You are on page 1of 42

Weighted Mean Calculatoris an online statistics tool for data analysis

programmed to calculate the Weighted Mean by giving different weights to some


of the individual values. The idea of weighted mean plays a role in descriptive
statistics and also occurs in a more general form in several other areas of
mathematics

Definition
Weighted Mean is an average computed by giving different weights to some of
the individual values. If all the weights are equal, then the weighted mean is the
same as the arithmetic mean. Whereas weighted means generally behave in a
similar approach to arithmetic means, they do have a few counter instinctive
properties. Data elements with a high weight contribute more to the weighted
mean than do elements with a low weight. The weights cannot be negative. Some
may be zero, but not all of them; since division by zero is not allowed. Weighted
means play an important role in the systems of data analysis, weighted
differential and integral calculus

Weighted Mean Formula


The Weighted mean for given set of non negative data {x1, x2, x3, ...xn} with non
negative weights {w1, w2, w3, ...wn} can be derived from the formula

where
x is the repeating value
w is the number of occurrences of x (weight)
x̄ is the weighted mean

The collection of tools employs the study of methods and procedures used for
gathering, organizing, and analyzing data to understand theory of Probability
and Statistics. The set of ideas which is intended to offer the way for making
scientific implication from such resulting summarized data. In many applications
it is necessary to calculate the weighted mean for a set of data with different
individual errors. With this online weighted mean calculator you can effortlessly
make your calculation for the set of given observations

2.5: The Empirical Rule and Chebyshev's


Theorem
1. Last updated
Sep 19, 2017
2. Share
Table of contents
SKILLS TO DEVELOP
 To learn what the value of the standard deviation of a data set implies about how the data
scatter away from the mean as described by the Empirical Rule and Chebyshev’s Theorem.
 To use the Empirical Rule and Chebyshev’s Theorem to draw conclusions about a data set.

You probably have a good intuitive grasp of what the average of a data set says about that data set. In
this section we begin to learn what the standard deviation has to tell us about the nature of the data
set.

The Empirical Rule


We start by examining a specific set of data. Table 2.5.12.5.1 shows the heights in inches
of 100100 randomly selected adult men. A relative frequency histogram for the data is shown
in Figure 2.5.12.5.1. The mean and standard deviation of the data are, rounded to two decimal
places, x¯=69.92x¯=69.92 and σ=1.70σ=1.70.

Table 2.5.12.5.1: Heights of Men

72.3 71.3 72.5 70.6 68.2 70.1 68.4

70.5 71.0 70.9 69.3 69.4 69.7 69.1

70.0 70.4 68.9 69.4 69.4 69.2 70.7

69.8 68.6 69.5 71.6 66.2 72.4 70.7

69.3 68.9 74.8 68.0 71.2 68.3 70.2


72.2 70.0 68.7 67.9 71.1 69.0 70.8

68.8 67.2 73.0 70.4 67.8 70.0 69.5

67.6 67.0 70.3 71.2 65.6 68.1 70.8

67.5 71.3 71.5 71.0 69.1 69.5 71.1

72.7 72.8 69.6 65.9 68.0 69.7 68.7

If we go through the data and count the number of observations that are within one standard deviation
of the mean, that is, that
are between 69.92−1.70=68.2269.92−1.70=68.22 and 69.92+1.70=71.6269.92+1.70=71.62 inches,
there are 6969 of them. If we count the number of observations that are within two standard deviations
of the mean, that is, that are
between 69.92−2(1.70)=66.5269.92−2(1.70)=66.52 and 69.92+2(1.70)=73.3269.92+2(1.70)=73.3
2 inches, there are 9595 of them. All of the measurements are within three standard deviations of the
mean, that is,
between 69.92−3(1.70)=64.82269.92−3(1.70)=64.822 and 69.92+3(1.70)=75.0269.92+3(1.70)=75
.02 inches. These tallies are not coincidences, but are in agreement with the following result that has
been found to be widely applicable.
Figure 2.5.12.5.1: Heights of Adult Men

The Empirical Rule


Approximately 68%68% of the data lie within one standard deviation of the mean, that is, in the
interval with endpoints x¯±sx¯±s for samples and with endpoints μ±σμ±σ for populations; if a data
set has an approximately bell-shaped relative frequency histogram, then (Figure 2.5.22.5.2)

 approximately 95%95% of the data lie within two standard deviations of the mean, that is, in
the interval with endpoints x¯±2sx¯±2s for samples and with endpoints μ±2σμ±2σ for
populations; and
 approximately 99.7%99.7% of the data lies within three standard deviations of the mean, that
is, in the interval with endpoints x¯±3sx¯±3s for samples and with endpoints μ±3σμ±3σ for
populations.
Figure 2.5.22.5.2: The Empirical Rule

Two key points in regard to the Empirical Rule are that the data distribution must be
approximately bell-shaped and that the percentages are only approximately true. The Empirical Rule
does not apply to data sets with severely asymmetric distributions, and the actual percentage of
observations in any of the intervals specified by the rule could be either greater or less than those
given in the rule. We see this with the example of the heights of the men: the Empirical Rule suggested
68 observations between 68.2268.22 and 71.6271.62inches, but we counted 6969.

EXAMPLE 2.5.12.5.1
Heights of 1818-year-old males have a bell-shaped distribution with mean 69.669.6 inches and
standard deviation 1.41.4 inches.

1. About what proportion of all such men are between 68.268.2 and 7171 inches tall?
2. What interval centered on the mean should contain about 95%95% of all such men?

Solution:

A sketch of the distribution of heights is given in Figure 2.5.32.5.3.

1. Since the interval from 68.268.2 to 71.071.0 has endpoints x¯−sx¯−s and x¯+sx¯+s, by the
Empirical Rule about 68%68% of all 1818-year-old males should have heights in this range.
2. By the Empirical Rule the shortest such interval has
endpoints x¯−2sx¯−2s and x¯+2sx¯+2s. Since

x¯−2s=69.6−2(1.4)=66.8(2.5.1)(2.5.1)x¯−2s=69.6−2(1.4)=66.8

and

x¯+2s=69.6+2(1.4)=72.4(2.5.2)(2.5.2)x¯+2s=69.6+2(1.4)=72.4

the interval in question is the interval from 66.866.8 inches to 72.472.4 inches.
Figure 2.5.32.5.3: Distribution of Heights

EXAMPLE 2.5.22.5.2
Scores on IQ tests have a bell-shaped distribution with mean μ=100μ=100 and standard
deviation σ=10σ=10. Discuss what the Empirical Rule implies concerning individuals with IQ scores
of 110110, 120120, and 130130.

Solution:

A sketch of the IQ distribution is given in Figure 2.5.32.5.3. The Empirical Rule states that

1. approximately 68%68% of the IQ scores in the population lie between 9090 and 110110,
2. approximately 95%95% of the IQ scores in the population lie between 8080 and 120120,
and
3. approximately 99.7%99.7% of the IQ scores in the population lie between 7070 and 130130.

Figure 2.5.32.5.3: Distribution of IQ Scores.


