
55-59 3
COMPUTATION (GROUPED DATA)
Solution:
CLASS
f X fX
INTERVAL
25-29 5 27 135
30-34 4 32 128
35-39 6 37 222
40-44 9 42 378
45-49 7 47 329
50-54 6 52 312
55-59 3 57 171
∑ 40 1675
COMPUTATION (GROUPED DATA)
Solution:
a. LONG METHOD
𝐤
𝐢=𝟏 𝐗𝐢 𝐟𝐢 𝟏𝟔𝟕𝟓
𝐱= = = 𝟒𝟏. 𝟖𝟖
𝐧 𝟒𝟎
COMPUTATION (GROUPED DATA)
Solution:
CLASS
f X fX D fD
INTERVAL
25-29 5 27 135 -15 -75
30-34 4 32 128 -10 -40
35-39 6 37 222 -5 -30
40-44 9 42 378 0 0
45-49 7 47 329 5 35
50-54 6 52 312 10 60
55-59 3 57 171 15 45
∑ 40 1675 -5
COMPUTATION (GROUPED DATA)
Solution:
a. LONG METHOD
𝐤
𝐢=𝟏 𝐗𝐢 𝐟𝐢 𝟏𝟔𝟕𝟓
𝐱= = = 𝟒𝟏. 𝟖𝟖
𝐧 𝟒𝟎
b. DEVATION METHOD
𝐤
𝐢=𝟏 𝐟𝐢𝐃𝐢 −𝟓
𝐱=𝐀+ = 𝟒𝟐 + = 𝟒𝟏. 𝟖𝟖
𝐧 𝟒𝟎
COMPUTATION (GROUPED DATA)
Solution:
CLASS
f X fX D fD U fU
INTERVAL
25-29 5 27 135 -15 -75 -3 -15
30-34 4 32 128 -10 -40 -2 -8
35-39 6 37 222 -5 -30 -1 -6
40-44 9 42 378 0 0 0 0
45-49 7 47 329 5 35 1 7
50-54 6 52 312 10 60 2 12
55-59 3 57 171 15 45 3 9
∑ 40 1675 -5 -1
COMPUTATION (GROUPED DATA)
Solution:
c. CODING METHOD
𝐤
𝐢=𝟏 𝐟𝐢𝐮𝐢 −𝟏
𝐱 = 𝐀+𝐜 = 𝟒𝟐 + (𝟓)( ) = 𝟒𝟏. 𝟖𝟖
𝐧 𝟒𝟎
COMPUTATION (GROUPED DATA)
MEDIAN
𝐧
− < 𝐜𝐟𝐛
𝐌𝐃 = 𝐋𝐌𝐃 + 𝐜 𝟐
𝐟𝐌𝐃
where: MD = median
LMD = lower class boundary of median class
c = class size
n = sample size
<cfb = less than cumulative frequency of the
class PRECEEDING (before) the median
class
fMD = frequency of the median class
COMPUTATION (GROUPED DATA)
MODE
𝐟𝐌𝐎 −𝐟𝟏
𝐌𝐎 = 𝐋𝐌𝐎 + 𝐜
𝟐𝐟𝐌𝐎−𝐟𝟏−𝐟𝟐
where: MO = mode
LMO = lower class boundary of the modal class
c = class size
fMO = frequency of the modal class
f1 = frequency PRECEEDING (before) the
modal class
f2 = frequency FOLLOWING (after) the
modal class
COMPUTATION (GROUPED DATA)
Example: Find the MEDIAN and the MODE of the
following distribution.
CLASS INTERVAL f
25-29 5
30-34 4
35-39 6
40-44 9
45-49 7
50-54 6
55-59 3
∑ 40
COMPUTATION (GROUPED DATA)
Solution: n/2 = 40/2 = 20
CLASS
INTERVAL
f <cf Median Class = 40-44
25-29 5 5
LMD = 39.5
<cfb = 15
30-34 4 9
fMD = 9
35-39 6 15
15 c=5
40-44 9 24 𝐧
− < 𝐜𝐟𝐛
45-49 7 31 𝐌𝐃 = 𝐋𝐌𝐃 + 𝐜 𝟐
50-54 6 37 𝐟𝐌𝐃
55-59 3 40 𝟒𝟎
− 𝟏𝟓
𝟐
∑ 40 = 𝟑𝟗. 𝟓 + 𝟓
𝟗
𝐌𝐃 = 𝟒𝟐. 𝟐𝟖
COMPUTATION (GROUPED DATA)
Solution: Modal Class = 40-44
CLASS
f
LMO = 39.5
INTERVAL fMO = 9
25-29 5 f1 = 6
30-34 4 f2 = 7
35-39 6 c=5
40-44 9 𝐟𝐌𝐎 − 𝐟𝟏
𝐌O = 𝐋𝐌𝐎 + 𝐜
