A pilot trial is conducted as a preliminary trial against the informations aggregation to place the jobs and avoid serious errors before the beginning of existent informations aggregation from our mark respondents. A pilot trial of 20 samples of lectors in UTAR Perak Campus is conducted in order to prove the dependability of the questionnaires set. Table 4.1 shows the dependability consequence of our pilot trial conducted. Harmonizing to Hair, Tatham, Anderson & A ; Black ( 1998 ) , the criterion lower edge for Cronbach ‘s alpha is 0.70. By and large, the dependability coefficients of all IVs and DV are above 0.70 and considered dependable. Therefore, there is no demand to amend the points of variables in questionnaires.

## 4.2 Descriptive Analysis

## 4.2.1 Demographic Profile of the Respondents

Table 4.2: Demographic Profile of the Respondents

## Frequency

## Percentage

## Gender

Male

Female

130

120

52.0

48.0

## Age

20-23

24-26

27-29

30 and above

114

44

25

67

45.6

17.6

10.0

26.8

## Marital Status

Single

Married

166

84

66.4

33.6

## Income Level

& lt ; RM1,000

RM1,000-RM1,999

RM2,000-RM2,999

RM3,000-RM3,999

RM4,000 and above

91

52

47

33

27

36.4

20.8

18.8

13.2

10.8

## Education Degree

SPM / STPM

Diploma

Degree

Others

44

96

98

12

17.6

38.4

39.2

4.8

Beginning: Developed for the research

Table 4.2 shows the demographic profile of the respondents. The tabular array includes gender, age, matrimonial position, income degree, and instruction degree. The entire sample is made up of 250 respondents. Male respondents are more than females which is 52 % for male and 48 % for female. There are 66.4 % of respondents are in individual position and 33.6 % of them are married. The consequences show that the age scope of respondents is chiefly within 20-23 old ages old, which consists of 45.6 % , followed by 30 old ages old and above, which is 26.8 % . Meanwhile, the scope of age within 24-26 old ages old and 27-29 old ages old are merely made up to 17.6 % and 10.0 % severally. For the income degree, most of the respondents autumn in the group of less than RM1, 000 for gross monthly income, which is 36.4 % , followed by 20.8 % , which is within RM1, 000-RM1, 999, after that are 18.8 % for RM2, 000-RM2, 999, 13.2 % for RM3, 000-RM3, 999, and 10.8 % for RM4, 000 and supra. For the instruction degree, most of the respondents are degree and sheepskin holders, which are 39.2 % and 38.4 % severally. While, 17.8 % of the respondents with SPM or STPM degree, and merely 4.8 % of the respondents fall under others such as with maestro grade and PHD degree degree.

## 4.2.2 Cardinal Tendencies Measurement of Constructs

Table 4.3: Mean and Standard Deviation of Questionnaire Items

## Items

## Mean

## Standard Deviation

TK 1

TK 2

TK 3

TK 4

TK 5

3.8200

3.7600

4.0400

4.0200

3.8600

0.95459

0.88631

1.07845

1.03143

1.07938

SN 1

SN 2

SN 3

SN 4

SN 5

3.5800

3.6200

3.1400

3.2000

3.4400

1.02362

1.03919

1.07938

1.04132

1.06345

LS 1

LS 2

LS 3

LS 4

LS 5

2.6800

3.5800

3.7400

3.2600

3.1000

1.08735

0.82846

1.01811

0.93590

1.06515

POF 1

POF 2

POF 3

POF 4

POF 5

3.4600

3.0800

3.4600

3.1400

3.1200

1.17213

0.91485

1.08309

1.04151

1.03435

TCI 1

TCI 2

TCI 3

TCI 4

TCI 5

3.3600

2.8400

3.4000

3.1600

3.7400

1.05587

0.98910

1.07919

1.14000

0.97787

Beginning: Developed for research

If the appraisal of standard divergence is less than 1, it means that the respondents of a study have the same sentiment or thought. While if the standard divergence is more than 3, it means the respondents have different sentiments. ( Hair, Babin, Money & A ; Samuel, 2003 ) .

For TK, all TK1, TK2, TK3, TK4, and TK5 are all fall more to hold class, which the mean values are 3.8200, 3.7600, 4.0400, 4.0200, and 3.8600 severally. Meanwhile, merely the standard divergences of TK1 and TK2 are less than 1, which are 0.95459 and 0.88631 severally. For TK3, TK4, and TK5, the standard divergences are greater than 1.

