<br><br><br> .center[.title[.large[Building Productive Habits <br> for Research Excellence]]] .sticker-float[] .bottom[ ## Bahman Rostami-Tabar <br> ] --- background-image: url("resources/hierarchy-left.jpeg") background-size: contain background-position: left class: middle .pull-right2[ ## Outline - Research questions - Forme - Fond - Effective and productive habits ] --- background-image: url("resources/hierarchy-left.jpeg") background-size: contain background-position: left class: middle .pull-right2[ ## Outline - .remember[Research questions] - .remember[Forme] - .remember[Fond] - Effective and productive habits ] --- ## Research questions <br> <br> - .remember[Constructing and formulating research questions] is one of the most, perhaps the .remember[most, critical] aspects of all research. - .remember[Good research questions], however, do not just exist they also .remember[need to be created and formulated]. --- ## What makes a question a good Research Question? .center[ <img src="figure/phdworkshop/menti-rq.png" width ="500px" > ] --- ## What makes a question a good Research Question? ### Very short answer - a question that can be researched, and answered - he key idea is that evidence must exist (or be possible to collect) that could provide a believable answer. - improve our understanding of how the world works - give us a reason for what we see in the world. Inform theory: A theory is just that—an explanation of why things are the way they are, whether big (like evolution) or small (like why bread prices rise). --- ## What makes a question a good Research Question? <br> <br> - .remember[Forme] relates to the structure, wording, and clarity of the research question. - .remember[Fond] pertains to the essence or substance of the question—its significance, relevance, and the intellectual weight it carries in contributing to knowledge or solving a problem. --- ## Forme - Clear definition & precision in wording - Focused scope - Researchable constructs - Neutrality - Relevance to audience --- ## Clear definition & precision in wording Explanation: Avoids vague, subjective, or undefined terms that could lead to multiple interpretations. -- ❌ How does AI improve healthcare? ("Improve" is too vague—what aspect of healthcare? For whom?) -- ✅ How does AI-assisted radiology affect diagnostic accuracy in lung cancer screening?" (Uses specific terms like "AI-assisted radiology" and "diagnostic accuracy" within a defined context.) --- ## Focused scope The research question should not be too broad or too narrow, ensuring it is researchable -- ❌ How does AI transform global healthcare? (Too broad—"transform" is vague, and "global healthcare" covers too many contexts.) -- ✅ How do AI-powered virtual assistants affect patient engagement in chronic disease management? (Specific technology, targeted effect, and a clear healthcare focus.) --- # Researchable constructs The question should contain terms that can be measured, analyzed, or observed within a research design. -- ❌ How does AI revolutionize patient care? ("Revolutionize" is too abstract—how do you measure it?) -- ✅ What factors influence the accuracy, speed, and user satisfaction of AI-based triage systems in emergency care settings? (Clear measurable factors—"effectiveness" can be studied through accuracy, speed, or user satisfaction.) --- ## Neutrality A research question should not assume an outcome or imply bias. -- ❌ Why is AI better than human doctors in diagnosing diseases? (Loaded with bias—assumes AI is superior.) -- ✅ What are the strengths and limitations of AI-driven decision-making in healthcare diagnostics? (Encourages an objective assessment of both positives and negatives.) --- ## Relevance to audience The question should address a meaningful problem that is significant to its intended academic, professional, or societal audience. -- ❌ Do doctors like AI in hospitals? (Too trivial—"like" is subjective, and the focus isn’t meaningful for decision-making.) -- ✅ How do AI-driven predictive models impact treatment plans in oncology? (Addresses a key issue in cancer treatment—highly relevant to medical professionals and researchers.) --- class: center middle ## Fond 1. Fundamental Significance 2. Scientific Contribution 3. Relevance to society 4. Research feasibility --- ## Fundamental Significance It examines how deeply rooted the research question is in fundamental principles. .small[ - Is this a question that persist(ed) over time and will be relevant over time? Does the question has some value over time? - Will answering this question lead to deeper insights into how something fundamentally works? - Could the findings challenge existing theories, assumptions, and understanding of the problem? - Is the question relevant to multiple disciplines? - Does it have broad implications beyond its immediate context making them relevant to other contexts, populations, or situations? ] --- ## Scientific Contribution It focuses on the research's ability to advance scientific knowledge or understanding. .small[ - How does this research contribute to advancing the field? - Does it fill a gap in the existing literature? - What new knowledge, methods, or theories might emerge from this work? - Will the findings change how the research audience view a particular problem or concept? - Can the findings establish new principles in the field? - How answering the question influence future research directions? - Are there leading researchers, conferences, or journals in the field that consider this topic important? ] --- ## Relevance to society It assesses how the research addresses pressing societal needs or challenges. .small[ - What real-world problem does this research address? - Does the