Instructions for Managing OnPoint Research
Definition: OnPoint is a mobile application of our data, designed to be issue focused and location specific.
Managing OnPoint from the research end can be summarized into three categories: Featured Issue Selection/Revision, Issue Descriptions, and Tag-Association Management
Featured Issue Selection/Review
Featured Issues are meant to best encapsulate the current political climate. We only want the most current and relevant issues to display on the app, which requires adding new featured issues as they arise and taking down issues that are no longer prevalent.Selection:
- Featured issues are selected on a rolling basis because a major issue can crop up at any time.
- Speeches sub-department will often be the most on top of what issues are currently being talked about, so they should be looking for potential featured issue additions to report to the Officials Research Director.
- Speeches, news articles, legitimate polls, and legislation should be gathered to be used as evidence that the issue should be incorporated into the app.
- Gallup can be a great resource here: http://news.gallup.com/poll/1675/most-important-problem.aspx
- Queries should be run on the topic to see the breadth and quality of the data we have on the issue.
- An on the fly meeting will then be had by Department Directors and National Director to decide if the issue will be incorporated.
- Once an issue is selected for the app, tag-ties and a description need to be made (info below)
- Research Directors, National Director and Tag Manager meet bi-monthly (once every two months...it's a flawed word! https://www.merriam-webster.com/dictionary/bimonthly) to review the current list of featured issues.
- Featured issues that are no longer relevant or widely followed should be brought up as possible cuts
- Criteria for cutting a featured issue can include but is not limited to: event related issues becoming outdated; issues being settled legislatively/judicially; issues that simply fall out of the news cycle.
- If it is decided that an issue should be cut from the current featured issues list, a mantis ticket needs to be filed for IT to un-release the issue
- It's very easy to un-release an issue, and the data (tag-ties & description) is not lost when we do this - the action required is a simple release_id change in the featured table.
Writing Featured Issue Descriptions
Descriptions provide an introduction to the issue for users. It is very challenging to keep informative, accurate and non-partisanWriting The Description - What To Include:
- Basic background on the issue
- The contested points of the issue
- The reasoning for both sides of the argument
- The current status of the debate
- Relevant statistics and factual information from outside sources
- Once written, descriptions should be reviewed for grammatical errors, spelling errors, and partisanship first by the Director of Research, then the Development and Communications department.
- Finally, descriptions should be converted to .csv format (one column for title and one column for text) and submitted to IT using mantis.
- They are stored in the description column of the featured table
Tag-Association Management
Data is pulled for OnPoint through our tagging system.How the Data Pulls
- Each featured issue has a list of tags that are associated with the topic, therefore ONLY data with tags found on these lists will be found on the results page.
- Ex: A statement on gun sales will appear under the "Guns" featured issue because it is tagged with "firearm-sales", a tag found on the Guns tag association list
- Goal is to accumulate list of specific tags associated with the featured issue
- If the tags are not specific, non-germane data will pull
- Tag Manager leads this project
- Queries and general issue/tag knowledge are used to create these lists
- Lists are revised and updated on a rolling basis, as we are always coming out with new tags.
- New/Updated lists are imported to the database via Mantis ticket import
- Tag association data is stored in the featured_tag table
Tables to Know
- featured - This table houses the featured issues, and links featured_id to a title and description. Release_id is the only other column in the table, and it works like any other release_id.- featured_tag - This table is where the tag associations are found for each issue. Each tag_id is tied to a specific featured_id (featured issue).
