JRNL 3165 - Assignment #4: Data Update 3

Chart

Unanswered Question

The dataset I used is quite comprehensive, but it is missing one data point that could lead to some very interesting analysis: the location of the issued ticket. One question that I would be very curious to see an answer to is related to the subset of tickets issued for 'Street Traffic' bylaws - Where in Port Moody are you the most likely to receive a parking ticket? Given the depth of the data that I already have, the analysis could go even deeper. Further topics like when you are the most likely to receive a parking ticket at a specific location, or why certain areas see more tickets than others could also be explored.

How would you get the data?

It is very likely that Port Moody already stores the ticket location alongside the rest of the bylaw ticket information that they have publicly available. I believe that it would be possible to obtain the location information as an addition to the existing dataset that I am using. Bylaw enforcement in Port Moody is carried out by their Bylaw division. As such, they would likely be the best place to initially reach out to regarding that addition to the dataset. Failing that, Port Moody is subject to Freedom of Information requests, so one could be submitted using the information provided on their website.

JRNL 3165 - Assignment #3: Data Update 2

Lead

The number of tickets issued for Port Moody's most frequent bylaw infractions saw monthly increases of up to 66% during the Summer, according to City of Port Moody data.

Excel Workbook and Explanation

Port Moody Bylaw Tickets - Raw data & Slice

The above Excel workbook contains two sheets, the first being the raw data from the City of Port Moody, and the second being a small slice of data which was used for analysis. The slice of data is based on total bylaw tickets issued for the 'Rocky Point boat launch & Parking Regulation' and 'Street Traffic & Public Places' categories. These categories were chosen as they were the two most frequently ticketed infractions in the data set.

The slice makes use of two tables: one for the monthly percentage of total bylaw tickets issued for each bylaw description, and the other for the change in percentage each month. Comparing these two tables allows for a reasonable analysis of which monthly increases were significant and not based on smaller numbers (e.g. for a 0.5% to 0.9% increase).

Original Dataset Link

I am using the 'Bylaw Tickets' dataset from Port Moody's Open Data Portal.

News Story Link and Brief Summary

A Tri City News article notes that Port Moody bylaw officers were moving from warnings to fines at Rocky Point during the summer of 2015. The article also notes that the city now has extra bylaw officers, one of which is stationed at Rocky Point to monitor parking on the weekends. This increase in enforcement can help explain why parking enforcement at Rocky Point was part of the second most frequently ticketed bylaw description.

JRNL 3165 - Assignment #2: Data Update 1

1. What dataset will you use for your final report?

The dataset I will use for my final report is titled 'Bylaw Tickets' from Port Moody's Open Data Portal.

2. Describe the dataset. What kind of data does it contain?

The Bylaw Tickets dataset contains over 38000 rows of data about bylaw tickets issued in Port Moody between January 1st 2008, and November 17th, 2020. The dataset contains columns such as: 

• Ticket numbers
• Ticket dates and times
• The ID number of the issuing officer
• Vehicle information (make, model, colour)
• Bylaw numbers, sections, and descriptions
• Dollar amounts for fines and discounts
• Long form descriptions with additional detail

The data about the tickets is broad and contains a wide variety of infractions such as improper parking, unlicensed dogs, noise, cutting down trees without permission, and graffiti.

3. Is there anything about your data that you don’t understand? How will you find this out?

As a whole, the data is actually quite straightforward and well described. The only column heading I am not currently sure of is the 'Offence_Code' column, which primarily contains whole numbers, but also includes entries like '4(A)', '16.18(F)', and a few numbers with decimals like '3.1', '4.1', and '4.2'.

To figure out exactly what is represented in the 'Offence_Code' column, I will likely need to look into information about Port Moody's bylaw enforcement. I will search through their website, and if I'm unable to find it there, I will have to look into external sources.

4. What are some questions you hope to answer with your data? List at least three.

Some of the questions I hope to answer are:

• Which officers issue bylaw tickets the most? Which bylaw infractions do they issue tickets the most for?
• At each hour of the day, which bylaw infraction is the most likely to generate a ticket?
• Which type of bylaw ticket has seen the most significant monthly increase or decrease over the length of the data set?
• Which type of bylaw ticket are the most common vehicle makes the most likely to receive?
• What is the average number of bylaw tickets issued each month?

There is a significant amount of data in this dataset, so depending on the question, I may be required to filter the data into a subset (10-20) of the most frequently seen data points.

JRNL 3165 - Assignment #1: Data Visualization Analysis

"Winners of Nobel prizes: top 5 nations", from The Guardian's 2016 article titled "Which Countries have had the most Nobel prize winners", additional annotations 'A', 'B', and 'C' added for reference.

As a whole, this chart does a poor job of clearly explaining the data it references. This chart shows a rough proportion for each type of Nobel prize won by five countries, but beyond the already numbered areas, it is essentially impossible to interpolate the remaining numbers exactly.

These charts are capable of answering some questions about the data, one such example being: "In each listed country, which Nobel prize category has the highest number of winners?". In most cases however, this series of charts is not capable of providing concrete answers. For all five charts, asking a question like "Which was the 3^rd most popular category for Nobel prize winners?" would not quickly or easily lead to a quantitative answer.

There are a few key areas which could be changed to improve the clarity of these charts:

• As seen in annotation 'A', each area should contain the numerical value being represented visually.
• In annotation 'B', the border extends outside the bounds of the area being represented, which has a side effect of making the areas of the surrounding boxes slightly unclear.
• Most importantly, in annotation 'C', these diagrams do not have a clear scale to reference. At a glance, these charts are highly based on visual proportion. If there is a difference between the green and pink boxes, it is unlikely a quantifiable number could be found.
• As a whole, the charts are highly inconsistent. Between each chart, individual boxes may scale horizontally or vertically, but for seemingly no reason as the shape of the charts can vary significantly like the 'France' chart. Improving consistency between charts would make it easier to compare the same categories across each of the five charts.

While these visualizations may look nice at a glance, for the purpose of actually visualizing data, a set of bar charts would be vastly superior. Each category would be easily compared to the others, and with a labelled axis, you would be able to quantify the exact number of prizes won in each category for each country. 

This particular visualization was not interactive, but it would certainly have potential to be a useful interactive visualization. For example:

• Hovering over each box could show the category title, and number of winners.
• The categories could potentially be rearranged to sort themselves numerically (by area) rather than randomly.
• Depending on implementation, the different charts could be more easily compared to each other.
• Clicking each box could lead to a list of the winners for each category.

Overall, this visual is not very usefully representative of the data, and has great potential for improvement.