Tesis Biostatistik(2)

1

CHAPTER 1

INTRODUCTION

1.1

BACKGROUND

According to Malaysian Road Safety Department,3.9% road death increases in 2008 compared to the previous year. Traffic signs represent one of the most common devices for controlling traffic in that they help regulate, warn, and guide road users. In spite of their importance, traffic signs are not always clear to the drivers (Tamar & David 2006). 1.2

METHODOLOGY JUSTIFICATION

i.

To increase awareness and driving etiquette on road.

ii.

To decrease road accidents risk.

iii.

To improve the knowledge regarding road signs among students.

iv.

To know the comprehension regarding road signs among students.

2

1.3

CONCEPTUAL FRAMEWORK

GOVERNMENT

Campaign

NONGOVERNMENT

Mass Media

Driving school

Surrounding

SOURCE ROAD SIGN KNOWLEDGE Gender factor

1.3.1

License factor

Area factor

Gender factor

We are taking male and female students. 1.3.2

License factor

The male and female students taken are divided between ones having license and the other without license.

3

1.3.3

Government

Government took part in holding campaigns and restricted the laws. 1.3.4

Media

Media serve by the television, radio, newspaper, and information computer and technology (ICT) 1.3.5

Education

The driving school and driving lesson 1.3.6

Surrounding factor

Urban and rural areas are taken into study. 1.4

OBJECTIVE

1.4.1

General Objective

To study the understanding about road signs among the first year FSKB’s students in UKM, KL session 2008/2009. 1.4.2

Specific Objective

i.

To identify the student’s knowledge about the road signs.

ii.

To determine the differences in knowledge between gender about the road signs.

iii.

To determine the differences in knowledge about the road signs between student having and not having license.

iv.

To determine the differences in knowledge about the road signs among the student who living in urban and rural area

4

v. 1.5

To identify the students’ sources of knowledge about the road sign. Hypothesis

i. There are differences between knowledge’s score and comprehension‘s score regarding the road signs among the first year students of FSKB .(question part B) ii. There are differences in the score of knowledge between gender, license availability and area. iii.There are differences in score of knowledge on road signs between gender. (question part A – no 1 , part B & C ) iv. There are differences in score of knowledge on road signs between students with and without driving license.(question part A – no 4, part B & C ) v. There are differences in score of knowledge on road signs between students living in urban and rural area. (question part A – no 3, part B & C ) iv.There are association between the gender/area/license with sources of knowledge on the road sign.(question part A-no 7)

student’s

5

CHAPTER 2

LITERATURE REVIEW

2.1

DEFINITION

Definition of sign is an indication, an event, an action and a fact that shows that something exists or may happen that you can find and see. Meanwhile, the definition of road is a hard surface built for vehicles (Oxford Advanced Learner’s Dictionary 6th Edition). The general definition of road signs is a sign near a road giving information or instruction to driver. Then, the specific definitions of road signs are used to give information about the location of either the driver or possible destinations and are considered a subset of informative sign group (Ross & Alan, 1992). 2.2

TYPES OF ROAD SIGN

Types of road signs are divided into three parts which is first, law road signs such as no entry, speed zone and stop. Secondly, warning road signs such as dangerous bend road, slippery road and accident spot. Lastly, direction road signs such as destination sign board and information sign board (Law T.H, 2004). Every each of these types should be distinct in its shape and colors (Tama and David, 2006). 2.3

FUNCTIONS OF ROAD SIGN

Functions of road signs are use to arrange traffic, to warn and act as guidance to road users. Besides, the designs of the road signs which are big, simple and similar are

6

easier for understanding, plus noticeable. This it can give enough time for road user to be ready for a certain unexpected condition such as sudden animal crossing. The roan signs are place in plain sight. Furthermore, the road signs are informative in terms of providing directions. (Marc Green & John Senders, 2004)

7

CHAPTER 3

RESEARCH METHODOLOGY

3.1

BACKGROUND

In order to do this research, we will survey the 1st year students of Faculty of Allied Health Sciences (FSKB), National University of Malaysia, Kuala Lumpur (UKMKL) intake 2008/2009. The selected students come from 12 different courses which are Biomedical Science Department, Audiology and Speech Science Department, Dietetic and Nutrition Department, Optometry Department, Diagnostic and Radiotherapy Program, Occupational Therapy Program, Physiotherapy Program, Environmental Health Program, Forensics Science Program and also Emergency Medicine Program. From all the 383 students of the 1st year in FSKB, only 192 are selected to be our respondents. Apart from that, we will later pick later based on the ratio of male to female from the answered questionnaires. We then will get the population of student either having or not having license. The license can be either `L` or` P` or even full license also known as Competent License that are registered under Malaysian Road Transport 3.2

Department

(JPJ).

RESEARCH DESIGN

We have selected the best way of designing our research. The cross-sectional study will be the best and suitable research design for us. Basically, we don`t refer to any other sources to get the result but we have to do the result based on our questionnaire to the respondent. Moreover, the result can be analyzed easily using the SPSS system. We also use open survey type question to gather all the respondent data.

8

3.3

SAMPLING METHOD

3.3.1. Target Population = Students of UKM, Kuala Lumpur. 3.3.2. Sample Population = 1st year students of FSKB intake 2008/2009 in UKMKL. 3.3.3. Sample Size = 192 students.(calculation from sample size of selected population formula below) Calculating sample size n = ____X2NP (1-P)____ ∆2(N-1) + X2P(1-P) = 192 Where X2 = 3.84, ∆ = 0.05, P = 0.5 But 10% would drop out so, n* = __192__ (1-0.1) = 213.3 = 214 Questionnaire  Distribution

 Received

: 214 : 195

 Not received: 19  Percentage of unreceived : 9.74%

9

3.4

METHOD OF GETTING THE DATA

i.

List name of first year students of FSKB UKM, KL are collected.

ii.

Using stratified sampling method in divide the student population according to gender. Followed by systematic random sampling to distribute the questionnaire.

iii.

Questionnaires consist of multiple choice, text open end and agreement scale (close end) test types are distribute among the samples.

