Analytics Vidhya
Published in

Analytics Vidhya

Common Skills of Data Analytics Tool- From a Noob, for the Noobs

DATA…DATA…DATA…Analyze..Transform…Predict…Model…Story Telling

For anyone who was interested in Data Science, Machine Learning, Analytics or ‘Mr. Robot’, they would’ve have heard these words multiple times. If you’re one of the unlucky ones who wanted to learn about it then you would’ve read or heard these so many times that yours have ears start to shrink. Worse still, there are so many tools-so so many and so many courses-so so many that it becomes difficult to pick up things fast. Every time you think you have learnt Tableau by watching their resource video, someone scares you away with a question like ‘Where is the trend line?’.

Up yours….that is where it is. –Anonymous Jerk

Nevertheless, I was trying to learn Tableau, R ,Python and Power BI and discovered a pattern which I would like to share. I hope it will cover the basics of what to look for while picking up a tool and start using it for a project. Later, if I survive the corona virus and the heavy work pressure, I would like to share how a new tool like Alteryx was picked up.

Please be reminded that I am a newbie. Also, the skills are not in the order.But without further ado, here we do-

1. Loading Data- Yes, this is the most basic skill. Duh!! But that is the truth. With a multitude of data sources, you can get overwhelmed. So just learn how to load an excel, a .csv and data from the server. There are numerous other sources. But be a proud noob and be ignorant about it. There will be time to pick them later.

2. Joins and Unions- Once you have that data loaded, you can go explode it. If you can pardon the bad attempt at rap, know that your data can be in form of multiple tables. For example, there can be table with data on what Mr. Trump had for breakfast and other table with data on when he threatened other countries. If you want to find out the relation between his breakfast and threats, you have got to JOIN these tables. So, learn how to join. If Mr. Trump eats a lot or threats a lot, a single table might not cover the complete eating/threatening and be spread over multiple tables. That’s where UNION comes in.

3. Pivoting Data- Summarizing your data is the best. And it may very well take the least effort if you a dataset which has showered properly. But being able to change the data from wide format to long format or vice versa may be the only critical thing you do to the dataset.

4. Creating new columns and Editing columns- Imagine being on a journey to being the best manager of moody rockstars. Lets say you have data on how much all the rockstars earned and your moody broody rockstar made the most. However, you know the best way to tell him is to showcase by how much he beat all the other rockstars. So instead of telling him how much he made, you tell him how much more he made from his competition. This skill of creating new information by transforming old information is the best.

5. Splitting data, Combining data and Extracting information- Sometimes the information plays hide and seek. Like the name ‘ Sir Spitsa Lot’ being stored in multiple columns taking all the fun out of it. Or on a serious note,the gender information being hidden in the prefix ‘Mr. /Mrs.’ of a names table. You will need to learn to split a string, combine multiple strings or extract parts of it to get your useful data. This is an important skill.

6. Mathematical operation on rows- Maths –Boo. For all that are as afraid of the word as I’m of the doll from the movie Annabelle should take it easy. First off, Math can be uglier than the doll and secondly, you do not need to know all of it. Yes, simple operation like summation, division, multiplication, finding averages and means is mostly what you need. And all the tools already have functions to calculate them. So, take a chill pill. If you’re afraid, please know that your fear is just a click of the button away from being destroyed.

7. Handling null values- Yes, there will be important places which may have been left empty in your data. And like the global warming situation, you can do something about it. Slackers like me can just drop the records. If the one inputting the information didn’t care, then why should I. But if you have a job at stake, just learn how to fill these spaces with things like Means, Most common values and other good things.

8. Filtering and Sorting data- The ability to choose what parts of data you need for your analysis is important. Step it up a notch to learn to apply functions to data which fits your criteria. Let us say you have a list of your friends and you want to assign all your friends whom who have known for more than 5 years as ‘My Bro’. That is the basic of filtering you want to learn. As for sorting, it is the heaven defying ability of putting Edison before Tesla because E comes before T. Yes, it is all about ordering and re-ordering. But it is also about grouping things. Don’t we all like restaurant menus where all foods Chinese are under the same tab. If you can add those Manchurian chilli nuggets to that group in your dataset, you are good to go.

9. Logging your workflow- As boring as it may sound, keeping documentation on your work and understanding how to maintain different versions is important. Just using comments, annotations, following a good file naming convention and folder organization go a long way. Besides, if you were to write a tale of your heroics you would be as explicit as possible. Have the same mindset here.

10. Visualizing information- Humans are visual learners. They are also greedy and a threat to the planet. But they are also your audience. So we need to familiarize ourselves with the different ways data can be visualized and presented using the tool. Be it a line chart, a bar chart or a scatter plot, much like Pokemon, you gotta get them all.

11. Outputting the result- Duh!! Again. Why go through all the hassle if you cannot parade around basking in glory of your genius.

There will definitely be more. But all good things take some time. There are also other skills being endorsed which are not dependent on the tools like presentation skills, choosing good themes etc. I am struggling with that as but the best way I can recommend is to try to be as cool as Barrack Obama. That Man can sure deliver any presentation.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store