DIFFERENCE BETWEEN MEANS

t TEST and Z TEST

 DIFFERENCE BETWEEN MEANS
Jegie 51

DIFFERENCE BETWEEN MEANS

❑ The difference between means denotes whether:

❖ there is a significant difference between

✓ perceptions about a certain situation

✓ characteristics of two samples or group
✓ reactions towards a certain issue

❖ there is significant gain after an experiment or after any activity

has been conducted.

❑ This statistical tool is used to test the null hypothesis that the
samples are drawn from:

✓ the same population

✓ from two different populations having the same means

DIFFERENCE BETWEEN MEANS

t TEST Z TEST

1. t test 1. Z test Sample Mean

2. t test The One-Tailed Test Compared with
Using a Single Group Population Mean
3. t test Compare Two 2. Z Test Testing
Correlated Groups Proportions
4. t test Compare Two
Independent Groups
 DIFFERENCE BETWEEN MEANS
t TEST
Jegie 52

t TEST

Definition

❖ t Test - is a statistical tool used to determine if there is a significant

difference between means when n  30.

DISADVANTAGE

▪ It does not allow us to determine the cause of the difference.

ADVANTAGES

▪ t allows us to accurately describe the shape of the sampling distribution for

means when the sample size is small ( N 30) and when we have no way of
directly knowing the population parameters.

▪ Determine if the sample mean was significantly different from a hypothesized

mean.

▪ Conduct research in which the behavior of one group of subjects is compared to

the behavior of another group of subjects.

▪ To determine if there is a systematic and reliable difference between members of

two populations.

✓ For instance, when we compare men’s and women’s abilities in English or

math, we are comparing two populations. From each population we take
a representative sample and then measure the appropriate dependent
variable.

▪ When the experimenter randomly assigns subjects to the groups, he or she is

creating two hypothetical populations.

✓ The first population may represent the subjects in the control condition
and the second population represents subjects exposed to the treatment
of interest. If the difference between the means is sufficiently large, we
assume that the population parameters from which the samples were
drawn are also different. We may also infer that the difference is the
result of the manipulated variables.
 t test

THE ONE-TAILED TEST USING A SINGLE GROUP

 PRE-TEST - Problem
t test The One-tailed Test Using a Single Group Jegie 53

Problem

An algebra teacher conducted a remedial teaching in algebra to 40 students. Before the

start of the remedial teaching, the teacher administered a pre-test. After one month of teaching,
the pre-test in another form was again given to the students. The results follow:

0.55 = the r between the pre-test scores and post-test scores.

83.8 = the mean of the students’ scores in the post-test
78.5 = the mean of the students’ scores in the pre-test
5.70 = the SD of the students’ scores in the post test
6.10 = the SD pf the students’ scores in the pre-test

Was the remedial teaching effective? How effective was it? Solve the problem by finding
the following:

1). Specific Questions

2). Null Hypothesis
3). Alternative Hypothesis
4). Level of Significance
5). Statistical Method Used
6). Rejection Region
7). Computation for the value of t
8). Decision
9). Conclusion
 PRE-TEST - Answers
t test The One-tailed Test Using a Single Group Jegie 54

I Specific Problem

Was the remedial teaching effective?

II Null Hypothesis

The remedial teaching was not effective.

III Alternative Hypothesis

The remedial teaching was effective.

IV Level of Significance

 = 0.01

V Statistical Method Used

Use t test One-Tailed Test Using a Single Group

VI Rejection Region

Since the hypothesis is directional, therefore the degrees of freedom is df = N-1 = 40-1
= 39. From the t test table, the corresponding value of t 0.01 = is equal to 2.71.

VII Computation

SEM1 = 0.964
SEM2 = 0.901
SEd = 0.886
t = - 5.98 Note: Pls. Neglect the - sign.

VIII Decision

Since the computed value t = 5.98 is greater than the critical value at 0.01 level of
significance t 0.01 = 2.71, therefore reject the null hypothesis and accept the alternative
hypothesis.

IX Conclusion

The remedial teaching was effective.

 t TEST
THE ONE-TAILED TEST USING A
SINGLE GROUP
Jegie 55

t test

THE ONE-TAILED TEST USING A SINGLE GROUP

Definition

It is used to determine if there is a gain after the group has been

exposed to an experimental variable.

How it should be done?

❑ First, the group is given a pre-test, a test given before the

group is exposed to the experimental variable.

❑ Then after the group is exposed to the experimental variable

for a sufficient period of time, it is given a post-test, the same
test given before the experiment but in different form to avoid
the effect of practice.

❑ The scores in the pre-test and in the post-test are then

correlated.
 t TEST
THE ONE-TAILED TEST USING A
SINGLE GROUP
Jegie 56

Formulas

1. Standard error of the pre-test and post test means

Pre-test Post Test

S1 S2
SEM1 = SEM2 =
N1 N2

 (X – X) 2

Where: S =
N

Where: S standard deviation

N number of samples
SEM standard error of the mean

2. Standard error of the difference between the two means (SEd)

SEd = (SEM1)2 + (SEM2)2 – (2) ( r ) (SEM1) (SEM2)

Where: SEM1 Standard error of the mean of the pre-test

SEM2 Standard error of the mean of the post test
r Coefficient of Correlation

3. t test

t = X1 - X2
SEd

Where: X1 pre-test mean

X2 post-test mean
SEd standard error of the difference between two
means
 t TEST
THE ONE-TAILED TEST USING A
SINGLE GROUP
Jegie 57

Problem

Forty weak students were given a remedial teaching in Mathematics. Before the
remedial teaching was conducted, the students were given a pre-test. After six months
of remedial teaching the students were given a post-test, the same test as the pre-test
but only in different form to avoid the effect of practice. The results follow:

0.70 the coefficient of correlation, r, between the pre-test and post-

test scores
55.6 the pre-test mean
69.83 the post-test mean
11.3 the standard deviation of the pre-test scores
10.5 the standard deviation of the post-test scores.
40 number of samples

Solution

I Specific Problem

Is the remedial teaching very effective?

II Null Hypothesis

The remedial teaching is not very effective.

III Alternative Hypothesis

The remedial teaching is very effective.