10

CHAPTER 4

RESEARCH ANALYSIS

4.1

RESEARCH ANALYSIS

The collected data from the completed survey forms are analyzed using the SPSS. The following are the tests used for this research: 1. Descriptive Statistic. 2. Independent t test

3. Chi square 4. Logistic Regression 4.2

Data analysis

Objective 1 :To identify the student’s knowledge about the road signs. Hypothesis 1: There are differences between knowledge’s score and comprehension’s score regarding the road sign among the first year FSKB’s students. Test: Descriptive Statistics

11

Road signs

Comprehension score (%) 1 2 3 4

5

Knowledge score (%) Correct Incorrect

60..0

19.0

5.1

6.7

9.2

90.8

9.2

39.0

31.3

16.9

9.7

3.1

45.6

54.4

65.6

19.5

4.6

2.1

8.2

95.9

4.1

72.8

13.8

2.6

2.1

8.7

92.3

7.7

73.3

16.4

1.0

2.1

7.2

97.4

2.6

56.9

22.1

8.7

6.2

6.2

96.4

3.6

50.8

23.6

11.8

7.2

6.7

90.3

9.7

51.3

30.3

7.7

5.1

5.6

87.7

12.3

31.8

35.4

17.9

10.8

4.1

45.6

54.4

53.3

27.7

8.7

5.1

5.1

89.2

10.8

Table 1.0: Student’s comprehension and knowledge score. Section B(a)1 vs Section B(b)1

12

70

64.4

comperhension (%)

60 50 40 27.8

30 20 10

22.2

18.6

22.2

16.7

4.5

4.5

11.1

7.9

0 correct

really understand a bit understand don't know

incorrect

understand little understanding

There are 64.4% students who answer correctly and really understand about this road sign and only 7.9% did not know about this road sign but still can answer correctly. There are only 16.7% said really understand but still answer incorrectly and 22.2% students who answer incorrectly and did not know about this road sign although this road sign quite common in used representing hospital.

13

Section B(a)2 vs Section B(b)2

50 44.3

45

comperhension (%)

40 35

32.6

32.6

30.2

30 25 20

20.2

15

14.2

12.4

10

7.5

5

3.8

2.2

0 correct

really understand a bit understand don't know

incorrect

understand little understanding

There are 32.6% students who answer correctly and really understand about this road sign and only 2.2% did not know about this road sign but still can answer correctly.However,44.3% who said really understand but answer incorrectly and only 3.8% students who answer incorrectly and do not understand about this road sign. It is because There are misunderstands this road sign with bumper sign.

Section B(a)3 vs Section B(b)3

14

80

70

66.8

comperhension (%)

60 50 37.5 37.5

40

30

20

18.7 12.5 12.5

10

8.6 4.3

1.6

0

0 correct really understand little understanding

incorrect understand don't know

a bit understand

There are 68.8% students who answer correctly and really understand about this road sign and only 8.6% did not know about this road sign but still can answer correctly. There are 37.5% who really understand but did answer incorrectly and 37.5% students who answer incorrectly and understand about this road sign.

Section B(a)4 vs Section B(b)4

15

80

73.9

70 60

comperhension (%)

60 50 40 30 20

13.9

13.3 8.9

10 2.2

13.3 6.7

6.7

1.1

0 correct

really understand a bit understand don't know

incorrect

understand little understanding

There are 73.9% students who answer correctly and really understand about this road sign and only 8.9% did not know about this road sign but still can answer correctly. But there are 60% who really understand but did answer incorrectly and 6.7% students who answer incorrectly and not understand about this road sign. Supposedly, student should know about this road sign because this road sign familiar for us and we can see this road sign in every parking lot.

16

Section B(a)5 vs Section B(b)5

90 80

80 73.2

comperhension (%)

70 60 50 40 30 20

20

16.3 7.4

10 1.1

2.1

0 correct

really understand a bit understand don't know

0

0

0

incorrect

understand little understanding

17

There are 73.2% students who answer correctly and really understand about this road sign and only 7.4% did not know about this road sign but still can answer correctly. Unfortunately,80% of students said really understand but still answer incorrectly and about 20% students who answer incorrectly but understand about this road sign.There is no students answer incorrectly and don’t know about this road sign.

Section B(a)6 vs Section B(b)6

18

60

57.1

56.9

comperhension (%)

50

40

28.6

30 21.8 20

14.3 10

8.5

6.4

6.4 0

0 correct

0

incorrect

really understand

understand

little understanding

don't know

a bit understand

There are 56.9% students who answer correctly and really understand about this road sign and only 6.4% did not know about this road sign but also answer correctly. On the other hand,a large percentage,that is around 57.1% who said really understand but

19

answer incorrectly and about 14.3% students who answer incorrectly but a bit understand about this road sign.

Section B(a)7 vs Section B(b)7

20

60

52.8

comperhension (%)

50 40 30 20

31.6 23.3 11.4

10

26.3 15.8 15.8

6.3 6.3

10.5

0 correct really understand a bit understand don't know

incorrect understand little understanding

There are 52.8% students who answer correctly and really understand about this road sign and only 6.3% did not know about this road sign but still can answer correctly.

21

There are 31.6% students who really understand but also answer incorrectly and about 10.5% students who answer incorrectly did not know about this road sign

Section B(a)8 vs Section B(b)8

22

60

55

comperhension (%)

50

45.8

40

30

28.1 25

20 12.5 10

7

5.3

12.5 4.2

0.7 0 correct

really understand a bit understand don't know

incorrect

understand little understanding

There are 55% students who answer correctly and really understand about this road sign and only small value that is 0.7% did not know about this road sign but still can answer correctly. There are 25% students who really understand and 45% said

23

understand but answer incorrectly. About 12.5% students who answer incorrectly and did not know about this road sign.

Section B(a)9 vs Section B(b)9

24

50 39.3

comperhension (%)

40 30 20

34.9

28.1 19.1

17 9

10

32.1

12.3

4.5

3.8

0 correct really understand a bit understand don't know

incorrect understand little understanding

There are only 28.1% students who answer correctly also really understand,39.3% said understand about this road sign and 4.5% did not know about this road sign but still can answer correctly. However, 34.9% students who said really understand and 32.1% said understand but answer incorrectly and only about 3.8% students who

25

answer incorrectly and did not know about this road sign. Maybe, students confuse between this road sign with do not parking road sign.

Section B(a)10 vs Section B(b)10

26

60

55.2

comperhension (%)

50

38.1

40

30

28.2 23.8 19

20

14.3 10

9.2 3.4

4

4.8

0 correct

really understand a bit understand don't know

incorrect

understand little understanding

There are 55.2% students who answer correctly and really understand about this road sign and only 4% did not know about this road sign but still can answer correctly. There are 38.1% students who really understand but also answer incorrectly and about

27

14.3% students who answer incorrectly and did not know about this road sign. It is because there misunderstands with the narrow bridge sign. Knowledge’s score and comprehension’s score about road signs are depend on types of road signs and evaluation of students about that road sign. So, there are differences between knowledge’s score and comprehension’s score regarding the road sign among the first year FSKB’s students. Objective 2: To determine the differences in knowledge between gender about the road signs. Hypothesis: There are differences in score of knowledge on road signs between gender. The score are taken through the answer of question part A – no 1, part Ba & C. •

HA, μ1≠μ2: There are differences in score of knowledge on road signs between gender.

Table 2.0: Test of normality for gender factor. Gender

score

Kolmogorov-Smirnov(a) Statistic

Df

Sig.

