Lesson 1: The Probability Value (p-value)

The Probability value or the p-value is the probability of committing Type I error. It is the
area of the tails of the normal table beyond z-statistic. If the p-value is large, then the probability
of committing Type I error is also large. And if the p-value is small, then the probability of
committing Type I error is also small. In testing the hypothesis, the researcher avoids committing
Type I error, thus a smaller p-value is preferable.
To determine the p-value, the Standard Normal Distribution Table is used. Note that
the first column entries are the z-statistic values. The entries in the body of the table give the
area of the tails beyond z-statistic. (See next page)

Steps in Finding the P-Value.

1. Compute the test statistic. Express the z-statistic into three-digit form, round off the z-
statistic to the hundredths place.
2. Using the Standard Normal Table, find the first two digits (the ones and tenths place)
on the first column with entries 0.0, 0.1, 0.2, 0.3, and so on to 3.5.
3. Match the third digit (hundredth place) with the appropriate column heading (first row)
with entries .00, .01, .02, .03, and so on to .09.
4. Read the area (or probability) at the intersection of the first two digits (row) and the
third digit (column).
5. Subtract this value from 0.5000. The result is the p-value.
6. For z-statistic more than 3.59, use 0.0001 as the p-value.
7. For negative z-statistic, use the value of the positive z-statistic as the normal curve is
symmetric. Thus, the left half have the same values as the right.

Examples: Find the corresponding p-value.

1. z = 2.56
 The first two digits 2.5, and the last digit .06.
 The intersection is .4948
 0.5000 – 0.4948 = 0.0052
 The p-value is 0.0052

2. z = -3.28
 Use 3.28 since the normal curve is symmetric.
 The first two digits 3.2, and the last digit .08.
 The intersection is .4995
 0.5000 – 0.4995 = 0.0005
 The p-value is 0.0005

1
 STANDARD NORMAL DISTRIBUTION:
Table Values Represent AREA between the mean (0) and the Z score.

Z .00 .01 .02 .03 .04 .05 .06 .07 .08 .09
.000 .016 .031
0 0 9
.039 .055 .071
0.0 8 .0040 .0080 .0120 7 .0199 .0239 .0279 4 .0359
0.1 .079 .0438 .0478 .0517 .094 .0596 .0636 .0675 .110 .0754
0.2 3 .0832 .0871 .0910 8 .0987 .1026 .1064 3 .1141
0.3 .117 .1217 .1255 .1293 .133 .1368 .1406 .1443 .148 .1517
0.4 9 .1591 .1628 .1664 1 .1736 .1772 .1808 0 .1879
0.5 .155 .1950 .1985 .2019 .170 .2088 .2123 .2157 .184 .2224
4 0 4
.191 .205 .219
5 4 0
.225 .238 .251
8 9 8
.258 .270 .282
0.6 .2291 .2324 .2357 .2422 .2454 .2486 .2549
0 4 3
0.7 .2612 .2642 .2673 .2734 .2764 .2794 .2852
.288 .299 .310
0.8 .2910 .2939 .2967 .3023 .3051 .3079 .3133
1 6 6
0.9 .3186 .3212 .3238 .3289 .3315 .3340 .3389
.315 .326 .336
1.0 .3438 .3461 .3485 .3531 .3554 .3577 .3621
9 4 5
.341 .350 .359
3 8 9
.364 .372 .381
3 9 0
.384 .392 .399
1.1 .3665 .3686 .3708 .3749 .3770 .3790 .3830
9 5 7
1.2 .3869 .3888 .3907 .3944 .3962 .3980 .4015
.403 .409 .416
1.3 .4049 .4066 .4082 .4115 .4131 .4147 .4177
2 9 2
1.4 .4207 .4222 .4236 .4265 .4279 .4292 .4319
.419 .425 .430
1.5 .4345 .4357 .4370 .4394 .4406 .4418 .4441
2 1 6
.433 .438 .443
2 2 0
.445 .449 .453
2 5 5
.455 .459 .462
1.6 .4463 .4474 .4485 .4505 .4515 .4525 .4545
4 1 5
1.7 .4564 .4573 .4582 .4599 .4608 .4616 .4633
.464 .467 .470
1.8 4649 .4656 .4664 .4678 .4686 .4693 .4706
1 1 0
1.9 .4719 .4726 .4732 .4744 .4750 .4756 .4767
.471 .473 .476
2.0 .4778 .4783 .4788 .4798 .4803 .4808 .4817

What’s

Direction: Find the corresponding p-value.

1. z = 1.76
2. z = 3.62
3. z = -2.50
4. z = 0.00

2
 Lesson 2: The Probability Value Approach

Another way of conducting hypothesis testing is by probability value approach. The risk of
making Type I error is the level of significance (α). That is, the p-value or the probability of
committing Type I error.
In the process, we evaluate the null hypothesis by assuming it to be true and test the
reasonableness of the assumption by calculating the probability of getting the results if chance
alone is operating. The obtained probability is the p-value.
After solving for the z-statistic, we compute the area of the tails beyond z-score based on
the normal table. Multiply the area by two for two-tailed test. (See Z-table) The corresponding
area of the tails beyond z-score is the p-value.

The corresponding decision rule is:

1. If the obtained probability is less than or equal to the level of significance, then we
reject the null hypothesis.
2. If the obtained probability is greater than the level of significance α, then we do not
reject the null hypothesis. In other words, we accept the null hypothesis.

While the traditional approach results in a reject-or-do-not-reject conclusion, the p-value

measures how confident we are in rejecting a null hypothesis. The traditional method compares
the test statistic to the critical values, whereas the p-value method compares the p-value to the
significance level. However, the two methods are equivalent in the sense that they yield the
same results.

p-value Interpretation

Highly statistically significant.

