IBM Data Science Professional Certificate Review

hello and welcome to data research labs where we discuss data related topics for today's leg of my data science journey we're going to discuss the ibm data science professional certificate that's out on coursera first we're going to level set by looking at the coursera site and see what this certificate is all about so we're going to do a different approach this time i wrote an article medium whose link i'll post in the description down below this video i'm going to bounce around between this article you see here and the coursera site and some other sites so this ibm data science professional certificate that you see up here on screen is one of many stepping stones on my journey to become a data scientist this certificate program is comprised of nine courses that we'll look at in a minute and there's already 235 428 students who've been enrolled and i googled it this site's been up around three years maybe a little longer so heavy usage and very well known and very well crafted and very well established i noticed something that i wanted to point out that speaks to the difficulty of this certification here in uh the course number one certificate we see 235 000 enrollees but if we go to the final ninth capstone project there's only 67 000 so 24 roughly seven so about two thirds of the people drop off by the time they get to the end they just don't complete it so that gives you an idea of the level of effort even though it's a beginner course by the time you get to the end it's intermediate to perhaps even advanced what do they say here yeah they say intermediate level this certificate program is comprised of nine courses taking the student from the basics of data science through to creating their own unique capstone project at the end i'm not going to go through the details i'll leave that to you i'll put the link to this coursera certificate down in the description field so you can browse to it but in a nutshell the nine courses are what is data science tools for data science data science methodology python for data science and ai here's where you start getting some hands-on using the tools database and sql for data science data analysis with python data visualization with python machine learning with python and then the big one the applied data science capstone where you put to use and exercise all the skills that you've learned up to that point this certification is a great starting point for a career shift in the data science and i highly recommend it to either beginners or intermediate level folks beginners receive a broad overview of the fundamentals of what data science entails mid-level students will appreciate the capstone project for the experience gained coming up with your own project sourcing your own data writing your own code as well as kickstarting your portfolio with a mandatory github account writing a research paper and writing blog articles curious students wondering whether a data science career might be a good fit should work through this program because if you can make it to the end and especially if you enjoy the last two courses then you'll know that data science is probably a good fit on the other hand if you don't make it to the end or if you don't enjoy the last two courses where you really get into the nuts and bolts of data science then until i contrary you'll know that maybe data science isn't a good fit for you next we'll discuss how much does this certification cost according to the coursera site it's 39 us dollars per month as you work your way through all nine courses also according to the coursera site this certification and all the courses are free so long as you can complete it within the seven day window but that's pretty unlikely you'd have to focus 10 hours a day to try and get that done in seven days and probably more than 10 hours a day we'll see in a minute how long it'll actually take quick tip you can certainly go into the seven day free window with a trial and error attitude and if this is not for you then drop it no harm no foul no money lost just be prepared to cancel i have an annual subscription so the charge was just added to my bill i assume those without an annual subscription would be prompted at that point after seven days to enter billing info and you just say no but i'm not sure how that works on the other hand if you didn't pay going in and everything's working out great then fine you would pay at the start of seven days to continue and all would be good next we'll discuss how much time does it take to complete the certificate so from the coursera site if we scroll down to any given course click on the course details link page comes up if we scroll down i have a lot of details on the course and it'll tell you approximately how many hours that given course will take but these numbers aren't entirely accurate you can do things like i suggest watching videos at double speed it keeps your concentration on and it gets you through the material a little bit faster your mind doesn't wander and then the labs sometimes you're a bit slower sometimes you're a bit faster so anyway we're gonna look at the total hours here is about 195 hours total if you add up all nine courses but next we're going to look at how long it actually took me the certificate took me 106 hours across 17 days i did it with christmas eve christmas day new year's day i had three eight hour days off holidays so that helped out get it done a little bit quicker but again according to the coursera documentation it takes an average person about 10 months 195 hours about double 106 that i took and for example the capstone course takes six weeks according to coursera for me the data science course number one five and a quarter hours course number two six course number three was a shorter course like a half to a third of the materials other courses two and a half hours four and a quarter etc these courses one through seven all took less than eight hours i could go through them pretty quickly course number eight there's a pretty involved lab and i ran into some issues that took quite