Male

0.194

45

0.000

female

0.171

150

0.000

Base on the test of normality Kolmogorov-Smirnov(a) , the significant level of p value is lower than 0.001. It is significant. Thus, the data is not normally distributed. Table 2.1: Descriptive table for gender factor. Gender

statistic

Std. dev.

28

score

male

female

skewness

-2.117

0.354

Kurtosis

7.680

0.695

skewness

-2.618

0.198

Kurtosis

12.599

0.394

Through the calculation of skewness, the standard deviation (0.354) multiply by 2 and the answer (0.708) is in the range of statistic value (-2.117 to +2.117). It shows that, the distribution of the data is normal for the male. For the female, standard deviation (0.198) multiply by 2 and the answer (0.396) is also in the range of statistic value (-2.618 to +2.618). It shows that, the distribution of the data is absolutely normal. It is a parametric analysis.

To compare the mean score of the two groups which are male and female student, independent sample t-test is use. Score as the test variable and gender as the grouping variable. Table 2.2: Statistical test for gender factor. Gender

n

Mean

Standard Deviation

p-value

Male

45

13.8667

2.24216

0.207

Female

150

14.3533

2.26481

Mean score of knowledge and standard deviation for male is 13.8667 and 2.24216 while for female is 14.3533 and 2.26481. On the output result, Levene’s test is higher than 0.05. It is assume that the data variances are relatively equal. Therefore, the upper row of the significant value is use. Base on it, the significance level of p value on the upper row is higher than 0.05. Thus, the mean score of knowledge of the two groups are not significantly different. t=1.267, df=193, p>0.05.

29

Objective 3: To determine the differences in knowledge about the road signs between student having and not having license. Hypothesis: There are differences in score of knowledge on road signs between student with and without driving license. The score are taken through the answer of question part A – no 4, part Ba & C.

•

HA, μ1≠μ2: There are differences in score of knowledge on road signs between student with and without driving license.

Table 3.0: Test of normality for license factor. License

score

Kolmogorov-Smirnov(a) Statistic

Df

Sig.

Yes

0.164

134

0.000

No

0.171

61

0.000

Base on the test of normality Kolmogorov-Smirnov(a) , the significant level of p value is lower than 0.001. It is significant. Thus, the data is not normally distributed. Table 3.1: Descriptive table for license factor. license

statistic

Std. dev.

30

score

yes

no

skewness

-1.968

0.209

Kurtosis

9.834

0.416

skewness

-2.060

0.306

Kurtosis

6.669

0.604

Through the calculation of skewness, the standard deviation (0.209) multiply by 2 and the answer (0.418) is in the range of statistic value (-1.968 to +1.968). It shows that, the distribution of the data is normal for student with license. For student without license, standard deviation (0.306) multiply by 2 and the answer (0.612) is also in the range of statistic value (-2.060 to +2.060). It shows that, the distribution of the data is absolutely normal. It is a parametric analysis.

To compare the mean score of the two groups which are student with license and student without license, independent sample t-test is use. Score as the test variable and gender as the grouping variable. Table 3.2: Statistical test for license factor. License

n

Mean

Standard Deviation p-value

Yes

134

14.5746

13.5082

No

61

1.72717

3.02557

0.012

Mean score of knowledge and standard deviation for student with license is 14.5746 and 13.5082 while for student without license is 1.72717 and 3.02557. On the output result, Levene’s test is lower than 0.05. It is assume that the data variances are relatively different. Therefore, the lower row of the significant value is use. Base on it, the significance level of p value on the lower row is lower than 0.05. Thus, the means score of the knowledge on road sign between student with and without driving license are different.

31

t=2.569, df=78.344, p<0.05. Objective 4: To observe the dissimilarities in knowledge about the road signs among the student who living in urban and rural area. Hypothesis: There are differences in score of knowledge on road signs between students living in urban and rural area. The score are taken through the answer of question part A – no 3, part Ba & C.

•

HA, µ1≠µ2 : There are differences in score of knowledge on road signs among the student who living in urban and rural area.

Table 4.0: Test of normality for area of residential factor. Gender

score

Kolmogorov-Smirnov(a) Statistic

Df

Sig.

Male

0.191

122

0.000

female

0.162

73

0.000

Base on the test of normality Kolmogorov-Smirnov(a) , the significant level of p value is lower than 0.001. It is significant. Thus, the data is not normally distributed. Table 4.1: Descriptive table for area of residential factor. area score

urban

rural

statistic

Std. dev.

skewness

-3.095

0.219

Kurtosis

15.932

0.435

skewness

-1.498

0.281

Kurtosis

4.348

0.555

32

Through the calculation of skewness, the standard deviation (0.219) multiply by 2 and the answer (0.438) is in the range of statistic value (-3.095 to +3.095). It shows that, the distribution of the data is normal for student live in urban. For student live in rural, standard deviation (0.281) multiply by 2 and the answer (0.562) is also in the range of statistic value (-1.498 to +1.498). It shows that, the distribution of the data is absolutely normal. It is a parametric analysis. To compare the mean score of the two groups which are student who living in urban area and student who living in rural area, independent sample t-test is use. Score as the test variable and area as the grouping variable. Table 4.2: Statistical test for area of residential factor. Area

n

Mean

Standard Deviation

p-value

Urban

122

14.3115

2.24927

0.575

Rural

73

14.1233

2.29701

Mean score of knowledge and standard deviation for student live in urban area is 14.3115 and 2.24927 while for student live in rural area is 14.1233and 2.29701.

On the output result, Levene’s test is higher than 0.05. It is assume that the data variances are relatively equal. Therefore, the upper row of the significant value is use. Base on it, the significance level of p value on the upper row is higher than 0.05. Thus, the mean scores of the knowledge on road sign among student who living in urban and rural area are no different. t=0.561, df=193, p>0.05.

33

Objective 5: To identify the sources of student’s knowledge on the road sign. Hypothesis: There are associations between the gender/area/license with sources of student’s knowledge on the road sign. •

Ha = Sources of student’s knowledge on the road sign dependent on gender.

Variables •

Sources of knowledge such as driving school or non-driving school (mass media, campaign, environment, others)

•

Gender of student

Chi-square test Case Processing Summary Valid Percent

N gender * sources3

166

85.1%

Cases Missing N Percent 29

14.9%

N

Total Percent 195

100.0%

Figure 1.0: The association graph of sources within gender.

