 Hi everyone, it's Monica Wahee here with some data science career advice for you. If you clicked on this video because you are thinking of starting the journey towards a career in data science, then perfect. Because this is the first video you should watch if that's what you are considering. And the reason why I say this is the first video you should watch is that the advice I give you now is totally conceptual and will help you map out your whole journey from beginning to end in a very big picture way. It's like planning your trip to a data science career in the beginning by just looking at the big highways you will drive on, not the little side streets you will take. Since I'm a LinkedIn Learning author of data science courses, a lot of new learners in data science connect with me on LinkedIn and write me messages asking my advice, especially my advice about what exactly to learn in data science, like what software to learn, what courses to take and in what order to do it. I realize that people need guidance not only on their topics of study but what order they should study them in. So in response, I wrote a blog post I titled healthcare data science newbie do it yourself starter kit because I'm in health data science so I made it specific to that but really it applies to deciding to learn data science in any field. I'll link you to it in the description but I thought I'd make this video to sum up more generically what I said in that blog post and that is how you chart the fastest course from learning data science to getting a job in data science. When I start out coaching a learner I start in a different place than most people rather than starting with the software and the data I start with the person. I realize there are roughly two kinds of people who are starting to learn data science. Those who have already had a career and those who are new to the workplace. Let me be clear about what I mean. Some people come to data science after already participating in some sort of career. Maybe they were a healthcare clinician but maybe they were a web designer or a marketing director. They could come from any field and they didn't even have to do that job very long but those are people who have already had a career even a short one and worked at some other job before coming to data science. The rest of the people I see coming to data science are completely new to the workplace. They don't have any experience working anywhere before and they want to train for a job in data science. The first job they hope to get is a data scientist job. Okay, so from a career and a data science standpoint what's the difference between these two learners? Exactly, the person who has already been in the workplace has subject matter expertise. I put a healthcare clinician on the slide as an example. She already knows the terminology in healthcare. She knows what kind of data are held in medical records and what lab pronounce look like and what terminology is used. ICD-10 codes and differential diagnoses and all this jargon. So she's already ahead of the game. So those of you who feel like you might be on the right side of the slide, learners who have never worked in a workplace before you really do have a lot of catching up to do. So while my advice is roughly the same for both of these groups, how you actually go about following the advice might be different. But the goal is the same. The goal is to follow a learning pathway that, at the end, makes it so you can get hired into the new job of data scientist, whether it's your first job or your 100th job. But you don't want to fall into the trap of being that person who takes every data science course available but doesn't actually know how to apply any of that knowledge in any field. Taking data science courses is a good place to start and I teach them so I endorse them. But they should fall in the context of an overall learning pathway that leads you to the final goal of getting a data science job because you look useful to a data science manager. So as like a rainbow umbrella overarching my code newbie data science do-it-yourself toolkit is a simple intuitive four step learning pathway. This four step framework that you can follow to make sure you aren't taking too many courses or learning the wrong things or otherwise doing something stupid along the way on your data science journey to the knowledge that should land you a job. And here's the first step which I'm pretty sure will surprise you. The first step is to actually pick a field in which you want to be a data scientist. That's right. You pick the field first, not last. Now, if you were that healthcare clinician on my last slide or you are someone who has already done a career or two if you can still stand that career it is smartest to become a data scientist in that field. But if you are a person who has never had a job before this first step can literally be the toughest. Still, I have three pieces of advice for you that will probably work well enough to get you going down a path that's mostly right for you. My first piece of advice is to take non-data courses in the topic. For example, if you think you want to be a FinTech data scientist take a few courses in accounting because that's who you are going to be working with accountants and their issues and their data. So if you think you like any topic astronomy, car design, weather, whatever if you take courses in the topic and you hate it then you know not to be a data scientist in it. In my blog post since I'm a healthcare subject matter expert I point out that if you are interested in becoming a healthcare data scientist and you don't have any background in healthcare you can start by watching my lecture series on the US healthcare system. I'll link you to it in the description to this video. My next advice is to talk to actual data scientists in the field you want to work in. Many learners have told me this is hard like they contact data scientists at workplaces they are thinking about and those people don't want to talk to them. First let me defend them by telling you honestly that many of these so-called data science or startup workplace environments are like way messed up. Bad management, total chaos and so on. So maybe that's why they don't want to talk to you. It might not be you they just might not know what to say to you. Well I talk to everyone who contacts me or I try to. They