That’s a misleading Y-axis. The difference (in self-reported subjective happiness rating 1–10 too) is not as significant as the graphic implies visually.
The difference (in self-reported subjective happiness rating 1–10 too) is not as significant as the graphic implies visually
Ah here is another one. So what? It makes the difference more distinguishable, which also the graph denotes numerically. Otherwise all Linux distros users would appear too flat to make any difference interpretable.
The fact that there are at least two such comments around here shows why teaching anything in schools is doomed to fail.
Even critical thinking skills are applied in a canned, thought-terminating fashion, similar to how XX/XY chromosomes are considered the only reality, in overconfident falsehood.
Ah the statistical significance, which as everybody knows is assessed …visually? Mic drop
BTW I have another comment here, totally irrelevant to this discussion, that I bring up statistical siGnifiCAnsE as an example of confident falsehood. Thanks for proving me right lol
Layman statistics is not the hill I would die on. Otherwise (being guilty of the fallacy myself) I now think that making a subject mandatory school lesson will only make people more confidently incorrect about it, so this is another hill I won’t die on for probability and statistics. See for instance the widespread erroneous layman use of “statistical significance” (like “your sample of partners is not statistical significant”) you see it is a lost cause. They misinterpret it because they were taught it. Also professionals have been taught it and mess it up more than regularly to the point we can’t trust studies or sth any more. So the solution you suggest is teach more of it? Sounds a bit like the war on drugs.
I’m not trying to do layman statistics. I’m stating that the graph obscures the magnitude of the difference between people’s ratings. You seem to be suggesting that I’m trying to do some kind of formal or specialist critique of the graph. I’m not. I didn’t go to school or anything and I am simply making a lay observation. If you disagree with it you can say without being so obfuscating yourself.
Although there is a common tip in critical thinking classes that manipulating the Y-axis range can lead to misleading presentation of a difference, I believe in this particular graph, which clearly provides numbers to compare, you can’t say it is misleading.
People can read and compare the values and draw their own conclusions. And I am saying that without any consideration of the distros discussed, since I am impartial to distros, I like all distros I have tried.
This “study” almost certainly must have way deeper assumptions- and metrics- related problems to start with, so even finding myself having this argument is preposterous. But I am just pointing out the misapplication of critical thinking guideline, and this is a valid point which I insist everyone who relies on to consider, if you care about critical thinking at all.
No one said you are doing layman statistics, the pasted comment is from another discussion, provided here for context, and for very good reasons. It aligns with obvious misconceptions about statistics that should be pointed out. Probability and statistics are thorny subjects that nonetheless are inevitable in order to understand the world surrounding us, material, social, and economic, so yes I will nitpick here and call out the misapplication of canned critical thinking thought-terminating cliches.
I am not trying to apply a “critical thinking guideline” I saw elsewhere. I’ve not taken any “critical thinking classes”. I’m more insulted that you think I couldn’t have possibly just thought of that comment myself. It’s not a particularly crazy comment to make, and I don’t see why any individual who knows how to read graphs couldn’t just happen to make that comment.
Anyway—sure, I never said the graph lied. Perhaps a better wording would be that, regardless of how the information is presented, I don’t think the difference in magnitude between people’s happiness ratings (ignoring the issues with how those ratings were collected and ascertained in the first place) is significant or particularly of note. The Y-axis is chosen so as to visually amplify this difference. I didn’t claim the data presented by the graph was untrue or that reading the graph correctly was too difficult if one wanted to read it properly.
I really did not mean to be insulting. I am just saying chart makers can choose to make a zoom in, and it is not automatically propaganda or something. All this has led people astray of the real issues, like WTF is measuring ‘happiness’ on a 1-10 scale, and what are the metric properties of this 1-10 scale. Then there are all the sampling issues and what have you. I just expected more people discussing this stuff rather than the Y-axis.
I didn’t say it was propaganda—the content of the graph reads as quite clearly silly to me and not trying to make a particularly serious or scientific point. I guess the same reason is why I pointed out the Y-axis instead of the sampling issues, because the sampling issues seem much more self-evident.
There’s not a lot of data to work with, and the kind of test used to determine significance is not the same across the board, but in this case you can do an analysis of variance. Start with a null hypothesis that the happiness level between distros are insignificant, and the alternative hypothesis is that they’re not. Here are the assumptions we have to make:
An alpha value of 0.05. This is somewhat arbitrary, but 5% is the go-to threshold for statistical significance.
A reasonable sample size of users tested for happiness, we’ll go with 100 for each distro.
A standard deviation between users in distro groups. This is really hard to know without seeing more data, but as long as the sample size was large enough and in a normal distribution, we can reasonably assume s = 0.5 for this.