The Association Graph of Sources Within Gender Percentage (%)

80 60 40

Driving School

20

Non-Driving School

0 Male

Female Gender

34

Table 5.0: Association of gender and sources of student knowledge. Sources

Driving school

Non-driving

Gender

Total

P value

Pearson ChiSquare value

P = 0.423

X 2 = 0.641

school

Male

25(64.1%)

14(35.9%)

39(100.00%)

Female

90(70.9%)

37(29.1%)

127(100.00%)

Total

115(69.3%)

51(30.7%)

166(100.00%)

The Pearson Chi-Square is 0.641. The p value is 0.423. We can conclude that is a not significant association between gender and sources of student’s knowledge on the road sign, and therefore we do not reject the null hypothesis. X 2 = 0.641, df = 1, p > 0.05 Hypothesis: • HA : Sources of student’s knowledge on the road sign dependent on area Variables •

Sources of knowledge such as driving school or non-driving school (mass media, campaign, environment, others)

•

Area of student such as urban and rural

Chi-square test Case Processing Summary Valid N Percent gender * sources3

166

85.1%

Cases Missing N Percent 29

14.9%

Total N Percent 195

100.0%

35

Percentage (%)

Figure 2.0: The association graph of sources within area of residential.

The Association of Sources Within Area

80 70 60 50 40 30 20 10 0

Driving School Non-Driving School

Urban

Rural Area

Table 5.1: Association of residential area and sources of student knowledge. Sources

Driving school

Area

Non-driving

Total

P value

Pearson ChiSquare value

P = 0.085

X 2 = 2.966

school

Urban

77(74.0%)

27(32.0%)

104(100.00%)

Rural

38(61.3%)

24(38.7%)

62(100.00%)

Total

115(69.3%)

51(30.7%)

166(100.00%)

The Pearson Chi-Square is 2.966. The p value is 0.085. We can conclude that is a not significant association between gender and sources of student’s knowledge on the road sign, and therefore we do not reject the null hypothesis. X 2 = 2.966, df = 1, p > 0.05 Hypothesis:

36

•

HA : Sources of student’s knowledge on the road sign dependent on license.

Variables •

Sources of knowledge such as driving school or non-driving school (mass media, campaign, environment, others)

•

License of student

Chi-square test Case Processing Summary

N gender * sources3

Valid Percent 166

Cases Missing N Percent

85.1%

29

N

14.9%

Total Percent 195

100.0%

Figure 3.0: The association graph of sources within license.

The Association Graph of Sources Within License 90 80 Percentage (%)

70 60 50 40 Driving School

30

Non-Driving School

20 10 0 Yes

No License

Table 5.2: Association of license and sources of student knowledge.

37

Sources

Driving school

Non-driving

License

Total

P value

Pearson ChiSquare value

P = 0.001

X 2 = 43.813

school

Yes

99(84.6%)

18(15.4%)

117(100.00%)

No

16(32.7%)

33(67.3%)

49(100.00%)

Total

115(69.3%)

51(30.7%)

166(100.00%)

The Pearson Chi-Square is 43.813. The p value is 0.001. We can conclude that is a not significant association between gender and sources of student’s knowledge on the road sign, and therefore license seems to be must factor to contribute sources of knowledge for student compare to gender and residential area. X 2 = 43.813, df = 1, p < 0.05

Objective 1: To identify the student’s knowledge about the road signs. Hypothesis: There are different in the score of knowledge between gender, license availability and area of residential. The score are taken through the answer of question part A-no 1,3,4, part Ba & C. Test: Binary logistic regression. Variable not in Equation Variable

Gender

Score .017

Df 1

Sig. .895

Area

.562

1

.453

License

7.417

1

.006

There are significant value show and it only on the availability of license that is 0.006 (p<0.05) compared to gender, 0.895 and area, 0.453 which is greater than 0.5. Variable in the Equation Step 1

Gender Area

B .658

Sig. .461

Exp(B) 1.931

-.088

.908

.916

38

License

-2.091

.018

.124

For the data on interaction on mean score of knowledge to the gender, residential area and availability of license to each student. From the table, there is negative value for data in column B. the negative value shows the opposite interaction of the second factor from the first factor to the score of knowledge Here, interpreted that the mean score of knowledge for the second factor (female student) is higher (due to positive value of B) 1.931 times (Exp(B) value) from the first value (male student). Also, the data for this factor showing a non-significant value, p=0.461 For area or residential, it shows that the mean score of knowledge of the second factor (rural area) is lower (due negative value of B) 0.916 times compared to student from urban area. The data of significant also shows there is no significant value for this factor, p=0.908 License showing the score of knowledge for the second factor (not having license) to be low than the first factor (having license) by 0.124 times less. But, the significant value show that there is a significant data to be observed, p=0.018 There are different in the score of knowledge on license but there is no different in the score of knowledge between genders and resident.

39

40

CHAPTER 5

DISCUSSION 5.1

DISCUSSION

A recent study that evaluated comprehension of traffic sign in four different countries show that comprehension level varies widely and is apparently related to the extend that the sign’s design incorporate ergonomic guidelines for good design(Shinar D. et al 2003). Based on our research, we found that comprehension and knowledge of students about the road sign is depend on types of the road sign. There are road signs that show high scores of comprehension and knowledge about the road sign but there are also road signs that show high score of comprehension but low in score of knowledge about the road sign and etc. What we can get from this situation is sign design should be guided by established ergonomics principles to enhance comprehension, especially for drivers who have not had prior encounters with specific signs (Tamar B. & Shinar D. 2006).

From the Chi-square test, it shows that the driving school is the main source in contribute to student knowledge in the road sign compared to mass media, campaign, environment and others. This is because, from the Kementerian Penerangan Malaysia, to get the license from driving school each individual need to pass road law test and usually the test is done by on-line. In this test, every participant must achieve the standard marks that standardized by Jabatan Pengangkuatan Jalan, Malaysia. After that, they will expose and apply their knowledge about the road sign during lesion and test of license. So, the experiences in the driving school help them to increase their knowledge about the road sign. In other words, the most factor that influence the score of knowledge is license compared to other factors that is gender and residential of students.

41

The knowledge about the road signs is very important because from Dr. Haji Mat Saad Abdul Rahman, Fellow Kanan Syariah Pusat Syariah, Undang-undang dan Sains Politik, Institute of Islamic Understanding Malaysia (IKIM), presence of road sign in certain location especially in danger zone is one of important matter to decrease the fatality rate in road accident.

42

CHAPTER 6

CONCLUSION

6.1

CONCLUSION

License is the most factors that influence the student’s knowledge compared to gender and residential area. Female students obtain higher knowledge more than male. Students that live in rural area obtain lower knowledge than urban area. Student without license obtain lower knowledge less than students with license. 6.2

SUGGESTION

To get the more accurate data, interview is the best way to evaluate student’s knowledge about road sign to reduce the bias. While developing questionnaire, more road signs should be added in questionnaire so that our result fulfill the objective in this research. Furthermore, this questionnaire also can help students to improve their knowledge about the road sign.

