01:25 - Teaching, Learning, and Education
06:16 - Becoming a Data Scientist
- Opportunities to Create New Knowledge
- Data Science Science
19:36 - Solving Bias in Data Science
23:36 - Recommendations for Aspiring Data Scientists
- Hire a Career Coach
- Creating and Maintaining a Portfolio * Make a Rosetta Stone
- Spend $$$/Invest on Transition
- Bet On Yourself
45:36 - Impostor Syndrome
- Immunity Boosts
- Know Your Baseline
- Disseminate Knowledge
- Confidence Leads to Confidence
- Dunning-Kruger Effect
- Johari Window
Mae: Checking out the metrics resources on Impostor Syndrome listed above.
Casey: Writing about software in a positive, constructive tone.
Mando: Investing in yourself. from:sheaserrano bet on yourself
Adam: Talking about career, data science, and programming in a non-technical way. Also, Twitter searches for book names!
To make a one-time donation so that we can continue to bring you more content and transcripts like this, please do so at paypal.me/devreps. You will also get an invitation to our Slack community this way as well.
MANDO: Good afternoon, everyone! Welcome to Greater Than Code. This is Episode number 241. I'm Mando Escamilla and I'm here with my friend, Mae Beale.
MANDO: Hi, there! And I am also here with Casey Watts.
CASEY: Hi, I am Casey! And we're all here with Adam Ross Nelson, our guest today.
ADAM: Hi, everyone! Thank you so much for having me. I'm so glad to be here.
CASEY: Since 2020, Adam is a consultant who provides research, data science, machine learning, and data governance services. Previously, he was the inaugural data scientist at The Common Application which provides undergraduate college application platforms for institutions around the world. He holds a PhD from The University of Wisconsin: Madison in Educational Leadership & Policy Analysis.
Adam is also formerly an attorney with a history of working in higher education, teaching all ages, and educational administration. He is passionate about connecting with other data professionals in-person and online. For more information and background look for his insights by connecting with Adam on LinkedIn, Medium, and other online platforms.
We are lucky we have him here today. So Adam, what is your superpower and how did you acquire it?
ADAM: I spent so much time thinking about this question, I really wasn't sure what to say. I hadn't thought about my superpower in a serious way in a very long time and I was tempted to go whimsy with this, but I got input from my crowd and my tribe and where I landed was teaching, learning, and education.
You might look at my background with a PhD in education, leadership, and policy analysis, all of my work in education administration, higher education administration, and teaching and just conclude that was how I acquired the superpower. But I think that superpower goes back much further and much deeper.
So when I was a kid, I was badly dyslexic. Imagine going through life and you can't even tell the difference between a lowercase B and a lowercase D. Indistinguishable to me. Also, I had trouble with left and right. I didn't know if someone told me turn left here, I'd be lucky to go – I had a 50/50 chance of going in the right direction, basically. Lowercase P and Q were difficult.
For this podcast, the greater than sign, I died in the math unit, or I could have died in the math unit when we were learning greater than, or less than. Well, and then another one was capital E and the number 3, couldn't tell a difference. Capital E and number 3. I slowly developed mnemonics in order to learn these things. So for me, the greater than, less than pneumonic is, I don't know if you ever think about it, but think of the greater than, or less than sign as an alligator and it's hungry. So it's always going to eat the bigger number. [laughs] It’s always going to eat the bigger quantity. So once I figured that mnemonic out and a bunch of other mnemonics, I started doing a little bit better.
My high school principal told my parents that I would be lucky to graduate high school and there's all kinds. We can unpack that for days, but.
ADAM: Right? Like what kind of high school principal says that to anybody, which resonates with me now in hindsight, because everything we know about student learning, the two most influential factors on a student's ability to learn are two things. One, teacher effectiveness and number two, principal leadership. Scholarship always bears out.
ADAM: Yeah. So the principal told my family that and also, my household growing up, I was an only child. We were a very poor household; low income was an understatement. So my disadvantages aside, learning and teaching myself was basically all I had. I was the kid who grew up in this neighborhood, I had some friends in the neighborhood, and I was always exploring adjacent areas of the neighborhoods. I was in a semi-rural area. So there were wooded areas, there were some streams, some rivers, some lakes and I was always the kid that found something new. I found a new trail, a new street, a new whatever and I would run back to my neighborhood and I'd be like, “Hey everybody, I just found something. Look what I found, follow me and I will show you also. I will show you the way and I'll show you how cool that is.”
ADAM: I love this thinking. [laughs]
MAE: I love that!
ADAM: I'm glad because when I'm in the classroom, when I'm teaching – I do a lot of corporate training now, too. When I'm either teaching in a traditional university classroom, or in corporate setting, that is me reliving my childhood playtime. It's like, “Hey everybody, look at this cool thing that I have to show you and now I'm going to show it to you, also.”
So teaching, learning, and education is my superpower and in one way, that's manifested. When I finished school, I finished my PhD at 37. I wasn't 40 years old yet, if you count kindergarten had been in school for 23 years. Over half of my life, not half of my adult life, half of my entire life I was in school [chuckles] and now that I'm rounding 41—that was last week, I turned 41. Now that I'm rounding 41 –
MAE: Happy birthday!
ADAM: Thank you so much. Now that I'm rounding 41, I'm finally a little more than half of my life not in school.
