Getting data into your statistical analysis system can be one of the most challenging parts of any data science project. Data must be imported and harmonized into a coherent format before any insights can be obtained. You will learn how to get data into R from commonly used formats and harmonizing different kinds of datasets from different sources. If you work in an organization where different departments collect data using different systems and different storage formats, then this course will provide essential tools for bringing those datasets together and making sense of the wealth of information in your organization.
This course introduces the Tidyverse tools for importing data into R so that it can be prepared for analysis, visualization, and modeling. Common data formats are introduced, including delimited files, spreadsheets and relational databases, and techniques for obtaining data from the web are demonstrated, such as web scraping and web APIs.
In this specialization we assume familiarity with the R programming language. If you are not yet familiar with R, we suggest you first complete R Programming before returning to complete this course.
Importing (and Exporting) Data in R
A basic data type in the tidyverse is the tibble. Tibbles store tabular data and are a modern take on the standard R data frame. They have many user-friendly features that are an improvement over standard data frames when doing interactive data analysis. The remainder of this module covers tabular data in spreadsheet formats like Excel, CSV, TSV, and other delimited files.
JSON, XML, and Databases
Data can come in non-tabular formats, especially unstructured data or data that otherwise would not fit into a table. JSON and XML are common formats for storing arbitrarily structured data and this module covers the packages used to read in those data formats. In addition, relational databases are common for storing very large collections of tables where you do not need to read in the entire dataset at once. There are many relational database formats and we will cover the SQLite format, which is a compact and simple to use format.
Web Scraping and APIs
Reading in data from various Internet sources can be a useful way to build analyses that need to be regularly updated. The rvest and httr packages are useful for connecting to web sites, web APIs and other online sources of data.
Foreign Formats, Images, and googledrive
Working with others in a data science project often involves reading output or data produced using other statistical analysis packages or other software. This module covers packages for reading in these foreign formats, as well as images and data from Google Drive.
Now we will demonstrate how to import data using our case study examples. When working through the steps of the case studies, you can use either RStudio on your own computer or Coursera lab spaces provided for each case study.
Project: Importing Data into R
This project will give you the opportunity to read in data from multiple sources and conduct some simple operations on those data.
Carrie Wright, PhD, Shannon Ellis, PhD, Stephanie Hicks, PhD and Roger D. Peng, PhD