43 BIBLIOGRAPHY Australia Road. Road safety Audit. Sydney, Australia.1996. Danish Road Directorate. Manual of Road Safety Audit. Ministry of Transportation. Copenhagen, Denmark. 1996. Public Works Department (JKR). Road Safety Audit. Guidelines for the Safety Audit of Roads Projects in Malaysia, Kuala Lumpur. 1997. Tamar Ben-Bassat, David Shinar. Ergonomic Guidelines for Traffic Sign Design Increase Sign Comprehension. Spring. 2006. http://www.jkr.gov.my [4 Feb 2009} Kurikulum Pendidikan Pemandu Panduan Pembelajaran. Jabatan Pengangkutan Jalan Malaysia. Kuala Lumpur. Edisi Ke 2. 2006.

44 APPENDIX Output SPSS Test: Descriptive Statistics Section B(a)1 vs Section B(b)1 Case Proce ssing Summary

Ba1 * Bb1

Cases Missing N Percent 0 .0%

Valid N Percent 195 100.0%

Total N Percent 195 100.0%

Ba1 * Bb1 Crosstabulation Bb1

Ba1

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba1 Bb1 Ba1 Bb1 Ba1 Bb1

really understand 114 64.4% 97.4% 3 16.7% 2.6% 117 60.0% 100.0%

understand 33 18.6% 89.2% 4 22.2% 10.8% 37 19.0% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 24.584 a 20.407 19.430

4 4

Asymp. Sig. (2-sided) .000 .000

1

.000

df

195

a. 4 cells (40.0%) have expected count less than 5. The minimum expected count is .92.

a bit understand 8 4.5% 80.0% 2 11.1% 20.0% 10 5.1% 100.0%

little understa nding 8 4.5% 61.5% 5 27.8% 38.5% 13 6.7% 100.0%

don't know 14 7.9% 77.8% 4 22.2% 22.2% 18 9.2% 100.0%

Total 177 100.0% 90.8% 18 100.0% 9.2% 195 100.0% 100.0%

45

Bar Chart Bb1

120

really understand understand a bit understand little understanding don't know

100

Count

80

60

40

20

0 correct

incorrect

Ba1

46

Section B(a)2 vs Section B(b)2

Case Proce ssing Summary

Ba2 * Bb2

Valid N Percent 195 100.0%

Cases Missing N Percent 0 .0%

N

Total Percent 195 100.0%

Ba2 * Bb2 Crosstabulation Bb2

Ba2

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba2 Bb2 Ba2 Bb2 Ba2 Bb2

really understand 29 32.6% 38.2% 47 44.3% 61.8% 76 39.0% 100.0%

understand 29 32.6% 47.5% 32 30.2% 52.5% 61 31.3% 100.0%

a bit understand 18 20.2% 54.5% 15 14.2% 45.5% 33 16.9% 100.0%

little understa nding 11 12.4% 57.9% 8 7.5% 42.1% 19 9.7% 100.0%

don't know 2 2.2% 33.3% 4 3.8% 66.7% 6 3.1% 100.0%

Total 89 100.0% 45.6% 106 100.0% 54.4% 195 100.0% 100.0%

47

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 4.375 a 4.396 2.062

4 4

Asymp. Sig. (2-sided) .358 .355

1

.151

df

195

a. 2 cells (20.0%) have expected count less than 5. The minimum expected count is 2.74.

Bar Chart Bb2

50

really understand understand a bit understand little understanding don't know

Count

40

30

20

10

0 correct

incorrect

Ba2

48 Section B(a)3 vs Section B(b)3

Case Proce ssing Summary

N Ba3 * Bb3

Cases Missing N Percent 0 .0%

Valid Percent 195 100.0%

N

Total Percent 195 100.0%

Ba3 * Bb3 Crosstabulation Bb3

Ba3

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba3 Bb3 Ba3 Bb3 Ba3 Bb3

really understand 125 66.8% 97.7% 3 37.5% 2.3% 128 65.6% 100.0%

understand 35 18.7% 92.1% 3 37.5% 7.9% 38 19.5% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 8.644 a 6.546 .611

4 4

Asymp. Sig. (2-sided) .071 .162

1

.435

df

195

a. 5 cells (50.0%) have expected count less than 5. The minimum expected count is .16.

a bit understand 8 4.3% 88.9% 1 12.5% 11.1% 9 4.6% 100.0%

little understa nding 3 1.6% 75.0% 1 12.5% 25.0% 4 2.1% 100.0%

don't know 16 8.6% 100.0% 0 .0% .0% 16 8.2% 100.0%

Total 187 100.0% 95.9% 8 100.0% 4.1% 195 100.0% 100.0%

49

Bar Chart Bb3 120

really understand understand a bit understand little understanding don't know

100

Count

80

60

40

20

0 correct

incorrect

Ba3

50 Section B(a)4 vs Section B(b)4

Case Proce ssing Summary

N Ba4 * Bb4

Cases Missing N Percent 0 .0%

Valid Percent 195 100.0%

N

Total Percent 195 100.0%

Ba4 * Bb4 Crosstabulation Bb4

Ba4

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba4 Bb4 Ba4 Bb4 Ba4 Bb4

really understand 133 73.9% 93.7% 9 60.0% 6.3% 142 72.8% 100.0%

understand 25 13.9% 92.6% 2 13.3% 7.4% 27 13.8% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 11.598 a 6.277 1.230

4 4

Asymp. Sig. (2-sided) .021 .179

1

.267

df

195

a. 6 cells (60.0%) have expected count less than 5. The minimum expected count is .31.

a bit understand 4 2.2% 80.0% 1 6.7% 20.0% 5 2.6% 100.0%

little understa nding 2 1.1% 50.0% 2 13.3% 50.0% 4 2.1% 100.0%

don't know 16 8.9% 94.1% 1 6.7% 5.9% 17 8.7% 100.0%

Total 180 100.0% 92.3% 15 100.0% 7.7% 195 100.0% 100.0%

51

Bar Chart Bb4 really understand understand a bit understand little understanding don't know

125

Count

100

75

50

25

0 correct

incorrect

Ba4

52 Section B(a)5 vs Section B(b)5

Case Proce ssing Summary

N Ba5 * Bb5

Cases Missing N Percent 0 .0%

Valid Percent 195 100.0%

N

Total Percent 195 100.0%

Ba5 * Bb5 Crosstabulation Bb5

Ba5

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba5 Bb5 Ba5 Bb5 Ba5 Bb5

really understand 139 73.2% 97.2% 4 80.0% 2.8% 143 73.3% 100.0%

understand 31 16.3% 96.9% 1 20.0% 3.1% 32 16.4% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value .598 a 1.107 .452

4 4

Asymp. Sig. (2-sided) .963 .893

1

.501

df

195

a. 7 cells (70.0%) have expected count less than 5. The minimum expected count is .05.