1. Since 68%68% of the IQ scores lie within the interval from 9090 to 110110, it must be the
case that 32%32% lie outside that interval. By symmetry approximately half of that 32%32%,
or 16%16% of all IQ scores, will lie above 110110. If 16%16% lie above 110110,
then 84%84% lie below. We conclude that the IQ score 110110 is the 84th84th percentile.
2. The same analysis applies to the score 120120. Since approximately 95%95% of all IQ scores
lie within the interval form 8080 to 120120, only 5%5% lie outside it, and half of them,
or 2.5%2.5% of all scores, are above 120120. The IQ score 120120 is thus higher
than 97.5%97.5% of all IQ scores, and is quite a high score.
3. By a similar argument, only 15/10015/100 of 1%1% of all adults, or about one or two in every
thousand, would have an IQ score above 130130. This fact makes the score 130130 extremely
high.

Chebyshev’s Theorem
The Empirical Rule does not apply to all data sets, only to those that are bell-shaped, and even then is
stated in terms of approximations. A result that applies to every data set is known as Chebyshev’s
Theorem.

CHEBYSHEV’S THEOREM
For any numerical data set,

 at least 3/43/4 of the data lie within two standard deviations of the mean, that is, in the interval
with endpoints x¯±2sx¯±2s for samples and with endpoints μ±2σμ±2σ for populations;
 at least 8/98/9 of the data lie within three standard deviations of the mean, that is, in the
interval with endpoints x¯±3sx¯±3s for samples and with endpoints μ±3σμ±3σ for
populations;
 at least 1−1/k21−1/k2 of the data lie within kk standard deviations of the mean, that is, in the
interval with endpoints x¯±ksx¯±ks for samples and with endpoints μ±kσμ±kσ for
populations, where kk is any positive whole number that is greater than 11.

Figure 2.5.42.5.4 gives a visual illustration of Chebyshev’s Theorem.


Figure 2.5.42.5.4: Chebyshev’s Theorem

It is important to pay careful attention to the words “at least” at the beginning of each of the three
parts of Chebyshev’s Theorem. The theorem gives the minimum proportion of the data which must lie
within a given number of standard deviations of the mean; the true proportions found within the
indicated regions could be greater than what the theorem guarantees.

EXAMPLE 2.5.32.5.3
A sample of size n=50n=50 has mean x¯=28x¯=28 and standard deviation s=3s=3. Without knowing
anything else about the sample, what can be said about the number of observations that lie in the
interval (22,34)(22,34)? What can be said about the number of observations that lie outside that
interval?

Solution:

The interval (22,34)(22,34) is the one that is formed by adding and subtracting two standard
deviations from the mean. By Chebyshev’s Theorem, at least 3/43/4 of the data are within this
interval. Since 3/43/4 of 5050 is 37.537.5, this means that at least 37.537.5 observations are in the
interval. But one cannot take a fractional observation, so we conclude that at least 3838observations
must lie inside the interval (22,34)(22,34).
If at least 3/43/4 of the observations are in the interval, then at most 1/41/4 of them are outside it.
Since 1/41/4 of 5050 is 12.512.5, at most 12.512.5 observations are outside the interval. Since again
a fraction of an observation is impossible, x(22,34)x(22,34).

EXAMPLE 2.5.42.5.4
The number of vehicles passing through a busy intersection
between 8:00a.m.8:00a.m. and 10:00a.m.10:00a.m. was observed and recorded on every weekday
morning of the last year. The data set contains n=251n=251 numbers. The sample mean
is x¯=725x¯=725 and the sample standard deviation is s=25s=25. Identify which of the following
statements must be true.

1. On approximately 95%95% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was
between 675675 and 775775.
2. On at least 75%75% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was
between 675675 and 775775.
3. On at least 189189 weekday mornings last year the number of vehicles passing through the
intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was
between 675675 and 775775.
4. On at most 25%25% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was either less
than 675675 or greater than 775775.
5. On at most 12.5%12.5% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was less than 675675.
6. On at most 25%25% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was less than 675675.

Solution:

1. Since it is not stated that the relative frequency histogram of the data is bell-shaped, the
Empirical Rule does not apply. Statement (1) is based on the Empirical Rule and therefore it
might not be correct.
2. Statement (2) is a direct application of part (1) of Chebyshev’s Theorem
because x¯−2sx¯−2s, x¯+2s=(675,775)x¯+2s=(675,775). It must be correct.
3. Statement (3) says the same thing as statement (2)
because 75%75% of 251251 is 188.25188.25, so the minimum whole number of observations
in this interval is 189189. Thus statement (3) is definitely correct.
4. Statement (4) says the same thing as statement (2) but in different words, and therefore is
definitely correct.
5. Statement (4), which is definitely correct, states that at most 25%25% of the time either fewer
than 675675 or more than 775775 vehicles passed through the intersection. Statement (5) says
that half of that 25%25% corresponds to days of light traffic. This would be correct if the
relative frequency histogram of the data were known to be symmetric. But this is not stated;
perhaps all of the observations outside the interval (675,775675,775) are less than 7575. Thus
statement (5) might not be correct.
6. Statement (4) is definitely correct and statement (4) implies statement (6): even if every
measurement that is outside the interval (675,775675,775) is less than 675675 (which is
conceivable, since symmetry is not known to hold), even so at most 25%25% of all
observations are less than 675675. Thus statement (6) must definitely be correct.

KEY TAKEAWAY
 The Empirical Rule is an approximation that applies only to data sets with a bell-shaped
relative frequency histogram. It estimates the proportion of the measurements that lie within
one, two, and three standard deviations of the mean.
 Chebyshev’s Theorem is a fact that applies to all possible data sets. It describes the minimum
proportion of the measurements that lie must within one, two, or more standard deviations of
the mean.

Contributor
 Anonymous
1. Back to top
2.
o 2.4: Relative Position of Data

o 2.E: Descriptive Statistics (Exercises)

Recommended articles

1. 7.1: Large Sample Estimation of a Population MeanA confidence interval for a population mean is
an estimate of the population mean together with an indication of reliability. There are different
form...
2. 2.E: Descriptive Statistics (Exercises)These are homework exercises to accompany the Textmap
created for "Introductory Statistics" by Shafer and Zhang. Complementary General Chemistry
quest...
3. 2.1: Three Popular Data DisplaysGraphical representations of large data sets provide a quick
overview of the nature of the data. A population or a very large data set may be repr...
4. 2.2: Measures of Central LocationThe mean, the median, and the mode each answer the question
“Where is the center of the data set?” The nature of the data set, as indicated by a relat...

5. 2.3: Measures of VariabilityThe range, the standard deviation, and the variance each give a
quantitative answer to the question “How variable are the data?”

1. The LibreTexts libraries are Powered by MindTouch® and are based upon work supported
by the National Science Foundation under grant numbers: 1246120, 1525057, and
1413739. Unless otherwise noted, the LibreTexts library is licensed under a Creative
Commons Attribution-Noncommercial-Share Alike 3.0 United States License. Permissions
beyond the scope of this license may be available at copyright@ucdavis.edu.

2.4: Relative Position of Data


1. Last updated
Sep 19, 2017
2. Share
Table of contents
SKILLS TO DEVELOP
 To learn the concept of the relative position of an element of a data set.
 To learn the meaning of each of two measures, the percentile rank and the zz-score, of the
relative position of a measurement and how to compute each one.
 To learn the meaning of the three quartiles associated to a data set and how to compute them.
 To learn the meaning of the five-number summary of a data set, how to construct the box
plot associated to it, and how to interpret the box plot.

