The Abstract Is Not The Conclusion

Everyone here seems to know how to work a search engine - the necessity of subject specific knowledge is irrelevant in this day and age when you can find a study to support any viewpoint you want in minutes using google.

That's bad.

To clarify before I begin, I'm not saying don't comment on an issue if you're not an expert on it. What I am saying is don't comment on an issue unless you know what you're talking about.

You don't need formal qualifications to make valid points on an issue, but you do need to know something about the issue.

I do not care about your feelings.

If you feel strongly on a subject, that's fine. That does not give you the right to argue about it. Reality is not emotion based. The fact is that no matter how much you want something, logic will always win in practice.

If conservative, read 1, if liberal, read 2.

1. Just because you think poor people should not receive any help from the government does not mean that they won't. They need it to survive, and people dying is bad. Your emotional response, i.e. socialism is bad will not change the facts, or anyone else's opinion.

2. Just because you think poor people should receive help from the government does not mean that they will. They don't need it to survive, they can work themselves out of poverty. Your emotional response i.e. socialism is great will not change the facts, or anyone else's opinion.

The main point I'm trying to convey here is that you need to argue your case based on evidence, not on what you believe in. I respect your beliefs, but unless there is evidence to support them, I will disagree with them.


READ THIS BIT


And that leads me nicely on to my pet peeve... People who use the abstract of a study to back up their opinions.

I am a scientist. I do science.

If you believe something, and someone asks you to back your claim up, DO NOT GOOGLE YOUR CLAIM AND USE THE FIRST STUDY WITH AN ABSTRACT AS EVIDENCE WHICH AGREES WITH YOU.

Recently, I've argued with people over several studies, they have only read the abstract which is a basic account of the findings and conclusions of a study. It is a paragraph long and does not fully explain the methodology, results and findings, or caveats, of a report.

When I've read the study in full, I've either found the methodology to be flawed, the conclusion to say something completely different to the abstract, or that the researchers involved were heavily funded by the people who the report favours.

I don't want to get too technical, so I'll use a political example, since everyone knows everything about politics. Here's a politcal analogy of what people are doing in the scientific forums on ATS.

Abstract: Iraq had WMDs, so we invaded.

ATS post: OMG IRAQ HAD WMDS SO WE INVADED

Methodology: "for this study, we polled 20 people on the street and asked them if they thought iraq had WMDs"

Results: "8 yes, 7 don't know, 5 no"

Conclusion: "the majority of people think iraq has WMDs, however, not an overall majority, less than significant results, data not based on realistic or common standards, results based on single experiment by one group, not repeated"

Basically, I'm trying to get this message across to the people in the scientific forums:


THE ABSTRACT IS ALWAYS WRONG

I think everyone is a bit guilty of this.

I dont disagree with what you're saying in the least but there are those of us out there that do have knowledge of the subject matter but just cite the first source that pops up out of lazy habit to back our opinions. Sometimes it turns out that source that popped up first was terribly presented and awfully shallow information.

I mean this is ats most people dont actually read the material that opposes their opinion anyways.

I could scream "obama punched a turtle" and then post a link to whatever shady source I find and there will be those that run with it

reply to post by lobotomizemecapin


Exactly.

I also want to add something else, science is not about proving yourself right, it's about proving yourself wrong. Everything you do in a laboratory or on paper is designed to prove your own hypothesis wrong. If you find conflicting evidence, you make a new theory, if you do not, you keep trying to prove yourself wrong. Finding evidence that backs your theory does not make it right - evolution is still a theory despite all the evidence supporting it. The only thing that matters in science is evidence that proves you wrong. Evidence that proves you right does not matter at all when presented with a single piece of peer reviewed, repeatable and accurate evidence that says your theory is wrong.

Another important point is: scientists lie. I've done it myself. (edit: yes, I'm ashamed, no, I shouldn't have done it, but good luck finding a scientist who hasn't withheld information or used a friendly set of statistics in the world) You can use statistics to prove anything. The only thing that matters is the results table. I can use any one of 10 statistical methods on my results - some of them will show a significant correlation, some of them won't. You need to look at the results, not the statistical analysis.