# Mini Project B2 - Creative Rationale ## By Group 12 :face_with_cowboy_hat: (Group 4000: Født I Går (Born Yesterday)) --- ## What are we researching? :microscope: In this mini project we are researching data from the initial speeches about the COVID-19 lockdown from different nation leaders. We are doing so in order to get an understanding of different countries’ standpoint regarding the spread of the virus as well as their initial strategies to stop the virus from spreading. By doing so, we intend to investigate how political and national relations and values are reflected in the speeches. --- ## RQ :pencil: * To what degree are political and/or national values refleced in nation leaders initial responses to the COVID-19 pandemic? --- ## Why did we choose this topic and why is it relevant? :chart_with_upwards_trend: Since the COVID-19 virus started to spread all over the world, we have seen different countries fighting the virus in different manners. We consider the fact that restrictions vary heavily from country to country as an interesting research topic because COVID-19 as a global pandemic requires global attention and collaboration in order to be brought down. Our interest regarding the topic is rooted in an interest in understanding how political values play a part in determining what approaches different countries have chosen to fight the pandemic and how these values are reflected in the initial speeches to the population of the individual countries. In an international crisis as the COVID-19 pandemic we find it very interesting to analyze different national leaders' approaches and handling of the pandemic. By analyzing the speeches from national leaders we can analyze the words, the context and what kind of emotions the national leaders are generating in the speech. Another reason that makes our research relevant is the topicality of the research. Speeches have often been compared through history to analyze comparisons and contrasts, but the topicality of our speeches makes it unlikely that anyone has compared the exact same speeches as us. We have searched the internet for comparison of speeches from nation leaders concerning the corona pandemic without result. The explanation for this could be that the pandemic is a new phenomenon and this could be a reason why comparison of these particular speeches haven't been made yet. --- ## What is our approach to the subject and what elements are we focusing on? :open_book: Our approach to researching the speeches was focused on the political and emotional elements of the speeches. We wanted to analyze the words used in the speeches to see which words were used the most and if the words as keywords were a part of certain programmes, anti-programmes of if they tried to stay neutral (Rogers, 2017). Afterwards we looked at which emotions was used in the speeches. --- ## How? :grey_question: #### Generel overview To start our research we found the first COVID-19 related speeches by national leaders from Denmark, USA, Germany, UK, Australia and Singapore. We chose 6 countries in order to get as broad an angle as possible, without making the research too comprehensive with regard to the frame of a mini project. #### Tools We analyzed the speeches with Voyant Tools and the IBM NLU. In Voyant Tools we added a list of stop-words in order to clean the gathered data for words that did not concern our topic of interest. This gave us a comparable overview of the most frequently used words in the different speeches. In addition to Voyant, we used IBM Natural Language Understanding for analysis of emotions and sentiment in the speches. We used the imbedded graphics from Voyant along with RAWGraphs, Visme and Canvas for visual representation of our results. --- ## Results :fireworks: * Emotions * Sentiment * Linguistics --- ## Emotions #### - generated via the IBM NLU <iframe src="//my.visme.co/_embed/31qn9ome-untitled-project" width="800" height="482" style="border:0" webkitallowfullscreen="true" mozallowfullscreen="true" allowfullscreen=""></iframe><p style="width:220px !important;border-radius:3px !important;padding:3px !important;font-size:12px !important;font-family:Arial, sans-serif !important;color:#314152 !important;white-space:nowrap !important">Made with <a href="https://www.visme.co/make-infographics?utm_source=CTA&amp;utm_medium=Embed" target="_blank" style="font-weight:600 !important;text-decoration:none !important;font-size:12px !important;font-family:Arial, sans-serif !important;color:#314152 !important;white-space:nowrap !important">Visme Infographic Maker</a></p> --- ## Sentiment #### - generated via the IBM NLU Score from -1 to 1 (-1 being purely negative, 1 being purely positive) <div style="position: relative; width: 100%; height: 0; padding-top: 75.0000%; padding-bottom: 48px; box-shadow: 0 2px 8px 0 rgba(63,69,81,0.16); margin-top: 1.6em; margin-bottom: 0.9em; overflow: hidden; border-radius: 8px; will-change: transform;"> <iframe style="position: absolute; width: 100%; height: 100%; top: 0; left: 0; border: none; padding: 0;margin: 0;" src="https:&#x2F;&#x2F;www.canva.com&#x2F;design&#x2F;DAEJLBvPQQg&#x2F;view?embed"> </iframe> </div> --- ## Linguistics (vocabulary density) #### - generated via Voyant Tools & RAWGraphs Score from 0,00 to 1,00 based on amount of unique words (1 being only unique words). ![](https://i.imgur.com/e1ScIpx.png) --- ## Linguistics (Word Use #1) #### - use of the most common word across the speeches ("Virus") #### - Generated via Voyant Tools <iframe style='width: 740px; height: 387px;' src='https://voyant-tools.org/tool/Bubblelines/?stopList=keywords-35e565ca6b5057c6db78b417b3523959&bins=300&query=vi&query=det&query=er&query=og&query=virus&docId=69b3b1e70f10b3e64230cf5f1fdff57d&docId=c33af69eaeadded997b2199e508ed935&docId=8b58480e07de43c831bd95d7e6f1f2d1&docId=8b9f0ef15363a5cff5477cb16842fed0&docId=68da89751bdabcb56f69b16f9c6e4e1c&docId=1ce3b38ce6e83c16d1b9e65747660eba&corpus=a97bd82ce8bbba4c495fe19eeb967d14'></iframe> --- ## Linguistics (Word Use #2) #### - generated via Voyant Tools * Most frequent words in the corpus: > virus (40); health (17); new (16); coronavirus (15); time (15) * Distinctive words (compared to the rest of the corpus): > Angela Merkel: life (4), federal (5), show (3), say (3), save (3). >Boris Johnson: patrick (3), nhs (3), latest (3), chris (3), uk (2). >Donald Trump: americans (9), states (6), europe (5), treatments (4), financial (4). >Lee Hsien Loong: sars (7), new (12), singaporeans (4), rate (3), point (3). >Mette Frederiksen: hinanden (10), danmark (9), dag (8), brug (8), uger (7). >Scott Morrison: australians (10), australia (9), plan (5), million (3), health (6). --- ## Linguistics (Word Use #3) #### - generated via Voyant Tools * Most used (unique) word by speaker > Angela Merkel: gouvernment / virus (6) > Boris Johnson: outbreak (5) > Donald Trump: virus (14) > Lee Hsien Loong: virus / sars (9) > Mette Frederiksen: hinanden (eachother) / danmark (Denmark) (9) > Scott Morrisson: australians / australia (9) --- ## Trends and vague (mini-) conclusions In the results we see different nation leaders approach to responding and informing their countrymen of the Corona Virus - with very different results. * The USA and Singapore (especially) both had a high frequency of mentioning the virus (the actual illness) as well as similar viruses. Thus focusing more on the things to avoid, not the things to embrace. This seems to inflict a higher level of fear in their language. * The Danish (especielly) and Australian national leaders had more focus on their nationality and how they should act. Mette Frederiksen emphasized "togethernes" instead of the actual virus, urging the Danish citizens to do something together. Similarly, Scott Morrisson (AUS) focused on the Australian citizens and their government plan, thus producing the most positive speech of the corpus. * There seem to be no correlation between vocabulary density and speech sentiment or emotion - except for the two (English) speeches that scored the lowest on vocabulary density similarly scored significantly higher on the "sadness" emotional category in their speech. ## Ethical Considerations Due to the creation and purpose of our research data, we have not had many ethical considerations to include in our small research project. All the speeches are written, delivered and published with publicity in mind, therefore we are not showcasing or highlighting any data that was not available for free viewing on either government websites or public newspapers beforehand. In addition, the analysis and comparison are focused on the emotions, sentiments and linguistics of the speeches and does not highlight or showcase any personal data from the speeches or about the authors /or state leaders themselves.