When you take an exam, what is often as important as your actual score on the exam is the way your
score compares to other students’ performance. If you made a 7070 but the average score (whether
the mean, median, or mode) was 8585, you did relatively poorly. If you made a 7070 but the average
score was only 5555 then you did relatively well. In general, the significance of one observed value
in a data set strongly depends on how that value compares to the other observed values in a data set.
Therefore we wish to attach to each observed value a number that measures its relative position.

Percentiles and Quartiles


Anyone who has taken a national standardized test is familiar with the idea of being given both a score
on the exam and a “percentile ranking” of that score. You may be told that your score was 625625 and
that it is the 85th85th percentile. The first number tells how you actually did on the exam; the second
says that 85%85% of the scores on the exam were less than or equal to your score, 625625.

DEFINITION: PERCENTILE OF DATA


Given an observed value xx in a data set, xx is the PthPth percentile of the data if P%P% of the data
are less than or equal to xx. The number PP is the percentile rank of xx.

EXAMPLE 2.4.12.4.1
What percentile is the value 1.391.39 in the data set of ten GPAs considered in a previous Example?
What percentile is the value 3.333.33?

Solution:

The data, written in increasing order, are

1.391.761.902.122.532.713.003.333.714.00(2.4.1)(2.4.1)1.391.761.902.122.532.713.003.333.714.
00

The only data value that is less than or equal to 1.391.39 is 1.391.39 itself. Since 11 out of ten,
or 1/10=10%1/10=10%, of the data points are less than or equal to 1.391.39, 1.391.39 is
the 10th10th percentile. Eight data values are less than or equal to 3.333.33. Since 88 out of ten,
or 8∕10=.80=80%8∕10=.80=80% of the data values are less than or equal to 3.333.33, the
value 3.333.33 is the 80th80th percentile of the data.
The Pth percentile cuts the data set in two so that approximately P%P% of the data lie below it
and (100−P)%(100−P)% of the data lie above it. In particular, the three percentiles that cut the data
into fourths, as shown in Figure 2.4.12.4.1, are called the quartiles of a data set. The quartiles are the
three numbers Q1Q1, Q2Q2, Q3Q3 that divide the data set approximately into fourths. The following
simple computational definition of the three quartiles works well in practice.

Figure 2.4.12.4.1: Data Division by Quartiles

DEFINITION: QUARTILE
For any data set:

1. The second quartile Q2Q2 of the data set is its median.


2. Define two subsets:
1. the lower set: all observations that are strictly less than Q2Q2
2. the upper set: all observations that are strictly greater than Q2Q2
3. The first quartile Q1Q1 of the data set is the median of the lower set.
4. The third quartile Q3Q3 of the data set is the median of the upper set.

EXAMPLE 2.4.22.4.2
Find the quartiles of the data set of GPAs of discussed in a previous Example.
Solution:

As in the previous example we first list the data in numerical order:

1.391.761.902.122.532.713.003.333.714.00(2.4.2)(2.4.2)1.391.761.902.122.532.713.003.333.714.
00

This data set has n=10n=10 observations. Since 1010 is an even number, the median is the mean of
the two middle observations: x~=(2.53+2.71)∕2=2.62x~=(2.53+2.71)∕2=2.62. Thus the second
quartile is Q2=2.62Q2=2.62. The lower and upper subsets are

Lower: L={1.39,1.76,1.90,2.12,2.53}L={1.39,1.76,1.90,2.12,2.53}, Upper:


U={2.71,3.00,3.33,3.71,4.00}U={2.71,3.00,3.33,3.71,4.00}.

Each has an odd number of elements, so the median of each is its middle observation. Thus the first
quartile is Q1=1.90Q1=1.90, the median of LL, and the third quartile is Q3=3.33Q3=3.33, the median
of UU.

EXAMPLE 2.4.32.4.3
Adjoin the observation 3.883.88 to the data set of the previous example and find the quartiles of the
new set of data.

Solution:

As in the previous example we first list the data in numerical order:

1.391.761.902.122.532.713.003.333.713.88,4.00(2.4.3)(2.4.3)1.391.761.902.122.532.713.003.333
.713.88,4.00

This data set has 1111 observations. The second quartile is its median, the middle value 2.712.71.
Thus Q2=2.71Q2=2.71. The lower and upper subsets are now

 Lower: L={1.39,1.76,1.90,2.12,2.53}L={1.39,1.76,1.90,2.12,2.53}
 Upper: U={3.00,3.33,3.71,3.88,4.00}U={3.00,3.33,3.71,3.88,4.00}.

The lower set LL has median, the middle value, 1.901.90, so Q1=1.90Q1=1.90. The upper set has
median 3.713.71, so Q3=3.71Q3=3.71.

In addition to the three quartiles, the two extreme values, the minimum xminxmin and the
maximum xmaxxmax are also useful in describing the entire data set. Together these five numbers are
called the five-number summary of a data set,

{ XminXmin, Q1Q1, Q2Q2, Q3Q3, XmaxXmax }

The five-number summary is used to construct a box plot, as in Figure 2.4.22.4.2. Each of the five
numbers is represented by a vertical line segment, a box is formed using the line segments
at Q1Q1 and Q3Q3 as its two vertical sides, and two horizontal line segments are extended from the
vertical segments marking Q1Q1 and Q3Q3 to the adjacent extreme values. (The two horizontal line
segments are referred to as “whiskers,” and the diagram is sometimes called a "box and whisker plot.")
We caution the reader that there are other types of box plots that differ somewhat from the ones we
are constructing, although all are based on the three quartiles.

Figure 2.4.22.4.2: The Box Plot

Note that the distance from Q1Q1 to Q3Q3 is the length of the interval over which the middle half of
the data range. Thus it has the following special name.

DEFINITION: INTERQUARTILE
The interquartile range IQRIQR is the quantity

IQR=Q3−Q1(2.4.4)(2.4.4)IQR=Q3−Q1

EXAMPLE 2.4.42.4.4
Construct a box plot and find the IQRIQR for the data in Example 2.4.32.4.3.

Solution:

From our work in Example 2.4.12.4.1, we know that the five-number summary is

 Xmin=1.39Xmin=1.39
 Q1=1.90Q1=1.90
 \[Q2 = 2.62\)
 Q3=3.33Q3=3.33
 Xmax=4.00Xmax=4.00

The box plot is:


The interquartile range is: IQR=3.33−1.90=1.43IQR=3.33−1.90=1.43.

zz-Scores
Another way to locate a particular observation xx in a data set is to compute its distance from the
mean in units of standard deviation. The zz-score indicates how many standard deviations an
individual observation xx is from the center of the data set, its mean. It is used on distributions that
have been standardized, which allows us to better understand its properties. If zz is negative then xx is
below average. If zz is 00 then xx is equal to the average. If zz is positive then xx is above the average

DEFINITION: zz-SCORE
The zz-score of an observation xx is the number zz given by the computational formula

z=x−μσ(2.4.5)(2.4.5)z=x−μσ
Figure 2.4.32.4.3: x-Scale versus z-Score

EXAMPLE 2.4.52.4.5
Suppose the mean and standard deviation of the GPA's of all currently registered students at a college
are μ=2.70μ=2.70 and σ=0.50σ=0.50. The zz-scores of the GPA's of two students, Antonio and
Beatrice, are z=−0.62z=−0.62 and z=1.28z=1.28, respectively. What are their GPAs?