MANDO: Congrats, man. That's an accomplishment. [laughs]
So I'm curious to know how you transitioned from that academic world into being a data scientist proper, like what got you to that point? What sets you down that path? Just that whole story. I think that'd be super interesting to talk about and dig into.
ADAM: Sure. I think context really matters; what was going on in the data science field at the time I finished the PhD. I finished that PhD in 2017. So in 2017, that was that the apex of – well, I don't know if it was, or maybe we're now at the apex. I don't know exactly where the apex was, or is, or will be, but there was a lot of excitement around data science as a field and as a career in about 3, or 4 years ago.
MANDO: For sure.
ADAM: So when I was finishing the PhD, I had the opportunity to tech up in my PhD program and gain a lot of the skills that others might have gained via other paths through more traditional computer science degrees, economics degrees, or bootcamps, or both. And then I was also in a position where I was probably—and this is common for folks with a PhD—probably one of the handful of people in the world who were a subject matter expert in a particular topic, but also, I had the technical skills to be a data scientist.
So there was an organization, The Common Application from the introduction, that was looking for a data scientist who needed domain knowledge in the area that I had my PhD and that's what a PhD does for you is it gives you this really intense level of knowledge in a really small area [chuckles] and then the technical skills.
That's how I transitioned into being a data scientist. I think in general, that is the template for many folks who have become a data scientist. Especially if you go back 3, or 4, or 5, or 6 years ago, before formal data science training programs started popping up and even before, and then I think some of the earliest bootcamps for data science were about 10 years ago. At least the most widely popular ones were about 10 years ago to be clear. And then there's another view that that's just when we started calling it data science because the skills for – all of the technologies and analytical techniques we're using, not all of them, many of them have been around for decades. So that's important to keep in mind.
So I think to answer your question, I was in the right place at the right time, there was a little bit of luck involved, and I always try and hold myself from fully giving all the credit away to luck because that's something. Well, maybe we'll talk about it later when it comes to imposter syndrome, that's one of the symptoms, so to speak, of imposter syndrome is giving credit for your success away to luck while you credit the success of others to skill, or ability.
But let me talk about that template. So the template is many data scientists become a data scientists with this three-step process. One, you establish yourself as an expert in your current role and by establishing yourself as an expert, you're the top expert, or one of very, very few people who are very, very skilled in that area. Then you start tackling business problems with statistics, machine learning, and artificial intelligence. You might not be called a data scientist yet, but by this point, you're already operating as a data scientist and then eventually, you be the data scientist, you become the data scientist. If it is a career path for you, you'll potentially change roles into a role that's formerly called, specifically called data science.
But one of the articles I wrote recently on Medium talks about the seven paths to data scientist and one of the paths talks about a fellow who really doesn't consider himself a data scientist, but he is a data scientist, been a data scientist for years, but he's really happy with this organization and his role as it’s titled as an engineer and he's great. He's good to go.
So maybe we'll talk about it a little bit later, too. I think as we were chatting and planning, someone asked about pedigree a little bit and one of the points I like to make is there's no right, or wrong way to do it. There's no right, or wrong way to get there just once you get there, have fun with it.
MAE: I love what you said, Adam, about the steps and they're very similar to what I would advise to any traditional coder and have advised is take all of your prior work experience before you become a programmer. It is absolutely relevant and some of the best ways to have a meaningful impact and mitigate one's own imposter syndrome is to get a job where you are programming and you already have some of that domain knowledge and expertise to be able to lend.
So you don't have to have been one of the rarefied few, but just having any familiarity with the discipline, or domain of the business you end up getting hired at, or applying to certainly is a way to get in the door a little easier and feel more comfortable once you're there, that you can contribute in lots of ways.
ADAM: And it gives you the ability to provide value that other folks who are on a different path, who are going into data science earlier—this is a great path, too don't let me discount that path—but those folks don't have the deep domain knowledge that someone who transitions into data science later in their career provides.
MAE: Exactly. Yeah, and the amazing teams have people with all the different versions, right?
MAE: Like we don't want a team with only one. Yeah.
ADAM: That's another thing I like to say about data science is it's a team sport. It has to be a teams – it has to be done in tandem with others.
CASEY: I just had a realization that everyone I know in data science, they tend to come from science backgrounds, or maybe a data science bootcamp. But I don't know anyone who moved from web development into data science and that's just so surprising to me. I wonder why.
MAE: I crossed the border a little bit, I would say, I worked in the Center for Data Science at RTI in North Carolina and I did do some of the data science there as well as just web programming, but my undergrad is biochem. So I don't break your role. [laughs]
MANDO: [chuckles] Yeah. I'm trying to think. I don't think I know any either. At the very least, they all come from a hard science, or mathematics background, which is interesting to me because that's definitely not my experience with web application developers, or just developers in general. There's plenty that come from comp side background, or an MIS background, or something like that, but there's also plenty who come from non-traditional backgrounds as well. Not just bootcamps, but just like, they were a history major and then picked up programming, or whatever and it doesn't seem to be as common, I think in data science. Not to say that you couldn't, but just for my own, or maybe our own experience, it's not quite as common.
ADAM: If there's anybody listening with the background that we're talking about, the other backgrounds, I would say, reach out probably to any of us and we'd love to workshop that with you.