IV Level of Significance

 = 0.01

V Statistical Method Used

t test The One-tailed test using a single group

VI Rejection Region

When df = n – 1 = 40 – 1 = 39. From the table the corresponding

critical value of t at 0.01 level of significance is equal to 2.71.
 t TEST
THE ONE-TAILED TEST USING A
SINGLE GROUP
Jegie 58

VII Computation

1. Compute the standard errors of the pre-test and post-test means

S
SEM =

Where: S standard deviation

N number of samples
SEM standard error of the mean

a. For pre-test

SEM1 = 11.3 = 11.3 = 1.79

6.32
40

b. For post-test

SEM2 = 10.5 = 10.5 = 1.66

6.32
40

2. Compute the standard error of the difference between the two means (SEd)

SEd = (SEM1)2 + (SEM2)2 – (2) ( r ) (SEM1) (SEM2)

SEd = (1.79)2 + (1.66)2 – (2) ( 0.70 ) (1.79) (1.66)

SEd = 3.2041 + 2.7556 – 4.15996

SEd = 1.79974

SEd = 1.34
 t TEST
THE ONE-TAILED TEST USING A
SINGLE GROUP
Jegie 59

3. Compute t using the formula

t = X1 - X2
SEd

Where: X1 pre-test mean

X2 post-test mean
SEd standard error of the difference between the twO
means

t = 55.6 - 69.83
1.34

t = -14.23
1.34

t = -10.62 (disregard the sign)

VIII Decision

Since the computed value of t = 10.62 is greater than the critical value which is equal to
2.71, therefore accept the alternative hypothesis.

IX Conclusion

The remedial teaching is very effective.

 POST TEST-Problem
t test The One-tailed Test Using a Single Group Jegie 60

Problem

A training director of a large chain of department stores in Makati wants to know of a

week’s training course will help 12 sales-ladies rapport with the customers. One month prior to
the training they were rated, and one month after the training they were also rated with the
following results:

0.75 the coefficient of correlation, r,

60.10 the mean pre-evaluation rating
85.75 the mean post evaluation rating
12.10 the standard deviation of the pre-evaluation rating
11.68 the standard deviation of the post-evaluation rating
12 the number of samples

Does training improve the rapport between the sales-ladies and the customers?
Solve the problem by finding the following:

1). Specific Questions

2). Null Hypothesis
3). Alternative Hypothesis
4). Level of Significance
5). Statistical Method Used
6). Rejection Region
7). Computation for the value of t
8). Decision
9). Conclusion
 POST TEST - Answers
t test The One-tailed Test Using a Single Group Jegie 61

I Specific Problem

Does training improve the rapport between the sales-ladies and the customers?

II Null Hypothesis

There is no improvement in the rapport between the sales-ladies and the customers after
the training.

III Alternative Hypothesis

There is an improvement in the rapport between the sales-ladies and the customers
after the training.

IV Level of Significance

 = 0.05

V Statistical Method Used

Use t test One-Tailed Test Using a Single Group

VI Rejection Region

Since the hypothesis is directional, therefore the degrees of freedom is df = N-1 = 12-1
= 11. From the t test table, the corresponding value of t 0.05 = is equal to 1.796.

VII Computation

SEM1 = 3.49
SEM2 = 3.37
SEd = 2.43
t = - 10.56 Note: Pls. Neglect the - sign.

VIII Decision

Since the computed value t = 10.56 is greater than the critical value at 0.05 level of
significance t 0.05 = 1.796, therefore reject the null hypothesis and accept the alternative
hypothesis.

IX Conclusion

There is an improvement in the rapport between the sales-ladies and the customers
after the training.
 t test

COMPARE TWO CORRELATED GROUPS

 PRE-TEST- Problem
t test Compare Two Correlated Groups Jegie 62

Problem

An experimental study was conducted on the effect of programmed materials in English

on the performance of 20 selected college students. Before the program was implemented the
pretest was administered and after 5 months the same instrument was used to get the posttest
result. The following is the result of the experiment:

Pretest (X1) Posttest (X2)

20 25
30 35
10 25
15 25
20 20
10 20
18 22
14 20
15 20
20 15
18 30
15 10
15 16
20 25
18 10
40 45
10 15
10 10
12 18
19 25

Use 0.05 level of significance. Solve the problem by getting the following requirements:
1). Specific Problem, 2). Null Hypothesis, 3). Alternative Hypothesis, 4). Statistical Method
Used, 5). Level of Significance, 6). Region of Rejection, 7). Computation of the value of t,
8). Decision, 9). Conclusion.
 PRE-TEST - Answers
t test Compare Two Correlated Groups Jegie 63

I - Specific Problem

Is there a significant difference between the pretest and the post test on the used of
programmed materials in English?

II - Null Hypothesis (Ho)

There is no significant difference between the pretest and the post test on the used of
programmed materials and did not affect the student’s performance in English.

III - Alternative Hypothesis (H1 or HA)

That post- test result is higher than the pretest result.

IV - Level of Significance:  = 0.05

V - Statistical Method Used

t Test for Correlated Samples

VI - Rejection Region

df = n - 1 at 0.05 level of significance

= 20 - 1 t = 1.729 for one tailed test
= 19

VII- Computation

t = -3.16 Neglect the - sign

VIII- Decision

Since the computed value is 3.16 and is greater than t = 1.729, therefore accept the
alternative hypothesis.

IX - Conclusion

The posttest result is higher than the pretest result. It implies that the use of the
programmed materials in English is effective.
 t TEST
COMPARE TWO CORRELATED
GROUPS
Jegie 64

t TEST

COMPARE TWO CORRELATED GROUPS

Definition

❖ If the same subjects are tested at two different times and we want
to look at the differences between the two means, we must use a
correlated groups t-test.

❖ The data is in an interval or ratio type.

Formula

D
t =

D2 -
(D)2
N

N(N-1)

Where:

D Difference between the value of X1 and X2

N Number of pairs of observations
D Summation of the Difference
D2 Summation of the square of the difference of
X1 and X2
D Summation of D divided by N
 t TEST
COMPARE TWO CORRELATED
GROUPS
Jegie 65

Problem

Students who received coaching to improve their ability to solve analogy problems. They
were tested at time 1, received several coaching sessions, and were tested again at time 2.

