*4.1. Pedestrian Profile*

The respondents' background information reflects the commercial and business status of Kuala Lumpur and the variety of justifications or attractions for residents to reside in and for tourists to visit the city (Table 5).


**Table 5.** Respondents profile (source: authors).

Note: RM represents Ringgit Malaysia, KL represents Kuala Lumpur, and CBD represents central business district.

The male and female genders are almost equally represented at 47.3% and 52.7%, respectively. The 20–29 years old age group has the highest number of respondents (198 or 49.5%). For monthly income, the group with the highest number of respondents is the RM2001–RM3000 income group (140 or 35%). Most of the respondents hold at least a first degree (56.8%). The three most cited reasons for visiting Kuala Lumpur CBD are for work/business purpose (40%), followed by entertainment/recreation (27%), and shopping (17.5%).

### *4.2. Perception of Fear and Street Crime Level*

The Safe City Program and its initiatives can be divided into two major categories, namely CPTED and CPSD. In correlation analysis, the authors found that CPTED had a stronger relationship with reducing street crime (0.592) compared to CPSD (0.562). CPTED and CPSD had the same level of moderate relationship with reducing fear of crime, with both obtaining a Pearson's correlation value of 0.628.

In detail, the results show that each of the independent variables had a significant relationship (*p* < 0.01) with the dependent variables, which are reducing crime and reducing the fear of crime. The three CPTED variables that had the strongest relationship with reducing street crime are landscaping (0.690), the appearance of the building, street, and city (0.686), and generate activities (0.675). Meanwhile, the three CPSD variables that had the strongest relationship with reducing street crime are full council meeting (0.654), watch group (0.653), and city status website (0.62). Overall, most of the relationships can be considered as moderately linear with the *r* values falling between 0.498 and 0.690.

Next, the three CPTED variables that had the strongest relationship with reducing the fear of crime are safety mirror (0.698), the appearance of the building, street, and city (0.692), and generate activities (0.679). Meanwhile, the three CPSD variables of city status website (0.673), watch group (0.656), and victimization/safety survey (0.653) had the strongest relationship with reducing the fear of crime. The range of *r* values for reducing the fear of crime (0.522–0.698) is slightly narrower than that for reducing street crime (0.498–0.690).

### *4.3. E*ff*ect of CPTED and CPSD*

In regression analysis, the R values of 0.630 and 0.638 for reducing street crime and reducing the fear of street crime, respectively, indicate a moderate level of prediction. The coefficient of determination (R2) is the proportion of variance in the dependent variable that can be explained by the independent variables. The R<sup>2</sup> value of 0.396 for reducing street crime indicates that the set of independent variables can explain only 39.6% of the variability in the dependent variable. Similarly, the R<sup>2</sup> value of 0.408 for reducing the fear of crime shows that the set of independent variables can explain only 40.8% of the variability in the dependent variable.

As for the results for the statistical significance of the regression models, the independent variables significantly predicted the dependent variables of reducing street crime (F(27,372) = 9.047, *p* < 0.0005) and reducing the fear of crime (F(27,372) = 9.483, *p* < 0.0005). Both regression models are a good fit for the data. However, the statistically significant level for the coefficient of each independent variable needs to be referred to.

Table 6 shows the estimated multiple regression model. Only three Safe City initiatives as the independent variables made a good prediction of Reducing Street Crime as the dependent variable, which can be statistically significantly predicted as:

"*Reducing Street Crime," F (27, 372)* = *1.124* + *(0.109* × *access control)* − *(0.091* × *full council meeting)* + *(0.021* × *city status website), with p* < *0.05*.


**Table 6.** Multiple regression results (source: authors).

Note: \* significant level < 0.05. CCTV stands for closed-circuit television, CPTED stands for crime prevention through environmental design, CPSD stands for crime prevention methods through social development, and GIS stands for geographic information system.

Meanwhile, notably, none of the independent variables could make a good prediction of Reducing the Fear of Crime at a significant level > 0.05. Therefore, no regression formula was formed to predict the outcome of Reducing the Fear of Crime.