MAE: Yes! Thank you for saying that. Absolutely.
MANDO: Yeah, the more stories we can amplify the better. We know y'all are out there; [chuckles] we just don't know you and we should.
MAE: Adam, can you tell us some descriptor that is a hobnobbing thing that we would be able to say to a data scientist? Maybe you can tell us what P values are, or just some little talking point. Do you have any favorite go-tos?
ADAM: Well, I suppose if you're looking for dinner party casual conversation and you're looking for some back pocket question, you could ask a data scientist and you're not a data scientist. I would maybe ask a question like this, or a question that I could respond to easily as a data scientist might be something like, “Well, what types of predictions are you looking to make?” and then the data scientists could respond with, “Oh, it's such an interesting question. I don't know if anybody's ever asked me that before!” But the response might be something like, “Well, I'm trying to predict a classification. I'm trying to predict categories,” or “I'm trying to predict income,” or “I'm trying to predict whatever it is that –” I think that would be an interesting way to go.
What's another one?
CASEY: Oh, I've got one for anyone you know in neuroscience.
ADAM: Oh, yeah.
CASEY: I was just reading a paper and there's this statistics approach I'm sure I did in undergrad stats, but I forgot it. Two-way ANOVA, analysis of variance, and actually, I don't think I know anyone in my lab that could explain it offhand real quickly really well because we just learn it enough to understand what it is and why we use it and then we have the computer do it. But it's an interesting word saying it and having someone say, “Yes, I know what that means enough. It’s a science, or neuroscience.”
ADAM: I would be interested in how neuroscience is used two-way ANOVA because I'm not a neuroscientist and two-way ANOVA is so useful in so many other contexts.
CASEY: I'm afraid I can't help today. Maybe 10 years ago, I could have done that.
CASEY: It's just something that you don't work with and talk about a lot. It's definitely fallen out of my headspace. I looked up the other day, I couldn't remember another word from my neuroscience background. Cannula is when you have a permanent needle into a part of the brain, or maybe someone's vein, same thing. I used to do surgeries on rats and put cannulas and I was like, “What's that thing? What was that thing I did?” I have no idea! It's just like time passes and it fades away. I don't do that anymore.
ADAM: So sometimes folks will ask me why I'm a data scientist and I love that question by the way, because I'm a major proponent of knowing what your why is in general, or just having a why and knowing a why, knowing what your why is. Why do you do what you do? What makes you excited about your career, about your work, about your clients, about your coworkers?
One of the main reasons I am a data scientist is because it's an opportunity to create new knowledge and that's the scientific process, really. That's the main output of science is new knowledge and if you think about that, that's really powerful. This is now at the end of this scientific process, if you implement it correctly, we now know something about how the world works, about how people in the world work, or something about the world in general that we didn't know before. I get goosebumps. We're on podcast so you can't see the goosebumps that I'm getting. But when I talk about this, I actually get goosebumps.
So for me, being a data scientist and then there's also the debate is data science, science and I say, absolutely yes, especially when you are implementing your work with this spirit’ the spirit of creating new knowledge. One of the reasons I am very adamant about keeping this why in the forefront of my mind and proposing it as a why for others who maybe haven't found their why yet is because it's also a really powerful guardrail that prevents us from working on problems that we already have answers to, that have been analyzed and solved, or questions asked and asked and answered.
I'm a major proponent of avoiding that type of work, unless you have a really good reason to replicate, or test replication, or you're looking for replication. That would be an exception, but in general, questions—analytical questions, research questions, and data science problems—that lead to new knowledge are the ones that excite me the most.
And then this goes back to what I was talking about a moment ago, my superpower teaching and learning. One of the reasons I really enjoy teaching data science in the classroom, or statistics in the classroom, or at corporate training is because then I can empower others to create new knowledge. That feels really good to me when I can help others create new knowledge, or give others the skills and abilities to do that as well.
MAE: I love that. Yeah. I do have one angle on that, but I hope this doesn't feel like putting you on the spot, but especially in the not revisiting a established—I'm going to do air quotes—facts and from undergrad, the scientific definition of fact has not yet been proven false.
But anyways, there is a growing awareness of bias inherent in data and we so often think of data as the epitome of objectivity. Because it's a bunch of numbers then therefore, we are not replicating, or imposing our thoughts, but there is the Schrodinger's cat, or whatever in place all the time about how those “facts” were established in the first place, where that data was called from? Like, the Portlandia episode where they ask where the chicken is from and they end up back at the farm.
The data itself, there's just a lot in there. So I'm curious if you have any thoughts about that accordion.
ADAM: There’s a lot. That's a big question. I will say one of the things that keeps me up at night is this problem, especially when it comes to the potential for our work in data science, to perpetuate, exacerbate social inequity, social inequality, racial inequality, gender inequality, economic inequality. This keeps me up at night and I am, like most, or like everyone – well, no, I don't know if everybody is interested in solving that problem. I think a lot of data scientists are, I think a lot of researchers are; I think many are interested in solving that particular problem and I count myself among those. But I would be ahead of myself if I purported to say that I had a solution.
I think in this format and in this context, one of the best things to do is to point folks towards others who have spent even more time really focusing on this and I think the go-to is Weapons of Math Destruction. Weapons of Math Destruction is a book. If you're on a bad connection, that's M-A-T-H. Weapons of Math Destruction and especially if you're just getting started on this concern, that's a good place to get started.
