There is an overwhelming number of data visualization solutions out there.  I really wish someone could put together a comparison report of a good number of them listing their features, appropriate application, and user critique.  In order to learn about the application and make things simple for me, especially during midterms week, I opted to visualize a time series chart of the Dow-Jones Industrial average between 2000 and 2014. This was, however, not an easy task since I still needed to pick the right software application and learn how to use it.

I opted to create my time series chart with Tableau because I believe it is used quite often out there in the “real world,” so it is a good skill to develop. Although it is a relatively simple tool to use, it did take me a bit to learn how to create my first chart. Once you get going become the least bit familiar with it, I really like that through usage you are able to find out what you can and cannot do with the data. Tableau will simply disable charts and options that do not make sense with your data set. Although this may seem logical, it is extremely helpful to analytic neophytes, like myself.

I decided to go beyond a simple time series chart and possibly add a bit more information and a political twist to it. Since we have been learning how charts and data visualizations can be manipulated to show, well, whatever the creator wants to show, I laid out the data points in a discrete (non-continuous) time series chart. On the y-axis I have the Dow-Jones Industrial Average Closing Prices and on the x-axis I the years broken up by financial quarters, which is what the business world uses as a standard to report financial information.

The political twist comes in pointing out that a person can potentially focus on one discrete year and report a trend to their benefit. For example, you had presidential elections in 2000, 2004, 2008, and 2012 within the data. Presidential candidates can point at the prior year’s trend upward or downward and make claims as to the competence of the incumbent in order to rhetorically attack them, while in truth holistic snapshot of the economy between 2000 and 2014 shows an overall upward trend. Furthermore, the time series can be cross-referenced with historic events that may have affected the Dow-Jones Industrial Average.

https://www.google.com/fusiontables/DataSource?docid=1uchWvXdmwwuk-4k0aLVyTJUjuVFoBUuYiW69u6p2

I chose to use the census dataset, which contains information about the population of the United States from every decade since 1790 until 2010. While the data on its own is relatively simple to read, it is difficult to conceptualize what all those numbers representing millions of people actually mean. When I put the data into a data visualization, in the case, a line graph, it allows users to see much more clearly the trend in the growing population of the country. Seeing this trend visual also makes it possible to easily predict future population growth patterns, which is massively helpful and important to know for those interested in public policy and government. Additionally, this graph shows some of the discrepancies in this overall trend; for instance, we can see a drop in the growth rate of the population in the 1940 census due to the Great Depression which occurred in the 1930’s. This line graph also has an additional feature which allows the user to highlight a certain time period to focus on and allows for closer analysis of fewer data points. Overall, the data visualization makes the data so much easier to work with and reveals the real meaning behind all the numbers.

I looked at the “Best City” data set which has an assortment of data for 20 cities in Florida relating to quality of life. The data involves everything from the number of restaurants and average income to the murder and rape rate of each city.

After looking at the dataset on a spreadsheet, I decided to look at Literacy and Income within all the cities via a scatterplot to see if I could see a relationship. I saw that once the average income in a city reaches 25K, the literacy rate is all over the board, 1.5 through 9.5–no obvious relationship. However, once the average income is greater than 30K, the literacy is always greater than 4.5. Before looking at the data visualization, I would have guessed that literacy would increase linearly as income increases, but the graph proved me wrong.

The dataset used to make the graph:

ControlPanel LA is the city’s source for information about expenditures, revenues, payroll,  and special funds.  The What We Buy   data shows which items L.A. spends our money on like   $21, 929 on custom  boots  for police on motorcycle patrol  or the $13,363 L.A. spent on the unfortunately named  L.U.S.T. Tax, short for Leaking Underground Storage Tank.

The data is presented in clickable card sets that provide relevant and easy to understand information on each expenditure .  The cards aren’t extremely in depth but provide a clear starting point for those who are interested in what L.A. does with tax revenue.

“(Ontologies)…act as objects that create and negotiate boundaries between groups…communities and states represent the realities around them through distinct ontologies, or systems of categories and their interrelations by which groups order and manage information about the people, places, things, and events around them.”

This site uses this definition of ontology to be easily understood by the community.The boundaries set are what information the L.A. Controller’s office chooses to give freely.