Table 3

Data from an Experiment in Which the

Same Subjects Are Tested on 2 Different Occasions

Subject Test 1 Test 2

X1 X2

1 10 7
2 17 18
3 15 20
4 8 15
5 18 19
6 10 13
7 16 17
8 12 16
9 13 18
10 9 16

Solution

I Specific Problem

Do the students who received coaching improve their ability to solve analogy problems?

II Null Hypothesis

Students who received coaching do not improve their ability in solving analogy
problems.

III Alternative Hypothesis

Students who received coaching improve their ability in solving analogy problems.

IV Statistical Method Used

t-test (Compare Two Correlated Groups)

 t TEST
COMPARE TWO CORRELATED
GROUPS
Jegie 66

V Level of Significance

 = 0.05, one tailed test

VI Critical Region for Rejection of H0 .

Degrees of freedom (df) = N – 1 = 10 – 1 = 9, t0.05  1.833 (One tailed).

The region is that portion of the area under the curve that includes the values of a
statistic that lead to rejection of Ho .

VII Computation

Steps:

1. Find the value of D by subtracting X2 from X1

2. Square the value of D

3. Find the sum of each column.

4. Present it in a tabular form.

Subject Test 1 Test 2 D D2

X1 X2 (X1 - X2)

1 10 7 3 9
2 17 18 -1 1
3 15 20 -5 25
4 8 15 -7 49
5 18 19 -1 1
6 10 13 -3 9
7 16 17 -1 1
8 12 16 -4 16
9 13 18 -5 25
10 9 16 -7 49

Total X1 = 128 X2 = 159 D = 31 D2 = 185

5. Find the value of D

D = D / N = 31 / 10 = 3.1
 t TEST
COMPARE TWO CORRELATED
GROUPS
6. Solve for the value of t. Jegie 67

t = D

 D2 - (D)2
N

N(N-1)

t = -3.1

185 - (31)2
10

10 (10-1)

t = -3.1

185 - 961
10

10 (9)

t = -3.1

185 - 96.1

90

t = -3.1

88.90

90
 t TEST
COMPARE TWO CORRELATE
GROUPS
Jegie 68

t = -3.1

0.987

t = -3.1

0.99

t = -3.13 neglect the - sign

VIII Decision

Since the computed value of t = 3.13 is greater than the critical value at 0.05 level of
significance = 1.833, we can reject the null hypothesis (H o) and accept the alternative
hypothesis.

IX Conclusion

Students who received coaching improve their ability in solving analogy Problems.
 POST TEST-Problem
t test Compare Two Correlated Groups Jegie 69

Problem

The table as shown below are scores of ten pairs of matched individuals on a test after
being exposed to two different teaching methods, A and B.

Pair Method A Method B

1 24 26
2 28 30
3 32 37
4 18 22
5 24 29
6 36 40
7 40 38
8 37 41
9 24 29
10 20 28

Solve the problem by getting the following requirements:

1). Specific Questions

2). Null Hypothesis
3). Alternative Hypothesis
4). Level of Significance
5). Statistical Method Used
6). Rejection Region
7). Computation for the value of t
8). Decision
9). Conclusion
 POST TEST-Answers
t test Compare Two Correlated Groups Jegie 70

I Specific Problem

In your own opinion, do you think those taught by method B did significantly better than
those taught by Method A?

II Null Hypothesis

There is no significant difference in the performance of the students being taught

By Method A or Method B.

III Alternative Hypothesis

Those students being taught by Method B do it better than those students being taught
by Method A.

IV Level of Significance

 = 0.01

V Statistical Method Used

t test Compare two Correlated Groups

VI Rejection Region

Since the hypothesis is directional, therefore the degrees of freedom is df = N-1 = 10-1
= 9. From the t test table, the corresponding value of t 0.01 = is equal to 2.821.

VII Computation

t = - 4.46 Pls. Neglect the - sign

VIII Decision

Since the computed t = 4.46 is greater than t 0.01 = 2.821, therefore reject the null
hypothesis and accept the alternative hypothesis.

IX Conclusion

Those students being taught by Method B do it better than those students being taught
by Method A.
.

t test

COMPARE TWO INDEPENDENT GROUPS

 PRE-TEST - Problem
t test Compare Two Independent Groups Jegie 71

Problem

The following are the scores of 10 male and 10 female engineering students in
spelling. Test the null hypothesis that there is no significant difference between the
performance of male and female engineering students in the said test. Use the t test at the
0.05 level of significance.

Male (X1) Female (X2)

14 12
18 9
17 11
16 5
4 10
14 3
12 7
10 2
9 6
17 13

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of t
8). Decision
9). Conclusion.
 PRE-TEST - Answers
t test Compare Two Independent Groups Jegie 72

I - Specific Problem

Is there a significant difference between the performance of male and female

Engineering students in spelling?

II - Null Hypothesis (Ho)

There is no significant difference between the performance of male and female

Engineering students in spelling.

III - Alternative Hypothesis (H1 or HA )

There is a significant difference between the performance of male and female

Engineering students in spelling.

IV - Level of Significance:

 = 0.05

V - Statistical Method Used

t Test for two Independent Samples

VI - Rejection Region

df = n1 + n2 – 2 for two-tailed test

= 10 + 10 – 2 t0.05 = 2.101
= 18

VII- Computation: t obs = 2.88

VIII- Decision

Since the computed value t = 2.88 is greater than t 0.05 = 2.101, therefore accept the
alternative hypothesis.

IX – Conclusion

This means that there is a significant difference between the performance of male and
female Engineering students in spelling. It implies that the male perform better than
the female students considering that the mean/ average score of the male student is
13.1 compared with the average score of female students of only 7.8.
 t TEST
COMPARE TWO INDEPENDENT
GROUPS
Jegie 73

t TEST

COMPARE TWO INDEPENDENT GROUPS

Definition

❖ It is used when samples of the two variables (n1 or n2) is less than 30
and variances are equal.