We can start with the total mean, this is pretty simple:
This makes a total sum of squares of 0.471. With our sample size of 100, this makes for a sum of squares between groups of 47.1. The degrees of freedom for between groups is one less than the number of groups (df1 = 10).
The sum of squares within groups is where it gets tricky, but using our assumptions, it would be:
<span style="color:#323232;">number of groups * (sample size - 1) * (standard deviation)^2
</span>
The degrees of freedom for this would be the number of groups subtracted from the sum of sample sizes for every group (df2 = 1089)
Now we can calculate the mean squares, which is generally the quotient of the sum of squares and the degrees of freedom:
<span style="color:#323232;"># MS (between)
</span><span style="color:#323232;">47.1 / 10 = 4.71 // Doesn't end up making a difference, but just for clarity
</span><span style="color:#323232;"># MS (within)
</span><span style="color:#323232;">272.25 / 1089 = 0.25
</span>
Now the F-statistic value is determined as the quotient between these:
To not bog this down even further, we can use an F-distribution table with the following calculated values:
df1 = 10
df2 = 1089
F = 18.84
alpha = 0.05
According to the linked table, the F-critical value is between 1.9105 and 1.8307. The calculated F-statistic value is higher than the critical value, which is our indication to reject the null hypothesis and conclude that there is a statistical significance between these values.
However, again you can see above just how many assumptions we had to make, that the distribution of the data within each group was great in number and normally varied. There’s just not enough data to really be sure of any of what I just did above, so the only thing we have to rely on is the representation of the data we do have. Regardless of the intentions of whoever created this graph, the graph itself is in fact misrepresent the data by excluding the commonality between groups to affect our perception of scale. There’s a clip I made of a great example of this:
This is an egregious example of this kind of statistical manipulation for the point of demonstration. In another comment I ended up recreating this bar graph with a more proper scale, which has a lower bound of 0 as it should. It’s suggested that these are values out of 10, so that should be the upper bound as well. That results in something that looks like this:
In fact, if you wanted you could go the other way and manipulate data in favor of making something look more insignificant by choosing a ridiculously high upper bound, like this:
But using the proper scale, it’s still quite difficult to tell. If these numbers were something like average reviews of products, it would be easy in that perspective to imagine these as insignificant, like people are mostly just rating 7/10 across the board. However, it’s the fact that these are Linux users that makes you imagine that the threshold for the differences are much lower, because there just aren’t that many Linux users, and opinions wildly vary between them. This also calls into question how that data was collected, which would require knowing how the question was asked, and how users were polled or tested to eliminate the possibility of confounding variables. At the end of the day I just really could not tell visually if it’s significant or not, but that graph is not a helpful way to represent it. In fact, I think Excel might be to blame for this kind of mistake happening more commonly, when I created the graph it defaulted the lower bound to 6. I hope this was helpful, it took me way too much time to write 😂
Oh sport, and I thought I was the one beating on a dead horse here. I understand why people claim to take issue with the Y-axis range. I am just saying chart makers can zoom in to make a point, and it is not automatically misleading. That is all. Anyway, thanks for writing this. Looks like a lot of effort, and some of it will make sense in my stats coursework, thanks!
This is very disappointing. I’m glad lemmy is federated and open source, so that I don’t have to endorse or support every single view of the creator to use it.
Nutomic’s view is very misguided. The section of the bourgeouisie that expresses support for trans people often do so out of opportunism (though some probably genuinely support it They’re human after all). The fact that transphobia is still rampant should be proof of the lack of an “agenda”. And what is this agenda about anyways? Acceptance? Or are people really still thinking that there’s a push to turn cis people trans the same way we have done the reverse since forever?
If you have 1 backup, you have no backups; If you have two backups you have one backup.
Just make a backup of your data, it’s the easiest way. Can it be done without a backup? Yes, but it’s complicated and error prone. Better go the safe route.
Yea but that’s boring, it’s more fun to YOLO it! Shrink the windows partition from Linux, create the new Linux partition and install, then move everything over from the NTFS partition to Linux, delete it then expand the Linux partition.
Ez-Pz and the added pressure makes it more fun! Lol
My opinion is that it comes down to people not caring as much.
Yeah, that seems obvious, but bear with me here.
Know what the biggest difference there is between knives and guns? It ain’t how they do the job as weapons, though that is a big one. It’s that guns all work the same, so if you start banning one kind, it ends up applying to far more than anyone that thinks of firearms as an extension of the right to defense, a core human right, is willing to accept.
You ban switchblades, and there’s still fixed blades, slipjoints, lockbacks, etc. Nobody has ever tried to make the kind of laws you see around knives in other countries, but if there was an attempt, I don’t think there would be as much passion against it as there are with guns because there’s just not as much concern about knives as part of the right to defense because guns exist. Range > melee 90% of the time.