a bit longer 16 and a quarter hours and then it took me 56 and a half hours on the applied data science capstone project an important caveat about the numbers above i just finished an intense year completing graduate degree in data science and consequently moved pretty fast through the course material labs so adjust your expectations accordingly also note that for courses one through eight i went about two to four times faster than most students but then i dropped it two times slower than most students for the applied did a capstone and uh this was intentional i planned on using the capstone to kick start my portfolio my project portfolio and i wanted to really understand the material and polish it off all the artifacts the jupiter notebook for code the pdf research paper and the powerpoint presentation so i knew going in i was going to spend a lot more time on this that's your decision what you want to do but that's what i did and why next we'll discuss whether the certification is worth the time money and effort to obtain and the answer is absolutely you cover a lot of material you have to practice your data science skills throughout the program especially during the capstone course where you get to heavily exercise what you've learned that capstone also requires that you start a portfolio on github something that i've been putting off but i'm very happy i had to complete for this project and here's my github account and it also required the capstone also requires that you write up a blog or do a presentation as well so we're going to drill down here into my github project so that you can see what github's all about if you don't already know so i really like this in my github project here i have a data research labs is my repo repository and then coursera ibm data science professional certificate is the project in this project i have a python i'm sorry a jupiter notebook with a python code in it i have my report word document and then i have my presentation powerpoint alternatively i could have done a blog but i just chose this because i needed to get it done faster the three of these i'm going to load them and show you in a minute but i also did them in pdf format just in case there's an issue sometimes loading a jupiter notebook can be slow but anyway let's go look at jupyter notebook i really like what those look like takes a little bit of time to load up here good github's not busy so load loaded in fine so there's my jupiter notebook which is a nice scientific approach to getting your code and data and results and notes all lined up so my markdown cells tell what's going on have a table of content have some instructions here's some code cells to install components some code cells to import the libraries pandas matplotlib etc python code data acquisition layer what i'm doing where i'm getting the data from all markdown cells that are instructions and then it starts to get into the nuts and bolts of it and what's neat is from github people don't have to have jupyter notebook installed github will automatically pre-format and run it for you so anybody can click this link and see the jupiter notebook with uh execution line 20 there's the blob of block of python code here's the output where i took the head the top five rows of that first run where i was loading in a csv and manipulating it a little bit anyway it's really neat it was nice to get that set up because as a data scientist it's really important to have a portfolio and show your work i can present all kinds of stuff and nobody cares nobody looks at it don't look at my resume don't look at linkedin but if i can get a portfolio with several sample projects out there then people can look at that and it's it's a valuable thing to have it reminds me of what architects have always had a portfolio of their work so anyway i really like github and that was a neat artifact oh almost forgot i want to also show you the report and the presentation i'm not going to do these two i'm going to do the pdfs so the report for the project it also renders in github in a pdf viewer so there it is it looks just like i set it up in word uh when you're doing it make sure you get you uh in google you can search for a common commercial what is it common something or other like actually let's go do that really quick boom pop up google here this is a good tangent for when you're doing papers and whatnot so if i want a picture of seattle and i go to images and i go to tools and i go to usage right and i go to creative commons license there we go and typically all of these are free and you can use them you can click it and go look at the details but that's where i source my images from so i know there's not a copyright problem i put the link there as well uh anyway tangent here's my research paper with all the details and a lot of this stuff was redundant i copy pasted it from my report back into my code as markdown cells in the jupyter notebook so that is the report and the powerpoint is going to be the presentation is going to be basically the same thing but summary level so as a pdf it's just slides black background easier on the eyes if you need more proof whether or not this ibm data science certification is worth the time and effort just google it and you will see all over the place that it's ranked in the top eight to 15 data science certificates so here is one and here is one and here is one we'll just pop open all five of them then we'll go look at them so data science certificates pay off wow my machine's going slow and if you go down ibm knock on it it jumped around a bit ibm data science there it is right there professional certificate and we go to the next one and anyway if you scroll down all of those it'll be right there and there's dozens of them if you search so it's a it's well known in the industry and you saw earlier at the beginning this course that there's hundreds of thousands of enrollees so great great program well worth doing next up we'll discuss what social media badges are awarded regarding social media badges there's two different styles or types one is directly from ibm so