❖ It is assume that the two samples are independent random samples

from populations having roughly the shape of normal distributions with
the same standard deviation.

Formula

t = X1 - X2

(N1 –1) S12 + (

N2 –1) S22 . [ 1 + 1 ]
N1 + N2 -2 N1 N2

where: X1 mean of sample 1

X2 mean of sample 2
S1 standard deviation of sample 1
S2 standard deviation of sample 2
N1 number of observations of sample 1
N2 number of observations of sample 2

Degrees of freedom (df)

df = N1 + N2 - 2
 t TEST
COMPARE TWO INDEPENDENT
GROUPS
Jegie 74

Problem

The following are the numbers of sales which a random sample of nine salesperson of
industrial chemicals in Pasay City and a random sample of six salesman of industrial chemicals in
Quezon City made over a fixed period of time.

Pasay City: 41 47 62 39 56 64 37 61 52

Quezon City: 34 63 45 55 24 43

Use the 0.01 level of significance to test whether the difference between the means of these two
samples is significant. Use t test.

Solution

I Specific Question

Is there a significant difference between the means of the number of sales of industrial
chemicals in Pasay City and Quezon City?

II Null Hypothesis

There is no significant difference between the means of the number of sales of industrial
chemicals in Pasay City and Quezon City.

III Alternative Hypothesis

There is a significant difference between the means of the number of sales of industrial
chemicals in Pasay City and Quezon City?

IV Level of Significance

 = 0.01

V Statistical Method Used

Independent t Test
 t TEST
COMPARE TWO INDEPENDENT
GROUPS
Jegie 75

VI Rejection Region

Degrees of Freedom (df) = n1 + n2 – 2 = 9 + 6 - 2 = 13

t 0.01/2 = t 0.005 = 3.012

VII Computation

Steps:

1. Find the sum of X1 and X2.

2. Compute for the value of the mean by dividing the summation of X1 and X2 by the
number of observations.

3. Find the difference of X1 and mean of X1. Same with X2.

4. Square the value of X1 – X1 and X2 – X2 .

Step1 Step 3 Step 4 Step 1 Step 3 Step 4

X1 X1 – X1 (X1-X1)2 X2 X2-X2 (X2-X2)2

41 41-51 = -10 100 34 34-44 = -10 100

47 47-51 = -4 16 63 63-44 = 19 361
62 62-51 = 11 121 45 45-44 = 1 1
39 39-51 = -12 144 55 55-44 = 11 121
56 56-51 = 5 25 24 24-44 = -20 400
64 64-51 = 13 169 43 43-44 = -1 1
37 37-51 = -14 196
61 61–51 = 10 100
52 52-51 = 1 1

 X1 = 459  (X-X1)2 =872  X2 = 264  (X-X2)2 = 984

Step 2 - Compute for the value of the Mean

X1 =  X1 X2 =  X2
N1 N2

X1 = 459 X2 = 264
9 6

X1 = 51 X2 = 44
 t TEST
COMPARE TWO INDEPENDENT
GROUPS
Jegie 76

5. Solve for the standard deviation by applying the formula.

S12 =  (X1-X1)2 S22 =  (X2-X2)2

N1 - 1 N2 - 1

S12 = 872 S22 = 984

9-1 6-1

S12 = 872 S22 = 984

8 5

S12 = 109 S22 = 196.80

6. Solve for the value of t by using the given formula.

t = X1 - X2

(N1 –1) S12 + (N2 –1) S22 . [ 1 + 1 ]

N1 + N2 -2 N1 N2

t = 51 - 44

(
9 –1) 109 + (6 –1) 196.80 . [ 1 + 1 ]
9+6-2 9 6

t = 7

( 8 ) 109 +
( 5) 196.80 . [ 1 + 1 ]
13 9 6
 t TEST
COMPARE TWO INDEPENDENT
GROUPS
Jegie 77

t = 7

872 +
984 . [ 1 + 1 ]
13 9 6

t = 7

1856 . [ 0.11 + 0.17 ]

13

t = 7

( 142.77 ) . [ 0.28 ]

t = 7

39.97

t = 7
6.32

t = 1.11

VIII Decision

Since the computed value of t = 1.11 is less than t 0.005 = 3.012, therefore accept the
null hypothesis.

IX Conclusion

There is no significant difference between the means of the number of sales of industrial
chemicals in Pasay City and Quezon City.
 POST TEST- Problem
t Test Compare Two Independent Groups Jegie 78

Problem

Two groups of experimental rate were injected with tranquilizer at a 1.0 mg, and 1.5
mg dose respectively. The time given in seconds that took them to fall asleep is given below.
Use the t test for independent samples at 0.01 to test the null hypothesis that the difference in
dosage has no effect on the length of time it took them to fall asleep.

1.5 mg dose 1.0 mg dose

12.0 0 9.80
7.40 13.20
9.80 11.20
11.50 9.50
13.00 13.00
12.50 12.10
9.80 9.80
10.50 12.30
13.50 7.90
10.20
9.70

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of t
8). Decision
9). Conclusion
 POST TEST - Answer
t Test Compare Two Independent Groups Jegie 79

I - Specific Problem

Is the difference in dosage has an effect on the length of time it took to fall asleep?

II - Null Hypothesis (Ho)

The difference in dosage has no effect on the length of time it took to fall asleep.

III - Alternative Hypothesis (H1 or HA )

The difference in dosage has an effect on the length of time it took to fall asleep.

IV - Level of Significance:

 = 0.01

V - Statistical Method Used

t Test for Independent Samples

VI - Rejection Region

df = n1 + n2 – 2 for two-tailed test

= 11 + 9 – 2 t0.01 = 2.878
= 18

VII- Computation

t = - 0.40 ( Neglect the - sign )

VIII- Decision

Since the computed value of t = 0.40 is less than the t critical value of 2.878 at
0.01 level of significance with 18 degrees of freedom, the null hypothesis is accepted.