research aim to solve a pressing issue or address a recognized challenge in society? - Does the research contribute to grand challenges such as those highlighted on Doughnut Economy or UN Sustainable Development Goals? - Who benefits from the results? Is there a current demand for solutions in this area? - Is the topic discussed in mainstream media, public forums, or social media? - Will the research produce findings that could be implemented in real-world settings (e.g., new technologies, policies, or practices)?] --- ## Research feasibility It assesses whether the research is practical within the constraints of time, resources, and available expertise. .small[ - Does the research question involve exploring uncharted territory (e.g., novel theories, poorly understood phenomena) which make take years to complete? - Is the research question realistic given the available time frame? - Does the project involve long-term experiments, data collection, or analysis that could extend beyond this period? - Are the necessary data, resources, or samples for this research readily available? - Are there clear and reliable ways to obtain data (e.g., fieldwork, public databases, or industry partnerships)? - Will you need permissions or ethical approvals that could take time to secure? - Do you have access to the necessary equipment, software, and resources to complete this research? ] --- ## Research topic evaluation - framework .center[ <img src="figure/phdworkshop/topic.png" width ="550px" > ] --- ## Research topic evaluation - examples .center[ <img src="figure/phdworkshop/topics-exmaples.png" width ="550px" > ] --- background-image: url("resources/hierarchy-left.jpeg") background-size: contain background-position: left class: middle .pull-right2[ ## Outline - Research questions - Forme - Fond - .remember[Effective and productive habits] ] --- ## Citing & reference management .pull-left[ ### [Zotero](https://www.zotero.org/) Set up account at www.zotero.org Download from mendeley.com <br> <img src="figure/phdworkshop/zotero.png" width ="500px" > ] .pull-right[ ### [Zotero Connector](https://chromewebstore.google.com/detail/zotero-connector/ekhagklcjbdpajgpjgmbionohlpdbjgc?pli=1) <img src="figure/phdworkshop/zotero_addin.png" width ="500px" > ] --- ## What to cite? Audience and prior research - Your research question is part of a much bigger puzzle - Identifying the audience talking about the puzzle - Reviewing prior research around the puzzle - Positioning your new research - the novel aspects of your research, such as a new perspective, a different foundation, an overlooked set of arguments, or new data and methods --- ## Reproducibility .center[ <img src="figure/phdworkshop/menti-rq.png" width ="500px" > ] --- ## Reproducibility .pull-left[ ❌ Data edited in a spreadsheet ❌ Data and analysis colour coded ❌ Click and point analysis ❌ Copy and paste graphs , results, and tables ❌ Tables typed by hand ] .pull-right[ ✅ All data edits scripted ✅ All analysis scripted ✅ Graphs and tables automatically pulled in to the thesis or papers ✅ Tables generated with scripts or read from csv/xls files ] --- ## Reproducibility Someone (or your future self) should be able to reproduce your work without having to guess what software you had installed, what versions, which files do what, etc. - Stay organized. - One system for doing this using R or python or both is to write your work in an Rmarkdown or Quarto file. - Track software versions --- ## Version control - thesis_v1, thesis_v2, etc., is not adequate version control. - You need to track changes over time, have a remote repository, and be able to roll back as required. - Your repository should contain everything required to produce your thesis including computer code, references, writing. - Your repository should have an obvious structure and be fully documented. - .remember[Github] solves these problems - Read “Happy git with R”: happygitwithr.com --- ## Version control with git - Git and github can be used with any text-based language including Python, LaTeX, R, Rmarkdown, Quarto, markdown, etc. - Git allows you to: - track changes - experiment in branches - undo - Github provides: - backup and restore - synchronization --- ## Transparency & ethics - Always conduct research and present findings .remember[truthfully], without fabrication, falsification, or selective reporting. - Be transparent about your research methodologies, including the tools, techniques, and processes you use, .remember[Embrace reproducibility] - Be accountable for your research and ensure that others can reproduce your results. - Ensure that research involving human subjects, or sensitive data is conducted ethically, with respect for privacy and rights. - Give proper acknowledgment and credit to all contributors, collaborators, and prior research. --- ## Writing - How to write alot? .pull-left[ - Create a writing schedule – and stick to it - Turn off distractions. Do not meet anyone, or do not allow yourself to be distracted - Set concrete, measurable writing goals - Separate writing from editing - Prioritise projects with due dates (paper revision, proofing, grant writing, etc) - You don’t need to wait for inspiration, you need to sit down and write - Monitor your writing progress ] .pull-right[ .center[<img src="figure/phdworkshop/howtowrite.jpeg" width ="400px" >] ] --- ## Writing ### Microsoft word ❌ Impossible to do proper version control ❌ Poor formatting of equations ❌ Impossible to make reproducible ❌ More difficult to automate cross-referencing, numbering ❌ Files more likely to be corrupted ❌ With One Drive , even worse --- ## Writing ### Quarto ✅ Combines R, Python , plain text and LaTeX into one system, reference management ✅ One tool to do all , writing, analysis, version control, file sharing, etc ✅ Render files to pdf, word, html, and latex formats ✅ Reproducible research ✅ See quarto.org for help --- ## Writing ### Overleaf - Real-time collaboration - Simplified LaTeX editing - Wide range of templates - Automatic version control --- ## Presentation .pull-left[ ### Creating - Create slides using Quarto/Rmarkdown in R or Python or use Beamer in latex - Tailor the content to the audience - Tell a story and follow a logical structure from problem, to questions, to method to finding - Use visuals (images, diagrams, etc) effectively, and reduce the text - Include a strong conclusion: key takeaways, Highlight contributions, Future work and next Steps ] -- .pull-right[ ### Delivering - Practice, Practice, Practice - You need to present in research seminar before presenting it outside, in conferences, etc - Before each presentation, you need to practice it several times - you can record it and watch it later or practice in front of a mirror - Engage the audience - Anticipate questions ] --- ## AI tools to support reserach .pull-left[ - Search engine - https://consensus.app/ - Literature review - https://www.connectedpapers.com/ - https://www.researchrabbit.ai/ - https://inciteful.xyz/ - https://scispace.com/ ] .pull-right[ - Writing Assistance - ChatGPT (by OpenAI) - Claude: https://claude.ai/new - https://notebooklm.google.com/ - Turn text to viual - https://www.napkin.ai/ ] --- ## Critique A critique is an evaluation of your work that highlights both strengths and areas for improvement. It helps .remember[identify gaps or weaknesses] you might have .remember[missed]. Build habits to embrace critique, why? -- - A willingness to learn and evolve, - Encourages intellectual growth and skill development, - Pushes you to think critically, - Improve your arguments, - Strengthen your work, - Making you a better researcher and individual. --- ## Handling critique .pull-left[ ### Presentation - Listen actively and stay .remember[open-minded] - Engage with the audience, .remember[don’t rush to defend] - Take Notes - Ask clarifying questions - Reflect and prioritize - .remember[Separate] personal .remember[feelings] from professional .remember[feedback] ] -- .pull-right[ ### Paper - Critique is about your work - Take time to read the feedback - Break down the feedback into categories—e.g., structure, argument clarity, or methodological issues - Acknowledge valid points, but stand your ground when necessary - Stay patient and view each revision as an opportunity to strengthen your work. - Stay professional and grateful - Use critique for future papers ] --- ## File organisation-Use cloud ### OneDrive (for forms, admin, and file sharing) Centralized storage for administrative documents: Use OneDrive to keep all important forms, official documents, and administrative files organized in one place. ### GitHub Efficient version control and collaborative research management: GitHub allows you to systematically track changes in your research projects, including code, data analysis scripts, journal correspondence, and written documents using Quarto --- ## File organisation .pull-left[ <img src="figure/phdworkshop/file_organise.jpg" width ="650px" > ] .pull-right[ - Keep your desk clean - Keep desktop and downloads folders empty in your computer - Use clear, descriptive file and folder Names - Regularly review and delete your files, if not necessary. ] --- ## File organisation .pull-left[ - Example structure: - administration - teaching - funding - award - training - temporary_files ] .pull-right[ - research - data - codes - papers - image - result - correspondence - presentation ] --- ## Email management .pull-left[ ### Stay organised - Prioritize emails - Use categories or labels - Create folders for organization - Use flags for follow-up - Practice inbox zero - Use auto-reply when you're away ] -- .pull-right[ ### Create good habits - Set specific times for checking emails - Don't start your day with emails - When writing emails, keep them brief and to the point - Respond in a timely manner - Never respond to difficult emails immediately - Do not install work email on your phone ] --- ## Work-life balance - Set boundaries and stick to a schedule - Create a consistent daily or weekly work schedule - Avoid emails or work-related messages outside of these hours - Practice deep work - Plan and prioritize tasks - focus on high-priority tasks and avoid overcommitting - Prioritize completing essential tasks within your set work hours - Set aside time for family, friends, hobbies and socializing - Schedule time to completely unplug from digital devices, especially those that connect you to work (laptops, emails, messaging apps) - Plan holidays in advance --- ## References 1. Alvesson, Mats, and Jorgen Sandberg. [Constructing research questions: Doing interesting research. (2013): 1-152.](https://methods.sagepub.com/book/mono/constructing-research-questions/toc) 2. Dorobantu, Sinziana, et al. ["]The AMJ management research canvas: A tool for conducting and reporting empirical research.](https://journals.aom.org/doi/pdf/10.5465/amj.2024.4005) Academy of Management Journal 67.5 (2024): 1163-1174. 3. Dorobantu, Sinziana, et al. [The AMJ management research canvas: A tool for conducting and reporting empirical research.](https://psycnet.apa.org/record/2011-15936-001) Academy of Management Journal 67.5 (2024): 1163-1174. 4. Silvia, Paul J. How to write a lot: A practical guide to productive academic writing. American Psychological Association, 2018. --- class: middle, center ## Slides and contact slides: https://www.bahmanrt.com/talks/ Bahman Rostami-Tabar- rostami-tabarb@cardiff.ac.uk