Project 1: Comparing how journalists or politicians use Twitter
For this project you must compare how either 6 journalists or 6 politicians use Twitter. The comparison must include journalists from at least 3 news organisations or politicians from at least 3 political parties. You can compare within Australia or beyond.
Research Question 1: What types of message do the journalists/politicians send out on Twitter, and how do they compare?
Research Question 2: Who do politicians/journalists mention or reply to in their tweets?
You must use NodeXL to collect 400 tweets in total made by your chosen politicians/journalists. You must analyse an equal number of tweets for each person (so if you have 6 journalists, you would analyse 66 tweets from each).
- Only collect and analyse the politicians'/journalists tweets (do not analyse mentions of them by others, replies to their tweets)
- You must have a rationale for how you select your 'cases' - the individual Twitter users. This might be purposive (e.g. you focus on party leaders) or random (e.g. randomly select from a list of politicians on Twitter)
- You must either randomly select tweets from each person, or analyse their most recent tweets.
- You must code the data using the Coding Analysis Toolkit: the method is content analysis and you will define your own codes for the contentanalysis
Research Question 3: What is the public sentiment towards either the political parties, media organisations, or the journalists/politicians? Who receives the most negative and positive Tweets, and on what topics?
You must use Senti-strength to analyse public sentiment about your chosen cases. Senti-strength will automatically analyse the sentiment in the Tweets.
- You must collect at least 600 Twitter mentions in total (equal numbers from each case e.g. 200 from Labour, Liberal, Green). You should use NodeXL to collect the Tweets.
- You must exclude the Tweets from you cases - this focuses solely on how other people talk about the them
- Senti-strength uses a scale of positive and negative sentiment.
- You must present a table with the number of Tweets in each sentiment category
- You must separate out the Tweets with the most positive (+5) and negative (-5) sentiment. You must qualitatively analyse these Tweets - what are the key themes or topics that people are positive/negative about. You do not need to use content analysis (it is likely to be a very small number of Tweets). If there are no highly positive or negative Tweets, you must discuss this finding.
Research Question 4: How do the Twitter social network maps for your chosen cases compare? (Potential sub-questions: Who holds the most important positions in the networks? What positions do the most positive and negative Tweeters hold? Is there political polarisation in the networks? Do journalists from different brands engage with each other? Now effective are users at spreading their messages?)
Project 2: Online Activism
In this project, you must analyse campaigns on the 38 Degrees website. This is a progressive organisation and is largely UK-based. 38 degrees hosts hundreds of campaigns across a wide range of topics. and allows people to comment on the campaigns. Most campaigns are submitted by users, and people are asked to show support for this by adding a digital signature. The petition creator sets the number of signatures which are required for a petition to be successful. (Note: we selected this website rather than say Getup! because it has more information on the website, and campaigns are largely crowd-sourced rather).
Research Question 1: Are campaigns started by local, national or International groups or Individuals? (Potential sub-questions: Do some users create multiple campaigns? Are campaigns created by males of females?)
Research Question 2: How many signatures are required for a campaign to be successful, and how close to this target is it? How many campaigns exceed their target?
Research Question 3: Do campaigns by groups or Individuals receive more support. and which is most likely to achieve their target?
Research Question 4: Who are campaigns directed at?
- You must randomly select a campaign section (e.g. health). You must select 100 campaigns (either random sample, or start with the newest/oldest)
- You must collect or 'scrape' relevant content from the websites using Outwit Hub Pro (the scraper to collect the data is provided).
- Use Excel or another spreadsheet to undertake the quantitative analysis
- To analyse who starts campaigns, and who they are directed at you must use content analysis. You must create a coding frame to categorise the data
Research Question 5: How many comments are received? What is the sentiment in comments towards the campaign? (Potential sub-questions: what campaign topics create the most positive and negative comments? Is sentiment worse or better in longer discussions?)
- Work with your existing 100 campaigns.
- Collect the number of comments (listed on the campaign page)
- Collect the comments that occur under each campaign.
- Use Senti-strength to analyse the sentiment in the comments.