usually contact me unlinked in. I think data scientists active on social media are more likely to talk to you. So that's my advice. Find ones who are active on social media that attracts data scientists like LinkedIn or maybe Stack Overflow or other sites data scientists posts on and contact those for advice. Finally, even if you've never been in a workplace you might have total subject matter expertise if you are way deep into a hobby or interest. I've been contacted by many sports fans who are interested in sports analytics for example and sports data are readily available to play with. Those of you thinking of marketing could just start by analyzing social media data. For example, Twitter has an analytics portal but also you can use software like the R package Twitter to analyze the data. I personally am interested in casinos so I made an example for you. I downloaded some casino data from the web and analyzed it mainly because I wanted the casinos where I live in Massachusetts to improve their restaurants and I wanted them to hire me to give them advice. Spoiler, it hasn't happened yet. Hope springs eternal. But since I go to the casino I know what customers look for in casinos so I could do a reasonable analysis with public data. I'll link you to it in the description. Oh, and by the way if you are wondering what software to start your learning with you need to first know what field you are in. That will determine what software to start learning. Okay, now that you have chosen the field in which you wanna be a data scientist the next step is learning how to make measurements in that field. What do I mean by measurements? Well, I mean data. There are two kinds of data. Data that already exist and data that you have to collect. If you want to get good at understanding data that already exist, curate them. What? You say you don't know what data curation is? No problem. You can take my data curation course on LinkedIn Learning. I'll link you to my LinkedIn Learning course list in the video description. I curated that casino data as soon as I found it on the web. That's how I learned something about how casinos do measurements. But let's say you need to understand data you will have to collect yourself like through a survey or maybe you will abstract data which is collecting data from data. Don't know what I'm talking about again? Well, you can start by taking my free online course in data collection. I give you an overview of how to do measurement using research forums, how to do data abstraction, how to set up data entry from paper forms, how to arrange to do online surveys, how to deal with instruments and how to write items. There's really a lot to learn about data collection. Okay, now that you've picked a subject matter and you've explored what data and measurement look like in that field like I did with the casino example, now you are ready to work with the data and measurements to try to make sense out of them. Now you are finally ready to learn how to design studies using data and measurements from that field. In my LinkedIn learning courses on study design with existing data, I teach you a set of study design tools you can apply to any topic. I also have a course in experimental design. A lot of people in data science jump into learning programming, but rarely do they realize early on that they have to also learn how to put all that programming knowledge together into a study design that leads to an analysis that produces knowledge at the end. Of course, I'll be happy to teach you how to program as well. You can learn R and SAS, which we use in the healthcare field by taking my courses on LinkedIn learning. My courses are different from other courses in that they walk you through doing an actual project on your own. I actually base those courses in R and SAS on what I normally do when I provide one-on-one coaching on Skype. We meet once a week or once every two weeks, usually for one or two hours. We start with a plan we set up and then each week, I give you homework to do in between our meetings. Then when we meet, I help you through any challenges and we plan your work for next week. That's how it all gets done. Okay, now you made it past step three. You picked a field, learned about measurement and data in the field, and learned study design concepts so you can do studies with data in that field. What's the last thing you need? You need to learn statistics and how to apply them to the field. Doesn't that sound backward? You actually learn how to do the statistics last. But those of you who already know statistics realize that you first need subject matter expertise before you can ask a reasonable question in any field that you can answer with statistics. My R and SAS courses teach statistics, but if you're just starting out, that might be too advanced for you. If you are new to statistics, you'll want to start with my YouTube playlist of lectures I used to give to my undergraduate nursing class. They have received excellent reviews from data scientists just starting out in the field. I'll link you to the playlist in the description. So that is my conceptual learning pathway for going from whatever you are right now to a hireable data scientist. Step one, you choose your subject matter, study it a little, and learn the terminology. Next, step two, you learn about measurement and data in the field. Step three is where you learn study design so you can design studies in that field. And finally, step four is when you learn statistics for that field and apply them to studies you design with the data from that field. Aren't you glad that this is the first video you watched to determine your pathway to going from whatever you are right now to having a job as a data scientist? Learners often start in data science getting tangled up in the side streets. They worry about what courses to take, what software to learn, and who to listen to. I'm giving you a big picture learning pathway that will help you stay out of dead ends and cul-de-sacs. If you pick your subject matter first, it is easier for you to target a job in that field. Then you can seek to arm yourself with the right software and practice applying study designs and statistics to data in that actual field. So you are ready to do a job in that field when the time comes. I hope you found this advice helpful. Thanks for watching and good luck on your data science journey.