jlibman / final_project Goto Github PK

Create interactive visual representation of data.

Requirements:

• Create an interactive platform using R. This could be a report, slideshow, or interactive visualization (preferably compiled into a webpage for enhanced accessibility)
• Choose a dataset of sufficient complexity to explore. This doesn't mean it has to be Big Data, but it should either be of notable size (more than a few hundred observations), or complexity (more than ~5 features of each observation)
• Demonstrate a nuanced understanding of the important features of the dataset. High-level insights (important descriptive information, major trends, notable outliers, etc.) should be prominent in your resource. Statistical analyses may be included if appropriate
• Devise a visual representation of your data (i.e., chart, map, etc.). Ideally, this should be an interactive visualization with ability to interact with the interface (hover, click, drag, etc.), or change the data being displayed (i.e., the chart responds to a set of controls, such as which data is being displayed)
• Tailor your resource to a specific target audience. The amount of framing you need to do for a scientific versus general audience is quite different

Project Proposal:

Project Description

We will be working with the Enrollment Status of the Population 3 Years Old and Over, by Sex, Age, Race, Hispanic Origin, Foreign Born, and Foreign-Born Parentage supplied by Child Protective Services of the United States. This dataset was updated in October 2014. This data was accessed though the census.gov website, routed through the United States’ public data domain, data.gov. This data was collected through the United States census. Our project will explore whether or not gender, race/ethnicity, or if students are foreign-born or have foreign-born parents appear to contribute to withdrawal from school, ages three and older. Our target audience includes education officials who will use the information to target students who we find to be more likely to withdraw from school.

Technical Description

We will be creating a HTML document. We will be comparing twelve different .csv files, separated by the ethnicity and the “born status” of students. In order to compare these files, we will need to reformat each file that has unnecessary rows, columns, and merged cells so that it will be easier to join the files. We are anticipating using some combination of ggplot2, ggvis, and htmlwidgets, in addition to several common libraries. We will not be using any outside statistical analysis or machine learning software to answer our questions. We will manipulate the data, present it visually, and draw conclusions. The major challenges we are anticipating are version control issues and figuring out the most effective visual representations to answer our questions.

Reformat and join the data sets.

Manipulate data to find trends and outliers.

Make summary tables and graphs so we can then find trends that we may want to explore further.

Compile the final HTML document.