IX – Conclusion

The difference in dosage has no effect on the length of time it took to fall asleep.
 Z test

SAMPLE MEAN COMPARED WITH POPULATION MEAN

 PRETEST- Problem
Z- test Sample Mean Compared with Population Mean
(One-Sample Mean Test) Jegie 80

Problem

It is claimed that the average weight of a bag of biscuit is 250 grams with a standard
deviation of 20.5 grams. Would you agree to this claim if a random sample of 50 bags of biscuits
showed an average weight of 240 grams, using a 0.05 level of significance?

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion.
 PRETEST- Answers
Z- test Sample Mean Compared with Population Mean
(One-Sample Mean Test) Jegie 81

I - Specific Problem

Would you agree to the claim that the average weight of a bag of biscuit is 250 grams?

II - Null Hypothesis

The manufacturer claims that the average weight of a bag of biscuit is 250 grams.

III - Alternative Hypothesis

The manufacturer claims that the average weight of a bag of biscuit is greater or less
than 250 grams.

IV - Level of Significance:

 = 0.05

V - Statistical Method Used

One Sample Mean Test ( Z – Test )

VI - Rejection Region

For two failed test at 0.05 level of significance, Z0.05 = 1.96

VII- Computation

Z = - 3.45 ( neglect the - sign )

VIII- Decision

Since the computed value of Z = 3.45 is greater than the critical value of Z 0.05 = 1.96,
therefore, accept H1 or HA .

IX – Conclusion

The claim of the manufacturer that the average weight per bag of biscuit is 250 grams is
not true.
 Z TEST
SAMPLE MEAN COMPARED WITH
POPULATION MEAN
Jegie 82

Z TEST

SAMPLE MEAN COMPARED WITH POPULATION MEAN

Definiton:

This is used when:

✓ The population standard deviation is known.
✓ Only the sample standard deviation is known but the sample size is
large in which case the sample standard deviation may be used as an
estimate of the population standard deviation.

Examples:

❖ To determine whether or not a certain product meets certain

specifications.
❖ To determine whether new techniques resulted in improved
production.

Formula

( X - ) n
Z =


where:  = population mean

X = sample mean
n = samples
 = population standard deviation
Z = z test

Table 4
CRITICAL VALUE OF Z

Test  0.1 0.05 0.025 0.01 0.005

One-Tailed  1.28  1.645  1.96  2.23  2.58

Two-tailed  1.645  1.96  2.24  2.58  2.81

 Z TEST
SAMPLE MEAN COMPARED WITH
POPULATION MEAN
Jegie 83

Problem

The mean and standard deviation in years of the lifetime of cameras produced by a
manufacturer are 4 and 1.6, respectively. The manufacturer claims that the mean lifetime of the
cameras has increased due to a new manufacturing technique. To determine the validity of such
a claim, a sample of 25 cameras was taken, the mean lifetime of which is 5 years. Is the claim
valid at an  = 0.05?

Solution

I Specific Question

Is there an increase in the mean lifetime of the cameras?

II Null Hypothesis

There is no increase in the mean lifetime of the cameras?

III Alternative Hypothesis

There is an increase in the mean lifetime of the cameras?

IV Level of Significance

 = 0.05

V Statistical Method Used

Since the population standard deviation is known, we use the Z test.

VI Rejection Region

Since H1 is directional, we use a one tailed test. at an  = 0.05, the critical Z value is
Z’ = 1.645 . Please refer to table 4 .
 Z TEST
SAMPLE MEAN COMPARED WITH
POPULATION MEAN
Jegie 84

VII Computation

( X - ) n
Z =


( 5 - 4) 25
Z =
1.6

Z = (1) (5)
1.6

Z = 3.125

VIII Decision

Since Z = 3.125 is greater than Z at 0.05 level of significance which is equal to 1.645,
we reject the null hypothesis. Since Ho is rejected it follows that the alternative
hypothesis is accepted.

IX Conclusion

Therefore there is an increase in the mean lifetime of the cameras. Hence, the claim is
valid.
 Z TEST
SAMPLE MEAN COMPARED WITH
POPULATION MEAN
Jegie 85

Problem

A machine which has been regulated dispenses an average of 330 ml fruit concentrate
per bottle. A random sample of 49 bottles filled by the machine has a mean content of 320 ml
and a standard deviation of 50 ml. Test the hypothesis that the machine is not in proper working
order at  = 0.01.

Solution

I Specific Problem

Is the machine in proper working order?

II Null Hypothesis

The machine is in proper working condition?

III Alternative Hypothesis

The machine is not in proper working condition.

IV Level of Significance

Use  = 0.01. Although the population standard deviation is not known, the sample
size is however large ( n = 49  30 ). Hence we use the Z test. We use the sample
standard deviation as an estimate of the population standard deviation.

V Statistical Method Used

Since the alternative hypothesis is non directional. We used a two tailed

test.

VI Rejection Region

At an  = 0.01 Z’ = + - 2.58 which means the acceptance region is -2.58  Z

2.58. If the computed Z is outside this region, we reject Ho and conclude that the
machine is not in proper working order.
 Z TEST
SAMPLE MEAN COMPARED WITH
POPULATION MEAN
Jegie 86

VII Computation

( X - ) n
Z =


( 320 - 330 ) 49
Z =
50

Z = (-10) ( 7)
50

Z = -70
50

Z = - 1.40 (neglect the – sign)

VIII Decision

Since Z falls within the acceptance region. We accept Ho.

IX Conclusion

The machine is in proper working order.

 POST TEST-Problem
Z Test Sample Mean Compared with population Mean
(One-Sample Mean Test) Jegie 87

Problem

In a study made on time and motion, it was found that a certain manual work can be
finished at an average time of 40 minutes with a standard deviation of 8 minutes. A group of 16
workers is given a special training and then found the average only 35 minutes. Can we
conclude that the special training can speed up the work using 0.01 level?

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion
 POST TEST- Answers
Z Test Sample Mean Compared with population Mean
(One-Sample Mean Test) Jegie 88

I - Specific Problem

Do the special training can speed up the work?

II - Null Hypothesis (Ho)

The special training is not a factor in speeding up the work.

III - Alternative Hypothesis (H1 or HA)

The special training is a factor in speeding up the work.

