Social Media Data Extraction: Managing Social Media Data

BEARdata is Baylor University's data repository.  Data extracted from social media is stored in a dataverse called Social Media Extraction Data which contains sub-dataverses for both Twitter and Facebook.

In order to upload your data, you must create an account in BEARdata.  Then email the BEARdata Admins and let them know whether you will be uploading Twitter and/or Facebook data. 

After you have access to the appropriate dataverse(s), you will be able to upload datasets according to the directions below.  Your datasets will not be published until and admin has approved it.  Once published, the metadata/information about your data will be available, but users will need to ask for permission to download the actual files.

Create a README for your data

1. Download the Twitter README file listed below
2. Edit the README file
   1. At the end of the file, replace KEYWORD_YYYY-MM-DD with the actual name of your file. 
   2. Replace ___ with the number of rows (not including the header) in your file.
   3. Add a line/repeat for every data file that you will upload

Upload the README and datafiles to BEARdata.

1. Go to the Twitter dataverse.
2. Click on the down arrow on the Add Data button and select New Dataset.
3. Use the Twitter template to enter the metadata for your dataset.  Replace everything in CAPITAL letters (including dates) with information about your dataset.  Starred (*) items are required.  It will seem like you are entering the same information more than once (that's okay).
   1. Title:  Replace "KEYWORD" with the keyword used for the search and replace YYYY-MM-DD with the date the search was run.  Keep the rest of the title intact.  If you ran multiple searches with the same keyword, you can generalize the date appropriately.
   2. Author:  Use the plus sign to add more authors.
      1. Enter your name with the format LAST, FIRST. 
      2. If you have an ORCID iD, select ORCID from the Identifier Scheme menu and add your ORCID iD.
   3. Contact
      1. Enter your name with the format LAST, FIRST.
      2. Add your Baylor email.
   4. Description:  Replace "KEYWORD" with the keyword used for the search and replace YYYY-MM-DD with the date the search was run.  Keep the rest of the base description intact.  If you ran multiple searches with the same keyword, include the dates for each run.  You can add additional information.
   5. Subject:  By default "Other" is chosen.  Change if appropriate.
   6. Keyword:  Replace KEYWORD with the keyword used for the search
   7. Production Place:  If the data were collected for a course, replace COURSE # with that information (e.g. MKT 3305)
   8. Depositor:  This will automatically insert your name.
   9. Date of Collection:  Enter the start and end date of the data you collected.  You will need to look at your data for this information.
   10. Kind of Data:  This will automatically fill in.
   11. Software:  This will automatically fill in.
4. Upload your README file
   1. Choose the Unrestrict status from the drop-down menu
   2. Click on Edit Tags
   3. Select Documentation
5. Upload your data file(s)
   1. Double check that the Restrict status is chosen
   2. Click on Edit Tags
   3. Select Data
6. Click on Save Dataset
7. A confirmation email will be sent to you.

Create a README for your data

1. Download the Facebook README file listed below
2. Edit the file
   1. At the end of the README, replace #####_YYYY-MM-DD with the actual name of your file. 
   2. Replace ___ with the number of rows (not including the header) in your file.
   3. Replace https://www.facebook.com/GROUP/ with the URL for the Facebook Group.
   4. Replace GROUPNAME with the name of the Facebook Group.
   5. Add a line/repeat for every data file that you upload

Upload the README and datafiles to BEARdata.

1. Go to the Facebook Dataverse
2. Click on the down arrow on the Add Data button and select New Dataset.
3. Use the Facebook template to enter the metadata for your dataset.  Replace everything in CAPITAL letters (including dates) with information about your dataset.  Starred (*) items are required.  It will seem like you are entering the same information more than once (that's okay).
   1. Title:  Replace "GROUPNAME" with the keyword used for the search and replace YYYY-MM-DD with the date the search was run.  Keep the rest of the title intact.  If you ran multiple searches with the same groupname, you can generalize the date appropriately.
   2. Author:  Use the plus sign to add more authors.
      1. Enter your name with the format LASTNAME, FIRSTNAME. 
      2. If you have an ORCID iD, select ORCID from the Identifier Scheme menu and add your ORCID iD.
   3. Contact
      1. Enter your name with the format LASTNAME, FIRSTNAME.
      2. Add your Baylor email.
   4. Description:  Replace "GROUPNAME" with the keyword used for the search and replace YYYY-MM-DD with the date the search was run.  Replace #### with the Facebook Group number and replace https://www.facebook.com/GROUP/ with the Facebook Group URL.  Keep the rest of the base description intact.  If you ran multiple searches with the same keyword, include the dates for each run.  You can add additional information.
   5. Subject:  By default "Other" is chosen.  Change if appropriate.
   6. Production Place:  If the data were collected for a course, replace COURSE # with that information (e.g. MKT 3305)
   7. Depositor:  This will automatically insert your name.
   8. Time Period Covered:  Enter the start of the Facebook Group and date you collected the data  You will need to look at your data for this information.
   9. Kind of Data:  This will automatically fill in.
   10. Software:  This will automatically fill in.
4. Upload your README file
   1. Choose the Unrestrict status from the drop-down menu
   2. Click on Edit Tags
   3. Select Documentation
5. Upload your data file(s)
   1. Double check that the Restrict status is chosen
   2. Click on Edit Tags
   3. Select Data
6. Click on Save Dataset
7. A confirmation email will be sent to you.