Timeline for Why does the proportion of native language speakers have an arcsine like distribution?
Current License: CC BY-SA 4.0
20 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 31, 2020 at 21:34 | comment | added | Sextus Empiricus | @StatsIT the majority of those average 20 per district do not make it to the plot. I guess that there is only 15-20k in the plot, and 80k are not plotted because they are below the 2% mark. | |
| May 31, 2020 at 21:12 | comment | added | Stats IT | No it doesnt make the plot almost entirely about the first and second sub district because on a average we have about 20 non-other languages in a sub-district. Yes will try to plot on a map, issue is unlike state borders, I don't know if there are geo coordinates for districts or sub districts in an interactive map for India | |
| May 31, 2020 at 20:56 | comment | added | Sextus Empiricus | You should make a map where you plot the percentages of the first language. This will give you an insight how the process works. | |
| May 31, 2020 at 20:54 | comment | added | Sextus Empiricus | @StatsIT that makes that plot almost entirely about only 1st and 2nd languages. Probably a few regions will have 3 or 4 large languages which is pushing that left side a bit higher up than the right side but the difference between left and right is not much. So what you see mostly is on the right side the proportion by which the first language is spoken, and the left side is nearly a mirror image of this. | |
| May 31, 2020 at 20:48 | comment | added | Stats IT | I found that in the data, they have treated outliers at the left side. If the populaiton of a language in sub-district is too small compared to others, they have been categorized as All Others in that sub-district. E.g. If a sub-district has 1 million hindi speakers, 500000 marathi speakers and 23 chinese speakers, then Chinese are counted as others since did not make sense to have separate statistics for just 23 chinese. The cutoff of others at sub-district level is below 2% of the total population of the sub-district. So the plot is essentially for proportion of the non-others languages. | |
| May 31, 2020 at 20:16 | comment | added | Sextus Empiricus | The left and the right half (divided by the 50% mark). The right side can contain at most 5923 points, so the left side must contain the rest of the 105961 points, which is at least 100038 points. So those two sides should not look like having equal weights (as is the case now). What you probably have is that the points with low percentage are not plotted (the plot is missing two bars close to 0%, but that is where the majority of the points are). So it is far from an arcsine curve. | |
| May 31, 2020 at 14:43 | comment | added | Stats IT | Left side of what?? Didn't get you? Or do you mean the leftmost? | |
| May 31, 2020 at 14:31 | comment | added | Sextus Empiricus | @StatsIT So how many points are on the left side? | |
| May 31, 2020 at 13:56 | comment | added | Stats IT | Yes theoretically, at max 5293 points can be on the right side.. All points were plotted. | |
| May 31, 2020 at 12:55 | comment | added | Sextus Empiricus | @StatsIT but at most 5923 points can be on the right side of the graph not? I would expect the left sight to have more weight, unless the graph represents only 10/20k points. It seems like the zero (with 80k or more points) is not plotted. | |
| May 31, 2020 at 12:48 | comment | added | Stats IT | Because there are 5923 sub-districts, but each one from this sub-districts can have between 1 and 105 languages to while one sub-district can have one data point for its only language, the next sub-district can give 105 data points, one for each of its language. | |
| May 31, 2020 at 12:28 | comment | added | Sextus Empiricus | @StatsIT How many points are there which have >50%. These can only be at most 5923 points right? Because there can never be two or more languages in a region that are simultaneously over 50%. Given that consideration, how can that curve represent 105k points if the right side, which looks almost half the points/weight represents at most 6k points (and might be less since the dominant language can be less than 50%). Could you provide more background about the data, or maybe the data itselve. Are most of those 105k points zero and not plotted? | |
| May 31, 2020 at 11:10 | history | edited | Sextus Empiricus | CC BY-SA 4.0 |
added 1706 characters in body
|
| May 31, 2020 at 9:18 | history | edited | Sextus Empiricus | CC BY-SA 4.0 |
added 207 characters in body
|
| May 31, 2020 at 9:14 | comment | added | Stats IT | I wonder if we can have theoretical derivation or justification to show that under such scenarios, we will get an arcsine distribution like in the case of Brownian motion. | |
| May 31, 2020 at 9:12 | history | edited | Nick Cox | CC BY-SA 4.0 |
added 8 characters in body
|
| May 31, 2020 at 9:12 | comment | added | Stats IT | My intuitive reasoning is exactly inline with your thoughts. There are distinct 5923 sub-districts. There is no bilingual because the census only counts 1 mother tongue per person. | |
| May 31, 2020 at 9:07 | history | edited | Sextus Empiricus | CC BY-SA 4.0 |
added 2 characters in body
|
| May 31, 2020 at 8:59 | history | edited | Sextus Empiricus | CC BY-SA 4.0 |
added 361 characters in body
|
| May 31, 2020 at 8:53 | history | answered | Sextus Empiricus | CC BY-SA 4.0 |