Solution:

Using the second formula right after the definition of zz-scores we compute the GPA's as

 Antonio: x=μ+zσ=2.70+(−0.62)(0.50)=2.39x=μ+zσ=2.70+(−0.62)(0.50)=2.39
 Beatrice: x=μ+zσ=2.70+(1.28)(0.50)=3.34x=μ+zσ=2.70+(1.28)(0.50)=3.34

KEY TAKEAWAYS
 The percentile rank and zz-score of a measurement indicate its relative position with regard
to the other measurements in a data set.
 The three quartiles divide a data set into fourths.
 The five-number summary and its associated box plot summarize the location and distribution
of the data.

Contributor
 Anonymous

1. Back to top
2.
o 2.3: Measures of Variability

o 2.5: The Empirical Rule and Chebyshev's Theorem

Recommended articles

1. 1.7: PercentilesA test score in and of itself is usually difficult to interpret. For example, if you
learned that your score on a measure of shyness was 35 out of a p...
2. 2.4: Measures of the Location of the DataThe values that divide a rank-ordered set of data into 100
equal parts are called percentiles and are used to compare and interpret data. For example,...
3. 2.6: Box Plots
4. 1.6: Examining Numerical DataIn this section we will be introduced to techniques for exploring and
summarizing numerical variables. Recall that outcomes of numerical variables are...
5. 2.5: Box PlotsBox plots are a type of graph that can help visually organize data. To graph a box plot
the following data points must be calculated: the minimum valu...

1. The LibreTexts libraries are Powered by MindTouch® and are based upon work supported
by the National Science Foundation under grant numbers: 1246120, 1525057, and
1413739. Unless otherwise noted, the LibreTexts library is licensed under a Creative
Commons Attribution-Noncommercial-Share Alike 3.0 United States License. Permissions
beyond the scope of this license may be available at copyright@ucdavis.edu.

Weighted Mean: Formula: How to Find Weighted Mean


Statistics Definitions > Weighted Mean
Watch the video or read the article below:

What is a Weighted Mean?


A weighted mean is a kind of average. Instead of each data point contributing equally to the final mean, some data
points contribute more “weight” than others. If all the weights are equal, then the weighted mean equals the arithmetic
mean (the regular “average” you’re used to). Weighted means are very common in statistics, especially when studying
populations.

The Arithmetic Mean.


When you find a mean for a set of numbers, all the numbers carry an equal weight. For example, if you want to find
the arithmetic mean of 1, 3, 5, 7, and 10:
1. Add up your data points: 1 + 3 + 5 + 7 + 10 = 26.
2. Divide by the number of items in the set: 26 / 5 = 5.2.
What do we mean by “equal weight”? The first sentence in some tests (like this one) is sometimes “All questions carry
an equal weight”). It’s saying that all the questions in the exam are worth the same number of points. If you have a 100
point exam and 10 questions, each question is worth 1/10th of the points. In the above question, you have of a set of five
numbers. You can think of each number contributing 1/5 to the total mean (as there are 5 numbers in the set).

The Weighted Mean.


In some cases, you might want a number to have more weight. In that case, you’ll want to find the weighted mean. To
find the weighted mean:
1. Multiply the numbers in your data set by the weights.
2. Add the results up.
For that set of number above with equal weights (1/5 for each number), the math to find the weighted mean would be:
1(*1/5) + 3(*1/5) + 5(*1/5) + 7(*1/5) + 10(*1/5) = 5.2.
Sample problem: You take three 100-point exams in your statistics class and score 80, 80 and 95. The last exam
is much easier than the first two, so your professor has given it less weight. The weights for the three exams are:
 Exam 1: 40 % of your grade. (Note: 40% as a decimal is .4.)
 Exam 2: 40 % of your grade.
 Exam 3: 20 % of your grade.
What is your final weighted average for the class?
1. Multiply the numbers in your data set by the weights:
.4(80) = 32
.4(80) = 32
.2(95) = 19
2. Add the numbers up. 32 + 32 + 19 = 83.
The percent weight given to each exam is called a weighting factor.

Weighted Mean Formula


The weighted mean is relatively easy to find. But in some cases the weights might not add up to 1. In those cases, you’ll
need to use the weighted mean formula. The only difference between the formula and the steps above is that you divide
by the sum of all the weights.

The image above is the technical formula for the weighted mean. In simple terms, the formula can be written as:
Weighted mean = Σwx/Σw
Σ = the sum of (in other words…add them up!).
w = the weights.
x = the value.
To use the formula:

1. Multiply the numbers in your data set by the weights.


2. Add the numbers in Step 1 up. Set this number aside for a moment.
3. Add up all of the weights.
4. Divide the numbers you found in Step 2 by the number you found in Step 3.
In the sample grades problem above, all of the weights add up to 1 (.4 + .4 + .2) so you would divide your answer (83)
by 1:
83 / 1 = 83.
However, let’s say your weighted means added up to 1.2 instead of 1. You’d divide 83 by 1.2 to get:
83 / 1.2 = 69.17.
Warning: The weighted mean can be easily influenced by outliers in your data. If you have very high or very low
values in your data set, the weighted mean may not be a good statistic to rely on.
------------------------------------------------------------------------------
If you prefer an online interactive environment to learn R and statistics, this free R Tutorial by Datacamp is a great way
to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try this
Statistics with R track.
Comments are now closed for this post. Need help or want to post a correction? Please post a comment on our Facebook
page and I'll do my best to help!
Weighted Mean: Formula: How to Find Weighted Mean was last modified: October 15th, 2017 by Stephanie Glen
By Stephanie Glen | January 23, 2014 | Statistics How To |
 ← Non Parametric Data and Tests (Distribution Free Tests)
 Line Graph →
34 thoughts on “Weighted Mean: Formula: How to
Find Weighted Mean”
1. iden kerryFebruary 11, 2015 at 2:06 am
well done thanks

2. Joanie MacuteMarch 4, 2015 at 6:34 pm


Nice one ! Thanks ! It helps a lot !

3. FatosMay 8, 2015 at 8:43 am


Very helpful, it has a lot to learn and repeat in a short time.

4. Joseph lunguJuly 3, 2015 at 1:42 pm


This is so easy to understand and interprete thanx

5. JoelJuly 25, 2015 at 12:09 am


very helpful.Thank you.

6. sophie moagiSeptember 28, 2015 at 12:41 am


thanx dats so useful

7. LorraineDecember 8, 2015 at 8:30 am


Very simple to understand,thanks

8. CherlotteDecember 30, 2015 at 12:39 am


Really helpful in understanding the basic difference !!
Thank u :)

9. rafey rafaqatFebruary 19, 2016 at 6:40 am


thanx sir

10. JoeMarch 13, 2016 at 3:29 am


Great….easy to understand. Thanks much. Itox JK – 14x

11. R.A.SaimaMay 14, 2016 at 9:29 am


Really helpful , thanks a lot ….

12. saranJune 7, 2016 at 10:11 pm


how did you calculate the weight of 0.4 is becoming 80

13. saranJune 7, 2016 at 10:12 pm


can you please explain , i dont know how to calculate Weight of the score.

pl explain some easy understanding of examples.

Regards,
R.Saran
9943041432

14. AndalePost authorJune 8, 2016 at 6:44 pm


Saran, do you mean you’re having trouble following the weighted mean example?