For SN, merely SN1 and SN2 are falls more to hold class as their mean values are 3.5800 and 3.6200 severally. Meanwhile, SN3, SN4, and SN5 are falls near to impersonal class, where their mean values are 3.1400, 3.2000, and 3.4400 severally. Their standard divergences are 1.02362, 1.03919, 1.07938, 1.04132, and 1.06345 severally, which all are more than 1.

In LS, merely LS2 and LS3 are falls near to hold class, as their mean values are 3.5800 and 3.7400 severally. For LS1, LS4, and LS5, all are falls in impersonal class. Their average values are 2.6800, 3.2600 and 3.1000 severally. Meanwhile, the standard divergences for LS1, LS3, and LS5 are all more than 1, merely exclude for LS2 and LS4 which are 0.82846 and 0.93590 severally for their standard divergences.

For POF, the mean values for POF1, POF2, POF3, POF4, and POF5 are 3.4600, 3.0800, 3.4600, 3.1400, and 3.1200 severally where all of them are fall in impersonal class. All of their standard divergences are greater than 1, merely except for POF2 where its standard divergence is 0.91485.

Last for TCI, merely TCI5 is falls in agree class, where its mean value is 3.7400. For TCI1, TCI2, TCI3, and TCI4 are all falls more to impersonal class, where their mean values are 3.3600, 2.8400, 3.4000, and 3.1600 severally. All of their standard divergences are more than 1 but except for TCI2 and TCI5, where their standard divergences are 0.98910 and 0.97787 severally.

## 4.3 Scale Measurement

## 4.3.1 Reliability Test

Table 4.4: Dependability Statisticss

Variables

Concepts

Cronbach ‘s Alpha

N of Items

IV 1

Tax Knowledge

0.731

5

IV 2

Social Norms

0.787

5

IV 3

Legal Sanctions

0.725

5

IV 4

Percept of Fairness

0.730

5

DV

Tax Compliance Intention

0.764

5

Beginning: Developed for the research

Dependability of the factors was calculated by composite dependability. The composite dependability was evaluated utilizing internal consistence of the measuring theoretical account ( Jayasingh and Eze, 2009 ) . Furthermore, the dependability analysis was carried out by utilizing Cronbach ; s alpha, which is a step of internal consistence ( Teo & A ; Pok, 2003 ) . Table 4.4 illustrates the Cronbach ‘s alpha for every IV and DV. The Cronbach ‘s alpha coefficients ranged from 0.725 to 0.787 bespeaking that all variables are above the criterion lower edge for Cronbach ‘s alpha which is 0.70 ( Hair et al. , 1998 ) . Therefore, it can be assumed that the study questionnaire distributed for measuring the factors impacting revenue enhancement conformity purpose of white neckbands in this survey is dependable.

## 4.3.2 Normality Assumption

Table 4.5: Normality Trial

## Trials of Normality

Kolmogorov-Smirnova

Shapiro-Wilk

Statistic

df

Sig.

Statistic

df

Sig.

Standardized Predicted Value

.055

250

.060

.988

250

.035

a. Lilliefors Significance Correction

Beginning: Developed for the research

Normality trial consists of Kolmogorov-Smirnov trial and Shapiro-Wilk trial. As the sample size for this research is 250, which is more than 50, the trial of informations normalcy among the independent variables was carried out utilizing Kolmogorov-Smirnov ‘s trial ( Chong, Ooi, $ Sohal, 2009 ; Hair, Black, Babin, Anderson & A ; Tatham, 2005 ) . In order to analyze the normalcy of dependent variable, the Kolmogorov-Smirnov statistic with Lilliefors significance rectification for measuring normalcy is used ( Chong et al. , 2009 ) . Harmonizing to Fatoko et Al. ( 2011 ) , normalcy of the informations can be assumed when the consequence of P & gt ; 0.05. Table 4.5 depicts the significance of Kolmogorov-Smirnov ‘s trial ( 0.060 ) is greater than 0.05 which implies that normalcy can be assumed.

## 4.4 Inferential Analysis

## 4.4.1 Pearson Correlation Analysis

Table 4.6: Correlations between Variables

## TK

## Tin

## Liter

## POF

## TCI

## TK

1

## Tin

0.019

1

## Liter

-0.170**

0.676**

1

## POF

0.002

0.133*

0.026

1

## TCI

0.172**

0.336**

0.430**

0.509**

1

** . Correlation is important at the 0.01 degree ( 2-tailed ) .

* . Correlation is important at the 0.05 degree ( 2-tailed ) .