Less than 0.01 There is a very strong evidence against
the null hypothesis.
Statistically significant.
0.01 to 0.05 There is an adequate evidence against
the null hypothesis.
Statistically insignificant.
Greater than 0.05 There is an insufficient evidence against
the null hypothesis.

µ = 81
Example 1:
x=80 σ =4 n=100
Left-tailed significance level α = 0.05

A. Computing the test statistic

x−µ 80−81
−5
z= σ = 4 = 2 = -2.50
√n √100
B. Find the p-value corresponding to -2.50
Using the value corresponding to 2.50 which is 0.4938.
Subtract 0.4938 from 0.5000, the result is 0.0062
ρ = 0.0062

C. Decision Rule: 0.0062 < 0.05; reject Ho

D. Conclusion: Highly statistically significant.
There is a very strong evidence to reject the null hypothesis.

µ = 33
Example 2:
x=35 σ =6 .5 n=36
two-tailed significance level α = 0.05

A. Computing the test statistic

3
 x−µ 35−33
24
z= σ = 6.5 = 13 = 1.85
√n √36
B. Find the p-value corresponding to 1.85
Using the value corresponding to 1.85 which is 0.4678.
Subtract 0.4678from 0.5000, the result is 0.0322
Multiply the area by two because it is a two-tailed test, 2 (0.0322)
ρ = 0.0644

C. Decision Rule: 0.0644 > 0.05; do not reject Ho

D. Conclusion: Statistically insignificant.

There is no adequate evidence to reject the null hypothesis.

What’s
Directions: Test the null hypothesis. Show your solution.
1. x=22 µ = 20 σ =5.5 n=36
H1: µ ≠ 20 two-tailed α = 0.05

2. x=98 µ = 95 σ =17 n=100

H1: µ > 95 right-tailed α = 0.01

Lesson 3: Hypothesis Testing Using P-Value

Approach

What is it…
In testing the hypothesis sing the p-value approach, always remember that for one-tailed
tests, the p-value is equal to the area beyond z in the same direction as the alternative
hypothesis. If the alternative hypothesis uses the greater than (>) symbol, the p-value is to the
right the observed value of z. And if the alternative hypothesis uses the less than (<) symbol, the
p-value is to the left the observed value of z.

After solving for the z-statistic, we compute the p-value.

The corresponding decision rule is:
1. If the obtained probability is less than or equal to the level of significance, then we
reject the null hypothesis.
2. If the obtained probability is greater than the level of significance α, then we do not
reject the null hypothesis. In other words, we accept the null hypothesis.

Example 1:
The barangay captain claimed that his constituents have average monthly income of
P5,000 with a standard deviation of P100. To test the claim, the DSWD conducted a census and
found that the average monthly income of 120 residents was P4,950. At 95% confidence, is the
barangay captain right?
A. The significance level α = 0.05 because confidence level is 95%
B. It is a two-tailed test because Ha: µ ≠ 5000
C. Computing the test statistic
x=4,950 µ = 5,000 σ =100n=120

x−µ 4950−5000
z= σ = 100 = -√ 30 or -5.48
√n √120
4
 D. Find the p-value corresponding to -5.48
For values of z above 3.59, use 0.4999 for the area
Subtract 0.4999 from 0.5000, the result is 0.0001
Multiply the area by two because it is a two-tailed test, 0.0002
ρ = 0.0002

E. Decision Rule: 0.0002 < 0.05; reject Ho

F. Conclusion: Highly statistically significant.

There is a very strong evidence to reject the null hypothesis.
The constituents do not have an average monthly income of P5,000.
The barangay captain is not right about his claim.

Notice that the same result can be arrived using the traditional method with the
Decision Rule: -5.48 < -1.96; reject H o. The two methods are equivalent in the sense
that they yield the same results.

Example 2:
The coordinator claimed that the mean age of incoming SHS students is less than 18
years. Mark interviewed 50 students and found the mean age of 17.8 years with a standard
deviation of 0.2 years. Is the claim true at 95% confidence?

A. The significance level α = 0.05 because confidence level is 95%.

B. It is a one-tailed, left-tailed test because H a: µ < 18
C. Computing the test statistic
x=17.8 µ = 18 s=0.2 n=50

x−µ 17.8−18
z= s = 0.2 = -5 √ 2 or -7.07
√n √50
D. Find the p-value corresponding to -5.48
For values of z above 3.59, use 0.4999 for the area
Subtract 0.4999 from 0.5000, the result is ρ = 0.0001

E. Decision Rule: 0.0001 < 0.05; reject Ho

F. Conclusion: Highly statistically significant.

There is a very strong evidence to reject the null hypothesis.
The mean age of incoming SHS students is less than 18 years. The
coordinator is right about his claim.

Notice that the same result can be arrived using the traditional method with the
Decision Rule: -7.07 < -1.96; reject H o. The two methods are equivalent in the sense
that they yield the same results.

What’s
Directions: Solve the following real-life problems involving test of hypothesis on the population
mean using the p-value approach. State your conclusions.
1. Counter.com claims that teenagers spend an average of 5 hours daily on Facebook. A survey
to 40 teenagers resulted with a mean time of 4.8 hours daily with a standard deviation of 1
hour. Is the claim true at 99% confidence?

2. A parent lists down the expenses he will incur if he sends his son to the university. He heard
that the average tuition fee is more than P20,000 per semester with a standard deviation of
P250. He asked 36 friends and got an average cost on university tuition fee of P20,050. Test
the hypothesis at 0.05 level of significance.

5
6