15. AndalePost authorJune 8, 2016 at 6:45 pm


0.4 is the same as 40%. The 80 is the number of points scored in the exam (from the question). You just calculate
the two.

16. Asmamaw y.June 17, 2016 at 4:55 pm


You have made it very easy to understtad! thanks a lot

17. MatthewJune 21, 2016 at 9:12 pm


Thanks explained perfectly

18. nisha chauhanJuly 27, 2016 at 4:02 am


it’s good but can you discribe weighted arithmetic mean by more examples & differnt tyeps of examples.

19. Manikanta alleAugust 26, 2016 at 6:36 am


Thank you very much
This problem is very clearly understanding

20. Abeerah TashfeenNovember 23, 2016 at 10:21 am


Superb! Thank you so much!

21. JarvilleJanuary 11, 2017 at 4:29 am


Teach me how to compute weighted mean

22. SteveJanuary 27, 2017 at 8:06 am


Not 19. That’s 17

23. AndalePost authorJanuary 27, 2017 at 8:38 am


Hi, Steve,
Where are you seeing this? The only place is I see a 19 is .2(95). Google tells me that’s 19 :)

24. AndalePost authorFebruary 13, 2017 at 6:57 am


Yes. Thanks for spotting that. It’s corrected.

25. ShrutiFebruary 14, 2017 at 10:46 pm


Hey,

The initial scores per the question statement were 80, 80 and 85.
“Sample problem: You take three 100-point exams in your statistics class and score 80, 80 and 85. ”

Why, then, is the calculation of 20% made on a score of 95 ?

26. AndalePost authorFebruary 15, 2017 at 8:26 am


Thanks for spotting that, Shruti. I fixed the typo in the article and I’ll fix the video later on (hopefully I’ll get to it
today :) ).

27. millyMarch 6, 2017 at 8:16 am


my problem solved ……. thanks a lot

28. HanzyusufMarch 7, 2017 at 8:57 pm


You can only use weights till 100% right? then the total weights must always add upto 1.0….please help me
understand this…..i mean u used (40%, 40%, 20%) right? but whatever u use it must always add upto 100%..if i
am mistaken plz help me understand……

29. AndalePost authorMarch 8, 2017 at 9:15 am


Weights can add up to more than 1. If they do, use the formula.
As a simple example of why, think of GPAs. Technically, a GPA can only got up to 4.0 (which would equal
getting 100% in all classes). But sometimes classes (like AP classes) are given more weight — say 5.0 instead of
4.0.

30. Veerendra Mc CoonMarch 14, 2017 at 10:01 am


Thanks, but, how about : The grades of a student and the credits that he received for the courses that he or she got
the grades in? Like this= English- 90, Math – 89, Science- 86 and Spanish- 89. The credits will be 3, 4, 3 and 4
respectively.
31. AndalePost authorMarch 14, 2017 at 11:31 am
I’m assuming you want the weighted mean for the grades. Use the formula, figuring out the weights like this:
90(.3)
89(.4)
86(.3)
89 (.4)
Then divide by 1.4 (the total weight)

32. BeatriceOctober 1, 2017 at 6:53 pm


Wow! I really love that.Okay but a little difficult to understand

33. BeatriceOctober 1, 2017 at 6:58 pm


Why is the total weight 1.4

34. AndalePost authorOctober 2, 2017 at 9:57 am


Where are you seeing the total weight of 1.4?

Important Statistics Formulas


This web page presents statistics formulas described in the Stat Trek tutorials. Each formula links to a web
page that explains how to use the formula.

Parameters

 Population mean = μ = ( Σ Xi ) / N

 Population standard deviation = σ = sqrt [ Σ ( Xi - μ )2 / N ]

 Population variance = σ2 = Σ ( Xi - μ )2 / N

 Variance of population proportion = σP2 = PQ / n

 Standardized score = Z = (X - μ) / σ

 Population correlation coefficient = ρ = [ 1 / N ] * Σ { [ (Xi - μX) / σx ] * [ (Yi - μY) / σy ] }

Statistics

Unless otherwise noted, these formulas assume simple random sampling.

 Sample mean = x = ( Σ xi ) / n

 Sample standard deviation = s = sqrt [ Σ ( xi - x )2 / ( n - 1 ) ]

 Sample variance = s2 = Σ ( xi - x )2 / ( n - 1 )

 Variance of sample proportion = sp2 = pq / (n - 1)

 Pooled sample proportion = p = (p1 * n1 + p2 * n2) / (n1 + n2)

 Pooled sample standard deviation = sp = sqrt [ (n1 - 1) * s12 + (n2 - 1) * s22 ] / (n1 + n2 - 2) ]

 Sample correlation coefficient = r = [ 1 / (n - 1) ] * Σ { [ (xi - x) / sx ] * [ (yi - y) / sy ] }


Correlation

 Pearson product-moment correlation = r = Σ (xy) / sqrt [ ( Σ x2 ) * ( Σ y2 ) ]

 Linear correlation (sample data) = r = [ 1 / (n - 1) ] * Σ { [ (xi - x) / sx ] * [ (yi - y) / sy ] }

 Linear correlation (population data) = ρ = [ 1 / N ] * Σ { [ (X i - μX) / σx ] * [ (Yi - μY) / σy ] }

Simple Linear Regression

 Simple linear regression line: ŷ = b0 + b1x

 Regression coefficient = b1 = Σ [ (xi - x) (yi - y) ] / Σ [ (xi - x)2]

 Regression slope intercept = b0 = y - b1 * x

 Regression coefficient = b1 = r * (sy / sx)

 Standard error of regression slope = sb1 = sqrt [ Σ(yi - ŷi)2 / (n - 2) ] / sqrt [ Σ(xi - x)2 ]

Counting

 n factorial: n! = n * (n-1) * (n - 2) * . . . * 3 * 2 * 1. By convention, 0! = 1.

 Permutations of n things, taken r at a time: nPr = n! / (n - r)!

 Combinations of n things, taken r at a time: nCr = n! / r!(n - r)! = nPr / r!

Probability

 Rule of addition: P(A ∪ B) = P(A) + P(B) - P(A ∩ B)

 Rule of multiplication: P(A ∩ B) = P(A) P(B|A)

 Rule of subtraction: P(A') = 1 - P(A)

Random Variables

In the following formulas, X and Y are random variables, and a and b are constants.