what you're looking at here is course by course by course there's six courses here's a seventh course and for whatever reason there's two courses that don't show up but anyway seven out of the nine individual courses ibm has specific badges for each of the coursera courses and they're put out by a claim and you can get a link i'll just do the machine learning by python you can get a link and it says it was issued by coursera ibm it was issued to me on this date etc i can click the share button i could click on one of these and have it published i usually don't do that instead i go grab where is it somewhere in here there's a copy link and i go grab that sometimes i just grab the url up here but usually there that one url and so i grab or i just hit the copy and now i have the url and to try it this is important too because i've messed this up before do a in private window if it's edge or do a incognito if it's chrome or whatever that way your credentials aren't there this is truly what a public view would be and sure enough this means that anyone could see this view of it so that is what an ibm course specific badge looks like now different from ibm which has badges for each individual course coursera has a badge for the entire certificate for the entire program and that's the one that i tend to use and put visibility towards because it's more meaningful it represents everything it's not as pretty as the individual ibm course badges but it's the one you should use to post in social media at least the one you should emphasize let's see here's mine i clicked on it um i need to go find the shareable link so why don't i i mean i could download certificate but i don't want that i want to share it so i'll click share certificate and right there it's the copy you want you could click facebook or linkedin and it'll automatically go post for you but you lose control it'll just post it and it's going to say whatever it's going to say whereas if you grab and copy the link then you have control over what you want to say when you post it so i generally copy the link next we'll look at what should be posted to linkedin and how it should be posted i happen to have a strong opinion on how to properly post ongoing certifications to linkedin i believe it is best to post a single overarching certificate representing the program or the single or final highest or highest ranking badge in a series but don't dilute the post by posting all nine badges from ibm post one post a summary don't dilute your message so in this example i got an mcse about a year ago so i didn't post all the individual exams that it took to do the mcse i made a note of them over here i like to emphasize the level of effort by listing out the number of hours i track in 15-minute increments as i'm doing the studying and taking the test so i knew that it took 106 hours i want to put that down look at the difference when i took the professional scrum master certificate that was only 12 hours of studying a i had a lot of experience in the past and read books on the subject and worked in the area for many years and b it just took 12 hours of prep time pass the exam boom have a certification with the credential id but that level of effort compared to three exams at microsoft for an mcse that's a big difference and if i don't specify that you have no idea how hard one is relative to the other and if i were to go take all nine of the courses the badges they're pretty badges i'd have a nice little circle badge here from ibm but it would dilute the message i would have literally i have 21 badges out on a claim and dozens of other badges all over the place so i'd have 30 badges littering this and it's just too many um and i'll show you in a minute an option but with with what to do with those now scratch what i said if you only have one or two or three badges and you're just getting started then by all means put those badges in but over time when you get to an end state endpoint certification put that and delete remove the other ones that all said there is value in posting those individual courses from ibm with the link that way resume crawlers google etc are aware of the details are aware of the specific course names but i don't put those in the licenses and certification sections of linkedin no i scroll down here down down down and in the accomplishments section you would think i would put it in courses but i don't because if i add a course i can't really do much there's i can put a name and a number and an associate with no instead what i do is i put it in the test scores so right now i have 21 test scores representing 21 different badges and for anyone who really wants to look they can hit the down arrow and they will get the details and they will get the actual credential that they can go copy the link doesn't work that's a bit of a bummer but they can copy paste it hit enter and it should pop up the badge so that's how i handle all the details and then in that way when it comes time to look for a job or when i want a google or something to resume crawler to walk through it'll find these details and it'll recognize oh he has data science methodology or python or etc so i do put the individual scores down here and i don't lose the information because if for whatever reason i decide i want to move this badge up at least i have the credentials here on file so that's optionally what you can do with those details i forgot to mention this in my medium article but the third thing that i post on social media is from for big events i'll actually go in here click it and paste the url and i'll type some notes and it looks something like this right here so uh it's not the greatest because you can only post one link which is a bummer but i'll post the link and then i'll post some details about it just finish this and talk about how much cost or how many hours it took etc and then if i could post other links up here i would but instead i end up having to post down in the comments section that here's my github repo here's the coursera link for those interested next we'll try to answer the question are there better