During my literature review, I’ve identified tools and techniques to manage my postgraduate research project. I’ve learned that to have peace of mind when working on a project, just a little planning and preparation can be the key to success. In this article, I want to share with you some of the tools and techniques that I’ve explored and why I’ve decided to use them.
Firstly, let’s define a project. According to Wingate (2015), a project is a long-term activity that aims to achieve a certain goal usually accomplished by reaching smaller but significant milestones. However, not all completed projects can be regarded as successful. Typically, a successful project can be described as an achievement of a particular aim that has been completed within a defined scope, budget, and timeframe, including meeting or exceeding project requirements and stakeholders’ expectations (Wingate, 2015). Therefore, taking the responsibility for the activities involved in the project, such as planning, scheduling, and responding to risks and change, is unavoidable in both individual and collaborative projects (Wingate, 2015). However, there are circumstances that will inevitably slow down the progress, hence employing relevant tools and techniques to plan, assess and evaluate the project is an important aspect of any project management.
Defining the Project Scope
Conducting a research project in sentiment analysis is a multidisciplinary project that encompasses activities such as literature review, keyword selection, data collection, analysis and visualisation, computer programming, critical evaluation and demonstration of results. The final milestone of such project is to collate eight months of work into one thesis written using an academic writing style and research methods. Together, those activities can be regarded as the project scope, or what will be done within the project (Wingate, 2015). To define the project scope, I’ve conducted a SWOT analysis of my own strengths, weaknesses, opportunities and threats that I think will have most impact. Below is a sample of SWOT analysis:
Strengths:
- 5 years in web industry (experience delivering projects),
- Previous experience with Twitter API,
- Undergraduate degree (exposure to academic research),
- Working individually (independence).
Weaknesses:
- No prior experience with NLP (need to read and practice more),
- Working individually (more time needed to complete all tasks).
Opportunities
- Studying part-time and online (more flexibility),
- Access to digital libraries (wealth of academic journals and papers),
- Access to other grey literature (Udemy, DataCamp, YouTube, Medium).
Threats
- Working full-time (unexpected overtimes, company meetings),
- Physically active (risk of injury),
- Friends and family (other commitments).
The SWOT analysis gave me a better picture of my own strengths and weaknesses, as well as what are the opportunities to address those weaknesses and what are the threats that could affect the schedule of the project.
Flexible vs. Traditional Project Management
There are several approaches, such as flexible and traditional project management. According to Wingate (2015), a flexible project management approach is an iterative process that is more suitable for software projects because their requirements can change at any time during the project lifecycle, for example, due to the client involvement. In terms of sentiment analysis, an iterative process can be applied to activities that involve data, such as data collection, pre-processing, analysis and evaluation because those activities are unlikely to provide satisfactory results and may need to be repeated for as long as there is enough time allocated to those activities. On the other hand, traditional approaches are more rigid and sequential. Therefore, they are less suitable for projects that might change at any point during the project lifecycle (Wingate, 2015).
Setting up the Goals
According to Boogaard (2021), to provide a sense of direction in the research project, it is important to take a step back and think about the goals of the project. A popular approach is to define goals that are specific, measurable, attainable, realistic and timely, or SMART (Boogaard, 2021). An example of SMART goals is shown below:
Specific: The student need to collect and pre-process Twitter data to conduct a geospatial sentiment analysis about military, medical and financial aid sent to Ukraine by the UK government.
Measurable: Collect Twitter data related to the topic of the research that can be used to conduct the sentiment analysis and visualise the results on the map of the UK.
Achievable: Submit a completed thesis on time and within the scope of the project. Include project deliverables such as infographic.
Relevant: Social media platforms like Twitter provide a wealth of information about people’s opinions, therefore studying them provides an opportunity to understand the public perception of the Russo-Ukrainian conflict.
Time-bound: The research project must be completed two weeks before the final submission to allow for any last minute changes or unexpected delays.
Conducting an academic research is a sequential process with many mutually dependent tasks that can be tracked using a project tracker. That being said, the waterfall method offered by the traditional approaches can be useful at visualising the project scope, including all the activities, the time allocated to complete them, and the key milestones that need to be met to achieve SMART goals. The waterfall can be regarded as a project baseline outlining what will be done throughout the project. The easiest way to create a waterfall is a simple spreadsheet (Phelps et al., 2007).
Assessing and Managing the Risk
The risk can be described as an obstacle that might get in a way to successfully complete the project (Wingate, 2015). The earlier SWOT analysis identified threats and weaknesses that could potentially affect the scheduling and management, however, the lack of technical skills and knowledge might be the biggest obstacle when conducting the sentiment analysis. Moreover, unavailability or insufficient amount of data could also constitute as a risk. The collection and analysis of user-generated content puts a significant pressure to adhere with the best practices to keep the data secure and inline with the Twitter developer policies.
To manage risks, a simple project risk register can be created using spreadsheets. The risk register is a list of known risks, categories, causes, effects and their probability, impact and priority expressed on a numerical scale. Moreover, the risk mitigation strategy can be applied to determine the risk treatment and help to identify what triggered the risk in the first place (Wingate, 2015). The risk assessment is an ongoing activity, as there are many unknown risks that could become more apparent as the project matures.
Conclusion
In summary, this article addressed some of the tools and techniques that I’ve applied to manage my own research project. The SWOT analysis helped me to define the scope and identify parts of the project that are likely to need more attention, whereas the SMART goals helped me narrow down the goals of the project. I found the traditional approach more appropriate for the academic research, especially the waterfall, or GANTT chart, as a useful technique to visualise all expected activities on an easy to follow timeline. However, I’ve integrated spiral approaches to all activities that involve sentiment analysis.
References
Boogaard, K. (2021). How to write SMART goals. [online] Work Life by Atlassian. Available at: https://www.atlassian.com/blog/productivity/how-to-write-smart-goals [Accessed 23 Jul. 2022].
Phelps, R., Fisher, K. and Ellis, A. (2007). Organizing and Managing Your Research Project, a Practical Guide for Postgraduates. SAGE Publications Ltd.
Wingate, L.M. (2015). Project Management for Research and Development, Guiding Innovation for Positive R&D Outcomes. CRC Press.