IV - Level of Significance:

 = 0.01

V - Statistical Method Used

Z-test Sample Mean Compared with population Mean or One Sample Mean Test

VI - Region of Rejection

Z 0.01 = 2.33

VII - Computation

Zobs = 2.50

VIII- Decision

Since the computed value Zobs = 2.50 is greater than the critical value at 0.01 level of
significance which is equal to 2.33 therefore accept the alternative hypothesis.

IX - Conclusion

The special training can really speed up the work.

 Z test

TESTING PROPORTIONS
 PRETEST- Problem
Z- test Testing Proportions
Jegie 89

Problem No. 1

A marketing expert for a pasta-making company believes that 40 % of pasta lovers

prefer lasagna. If 9 out of 20 pasta lovers choose lasagna over other pastas, what can be
concluded about the expert’s claim? Use a 0.01 level of significance.

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion
 PRETEST- Answers
Z- test Testing proportions
Jegie 90

I - Specific Problem

Does the marketing expert’s claim is true that 40 % of pasta lovers prefer lasagna?

II - Null Hypothesis (Ho): P = 0.40

The marketing expert’s claim is true that 40 % of pasta lovers prefer lasagna?

III - Alternative Hypothesis (H1 or HA): P < 0.40

The marketing expert’s claim is NOT true that 40 % of pasta lovers prefer lasagna?
(It means it is less than 40 % of pasta lovers prefer lasagna).

IV - Level of Significance:

 = 0.01

V - Statistical Method Used

Z-test Testing Proportions

VI - Region of Rejection

Z 0.01 = 2.33 for one-tailed test, please refer to table 4.

VII - Computation

Po = 0.45 Q = 0.60 Qp = 0.1095 Zobs = 0.457 = 0.676

VIII- Decision

Since the computed value Zobs = 0.676 is less than the critical value at 0.01 level of
significance which is equal to 2.33 therefore accept the null hypothesis.

IX - Conclusion

The marketing expert’s claim is true that 40 % of pasta lovers prefer lasagna?
 PRETEST- Problem
Z- test Testing Proportions
Jegie 91

Problem No. 2

Dr. Lopez claims that of 1000 civil engineering graduates of a college of which he is the
dean, at least 80 % passed the board examinations. If out of the 150 examinees, only 100
passed the board, determine the validity of the claim at an  = 0.05 level of significance.

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion
 PRETEST- Answers
Z- test Testing proportions
Jegie 92

I - Specific Problem

Is the claim of Dr. Lopez valid that 80 % of the civil engineering graduates passed the
board examinations.

II - Null Hypothesis (Ho): P = 0.80

Dr. Lopez claim that 80 % of the civil engineering graduates passed the board
examinations is valid.

III - Alternative Hypothesis (H1 or HA): P < 0.80

Dr. Lopez claim that 80 % of the civil engineering graduates passed the board
examinations is NOT valid.

IV - Level of Significance:

 = 0.05

V - Statistical Method Used

Z-test Testing Proportions

VI - Region of Rejection

Z 0.05 = 1.645 for one-tailed test, please refer to table 4.

VII - Computation

Qp = 0.03 Po = 0.6667 or 66.67 % P = 0.80 or 80 %

Z = - 4.44 neglect the - sign = 2.11

VIII- Decision

Since the computed value Zobs = 2.11 is greater than the critical value at 0.05 level
of significance which is equal to 1.645 therefore accept the alternative hypothesis.

IX - Conclusion

Dr. Lopez claim that 80 % of the civil engineering graduates passed the board
examinations is NOT valid.
 PRETEST- Problem
Z- test Testing Proportions
Jegie 93

Problem No. 3

The records of a hospital show that 52 men in a sample of 1000 men versus 23 women
in a sample of 1000 women were admitted because of heart disease. Do these data present
sufficient evidence to indicate a higher rate of heart disease among men admitted to the
hospital? Use  = 0.05.

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion
 PRETEST- Answers
Z- test Testing proportions
Jegie 94

I - Specific Problem

Do these data present sufficient evidence to indicate a higher rate of heart disease
among men admitted to the hospital?

II - Null Hypothesis (Ho): Pm = Pf

The data presented is not sufficient evidence to indicate that a higher rate of heart
disease among men admitted to the hospital.

III - Alternative Hypothesis (H1 or HA): Pm > Pf

The data presented is sufficient evidence to indicate that a higher rate of heart
disease among men is admitted to the hospital.

IV - Level of Significance:

 = 0.05

V - Statistical Method Used

Z-test Testing Proportions

VI - Region of Rejection

Z 0.05 = 1.645 for one-tailed test, please refer to table 4.

VII - Computation

P1 = 0.052 P2 = 0.023 P = 0.0375 q = 0.9625 qp = 0.008496

Z= 3.41 = 1.85

VIII- Decision

Since the computed value Zobs = 1.85 is greater than the critical value at 0.05 level
of significance which is equal to 1.645 therefore accept the alternative hypothesis.

IX - Conclusion

The data presented is sufficient evidence to indicate that a higher rate of heart disease
among men is admitted to the hospital.
 Z TEST
TESTING PROPORTIONS
Jegie 95

Z Test

TESTING PROPORTIONS

Definition

❖ This is used when we wanted to get the proportion of

one variable compared to another variable.

❖ When the sample is small, the test used is the binomial

probability distribution. When the sample size is large ( n 
100 ), we use this test

Examples:

Studies are sometimes concerned with proportions such as

the proportion of voters in favor of a particular candidate

the proportion of viewers who prefer a certain TV program
the proportion of consumers who go to for a certain brand of a
product.

Formula

Z = Po - P
p

where p = PQ
n

Here: Po = sample population

P = population proportion
p = standard error
n = sample size
Q = 1 - P
 Z TEST
TESTING PROPORTIONS
Jegie 96

TAKE NOTE

In case where the sampling fraction n/N is greater than 5%, we use a
finite population correction.

p = PQ N-n
n N–1

Note that Po = n’
n

where n’ being the actual number of successes in a sample.

n samples
N population

Problem 1

An exporter of prawns claims that 90 % of his harvest meets certain specifications. In a sample
of 320 prawns, only 280 meets the specifications. At an  = 0.01, determine the legitimacy
of the exporter’s claim.

