Measures of Position (Quartiles, Deciles, and Percentiles)
Uploaded by
ScribdTranslationsMeasures of Position (Quartiles, Deciles, and Percentiles)
Uploaded by
ScribdTranslationsJuan Daniel Herrera Martínez
20162269
Measures of position (quartiles, deciles, and percentiles).
Measures of position.
The Measures of Position, also known as Other Measures of
Dispersion, are other measures ormethodswhich turn out to be more practical
to clarify certain situations in which the variation is sought to be described
or dispersion in a set ofdata.
Quartiles:
The quartiles are the three values that divide the data set.
ordered in four equally percentage parts.
There are three quartiles usually denoted as Q1, Q2, Q3. The second quartile is
precisely the median. The first quartile is thevaluein which or by
below which remains a quarter (25%) of all the values of the sequence
the third quartile, is the value at or below which
three quarters (75%) of the data remain.
Grouped Data
How quartiles gain their greatest importance when we count a
large number of data and we take into account that in these cases
Generally, the data is summarized in a frequency table. The
formula for thecalculationof the quartiles when it comes to data
grouped is the following:
k= 1,2,3
Where:
Lk = Lower real limit of theclassof the k-th quartile
n = Number of data
Fk = Cumulative frequency of the class preceding the k-th quartile class.
fk = Frequency of the class of quartile k
c = Length of the class interval of the k-th quartile
If you want to calculate each quartile individually, using another formula
the following is given:
• The first quartile Q1 is the smallest value that is greater than a quarter.
part of the data; that is, the value of the variable that exceeds 25%
of the observations and is surpassed by 75% of the observations.
Formula for Q1, for grouped data series:
Where:
L1 = lower limit of the class that contains it
P = value that represents the position of the measurement
f1 = the frequency of the class that contains the requested measurement.
Fa-1 = accumulated frequency before that contains the measure
requested.
class interval
• Thesecond quartile Q2 (coincides, is identical or similar to the median, Q2 =
Md), is the smallest value that is greater than half of the data, that is,
50% of the observations are greater than the median and 50% are
minors.
Formula for Q2, for grouped data series:
Where:
L1 = lower limit of the class that contains it
P = value that represents the position of the measurement
f1 = the frequency of the class that contains the requested measurement.
Fa-1 = accumulated frequency before the one that contains the measurement
requested.
class interval
• The third quartile Q3 is the smallest value that is greater than three quarters.
parts of the data, that is, the value of the variable that exceeds the
75% is surpassed by 25% of the observations.
Formula for Q3, for grouped data series:
Where:
L1 = lower limit of the class that contains it
P = value that represents the position of the measure
f1 = the frequency of the class that contains the requested measurement.
Fa-1 = accumulated frequency before the one that contains the measure
requested.
Ic = class interval.
Another way to see it is to start from the idea that all measurements are nothing but cases.
particulars of the percentile, since the first quartile is the 25th percentile and the
third quartile 75% percentile.
For Ungrouped Data
If there is a series of values X1, X2, X3 ... Xn, it is located by the
following formulas:
The first quartile:
When n is even:
When n is odd:
• For the third quartile
When n is even:
When n is odd:
DECILES.
The deciles are certain numbers that divide the sequence of data.
divided into ten equally percentage parts. These are the nine values
which divide the ordered dataset into ten equal parts, are
also a particular case of percentiles. The deciles are denoted D1,
D2,..., D9, which are read as first decile, second decile, etc.
Deciles, like quartiles, are widely used to set
academic performance.
Grouped Data
For grouped data, the deciles are calculated using the formula.
k = 1, 2, 3,... 9
Where:
Lk = Lower real limit of the class of decile k
n = Number of data
Fk = Cumulative frequency of the class preceding the class of the decile k.
Frequency of the class of decile k
c = Length of the class interval of the k-th decile
Another formula to calculate the deciles:
• Thefourth decile is the value of the variable that exceeds 40% of the
observations and is surpassed by 60% of the observations.
• The fifth decile corresponds to the median.
