Skip to content
The methodOpen data
Back to the collection

How Community Feedback Shapes User Behavior

Justin Cheng, Cristian Danescu-Niculescu-Mizil, Jure Leskovec

2014 · arXiv

Psychology
Read the paper
HN score
0
Mentions
3
Readers
3
Months
3
What’s in a score?

From the abstract

Social media systems rely on user feedback and rating mechanisms for personalization, ranking, and content filtering. However, when users evaluate content contributed by fellow users (e.g., by liking a post or voting on a comment), these evaluations create complex social feedback effects. This paper investigates how ratings on a piece of content affect its author's future behavior. By studying four large comment-based news communities, we find that negative feedback leads to significant behavioral changes that are detrimental to the community. Not only do authors of negatively-evaluated content contribute more, but also their future posts are of lower quality, and are perceived by the community as such. Moreover, these authors are more likely to subsequently evaluate their fellow users negatively, percolating these effects through the community. In contrast, positive feedback does not carry similar effects, and neither encourages rewarded authors to write more, nor improves the quality of their posts. Interestingly, the authors that receive no feedback are most likely to leave a community. Furthermore, a structural analysis of the voter network reveals that evaluations polarize the community the most when positive and negative votes are equally split.

Metadata: arXiv. Abstracts retain their original authors’ rights.

In the conversation

Original HN comments, with labels from simple text rules. A label describes the whole comment and may miss context or irony. Read the discussion to make up your own mind. About the labels

3 selected excerpts · 3 total mentions

intended
Neutral mention · rule
People attribute to malice what is best attributed to entropy The old reddit, allowed actual pedophiles to describe their issues in ask reddit threads. Theres even a dark ama where a relation which would make people ill, was discussed. ---- Unfortunately what social media is exposing, is weaknesses in how human beings agglomerate. If someone makes a forum for dead baby jokes - distasteful to many, but not harmful in private - it eventually attracts people who think that dead baby jokes are the norm. Or in worse terms - a sub for making offcolor jokes soon gets overrun by people who think off color jokes are not a risque deviation from normal behavior, but is actually…
Read the full comment
normal behavior. ---- On top of this, the internet and social communities are extremely low on contextual information - which is critical for most people to understand a conversation in real life. This means that the only way someone can know what you mean when you are being vague, is if you make an constant effort for tonal and idea accuracy in your comments. Obviously this is challenging, leading to a cascade of misunderstandings, which only serve to polarize groups more. ---- Theres a lot of research being done on how people behave online, and its just a grim picture. How Community Feedback Shapes User Behavior (https://arxiv.org/abs/1405.1429) The spreading of misinformation online( http://www.pnas.org/content/113/3/554.full) The effect of the ban on hate subs on reddit (http://comp.social.gatech.edu/papers/cscw18-chand-hate.pdf)
Original comment Reddit and the Struggle to Detoxify the Internet
indubitable
Neutral mention · rule
You are making many assumptions that need support. They are directly contradicted by relevant research. This study on the community effect of downvotes [1] came to rather different conclusions. They conclude that downvoting has a paradoxical effect. It actually incentivizes the poster to post even more content, of even lower quality. And just as importantly, upvotes similarly have no positive effects. They do not result in users posting more regularly, nor generating more well received content. Upvoted users actually tend to leave the community more quickly than downvoted users. The one and only response that negatively incentivizes users is no response at all - which…
Read the full comment
causes them to leave much earlier than up or down votes. And it gets even better. Downvotes tend to spread. Downvoted users tend to downvoter others who in turn begin downvoting others as well. All of this along with the consequential effects downvoting have on quality. I'm not sure the content voting system has been a success. The nominal idea is to enable the masses to sort content by quality. But on most sites when the systems are not being gamed, it just results in people upvoting things that stir their emotions (sensationalism, controversy, politics, cat videos). And when it is being gamed, what little meaningfulness that remained is completely removed. Even on a site like this which is going to have a more well educated and technical userbase than average who clearly make some effort to upvote more on quality than the above metrics, compare the front page to the first page of new. The frontpage should be on the orders of magnitude more interesting and higher quality, since you're comparing a heavily filtered selection to a random selection. Is it? [1] - https://arxiv.org/abs/1405.1429
Original comment Facebook’s 2016 Election Team Gave Advertisers a Blueprint to a Divided US
maxerickson
Neutral mention · rule
From the arXiv blog linked to by slashdot: Cheng and co began by compiling a dataset of the comments associated with news articles on four online communities: CNN.com, a general news site; Breitbart.com, a political new site; IGN.com, a computer gaming news site; and Allkpop.com, a Korean entertainment site. The data includes 1.2 million threads with 42 million comments and 114 million votes from 1.8 million different users. Are any of those sites famous for their effective community management? The arXiv submission: http://arxiv.org/abs/1405.1429
Original comment Data Mining Shows How Down-Voting Leads To Vicious Circle of Negative Feedback

Keep following the thread.

More in Psychology

More from the same topics.

2018arXiv

Talent vs Luck: the role of randomness in success and failure

A. Pluchino. A. E. Biondo, A. Rapisarda

The largely dominant meritocratic paradigm of highly competitive Western cultures is rooted on the belief that success is due mainly, if not exclusively, to personal qualities such as talent, intelligence, skills, efforts or risk taking.

Psychology
16 mentions14 months
2.5HN score