Comprehensive Analysis of London Crime Dataset: Trends and Patterns

This report presents an analysis of the London crime dataset spanning from January 2008 to December 2016. The research aims to determine the relationship between the spatial accuracy of the data and the types of crimes, with the goal of identifying areas prone to crime for easier detection. The report employs a positivism research philosophy, using descriptive statistics, data visualization, correlation analysis, and significance testing, including linear regression with SPSS. Key findings include the analysis of major and minor category crimes, district-wise crime comparisons, and correlation of crime values over the years. The discussion evaluates the model's statistical significance and its implications, concluding that the created model is not statistically significant. The report includes a detailed methodology, results, and discussion, along with relevant references.

University
Team Research and Development
Student Name:
Register ID:

Table of Contents
1. Introduction............................................................................................................................1
2. Background.............................................................................................................................1
2.1 Research Questions.........................................................................................................1
2.2 Peer Evaluation...............................................................................................................2
3. Method.....................................................................................................................................2
4. Results......................................................................................................................................3
4.1 Project Planning & Management..................................................................................3
4.2 Descriptive Statistics.......................................................................................................4
4.3 Data Visualization...........................................................................................................5
4.4 Correlation.......................................................................................................................8
4.5 Significance......................................................................................................................8
5. Discussion................................................................................................................................9
6. Conclusion.............................................................................................................................10
References.....................................................................................................................................11

1. Introduction
In London, the number of crimes have increased which are categorized in major and
minor category crimes. This objective of this research is to determine the relationship between
the spatial accuracy of London crime dataset, with the type of crimes, to find which place crimes
can occur for easy detection. The London crime dataset from January 2008 to December 2016
will be analyzed.
2. Background
Like in various cities, the cities of London also have a hard time with many crimes taking
place. The police is finding difficulty to take measures to stop the threats of crime. Moreover, as
London has many districts and it is highly complicated to spot the right place. In this case, the
police must consider the recorded crime statistics to take necessary steps stop the local crimes.
Only this type of research can ensure to help the police to figure out the extreme crime places, so
as to be alert and stop the crimes locally. The police have to take measures to help the
community, but how this is possible is the question. Additionally, it is about the people’s
credibility on law.
2.1 Research Questions
The research questions are listed below:
1) How are the major and minor category crimes from January 2008 to December
2016 analyzed in London?
2) Which district has the highest and lowest major category crime among all the
districts?
3) Which district has the highest and lowest minor category crime among all the
districts?
4) Can the statistics help the public to participate in the efforts of preventing crime?
5) What the crime statistics have impact on public's credibility on the law?
6) Can the police serve be improved to help the community?
7) What is the crime situation after 2016?
8) Various crime exhibits different geographical patterns and different spatial
1

clustering. In this case, how is the spatial accuracy of the data examined for
various categories of crime?
9) In the spatial pattern, what factors lead to systematic inconsistencies?
10) What is the relation between the spatial accuracy of London crime dataset, with
the type of crimes, conditional on the degree to which they occur at places that
can be addressed easily.
2.2 Peer Evaluation
The below represented table contains peer contribution (Bingham and Fry, 2010).
3. Method
For the London crime dataset, the research analysis method utilized is discussed in this
section. This research can use the positivism research philosophy, as it requires the real facts
observed from the research. Subsequently, the research questions are created with the help of the
collected real facts. The patterns of the crimes can be observed using the abduction research
approach. As a whole the qualitative research methodology can be used, where the descriptive
statistics, correlation, data visualization, and significance analysis can be conducted. This
research can benefit with linear regression, for which the SPSS tool could be used as it helps to
identify the significance of data (Blunch, 2013).
2

4. Results
4.1 Project Planning & Management
The WBS is created based on the project plan and its management. The WBS is shown in
the following figure. The project progress is also represented below.
The below table shows the sprint back log (Fox, 2016).
Team
Members
Cards Tasks Estimation Status
Team
Member-1
One card-
Internal
Review
Introductions
The first member presents the
introduction, including the project’s aim
and objective.
1 Sprinting-
Completed
One card-
Internal
Review
Problem Statement
Also, the first member of the team
determines the problem in this project.
1 Sprinting –
Completed
Team
Member-2
One card-
Internal
Review
Background
The second member outlines the
background of the project.
1 Sprinting –
Completed
3

One card-
Internal
Review
Research Question
The second member also develops the
appropriate research questions.
1 Sprinting -
Completed
Team
Member-3
One card-
Internal
Review
Methodology
The third member identifies the
suitable research analysis methods for
collecting the data and answering the
research questions.
1 Sprinting –
Completed
Team
Member-4
One card-
Internal
Review
Results
The fourth member shows the result
of the implemented methods by
answering the research questions, and
provides the suitable data
visualization plots.
4 Sprinting –
Completed
Team
Member-5
One card-
Internal
Review
Discussion
The fifth member facilitates the
discussion part and interprets the
results.
4 Sprinting –
Completed
Team
Member-5
One card-
Internal
Review
Conclusion
The fifth member concludes the
project report.
4 Sprinting –
Completed
4.2 Descriptive Statistics
The conducted SPSS data analysis is depicted in the following figure.
4

The descriptive statistics helps to give the statistical information of the value of the
London crime data.
4.3 Data Visualization
The below shown graph visualization shows the major category crimes from January
2008 to December 2016 analyzed in London (Graham, 2011).
5

The graph visualization shows the minor category crimes from January 2008 to
December 2016 analyzed in London.
6

The below represented graph visualization shows the highest and lowest major category
crime among all the districts.
The below represented graph visualization shows the highest and lowest minor category
crime among all the districts (Hand, 2010).
7

4.4 Correlation
The crime value and years are correlated for measuring, formulating and classifying the
London crime data.
As per the correlation, the .sig value is 0.396 where the valued year are not statistically
significant, and it measures the crimes in London.
4.5 Significance
Linear regression in SPSS is used for find the data’s significance. The statistical
significance of the models created is determined as shown below (HARRELL, 2016).
8

The result show the linear regression implications.
5. Discussion
This research’s descriptive statistics supports to get the complete statistical details of the
particular dataset. The data visualization shows the below listed aspects (Heck, Thomas and
Tabata, 2012):
1) Represents the major and minor category crimes from January 2008 to December
2016 analyzed in London.
2) Represents the highest and lowest major category crime among all the districts.
The crime value and years are correlated, and linear regression is utilized for the
determination of significance of data. From the linear regression, over three tables are provided
9

where the model summary table clarifies whether the model is a good fit model or a bad fit
model. Based on the provided context, the R value is .000 so it gives 0% of variability depending
on the crime value. The ANNOVA table contains .792 .sig values and it doesn’t satisfy <0.005.
Thus, the model that is created is believed to be statistically insignificant for determining the
public to participate in the efforts of preventing crime, and crime statistics have impact on
public's credibility on the law with the type of crimes, conditional on the degree to which they
occur at places that can be addressed easily (Mcgrath, 2018). Also, the coefficient tables contains
sig value i.e., .525 depending on the crime values that are utilized for justifying the London
crime data (Lee, 2012).
6. Conclusion
The London crimes from January 2008 to December 2016 are analyzed. The results
conclude that the model created is not statistically significant, and thus it is a bad fit model for
London data justification. This research implemented the descriptive statistics, data visualization,
correlation, including the significance analysis. The linear regression with SPSS was used for the
determination of statistical significance.
10