a bit understand 2 1.1% 100.0% 0 .0% .0% 2 1.0% 100.0%

little understa nding 4 2.1% 100.0% 0 .0% .0% 4 2.1% 100.0%

don't know 14 7.4% 100.0% 0 .0% .0% 14 7.2% 100.0%

Total 190 100.0% 97.4% 5 100.0% 2.6% 195 100.0% 100.0%

53

Bar Chart Bb5 really understand understand a bit understand little understanding don't know

125

Count

100

75

50

25

0 correct

incorrect

Ba5

54 Section B(a)6 vs Section B(b)6

Case Proce ssing Summary

N Ba6 * Bb6

Cases Missing N Percent 0 .0%

Valid Percent 195 100.0%

N

Total Percent 195 100.0%

Ba6 * Bb6 Crosstabulation Bb6

Ba6

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba6 Bb6 Ba6 Bb6 Ba6 Bb6

really understand 107 56.9% 96.4% 4 57.1% 3.6% 111 56.9% 100.0%

understand 41 21.8% 95.3% 2 28.6% 4.7% 43 22.1% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 1.292 a 2.101 .327

4 4

Asymp. Sig. (2-sided) .863 .717

1

.567

df

195

a. 5 cells (50.0%) have expected count less than 5. The minimum expected count is .43.

a bit understand 16 8.5% 94.1% 1 14.3% 5.9% 17 8.7% 100.0%

little understa nding 12 6.4% 100.0% 0 .0% .0% 12 6.2% 100.0%

don't know 12 6.4% 100.0% 0 .0% .0% 12 6.2% 100.0%

Total 188 100.0% 96.4% 7 100.0% 3.6% 195 100.0% 100.0%

55

Bar Chart Bb6

120

really understand understand a bit understand little understanding don't know

100

Count

80

60

40

20

0 correct

incorrect

Ba6

56 Section B(a)7 vs Section B(b)7

Case Proce ssing Summary

N Ba7 * Bb7

Cases Missing N Percent 0 .0%

Valid Percent 195 100.0%

N

Total Percent 195 100.0%

Ba7 * Bb7 Crosstabulation Bb7

Ba7

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba7 Bb7 Ba7 Bb7 Ba7 Bb7

really understand 93 52.8% 93.9% 6 31.6% 6.1% 99 50.8% 100.0%

understand 41 23.3% 89.1% 5 26.3% 10.9% 46 23.6% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 4.522 a 4.152 3.746

4 4

Asymp. Sig. (2-sided) .340 .386

1

.053

df

195

a. 4 cells (40.0%) have expected count less than 5. The minimum expected count is 1.27.

a bit understand 20 11.4% 87.0% 3 15.8% 13.0% 23 11.8% 100.0%

little understa nding 11 6.3% 78.6% 3 15.8% 21.4% 14 7.2% 100.0%

don't know 11 6.3% 84.6% 2 10.5% 15.4% 13 6.7% 100.0%

Total 176 100.0% 90.3% 19 100.0% 9.7% 195 100.0% 100.0%

57

Bar Chart Bb7

100

really understand understand a bit understand little understanding don't know

Count

80

60

40

20

0 correct

incorrect

Ba7

58 Section B(a)8 vs Section B(b)8

Case Proce ssing Summary

Ba8 * Bb8

Cases Missing N Percent 0 .0%

Valid N Percent 195 100.0%

Total N Percent 195 100.0%

Ba8 * Bb8 Crosstabulation Bb8

Ba8

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba8 Bb8 Ba8 Bb8 Ba8 Bb8

really understand 94 55.0% 94.0% 6 25.0% 6.0% 100 51.3% 100.0%

understand 48 28.1% 81.4% 11 45.8% 18.6% 59 30.3% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 9.035 a 8.916 5.282

4 4

Asymp. Sig. (2-sided) .060 .063

1

.022

df

195

a. 3 cells (30.0%) have expected count less than 5. The minimum expected count is 1.23.

a bit understand 12 7.0% 80.0% 3 12.5% 20.0% 15 7.7% 100.0%

little understa nding 9 5.3% 90.0% 1 4.2% 10.0% 10 5.1% 100.0%

don't know 8 4.7% 72.7% 3 12.5% 27.3% 11 5.6% 100.0%

Total 171 100.0% 87.7% 24 100.0% 12.3% 195 100.0% 100.0%

59

Bar Chart Bb8

100

really understand understand a bit understand little understanding don't know

Count

80

60

40

20

0 correct

incorrect

Ba8

60 Section B(a)9 vs Section B(b)9

Case Proce ssing Summary

N Ba9 * Bb9

Cases Missing N Percent 0 .0%

Valid Percent 195 100.0%

N

Total Percent 195 100.0%

Ba9 * Bb9 Crosstabulation Bb9

Ba9

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba9 Bb9 Ba9 Bb9 Ba9 Bb9

really understand 25 28.1% 40.3% 37 34.9% 59.7% 62 31.8% 100.0%

understand 35 39.3% 50.7% 34 32.1% 49.3% 69 35.4% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 2.090 a 2.098 .079

4 4

Asymp. Sig. (2-sided) .719 .718

1

.778

df

195

a. 2 cells (20.0%) have expected count less than 5. The minimum expected count is 3.65.

a bit understand 17 19.1% 48.6% 18 17.0% 51.4% 35 17.9% 100.0%

little understa nding 8 9.0% 38.1% 13 12.3% 61.9% 21 10.8% 100.0%

don't know 4 4.5% 50.0% 4 3.8% 50.0% 8 4.1% 100.0%

Total 89 100.0% 45.6% 106 100.0% 54.4% 195 100.0% 100.0%

61

Bar Chart Bb9

40

really understand understand a bit understand little understanding don't know

Count

30

20

10

0 correct

incorrect

Ba9

62 Section B(a)10 vs Section B(b)10 Case Proce ssing Summary

N Ba10 * Bb10

Cases Missing N Percent 0 .0%

Valid Percent 195 100.0%

N

Total Percent 195 100.0%

Ba10 * Bb10 Crosstabulation Bb10

Ba10

correct

incorrect

Total

Count % within % within Count % within % within Count % within % within

Ba10 Bb10 Ba10 Bb10 Ba10 Bb10

really understand 96 55.2% 92.3% 8 38.1% 7.7% 104 53.3% 100.0%

understand 49 28.2% 90.7% 5 23.8% 9.3% 54 27.7% 100.0%

Chi-Square Te sts

Pearson Chi-Square Likelihood Ratio Linear-by-Linear Association N of Valid Cases

Value 14.315 a 10.240 8.299

4 4

Asymp. Sig. (2-sided) .006 .037

1

.004

df

195

a. 3 cells (30.0%) have expected count less than 5. The minimum expected count is 1.08.

