I wasn't sure whether I should bump this again as I'm already in the process of writing about this experiment, but I suppose getting some more data while writing won't do any harm.
Several hours and over 2500 words later, I've written up a report on this experiment! http://folk.simpsite.nl/pick-a-number
Well done. Your method is already quite developed. You are in highschool now? Ive seen bachelor students not being able to come up with ideas how to make such tests. For future work I suggest starting with a thesis. What you did is the test you would normally do to prove or disprove a thesis. This is not critc, you just did it for fun. But since you had this "can this be used for anything" I figured I mention it. Scientific paper usually go I. introduction to the problem and motivation (why is this important to know) II. List of hypothesis (what question will be answered by the paper -> doesnt matter of proven or disproven) III. Explaination how an experiment would work that would give the answer (Thats where you started) IV. Review of the results and interpretation (Also done very well by you) V. Summary and Outlook (Also done quite well, however due to the lacking thesis no real answer could be answered) Searching for the thesis after the experiment (or rather look for something interesting to happen) is usually bad practise as it could lead to biased insights. Again overall good work ! Getting used with the tools like these analyticcs tools before going to college/ university is pretty good. It will help you alot.
Haha, thank you very much! I know, but thanks for outlining it that well! I actually had to do a real 'thing' that did follow these rules for school earlier, I did it about RSA-cryptography, and it went quite well. This time though, I did it just for fun, and when I started I didn't know I would take it so seriously later on. You're right though, I think it's definitely good to have some practice doing things like this, especially when looking at possible future studies in university! And hey, if I can combine fun with it too, that's great, right?
Neat Pi = 3.14159265359 You could've decided to count that as 3.1 or 3.14 Hmmmmmm, you are not by any chance meaning me here, right?
I can't say I really agree with your interpretation of the data in a lot of instances, sorry. It feels like you looked at the data and then just kinda made up what you think it means. The lack of an initial hypothesis only contributes to this. You identify a pattern well enough, but you pull an explanation for it out of thin air. You notice that '6' is a relatively unpopular choice in the "1-10" interval, and your conclusion is it has to do with '666', the "Number of the Beast"? The number is '6', not '666', and I bet you '666' would've been a very popular choice if it was included in any of the intervals. You even make the comment later that '99' "...might also be appealing because it features twice the same digit". Also, in the very next interval, '6' is actually in the top five choices, but you fail to notice that. You make the assumption that people didn't understand that border numbers were allowed, arguing that this explains the unpopularity of '10' in the "1-10" interval. You then immediately point out how popular the other border number is in that same interval. My biggest problem, though, is with your very final thought. You say that "...humans are bad at randomness. Even when trying to be random, there are always certain tendencies at play". This is your takeaway from the experiment, your conclusion. Well, when did you ever ask us to pick "random" numbers? All you asked for was numbers within the given intervals. There was no expectation for "randomness". Of course I, and I'm sure many others, picked numbers deliberately. I picked numbers that I like, or have some kind of special meaning to me. If the prompt was to pick "random" numbers, I bet there would've been a lot less 7's, in fact maybe unnaturally so, since people might try to avoid it to "appear random". That could be an experiment on human randomness. This was not that, though. This was more about human psychology, and unfortunately I'm not exactly sure what we've learned from it. Sorry if this comes off as rude.
Nah, it doesn't come off as rude, just like you're looking too much into it. Do you think I should add a disclaimer in at the top, stating this isn't at all a scientific study and just a write-up I did for fun?
I liked this! Please, please,please Corq: do not edit your post. Same for you 607. I agree with Corq up there, and I seriously think he added a lot of value to the thread with his comments. Sure: the intent was nothing serious, but that doesn't mean that there's nothing to learn from Corq post (a classic mistake (too (IMO!!) many people make). So please leave it in there! I laughed so hard here: (left the emphasis in). Oh my... This brings back memories I recognize the 0,5 Guilty by association! (it makes no sense here, but considering 607, Jelle, Tom & me are all Dutch it has a nice ring to it) Be warn EMC, us Dutch are taking over (a little bit at least, please? ) Because as we say in Holland: Als de niet niet niet dan niet de niet niet! (take that!). Translation () When the staple doesn't knit then the staple doesn't knit See how our languague rules here? (uhm.... Let's not take this too seriously, even I could get a headache from reading that Dutch mess) Lottie rules!! ha ha ha, and I love the 42 too You might not see Lottie as often as me on the forums but she's there when it counts (and I mean every part of that seriously!). As to the rest.... lots of numbers, for a Sunday evening We'll see next week
I just noticed that the results of the poll make an almost perfect bell curve. EDIT: So I started looking at your documentation, specifically for the first few graphs. While I do agree with some of your statements about why some numbers were chosen more than others, I feel like there's more to it. In the interval [1, 10], I believe that people are less likely to choose border numbers for two reasons: The reason you stated, that they may not have known that the interval is inclusive They want their responses to be as random as possible. Numbers that are more friendly looking and common like 1, 2, and 10 were chosen less often because those responses wouldn't be as "random" as another number like 7 or 9. I also find it surprising that there was such a large amount of outliers, such as pi or those that don't fit within the interval. I feel that these should have been documented on their own (They may have, but I haven't read the whole study yet) because of behavioral traits. For example, there's always going to be that one person that wants to be "different" or "funny" so they choose an oddball number that would stand out from the rest. I know that these behavioral traits do exist in people, but I didn't know just how common it would be. This is just my small personal input on what I've read so far.
Them's figjhtin' words Disclaimer: I have no idea where this pic came from. This dates from the time (uh,oh, granpa Shell talking ) where there was hardly Internet and we we're all happy in our IRC channel kicking the n00bs out
Thank you, I agree! I don't mind Corq's comments at all, it was quite interesting to read his view. However, to me it did come off a little as 'reprimanding' me at first, but that isn't a problem, we figured it out alright. I agree! It would've been good to look into that. But well, at this point I had already spent quite some hours on this little thing I did for fun, so I chose for a more superficial review.
