Engineering Data
Analysis: Part 2
May 2022 Civil Engineering Board Exam Refresher
Nov
Engr. Jaydee
Hypothesis testing
Hypothesis testing is an act in statistics whereby an analyst tests an assumption
regarding a population parameter. Hypothesis testing is used to assess the
plausibility of a hypothesis by using sample data.
● Null hypothesis (H0) is the statement that proposes that no statistical
significance exists in a set of given observations; the statement being tested
for.
● Alternative hypothesis (Ha) is the statement that proposes that there is
statistical significance in a set of given observations; the statement being
tested against.
Steps in hypothesis testing
Types of errors in hypothesis testing
● Level of significance (𝛼 ). The probability
of a Type I error.
● (𝛽 ). The probability of a Type II error.
● Confidence level ( 1 − 𝛼 ). The probability
that a true null hypothesis is not rejected.
One-tail and two-tail testing
Confidence Intervals
Confidence interval shows the most
probable range the value of the
parameter lies based on the statistic,
assuming the sample is representative
z- and t-distributions 𝜒 -distribution
of the population.
Statistic Constraint Confidence Interval
Population standard deviation known. 𝜎
mean 𝜇 = 𝑥±𝑧
Sample size at least 30. 𝑛
Population standard deviation known. 𝜎
mean 𝜇 =𝑥±𝑡
Sample size less than 30. 𝑛
Population standard deviation not 𝑠
mean 𝜇 =𝑥±𝑡
known. 𝑛
proportion
variance One variance.
Example 8
Find the 95% confidence interval of the true mean yield strength of rebars for the
following scenarios.
1. Sample mean = 37.5 psi, population standard deviation = 3.0 psi, 50 bars were
tested
2. Sample mean = 37.5 psi, population standard deviation = 3.0 psi, 25 bars were
tested
3. Sample mean = 37.5 psi, sample standard deviation = 3.0 psi, 50 bars were tested
Solution
Confidence One-tail Two-tail
Interval z-value z-value
90% 1.28 1.645
95% 1.645 1.96
97.5% 1.96 2.24
99% 2.326 2.576
Solution
Example 9
In a traffic survey, the speeds of vehicles are measured by laser guns for the purpose
of determining the mean vehicle speed on a particular city street. It is known that the
standard deviation of city streets with the same posted speed limit is 3.58 kph. If we
wish to determine the mean vehicle speed to within ±1 kph with a 99% confidence,
what should be the sample size of our observations?
Confidence One-tail Two-tail
Interval z-value z-value
90% 1.28 1.645
95% 1.645 1.96
97.5% 1.96 2.24
99% 2.326 2.576
Example 10
In order to control the quality of the compaction of the subgrade in a highway
pavement project, 50 soil specimens were prepared and tested. Three of the samples
were found to have compaction below the CBR requirement. What is the 95%
confidence interval of the proportion of well-compacted pavement?
Confidence One-tail Two-tail
Interval z-value z-value
90% 1.28 1.645
95% 1.645 1.96
97.5% 1.96 2.24
99% 2.326 2.576
Example 11
From a data record of 25 storms, it is estimated that the sample variance of the runoff
on the Monocracy River is 0.36 in2. What is the upper 95% confidence limit of the
population variance for the runoff of the river?
Example 12
Suppose the specification for the yield strength of rebars required a mean value of 38
psi. It is, therefore, essential that the population of rebars to be used in the
construction of a reinforced concrete structure has the required mean strength. From
the rebars delivered to the construction site by the supplier, the engineer ordered that
a sample of 50 rebars be randomly selected and tested for yield strengths. The sample
mean from the 50 tests yielded a value of 37.5 psi. It is known that the standard
deviation of rebar strength from the supplier is 3.0 psi.
Check if the rebars sampled from the supplier satisfy the required yield strength of the
specification. Use 5% level of significance.
Confidence One-tail Two-tail
Interval z-value z-value
90% 1.28 1.645
95% 1.645 1.96
97.5% 1.96 2.24
99% 2.326 2.576
Example 13
Chi-square goodness of fit test: one-way table
Jerome claims that 60% of the engineering students in a university with a population
of 1000 love math while 30% love physics and 10% love PE. To prove his claim, he asks
100 students and found out that 57 love math, 40 love physics, and 3 love PE.
1. Calculate the 𝜒 -statistic.
2. At 5% level of significance, is Jerome’s claim true?