MAE: Thank you. Thanks for speaking to that, Adam.
CASEY: There's a piece of the question you asked me that I always think about is the data true and I like to believe most data is true in what it measured, but it's not measuring truth with a T-H.
ADAM: That’s true.
ADAM: I think you could spend a lot of time thinking this through and noodling through this, but I would caution you on something you said it's true as to what you measured. Well, you have measurement error. We have entire – actually, I happen to have social statistics handbook handy. In any statistics handbook, or statistics textbook is going to have either an entire chapter, or a major portion of one of the introductory chapters on error, the types of error, and measurement error is one of them, perception error, all of the – and I'm on the spot to name all the errors. I wish I could rattle those off a little bit better.
ADAM: But if you're interested, this is an interesting topic, just Google data errors, or error types, or statistical errors and you will get a rabbit hole that will keep you occupied for a while.
MAE: Love it. I will be in that rabbit hole later. [laughs]
ADAM: Yeah. I'm going to go back down that one, too myself.
MANDO: So Adam, we have people who are listening right now who are interested in following one of your paths, or one of the paths to becoming a data scientist and maybe they have domain expertise in a particular area, maybe they don't. Maybe they're just starting out. Maybe they're coming from a bootcamp, or maybe they're from a non-traditional background and they're trying to switch careers. If you were sitting there talking to them one-on-one, what are some things that you would tell them, or what are some starting points for them? Like, where do you begin?
ADAM: Well, one, admittedly self-serving item I would mention is consider the option of hiring a career coach and that's one of the things that I do in my line of consulting work is I help folks who are towards the middle, or latter part of their career, and they're looking to enter into, or level up in data science. So a career coach can – and I've hired career coaches over the years.
Back to, Mando, one of the questions you asked me earlier is how did you end up in data science? Well, part of that story, which I didn't talk to then is, well, I went into data science route when the faculty route didn't open up for me and I'm a huge fan. I had two career coaches helping me out with both, faculty and non-faculty work for a while. So having been the recipient and the beneficiary of some great career coaching, I have also recently become a career coach as well.
Probably something more practical, though. Let me give some practical advice. A portfolio, a professional portfolio for a data scientist is probably one of the most essential and beneficial things you can do for yourself in terms of making that transition successfully and then also, maintaining a career. If you're interested in advancing your career in this way, maintaining a career trajectory that keeps you going so having and maintaining a portfolio.
I'll go through four tips on portfolio that I give folks and these tips are specifically tips that can help you generate content for your portfolio, because I know one of the hardest things to do with the portfolio is, well, let me just do some fictional hypothetical project for my portfolio, so hard to do and also, can end up being sort of dry, stale, and it might not really connect with folks. These are four ways you can add to, or enhance your portfolio. I wouldn't call them entire projects; maybe they're mini projects and they're great additions to your portfolio.
The first one is: make a Rosetta Stone. This one is for folks who have learned one computer programming language, and now it's time for them to learn another computer programming language, or maybe they already know two computer programming languages. In fact, the Rosetta Stone idea for your portfolio doubles as a way to build on and expand your skills.
So here's what a Rosetta Stone is. You have a project; you've done it from start to finish. Let's say, you've done a project from start to finish in Python. Now port that entire project over to R and then in a portfolio platform—I usually recommend GitHub—commit that work as git commits as a Rosetta Stone side-by-side examples of Python and R code that produce the same results and the same output.
I love this piece of advice because in doing this, you will learn so much about the language that you originally wrote the program in and you will learn a lot about the target language. You're going to learn about both languages and you're going to have a tangible artifact for your portfolio and you might even learn more about that project. You might encounter some new output in the new language, which is more accessible for that language, that you didn't encounter in the old language and now you're going to have a new insight about whatever your research project was.
The next piece of advice I have is make a cheat sheet and there's tongue in cheek opinion about cheat sheets. I think sometimes folks don't like to call them cheat sheets because the word cheat has negative connotations, but whatever you're going to call it, if it's a quick reference, or if it's a cheat sheet, a well-designed cheat sheet on any tool, platform, tool platform, language that you can think of is going to be a really nice addition to your portfolio.
I recommend folks, what you do is you just find the things that you do the most frequently and you're constantly referencing at whatever website, make a cheat sheet for yourself, use it for a while, and then polish it up into a really nice presentable format. So for example, I have a cheat sheet on interpreting regression. I also have a cheat sheet that is a crosswalk from Stata, which is a statistical programming language, to Python. So actually there, I've put the two of them together. I've made this cheat sheet, which is also a Rosetta Stone.
If you're looking for those, you can find those on my GitHub, or my LinkedIn, I have cheat sheets on my LinkedIn profile as well and you can see examples. I do have on YouTube, a step-by-step instructional video on how to make a cheat sheet and they're actually really easy to do. So if you even if you consider yourself not graphically inclined, if you pick the right tools—and the tools that you would pick might not be your first choice just because they're not marketed that way—you can put together a really nice cheat sheet relatively easily.
The third tip is to write an article… about a piece of software that you dislike. So write an article about a piece of software that you dislike and this has to be done with, especially in the open source community, do this one carefully, possibly even contact the creators, and also, be sure not to blame anybody, or pass judgment. Just talk about how and why this particular project doesn't quite live up to your full aspiration, or your full expectation.
