TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

How Should We Critique Research? (2019)

51 点作者 lazyjeff12 个月前

5 条评论

notjoemama12 个月前
Its too bad research papers can&#x27;t be organized like a git history. We&#x27;d see many forks that never end up as pull requests that are merged back to main. And probably forks of forks that stray too far from the founding paper&#x27;s intent. It would be nice to more easily identify original versus derivative research. Maybe that solves a different problem. I like their suggestion though:<p>&quot;I offer a prag­matic cri­te­rion: what makes a crit­i­cism im­por­tant is how much it could change a re­sult if cor­rected and how much that would then change our de­ci­sions or ac­tions: to what ex­tent it is a “dif­fer­ence which makes a dif­fer­ence”. This is why is­sues of re­search fraud, causal in­fer­ence, or bi­ases yield­ing over­es­ti­mates are uni­ver­sally im­por­tant: be­cause a ‘causal’ ef­fect turn­ing out to be zero ef­fect or grossly over­es­ti­mated will change al­most all de­ci­sions based on such re­search; while on the other hand, other is­sues like mea­sure­ment error or dis­tri­b­u­tional as­sump­tions, which are equally com­mon, are often not im­por­tant: be­cause they typ­i­cally yield much smaller changes in con­clu­sions, and hence de­ci­sions.&quot;<p>So, 2 papers, both with data and claims.<p>The first is critiqued on its claim because the data, while correct with quality methodology, doesn&#x27;t support the extent made in the claim. This critique is more meaningful because it changes the outcome of the paper and any decisions following its publication.<p>The second&#x27;s claim is within the bounds of the data but there is a discrepancy in the data collection which is the source its its critique. Fixing that doesn&#x27;t change the claim but may indicate more research is needed. This critique <i>could</i> change decisions made from publishing, but if the claim is still within the reason of the data, then likely not.<p>I had to think through that and I think I like it.
评论 #40566280 未加载
评论 #40438989 未加载
评论 #40437691 未加载
评论 #40439411 未加载
dang12 个月前
Related:<p><i>How Should We Critique Research?</i> - <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=26834499">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=26834499</a> - April 2021 (51 comments)<p><i>How should we critique research?</i> - <a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=19981774">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=19981774</a> - May 2019 (20 comments)
verisimi12 个月前
It seems to me that the most important factor is not being mentioned here. That is money - who funds the research.<p>Its a simple enough problem - eg if you wanted papers to show anything - eg &#x27;that koalas cause forest fires in Australia&#x27; (I know that&#x27;s ridiculous!) - then you simply fund a bunch of papers. If you have 10 papers, and 2 are supportive, 2 are against and the remaining are ambiguous - you have a start! You take the supportive ones, and fund similar studies. Soon you have a lot of data that seems to say something in support of the thesis you like, but this is nothing to do with uncovering some underlying principle.<p>If you have a big enough pocket, you get the science you pay for.<p><a href="https:&#x2F;&#x2F;www.threads.net&#x2F;@tmurrayhimself&#x2F;post&#x2F;C56uXtKM0y0" rel="nofollow">https:&#x2F;&#x2F;www.threads.net&#x2F;@tmurrayhimself&#x2F;post&#x2F;C56uXtKM0y0</a><p>&quot;studies show that all studies can be traced back to the guy with the most money&quot;
评论 #40440062 未加载
Joel_Mckay12 个月前
With data quality standards, research ethics, and random audits.<p><a href="https:&#x2F;&#x2F;mchankins.wordpress.com&#x2F;2013&#x2F;04&#x2F;21&#x2F;still-not-significant-2&#x2F;" rel="nofollow">https:&#x2F;&#x2F;mchankins.wordpress.com&#x2F;2013&#x2F;04&#x2F;21&#x2F;still-not-signifi...</a><p>The primary issue is circular references in citation, and non-cascaded retractions for known errata.<p>Typically thesis work tends to be reproducible most of the time, but around 12% to 17% of the hundreds of papers I read every month were nongeneralizable or worse outright BS.<p>It can be really disillusioning for many students...<p>Now I eat cheese goldfish crackers, and no longer care either way. Have a wonderful day =3
jsemrau12 个月前
I am reviewing papers for my substack regularly and there is a structure I believe works well that focuses on potential impact, structure, reading comprehension, github availability (where applicable), and problem relevance.