Tests of Hypotheses

Lecture 10
 Statistical Hypotheses

• A statistical hypothesis is an assumption or

statement which may or may not be true
concerning one or more populations.
• We should make it clear at this point that the
acceptance of a statistical hypothesis is a result
of insufficient evidence to reject it and does not
necessarily imply that it is true.
• Hypotheses that we formulate with hope of
rejecting are called null hypotheses denoted by
H0. The rejection of H0 leads to the acceptance of
an alternative hypothesis denoted by H1.
 Type I And Type II Errors
• A type I error has been committed if we reject
the null hypothesis when it is true.
• A type II error has been committed if we accept
the null hypothesis when it is false.
• The probability of committing a type I error is
called the level of significance of the test and is
denoted by α.
• The probability of committing a type II error, is
denoted by β, is impossible to compute unless we
have a specific alternative hypothesis.
• The following are some of important properties
regarding type I and type II errors:
• 1) The type I error and Type II error are related. A
decrease in probability of one generally results in
an increase in the probability of other.
• 2) The size of the critical region, and therefore
the probability of committing type I error, can
always be reduced by adjusting the critical
value(s).
• 3) An increase in the sample size n will reduce
α and β simultaneously.
• 4) If the null hypothesis is false, β is maximum
when the true value of the parameter is close
to the hypothesized value. The greater the
distance between the true value and the
hypothesized value, the smaller β will be.
 One-Tailed And Two Tailed Tests

• A test of any statistical hypothesis where the

alternative is one-sided such as
• H0 : θ = θ0
• H1 : θ ˂ θ0
• or perhaps
• H0 : θ = θ0
• H1 : θ ˃ θ0
• is called one-tailed test.
• A test of any statistical hypothesis where the
alternative is two-sided such as
• H0 : θ = θ 0
• H1 : θ ≠ θ 0
• is called a two-tailed test.
• The alternative hypothesis states that either θ ˂ θ0 or
θ ˃ θ0.
• A test is said to significant if the null hypothesis is
rejected at 0.05 level of significance, and considered
highly significant if the null hypothesis is rejected at
the .01 level of significance.
Tests Concerning Means And Variances

• The steps for testing a hypothesis concerning a population

parameter θ against some alternative hypothesis may be
summarized as follows:
• 1. H0 : θ = θ0.
• 2. Alternatives are θ ˂ θ0, θ ˃ θ0, or θ ≠ θ0.
• 3. Choose a level of significance equal α.
• 4. Select the appropriate test statistic and establish the
critical region.
• 5. Compute the value of the statistic from a random sample
of size n.
• 6. Conclusion: Reject H0 if the statistic has a value in the
critical region; otherwise accept H0.
 Example

• A manufacture of sports equipment has

developed a new synthetic fishing line that he
claims has a mean breaking strength of 8
kilograms with standard deviation of 0.5
kilogram. Test the hypothesis that μ = 8
kilograms against the alternative μ ≠ 8
kilograms if a random sample of 50 lines is
tested and found to have a mean breaking
strength of 7.8 kilograms. Use 0.01 level of
significance.
 Solution

• 1. H0 : μ = 8 kilograms.
• 2. H1 : μ ≠ 8 kilograms.
• 3. α = 0.01
• 4. Critical region: Z < -2.58 and Z ˃ 2.58, where
• Z = (X͞ -μ0 )/(σ/√n)
• 5. Computation: ͞x = 7.8 kilograms n = 50, z =(7.8-
8)/(0.5/√5)=-2.828
• 6. Conclusion: Reject the H0 and conclude that
the average breaking strength is not equal to 8
but is in fact, less than 8 kilograms.
 Example

• The average length of time for students to

register at a certain college has been 50 minutes
with standard deviation of 10 minutes. A new
registration procedure using modern computing
machines is being tried. If a random sample of 12
student had an average registration time of 42
minutes with standard deviation of 11.9 minutes
under the new system, test the hypothesis that
the population mean is now less than 50, using a
level of significance of (1) 0.05, and (2) 0.01.
Assume the population of times to be normal.
 Solution

• 1. H0 : μ = 50 minutes.
• 2. H1 : μ < 50 minutes.
• 3. (a) α = 0.05 (b) α =0.01
• 4. Critical region: (1)T < -1.796 (2) T < -2.718, where
• T = (X͞ -μ0 )/(s/√n) with ν =11 degrees of freedom.
• 5. Computation: ͞x = 42 minutes, s = 11.9 minutes n = 12, t
=(42-50)/(11.9/√12)=-2.33.
• 6. Conclusion: Reject the H0 at .05 level of significance but
not at .01 level. This essentially means that the true mean
is likely to be less than 50 minutes but does not differ
sufficiently to warrant the high cost that would be required
to operate a computer.
 Example

