Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

“The closer to the train station, the worse the kebab” – a “study”

jmspae.se

Post reply on HN
Print view View original

241–250 of 374 posts

Re: “The closer to the train station, the worse the kebab” – a “study”

#241

Post by jval43 »

mxfh wrote: post #160
macrael wrote: post #60

Earlier quoted context omitted.

OP found no correlation between railway proximity and quality

The point is, that quality is not the metric here; the metric is google ratings . I would take a place with a solid 4.6 but hundreds of ratings over a low double digit 4.9 any time.

Even Google ratings are sometimes gamed nowadays. This wasn't always the case, they used to be reliable. Tripadvisor ratings on the other hand were always garbage.

I recently had some really bad experiences with some fast food places in my corner of the world, at a train station as well.

They all had 4.9 stars but lots of 1 star reviews matching my experience. But also tons and tons of eerily similar 5 star reviews with a generic photo of the counter (no faces, no food) and a random name and glowing review of who "served" them. Which is impossible at those places.

Re: “The closer to the train station, the worse the kebab” – a “study”

#242

Post by FabHK »

myhf wrote: post #38
janwillemb wrote: post #9

He didn't find a correlation, or rather found that there is no correlation, between proximity to a railway station and how the kebab is reviewed. It's a nice study for a statistics class!

There may not be a correlation, but you can clearly see that the bottom-right quadrant of the plot is basically empty, which is an important insight. A more accurate aphorism would be "You can sell good kebabs anywhere, but you can only sell bad kebabs near a train station." And if you look at the "minimum viable quality" instead of the overall quality, there does seem to be a linear correlation with the distance. Yo…

Wonder whether that's an instance of Berkson's paradox.

Basically nobody bothering to report about Kebab shops that are neither good nor conveniently located. Probably not, but "one quadrant empty" is a handy red flag that you might encounter Berkson's.

https://en.wikipedia.org/wiki/Berkson%27s_paradox

Re: “The closer to the train station, the worse the kebab” – a “study”

#243

Post by lqet »

JmsPae wrote: post #86

Hi there, "OP" here. First off, it's been fun to see this post spread across the interwebs since I first wrote it one caffeine-fueled day just over a week ago (first Menéame, now here) and for the heck of it, I thought I would clarify a few things; This post was (sortof) a meme. Sure, I "understood the assignment" and performed the quick "study" ("analysis" might be more fitting) for the sake of the original post ove…

Why did you include metro stations in your study? The original quote only included train stations. The metro originally largely replaced a tram network, so the metro station density in Paris is higher than for other underground systems. I think it is very hard to find kebap shops in Paris that are not within 500 m of a metro station. This kind of dilutes the significance of your study, because I never really noticed that metro stations had a negative impact on their surrounding area (the opposite, really).

I think your study would have been much more interesting (but not that algorithmically challenging...) if you only included the large train stations in Paris.

Re: “The closer to the train station, the worse the kebab” – a “study”

#244

Post by ant6n »

cloudbonsai wrote: post #217
macrael wrote: post #14

LOL we may need to update the title of this post, half the top level comments right now are assuming the study confirmed the hypothesis. > With a mighty Pearson's correlation of 0.091, the data indicates that this could > be true! If you ignore the fact that the correlation is so weak that calling it 'statistically > insignificant' would be quite generous.

> be true! If you ignore the fact that the correlation is so weak that calling it 'statistically insignificant' would be quite generous. I actually came to a different conclusion than the author. Here is the way I'm thinking about the presented statistics: 1. There is 17 Kebab shops (out of 400 samples) with a google review lower than 3 stars. Let's call them "bad kebabs". 2. All those "bad kebabs" actually located w…

Bad kebab shops may survive if located near a train station.

Re: “The closer to the train station, the worse the kebab” – a “study”

#245

Post by lqet »

> I could've absolutely just routed to every single entrance for every single restaurant to get the nearest... But that would've taken several decades.

Hm, this seems like a standard many-to-many routing problem on a relatively small road network (only Paris). Why would this take several decades? Even if you just implemented a straightforward Dijkstra without any speed-up technique, it should take not more than a minute (you can trivially compute routes from N to M places using N Dijkstra runs).

Re: “The closer to the train station, the worse the kebab” – a “study”

#246

Post by jillesvangurp »

Here in Berlin there are an enormous amount of kebab places (thousands). Basically, most places are embedded in neighborhoods and have a crowd of locals that go there. The quality varies wildly as a lot of these places are obviously aimed at the money laundering business (cash only) rather than serving food. Poor hygiene, indifferent service, etc. Many of these places are completely pointless from a business point of view. Other than the money laundering thing. They are easy to recognize since they look kind of old and dirty and there are usually not a lot of customers. If it's lunch/dinner time and nobody is eating, maybe skip it.

And then there are some really outstanding places with proper char coal grills, staff that knows what they are doing and are friendly and laser focused on customer service. These places are wonderful and generally good value. Some of these marinade their own meat as well and they venture beyond just serving the standard kebabs. If you find yourself in Berlin, worth getting some advice on where to go in your area. There are some amazing kebab restaurants and most of those places have been around for ages.

And then there's a third category of places that you find in the usual late night and travel hot spots targeting travelers or people that are drunk that will probably not be regulars. Competition for this audience is fierce and there are some good choices here that people will take detours for to get to. Many of these places are well run and optimize for throughput and consistency (but not necessarily high quality).

Re: “The closer to the train station, the worse the kebab” – a “study”

#247

Post by Galanwe »

cloudbonsai wrote: post #217
macrael wrote: post #14

LOL we may need to update the title of this post, half the top level comments right now are assuming the study confirmed the hypothesis. > With a mighty Pearson's correlation of 0.091, the data indicates that this could > be true! If you ignore the fact that the correlation is so weak that calling it 'statistically > insignificant' would be quite generous.

> be true! If you ignore the fact that the correlation is so weak that calling it 'statistically insignificant' would be quite generous. I actually came to a different conclusion than the author. Here is the way I'm thinking about the presented statistics: 1. There is 17 Kebab shops (out of 400 samples) with a google review lower than 3 stars. Let's call them "bad kebabs". 2. All those "bad kebabs" actually located w…

> 2. All those "bad kebabs" actually located within 500m from the nearest station. No kebab located in further than >500m is bad.

Right, but this is selection bias. There will always exist a distance D from which all bad kebabs are located.

Unless D is provenly chosen _before_ looking at the data, this has no meaning.

One also has to take the kebab density into account.

Re: “The closer to the train station, the worse the kebab” – a “study”

#248

Post by asah »

hattmall wrote: post #176
asah wrote: post #52

Earlier quoted context omitted.

They are not mutually exclusive. Counter examples: - Katz's deli in NYC is incredibly famous, in a great location, and actually has kickass pastrami. The trade-off are relatively high prices and lines down the block - restaurants with exclusive relationships. - restaurants that make money another way, e.g. gambling. - family owned restaurants with legacy rent deals. - restaurants that cater to niche audiences e.g. sm…

I think it's probably even more close to being the opposite. Well known Restaurants in great locations tend to actually be very good. I think being good and in a good location leads to restaurants being well known. But I also think people claim Chick-fil-a and McDonald's aren't good are lying to themselves. Those restaurants routinely focus-group their food and make sure that it's ranked very high for taste and not j…

"times square olive garden" hahaha
Post reply on HN