45-49 7 𝟐𝐟𝐌𝐎 − 𝐟𝟏 − 𝐟𝟐
50-54 6 𝟗−𝟔
55-59 3 = 𝟑𝟗. 𝟓 + 𝟓
𝟐 𝟗 −𝟔 −𝟕
∑ 40 𝐌𝐎 = 𝟒𝟐. 𝟓𝟎
WEIGHTED MEAN (x)
► kind of average in which instead of each data point
contributing equally to the final mean, some data
points contribute more “weight” than others
► a numerical multiplier is multiplied to the variable in order
to indicate the relative importance
► enables to calculate the average that takes into account
the importance of each value to the overall total
𝐧
𝐢=𝟏 𝛚𝐢 𝐱𝐢
𝐱= 𝐧
𝐢=𝟏 𝛚𝐢
Solution:
Σ𝜔𝑥 .4 80 + .4 80 + .2 95
x= = = 83
Σ𝜔 .4 + .4 + .2
FRACTILES or QUANTILES
► measures of location that describes or locates the position
of certain pieces of data relative to the entire set of
data
► unlike median that divides the distribution to 2 equal
parts, fractiles or quantiles divide a given distribution
into N equal parts
► applicable to grouped data
PERCENTILES P10 P20 P30 P40 P50 P60 P70 P80 P90 P100
DECILES D1 D2 D3 D4 D5 D6 D7 D8 D9 D10
QUARTILES Q1 Q2 Q3 Q4
MEDIAN MD
WHOLE
DISTRIBUTION
FRACTILES or QUANTILES
𝐧
− <𝐜𝐟𝐛
MEDIAN 𝐌𝐃 = 𝐋𝐌𝐃 + 𝐜 𝟐
𝐟𝐌𝐃
𝐤𝐧
− <𝐜𝐟𝐛
QUARTILES 𝐐𝐤 = 𝐋𝐐 + 𝐜 𝟒
𝐟𝐐
𝐤𝐧
− <𝐜𝐟𝐛
DECILES 𝐃𝐤 = 𝐋𝐃 + 𝐜 𝟏𝟎
𝐟𝐃
𝐤𝐧
− <𝐜𝐟𝐛
PERCENTILES 𝐏𝐤 = 𝐋𝐏 + 𝐜 𝟏𝟎𝟎
𝐟𝐏
FRACTILES or QUANTILES
𝐤𝐧
− <𝐜𝐟𝐛
QUARTILES 𝐐𝐤 = 𝐋𝐐 + 𝐜 𝟒
𝐟𝐐
𝐤𝐧
− <𝐜𝐟𝐛
DECILES 𝐃𝐤 = 𝐋𝐃 + 𝐜 𝟏𝟎
𝐟𝐃
𝐤𝐧
− <𝐜𝐟𝐛
PERCENTILES 𝐏𝐤 = 𝐋𝐏 + 𝐜 𝟏𝟎𝟎
𝐟𝐏
CLASS INTERVAL f
25-29 5
30-34 4
35-39 6
40-44 9
45-49 7
50-54 6
55-59 3
∑ 40
FRACTILES or QUANTILES
Solution: kn/4 = 1(40)/4 = 10
CLASS
INTERVAL
f <cf Q1 Class = 35-39
LQ = 34.5
25-29 5 5
<cfb = 9
30-34 4 9 fQ = 6
35-39 6 15 c=5
40-44 9 24 𝐤𝐧
− < 𝐜𝐟𝐛
45-49 7 31 𝐐𝟏 = 𝐋𝐐 + 𝐜 𝟒
50-54 6 37 𝐟𝐐
55-59 3 40 (𝟏)(𝟒𝟎)
− 𝟗
𝟒
∑ 40 = 𝟑𝟒. 𝟓 + 𝟓
𝟔
𝐐𝟏 = 𝟑𝟓. 𝟑𝟑
FRACTILES or QUANTILES
Solution: kn/10 = 7(40)/10 = 28
CLASS
INTERVAL
f <cf D7 Class = 45-49
25-29 5 5
LD = 44.5
<cfb = 24
30-34 4 9
fD = 7
35-39 6 15 c=5
40-44 9 24 𝐤𝐧
45-49 7 31 𝟏𝟎 − < 𝐜𝐟𝐛
𝐃𝟕 = 𝐋𝐃 + 𝐜
50-54 6 37 𝐟𝐃
55-59 3 40 (𝟕)(𝟒𝟎)
− 𝟐𝟒
𝟏𝟎
∑ 40 = 𝟒𝟒. 𝟓 + 𝟓
𝟕
𝐃𝟕 = 𝟒𝟕. 𝟑𝟔
FRACTILES or QUANTILES
Solution: kn/100 = 30(40)/100 = 12
CLASS
INTERVAL
f <cf P30 Class = 35-39
25-29 5 5
LP = 34.5
<cfb = 9
30-34 4 9
fP = 6
35-39 6 15 c=5
40-44 9 24 𝐤𝐧
45-49 7 31 𝟏𝟎𝟎 − < 𝐜𝐟𝐛
𝐏𝟑𝟎 = 𝐋𝐏 + 𝐜
50-54 6 37 𝐟𝐏
55-59 3 40 (𝟑𝟎)(𝟒𝟎)
− 𝟗
𝟏𝟎𝟎
∑ 40 = 𝟑𝟒. 𝟓 + 𝟓
𝟔
𝐏𝟑𝟎 = 𝟑𝟕. 𝟎