This data set could be more specific. The exact time range for purchases are missing. The ” What is this” and “Why did we buy this” descriptions can come off a bit patronizing in sections like the LA Fire Department where they explain “firefighters use hoses to pump water from trucks and hydrants… fire hoses are essential tools for fighting fires protecting property and saving lives”. We know that but why did our fire hoses cost $1,348,566 and what exact date was that purchase or when will you have to purchase more because that only paid for 7,617 hoses.

If I was starting over, I would allow for more links to deeper examination of spending.

L.A. Controller’s Office breaks down the total earnings of each city department in this data set.  The Office uses monetary data, each record consisting of the total earning of a city department, to increase transparency about the department that most heavily relies on the tax payers’ money. The data set can be visualized in multiple ways: pie chart, bar graph, line graph, donut chart, etc. Users can also customize the filter to reorganize the data by year, projected annual salary, quarterly payments,  department title, etc.

According to Wallack and Srinivasan, a dataset’s ontology is a “system of categories and their interrelations by which groups order and manage information about what’s around them.” In other words, ontology is a way of organizing information in a dataset through categorization. Ontologically speaking, this particular dataset is categorized by department title, which serves as the “boundary” that enables us to compare groups of information. This ontology makes the most sense from the point of view of U.S. citizens who reside in Los Angeles. It is through their tax money that the city’s departments sustain operations that benefit L.A. residents, making them the stakeholders of the departments’ finances. It is critical that they are aware of how much each department makes because only then can you make judgements about your “money’s worth.”

According to the data set, LAPD has the highest payroll with total earnings of $1,299,609,453.87, followed by the Department of Water and Power, with total earnings of $1,114,504,612.43.

More than half of the total sum of departmental payrolls goes into supporting these two departments, which hints upon the high crime rate and severe drought in Los Angeles. The task of curbing crime in a major metropolis such as Los Angeles is costly. Also, it makes sense that a large portion of the tax payers’ money is going into providing water for local communities, given that water is in high demand and short supply in Los Angeles these days.

The data set leaves information about why payroll is distributed in such way. From the perspective of a politician who may be interested in learning more about how to more efficiently distribute funds to each department, it would be helpful to organize the data according to use – use meaning, the reason for which the department was designated a certain amount of payroll in comparison to other departments.

The LA Controller’s Office features data from the City of LA’s Bureau of Street Services that composes an interactive map containing a data set called Street Grades. This data set provides information about the quality and state of LA’s roads, and can be manipulated using this interactive map. The data presented on the map can be filtered using five different criteria: council district, neighbourhood council, road repair, street pavement condition, and 2014-15 road repair, and filters can be applied at the same time, if the user desires. Within these categories there are only three data types: geographic boundaries based on city council, road conditions, and road repairs, both past and scheduled. Therefore a record is a single road and its variables represent the data associated with that record.

Wallack and Srinivasan define ontology as a system of categories by which groups order and manage information about the things around them. In this case therefore, the Street Pavement Condition is the dataset’s ontology, as is provides clearly defined categories into which roads can be assigned. The road conditions are quantified using the Pavement Condition Index (PCI) which runs from 1-100 and is the sum of the road’s present physical condition and how much repair is necessary to restore the road to perfect conditions. The PCI then categorizes each road into five groups based on score: failed, poor, fair, satisfactory and good.

In my opinion, the people who will find this information most useful and illuminating are industries that rely on road transportation networks, such as taxi, courier and shipping services and navigation software. This is because roads that are in good condition tend to provide more efficient traffic flow than, since they are easier and less challenging to drive on. On the other hand, roads that are in poor repair slow down traffic due to drivers avoiding potholes and other nuisances in the road. Additionally, real estate companies can also find this information useful as they can rate geographic regions’ quality of living using this information.

This dataset does a very good job at describing that many of LA’s roads are in poor condition and are in great need of repair. It also succeeds in showing that while many are in need of repair, very few of these roads are actually scheduled for such repair. It also illuminates that more affluent neighbourhoods tend to have better quality roads.

There are some cities within LA that withhold their data on their roads conditions, such as Santa Monica and Beverly Hills. This is an inconvenient gap in the data and it would be nice to have the entire map covered with interactive data.

I chose to look at the LA Control Panel’s Payroll By Department.  I was somewhat surprised to see how funds were distributed and what departments required the most funding. The dataset shows the amount of funding each city department received from 2011 to 2013. It shows how much funding the departments have received over time in different graph formats.