a bit understand 16 9.2% 94.1% 1 4.8% 5.9% 17 8.7% 100.0%

little understa nding 6 3.4% 60.0% 4 19.0% 40.0% 10 5.1% 100.0%

don't know 7 4.0% 70.0% 3 14.3% 30.0% 10 5.1% 100.0%

Total 174 100.0% 89.2% 21 100.0% 10.8% 195 100.0% 100.0%

63

Bar Chart Bb10

100

really understand understand a bit understand little understanding don't know

Count

80

60

40

20

0 correct

incorrect

Ba10

64 Output SPSS Test: Independent t-test GENDER TEST OF NORMALITY

Case Processing Summary

gender

Cases Valid N

score

male female

Missing Percent

N

Total

Percent

N

Percent

45

100.0%

0

.0%

45

100.0%

150

100.0%

0

.0%

150

100.0%

Descriptives

gender

Statistic

Std. Error

65 score

male

Mean 95% Confidence Interval for Mean

13.8667 Lower Bound Upper Bound

14.5403 14.0802

Median

14.0000 5.027

Std. Deviation

2.24216

Minimum

4.00

Maximum

17.00

Range

13.00

Interquartile Range

female

13.1930

5% Trimmed Mean

Variance

.33424

2.00

Skewness

-2.117

.354

Kurtosis

7.680

.695

14.3533

.18492

Mean 95% Confidence Interval for Mean

Lower Bound Upper Bound

13.9879 14.7187

5% Trimmed Mean

14.5852

Median

15.0000

Variance

5.129

Std. Deviation

2.26481

Minimum

.00

Maximum

17.00

Range

17.00

Interquartile Range

3.00

Skewness

-2.618

.198

Kurtosis

12.599

.394

Tests of Normality

66 gender

Kolmogorov-Smirnov(a) Statistic

score

df

Shapiro-Wilk

Sig.

Statistic

df

Sig.

male

.194

45

.000

.824

45

.000

female

.171

150

.000

.787

150

.000

67

STATISTICAL TEST

Group Statistics

gender score

N

Mean

male female

Std. Deviation

Std. Error Mean

45

13.8667

2.24216

.33424

150

14.3533

2.26481

.18492

Independent Samples Test

Levene's Test for Equality of Variances

t-test for Equality of Means

Sig. (2-tailed) F score

Equal variances assumed Equal variances not assumed

Sig. .000

.996

t

Mean Difference

Std. Error Difference

df

95% Confidence Interval of the Difference Lower

upper

-1.267

193

.207

-.48667

.38407

-1.24418

.27084

-1.274

73.038

.207

-.48667

.38199

-1.24796

.27462

68 AREA TEST OF NORMALITY Case Processing Summary

area

Cases Valid N

score

Missing Percent

N

Total

Percent

N

Percent

urban

122

100.0%

0

.0%

122

100.0%

rural

73

100.0%

0

.0%

73

100.0%

Descriptives

area

Statistic

Std. Error

69 score

urban

Mean 95% Confidence Interval for Mean

14.3115 Lower Bound Upper Bound

14.7146 14.5364

Median

15.0000 5.059

Std. Deviation

2.24927

Minimum

.00

Maximum

17.00

Range

17.00

Interquartile Range

rural

13.9083

5% Trimmed Mean

Variance

.20364

3.00

Skewness

-3.095

.219

Kurtosis

15.932

.435

14.1233

.26884

Mean 95% Confidence Interval for Mean

Lower Bound Upper Bound

13.5874 14.6592

5% Trimmed Mean

14.3242

Median

14.0000

Variance

5.276

Std. Deviation

2.29701

Minimum

4.00

Maximum

17.00

Range

13.00

Interquartile Range

3.00

Skewness Kurtosis

Tests of Normality

-1.498

.281

4.348

.555

70

area

Kolmogorov-Smirnov(a) Statistic

score

df

Shapiro-Wilk

Sig.

Statistic

df

Sig.

urban

.191

122

.000

.739

122

.000

rural

.162

73

.000

.878

73

.000

71

STATISTICAL TEST

Group Statistic

area score

N

Mean

Std. Deviation

Std. Error Mean

urban

122

14.3115

2.24927

.20364

rural

73

14.1233

2.29701

.26884

Independent Samples Test

Levene's Test for Equality of Variances

t-test for Equality of Means

Sig. (2-tailed) F score

Equal variances assumed Equal variances not assumed

Sig. .919

.339

t

Mean Difference

Std. Error Difference

df

95% Confidence Interval of the Difference lower

upper

.561

193

.575

.18819

.33548

-.47349

.84986

.558

149.113

.578

.18819

.33726

-.47824

.85462

72 LICENSE TEST OF NORMALITY Case Processing Summary

lisence

Cases Valid N

score

Missing Percent

N

Total

Percent

N

Percent

Yes

134

100.0%

0

.0%

134

100.0%

No

61

100.0%

0

.0%

61

100.0%

Descriptives

lisence

Statistic

Std. Error

73 score

yes

Mean 95% Confidence Interval for Mean

14.5746 Lower Bound Upper Bound

14.8697 14.6824

Median

15.0000 2.983

Std. Deviation

1.72717

Minimum

4.00

Maximum

17.00

Range

13.00

Interquartile Range

no

14.2795

5% Trimmed Mean

Variance

.14920

2.00

Skewness

-1.968

.209

Kurtosis

9.834

.416

13.5082

.38739

Mean 95% Confidence Interval for Mean

Lower Bound Upper Bound

12.7333 14.2831

5% Trimmed Mean

13.8197

Median

14.0000

Variance

9.154

Std. Deviation

3.02557

Minimum

.00

Maximum

17.00

Range

17.00

Interquartile Range

3.00

Skewness

-2.060

.306

Kurtosis

6.669

.604

Tests of Normality

74 lisence

Kolmogorov-Smirnov(a) Statistic

score

df

Shapiro-Wilk Sig.

