instaRanalysis

Guide and requirements for analysing posts on Instragram

1. Install required software
   
2. Scraping
   
3. Data manipulation in R and export to Microsoft Excel
   

Required software:

• Python
• Instagram Scraper
• R
• RStudio (optional)

Install Python on Windows:

1. Download
   
2. Open Command Prompt (on some corporate networks you need to run the command prompt as administrator)
3. setx PATH "%PATH%;C:\Python27\Scripts"

Install Python on Mac:

1. Install Xcode tools
   

   1. Open Terminal and copy/paste the following code:
      xcode-select --install

2. Install Homebrew
   

   1. Paste following code into the terminal window:
      /usr/bin/ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"

   2. Add Homebrew to your path. Paste following code into the terminal windows:
      export PATH="/usr/local/bin:/usr/local/sbin:$PATH"

3. Install Python (2.7) via terminal:
   brew install python@2

   1. Add Python to your path:
      export PATH="/usr/local/opt/python@2/libexec/bin:$PATH"

Install Instagram scraper:

1. Open command prompt (windows) or terminal (mac)
2. Paste the following command:
   pip install instagram-scraper

Install R and RStudio

1. Download R

2. Download RStudio

3. Install packages for R:
   

   • jsonlite
   • stringr
   • tidyr
   • dplyr
   • openxlsx
   • plyr
   • repurrrsive
   • purrr
   • webshot

   e.g.:
   install.packages("jsonlite")

   1. Install PhantomJS for R:
      webshot::install_phantomjs()

Run Instagram scraper:

To scrape a user's account:
instagram-scraper username -u yourusername -p yourpassword –-media-metadata –-comments –d path

To scrape a hashtag:
instagram-scraper hashtag --tag

The program will produce images for each instagram post and a json file with all metadata (tag, post, likes, comments, etc.).

Check full documentation here

Data manipulation in R

This operation outputs a readable Microsoft Excel file.

1. Download this repos and save to a location where the scraper downloaded images and json file
2. Open RStudio
3. Create new project and choose the same save location
4. Open file: instaRanalysis.Rmd
5. Change the variable input to json filename (without the .json extension)
   1. E.g. input <- "somejsonfile"
6. Change the variable fileLoc to the the full path of where the images from the scrape are saved:
   1. E.g. fileLoc <- "C:\myfiles\"
   2. Works also with online drives such as Microsoft Onedrive or Sharepoint:
      • E.g. fileLoc <- https://corpname.sharepoint.com/Sites/sitename/Shared%20Documents/images/
7. Run both "chunks" (wait for one to finish before you start the second)
   1. the command:
      webshot(c(df$url),delay = 3, file="InstaShot.png")
      downloads screenshots of instagram post. This process might take a very long time for especially big datasets. This command may also exit with an error saying it cannot open a specific link. Usually this can be resolved by running this line again.