Beginning: Developed for the research

Pearson correlativity analysis was conducted to analyze the relationship between the variables ( Wong et al. , 2005 ; Jahangir & A ; Begum, 2008 ) . In order to avoid multicollinearity job, correlativity coefficient should non travel beyong 0.8 ( Field, 2005 ) . In Table 4.6, all the correlativity coefficients are less than 0.8, where the highest is 0.676 ( correlativity between LS and SN ) , we assumed that there is no multicollinearity job among all the variables in this research.

Most of the affiliated braces of variables are important at the 0.01 degree, which include LS and TK, LS and SN, TCI and TK, TCI and SN, TCI and LS, and TCI and POF. Meanwhile, merely a brace of variable which is POF and SN is important at the 0.05 degree. For the remainder, SN and TK, POF and TK, and POF and LS are non important at all. This analysis consequence implies that TK ( r = 0.172, P & lt ; 0.01 ) , SN ( r = 0.336, P & lt ; 0.01 ) , LS ( r = 0.430, P & lt ; 0.01 ) , and POF ( r = 0.509, P & lt ; 0.01 ) are all positively and significantly correlated with TCI.

Based on Table 4.6, the strongest correlativity between IVs and TCI ( DV ) is between POF and TCI ( r = 0.509, P & lt ; 0.01 ) . This is followed by the correlativity between LS and TCI ( r = 0.430, P & lt ; 0.01 ) , following is the correlativity between SN and TCI ( r = 0.336, P & lt ; 0.01 ) . Last but non least, it is followed by the correlativity between TK and TCI ( r = 0.172, P & lt ; 0.01 ) .

## 4.4.2 Multiple Regression Analysis

Table 4.7: Consequence of Model Summary

## Model Summaryb

Model

Roentgen

R Square

Adjusted R Square

Std. Mistake of the Estimate

1

.705a

.497

.489

.48043

a. Forecasters: ( Constant ) , LS_AVERAGE, POF_AVERAGE, TK_AVERAGE, SN_AVERAGE

B. Dependent Variable: TCI_AVERAGE

The R2 value is 49.7 % which means that the 4 IVs in the theoretical account have 49.7 % influence on TCI of single taxpayers in Malaysia. It is besides means that about 50 % of the single taxpayers in Malaysia have TCI could be significantly explained by the 4 IVs which are TK, SN, LS, and POF.

Table 4.8: ANOVA Table

## ANOVAb

Model

Sum of Squares

df

Mean Square

F

Sig.

1

Arrested development

55.950

4

13.988

60.600

.000a

Residual

56.550

245

.231

Entire

112.500

249

a. Forecasters: ( Constant ) , POF, TK, SN, LS

B. Dependent Variable: TCI

F-statistic is valued at 60.600 and important at the degree of 0.000. This implies that all the IVs are dependable forecasters of TCI of single taxpayers in Malaysia.

Table 4.9: Multiple Linear Regression for Independent Variables and Dependent Variable

## Coefficientsa

Model

Unstandardized Coefficients

Standardized Coefficients

T

Sig.

Bacillus

Std. Mistake

Beta

1

( Constant )

-.590

.281

-2.103

.037

TK

.254

.046

.261

5.583

.000

Tin

-.086

.061

-.090

-1.420

.157

Liter

.498

.061

.522

8.193

.000

POF

.481

.044

.507

11.041

.000

a. Dependent Variable: TCI

Table 4.9 shows that TK ( P & lt ; 0.001 ) , LS ( P & lt ; 0.001 ) , and POF ( P & lt ; 0.001 ) are significantly affect the TCI of single taxpayers in Malaysia. The strongest determiner is LS which has 52.2 % . Meanwhile, SN ( p = 0.157 ) was found to has no significance linked with TCI of single taxpayers in Malaysia.

The equation of the theoretical account in this survey can be written as:

TCI = -0.590 + 0.254 TK – 0.086 SN + 0.498 LS + 0.481 POF

## 4.5 Decision

In decision, most of our respondents were aged between 20-23 old ages old, male and in individual position. We found out that TK, LS, and POF were significantly in foretelling TCI while SN was non. The information collected was normal in distribution and there was no multicollinearity job exists between the variables.