Solution

I Specific Problem

Is there a legitimacy in the exporter’s claim that 90 % of his harvest meets certain
specifications?

II Null Hypothesis Ho: or p = 0.90

There is a legitimacy in the exporter’s claim that 90 % of his harvest meets certain
specifications.
 Z TEST
TESTING PROPORTIONS
Jegie 97

III Alternative Hypothesis: p < 0.90

There is no legitimacy in the exporter’s claim that 90 % of his harvest

meets certain specifications.

IV Level of Significance

 = 0.01

V Statistical Method Used

Z test in testing proportions

VI Rejection Region

The test we use should be one-tailed since H1 is directional. At an alpha () = 0.01,
z’ = -2.33. If the computed z is < -2.33 we reject Ho. Take note, neglect the sign or
take the absolute value.

VII Computation

a) Po = n’ b) Q = 1 - P where P = 90%
n
Q = 1 - 0.90
Po = 280
320 Q = 0.10

Po = 0.875

c) Qp = PQ d) Z = Po - P
n Qp

Qp = (0.9) (0.1) Z = 0.875 – 0.9

320 0.017

Qp = (0.09) Z = -1.47 neglect -

320 the sign

Qp = 0.00028125 Z = 1.21

Qp = 0.017
 Z TEST
TESTING PROPORTIONS
Jegie 98

VIII Decision

Since the computed value of Z = 1.21 ( Zcomp = 1.21) is less than the critical value
at 0.01 level of significance ( Z0.05 = 2.33 ), therefore accept the null hypothesis.

IX Conclusion

Therefore, there is a legitimacy in the exporter’s claim that 90 % of his harvest meets
certain specifications.

Problem 2

Dr. Matalino claims that of 2,000 accounting graduates of a college of which he is the
dean, at least 85 % passed the board examinations. If out of the 200 examinees, only 154
passed the board, determine the validity of the claim at an  = 0.05 level of significance.

Solution

I Specific Problem

Is the claim of Dr. Matalino that during the time when he was still the dean of the
college, out of 2,000 accounting graduates, 85 % of them passed the board
examinations?

II Null Hypothesis Ho or (P = 0.85)

Dr. Matalino’s claim that 85 percent of the graduates passed the board examinations is
true.

III Alternative Hypothesis ( P < 0.85)

Dr. Matalino’s claim that 85 percent of the graduates passed the board examinations is
NOT true.
 Z TEST
TESTING PROPORTIONS
Jegie 99

IV Level of Significance

 = 0.05

V Statistical Method Used

Z test in testing proportions

VI Rejection Region

The alternative hypothesis ( H1 ) is directional so we use a one-tailed test. At an

 = 0.05, Z’ = -1.645 so that if Z < -1.645 we reject the Null Hypothesis ( Ho ).

VII Computation

Since the population is finite with n/N = 200/2000 = 0.1 or 10 % which is greater than
5 %. Hence, we use

Formula: Z = Po – P
p

where:
1) p = PQ N-n
n N-1

Here P = 0.85 or 85 % Q = 1 - P = 1 – 0.85 = 0.15

n = 200 N = 2000

p = (0.85) (0.15) 2000 - 200

200 2000 - 1

p = 0.1275 1800
200 1999

p = 0.0006375 0.900

p = (0.025) ( 0.949 )

p = 0.0237
 Z TEST
TESTING PROPORTIONS
Jegie 100

2) Po = n’
n
Here
n’ = 154

n = 200

Po = 154 = 0.77
200

3) Substitute the solved values to the given equation.

Z = Po – P
p

where: Po = 0.77

P = 0.85 or 85 %

Z = 0.77 - 0.85
0.0237

Z = - 0.08
0.0237

Z = - 1.84 (neglect the - sign)

VII Decision

Since Z = 1.84 is greater than 1.645, we reject the null hypothesis.

VIII Conclusion

Hence, the dean’s claim is NOT valid that 85 percent of the graduates passed the board
examinations is true.
 Z TEST
PROPORTIONS
Jegie 101

Z Test

PROPORTIONS
Definition

This is used if certain studies are concerned with the difference of two
proportions. Given are the sample proportions, sizes of two samples and
population proportions.

Certain studies are concerned with the difference of two proportions such as:

❖ difference in the proportion of male faculty members and the

proportion of female faculty members who are in favor of divorce.

❖ difference in the proportion of students who passed an examination in

groups A and B.

If P1 and P2 are the sample proportions, n1 and n2 are the sizes of two
samples and P1 and P2 are population proportions then we can make use of
the Z test PROPORTIONS.

Formula Number 1

Z = P 1 - P2

P1 ( 1 - P1) + P2 ( 1 - P 2 )
n12 n22

p1 - p2
Z =

P1 Q1 + P2 Q2
n12 n22

where P1 = sample proportion for A

P2 = sample proportion for B
n1 = number of samples for A
n2 = number of samples for B
 = population proportion
Q =  - 1
 Z TEST
PROPORTIONS
Jegie 102

Formula Number 2

Note When the population proportions are known. Very often however,
they are not known so use

Z = P1 - P2
p

and p = pq ( 1 + 1 )
n1 n2

p = N1 P1 + N2P2
N1 + N2

q = 1 - p

where:
P1 = proportion for A
P2 = proportion for B
n1 = number of samples for A
n2 = number of samples for B
 = population proportion
q = 1 - p
 Z TEST
PROPORTIONS
Jegie 103

Problem

An examination was administered to two groups of students. In group A, 10 out of 50

failed while in group B, there were 15 out of 60 students who failed. Is there a difference in
the performance out of the two groups of students at  = 0.01 ?

Solution

I Specific Problem

Is there a difference in the performance out of the two groups of students?

II Null Hypothesis

There is no significant difference in the performance out of the two groups.

III Alternative Hypothesis

There is no significant difference in the performance out of the two groups.

IV Level of Significance

 = 0.01

V Statistical Method Used

Since the population proportions are not known, we use Z test Proportions

VI Rejection Region

We use a two tailed test. At an  = 0.01, Z’ = + 2.33 which means the acceptance
region is -2.33  Z  2.33. If Z fails outside of this region we reject Ho.

