Module 3 (301 SI-1)
Module 3 (301 SI-1)
Module 3 (301 SI-1)
MODULE-III
STATISTICAL INFERENCE 1
Hypothesis Testing
Introduction:
Statistical Inference is a branch of Statistics which uses probability concepts to deal with
uncertainty in decision making. There are a number of situations where in we come across
problems involving decision making. For example, consider the problem of buying 1 kilogram
of rice, when we visit the shop, we do not check each and every rice grains stored in a gunny
bag; rather we put our hand inside the bag and collect a sample of rice grains. Then analysis
takes place. Based on this, we decide to buy or not. Thus, the problem involves studying whole
rice stored in a bag using only a sample of rice grains.
rejection under the assumption that it is true. Next we set up another hypothesis called
alternate statement which is just opposite of null statement; denoted by H1 which is just
complimentary to the null hypothesis. Therefore, if we start with H0 :μ=μ0 then alternate
As we are studying population parameter based on some sample study, one can not do the job
with 100% accuracy since sample is drawn from the population and possible sample may not
represent the whole population. Therefore, usually we conduct analysis at certain level of
significance (lower than 100%. The possible choices include 99%, or 95% or 98% or 90%.
Usually we conduct analysis at 99% or 95% level of significance, denoted by the symbol
. We test H0 against H1 at certain level of significance. The confidence with which a
person rejects or accepts H0 depends upon the significance level adopted. It is usually
expressed in percentage forms such as 5% or 1% etc. Note that when is set as 5%, then
probability of rejecting null hypothesis when it is true is only 5%. It also means that when the
hypothesis in question is accepted at 5% level of significance, then statistician runs the risk of
taking wrong decisions, in the long run, is only 5%. The above is called II step of hypothesis
testing.
Critical values or Fiducial limit values for a two tailed test:
Sl. No Level of significance Theoretical Value
1 = 1% 2.58
2 = 2% 2.33
3 = 5% 1.96
Critical values or Fiducial limit values for a single tailed test (right and test)
Tabulated value = 1% = 5% = 10%
Setting a test criterion: The third step in hypothesis testing procedure is to construct a test
criterion. This involves selecting an appropriate probability distribution for the particular test
i.e. a proper probability distribution function to be chosen. Some of the distribution functions
used are t, F, when the sample size is small (size lower than 30). However, for large samples,
normal distribution function is preferred. Next step is the computation of statistic using the
sample items drawn from the population. Usually, samples are drawn from the population by
a procedure called random, where in each and every data of the population has the same chance
of being included in the sample. Then the computed value of the test criterion is compared
with the tabular value; as long the calculated value is lower then or equal to tabulated value,
we accept the null hypothesis, otherwise, we reject null hypothesis and accept the alternate
hypothesis. Decisions are valid only at the particular level significance of level adopted.
During the course of analysis, there are two types of errors bound to occur. These are (i) Type
– I error and (ii) Type – II error.
Type – I error: This error usually occurs in a situation, when the null hypothesis is true, but
we reject it i.e. rejection of a correct/true hypothesis constitute type I error.
Type – II error: Here, null hypothesis is actually false, but we accept it. Equivalently,
accepting a hypothesis which is wrong results in a type – II error. The probability of
committing a type – I error is denoted by where
= Probability of making type I error = Probability [Rejecting H0 | H0 is true]
On the other hand, type – II error is committed by not rejecting a hypothesis when it is false.
The probability of committing this error is denoted by . Note that
= Probability of making type II error = Probability [Accepting H1 | H1 is false]
Critical region:
A region in a sample space S which amounts to Rejection of H0 is termed as critical region.
trial, q stands for the chance that there is a failure in the trial and n refers to the size of
the sample.
2. The magnitude of standard error gives an index of the precision of the parameter.
Probable limits of population mean are:
σ
X 1.96
95% fiducial limits of population mean are n
99% fiducial limits of population mean are X 2.58 σ . Further, test criterion z cal =
x-μ
n S.E.
