Built Environment and Crimes in Residential Areas

I am struggling with my research paper which I need to publish However, I have got comments from reviewers should i work on in fact, the notes were focusing on statistical issues related to research method and data analysis which i have lack on and need a help in this regard almost 25 with references pages the academic field is urban planning i utilised GIS and regression analysis  I need you to have mathematical and statistical services by which the research method and data analysis sections can be improved the subject is titled by ” built environment and crimes in residential areas By the way, the paper is extracted from my Ph.D thesis as you know the publishing is a requirement for getting the Degree the comments on the paper are summarised as follows: a. Novelty of the content: Partially met b. Potential impact of the manuscript in the relevant field of research: Partially met c. Standard of English (expression, grammar and spelling): Poor, which made it difficult to assess. d. Suitability for ISPRS International Journal of Geo-Information: Suitable e. Study design and methodology: Partially met. f. Ethical issues: Met. This study is mainly of interest as it presents findings about the relationship between built environment features and crime in a part of the world that is under-represented in the literature. The methods are not novel. It might be better suited to a general issue. A concern is that the research questions are explicitly framed as causal but the regression method used cannot answer causal, only association, questions. Thus either the research questions or the method are inappropriate and would need to be revised. The accuracy and precision of the crime locations and their geocoding process is not described. A large spatial scale is used but potential issues with larger scale (MAUP) are not explicitly addressed, nor does it appear that spatial autocorrelation was managed in the regression model (e.g. through a spatial lag term). It is not clear how the Getis-Ord hotspot analysis that identifies concentrations of crime relates to the regression analysis where the outcome variable is a different measure of crime density (n crimes / neighbourhood area). The results described in the text are not consistent with the tables (e.g. claiming there are relationships when the p-values are not under 0.05 in the full model). The paper would benefit from the use of an English language editing service. I have the following additional minor concerns: – the abstract exceeds the stated maximum for the journal. – the citations are not in a proper format and contain errors e.g. Brantingham 1981 instead of Brantingham & Brantingham 1981. – many statements about the findings of prior studies are not followed with citations. – contradictions e.g. p3 line 85 “Newman has excluded micro scales” vs p12 line 291 “Newman was focusing on micro scales” and other references to Newman. – the maps appear to have been stretched, violating basic cartographic principles – either figure 1 or table 1 is redundant. – the authors have not specified in the cover letter which of the special issue topics the paper relates to, although the keywords do include the special issue topic ‘hot spot analysis’ and similar topics (‘built environment attributes’). Initial review 2: Statistical analysis is very weak, authors does not present enough statistical proofs to support the use of the variables that they employed on their analysis. Particularly, Linear Regression are not the most useful statistical tool for this kind of analysis. Spatial analysis is not concluding, I mean, authors presented concentration conclusions with a huge bandwidth, obviously, crimes will be “concentrated” when the bandwidth is 4 to 6 kms. Theoretically, there are a lot of opportunities to improve the analysis. I mean, authors mix crime science theories with urbanism and it is not very clear The main idea behind this paper is very interesting but the authors need to improve their statistical and spatial analysis to support the theoretical ideas that they are exposing. Personally, I do not find this paper good enough to be published on this Journal (or any other), at leat, not ye