• The ninth decile exceeds 90% and is surpassed by the remaining 10%.
Where (for everyone):
L1 = lower limit of the class that contains it
P = value that represents the position of the measurement
f1 = the frequency of the class that contains the requested measurement.
Fa-1 = cumulative frequency prior to the measure it contains
requested.
Ic = class interval.
Ungrouped Data Formulas
If you have a series of values X1, X2, X3 ... Xn, it is located by the
following formulas:
When n is even:
When n is odd:
Let A be the decile number.
PERCENTILES
Percentiles are perhaps the most used measures for purposes of
location or classification of people when attending to characteristics
such as weight, height, etc.
Percentiles are certain numbers that divide the sequence of data.
divided into one hundred equal percentage parts. These are the 99
values that divide the ordered data set into one hundred equal parts.
The percentiles (P1, P2,... P99), read as first percentile,..., percentile 99.
Grouped Data
When the data is grouped in a frequency table, calculations are made
by means of the formula:
k = 1, 2, 3,... 99
Where:
Lk = Lower real limit of the k-th decile class
Number of data
Fk = Cumulative frequency of the class that precedes the class of decile k.
Frequency of the class of decile k
c = Length of the interval of the decile class k
Another way to calculate percentiles is:
• First
percentile, which exceeds one percent of the values and is
overcome by the remaining ninety-nine percent.
• The60th percentile is the value of the variable that exceeds 60% of the
observations and is surpassed by 40% of the observations.
• The 99th percentile surpasses 99% of the data and is in turn surpassed by the 1%.
remaining.
Non-Grouped Data Formulas
If you have a series of values X1, X2, X3 ... Xn, it is located by means of the
following formulas:
For percentiles, when n is even:
When n is odd:
Being A, the percentile number.
It is easy to see that the first quartile corresponds to the 25th percentile; the second
quartile with the 50th percentile and the third quartile with the 75th percentile.
EXAMPLE 1:
Determination of the first quartile, the seventh decile, and the 30th percentile, of the
next table:
Salaries No. Of fa
(I. Of Classes) Employees (f1)
200-299 85 85
300-299 90 175
400-499 120 295
500-599 70 365
600-699 62 427
700-800 36 463
Since they are grouped data, the formula is used
Being,
The position of the first quartile.
The position of the 7th decile.
The position of the 30th percentile.
So,
The first quartile:
115.5 - 85 = 30.75
Li = 300, Ic = 100 , fi = 90
The 7th decile:
Position:
324.1 - 295 = 29.1
Li = 500, fi = 70
The 30th percentile
Position:
138.9 - 85 = 53.9
fi = 90
These results indicate that 25% of employees
they winsalariesbelow $334; that below 541.57 earns 57% of the
employees and over $359.88, earn 70% of the employees.
Example 2:
From the 20 students we have their evaluations of an exam to calculate the 1,
5 75.
5,5,8,7,9,10,7,6,8,7,8,9,10,10,8,7,6,5,9,6.
First, the numbers must be arranged in ascending order.
5,5,5,6,6,6,7,7,7,7,8,8,8,8,9,9,9,10,10,10.
1 = (20+1) 4 = 5.25 this is the Position and we are going to look for it in the series of
data already sorted.
=6
Now we are going to calculate the 5 5,5,5,6,6,6,7,7,7,7,8,8,8,8,9,9,9,10,10,10.
5 = 5( + 10 = 10.5
This is the position and we are going to look for it in the already sorted data series.
Therefore, it is calculated after finding the position that is
The following is done at 10.5:
7+8 2 = 7.5 therefore the half is:
5 = 7.5
First, the numbers must be arranged in ascending order.
5,5,5,6,6,6,7,7,7,7,8,8,8,8,9,9,9,10,10,10
75 = 75(20+1) 100 = 15.75 this is the Position and we are going to search for it in the
sorted data series.
=9
BIBLIOGRAPHY
• TextStatisticsfortheSciencesAdministrative.
• Martinez, Ciro. Statistics andSampling.Ecoe Editions. Bogotá.
11th.Edition.
• Internet consultation
