Total Posts:8|Showing Posts:1-8
Jump to topic:

Poor Academic Articles

ConservativePolitico
Posts: 8,210
Add as Friend
Challenge to a Debate
Send a Message
5/16/2013 8:28:57 PM
Posted: 3 years ago
Recently I have been a series of poor academic articles by my professors and they're starting to really bug me.

First of all, one article I read had dates for half of the example (in 1987... in 1999) but not the other half meaning the writer didn't want half of the examples put into a temporal context.

Secondly, THEY DON'T HAVE SOURCES! How can I trust the damn thing if I can't see where they're getting they're stats?

They take quotes out of context.

/rant

I'm reading these articles and am thinking to myself "really, educated people have to acknowledge how poorly these are written".

Irritating...
YYW
Posts: 36,252
Add as Friend
Challenge to a Debate
Send a Message
5/16/2013 9:00:00 PM
Posted: 3 years ago
At 5/16/2013 8:28:57 PM, ConservativePolitico wrote:
Recently I have been a series of poor academic articles by my professors and they're starting to really bug me.

First of all, one article I read had dates for half of the example (in 1987... in 1999) but not the other half meaning the writer didn't want half of the examples put into a temporal context.

Secondly, THEY DON'T HAVE SOURCES! How can I trust the damn thing if I can't see where they're getting they're stats?

They take quotes out of context.

/rant

I'm reading these articles and am thinking to myself "really, educated people have to acknowledge how poorly these are written".

Irritating...

Would you post the citations of the one's your talking about? You've peaked my curiosity.
Tsar of DDO
ConservativePolitico
Posts: 8,210
Add as Friend
Challenge to a Debate
Send a Message
5/16/2013 10:56:07 PM
Posted: 3 years ago
At 5/16/2013 9:00:00 PM, YYW wrote:
At 5/16/2013 8:28:57 PM, ConservativePolitico wrote:
Recently I have been a series of poor academic articles by my professors and they're starting to really bug me.

First of all, one article I read had dates for half of the example (in 1987... in 1999) but not the other half meaning the writer didn't want half of the examples put into a temporal context.

Secondly, THEY DON'T HAVE SOURCES! How can I trust the damn thing if I can't see where they're getting they're stats?

They take quotes out of context.

/rant

I'm reading these articles and am thinking to myself "really, educated people have to acknowledge how poorly these are written".

Irritating...

Would you post the citations of the one's your talking about? You've peaked my curiosity.

The one that set me off was "Capital Punishment and the Legacy of Racial Bias in America" by Bryan Stevenson.
YYW
Posts: 36,252
Add as Friend
Challenge to a Debate
Send a Message
5/16/2013 11:57:24 PM
Posted: 3 years ago
At 5/16/2013 10:56:07 PM, ConservativePolitico wrote:
At 5/16/2013 9:00:00 PM, YYW wrote:
At 5/16/2013 8:28:57 PM, ConservativePolitico wrote:
Recently I have been a series of poor academic articles by my professors and they're starting to really bug me.

First of all, one article I read had dates for half of the example (in 1987... in 1999) but not the other half meaning the writer didn't want half of the examples put into a temporal context.

Secondly, THEY DON'T HAVE SOURCES! How can I trust the damn thing if I can't see where they're getting they're stats?

They take quotes out of context.

/rant

I'm reading these articles and am thinking to myself "really, educated people have to acknowledge how poorly these are written".

Irritating...

Would you post the citations of the one's your talking about? You've peaked my curiosity.

The one that set me off was "Capital Punishment and the Legacy of Racial Bias in America" by Bryan Stevenson.

Ok. If there are others, please post them here like this

Title of Publication:
Author(s):
Type of Publication:
Source:
Pages:
Date:

That will enable me to find and review them.
Tsar of DDO
YYW
Posts: 36,252
Add as Friend
Challenge to a Debate
Send a Message
5/17/2013 12:01:30 AM
Posted: 3 years ago
Or better yet, if you have PDF links... that would be really awesome. I'd also be curious what classes they were for.

I'm a nerd like that.
Tsar of DDO
Wnope
Posts: 6,924
Add as Friend
Challenge to a Debate
Send a Message
5/17/2013 12:06:26 AM
Posted: 3 years ago
At 5/16/2013 8:28:57 PM, ConservativePolitico wrote:
Recently I have been a series of poor academic articles by my professors and they're starting to really bug me.