• An experiment was performed to compare the abrasive

wear of two different laminated materials. Twelve
pieces of material I is tested by exposing each piece to
a machine measuring wear. Ten pieces of material II
were similarly tested. In each case, the depth of wear
was observed. The samples of material I gave an
average (coded) wear of 85 units with standard
deviation of 4, while the samples of material II gave an
average of 81 and standard deviation of 5. Test the
hypothesis that the two types of material exhibit the
same mean abrasive wear at the 0.1 level of
significance. Assuming the populations to be
approximately normal with equal variances.
 Solution

• Let μ1 and μ2, represent the population means of material I and

material II, respectively. Using the six steps procedure, we have
• 1. H0 : μ1 = μ2 or μ1 – μ2 = 0.
• 2. H1 : μ1 ≠ μ2 or μ1 - μ2 ≠ 0.
• 3. α =0.1.
• 4. Critical region: (1)T < -1.725 (2) T ˃ 1.725, where
• T = ((X1͞ –X2͞ ) -d0 )/(sp(s/√(1/n1 + 1/n2)) with ν = 20 degrees of
freedom.
• 5. Computation: ͞x1 = 85, s1 = 4, n1 = 12, : ͞x2 = 861, s2 = 5, n2 = 10
• Hence sp =√((11x16 + 9x25)/(12 + 10 -2))= 4.478
• t = ((85-81) – 0)/4.478√‾((1/12) + (1/10)) = 2.07
• 6. Conclusion: Reject H0 and conclude that the two materials do not
exhibit the same abrasive wear.
 Example

• Five samples of a ferrous-type substance are

to be used to determine if there is a difference
between a laboratory chemical analysis and
an X-ray fluorescence analysis of the iron
content. Each sample was split into two
subsamples and the two types of analysis was
applied. The following are the coded data
showing the iron
 Sample
Analysis 1 2 3 4 5
X-ray 2.0 2.0 2.3 2.1 2.4
Chemical 2.2 1.9 2.5 2.3 2.4
• Assuming the populations normal test at the
.05 level of significance whether the two
methods of analysis give, on the average, the
same result.
 Solution

• 1. H0 : μ1 = μ2 or μD = 0.
• 2. H1 : μ1 ≠ μ2 or μD ≠ 0.
• 3. α =0.05.
• 4. Critical region: (1)T < -2.776 (2) T ˃ 2.776,
• T = ((D͞ -d0 )/(sd/√n) with ν = 4 degrees of
freedom.
• 5. Computation:
X-ray Chemical di di2
2.0 2.2 -0.2 0.04
2.0 1.9 0.1 0.01
2.3 2.5 -0.2 0.04
2.1 2.3 -0.2 0.04
2.4 2.4 0.0 0.00
-0.5 0.13
• We find d͞ = -0.5/5 = -0.1, sd2 = (5x0.13 – (-
0.5)2)/(5)(4) = 0.02
• t = (-0.1 – 0)/(0.1412/√5) = - 1.6
• 6. Conclusion: Accept H0 and conclude that
the two methods of analysis are not
significantly different.
 Example

• A manufacture of car batteries claims that the

life of his batteries is approximately normally
distributed with a standard deviation equal to
0.9 year. If a random sample of 10 of these
batteries has standard deviation of 1.2 years,
do you think that σ ˃ 0.9 year? Use a 0.05
level of significance.
 Solution

• 1. H0 : σ2 = 0.81.
• 2. H1 : σ2 ˃ 0.81.
• 3. α = 0.05
• 4.Critical region: Χ2 ˃ 16.919 where Χ2 = (n-1)S2
/σ02 with ν = 9 degrees of freedom.
• 5. Computation: s2 = 1.44, n = 10, and
• Χ2 = 9x1.44/0.81 = 16.0
• 6. Conclusion: Accept H0 and conclude that there
no reason to doubt that the standard deviation is
0.9 year.
Goodness-OF-Fit Test

A goodness-of-fit test between observed

and expected frequencies is based on the quantity

Where χ2 is a value of the random variable Χ2

whose sampling distribution is approximated
very closely by the chi-square distribution.
The symbols oi and ei represent the observed
and expected frequencies, respectively, for the ith cell.
• The number of degrees of freedom in a chi-
square goodness-of-fit test is equal to the
number of cells minus the number of
quantities obtained from the observed data
that are used in the calculations of the
expected frequencies.