Challenge #141: Examination Data Simulation

[JoeM]

Generating sample data to represent a population is key for many types of analyses! In this workflow, we have a list of 1000 people and we want to simulate a a test score for each person. For each person, randomize a score for the exam, but we're not going to make it that easy!

You'll need to generate the simulated scores under these conditions:
1) The lowest possible test score is a 50%, while the highest is a 90%
2) The simulated score must be 'randomly' generated as a normal distribution, where the population mean is 70%.
3) After creating the simulated test results, build a chart view to prove you have achieved a 'random' normal distribution by binning each score to the nearest point.

Stuck? Hint 1

Spoiler

Did I say 'simulation'?

Hint 2

Spoiler

You might have a hard time without the free predictive tools installed.

Hint 3

Spoiler

Oh, cool! There's a tool that simulates sample data! *Cough* Simulation Sampling *Cough* *Cough*

[patrick_digan]

Spoiler

[neilgallen]

Spoiler

[Justin_0]

Spoiler

Thanks to that one tool, that was much simpler than I thought it would be.  Just had to figure out how to get it configured, then it was all down hill.

[MichalM]

Great example of an iterative development...build, run, check, tweak...run again, check, tweak etc etc  Got there eventually.

[ggruccio]

Mine looks a lot like the picture - first time using this tool!

Spoiler

Not sure why 5 standard deviations seemed to work. It just was where the extremes became 50 and 90.  Was trying to calculate what it 'should' be, but began experimenting with the setup...again limited experience with this tool....will have to read more about it!

[kat]

It was time to try out the Python tool :)

Spoiler