The binomial distribution is regarded as the sampling distribution of the number of successes
in the sample. We know that the mean of this distribution is np and S.D is √𝑛𝑝𝑞. S.E proportion
𝑝𝑞
of successes √ 𝑛 .
𝑥−𝑛𝑝
The associated standard normal variate Z be defined by 𝑧 = . The probable limits 𝑝 ±
√𝑛𝑝𝑞
𝑝𝑞
2.58√ 𝑛 .
Problems:
1. A coin is tossed 400 times and the head turned up 216 times. Test the hypothesis that the
coin is un– biased?
Solution: First we construct null and alternate hypotheses set up H0 : The coin is not a biased
one. Set up H1 : Yes, the coin is biased. As the coin is assumed be fair and it is tossed 400
times, clearly we must expect 200 times heads occurring and 200 times tails. Thus, expected
number of heads is 200. But the observed result is 216. There is a difference of 16. Further,
standard error is σ= npq . With p = ½, q = ½ and n = 400, clearly = 10 . The test criterion
2. A person throws a 10 dice 500 times and obtains 2560 times 4, 5, or 6. Can this be attributed
to fluctuations in sampling?
Solution: As in the previous problem first we shall set up H 0 : The die is fair and H1 : The
die is unfair. We consider that problem is based on a two – tailed test. Let us choose level of
significance as = 5% then, the tabulated value is 1.96. Consider computing test criterion,
Expected value - observed result
zcal = ; here, as the dice is tossed by a person 5000 times, and
standard error
on the basis that die is fair, then chance of getting any of the 6 numbers is 1/6. Thus, chance
of getting either 4 or 5, or 6 is p = ½. Also, q = ½. With n = 5000, standard error, σ= npq
limits to the percentage of foggy days is given by p±3 pqn . Using the data available in this
problem, one obtains the answer as 0.12 3 0.12 88 1000 . Equivalently, 8.91% to 15.07%.
4. A die was thrown 9000 times and a throw of 5 or 6 was obtained 3240 times. On the
assumption of random throwing, do the data indicate that die is biased?
Solution: We set up the null hypothesis as H0 : Die is un - biased. Also, H1 : Die is biased. .
Let us take level of significance as α=5% . Based on the assumption that distribution is
normally distributed, the tabulated value is 1.96. The chance of getting each of the 6 numbers
is same and it equals to 1/6 therefore chance of getting either 5 or 6 is 1/3. In a throw of 9000
1
times, getting the numbers either 5 or 6 is ×9000 = 3000 . Now the difference in these two
3
results is 240. With p = 1/3, q = 2/3, n = 9000, S.E.= npq = 44.72. Now consider the test
Difference 240
criterion zcal = = = = 5.367 which is again more than the tabulated value.
S.E. 44.72
Therefore, we reject null hypothesis and accept the alternate that die is highly biased.
σ
population is unknown, then S.E (X) = where is the standard deviation of the sample.
n
Fiducial limits of population mean are:
σ
95% fiducial limits of population mean are X 1.96
n
σ x-μ
99% fiducial limits of population mean are X 2.58 . Further, test criterion z cal =
n S.E.
Problems:
1. A sample of 100 tyres is taken from a lot. The mean life of tyres is found to be 39, 350 kilo
meters with a standard deviation of 3, 260. Could the sample come from a population with
mean life of 40, 000 kilometers? Establish 99% confidence limits within which the mean life
of tyres is expected to lie.
Solution: First we shall set up null hypothesis, H0 : = 40,000 , alternate hypothesis as
H1 : 40,000 . We consider that the problem follows a two tailed test and
chose = 5% . Then corresponding to this, tabulated value is 1.96. Consider the expression
x -
for finding test criterion, zcal = . Here, =40, 000, x = 39, 350 and = 3, 260 , n =
S.E.
