Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Winning a competition predicts dishonest behavior (pnas.org)
91 points by pavornyoh on Feb 10, 2016 | hide | past | favorite | 25 comments


Applied to democracy, this means that politicians are more dishonest once voted into office.

Maybe there is a reason why the ancient Greeks drew their leaders randomly from their population

https://en.wikipedia.org/wiki/Athenian_democracy


> Maybe there is a reason why the ancient Greeks drew their leaders randomly from their population

I've actually advocated doing this in the US, only half-facetiously.

As an aside, though, it's common when discussing democracy to look back at Athens as an archetypal, Platonic ideal of democracy (pun intended). However, Athenian democracy actually had a lot of characteristics that we would find completely objectionable and deplorable today.

Futhermore, it wasn't particularly long-lasting. It's hard to get an exact date range, because what people mean when they say "Athenian democracy" isn't very well-defined (there were a few different stages that the Athenian democratic systems[0] passed through, so it depends on where you draw the line). But by almost any measure, Athenian democracy didn't even last as long as the US democratic system[1], and it was stable for a lot less time than that[2].

[0] systems - there were more than one throughout Athenian history

[1] In before "we have a republic, not a democracy" - this is a contested point that really boils down to which definition of 'democracy' and 'republic' you use. For example, the definitions used by the Federalist Papers were not entirely self-consistent, and you can cite some that would support this belief, but you can also cite ones that don't. And at the end of the day, it's really just quibbling over antiquated definitions of words that don't really represent how they're used today.

[2] I've heard one classicist (a Hellenist, to boot!) conclude that the stable period of Athenian democracy was as short as 35-50 years. This is the lower bound of what most classical historians would respond if you ask them, but it's just as easy to measure it in decades as it is in centuries - that's how short it was.


In response to [1], I must insist that "democracy" has an obvious colloquial meaning, that is, the power of the people. While I reckon the "will of the people" is not exactly trivial to define (if there's even such a thing), we do have a few instances of supposedly democratic government go directly against the wishes of the majority of its people (the adoption of the EU constitution by France comes to mind).

Come to think of it, the difference between a democracy and a republic is quite obvious: the first is an ideal, while a republic is a set of institutions. Whether a particular set of institutions achieves democracy or not is the hard question.


Nitpick:

Applied to democracy, this means that politicians who are voted into office are more dishonest than all the others who aren't.

Being voted into office doesn't cause dishonesty. It's the reverse: dishonesty helps with being voted into office.


You may have made a valid observation about causes of dishonesty in politics. But that conclusion does not follow from the study. In the study dishonest behaviour is shown to be an effect of winning. Not a cause of winning.


Oh, that's new.

Reading the abstract more carefully, I can see that. But then, why does the title says "predict" instead of "causes"?


From the abstract: Studies 1 and 2 demonstrate that winning a competition increases the likelihood of winners to steal money

So while you're not wrong, it doesn't mean my above post is wrong either ;)

> Being voted into office doesn't cause dishonesty. It's the reverse: dishonesty helps with being voted into office.

This makes intuitive sense (and I even agree with it) but I fail to see this stated in the abstract (didn't read the PDF though)


You're both right. I think the study is on what it is because that is the less obvious of the two, and the more academically interesting. Cheaters winning is rather obvious. But it's also a chicken and the egg situation. Which came first? Cheating or winning? To each their own probably. The stories of our lives...


In Australia we've had a couple of 'accidental politicians' get in due to lucky preference votes where no-one expected them to be taken seriously. I've seen both good & bad from this. One consistency is they seem to approach things with a more openness and honesty. So while I like democracy, I'd also love to see some experiment where 10% of politicians were taken from the populace at random, kinda like Jury duty. I think it could be a good thing to combat political dynasty and keep things fresh.


> I'd also love to see some experiment where 10% of politicians were taken from the populace at random, kinda like Jury duty

It's called Sortition, and I very much agree.

https://en.wikipedia.org/wiki/Sortition


If nothing else, it would make people more conscious about the quality of education.


Reminds me of one of my favorite quotes from Hitchhiker's Guide: “Anyone who is capable of getting themselves made President should on no account be allowed to do the job.”

It applies to cheating and, more generally, to the fact that political candidates—especially ones that are legitimately "electable"—tend to be incredibly self-selected, in ways that aren't necessarily great for our actual leadership.


It would be interesting to do a parallel study so with the follow-up game being more winnable by extra effort instead of by cheating.

My hypotheses is that winning begets a desire to continue winning -- if the only way to win is cheating, the winners are more likely to cheat; however, if the only way to win is to try harder, perhaps winners would do that too.

That would be an interesting outcome because it would suggest a model where the likelihood of cheating by winners is primarily governed by whether the only way to win is to cheat (i.e. the dice game used in the study)


ars technica reviewed the article a couple of days ago: http://arstechnica.com/science/2016/02/winners-act-as-thick-...