certifications available for data science sure there's better sites harvard and mit both have statistics data science certificates out on edx johns hopkins university has one on corsair there's several excellent python r statistics focused certifications for from duke university university of michigan and university of washington all out on coursera and there's other large institutions that have it on either on edx coursera or on their own uh university sites and i plan on taking many of them so we can look at harvard's for example on edx and i'll again post the link to this medium article so you can go grab these links yourself if you're interested in those courses the harvard edx 800 bucks that would be very prestigious to have on your resume it takes it's self-paced says it takes one year uh five months two to three hours a week what is that 52 weeks and another 25 75 weeks times two to three hours that'd be 150 to 300 hours so i would get it's harvard though and it's i don't see it as it i'm guessing it's intermediate or more difficult probability wrangling linear regression yeah it's similar it's similar to the ibm one so maybe this is 150 to 200 hours so maybe if you just went full bore take three to four weeks to do i don't know we'll see i'm gonna go look at that one later but that's a neat one mit has one johns hopkins that one's on corsair i'm probably gonna do that one next it has 416 000 already enrolled i think that's even more than the ibm one so very neat course and beginner level anyway there's there's lots of neat ones and the point though isn't are these better than the ibm certificate i don't think figuring out which one's better or worse isn't it is the issue i think to me the objective should be to hit the material from a diverse set of perspectives and each time you go through the material from a different viewpoint you learn more and more to me an example is i just spent the last year been graduated with a degree graduate degree on data science and yet taking this ibm certification i still had several excellent takeaways new ideas new tools new techniques from the certification getting github set up using the folio for the mapping using the clustering the way that they did it that's all new to me it's all great so and if you think about a data scientist working 2 000 hours a year i just did 1200 hours to get a master's i just did a 106 hours on this certification obviously i have hundreds and hundreds of more hours thousands of hours to get up to speed and transition so i'm going to be taking lots of these courses to get my skill set up get my portfolio up i i'm not a believer in the fake it till you make it stuff i don't i don't like that i think you should know the material really thoroughly and be well versed in it before you transition so that's what i'm gonna do is just grind through all of these different certifications and get the skill set up so to me yeah one's not better than the other they're all great they're all worthwhile anything else i want to mention on that yeah if you're completely new to the field then it probably does matter you probably don't want to start with some of the harder ones you probably don't even want to start with the ibm one you may want to start with some of the smaller taking onesie twozy classes and then work your way up so that's probably the best way to look at these courses is their different levels of difficulty different levels of prestige but they're all excellent just figure out what's the proper order for you and which to take them and finally i'm going to double up here and discuss pretty briefly how the material has stood up over time over the past few years since it was generated for the course and then a conclusion so how's the material stood up has it aged out since it's three years old well the answer is a mix of good and bad i'm not going to belabor the point but in general it's fine the core concepts they're all building blocks of data science they've held up fine and they'll continue to do so for a long time to come the course will be great that said with most technology specific how to's and detailed description guides as the underlying software in this case we use the ibm watson a lot it changed and some of the instructions aren't accurate and yeah it can be a little bit frustrating at times i think in total out of 106 hours about two hours i lost on three different occasions waiting for things that weren't working and researching them then finding a workaround and i noticed folks had complaints the bottom line here is that a former professor of mine once was asked why is the documentation so sparse and some of the labs outdated and broken that makes things much more difficult and time-consuming than otherwise could be yeah that's you know a valid point but his response was fantastic and it stuck with me he said this is a graduate level course your job is not just to find the answers but it's also to figure out the right questions to get the best answers so that really stuck with me and resonated it's it's frustrating but you know what that's life i toss this in down here in the real world you're paid to solve problems you're not paid to complain about them you just lots of other students have figured out how to complete the course there's a few little hiccups because the materials getting dated and instructions aren't accurate and haven't been updated but it's okay the work world is like that in spades and it's just part of the process you just have to work through it so in conclusion this certificate is well worth the time effort and minimal cost and i highly recommend it if you're on a path to becoming a data scientist and so thanks i hope this video was helpful and if it was great you can hit me up on linkedin or subscribe to this channel thanks thank you for watching and please if you found this video helpful click like and be sure to subscribe below you

Channel: DataResearchLabs

Views: 8,186

Rating: 4.9290781 out of 5

Keywords: data science, ibm data science, coursera

Id: wl0X8cCEz00

Length: 25min 7sec (1507 seconds)

Published: Wed Jan 06 2021