First of all, one article I read had dates for half of the example (in 1987... in 1999) but not the other half meaning the writer didn't want half of the examples put into a temporal context.

Secondly, THEY DON'T HAVE SOURCES! How can I trust the damn thing if I can't see where they're getting they're stats?

They take quotes out of context.

/rant

I'm reading these articles and am thinking to myself "really, educated people have to acknowledge how poorly these are written".

Irritating...

Was it a peer reviewed journal? Because a lot of magazines I know won't accept citations (don't wanna waste the space).
RoyLatham
Posts: 4,488
Add as Friend
Challenge to a Debate
Send a Message
5/19/2013 1:54:02 PM
Posted: 3 years ago
The bothersome pattern I see is journal articles that abuse statistics by applying corrections to get rid of contrary data.

An oft-cited study showed that school uniforms had no effect on school performance. The authors found reasons for throwing out Catholic schools and virtually every other case where uniforms showed a positive effect.

In a recent debate on the death penalty, a study showed the death penalty had no deterrent effect. The authors noted that Texas was the only state in which he death penalty was consistently applied in significant numbers cases. So they excluded Texas and found that other states showed no effect, then made their claim based on the states that rarely used the death penalty.

The most famous case is the global warming hockey stick. There are many kinds of measurements that serve as proxies for temperature in climate. Just combining all the proxies provides a picture of past climate that includes the Medieval Warm Period followed by the Little Ice Age. The hockey stick was derived by introducing a weighting system that got rid of anything that didn't look like a hockey stick.

Another bogus study claimed to show violent movies caused violence, but merely showed a short term effect, just being excited after watching an action movie. Long term studies, of which there are few and which are rarely cited, show no effect.

The pattern is to keep twiddling the statistical knobs until the desired result is obtained, then stop. I don't think it's deliberate fraud, it's confirmation bias.
YYW
Posts: 36,252
Add as Friend
Challenge to a Debate
Send a Message
5/19/2013 2:10:33 PM
Posted: 3 years ago
At 5/19/2013 1:54:02 PM, RoyLatham wrote:
The bothersome pattern I see is journal articles that abuse statistics by applying corrections to get rid of contrary data.

An oft-cited study showed that school uniforms had no effect on school performance. The authors found reasons for throwing out Catholic schools and virtually every other case where uniforms showed a positive effect.

In a recent debate on the death penalty, a study showed the death penalty had no deterrent effect. The authors noted that Texas was the only state in which he death penalty was consistently applied in significant numbers cases. So they excluded Texas and found that other states showed no effect, then made their claim based on the states that rarely used the death penalty.

The most famous case is the global warming hockey stick. There are many kinds of measurements that serve as proxies for temperature in climate. Just combining all the proxies provides a picture of past climate that includes the Medieval Warm Period followed by the Little Ice Age. The hockey stick was derived by introducing a weighting system that got rid of anything that didn't look like a hockey stick.

Another bogus study claimed to show violent movies caused violence, but merely showed a short term effect, just being excited after watching an action movie. Long term studies, of which there are few and which are rarely cited, show no effect.

The pattern is to keep twiddling the statistical knobs until the desired result is obtained, then stop. I don't think it's deliberate fraud, it's confirmation bias.

As a rule this is a general trend in social science; which is one of the most offensive crimes against academic integrity one can commit. One part of me want's to blame poor/sloppy/biased statistical methodology on dimwitted people -and sometimes that's true- but the real problem is the extent to which funding for projects necessarily will be cut if the answers it produces are out of line with the preconceptions of the "movers and shakers" in academia.

The worst example of this I've ever seen (which made huge waves) was from a paper published by the University of Chicago a few years ago which essentially claimed that the use of drone strikes actually causes terrorism. Anyone with a even a basic example of statistical modeling could pick up on the inconsistencies, errors, and just blatantly absurd assumptions throughout the paper -and yet it was published because it appeared to criticize a policy designed to protect US national security.

The broader point is about the legitimacy of social science, as a science -and the extent to which errors associated with a more broad confirmation bias erode that legitimacy. It's funny how really "conservative" in the purest sense of the definition of the word (meaning, averse to change) academia really is in that anything which stands in the face of their prior understanding of the world is met with exceptional resistance and is often silenced.
Tsar of DDO