According to Wallack’s and Srinivasan’s definition, the ontology of the dataset is the City government and the larger United States government because the data displayed is the city government’s payrolls and allocation of funds which is also representative of the US government as a whole . The site clearly maps out where all of its funding is distributed and in that way it represents its interests and activities of the city. The ontology of the data represents the interests of the government because it was sorted and published by a government agency. The way they display their data is cognizant of how they would like to be perceived. However, the data is held accountable by its viewers, the public, and there is a section for leaving comments so in that way the community also contributes to the site’s ontology. I found the data very illuminating because it was so clearly laid out and did not seem to have much of an agenda. However, I did find that the lack of breakdown in what the money is going towards in each department was troubling

The Dataset shows me where my taxes are going and how much funding each department receives. It is clear from the charts that LAPD, Water and Power, and Airports receive the most funding while many other departments that seem to be struggling receive a small fraction of funding. There is no description of what the section labeled other is. I am not sure what departments are in that section and the fact that they receive the least funding and are not named is worrying. There are also no descriptions of the agencies or what they do, although a quick google search can show you.

I would gather information from different communities and see how they thought or wished their tax money was being spent and then compare it to how it is actually being spent. Or I would get the different departments budgets and cost needs and reference that against the funds being allocated. It would be interesting to see how the departments divided up the funding if they all had a say in it.

This week I decided to analyze the data identifying the Top City Earners of Los Angeles. This dataset shows the payroll information for all Los Angeles City Departments, from fire chief, to police deputy chief and etc. The data type for this dataset shows each positions salary and salary type, and is organized in a way that you can compare each position with each other. The salary types are separated into categories such as base pay, permanent bonus pay, temporary bonus pay, overtime, and so on. This dataset also has filters in case someone wanted to look more in depth at the information and find out a specific job title’s specific pay in a specific category.

Wallack and Srinivasan described ontologies as a way to look at data and find particular relationships between the different categories. When looking at this dataset, it can be said that the ontology are the different positions and the salaries for each position. This dataset would make the most sense for anyone interested in looking at how much city officials are being paid. As long as you are interested in how much city officials are being paid, the data is relevant and by being able to compare the salaries next to each other, it is also easy to see how different salaries line up next to each other. The dataset is fairly easy to understand and after playing around on the site for a bit it becomes easy to see what the data is showing. The different filters are a little confusing, but after playing around with the site, it becomes better to understand.

As thought out and well done as this dataset is, its not the best way to find out why certain jobs are paid more than others. The data would be more helpful is there was a way to understand why certain jobs make more money than others, as well as what they do in their certain positions. Why does one Fire Captain make more money than the other? What and where in Los Angeles are they working. By adding these constraints, the information would be beneficial to a different ontology.

https://controllerdata.lacity.org/Payroll/Top-City-Earners/78mt-gezm

This dataset lists all of the top paid public employees of Los Angeles county, and also has a few different methods of visualization and comparison that the user can use to get a better sense of how the county government is allocating its funds. The data types are primarily monetary, as the dataset is essentially the payroll of all the county’s highest earners. A single record in the set consists of a job title, along with their annual pay, which is then broken up into smaller categories including base pay, bonus pay, overtime, and lump sum pay. By default, when the site is opened it presents the data in a stacked bar graph of the types of pay of the employees, ranking from the highest down.

Wallack and Srinivasan define ontology as a system of categories and their interrelations by which groups order and manage information about what’s around them. The ontology of this dataset comes from the county government to be as transparent as possible in showing taxpayers where their money is going by creating varieties of visualizations that present the allocation of funds. While they do a decent job of making the information easily accessible and understandable, it is definitely geared toward the public and the citizens of Los Angeles who may be wanting information about how their government is running finances.

However, some aspects of the information are quite vague and it is not easily discernible what the job titles explicitly mean and what these employees are actually doing to earn their salary. Nonetheless, if the user does a bit more research this dataset gives great information about what is most important to the running of the county; for instance, the single highest paid employee is the Chief Port Pilot who runs Long Beach port, and the most funds are allocated to the police department. While this ontology doesn’t easily allow input from the actual citizens of Los Angeles on whether or not they believe this is the proper way to organize funds, the effort toward transparency is a step toward including the public more.

That being said, if I were to reorganize this dataset with a different ontology seeking to get more interaction from the people funding public jobs, I would add a feature that compared the reality of the salaries to the public’s idea of what would be appropriate, thereby allowing more collaboration on the issue and clearly showing every party involved how to best balance the financial wants and needs of Los Angeles.