I like that they have to use median, because I'm often having to find examples of why you'd use that over mean. My wife for instance is over 5k miles from her mom, and I imagine that's true of most immigrants. So the skew is probably really, really strong if you didn't use a non parametric measure like the median.
In other words, I'm stealing this for my stats class.
Yeah but you can't really do that with a map. In a table you could. A report would likely report both, but also differentiate groups because you don't usually want to report skewed data without explaining why.
Coincidentally, 18 miles is almost exactly the maximum distance you can be from the sea and still be in Denmark.
Not advocating throwing your parents in the sea, just bragging about one of the advantages of living in a tiny country consisting mostly of sea and islands 😁
Would be interesting to see the methodology on this, or to see how much it has changed since this was published, or to see it broken down at a county level.
I also want to see the median, mode, and quartiles... for the last 25 years. I wonder if the recent rise in individuals living with parents/relatives has had an influence.
It makes some sort of sense imo. What the graph is essentially saying is that most people stay in the same city as their parents. An 18 mile rage around a house covers most of a city. It's probably thousands of people who live in the same neighborhood as their parents skewed by the handful that move out of their home city.
You should see Belgium, many children refuse to move out of the same village as their mothers, much less 50km. It is probably similar in italy for example. Would be interesting to see state on Europe and Asia too!
The median distance without anything else is worthless. The median only represents the "typical American" if we're working with some kind of bell curve, which can't be true since a sizable chunk of people still lives with their parents. For all I know 49% of people could be living with their parents and another 49% could have moved far away - the median still only shows a data point from the remaining 2% of people.
If the data is normally distributed (as in bell shaped), the mean and median will be the same. The problem is that the data is probably very skewed, so the median is probably a better representation of the central tendency than the mean. Also, it's incorrect to say it only represents 2% of people; it's more that every person is only represented by their position in the set. Quintiles, quartiles, and percentiles work the same way. I like to think of it as everyone being a single vote, unweighted by value.
That said, if you wanted another acceptable alternative, you can also remove all outliers and likely return the curve back to normal. The problem there is you'd probably be removing every immigrant, so you wouldn't be representing all Americans. Pros and cons, but given medians are almost only used to describe data and not analyze it, since it's not compatable with a lot of statistics. A real analyst would probably just dummy code immigration in a regression and provide coefficients for both groups, anyway.
And this is in 2015, after we'd recovered from the great recession but before the housing/rental market forced a lot of families back into multi-generational housing situations.
It's probably also not uncommon for people who can work from home to have moved outward to lower cost-of-living areas during this time, but I bet that pales in comparison to the increase of young adults living with their parents.
I inherited my childhood home and I'm still over 1,600 miles away from my mother. But she preferred her own childhood home on the other side of the country and moved there as soon as she divorced my dad. I'm still very close with my mother. Just not... physically.
Does this factor in city dwellers? A lot of cities, you've got people that can't just up and move to the next town over like we can out here in the boonies. You move from some parts of NYC 18 miles away and still be in the same city, but still effectively be like moving to another town.
I'd be interested to see what would happen to the numbers if you split out cities at given population levels.
So, if you take new York State as an example. If the average for the state is 18 miles (I can't see the post when replying to a comment in this app, so that's just using the same number as an example), do the cities skew that?
Like, if NYC and Albany have their own average of five miles, but people outside cities average 25 (or whatever it would be, I'm not doing math for this lol), that's an interesting thing.
Right, but that's exactly my point, it likely wouldn't just be one flat color. If you scale it by population density you get a map displaying the average distance between kids and parents compared to the average distance between any two people which I would expect to be 1. non-uniform and 2. more meaningful than raw kid-parent distance. The current map is useful and accurate, but I think the more interesting contributing factors are being drowned out by raw population density. Deciding what factors to control for (ie. pop. density, wealth level, etc.) results in a different meaningful outcome and is very important to consider when making conclusions based on the map. The image's scale is probably too granular to do this analysis but if the raw data is finer-grained I would love to see a density-controlled version.
They mention a few major influences, and population density is one of them. In areas with more sprawl and land, it's more likely for people to drive longer distances. (This probably explains the Midwest and West)
They also mention poverty being a factor, where it's more common for families to live together, or very close, in order to help support each other. (so probably explains the South)
Another thing to consider is grandmothers helping when couples have young children. I bet if we overlayed a map of locations where people are more likely to have kids, we'd see a trend too.
It would be neat to have an interactive version where you can select different factors to control for, including pop. density, wealth level, children per family, etc.
If you assume the number of people who live across the country from their mother is the same number of people who live with their mother, the balance is completely in favor of the across country people. the difference between 0 and 18 is way less than the difference between 18 and 1200.
Most Americans live within an hour of where they grew up, so I'm not surprised. I think it's bonkers, I mean there's so much more out there, but from my own friend group yeah I'm the only one who moved away
I get it. My parents/hometown is a days travel away so I only visit ~two times a year. It's hard to stay in touch with all your old friends when you rarely see them. If you're just an hour away it's much easier to keep touch with your old circle.
The number of people who live and die in the city where they were born is too high. People should desire change and expansion of opportunities. I can’t imagine staying put. Complacency is accepting boredom.
Agreed, but like how do you fix it? Moving is expensive and difficult the farther away from your starting point you go. That's something that has to be prioritized and even then, it's not always attainable.
Without at least some economic privilege and luck, it's not an option for most people. Certainly there are a group of people who would like to move and cannot because of reasons they find more compelling.
The people who would move but can't, aren't all of the people that don't move far from where they are born. It could be complacency, but it could also be contentedness. The people who stay put usually have much stronger social structures than people who move around, which is not nothing.
Because lots of us live in the same city we grew up in. I never moved to a new city. Neither did my parents. Therefore we live relatively close to each other.