We all know about the importance of hypothesis testing in the science world.
Ok! I can sense some uneasiness in the room 😊
Do not worry. We are not going to talk about boring science experiments here.
We are just going to borrow the word from them and use it to illustrate its importance.
Now, as I gain more experience in the corporate world, one thing is beginning to stand out. You need to know what you are looking for. Or at least have an idea. This idea can change. In many cases, it will change.
So where is the connection then to having a hypothesis?
In today’s business world, there are many software out there that make our lives easier. However, in many cases, it is more or less garbage in garbage out. You get what you give.
So how can you avoid the “garbage in garbage out” syndrome and make sure you stand out in the workplace?
Well, you need to make sure you move from just running the simulation or working with the software to a combination of understanding the software and also being vast in the interpretation of the results. Remember, you could rightfully draw logical conclusions on the wrong output.
This is where having a hypothesis comes in.
Having a hypothesis is powerful in the sense that it prevents you from making conclusions on the wrong output, which could have severe consequences on your reputation for instance.
Let us look at an example.
Would bring in the old enemy here – maths. But trust me, it would be easy to follow.
Let us assume that I have a linear equation:
This quite straight forward. My output should be a straight line (with a positive slope) with a y intercept of 2. Therefore, option A (see image below)
Now, this is your hypothesis. Before you even start your analysis, you have an idea about the shape of the output. This is important because in the case, where the output is not what you anticipated or if there was a mistake in the data entry, you would be able to call it out.
Of course, I know that there are instances, where this is practically impossible, but in the cases where you can, do it – try to think of what the outcome should look like. No need to be precise here.
With this in mind, I get to work.
For this case, I would use Wolframalpha as my online software. (Just as a side note, in case you have never heard of this, you just hit gold
). Say bye-bye to your maths worries. An alternative option is Cymath. They basically do almost the same thing.
Back to the topic.
I put in the formula in Wolframalpha and I get the graph shown in the image below.
As you can see above, this fits well with what I was expecting. Now I can go on to analyse whatever I want to analyse.
Let us assume I had no idea what to expect. So classical garbage in garbage out situation and unfortunately, in my data input, I mistakenly typed in “minus” in front of the 5. That is instead of:
I typed in:
The minus in front changes everything, giving me an output shown in image below.
Instead of a positive slope you get a negative slope. Fundamentally, both solutions are correct, but you are not looking for a negative slope. The output is as a result of a mistake.
If you had a hypothesis of what you expected the outcome to “look like”, you would challenge this result and avoid embarrassing yourself in front of your boss. However, if you did not have any clue what the output was supposed to look like, you would go ahead and analyse a wrong output.
In general, though not always easy, going in with an idea of what to expect would keep you in check and is one of those skills required to make you stand out.
Try to put this into practice at work, school or whatever you do.
Till then, Dan
Leave a Reply
You must be logged in to post a comment.