I've done this a couple of times in a variety of ways. I didn't in the title specifically say, “I don't like this,” or “I don't like that,” but in at least one case, one of the articles I wrote, I was able to later submit as a cross-reference, or an additional reference on an issue in GitHub and this was specifically for Pandas. So there was a feature in Pandas that wasn't working the way I wanted it to work. [chuckles]
ADAM: Yeah, Pandas is great, right? So there's a feature in Pandas that wasn't working in quite the way that I wanted it to. I wrote an article about it. Actually, I framed the article, the article title is, “How I broke Pandas.” Actually, several versions of Pandas back, the issue was it was relatively easy to generate a Pandas data frame with duplicate column names. Having duplicate column names in a Pandas data frame obviously can cause problems in your code later because you basically have multiple keys for different columns. Now, there's a setting in Pandas that will guard against this and it's an optional setting—you have to toggle it on and off. This article, I like to say, helped improve Pandas.
So write an article about software you dislike and also, like I said, be diplomatic and in this case, I was diplomatic by framing the article title by saying, “A few times, I managed to break Pandas,” and then –
MANDO: This reminds me a lot of Kyle Kingsbury and his Jepsen tests that he used to do. He was aphyr on Twitter. He's not there anymore, but he would run all these tests against distributed databases and distributed locking systems and stuff like that and then write up these large-scale technical explanations of what broke and what didn't. They're super fascinating to read and the way that he approached them, Adam, it's a lot like you're saying, he pushed it with a lot of grace and what I think is super important, especially when you're talking about open source stuff, because this is what people, they're pouring their heart and soul and lives into. You don't have to be ugly about it.
ADAM: Oh, absolutely.
MANDO: [chuckles] And then he ended up like, this is what he does now. He wrote this framework to do analysis of distributed systems and now companies hire him and that's his job now.
I'm a big fan of the guy and I miss him being on Twitter and interacting with him and his technical expertise and also, just his own personality. Sorry, your topic, or your little cheat there reminded me of that. We'll put some links—thanks, Casey—and in the show notes about his posts so if people haven't come across this stuff yet, it's a fascinating read. It's super helpful even to this day.
ADAM: I'm thankful for the connection because now I have another example, when I talk to people about this, and it's incredible that you say built an entire career out of this. I had no idea that particular tip was so powerful.
MAE: So cool.
MANDO: [chuckles] So I think you said you have one more, Adam?
ADAM: The fourth one is: contribute to another project. One of the best examples of this is I wrote an article on how to enhance your portfolio and someone really took this fourth one to a whole new level. I'm sure others have as well, but one person—we’ll get links in, I can get some links in the show notes—what he did was he found a package in R that brings data for basically sample datasets for our programmers and citizens working and data scientists working with R. But he was a Python person.
So he suggested, “Hey, what about making this?” I remember he contacted me and he said, “I read your article about adding to my portfolio. I really think it might make sense to port this project over to Python,” and so, he was combining two of them. He was making a Rosetta Stone and he was contributing someone else's project.
Now this data is available both in R and in Python and the author of this project has posted about it. He posted about it in May, early May, and it's constantly still a month and a half later getting comments, likes, and links. So he's really gotten some mileage out of this particular piece, this addition to his portfolio and the original author of the original software also has acknowledged it and it's really a success. It's really a success. So contribute to another project is my fourth tip.
Oh, one more idea on contributing to another project. Oh, I have an article on that lists several projects that are accepting contributions from intermediate and beginners. The point there is identify specific projects that are accepting beginner and immediate submissions on contributions, mostly via GitHub. But if you go to GitHub and if you're newer to GitHub, you can actually go to a project that you like, go to its Issues tab, and then most projects have tags associated with their issues that are identified as beginner friendly. That is an excellent place to go in order to get started on contributing to another project, which makes the world a better place because you're contributing to open source and you have an addition to your portfolio.
MANDO: Oh, these are fantastic tips. Thank you, Adam.
ADAM: I'm glad you like them. Can I give another one? Another big tip? This one's less portfolio, more –
MANDO: Yeah, lay it on us.
MAE: Do! By all means.
ADAM: And I'd be interested, Mae, since you also made a similar career transition to me. I made an investment. I think I know what you might say on this one, but I spent money. I spent money on the transition. I hired consultants on Fiverr and Upwork to help me upgrade my social media presence. I hired the career coaches that I mentioned. Oh, actually the PhD program, that was not free. So I spent money on my transition and I would point that out to folks who are interested in making this transition, it's not a transition that is effortless and it's also not a transition that you can do, I think it's not one that you can do without also investing money.
MAE: Yeah. [chuckles] Okay, I'm going to tell you my real answer on this.
MAE: Or corollary. I had a pretty good gig at a state institution with a retirement, all of these things, and I up and left and went to code school. I had recently paid off a lot of debt, so I didn't have a lot of savings. I had no savings, let's just say that and the code school had offered this like loan program that fell through. So I'm in code school and they no longer are offering the ability to have this special code school loan. I put code school on my credit card and then while in code school, my 10-year-old car died and I had to get a new car.
