Testing of Hypotheses

Introduction
• Significance tests (or Tests of Hypotheses) is an integral part of statistical
inference / Inferential analytics.

• Any significance test procedure has five distinct steps viz.

1. Making assumptions : random sampling, ……

2. Constructing hypotheses

3. Determining the test statistic

4. P-values

5. Drawing conclusion
Steps: Constructing hypotheses
• 
Each significance test is composed of two hypotheses:

Null hypotheses (H0) : It is a statement that specifies a particular value for the
parameter (in our case, p or ) which is pre-determined from experience and/or
prior belief.

Eg : You believe that on an average a WeSchool student spends 15 hours per week
discussing/solving cases outside of class i.e the null hypotheses will be ……..
Steps: Constructing hypotheses

Alternative hypotheses (H𝛼) : It states that the population parameter

takes values in some alternative parameter space (than what is stated
by the null). It may be one or two sided.

• Eg : Suppose on the contrary, your friend believes that a WeSchool

student spends more than 15 hours per week solving cases i.e the
alternative hypotheses will be ……………..
Steps: Determining the test statistic

• A test statistic measures how close the point estimate of the

population parameter is to the null hypotheses value (of the
parameter).

• This “closeness“ is measured in terms of the standard error of the

point estimate.
Steps: P-values
• 
• p-values represent the amount of evidence against the null
hypotheses based on the available data.

• Smaller the p-value, stronger is the evidence against the null

hypotheses and vice versa.

• The smallness of the p-value is measured with respect to the

significance level ().
Steps: Drawing conclusion
• 
• In order to come to a definite conclusion (about rejecting or not
rejecting the null hypotheses), we compare the p-value with the
significance level (denoted by )

• The significance level is usually set at 0.05, 0.1 or 0.01

• We would reject H0 at a given significance level , if p value and fail to

reject H0 if if p value
 Significance Tests for Population
Proportion
•1.  Assumptions :
• Random sample.
• Sample size (n) should be large enough such that np0 10 and n(1 − p0) .
2. Hypotheses :
• Null : p = p0 , p p0 , p p0
• Alternative : p p0 , p p0 , p p0
where p0 is known as the null value of p.
3. Test statistic, Zobs : Sample statistic – Null Value
Standard error of sample statistic
Significance Tests for Population
Proportion
•4.  P-value : The p-value would depend on the direction of the alternative
hypothesis as follows :
• If H𝛼 : p > p0 , p-value will be the right tailed area above the observed value of
the test statistic (Zobs) under the standard normal curve.
• If H𝛼 : p p0 , p-value will be the left tailed area below the observed value of
the test statistic (Zobs) under the standard normal curve.
• If H𝛼 : p p0, p-value will be the two tailed area beyond the observed value of
the test statistic under the standard normal curve. Since the normal curve is
symmetric, it can also be calculated as twice the one-tailed area above (or
below) the observed value of the test statistic.
Significance Tests for Population
Proportion
• 
5. Drawing conclusion : We will reject H0 if p value and fail to reject H0
otherwise.
Critical Values for Two-tailed and
One-tailed
Significance Two-tailed test Left-tailed test Right-tailed test
level

0.10 1.645 -1.28 1.28

0.05 1.96 -1.645 1.645

0.01 2.575 -2.33 2.33

Example

• Female managers : Traditionally, the percentage of managers who are

female in the Indian corporate sector has been pretty low, about 18%.
The HRD ministry wants to know whether the percentage has
improved during recent times. Accordingly, a random sample of 100
managers were chosen and 25 of them were females. Perform an
appropriate test of hypotheses for the above problem.
Example
• According to Indian census, the percentage of workers engaged in
agriculture among all workers is 56.6%. A sample of 100 workers were taken
recently and the percentage of workers engaged in agriculture was found to
be 53.6%.

• A) Test whether the proportion of workers engaged in agriculture has

changed over the last few years. Take a significance level of 0.05.

• B) test whether the proportion of workers engaged in agriculture has

decreased over the last few years with a significance level of 0.05.
 Significance Tests for Population
Mean
•1.  Assumptions :
• Random sample.
• Sample size (n) should be large enough such that population distribution
is approximately normal.
2. Hypotheses :
• Null : = 0 , 0 , 0
• Alternative : 0, 0, 0

where 0 is known as the null value of .

3. Test statistic, tobs : Sample statistic – Null Value
Standard error of sample statistic
 Significance Tests for Population
Mean
•4.  P-value : The p-value would depend on the direction of the alternative
hypothesis as follows :
• If H𝛼 : > 0 , p-value will be the right tailed area above the observed value of the
test statistic (tobs) under a t n-1 curve.
• If H𝛼 : 0 , p-value will be the left tailed area below the observed value of the
test statistic (tobs) under a t n-1 curve.
• If H𝛼 : , p-value will be the two tailed area beyond the observed value of the
0

test statistic under the t n-1 curve. Since the t distribution is symmetric, it can
also be calculated as twice the one-tailed area above (or below) the observed
value of the test statistic.
 Significance Tests for Population
Mean
•  5. Drawing conclusion : We will reject H0 if value and fail to reject H0
otherwise.
Critical Values for Two-tailed and
One-tailed
Significance Two-tailed test Left-tailed test Right-tailed test
level

0.10 1.645 -1.28 1.28

0.05 1.96 -1.645 1.645

0.01 2.575 -2.33 2.33

Example
• An automatic filling machine has to fill eye drops with a mean of 10 ml per
bottle. Both overfilling and under filling are not desirable. A quality control
inspector takes a sample of 30 bottles in every half an hour in order to
decide whether he has to stop operation in the case of overfilling and
under filling beyond a certain level for adjusting the machine. If the sample
has a mean of 9.8 ml with a standard deviation of 0.5 ml, give the decision
rule for the quality control inspector taking 0.05 as the significance level.
Example

• 1,500 women followed a particular diet for a month. A random

sample of 39 women gained an average of 6.7 pounds. Test the
hypothesis that the average weight gain per woman for the month
was over 5 pounds. The standard deviation for all women in the group
was 7.1.
 Errors in Hypotheses Tests

• Type I error : We commit a type I error when we mistakenly reject H0

when it is true.

• Type II error : We commit a type II error when we fail to reject H0 even

when it is false.
Conclusions in Test of Hypothesis

  Do Not Reject H0 Reject H0

H0 is True Correct Decision Type I Error
H0 is False Type II Error Correct Decision