Firearms are the gold standard for personal weaponry. So out of the peeps that care about the enumerated right to keep and bear arms simply don’t think about knives as much. Same as they don’t think about bows, or swords or halberds. They don’t care because the fight isn’t as relevant to them.
And, on the other side of things, because guns are the gold standard, you don’t have as many people targeting knives. There’s less to fight against
Now, if guns disappeared, see how quickly crossbows and swords started being the target as people shifted to them instead of guns. That’s where the fight would go to. Ban those, and there’s going to be steps all the way down until the debate is about how big a rock you are allowed to have.
Also, because of that lack of give-a-shit, knife laws aren’t draconian everywhere. There’s some states where you can own and carry any knife you want. Others only ban knives that can shoot across the room, or other specific, niche types
Also, I think you’re underestimating how easy it is to get a pistol. If you go in without the work done ahead of time, you ain’t buying a pistol in half an hour. I don’t think even Texas is that loose. Background checks take at least that long. Maybe I’m behind the times om something, but even “shall issue” permit states, you have to count getting the permit in the time factor, imo.
As I understand it, NICS checks take minutes now that everything is done electronically. The more comprehensive so-called “universal background checks” are only required in a minority of states.
30 minutes is probably on the longer end for the whole process. So, you may be behind the times a little bit on this one, but idk for sure, and obviously it varies from state to state.
I know my friend bought a hunting rifle in about 15min last year to go deer hunting for the first time, because I went with him out of morbid curiosity. I think if you’re over 21, a handgun purchase is identical.
The background check was so fast I didn’t even know they did one until I told him I thought they need to do a background check and he told me they did it while they were packaging his gun and ringing it up. I thought it was like in The Simpsons where it’s a few days, where Homer goes “aww, but I’m mad now!!” Lol
That scene with robots trying to get into the Infinity is basically copy-pasted from The Matrix (a homage, whatever you want to call it). Even the weird electrical pylons on the Infinity look like Nebuchadnezzar and other ships in The Matrix.
Well… You’ve been dealt a bad hand and I’m not sure how to help you. You can do the IT career path via educating yourself but that’s not easy. It’s hard to cut yourself above the rest although it’s possible.
First thing you could do though is to accept things for how they are. People are going to throw trash until someone with authority will tell them it’s wrong and then you’ll need a generation for the change to happen.
You’ll not be able to fix India by yourself and I’m not sure if anyone can but over time things will slowly get better. Emigrating is an option, although hard and going into politics can work but that’s even harder.
But honestly, I feel like this is all things you already know. These are my completely useless two cents.
I need to read Man’s search for meaning. Ughh… sorry, that came out of nowhere, what I meant was, I need to know how to deal with things when I have been clearly dealt a bad hand by reading the work of a man who had the least reasons to see hope!
These are my completely useless two cents.
Trust me, just the acknowledgement of adversity of another helps (as long as the victim tries to acknowledge/accept the reality and work hard on improving it rather than drowning in their own miseries and the pleasures of being a victim)
For sure, realising your situation for what it is is an advantage. It sounds like you’re the type of person to get an advantage out of it. Like for example strategic land buying in India if you have the money can make you relatively wealthy over time. I want to say money isn’t everything but given I’m from Europe I feel like my opinion doesn’t apply.
Either way, I’m sure you will at the bare minimum have an average life given how much aware you are so you got that going. Best of luck man
Holy jam, I saw this and had to comment. So idk about rooting a Motorola because it's not a "mainstream" Android (as in pretty much Samsung) nowadays. But perhaps Legacy in that situation. But I did have a solution for you if you were interested that doesn't require rooting (as far as I know). I was goobing around on the internet and found this man who was talking about the perfect minimalist phone. And figured when I gave a shit I would mess around with ADK (I believe it is) and this git-repo. But the lovely Jose Briones apparently also supports this. So give it a check if you want because I remember seeing Motorola explicitly being used in both the OG and the offshoot.
Anyways - less invasive phone while not being totally degoogled and the battery lasts longer. Hope you dig in cause I think you'll end up really liking it. GL!
I'll take persona, although it's been way too many games with the same setup. Ditto for the Trails series.
Honestly, I don't think it got any better than ATB systems in FF 6 and 7. Everybody else is either riffing on those or spending so much money they think they can't be those and need to be Devil May Cry instead.
Not sure if it’s already covered in another comment or if you are already aware, however … Depending on where in the world you live, there may be a semi-local official UPS drop off location at a store or business near you. I live outside of a very small rural town, yet there are UPS drop off options here. They even print, text, or email a receipt to confirm they’ve received it.
Still not as convenient as an at-home pick-up, but it’s something to consider if you didn’t already know.
kbin.life
Active