 Expected value of X = E(X) = μx = Σ [ xi * P(xi) ]

 Variance of X = Var(X) = σ2 = Σ [ xi - E(x) ]2 * P(xi) = Σ [ xi - μx ]2 * P(xi)

 Normal random variable = z-score = z = (X - μ)/σ

 Chi-square statistic = Χ2 = [ ( n - 1 ) * s2 ] / σ2

 f statistic = f = [ s12/σ12 ] / [ s22/σ22 ]

 Expected value of sum of random variables = E(X + Y) = E(X) + E(Y)


 Expected value of difference between random variables = E(X - Y) = E(X) - E(Y)

 Variance of the sum of independent random variables = Var(X + Y) = Var(X) + Var(Y)

 Variance of the difference between independent random variables = Var(X - Y) = Var(X) + Var(Y)

Sampling Distributions

 Mean of sampling distribution of the mean = μx = μ

 Mean of sampling distribution of the proportion = μp = P

 Standard deviation of proportion = σp = sqrt[ P * (1 - P)/n ] = sqrt( PQ / n )

 Standard deviation of the mean = σx = σ/sqrt(n)

 Standard deviation of difference of sample means = σ d = sqrt[ (σ12 / n1) + (σ22 / n2) ]

 Standard deviation of difference of sample proportions = σ d = sqrt{ [P1(1 - P1) / n1] + [P2(1 - P2) / n2] }

Standard Error

 Standard error of proportion = SEp = sp = sqrt[ p * (1 - p)/n ] = sqrt( pq / n )

 Standard error of difference for proportions = SEp = sp = sqrt{ p * ( 1 - p ) * [ (1/n1) + (1/n2) ] }

 Standard error of the mean = SEx = sx = s/sqrt(n)

 Standard error of difference of sample means = SEd = sd = sqrt[ (s12 / n1) + (s22 / n2) ]

 Standard error of difference of paired sample means = SEd = sd = { sqrt [ (Σ(di - d)2 / (n - 1) ] } / sqrt(n)

 Pooled sample standard error = spooled = sqrt [ (n1 - 1) * s12 + (n2 - 1) * s22 ] / (n1 + n2 - 2) ]

 Standard error of difference of sample proportions = sd = sqrt{ [p1(1 - p1) / n1] + [p2(1 - p2) / n2] }

Discrete Probability Distributions

 Binomial formula: P(X = x) = b(x; n, P) = nCx * Px * (1 - P)n - x = nCx * Px * Qn - x

 Mean of binomial distribution = μx = n * P

 Variance of binomial distribution = σx2 = n * P * ( 1 - P )

 Negative Binomial formula: P(X = x) = b*(x; r, P) = x-1Cr-1 * Pr * (1 - P)x - r

 Mean of negative binomial distribution = μx = rQ / P

 Variance of negative binomial distribution = σx2 = r * Q / P2

 Geometric formula: P(X = x) = g(x; P) = P * Qx - 1

 Mean of geometric distribution = μx = Q / P

 Variance of geometric distribution = σx2 = Q / P2


 Hypergeometric formula: P(X = x) = h(x; N, n, k) = [ kCx ] [ N-kCn-x ] / [ NCn ]

 Mean of hypergeometric distribution = μx = n * k / N

 Variance of hypergeometric distribution = σx2 = n * k * ( N - k ) * ( N - n ) / [ N2 * ( N - 1 ) ]

 Poisson formula: P(x; μ) = (e-μ) (μx) / x!

 Mean of Poisson distribution = μx = μ

 Variance of Poisson distribution = σx2 = μ

 Multinomial formula: P = [ n! / ( n1! * n2! * ... nk! ) ] * ( p1n1 * p2n2 * . . . * pknk )

Linear Transformations

For the following formulas, assume that Y is a linear transformation of the random variable X, defined by the
equation: Y = aX + b.

 Mean of a linear transformation = E(Y) = Y = aX + b.

 Variance of a linear transformation = Var(Y) = a2 * Var(X).

 Standardized score = z = (x - μx) / σx.

 t statistic = t = (x - μx) / [ s/sqrt(n) ].

Estimation

 Confidence interval: Sample statistic + Critical value * Standard error of statistic

 Margin of error = (Critical value) * (Standard deviation of statistic)

 Margin of error = (Critical value) * (Standard error of statistic)

Hypothesis Testing

 Standardized test statistic = (Statistic - Parameter) / (Standard deviation of statistic)

 One-sample z-test for proportions: z-score = z = (p - P0) / sqrt( p * q / n )

 Two-sample z-test for proportions: z-score = z = z = [ (p1 - p2) - d ] / SE

 One-sample t-test for means: t statistic = t = (x - μ) / SE

 Two-sample t-test for means: t statistic = t = [ (x1 - x2) - d ] / SE

 Matched-sample t-test for means: t statistic = t = [ (x1 - x2) - D ] / SE = (d - D) / SE

 Chi-square test statistic = Χ2 = Σ[ (Observed - Expected)2 / Expected ]

Degrees of Freedom
The correct formula for degrees of freedom (DF) depends on the situation (the nature of the test statistic, the
number of samples, underlying assumptions, etc.).

 One-sample t-test: DF = n - 1

 Two-sample t-test: DF = (s12/n1 + s22/n2)2 / { [ (s12 / n1)2 / (n1 - 1) ] + [ (s22 / n2)2 / (n2 - 1) ] }

 Two-sample t-test, pooled standard error: DF = n1 + n2 - 2

 Simple linear regression, test slope: DF = n - 2

 Chi-square goodness of fit test: DF = k - 1

 Chi-square test for homogeneity: DF = (r - 1) * (c - 1)

 Chi-square test for independence: DF = (r - 1) * (c - 1)

Sample Size

Below, the first two formulas find the smallest sample sizes required to achieve a fixed margin of error, using
simple random sampling. The third formula assigns sample to strata, based on a proportionate design. The
fourth formula, Neyman allocation, uses stratified sampling to minimize variance, given a fixed sample size.
And the last formula, optimum allocation, uses stratified sampling to minimize variance, given a fixed budget.

 Mean (simple random sampling): n = { z2 * σ2 * [ N / (N - 1) ] } / { ME2 + [ z2 * σ2 / (N - 1) ] }

 Proportion (simple random sampling): n = [ ( z2 * p * q ) + ME2 ] / [ ME2 + z2 * p * q / N ]

 Proportionate stratified sampling: nh = ( Nh / N ) * n

 Neyman allocation (stratified sampling): nh = n * ( Nh * σh ) / [ Σ ( Ni * σi ) ]

 Optimum allocation (stratified sampling):


nh = n * [ ( Nh * σh ) / sqrt( ch ) ] / [ Σ ( Ni * σi ) / sqrt( ci ) ]

Statistics Tutorial

Common Statistical Formulas


Statistical formulas are used to calculate values related to statistical concepts or analyses. Here we will
discuss common formulas and what they stand for.
Population Mean

The term population mean, which is the average score of the population on a given variable, is
represented by:

μ = ( Σ Xi ) / N
The symbol ‘μ’ represents the population mean. The symbol ‘Σ Xi’ represents the sum of all scores
present in the population (say, in this case) X1 X2 X3 and so on. The symbol ‘N’ represents the total
number of individuals or cases in the population.
Population Standard Deviation
The population standard deviation is a measure of the spread (variability) of the scores on a given
variable and is represented by:

σ = sqrt[ Σ ( Xi – μ )2 / N ]
The symbol ‘σ’ represents the population standard deviation. The term ‘sqrt’ used in this statistical
formula denotes square root. The term ‘Σ ( Xi – μ )2’ used in the statistical formula represents the sum of
the squared deviations of the scores from their population mean.
Population Variance

The population variance is the square of the population standard deviation and is represented by:

σ2 = Σ ( Xi – μ )2 / N
The symbol ‘σ2’ represents the population variance.
Sample Mean

The sample mean is the average score of a sample on a given variable and is represented by:
x_bar = ( Σ xi ) / n
The term “x_bar” represents the sample mean. The symbol ‘Σ xi’ used in this formula represents the
represents the sum of all scores present in the sample (say, in this case) x1 x2 x3 and so on. The symbol
‘n,’ represents the total number of individuals or observations in the sample.
Sample Standard Deviation

