Evaluating fear of crime using the structural equation model / Siti Rasidah Md Sakip … [et at.]

Fear of crime is based on three preferences which are crime-specific, crime problems in neighbourhood and environmental factors such as physical disorder, social disorder and victimization. Most findings by the researcher found that those dimensions show a high level of reliability to measure the fear of crime. Therefore, in this paper the dimensions of fear of crime were tested using confirmatory factor analysis with a series of measurement model. The validation and confirmation of the fear of crime construct was done using Confirmatory Factor Analysis (CFA) via AMOS. 19 items were initially involved in measuring the three fears of crime dimensions, but 6 items were excluded from the list of variable indicators of the fear of crime
dimensions because these items have a factor loading below 0.3. The results of this study indicate that the crime problems in neighbourhood (CPN) and environmental factors (EF) dimensions achieved good fit indices where the values for GFI, TLI and CFI exceeded 0.90 and the RMSEA value was
less than 0.05. The CPN dimension on the other hand, was found to be the best indicator to measure fear of crime in neighbourhoods with the value of standardized coefficients at r=0.91.