 This is a short summary of the strategies and principles I've found to be effective for upskilling within academia. There are three main principles I'll explain each and give examples. If you're going to dedicate time and effort to learning something new, it should ideally be an evergreen skill. This means it's not something you'll learn for the purpose of one study or project and then never use again. It should be something that will be continually useful moving forward in your career. That helps me decide what to focus on or to prioritize possible skills to learn. Ideally, this also shouldn't be a standalone skill. It should be something that compounds with other skills you've learned previously or that you'll go on to learn in the future. This enables the development of a combined skill set in which the overall benefits or capabilities are greater than the sum of the individual parts. If strategically targeted, each new skill can enable you to get more out of other skills that you already have. While you're developing a set of evergreen and compounding skills, it's well worth taking the extra time to automate anything possible or to set up templates that can be reused in the future. Even if it's quicker to do something manually, consider the benefits of automation when you repeat this evergreen skill throughout your career, especially when your future self is likely to be using the skill under increasing time pressure or trying to compound additional skills on top of it. But how do you go about learning new skills? This quote sums it up well. You don't have to be great to start, but you do have to start to be great. For me, upskilling is all about breaking things down into the smallest possible tasks or sub skills. What's the very first step you need to take and what do you need to do or learn in order to actually take that first step? With all of the tutorials on my channel, there are people who know far more about the topics than I do. So I believe the reason some of them have helped so many people is because of the simplistic approach and focus. I try to give people just the necessary first steps to overcome that first hurdle and start to figure things out for themselves with no assumed prior knowledge. It requires splitting things into a series of very small tasks or bite sized chunks. However, it does require a continual effort to improve in these small steps throughout every project. So it's worth exploring what systems you can put in place to encourage you to do this. Booking out time in your calendar might work for you or you might need to do more. For example, one of the main benefits of my monthly best things I've read this month newsletter is that it forces me to read a range of topics so that I've got something to write about each month. The same goes for sharing useful information on social media. And one of the main benefits of creating a YouTube video is that it forces me to break down a subject and make sure I can explain it from the foundations upwards. Another strategy I use is to learn and apply one new skill for each first author study. This should be a skill that is evergreen, compounds with other skills and can eventually be automated. For example, my most recent study gave me an opportunity to learn to simulate data sets in R. This is something I'd wanted to learn for a while that hadn't had the time or opportunity. This is an evergreen skill that I will now use in some future studies. It links to previous skills, such as the ability to perform an a priori power analysis, first learned for a different study. The two skills combined enable a power analysis to be performed using simulated data. And it's done in a way that enables me to quickly reuse and adapt the code in future studies without the initial time investment. A second example is my experience of upskilling in data visualization. One of the best tips I've seen for creating figures is to Google the plot type and find the code for the one you prefer. That's exactly what I did for violin plots in R, but I needed an excuse to try it out. Before implementing it in a study, I used the weekly tidy Tuesday challenge to try out some plot types on example data and learn from others code over a couple of weeks. I then implemented it within one study, forming an evergreen skill that I was able to build upon in another study. It can now link in with the previous examples to plot simulated data in a more insightful manner and has resulted in code that can automate the process for future data sets. As a final example, I'll talk about learning to code. The first way I learned to code was by taking somebody else's script and changing one thing at a time to see what happens or by trying to get someone else's code to run on my new data. The second way was by googling everything. This forces you to split the task into the smallest possible subtasks and it's much easier to learn that one subtask than an entire coding language. You could read a textbook cover to cover or you could Google import data from Excel to MATLAB. Even with my typing error, the first result that comes up shows how to do it both manually and through code. If you do this manually, you'll have to do it for all trials by all participants, redo it if anything changes and start from the beginning with each new study. But if you automate it through code, you can save time in the future. You can also repeat the process for multiple data files and so that might be your next small step to search for. You'll figure out what works for you, but I found that it helped me greatly to identify one small learning or upskilling opportunity within each project I work on and to then split that process into the smallest subtasks possible. If I do it well, that should provide me with a skill set that remains useful throughout my research career, can be added to by future skills and can save me time and effort in the long run. If you're looking to learn a new skill, check out the videos in my tutorials playlist. We'll see the individual videos linked in the description below.