The statistic called sample standard deviation, is a measure of the spread (variability) of the scores in the
sample on a given variable and is represented by:

s = sqrt [ Σ ( xi – x_bar )2 / ( n – 1 ) ]
The term ‘Σ ( xi – x_bar )2’ represents the sum of the squared deviations of the scores from the sample
mean.
Sample Variance

The sample variance is the square of the sample standard deviation and is represented by:

s2 = Σ ( xi – x_bar )2 / ( n – 1 )
The symbol ‘s2’ represents the sample variance.
Pooled Sample Standard Deviation

The pooled sample standard deviation is a weighted estimate of spread (variability) across multiple
samples. It is represented by:

sp = sqrt [ (n1 – 1) * s12 + (n2 – 1) * s22 ] / (n1 + n2 – 2) ]


The term ‘sp’ represents the pooled sample standard deviation. The term ‘n1’ represents the size of the
first sample, and the term ‘n2’ represents the size of the second sample that is being pooled with the first
sample. The term ‘s12’ represents the variance of the first sample, and ‘s22’ represents the variance of the
second sample.
Tools to Calculate Sample Size and Power Analysis
Statistics Solutions offers free tools to calculate appropriate sample sizes and conduct power analysis for
your dissertation or research study. Our sample size calculator is available for free when you sign up for
the dissertation toolbox membership.
The free toolbox membership also includes access to our Sample Size / Power Analysis Write-up
Generator with References tool.

The Sample Size/Power Analysis Calculator with Write-up is a tool for anyone struggling with power
analysis. Simply identify the test to be conducted and the degrees of freedom where applicable
(explained in the document), and the sample size/power analysis calculator will calculate your sample
size for a power of .80 of an alpha of .05 for small, medium and large effect sizes. The calculator then
presents the write-up with references which can easily be integrated in your dissertation
document. Click here for a sa

2.5: The Empirical Rule and Chebyshev's


Theorem
1. Last updated
Sep 19, 2017
2. Share
Table of contents
SKILLS TO DEVELOP
 To learn what the value of the standard deviation of a data set implies about how the data
scatter away from the mean as described by the Empirical Rule and Chebyshev’s Theorem.
 To use the Empirical Rule and Chebyshev’s Theorem to draw conclusions about a data set.

You probably have a good intuitive grasp of what the average of a data set says about that data set. In
this section we begin to learn what the standard deviation has to tell us about the nature of the data
set.

The Empirical Rule


We start by examining a specific set of data. Table 2.5.12.5.1 shows the heights in inches
of 100100 randomly selected adult men. A relative frequency histogram for the data is shown
in Figure 2.5.12.5.1. The mean and standard deviation of the data are, rounded to two decimal
places, x¯=69.92x¯=69.92 and σ=1.70σ=1.70.

Table 2.5.12.5.1: Heights of Men

72.3 71.3 72.5 70.6 68.2 70.1 68.4


70.5 71.0 70.9 69.3 69.4 69.7 69.1

70.0 70.4 68.9 69.4 69.4 69.2 70.7

69.8 68.6 69.5 71.6 66.2 72.4 70.7

69.3 68.9 74.8 68.0 71.2 68.3 70.2

72.2 70.0 68.7 67.9 71.1 69.0 70.8

68.8 67.2 73.0 70.4 67.8 70.0 69.5

67.6 67.0 70.3 71.2 65.6 68.1 70.8

67.5 71.3 71.5 71.0 69.1 69.5 71.1

72.7 72.8 69.6 65.9 68.0 69.7 68.7

If we go through the data and count the number of observations that are within one standard deviation
of the mean, that is, that
are between 69.92−1.70=68.2269.92−1.70=68.22 and 69.92+1.70=71.6269.92+1.70=71.62 inches,
there are 6969 of them. If we count the number of observations that are within two standard deviations
of the mean, that is, that are
between 69.92−2(1.70)=66.5269.92−2(1.70)=66.52 and 69.92+2(1.70)=73.3269.92+2(1.70)=73.3
2 inches, there are 9595 of them. All of the measurements are within three standard deviations of the
mean, that is,
between 69.92−3(1.70)=64.82269.92−3(1.70)=64.822 and 69.92+3(1.70)=75.0269.92+3(1.70)=75
.02 inches. These tallies are not coincidences, but are in agreement with the following result that has
been found to be widely applicable.
Figure 2.5.12.5.1: Heights of Adult Men

The Empirical Rule


Approximately 68%68% of the data lie within one standard deviation of the mean, that is, in the
interval with endpoints x¯±sx¯±s for samples and with endpoints μ±σμ±σ for populations; if a data
set has an approximately bell-shaped relative frequency histogram, then (Figure 2.5.22.5.2)

 approximately 95%95% of the data lie within two standard deviations of the mean, that is, in
the interval with endpoints x¯±2sx¯±2s for samples and with endpoints μ±2σμ±2σ for
populations; and
 approximately 99.7%99.7% of the data lies within three standard deviations of the mean, that
is, in the interval with endpoints x¯±3sx¯±3s for samples and with endpoints μ±3σμ±3σ for
populations.
Figure 2.5.22.5.2: The Empirical Rule

Two key points in regard to the Empirical Rule are that the data distribution must be
approximately bell-shaped and that the percentages are only approximately true. The Empirical Rule
does not apply to data sets with severely asymmetric distributions, and the actual percentage of
observations in any of the intervals specified by the rule could be either greater or less than those
given in the rule. We see this with the example of the heights of the men: the Empirical Rule suggested
68 observations between 68.2268.22 and 71.6271.62inches, but we counted 6969.

EXAMPLE 2.5.12.5.1
Heights of 1818-year-old males have a bell-shaped distribution with mean 69.669.6 inches and
standard deviation 1.41.4 inches.

1. About what proportion of all such men are between 68.268.2 and 7171 inches tall?
2. What interval centered on the mean should contain about 95%95% of all such men?

Solution:

A sketch of the distribution of heights is given in Figure 2.5.32.5.3.

1. Since the interval from 68.268.2 to 71.071.0 has endpoints x¯−sx¯−s and x¯+sx¯+s, by the
Empirical Rule about 68%68% of all 1818-year-old males should have heights in this range.
2. By the Empirical Rule the shortest such interval has
endpoints x¯−2sx¯−2s and x¯+2sx¯+2s. Since

x¯−2s=69.6−2(1.4)=66.8(2.5.1)(2.5.1)x¯−2s=69.6−2(1.4)=66.8

and

x¯+2s=69.6+2(1.4)=72.4(2.5.2)(2.5.2)x¯+2s=69.6+2(1.4)=72.4

the interval in question is the interval from 66.866.8 inches to 72.472.4 inches.
Figure 2.5.32.5.3: Distribution of Heights

EXAMPLE 2.5.22.5.2
Scores on IQ tests have a bell-shaped distribution with mean μ=100μ=100 and standard
deviation σ=10σ=10. Discuss what the Empirical Rule implies concerning individuals with IQ scores
of 110110, 120120, and 130130.

Solution:

A sketch of the IQ distribution is given in Figure 2.5.32.5.3. The Empirical Rule states that

1. approximately 68%68% of the IQ scores in the population lie between 9090 and 110110,
2. approximately 95%95% of the IQ scores in the population lie between 8080 and 120120,
and
3. approximately 99.7%99.7% of the IQ scores in the population lie between 7070 and 130130.