MAE: In that moment, I was struggling to get some fundamental object-oriented programming concepts that I'm like, “Holy cow, I've got a mortgage. I no longer have a car.” Now I'm in a real bind here, but I be leaving myself. I know I made these choices after a lot of considered thought and consultation. I, too had hired a career coach and I was like, “I've already made this call. I'm going to make the best of it. I'm just going to do what I can and see what happens.”
I really have a test of faith on that original call to make those investments. I would not recommend doing it the way I did to anyone!
MAE: And I went from a pretty well-established career and salary into – a lot of people when they go into tech, it's a huge jump and I had the opposite experience. That investment continued to be required of me for several years. Even still, I choose to do things related to nonprofits and all kinds of things, but it takes a lot of faith and commitment and money often, in some form, can be helpful. There are a lot of, on the programming side, code schools that offer for you to pay a percentage once you get a salary, or other offsetting arrangements.
So if somebody is listening, who is considering programming, I have not seen those analogs in data science, but on the programming side, especially if you're from a group underrepresented in tech, there's a number of different things that are possible to pursue still.
ADAM: Here we are talking about some of the lesser acknowledged aspects of this transition.
ADAM: Some of the harder to acknowledge.
MANDO: Yeah, I really liked what you said, Mae about the need to believe in yourself and Adam, I think what you're saying is you have to be willing to bet on yourself.
MANDO: You have to be willing to bet on yourself and sometimes, in some forms, that's going to mean writing a check, or [chuckles] in Mae’s example, putting it on your credit card, but.
Sometimes that's what it means and that's super scary. I'm not a 100% convinced that I have enough faith in my ability to run the dishwasher some days, you know what I mean? Like, I don't know if I'm going to be able to do that today, or not.
This is going to be really silly and stupid, but one of my favorite cartoons is called Avatar: The Last Airbender.
MANDO: It's a series on Cartoon Network, I think. No, Nickelodeon, I watched it with my kids when they were super little and it's still a thing that we rewatch right now, now that they're older. There's this one episode where this grandfatherly wizened uncle is confronted [chuckles] by someone who's trying to mug him [chuckles] and the uncle is this super hardcore general guy. He critics his mugging abilities and he corrects him and says, “If you stand up straight and you change this about the way that you approach it, you'll be much more intimidating and probably a more successful mugger,” and he's like, “But it doesn't seem that your heart is into the mugging.” [chuckles]
So he makes this guy a cup of tea and they talk about it and the guy's like, “I don't know what I'm doing. I'm lost. I'm all over the place. All I want to do is become a masseuse, but I just can't get my stuff together.” Something that the uncle said that really, really struck with me was he said, “While it's important and best for us to believe in ourselves, sometimes it can be a big blessing when someone else believes in you.”
MAE: So beautiful.
MANDO: “And sometimes, you need that and so, I get it. You can't always bet on yourself, or maybe you can bet on yourself, but sometimes you don't have that backup to actually follow through with it.” That's why community is so important. That's why having a group of people. Even if it's one person. Someone who can be like that backstop to be, “You don't believe in yourself today. Don't worry about it. I believe in you. It's okay. You can do it. You're going to do it.”
ADAM: Community is just massive. Absolutely massive.
ADAM: Having a good, strong community is so important. Also, I think I could add to what you're saying is about betting on yourself. I don't know if I love the analogy because it's not a casino bet.
ADAM: The odds are not in favor of the house here. If you have done the right consultation, spoken with friends and family, leveraged your community, and done an honest, objective, accurate assessment of your skills, abilities, and your ambition and your abilities, et cetera. It's a bet. It's a wager, but it's a calculated risk.
MAE: Yes! That is how I have described it also. Yes, totally. I loved that story from Airbender and it ties in a few of our topics. One is one of the things Adam said originally, which is being deeply in touch with your why really helps. It also ties in the whole teaching thing and often, that is one of the primary roles is to offer faith and commitment to your pursuits.
If I had had different code school teachers, the stress of my entire livelihood being dependent on my understanding these concepts in week two of bootcamp that I was struggling with, and I had made a calculated bet and I thought I was going to be awesome, but I was not. It was like the classic Peanuts teacher is talking, “Wah wah woh wah wah.”
I had to lean into my teachers, my school, my peers, believe in me. I believed in me before, even if I don't in this moment and I just have to let that stress move to the side so that I can reengage. That was really the only way I was able to do it was having a similar – well, I didn't try to mug anybody, [laughs] but I had some backup that really helped me make that through.
MANDO: Yeah, and those credit card folks call like, it’s tricky.
MAE: Yeah, and then I had to buy a car and those people were calling me and they just did an employment verification. They said, “You don't have a job!” I was like, “Oh my god. Well, you [inaudible] get my car back, but I have really good credit. How about you talk to your boss and call me back?”
So anyway, these things all tie into, if we have time to talk about something, I was hoping we would cover is this thing about imposter syndrome and believing in oneself, but also not believing in oneself simultaneously and how to navigate that. I don't know, Adam, if you have particular advice, or thoughts on that.
ADAM: I do have some advice and thoughts on that. Actually, just yesterday, I hosted a live webinar on this particular topic with another career coach named Sammy and she and I are very passionate about helping folks. When we work with clients, we work with folks intentionally to evaluate whether imposter syndrome might be part of the equation. Actually, in this webinar, we talked about three immunity boosts, or three ways to boost your immunity against imposter syndrome and in one way, or another, I think we've touched on all three with the exception of maybe one of them.
