My 2nd Weekly Reflection: How Being Curious Opens the Door for New Knowledge

My 2nd-week reflection as Data Analyst participant in Generasi Gigih by Yayasan Anak Bangsa Bisa (YABB)

Ilyas Perlindungan
4 min readJul 18, 2021
Photo by Joakim Honkasalo on Unsplash

I can not believe that my second week of the Intermediate level in Program Generasi Gigih by Yayasan Anak Bangsa Bisa (YABB) is over. Time is flying so fast but I am so glad and happy that what taking up all my time is so advantageous (Being part of learning community, this program). In case you don’t know what this program is you can check my previous article since I explained a little bit about this program there, check that out.

Last week, I was very excited knowing that one of the session’s materials is Statistic. Since senior high school, I have always thought I have been fairly good at Statistics. However, that is when I have to attend a statistic session last week. After only a few minutes the session begins, I realized that Statistic that we have in this session is on a whole other level compared to Statistics that I used to know. There is so much more than just mean, median, mode, percentile, range, standard deviation that I’ve known before the class. Not to mention, there are things like p-value, hypothesis testing, normal distribution, how to design an experiment, how to differentiate whether a certain type of study is experimental or observational, how to use standard deviation, and a lot of other things that I can not list here one by one. But the point is, there are so much more that I still don’t know regarding Statistics.

There’s one time when the instructors opened the Q&A session and I asked this one question:

When we can safely assume that our data is normally distributed?

Source: Image from here

Actually, the real question is quite longer than that, involving things like p-value and significance level. But, that’s the main idea from the question. Kak Adji, one of the instructors, gives a fairly good answer and a bit teasing about this question that this question is quite technical and experience will give me more understanding on this. I think he is right.

To be honest, the whole session is so fun and compact. While also there are lots of questions from other students that I think it makes maybe a 3-hours session not enough to answer it all. But, the instructors are so kind they please anyone to reach out after the session. Ask them if there is any confusion. So, one day after the session I sent a message to Kak Kevin, one of the instructors, about my question (He is so kind :D, actually he’s the one who reminded me about my question to ask it again hopefully he can give a better explanation).

Turned out that my question has something to do with sampling distribution. It’s going to be so long if I explain it here. But in a nutshell, what Kak Kevin means is that real-world data is not always normal (hmm, I also began to think maybe there are a lot more kinds of distribution) and it’s impossible to collect and process the whole population. So with that in mind, we can do sampling (sampling distribution) a certain times with each sampling has a certain size. Let’s say 100. This sampling process that we do certain times we take its mean. The mean will be close to normal (the mean of the sampling means will be close to the real population’s mean). The sampling distribution means has mean that is close to the population mean, also its standard deviation. We can take these two as parameter estimation for further analysis. (Note: the more times we do sampling the closer its mean of means to the population’s mean)

Source: Image from here

As I said before, Kak Kevin said data is not always normal and there is a lot more kinds of distributions other than just this one normal distribution. Intrigued, I decided to learn more. I found there are a lot more things like poisson distribution, exponential distribution, discrete uniform distribution, continuous uniform distribution, binomial distribution, probability mass function (pmf), cumulative distribution function (cdf), maximum likelihood, etc.

Wow, that is a lot.

Summary

Okay, to cover up I will list some points as a summary from this weekly reflection:

  • Let the curiosity arise since in the end that will give you much more understanding.
  • Never shy to communicate what you’re curious about.
  • Ask that. Do not fear being judged that your question might sound ridiculous. All questions are valid.
  • Do not feel enough with what you’ve learned from the class (we called it a session here), do more research (more learning) after the class. Since most of the class can not cover all the materials due to time constraints.

Alright, maybe that’s for this week’s reflection. I would like to put this one quote here. Thanks for reading.

Reflective thinking turns experience into insight.

— John C. Maxwell

--

--