Figure 2.5.32.5.3: Distribution of IQ Scores.


1. Since 68%68% of the IQ scores lie within the interval from 9090 to 110110, it must be the
case that 32%32% lie outside that interval. By symmetry approximately half of that 32%32%,
or 16%16% of all IQ scores, will lie above 110110. If 16%16% lie above 110110,
then 84%84% lie below. We conclude that the IQ score 110110 is the 84th84th percentile.
2. The same analysis applies to the score 120120. Since approximately 95%95% of all IQ scores
lie within the interval form 8080 to 120120, only 5%5% lie outside it, and half of them,
or 2.5%2.5% of all scores, are above 120120. The IQ score 120120 is thus higher
than 97.5%97.5% of all IQ scores, and is quite a high score.
3. By a similar argument, only 15/10015/100 of 1%1% of all adults, or about one or two in every
thousand, would have an IQ score above 130130. This fact makes the score 130130 extremely
high.

Chebyshev’s Theorem
The Empirical Rule does not apply to all data sets, only to those that are bell-shaped, and even then is
stated in terms of approximations. A result that applies to every data set is known as Chebyshev’s
Theorem.

CHEBYSHEV’S THEOREM
For any numerical data set,

 at least 3/43/4 of the data lie within two standard deviations of the mean, that is, in the interval
with endpoints x¯±2sx¯±2s for samples and with endpoints μ±2σμ±2σ for populations;
 at least 8/98/9 of the data lie within three standard deviations of the mean, that is, in the
interval with endpoints x¯±3sx¯±3s for samples and with endpoints μ±3σμ±3σ for
populations;
 at least 1−1/k21−1/k2 of the data lie within kk standard deviations of the mean, that is, in the
interval with endpoints x¯±ksx¯±ks for samples and with endpoints μ±kσμ±kσ for
populations, where kk is any positive whole number that is greater than 11.

Figure 2.5.42.5.4 gives a visual illustration of Chebyshev’s Theorem.


Figure 2.5.42.5.4: Chebyshev’s Theorem

It is important to pay careful attention to the words “at least” at the beginning of each of the three
parts of Chebyshev’s Theorem. The theorem gives the minimum proportion of the data which must lie
within a given number of standard deviations of the mean; the true proportions found within the
indicated regions could be greater than what the theorem guarantees.

EXAMPLE 2.5.32.5.3
A sample of size n=50n=50 has mean x¯=28x¯=28 and standard deviation s=3s=3. Without knowing
anything else about the sample, what can be said about the number of observations that lie in the
interval (22,34)(22,34)? What can be said about the number of observations that lie outside that
interval?

Solution:

The interval (22,34)(22,34) is the one that is formed by adding and subtracting two standard
deviations from the mean. By Chebyshev’s Theorem, at least 3/43/4 of the data are within this
interval. Since 3/43/4 of 5050 is 37.537.5, this means that at least 37.537.5 observations are in the
interval. But one cannot take a fractional observation, so we conclude that at least 3838observations
must lie inside the interval (22,34)(22,34).
If at least 3/43/4 of the observations are in the interval, then at most 1/41/4 of them are outside it.
Since 1/41/4 of 5050 is 12.512.5, at most 12.512.5 observations are outside the interval. Since again
a fraction of an observation is impossible, x(22,34)x(22,34).

EXAMPLE 2.5.42.5.4
The number of vehicles passing through a busy intersection
between 8:00a.m.8:00a.m. and 10:00a.m.10:00a.m. was observed and recorded on every weekday
morning of the last year. The data set contains n=251n=251 numbers. The sample mean
is x¯=725x¯=725 and the sample standard deviation is s=25s=25. Identify which of the following
statements must be true.

1. On approximately 95%95% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was
between 675675 and 775775.
2. On at least 75%75% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was
between 675675 and 775775.
3. On at least 189189 weekday mornings last year the number of vehicles passing through the
intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was
between 675675 and 775775.
4. On at most 25%25% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was either less
than 675675 or greater than 775775.
5. On at most 12.5%12.5% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was less than 675675.
6. On at most 25%25% of the weekday mornings last year the number of vehicles passing
through the intersection from 8:00a.m.8:00a.m. to 10:00a.m.10:00a.m. was less than 675675.

Solution:

1. Since it is not stated that the relative frequency histogram of the data is bell-shaped, the
Empirical Rule does not apply. Statement (1) is based on the Empirical Rule and therefore it
might not be correct.
2. Statement (2) is a direct application of part (1) of Chebyshev’s Theorem
because x¯−2sx¯−2s, x¯+2s=(675,775)x¯+2s=(675,775). It must be correct.
3. Statement (3) says the same thing as statement (2)
because 75%75% of 251251 is 188.25188.25, so the minimum whole number of observations
in this interval is 189189. Thus statement (3) is definitely correct.
4. Statement (4) says the same thing as statement (2) but in different words, and therefore is
definitely correct.
5. Statement (4), which is definitely correct, states that at most 25%25% of the time either fewer
than 675675 or more than 775775 vehicles passed through the intersection. Statement (5) says
that half of that 25%25% corresponds to days of light traffic. This would be correct if the
relative frequency histogram of the data were known to be symmetric. But this is not stated;
perhaps all of the observations outside the interval (675,775675,775) are less than 7575. Thus
statement (5) might not be correct.
6. Statement (4) is definitely correct and statement (4) implies statement (6): even if every
measurement that is outside the interval (675,775675,775) is less than 675675 (which is
conceivable, since symmetry is not known to hold), even so at most 25%25% of all
observations are less than 675675. Thus statement (6) must definitely be correct.

KEY TAKEAWAY
 The Empirical Rule is an approximation that applies only to data sets with a bell-shaped
relative frequency histogram. It estimates the proportion of the measurements that lie within
one, two, and three standard deviations of the mean.
 Chebyshev’s Theorem is a fact that applies to all possible data sets. It describes the minimum
proportion of the measurements that lie must within one, two, or more standard deviations of
the mean.

Contributor
 Anonymous
1. Back to top
2.
o 2.4: Relative Position of Data

o 2.E: Descriptive Statistics (Exercises)

Recommended articles

1. 7.1: Large Sample Estimation of a Population MeanA confidence interval for a population mean is
an estimate of the population mean together with an indication of reliability. There are different
form...
2. 2.E: Descriptive Statistics (Exercises)These are homework exercises to accompany the Textmap
created for "Introductory Statistics" by Shafer and Zhang. Complementary General Chemistry
quest...
3. 2.1: Three Popular Data DisplaysGraphical representations of large data sets provide a quick
overview of the nature of the data. A population or a very large data set may be repr...
4. 2.2: Measures of Central LocationThe mean, the median, and the mode each answer the question
“Where is the center of the data set?” The nature of the data set, as indicated by a relat...
5. 2.3: Measures of VariabilityThe range, the standard deviation, and the variance each give a
quantitative answer to the question “How variable are the data?”

1. The LibreTexts libraries are Powered by MindTouch® and are based upon work supported
by the National Science Foundation under grant numbers: 1246120, 1525057, and
1413739. Unless otherwise noted, the LibreTexts library is licensed under a Creative
Commons Attribution-Noncommercial-Share Alike 3.0 United States License. Permissions
beyond the scope of this license may be available at copyright@ucdavis.edu.