Statistic

df

Sig.

yes

.164

134

.000

.845

134

.000

no

.171

61

.000

.821

61

.000

75

STATISTICAL TEST. Group Statistics

lisence score

N

Mean

Std. Deviation

Std. Error Mean

Yes

134

14.5746

1.72717

.14920

No

61

13.5082

3.02557

.38739

Independent Samples Test Levene's Test for Equality of Variances F

Sig.

t-test for Equality of Means t

df

Sig. (2-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference Lower

score

Upper

Equal variances assumed 12.214

Equal variances not assumed

.001

3.119

193

.002

1.06643

.34195

.39198

1.74088

2.569

78.344

.012

1.06643

.41513

.24004

1.89282

76

Output SPSS Test: Chi Square test GENDER Case Proce ssing Summary

N gender * sources3

Cases Missing N Percent 29 14.9%

Valid Percent 166 85.1%

N

Total Percent 195 100.0%

ge nde r * sources3 Crosstabulation

gender

male

female

Total

Count Expected Count % within gender % within sources3 % of Total Count Expected Count % within gender % within sources3 % of Total Count Expected Count % within gender % within sources3 % of Total

sources3 non-driving driving school school 25 14 27.0 12.0 64.1% 35.9% 21.7% 27.5% 15.1% 8.4% 90 37 88.0 39.0 70.9% 29.1% 78.3% 72.5% 54.2% 22.3% 115 51 115.0 51.0 69.3% 30.7% 100.0% 100.0% 69.3% 30.7%

Total 39 39.0 100.0% 23.5% 23.5% 127 127.0 100.0% 76.5% 76.5% 166 166.0 100.0% 100.0% 100.0%

Chi-Square Te sts

Pearson Chi-Square a Continuity Correction Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases

Value .641 b .363 .629 .637

df 1 1 1 1

Asymp. Sig. (2-sided) .423 .547 .428

Exact Sig. (2-sided)

Exact Sig. (1-sided)

.433

.271

.425

166

a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 11. 98.

77

Symme tric M e asure s

Nominal by Nominal

Value -.062 .062 166

Phi Cramer's V

N of Valid Cases

Approx. Sig. .423 .423

a. Not assuming the null hypothesis. b. Using the asymptotic standard error assuming the null hypothesis.

Bar Chart sources3

100

driving school non-driving school

Count

80

60

40

20

0 male

female

gender

78

AREA Case Proce ssing Summary Cases Missing N Percent 29 14.9%

Valid N Percent 166 85.1%

area * sources3

Total N Percent 195 100.0%

are a * source s3 Crosstabulation

area

urban

rural

Total

Count Expected Count % within area % within sources3 % of Total Count Expected Count % within area % within sources3 % of Total Count Expected Count % within area % within sources3 % of Total

sources3 non-driving driving school school 77 27 72.0 32.0 74.0% 26.0% 67.0% 52.9% 46.4% 16.3% 38 24 43.0 19.0 61.3% 38.7% 33.0% 47.1% 22.9% 14.5% 115 51 115.0 51.0 69.3% 30.7% 100.0% 100.0% 69.3% 30.7%

Total 104 104.0 100.0% 62.7% 62.7% 62 62.0 100.0% 37.3% 37.3% 166 166.0 100.0% 100.0% 100.0%

Chi-Square Te sts

Pearson Chi-Square a Continuity Correction Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases

Value 2.966 b 2.397 2.926 2.948

df 1 1 1 1

Asymp. Sig. (2-sided) .085 .122 .087

Exact Sig. (2-sided)

Exact Sig. (1-sided)

.117

.061

.086

166

a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 19. 05.

79

Symme tric M e asure s

Nominal by Nominal

Phi Cramer's V

N of Valid Cases

Value .134 .134 166

Approx. Sig. .085 .085

a. Not assuming the null hypothesis. b. Using the asymptotic standard error assuming the null hypothesis.

Bar Chart sources3

80

driving school non-driving school

Count

60

40

20

0 urban

rural

area

80

LICENSE

Case Proce ssing Summary Cases Missing N Percent 29 14.9%

Valid N Percent 166 85.1%

lisence * sources3

Total N Percent 195 100.0%

lise nce * source s3 Crosstabulation

lisence

yes

no

Total

Count Expected Count % within lisence % within sources3 % of Total Count Expected Count % within lisence % within sources3 % of Total Count Expected Count % within lisence % within sources3 % of Total

sources3 non-driving driving school school 99 18 81.1 35.9 84.6% 15.4% 86.1% 35.3% 59.6% 10.8% 16 33 33.9 15.1 32.7% 67.3% 13.9% 64.7% 9.6% 19.9% 115 51 115.0 51.0 69.3% 30.7% 100.0% 100.0% 69.3% 30.7%

Total 117 117.0 100.0% 70.5% 70.5% 49 49.0 100.0% 29.5% 29.5% 166 166.0 100.0% 100.0% 100.0%

Chi-Square Te sts

Pearson Chi-Square a Continuity Correction Likelihood Ratio Fisher's Exact Test Linear-by-Linear Association N of Valid Cases

Value 43.813b 41.405 42.432 43.549

df 1 1 1 1

Asymp. Sig. (2-sided) .000 .000 .000

Exact Sig. (2-sided)

Exact Sig. (1-sided)

.000

.000

.000

166

a. Computed only for a 2x2 table b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 15. 05.

81

Symme tric M e asure s

Nominal by Nominal

Phi Cramer's V

N of Valid Cases

Value .514 .514 166

Approx. Sig. .000 .000

a. Not assuming the null hypothesis. b. Using the asymptotic standard error assuming the null hypothesis.

Bar Chart sources3

100

driving school non-driving school

Count

80

60

40

20

0 yes

no

lisence

82

Output SPSS Test: Binary logistic regression LOGISTIC REGRESSION De pe ndent Variable Encoding Original Value Internal Value 1.00 0 2.00 1

Block 0: Beginning Block Classification Tablea,b Predicted skor2 Step 0

Observed skor2

1.00 1.00 2.00

2.00 0 0

8 187

Overall Percentage

Percentage Correct .0 100.0 95.9

a. Constant is included in the model. b. The cut value is .500

Variable s in the Equation Step 0

Constant

B 3.152

S.E. .361

Wald 76.204

df 1

Sig. .000

1 1 1 3

Sig. .895 .453 .006 .049

Variable s not in the Equation Step 0

Variables

Overall Statistics

gender area lisence

Score .017 .562 7.417 7.868

df

Exp(B) 23.375

83

Block 1: Method = Enter

Omnibus Te sts of M ode l Coe fficie nts Step 1

Chi-square 7.291 7.291 7.291

Step Block Model

df 3 3 3

Sig. .063 .063 .063

M ode l Summary Step 1

-2 Log Cox & Snell likelihood R Square 59.473 a .037

Nagelkerke R Square .127

a. Estimation terminated at iteration number 7 because parameter estimates changed by less than .001.

Classification Tablea Predicted skor2 Step 1

Observed skor2

1.00 1.00 2.00

2.00 0 0

8 187

Overall Percentage

Percentage Correct .0 100.0 95.9

a. The cut value is .500

Variable s in the Equation Step a 1

gender area lisence Constant

B .658 -.088 -2.091 5.313

S.E. .893 .758 .885 2.120

Wald .542 .013 5.580 6.280

a. Variable(s) entered on step 1: gender, area, lisence.

df 1 1 1 1

Sig. .461 .908 .018 .012

Exp(B) 1.931 .916 .124 202.868

84