So if you're interested in that topic reached out to me as well. I have a replay available of that particular webinar and I could make the replay available on a one-on-one basis to folks as well, who really want to see that material, and the section –
MANDO: [inaudible] that.
ADAM: Yeah, please reach out and LinkedIn. Easiest way to reach me is LinkedIn, or Twitter. Twitter actually works really well, too these days.
MANDO: We’ll put both of those in the show notes for folks.
ADAM: Okay. Yeah, thank you so much. I look forward to potentially sharing that with folks who reach out.
The community was the second immunity boost that we shared and actually, Mando and Mae, both just got done talking extensively about community. And then the first immunity boost we shared was know your baseline. We called it “know your baseline” and I know from our planning that we would put in this program notes, a link to an online assessment that's named after the original scientist, or one of the two original scientists who really began documenting imposter syndrome back in the 70s and then they called it imposter phenomenon.
Oh, the history of this topic is just fascinating. Women scientists, North Carolina, first documented this and one of the two scientists is named Pauline Clance. So the Clance Imposter Phenomenon Scale, that'll be in the show notes. You can take the Imposter Phenomenon Scale and then objectively evaluate based on this is imposter syndrome a part of your experience, if it is what is the extent of that, and just knowing your baseline can be a really good way, I think to protect you from the effects of the experience. It's also, I think important to point out that imposter syndrome isn't regarded as a medical, or a clinical diagnosis. This is usually defined as a collection of thoughts and actions associated with career, or other academic pursuits.
And then the third immunity boost is disseminate knowledge and I love the disseminate knowledge as an immune booster because what it does is it flips the script. A lot of times folks with imposter syndrome, we say to ourselves, “Gee, if I could get one more degree, I could probably then do this,” or “If I got one more certification,” or “I can apply for this job next year, I could apply for that permission next year because I will have completed whatever certification program,” or “If I read one more –”
MANDO: One more year of experience, right?
ADAM: Yeah. One more year of experience, or one more book, or one more class on Udemy. Especially for mid and late career professionals and we talked about this earlier, Mae the bank of experience and domain knowledge that mid and late career professionals bring, I promise nobody else has had your experience. Everybody has a unique experience and everybody has something to offer that is new and unique, and that is valuable to others. So I say, instead of signing up for the seminar, host the seminar, teach the seminar.
ADAM: Right? Again, there's nothing wrong with certifications. There's nothing wrong with Udemy classes, I have Udemy classes that you could should go take. There's nothing wrong with those, but in measure, in measure and then also, never, never, never, never forget that you already have skills and abilities that is probably worth sharing with the rest of the world. So I recommend doing that as a boost, as an immunity boost, against imposter syndrome.
MANDO: Yes, yes, and yes!
CASEY: Now, I took the Clance Imposter Phenomenon Scale test myself and I scored really well. It was super, super low for me. I'm an overconfident person at this point, but when I was a kid, I wasn’t.
I was super shy. I would not talk to people. I'd read a book in a corner. I was so introverted and it changed over time, I think by thinking about how confidence leads to confidence.
CASEY: The more confident you are, the more confident you act, you can be at the world and the more reason you have to be competent over time and that snowballed for me, thank goodness. It could happen for other people, too gradually, slowly over time the more you do confidence, the more you'll feel it and be it naturally.
MANDO: I think it works the other direction, too and you have to be real careful about that. Like Adam, you were talking about flipping the script. If you have a negative talk script of just one more, just this one thing, I'm not good enough yet and I'm not you know. That can reinforce itself as well and you just never end up getting where you should be, or deserve to be, you know what I mean?
It's something that I struggle with. I've been doing this for a really, really long time and I still struggle with this stuff, it’s not easy. It's not easy to get past sometimes and some days are better than others and Casey, like you said, it has gotten better over time, but sometimes, you need those daily affirmations in the morning in the mirror [laughs] to get going, whatever works for you. But that idea, I love that idea, Casey of confidence bringing more confidence and reinforcing itself.
MAE: And being mindful of Dunning-Kruger and careful of the inaccuracy of self-assessment. I like a lot of these ways in which making sure you're doing both, I think all the time as much as possible. Seeing the ways in which you are discounting yourself and seeing the ways in which you might be over crediting.
ADAM: Right. Like with a lot of good science, you want to take as many measurements as possible.
ADAM: And then the majority vote of those measurements points to some sort of consensus. So the IP scale is one tool you can use and I think to your point, Mae it'd be a mistake to rely on it exclusively. You mentioned Dunning-Kruger, but there's also the Johari window.
MAE: Oh, I don’t know. What’s that?
ADAM: Oh, the Johari window is great. So there's four quadrants and the upper left quadrant of the Johari window are things that you know about yourself and things that other people know about yourself. And then you also have a quadrant where things that you know about yourself, but nobody else knows. And then there's a quadrant where other people know things about you that you don't know. And then there's the complete blind spot where there are things about you that you don't know that other people don't know. And then of course, you have this interesting conversation with yourself. So that quadrant that I don't know about it and nobody else knows about it, does it really exist? Does the tree falling in the woods make a sound when nobody's there to hear it?