3,260
100. S.E. = = = 326 . Thus, zcal = 1.994. As this value is slightly greater than 1.96,
n 100
we reject the null hypothesis and conclude that sample has not come from a population of 40,
000 kilometers.
The 99% confidence limits within which population mean is expected to lie is given as
H1 : 1,600 hours . We consider that the problem follows a two tailed test and chose = 5%
. Then corresponding to this, tabulated value is 1.96. Consider the expression for finding test
x -
criterion, zcal = . Here, =1, 600, x = 1, 570, n = 400 , = 150 hours so that using all
S.E.
these values above, it can be seen that zcal = 4.0 which is really greater than 1.96. Hence, we
𝑋̅ − 𝜇 1180 − 1200
𝑍= 𝜎 = = −1.41
100
√𝑛 √50
The critical (tabulated) values at 5% level of significance are 𝑧𝛼 = −1.96.
Hence, we have to accept the null hypothesis and to reject the alternate hypothesis.
4. A manufacturer of ball point pens claims that a certain pen manufactured by him has a mean
writing-life at least 460 A-4 size pages. A purchasing agent selects a sample of 100 pens and put
them on the test. The mean writing-life of the sample found 453 A-4 size pages with standard
deviation 25 A-4 size pages. Should the purchasing agent reject the manufacturer’s claim at 1%
level of significance?
Solution: Here, we are given that
Specified value of population mean = μ = 460,
Sample size = n = 100,
Sample mean =𝑋̅ = 453,
Sample standard deviation = σ = 25
𝐻0 : 𝜇 ≥ 460
𝐻1 : 𝜇 < 460
Thus, for testing the null hypothesis the test statistic is given by
𝑋̅ − 𝜇 453 − 460
𝑍= 𝜎 = = −2.8
25
√𝑛 √100
The critical (tabulated) values at 1% level of significance is 𝑧𝛼 = −2.33.
Hence, we have to reject the null hypothesis and to accept the alternate hypothesis.
Consider two populations P1 and P2. Let S1 and S2 be two samples drawn at random from
these two different populations. Suppose we have the following data about these two samples,
say
S1 n1 x1 1
S2 n2 x2 2
then standard error of difference between the means of two samples S1 and S2 is
2 22 Difference of sample means
S.E = 1
+ and the test criterion is Zcal = . The rest of
n1 n2 Standard error
the analysis is same as in the preceding sections.
When the two samples are drawn from the same population, then standard error is
When the standard deviations are un – known, then standard deviations of the two samples
s12 s 22
must be replaced. Thus, S.E = + where s1 and s2 are standard deviations of the
n1 n 2
two samples considered in the problem.
Problems:
1. Intelligence test on two groups of boys and girls gave the following data:
Girls 70 20 250
Is there a significant difference in the mean scores obtained by boys and girls?
Solution: We set up null hypothesis as H0 : there is no significant difference between the mean
scores obtained by boys and girls. The alternate hypothesis is considered as H1 : Yes, there is
a significant difference in the mean scores obtained by boys and girls. We choose level of
Difference of means
significance as = 5% so that tabulated value is 1.96. Consider zcal =
Standard Error
152 202
The standard error may be calculated as S.E = + =1.761 , The test criterion is
150 250
75 - 70
zcal = = 2.84 . As 2.84 is more than 1.96, we have to reject null hypothesis and to accept
1.761
alternate hypothesis that there are some significant difference in the mean marks scored by
boys and girls.
2. A man buys 50 electric bulbs of “Philips” and 50 bulbs of “Surya”. He finds that Philips
bulbs give an average life of 1,500 hours with a standard deviation of 60 hours and Surya bulbs
gave an average life of 1, 512 hours with a standard deviation of 80 hours. Is there a significant
difference in the mean life of the two makes of bulbs?