VII Computation

1. Solve for the value of p

p = N1 P1 + N2P2
N1 + N2
 Z TEST
PROPORTIONS
Jegie 104

p = 50 (10/50) + 60 (15/60)
50 + 60

p = 50 (10/50) + 60 ( 15/60)
50 + 60

P = (50) (0.20) + 60 (15/60)

110

P = 10 + 15
110

25
P =
110

P = 0.23

2. Then solve for the value of q.

q = 1 - p
q = 1 - 0.23
q = 0.77

3. Find the value of p.

p = pq ( 1 + 1 )
n1 n2

p = (0.23) (0.77) ( 1 + 1 )
50 60

p = ( 0.23) (0.77) ( 0.02 + 0.017)

p = (0.23) (0.77) (0.037)

 Z TEST
PROPORTIONS
Jegie 105

p = 0.0065527

p = 0.081

3. Compute for the value of Z.

Z = p1 - p 2
p

here : p1 = 10 = 0. 20
50

p2 = 15 = 0.25
50

Z = 0.20 - 0.25
0.081

Z = - 0.05
0.081

Z = - 0.62 neglect the - sign

Z = 0.79

VIII Decision

Since the computed value of Z = 0.0.79 is less than the critical value which is
Z0.01 = 2.33 therefore accept the null hypothesis.

IX Conclusion

There is no significant difference in the performance out of the two groups.

 POST TEST-Problem
Z Test Testing Proportion
Jegie 106

Problem No. 1

It is believed that at least 60 % of the residents in a certain area favor an annexation

suit by a neighboring city. What conclusion would you draw if only 110 in a sample of 200 voters
favor the suit? Use a 0.05 level of significance.

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion
 POST TEST- Answers
Z Test Testing Proportions
Jegie 107

I - Specific Problem

Do you believe that at least 60 % of the residents in a certain area favor an annexation
suit by a neighboring city.

II - Null Hypothesis (Ho): P = 0.60

Sixty percent of the residents in a certain area do not favor an annexation suit by a
neighboring city.

III - Alternative Hypothesis (H1 or HA) : P < 0.60

Sixty percent of the residents in a certain area favor an annexation suit by a neighboring
city.

IV - Level of Significance:

 = 0.05

V - Statistical Method Used

Z-test Testing proportions

VI - Region of Rejection

Z 0.05 = 1.645

VII - Computation

Po = 0.55 Q = 0.40 Qp = 0.035 Zobs = - 1.43 = 1.20 neglect the –

sign

VIII- Decision

Since the computed Zobs = 1.20 is less than the critical value at 0.05 level of
significance which is equal to 1.645 therefore accept the null hypothesis.

IX - Conclusion

Sixty percent of the residents in a certain area favor an annexation suit by a neighboring
city.
 POST TEST-Problem
Z Test Testing Proportion
Jegie 108

Problem No. 2

The president of a certain university claims that 75 % of 500 graduates of the college of
nursing passed the board examinations. If out of 150 examinees only 75 passed the board,
determine the validity of the claim at an  = 0.01 level of significance.

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion
 POST TEST- Answers
Z Test Testing Proportions
Jegie 109

I - Specific Problem

Is the claim of the president of a certain university valid that out of the 500 graduates
of the college of nursing 75 % passed the board examinations?

II - Null Hypothesis (Ho): P = 0.75

The claim of the president of a certain university valid that out of the 500 graduates of
the college of nursing 75 % passed the board examinations.

III - Alternative Hypothesis (H1 or HA) : P < 0.60

The claim of the president of a certain university is NOT valid that out of the 500
graduates of the college of nursing 75 % passed the board examinations?

IV - Level of Significance:

 = 0.05

V - Statistical Method Used

Z-test Testing proportions

VI - Region of Rejection

Z 0.01 = 2.33

VII - Computation

qp = 0.0296 Po = 0.50 P = 0.75 Z = - 8.45 neglect the - sign

VIII- Decision

Since the computed Zobs = 8.45 is greater than the critical value at 0.05 level of
significance which is equal to 2.33 therefore accept the alternative hypothesis.

IX - Conclusion

The claim of the president of a certain university is NOT valid that out of the 500
graduates of the college of nursing 75 % passed the board examinations?
 POST TEST-Problem
Z Test Testing Proportion
Jegie 110

Problem No. 3

A vote is to be taken among the residents of a town and the surrounding country to
determine whether a proposed chemical plant should be constructed. The construction site is
within the town limits and for this reason many voters in the country feel that the proposed will
pass because of the large proportion of town voters who favor the construction. To determine if
there is a significant difference in the proportion of town voters and country voters favoring the
proposal, a poll is taken. If the 120 of 200 town voters favor the proposal and 240 of 500 country
residents favor it, would you agree that the proportion of town voters favoring the proposal is
higher than the proportion of country voters? Use  = 0.025 level of significance.

Solve the problem by getting the following requirements:

1). Specific Problem

2). Null Hypothesis
3). Alternative Hypothesis
4). Statistical Method Used
5). Level of Significance
6). Region of Rejection
7). Computation of the value of Z
8). Decision
9). Conclusion
 POST TEST- Answers
Z Test Testing Proportions
Jegie 111

I - Specific Problem

Do you agree that the proportion of town voters favoring the proposal is higher than the
proportion of country voters?

II - Null Hypothesis (Ho): P1 = P2

The proportion of town voters favoring the proposal is NOT higher than the proportion of
country voters?

III - Alternative Hypothesis (H1 or HA) : P1 > P2

The proportion of town voters favoring the proposal is higher than the proportion of
country voters?

IV - Level of Significance:

 = 0.025

V - Statistical Method Used

Z-test Testing proportions

VI - Region of Rejection

Z 0.025 = 1.96

VII - Computation

P1= 0.60 P2 = 0.48 P = 0.514 q = 0.486 qp = 0.0418

Z = 2.87 Z = 1.69

VIII- Decision

Since the computed Zobs = 2.87 is lesser than the critical value at 0.025 level of
significance which is equal to 1.96 therefore accept the alternative hypothesis.

IX - Conclusion

The proportion of town voters favoring the proposal is higher than the proportion of
country voters?