You can have a lot of fun with Johari window as well and I think it also definitely connects with what you were just saying a moment ago about accuracy of self-assessments, then it gets back to the measurement that we were talking about earlier, the measurement errors. So there's perceptual error, measurement error—shucks, I had it, here it is—sampling error, randomization, error, all kinds of error. I managed to pull that book out and then get some of those in front of me.
CASEY: There are some nice nicknames for a couple of the windows, Johari windows. The blind spot is one of those four quadrants and façade, I like to think about is another one. It's when you put on the front; people don't know something about you because you are façading it.
MANDO: So now we'll go ahead and transition into our reflection section. This is the part where our esteemed panelists and dear friends reflect on the episode and what they learned, what stuck with them, and we also get reflection from our guest, Adam as well, but Adam, you get to go last.
ADAM: Sounds good.
MANDO: You can gauge from the rest of us. Who would like to go first?
MAE: I can! I did not know that there was an evaluative measure about imposter phenomenon, or any of that history shared and I'm definitely going to check that out. I talk with and have talked and will talk with a lot of people about that topic, but just having some sort of metric available for some self-assessment, I think is amazing. So that is a really fun, new thing that I am taking away among many, many other fun things. How about you, Casey?
CASEY: I like writing about software you dislike in a positive, constructive tone. That's something I look for when I'm interviewing people, too. I want to know when they get, get feedback, when they give feedback, will it be thoughtful, unkind, and deep and respectful of past decisions and all that. If you've already done that in an article in your portfolio somewhere, that's awesome. That's pretty powerful.
MANDO: Oh, how fantastic is that? Yeah, I love that!
CASEY: I don't think I've ever written an article like that. Maybe on a GitHub issue, or a pull request that's longer than it feels like it should be.
Maybe an article would be nice, next time I hit that.
MANDO: Oh, I love that. That's great.
I guess I’ll go next. The thing that really resonated with me, Adam was when you were talking about investing in yourself and being willing to write that check, if that's what it means, or swipe that credit card, Mae, or whatever. I'm sorry, I keep picking on you about that.
MAE: It’s fine. [laughs] It’s pretty wild!
MANDO: I love it. I love it, and it reminded me, I think I've talked about it before, but one of my favorite writers, definitely my favorite sports writer, is this guy named Shea Serrano. He used to write for Grantland and he writes for The Ringer and he's a novelist, too and his catchphrase—this is why I said it earlier in the episode—is “bet on yourself.”
Sometimes when I'm feeling maybe a little imposter syndrome-y, or a little like, “I don't know what I'm going to do,” I click on the Twitter search and I type “from:sheaserrano bet on yourself” and hit enter and I just see hundreds and hundreds and hundreds of tweets of this guy that's just like, “Bet on yourself today.” “Bet on yourself” “Bet on yourself today, no one else is going to do it.” “No one's coming to save you, bet on yourself,” stuff like that and thank you, Adam for that reminder today. I needed that.
ADAM: You're welcome. I'm so happy that you've got that takeaway. Thank you so much for sharing the takeaway.
I have, I think two reflections. One, what a breath of fresh air, the opportunity to talk about life, career, but career in data science, and programming in a non-technical way. I think the majority of our conversation was non-technical.
We briefly went into some technicalities when we talked about how you can sometimes have duplicate heading names in a Pandas data frame. That was a little bit technical. Otherwise, we really just spoke about the humanistic aspects of this world. So thank you so much for that and I got a research tip!
Mando, what a brilliant idea. If you're ever looking for more background on a book, do a Twitter search for the book name and then anybody who's been speaking about that book –
MANDO: Oh, yes!
ADAM: Yeah, right? You could extend that to a research tip. [overtalk]
MANDO: That’s fantastic! Absolutely. Yeah.
ADAM: So today, I learned a new way to get additional background on any book. I'm just going to go to Twitter, Google, or not Google that, search the book title name, and I'm going to see what other people are saying about that book. And then I can check out their bios. I can see what else they're sharing. They might have insights that I might not have had and now I can benefit from that. Thank you. Thank you so much for the research tip.
MANDO: Yeah, and I think it dovetails really well into what you were talking about earlier, Adam, about publishing data. Like building out this portfolio, writing your articles, getting it out there because someone's going to go to Google, or Twitter and type into the search bar a Pandas data frame, column, same name, you know what I mean and now they're going to hit “A few times, I managed to break Pandas,” your article.
But it could be about anything. It could be about that stupid Docker thing that you fought with yesterday, or about the 8 hours I spent on Monday trying to make an HTTP post with no body and it just hung forever and I couldn't. 8 hours, it took me to figure out why it wasn't working and it's because I didn't have one line in and I didn't call request that set body. I just didn't do it. I've done this probably more than a million times in my career and I didn't do it and it cost me 8 hours of my life that I'm never getting back, but it happens. That's part of the job is that – [overtalk]
MAE: Yeah, sure.
MANDO: And you cry about it and you eat some gummy worms and then you pick yourself back up and you're good to go.
ADAM: Yeah, another common one that people are constantly writing about is reordering the columns in a Pandas data frame. There's like a hundred ways to do it and none of them are efficient.
MANDO: [laughs] Mm hm.
ADAM: So I love [inaudible], of course.
MANDO: Yeah, you hit the one that works for you, write a little something about it. It’s all right.
ADAM: Exactly, yeah.
MANDO: All right. Well, thanks so much for coming on, loved having you on.Support Greater Than Code