Solution: we set up null hypothesis, H0 : there is no significant difference between the bulbs
made by the two companies, the alternate hypothesis can be set as H1 : Yes, and there could
be some significant difference in the mean life of bulbs. Taking = 1% and =5% , the
respective tabulated values are 2.58 and 1.96. Consider standard error is
3. A random sample of size N=100 is taken from a population with standard deviation 𝜎 =
5.1. Given that the sample mean is 𝑋̅ = 21.6. Obtain the 95% confidence interval for the
population mean 𝜇.
Solution: N=100, 𝜎 = 5.1 , 𝑋̅ = 21.6
Confidence limits for the population mean are
𝑠 5.1
̅𝑋
̅̅̅ ± 𝑍𝑐 = 21.6 ± 𝑍𝑐
√𝑁 √100
4. One type of aircraft is found to develop engine trouble in 5 flights out of 100 flights and
another type in 7 flights out of 200 flights. Is there a significant difference in the two types of
air crafts so far as engine defects are concerned?
Solution: Let P1 be the proportion of type 1 aircrafts that develop engine trouble, then
P1= 5/100 = 0.05 in the sample of size N1 = 100.
Let P2 be the proportion of type 2 aircrafts that develop engine trouble, then
P2= 7/200 = 0.035 in the sample of size N2 = 200.
Let us make an hypothesis that there is no difference in the two types of aircrafts.
Then the mean of the distribution of differences in proportions is zero; that is
𝜇(𝑝1 −𝑝2 ) = (𝑝1 − 𝑝2 ) = 0and the standard deviation of the distribution is
𝑝1 𝑞1 𝑝2 𝑞2 0.05(0.95) 0.035(0.965)
𝜎(𝑝1 −𝑝2 ) = √ + =√ + = 0.0254.
𝑁1 𝑁2 100 200
This Z − score is less than Zc = 1.96 and Zc = 2.58. Therefore, the hypothesis
cannot be rejected. This means that the difference between the two types of aircrafts is
not significant.
5. A sample height of 6400 soldiers has a mean of 172.34 cms and a standard deviation of 6.5
cms, while a sample of heights of 1600 sailors has a mean of 174.12 cms and a standard
deviations of 6.4 cms. Does the data indicate that the sailors are on the average taller than
soldiers. Use the left – tailed test.
Solution: Here,
𝑁1 = 6400, 𝑋̅1 = 172.34, 𝑠1 = 6.5, 𝑁2 = 1600, 𝑋̅2 = 174.12, 𝑠2 = 6.4 .
We test the Null hypothesis
H : There is no difference in the heights of soldiers and sailors(on the average), against the
alternative hypothesis.
H1 : Sailors are taller than soldiers, on the average; that is 𝜇2 > 𝜇1 , 𝑜𝑓 𝜇1 − 𝜇2 < 0.
Under the hypothesis H, we have
𝜇(𝑋̅1 −𝑋̅2) = 𝜇1 − 𝜇2 = 0 and
𝑠 2 𝑠2 2 (6.5)2 (6.4)2
𝜎(𝑋̅1 −𝑋̅2) = √ 𝑁1 + = √ 6400 + = 0.179.
1 𝑁2 1600
This score lies in the critical region 𝑧 < −𝑧𝑐 for the left – tailed test at both of 0.01 and 0.05
levels of significance. Therefore, we reject the null hypothesis H. Consequently, we accept the
alternative hypothesis H1. Thus, the data indicates that, on the average, the sailors are taller
than soldiers.
6. In a sample of 500 men it was found that 60% of them had overweight. Find the 99% of
confidence limits for the proportion of men in the population having overweight.
Solution: Probability of persons having over weight p = 60% = .60
q = 40% = .40
7. In two samples of women from Punjab and Tamilnadu, the mean height of 1000 and 2000 women
are 67.6 and 68.0 inches respectively. If population standard deviation of Punjab and Tamilnadu
are same and equal to 5.5 inches then, can the mean heights of Punjab and Tamilnadu women be
regarded as same at 1% level of significance?