Wow, this is really relevant to some experiences I've dealt with, in watching contests that seemingly could only (most easily?) be "won" through dishonest means. I'm specifically thinking about "music contests" where an initial threshold is basically a view-count. Could using dishonest means have down-the line effects on a band or music act?

Here's some sample text from the Vince Gill contest being put on by Guitar Center:

>ROUND ONE:During the Contest Period, the Contest platform will track each Submission’s popularity using a proprietary ranking algorithm ("online activity"). Online activity tracking will be measured weekly throughout the Contest Period. Each week during the Contest Period, the Contest platform will identify the ten (10) Entrants with the highest online activity during that week (the "Top Ten Weekly" and the "Weekly Ranking" respectively). Each Entrant who’s Submission (i) is in the Top Ten Weekly (in the highest Weekly Ranking) as determined by Administrator and (ii) also meets the Submission Requirements is guaranteed to have at least one (1) Submission be reviewed by the Judging Panel in the Finalist Round. In the event that an Entrant has a Top Ten Weekly ranking in the Weekly Ranking on more than one occasion, the Judging Panel shall also review the next highest ranked Entrant so that there are a total of one hundred (100) Entrants that are guaranteed to have their Backing Track video in their Submission be reviewed by the Judging Panel in the Finalist Round, additional Submission videos, may or may not be reviewed. Any attempt by a Designated Agent or an Entrant and/or his or her family/friends to use fraudulent mechanisms to unfairly affect the outcome, as determined by Sponsor in its sole discretion, shall give Sponsor the right to disqualify the Entrant. Entrants will not be notified if they are being judged in the Finalist Round.

I participated in a similar contest earlier using the same platform (Zedd if you're at all curious). There was rampant click fraud being done by numerous contestants, most of it simple click-for-click type stuff, but there were obvious attempts to get way ahead.

Considering how frequently I come across "services" that guarantee listens/likes/clicks through different platforms (Twitter, SoundCloud, DatPiff, YouTube, etc) I'm highly skeptical of how well such contests are policed.

In returning to the OP article, if dishonest means were the method to winning, and it follows that dishonesty will rise again, I wonder how that plays out in the music environment.


The thing is how long does that effect last?

Maybe the winning triggers a bit of feel-good hormone so that the participant finds himself in a situation of enjoying a little bit risk.

There are many behavioral experiments these days. And they'd like to produce "common sense" results.

These results can't be generalised and i think it'd be even amoral to use it to justify hating people. Like, seriously, ethical doings are not that clearly defined anymore. The world has changed so much.


The paper isn't explicit about it, but I'm guessing "...among Ivy League college students" is an implicit part of the title.

Drawing from a population that has spent most of their lives competing and winning may skew the results somewhat, relative to other populations...


So... I must be the most honest man in the world.


So you won the honesty competition?


I ... refuse to answer that without my lawyer present.


The study used individual competition. I wonder if they would observe the same phenomenon with team competition.


This says so much about the dangers of winner-take-all economics.


I smell bullshit when even the first sentence of the abstract is not convincing. But it's not really bullshit, it's only common sense.

> "Competition is prevalent. People often resort to unethical means to win (e.g., the recent Volkswagen scandal)."

The recent "Volkswagen scandal", where the authors probably mean the recent Bosch/Audi scandal who got caught cheating the diesel emission tests, had nothing to do with competition. When everyone is cheating these tests since centuries (starting with the US companies in the 90ies, and then the other car makers later), it is a purely political problem, but has nothing to do with competition. All the engines are using the cheating device, everybody cheats, there's no competition, the game is rigged.

So when everyone cheats, just as also in competitive cycling, swimming, athletics, or in politics, it's more a scenario how to play the game properly and just avoid getting caught.

But here we went full-cycle, and declare favorable dishonest behavior based on an environment where everyone cheats. The problem is not the winner as declared in this paper, the problem is the whole game, at least the cabal of the top 10 who mostly conspire to keep quiet. But in reality not only the players of the game, also the whole training staff, the media, the judges, the federation.

The rest of the paper wants to disprove the false image of the "good sport", which is sometimes based on unfairness. And the new finding is that winning a competition will favor dishonesty, and not only the other way round. Which explains "corruption".


1) Just because everyone is cheating doesn't mean there isn't a competition. The competition is for clean, high-mileage, high-performing vehicles and it very much is a competition.

2) I'm not sure why you decided to rename it a Bosch/Audi scandal as it is a Volkswagen/Bosch/Audi/Porsche scandal. Volkswagen is simply the combination of most highly distributed manufacturer and highest name recognition in that set.

3) Everyone cheats is not an excuse for cheating. It just means better testing and regulation enforcement is required.

4) I do agree that the paper has it backwards. Dishonesty leads to Competition "Winning" rather than the other way around.


Evidence has come to public that VW deliberately scammed the emissions tests. Do you have evidence to support your argument that "everyone is cheating these tests since centuries"?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: