Residents’s Subjective Assessment of Walkability Attributes in Objectively Assessed Neighbourhoods

The attributes of the built environment may influence walking in neighbourhood. The objective of this study is to find the association between objectively assessed and perceived built environment attributes. To achieve the objective of the study three neighbourhood of Putrajaya, Malaysia Precinct 8, 9 and 18 were selected. This study used NEWS (Neighbourhood Environment Walkability Scale) abbreviated version for the comparison of the residents’ perceptions regarding attributes of three neighbourhoods of Putrajaya in Malaysia that differ on objective measures derived from a GIS (Geographical Information System) data basis and Putrajaya master plan. High-walkable neighbourhood residents reported built environment attributes ratings persistently higher as compared to the residents of the moderate and lowwalkable neighbourhood. Results from the high walkable neighborhood shows that on a 5-point Likert scale most residents perceived high ratings of within the range of “3” except for residential density as there is a different scoring procedure for it. There are lowest ratings perceived within the range of “1” for the attribute crime and according to the scoring procedure for crime lowest score reflects the highest level of walking.There is no significant difference found in the ratings for traffic hazards. Subjective measures of built environment attributes had moderate to high alpha () value. Hence theneighbourhood environment walkability scale can be used for subjective assessment in the tropical context.


INTRODUCTION
T he importance of the influence of the built environment attributes on walking is creating more and more interest with the passage of time among the researchers [1][2][3]. Several studies showed that people who live near to destinations such as parks or commercial areas have easy access to them to that, it has been a focus of considerable research.
Neighbourhood's built environment attributes may encourage walking for the various reasons like walking for transport, leisure or exercise. The subjective assessment of neighbourhood built environment attributes influencing walkability has been done by several methods yet there is limited study found on how these subjective assessments could have been related to corresponding attributes assessment based on objectives measures [7].
During the last few decades the two planning groups actively conducting research on walking-related topics are transportation planners and urban designer. The transportation planner group is interested in objective assessment of the built environment and its association to higher rates of walking trips for transportation [8]. The urban designers on the other hand, are more interested in the subjective assessment of built environment attributes and its influence on walking [9]. For the conduct of common beneficial research those two groups could easily come together to test the effect of the walking environment on walking travel behavior. Therefore, this paper presents comprehensive approaches for assessing walkability.

METHODOLOGY
In

Subjects
A two-stage cluster sampling was done to select 402 children and adults aged between seven to sixty-five.

Procedures
Sample of population was selected from low walkable Precinct 8, moderate walkable Precinct 9, and highly walkable Precinct 18. Selected neighbourhoods were considered as high, moderate and low walkable neighbourhood based on their walkability index. All the data on land use and each precinct specific data to be use for the indices were obtained from the Putrajaya [ GIS Unit Perbadanan Putrajaya (Fig. 1). According to the data obtained it was found that Precinct 18 had higher residential density higher diversity of land use mix, higher distances to destinations and higher street connectivity.
Whereas it was found from the data obtained from the report that Precinct 9 is medium dense, medium diversity in terms of its land use mixes, moderate distance to destinations but lower street connectivity, it was also found that Precinct 8 has a low residential density, low diversity in terms of its mix uses, low distance to destinations and low street connectivity.
High walkable Precinct 18 has a high gross residential density mainly due to residential type there were apartment of up to fifteen stories and high diversity in terms of land uses as these 12 storey apartments has different kinds of shops as discussed in the master plan of Putrajaya [11]. Street connectivity is also found higher in this precinct compare to the other 2 selected neighbourhood for studies. Therefore, it is found in this precinct that there are shorter distances to different facilities like school, mosque, food courts etc. Almost all or most of the facilities are required to be falling within four hundred meters by walk as this precinct is dominated  by high-rise residential blocks and also multi-unit residential blocks of 4-6 stories are also found in this precinct in the form of terrace houses and low rise apartment. This precinct is characterized by diverse topography. Numerous public facilities are provided such as schools, schools for disable, mosque, market place, library and health care center [12][13][14]. refers to highly agree. Residential density items inquired about the types of dwellings, from detached single-family residents to more than 13-stories apartments or condominium, and the responses were ranging from 1 refers to none to 5 which refers to all. Residential density items were weighted relative to the average density of detached houses (e.g. Link houses were considered "12" times more dense in comparison with the detached houses. Apartment of up to 4 to 6 stories are rated to have "25" times more dense in comparison with the detached houses, on the other hand apartment of up to 7-12-levels were evaluated to be "50"times more dense in comparison with the detached apartments and lastly apartments of 13 levels and beyond are evaluated to be "75" times more dense in comparison with the detached houses) and conclusively all the weighted values were added to formulate a score consisting of density for residential subscale. These numbers (12,25,50,75) are weighted for calculating the perceived residential density score what you get is a relative score reflecting (perceived) residential density, but not actual residential density. Whereas a diversity in terms of its land use mix was evaluated with the walking proximity from residences to different kind of facilities and business, the answers for the questionnaire varied from one to five minutes walking distance (coded as 1) whereas thirty minutes walking distance (coded as 4). If higher scores are obtained on diversity of mix uses that means business or other facilities are in close proximity [15].

RESULTS AND DISCUSSION
Data were coded, entered and analyzed using SPSSv20.0 to compare mean subscale scores between occupants of the three different neighbourhoods. Individual reliabilities alpha values were also checked for each item. Mean subscale score of high, middle and low walkable neighbourhood are given in Table 3.