Solution: We are given
𝑛1 = 1000, 𝑛2 = 2000, 𝑥̅ = 67.6, 𝑦̅ = 68.0, 𝜎1 = 𝜎2 = 5.5
𝐻0 : 𝜇1 = 𝜇2
𝐻1 : 𝜇1 ≠ 𝜇2
Thus, for testing the null hypothesis the test statistic is given by
𝑥̅ − 𝑦̅
𝑧=
𝜎2 𝜎2
√ 1 + 2
𝑛1 𝑛2
67.6 − 68.0
𝑧= = −1.88
2 2
√(5.5) + (5.5)
1000 2000
80.4 − 74.3
𝑧= = 2.23
2 2
√(12.8) + (20.5)
50 100
Here, we want to test that the die is unbiased and we know that if die is unbiased then proportion
or probability of getting 2 or 5 is
𝑝0 be probability of getting a 2 or 5
= Probability of getting 2 + Probability of getting 5
2 1 2
𝑝0 = = ; 𝑞0 = 1 − 𝑝0 =
6 3 3
𝐻0 : 𝑝 = 𝑝0
𝐻1 : 𝑝 ≠ 𝑝0
Thus, for testing the null hypothesis the test statistic is given by
𝑝 − 𝑝0 0.3444 − 0.3333
𝑧= = = 2.22
𝑝 𝑞
√ 0 0 √0.3333 × 0.6667
𝑛 900
11. In a random sample of 100 persons from town A, 60 are found to be high consumers of wheat.
In another sample of 80 persons from town B, 40 are found to be high consumers of wheat. Do
these data reveal a significant difference between the proportions of high wheat consumers in town
A and town B (at α = 0.05)?
Solution: Here, attribute under study is high consuming of wheat. And we define our success and
failure as getting a person of high consumer of wheat and not high consumer of wheat respectively.
We are given that
𝑛1 = total number of persons in the sample of town A = 100
𝑛2 = total number of persons in the sample of town B = 80
𝑋1= number of persons of high consumer of wheat in town A = 60
𝑋2= number of persons of high consumer of wheat in town B = 40
The sample proportion of high wheat consumers in town A is
𝑋1 60
𝑝1 = = = 0.60
𝑛1 100
and the sample proportion of wheat consumers in town B is
𝑋2 40
𝑝2 = = = 0.50
𝑛2 80
𝐻0 : 𝑝1 = 𝑝2
𝐻1 : 𝑝1 ≠ 𝑝2
The estimate of the combined proportion (P) of high wheat consumers in two towns is given by
𝑋1 + 𝑋2 60 + 40 5 4
𝑃= = = ; 𝑄 =1−𝑃 =
𝑛1 + 𝑛2 100 + 80 9 9
Thus, for testing the null hypothesis the test statistic is given by
𝑝1 − 𝑝2 0.60 − 0.50
𝑧= = = 1.34
1 1
√𝑃𝑄 ( + ) √5 × 4 ( 1 + 1 )
𝑛1 𝑛2 9 9 100 80
𝑋2 30
𝑝2 = = = 0.10
𝑛2 300
𝐻0 : 𝑝1 ≤ 𝑝2
𝐻1 : 𝑝1 > 𝑝2
Since P is unknown, so the pooled estimate of proportion is given by
𝑋1 + 𝑋2 60 + 30 9 61
𝑃= = = ; 𝑄 = 1−𝑃 =
𝑛1 + 𝑛2 400 + 300 70 70
Thus, for testing the null hypothesis the test statistic is given by
𝑝1 − 𝑝2 0.15 − 0.10
𝑧= = = 1.95
1 1
√𝑃𝑄 ( + ) √ 9 × 61 ( 1 + 1 )
𝑛1 𝑛2 70 70 400 300
Video links: