Total Posts:22|Showing Posts:1-22
Jump to topic:

Confirmation bias in scientific research

tejretics
Posts: 6,089
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 9:04:14 AM
Posted: 7 months ago
Confirmation Bias in Studies

There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.

(1) LGB parenting

One example of such a study is Mark Regnerus's study on the effects of same-sex parenting. Here is Ana Samuel's summary of Regnerus's study [http://www.familystructurestudies.com...] and the study itself. [https://www.ncbi.nlm.nih.gov...] The study concluded that children who saw their parents engage in same-gender relationships were at greater risk of adverse outcomes (e.g. "being on public assistance, being unemployed, and having poorer educational attainment." [http://pediatrics.aappublications.org...]).

The worst part is, in multiple of Regnerus's media appearances, he used the study as proof that same-sex parenting should be made illegal. [http://www.slate.com...] That is truly pathetic, because (1) his study never established a causal link between the same-sex parents and harms to children (merely establishing a correlation) and (2) his study wasn't even about same-sex parenting--it was about children witnessing their parents in same-gender relationships, which were often broken apart. The same results would come on in opposite-gender relationships when you take into account that only two of Regnerus's subjects were actually raised by same-gender parents--and both of those subjects had positive outcomes. [http://thinkprogress.org...] Mark Oppenheimer explains about Regnerus's University of Notre Dame profile: "'Mark alluded to the fact that his academic interest in family formation trends and processes had arisen while still an evangelical, and his recent entrance into the Catholic Church has shaped his own thinking about fertility and family life,' the profile says. Dr. Regnerus 'also hinted at future contributions that his academic research could potentially make to the larger Catholic Church.'" [http://www.nytimes.com...]

(2) Young earth

Pro-creationist studies almost always carry confirmation bias in them. The RATE model of creationist physicist Russell Humphreys [https://en.wikipedia.org...] is the most frustrating. The RATE project study by Dr. Humphreys essentially argues that the presence of helium in zircon crystals prevents uranium-lead dating to function. A bit of background, uranium lead dating works by the following formula:

Nnow = Norig * e^[(-lambda)*t]

Here, "Nnow" is the number of uranium atoms currently present in the sample of zircon rock or zircon crystals. "Norig" refers to the number of uranium atoms originally in the sample, which is equal to the sum of lead and uranium atoms in the sample, since lead atoms displace some uranium atoms in the sample. Lambda is the decay rate of uranium. The age of zircon rock is t. If we have methods to find all variables other than t, then we can find the age of zircon rock.

Humphrey's study questioned the ability of researchers to find the decay rate of uranium because of the presence of helium in zircon rock. Here's why the study is complete nonsense. Dr. Humphreys' RATE model fails to fulfill Quality Assurance or Quality Control Procedures. "Essential QC/QA Concedures include properly collecting, identifying, labeling, storing and monitoring all samples. If the collection site of a specimen is unknown or if it has been improperly stored for several decades, any resulting data are often useless." [http://www.talkorigins.org...] Humphreys insists that most of the Precambrian sections of the Fenton Hill cores are "granodiorites." This is flawed, since they are, in fact, gneisses. The entire RATE study assumes that the Precambrian sections consisted of *only one* rock unit. Humphreys admits that if it was not only one rock unit, or if the experiment failed to identify rock types, it would be inappropriate for their modeling efforts.

When studies fail to follow the basic scientific method just so they can reach the conclusion the researchers want to reach, it is immensely frustrating. Science and the scientific method is all about questioning preexisting assumptions and displaying skepticism, and such studies are an insult to scientific credibility.
"Where justice is denied, where poverty is enforced, where ignorance prevails, and where any one class is made to feel that society is an organized conspiracy to oppress, rob and degrade them, neither persons nor property will be safe." - Frederick Douglass
user13579
Posts: 822
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 9:37:16 AM
Posted: 7 months ago
At 5/9/2016 9:04:14 AM, tejretics wrote:
(2) Young earth

Pro-creationist studies almost always carry confirmation bias in them. The RATE model of creationist physicist Russell Humphreys [https://en.wikipedia.org...] is the most frustrating. The RATE project study by Dr. Humphreys essentially argues that the presence of helium in zircon crystals prevents uranium-lead dating to function. A bit of background, uranium lead dating works by the following formula:

Nnow = Norig * e^[(-lambda)*t]

Here, "Nnow" is the number of uranium atoms currently present in the sample of zircon rock or zircon crystals. "Norig" refers to the number of uranium atoms originally in the sample, which is equal to the sum of lead and uranium atoms in the sample, since lead atoms displace some uranium atoms in the sample. Lambda is the decay rate of uranium. The age of zircon rock is t. If we have methods to find all variables other than t, then we can find the age of zircon rock.

Humphrey's study questioned the ability of researchers to find the decay rate of uranium because of the presence of helium in zircon rock. Here's why the study is complete nonsense. Dr. Humphreys' RATE model fails to fulfill Quality Assurance or Quality Control Procedures. "Essential QC/QA Concedures include properly collecting, identifying, labeling, storing and monitoring all samples. If the collection site of a specimen is unknown or if it has been improperly stored for several decades, any resulting data are often useless." [http://www.talkorigins.org...] Humphreys insists that most of the Precambrian sections of the Fenton Hill cores are "granodiorites." This is flawed, since they are, in fact, gneisses. The entire RATE study assumes that the Precambrian sections consisted of *only one* rock unit. Humphreys admits that if it was not only one rock unit, or if the experiment failed to identify rock types, it would be inappropriate for their modeling efforts.

When studies fail to follow the basic scientific method just so they can reach the conclusion the researchers want to reach, it is immensely frustrating. Science and the scientific method is all about questioning preexisting assumptions and displaying skepticism, and such studies are an insult to scientific credibility.

YEC can explain away any results of any experiment. If Earth looks billions of years old, creationists can just say that the creator created Earth to look a lot older than it is! Totally unfalsifiable and not science!
Science in a nutshell:
"Facts are neither true nor false. They simply are."
"All scientific knowledge is provisional. Even facts are provisional."
"We can be absolutely certain that we have a moon, we can be absolutely certain that water is made out of H2O, and we can be absolutely certain that the Earth is a sphere!"
"Scientific knowledge is a body of statements of varying degrees of certainty -- some most unsure, some nearly sure, none absolutely certain."
Rukado
Posts: 527
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 3:10:07 PM
Posted: 7 months ago
At 5/9/2016 9:04:14 AM, tejretics wrote:
Confirmation Bias in Studies

There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.

(1) LGB parenting

Yeah, I wouldn't expect two mentally ill people engaged in unnatural relations would have a negative influence children. Anyone who says different must be guilty of confirmation bias.
Ramshutu
Posts: 4,063
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 3:15:43 PM
Posted: 7 months ago
At 5/9/2016 3:10:07 PM, Rukado wrote:
At 5/9/2016 9:04:14 AM, tejretics wrote:
Confirmation Bias in Studies

There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.

(1) LGB parenting

Yeah, I wouldn't expect two mentally ill people engaged in unnatural relations would have a negative influence children. Anyone who says different must be guilty of confirmation bias.

He was talking about LGB parenting, I don't understand how bringing evangelical fundamentalism into the argument is relevant.
Rukado
Posts: 527
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 3:49:04 PM
Posted: 7 months ago
At 5/9/2016 3:15:43 PM, Ramshutu wrote:
He was talking about LGB parenting, I don't understand how bringing evangelical fundamentalism into the argument is relevant.

There's much you don't understand, including the fact that no one brought up evangelical fundamentalism.
slo1
Posts: 4,333
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 3:51:01 PM
Posted: 7 months ago
At 5/9/2016 9:04:14 AM, tejretics wrote:
Confirmation Bias in Studies

There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.

(1) LGB parenting

One example of such a study is Mark Regnerus's study on the effects of same-sex parenting. Here is Ana Samuel's summary of Regnerus's study [http://www.familystructurestudies.com...] and the study itself. [https://www.ncbi.nlm.nih.gov...] The study concluded that children who saw their parents engage in same-gender relationships were at greater risk of adverse outcomes (e.g. "being on public assistance, being unemployed, and having poorer educational attainment." [http://pediatrics.aappublications.org...]).

The worst part is, in multiple of Regnerus's media appearances, he used the study as proof that same-sex parenting should be made illegal. [http://www.slate.com...] That is truly pathetic, because (1) his study never established a causal link between the same-sex parents and harms to children (merely establishing a correlation) and (2) his study wasn't even about same-sex parenting--it was about children witnessing their parents in same-gender relationships, which were often broken apart. The same results would come on in opposite-gender relationships when you take into account that only two of Regnerus's subjects were actually raised by same-gender parents--and both of those subjects had positive outcomes. [http://thinkprogress.org...] Mark Oppenheimer explains about Regnerus's University of Notre Dame profile: "'Mark alluded to the fact that his academic interest in family formation trends and processes had arisen while still an evangelical, and his recent entrance into the Catholic Church has shaped his own thinking about fertility and family life,' the profile says. Dr. Regnerus 'also hinted at future contributions that his academic research could potentially make to the larger Catholic Church.'" [http://www.nytimes.com...]

(2) Young earth

Pro-creationist studies almost always carry confirmation bias in them. The RATE model of creationist physicist Russell Humphreys [https://en.wikipedia.org...] is the most frustrating. The RATE project study by Dr. Humphreys essentially argues that the presence of helium in zircon crystals prevents uranium-lead dating to function. A bit of background, uranium lead dating works by the following formula:

Nnow = Norig * e^[(-lambda)*t]

Here, "Nnow" is the number of uranium atoms currently present in the sample of zircon rock or zircon crystals. "Norig" refers to the number of uranium atoms originally in the sample, which is equal to the sum of lead and uranium atoms in the sample, since lead atoms displace some uranium atoms in the sample. Lambda is the decay rate of uranium. The age of zircon rock is t. If we have methods to find all variables other than t, then we can find the age of zircon rock.

Humphrey's study questioned the ability of researchers to find the decay rate of uranium because of the presence of helium in zircon rock. Here's why the study is complete nonsense. Dr. Humphreys' RATE model fails to fulfill Quality Assurance or Quality Control Procedures. "Essential QC/QA Concedures include properly collecting, identifying, labeling, storing and monitoring all samples. If the collection site of a specimen is unknown or if it has been improperly stored for several decades, any resulting data are often useless." [http://www.talkorigins.org...] Humphreys insists that most of the Precambrian sections of the Fenton Hill cores are "granodiorites." This is flawed, since they are, in fact, gneisses. The entire RATE study assumes that the Precambrian sections consisted of *only one* rock unit. Humphreys admits that if it was not only one rock unit, or if the experiment failed to identify rock types, it would be inappropriate for their modeling efforts.

When studies fail to follow the basic scientific method just so they can reach the conclusion the researchers want to reach, it is immensely frustrating. Science and the scientific method is all about questioning preexisting assumptions and displaying skepticism, and such studies are an insult to scientific credibility.

You are right, it is ridiculous. I wasn't able to get to his methods to even see if his study was blinded. For all we know he tampered with the data. The good news is that his study was ripped apart. See the blurb from this study that evaluated the means and methods of gay/lesbian studies.

Unfortunately a lawyer can call any expert witness they want to call. I don't think people have a good understanding why the gold standard of research is double blinded. I almost think critical thinking which includes evaluation of study design and execution should be mandatory class in high school to graduate, so crack pots can be outed by the general population. It would be interesting to see if the opposition lawyers were able to adequately discredit this man's study to the juror and judge.

http://www.ncbi.nlm.nih.gov...
For example, a recent study (Regnerus, 2012), which purportedly showed adverse effects of same-sex parents on children, has been widely criticized for using retrospective proxy reports from adult children to identify a parent as having ever been involved in a same-sex relationship (for a critique, see Perrin et al., 2013). Although the findings from this study have been largely discredited (Perrin et al., 2013), the results have been used as evidence in legal proceedings geared toward forestalling same-sex partners" efforts to adopt children or legally marry (e.g., American Sociological Association, 2013; DeBoer v. Snyder, 2014; Hollingsworth v. Perry, 2013). This use of social science research highlights the importance of adhering to best practices for research on same-sex relationships (which several U.S.-based surveys are implementing), including directly asking respondents if they have a same-sex partner and allowing for multiple response options for union status (e.g., legal marriage, registered domestic partnership, civil union, cohabitation, and living-apart-together relationships; Bates & DeMaio, 2013; Festy, 2008).
NothingSpecial99
Posts: 369
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 4:12:48 PM
Posted: 7 months ago
At 5/9/2016 9:04:14 AM, tejretics wrote:

As a YEC, I am somewhat familiar to Humphrey's work along with RATE. I used to use his work as evidence for young-earth creationism before reading all of the "refutations" of his work. I'm willing to admit that I know little to nothing when dealing with the subject at hand. I prefer more biological arguments over geological or radiometric arguments when arguing for Young Earth Creationism.
"Check your facts, not your privilege" - Christina Hoff Summers

If you go to jail for Tax Evasion, you're living off of Taxes as a result of not paying Taxes

"Facts don't care about your feelings" - Ben Shapiro
tejretics
Posts: 6,089
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 4:42:03 PM
Posted: 7 months ago
At 5/9/2016 3:10:07 PM, Rukado wrote:
Yeah, I wouldn't expect two mentally ill people engaged in unnatural relations would have a negative influence children. Anyone who says different must be guilty of confirmation bias.

That is a hilariously foolish position.
"Where justice is denied, where poverty is enforced, where ignorance prevails, and where any one class is made to feel that society is an organized conspiracy to oppress, rob and degrade them, neither persons nor property will be safe." - Frederick Douglass
keithprosser
Posts: 1,978
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 6:25:32 PM
Posted: 7 months ago
Confirmation Bias in Studies
There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.


I think you are using the term 'confirmation bias' wrongly. Confirmation bias is about how people unconsciously prefer information that matches their preconceptions. Deliberately (ie consciously, with malice aforthought) crafting a study to further an agenda isn't confirmation bias - it is a form of scientific fraud.

http://www.regnerusfallout.org...

Such papers - I would suggest - are not intended for other scientists to read but to provide a fig-leaf of respectability to a political agenda. It was published in a respected journal so for their 800,000 dollar investment the right-wing Witherspoon Foundation got what they wanted - something they could point to as a 'peer-reviwed academic paper' that seems to support their odious world-view.

As for Regnerus himself, "This year he joined the nascent Austin Institute for the Study of Family and Culture, a new academic initiative "dedicated to research on the family, marriage, and contemporary relationships."

That's right - the right-wing rewarded him for selling out with a nice cushy job. Put altogether it's a tawdry tale of corruption, scientific fraud and ideological agenda pushing. Unconscious 'confirmation bias' (which we are all prone to and should watch out for) had nothing to do with it.
Rukado
Posts: 527
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 7:03:14 PM
Posted: 7 months ago
At 5/9/2016 4:42:03 PM, tejretics wrote:
At 5/9/2016 3:10:07 PM, Rukado wrote:
Yeah, I wouldn't expect two mentally ill people engaged in unnatural relations would have a negative influence children. Anyone who says different must be guilty of confirmation bias.

That is a hilariously foolish position.

You think being mentally ill is not a hindrance to parenting?
You think homosexuality is a choice?
Just what are you being an ignorant bigot about?
Rukado
Posts: 527
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 7:06:09 PM
Posted: 7 months ago
At 5/9/2016 6:25:32 PM, keithprosser wrote:
Confirmation Bias in Studies
There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.


I think you are using the term 'confirmation bias' wrongly. Confirmation bias is about how people unconsciously prefer information that matches their preconceptions. Deliberately (ie consciously, with malice aforthought) crafting a study to further an agenda isn't confirmation bias - it is a form of scientific fraud.

I agree with you. The author of the OP is just all around ignorant.
someloser
Posts: 1,377
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 7:49:31 PM
Posted: 7 months ago
Different but similar: http://www.theatlantic.com...

Also, one on publication bias: http://andrewgelman.com...
Ego sum qui sum. Deus lo vult.

"America is ungovernable; those who served the revolution have plowed the sea." - Simon Bolivar

"A healthy nation is as unconscious of its nationality as a healthy man of his bones. But if you break a nation's nationality it will think of nothing else but getting it set again." - George Bernard Shaw
Ramshutu
Posts: 4,063
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 7:51:28 PM
Posted: 7 months ago
At 5/9/2016 3:49:04 PM, Rukado wrote:
At 5/9/2016 3:15:43 PM, Ramshutu wrote:
He was talking about LGB parenting, I don't understand how bringing evangelical fundamentalism into the argument is relevant.

There's much you don't understand, including the fact that no one brought up evangelical fundamentalism.

Actually, I was facetiously commenting on your position for the purposes of ridiculing it.
RuvDraba
Posts: 6,033
Add as Friend
Challenge to a Debate
Send a Message
5/9/2016 8:29:27 PM
Posted: 7 months ago
At 5/9/2016 9:04:14 AM, tejretics wrote:
There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.
While I haven't looked at these papers in detail, Tej, what you describe seems not so much bias as deliberate, systematic prejudice, which makes it pseudoscience rather than science.

At 5/9/2016 6:25:32 PM, keithprosser wrote:
Deliberately (ie consciously, with malice aforthought) crafting a study to further an agenda isn't confirmation bias - it is a form of scientific fraud.
I agree, Keith. Increasingly, government and management relies on science to make informed, accurate, accountable policy decisions. So lobby groups, who focus on producing communications and peddling influence, have themselves sought to influence government with what appear to be scientific credentials and scientific papers, but which lack scientific values, rigour and accountability.

At 5/9/2016 3:51:01 PM, slo1 wrote:
Unfortunately a lawyer can call any expert witness they want to call.
While that's true, Slo, the evidentiary sense in the judiciary is sometimes very impressive. If you'd like cause for hope, take a look at how meticulous are both the expert testimonies [http://ncse.com...] and the judicial findings [http://ncse.com...] in the "Dover Panda Trial" (Kitzmiller vs. Dover Area School District", 2005 [https://en.wikipedia.org...].) It's really very impressive how much scientific insight is presented to legal generalists to help them make clear, well-founded decisions.

As long as judges are scientifically literate -- or can quickly become so -- then the ploy of faking independent inquiries to manipulate judicial findings will largely fail.

Unfortunately, the question of manipulating legislature is a different matter. Politicians aren't known for great science literacy or great courage. They are as often led by public sentiment as leading by clear evidence. Getting scientific literacy and accountability into legislature is a more vexed and fraught issue I think. It needs to be done, but I suspect it must be through greater science literacy in the general population.

At 5/9/2016 7:03:14 PM, Rukado wrote:
At 5/9/2016 4:42:03 PM, tejretics wrote:
At 5/9/2016 3:10:07 PM, Rukado wrote:
Yeah, I wouldn't expect two mentally ill people engaged in unnatural relations would have a negative influence children. Anyone who says different must be guilty of confirmation bias.
That is a hilariously foolish position.
You think being mentally ill is not a hindrance to parenting?
You think homosexuality is a choice?
Just what are you being an ignorant bigot about?
This is an example of a member who doesn't understand the standard of evidence, transparency and accountability needed for scientific debate.

Rukado, if you'd like an explanation of why your arguments are falling far short of anything that could be respected under scientific methodology, please feel free to ask in another thread, and I'll be happy to tell you. However at the moment, nothing you've argued is worthy of serious consideration or substantive response as a contribution to scientific discussion.
tejretics
Posts: 6,089
Add as Friend
Challenge to a Debate
Send a Message
5/10/2016 2:30:16 AM
Posted: 7 months ago
At 5/9/2016 7:03:14 PM, Rukado wrote:

Prove that homosexuality is a mental illness, and one that has adverse effects on parenting.
"Where justice is denied, where poverty is enforced, where ignorance prevails, and where any one class is made to feel that society is an organized conspiracy to oppress, rob and degrade them, neither persons nor property will be safe." - Frederick Douglass
tejretics
Posts: 6,089
Add as Friend
Challenge to a Debate
Send a Message
5/10/2016 2:33:08 AM
Posted: 7 months ago
At 5/9/2016 8:29:27 PM, RuvDraba wrote:
At 5/9/2016 9:04:14 AM, tejretics wrote:
There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.

While I haven't looked at these papers in detail, Tej, what you describe seems not so much bias as deliberate, systematic prejudice, which makes it pseudoscience rather than science.

Regnerus's paper (a) assumes a conclusion, and (b) finds actual correlations (read: not causation) to back the conclusion. That's "bias," not "pseudoscience." The latter paper is pseudoscience, sure, but it's also an instance of confirmation bias in that it assumes the conclusion and tries to find evidence to back it up.

"Confirmation bias, also called confirmatory bias or myside bias, is the tendency to search for, interpret, favor, and recall information in a way that confirms one's preexisting beliefs or hypotheses, while giving disproportionately less consideration to alternative possibilities."

https://en.wikipedia.org...

Both examples are clear instances of interpreting information in a way that confirms one's preexisting beliefs. The latter is pseudoscience, agreed, but the former isn't "pseudoscience" so much as an incorrect conclusion.
"Where justice is denied, where poverty is enforced, where ignorance prevails, and where any one class is made to feel that society is an organized conspiracy to oppress, rob and degrade them, neither persons nor property will be safe." - Frederick Douglass
tejretics
Posts: 6,089
Add as Friend
Challenge to a Debate
Send a Message
5/10/2016 2:36:12 AM
Posted: 7 months ago
At 5/9/2016 6:25:32 PM, keithprosser wrote:

I get what you're saying--the second example doesn't work, then. But the first one does, because Regnerus clearly believed the results of his study while subconsciously favoring incorrect results. But your source might be correct....
"Where justice is denied, where poverty is enforced, where ignorance prevails, and where any one class is made to feel that society is an organized conspiracy to oppress, rob and degrade them, neither persons nor property will be safe." - Frederick Douglass
Axonly
Posts: 1,802
Add as Friend
Challenge to a Debate
Send a Message
5/10/2016 9:08:58 AM
Posted: 7 months ago
At 5/9/2016 9:37:16 AM, user13579 wrote:
At 5/9/2016 9:04:14 AM, tejretics wrote:
(2) Young earth

Pro-creationist studies almost always carry confirmation bias in them. The RATE model of creationist physicist Russell Humphreys [https://en.wikipedia.org...] is the most frustrating. The RATE project study by Dr. Humphreys essentially argues that the presence of helium in zircon crystals prevents uranium-lead dating to function. A bit of background, uranium lead dating works by the following formula:

Nnow = Norig * e^[(-lambda)*t]

Here, "Nnow" is the number of uranium atoms currently present in the sample of zircon rock or zircon crystals. "Norig" refers to the number of uranium atoms originally in the sample, which is equal to the sum of lead and uranium atoms in the sample, since lead atoms displace some uranium atoms in the sample. Lambda is the decay rate of uranium. The age of zircon rock is t. If we have methods to find all variables other than t, then we can find the age of zircon rock.

Humphrey's study questioned the ability of researchers to find the decay rate of uranium because of the presence of helium in zircon rock. Here's why the study is complete nonsense. Dr. Humphreys' RATE model fails to fulfill Quality Assurance or Quality Control Procedures. "Essential QC/QA Concedures include properly collecting, identifying, labeling, storing and monitoring all samples. If the collection site of a specimen is unknown or if it has been improperly stored for several decades, any resulting data are often useless." [http://www.talkorigins.org...] Humphreys insists that most of the Precambrian sections of the Fenton Hill cores are "granodiorites." This is flawed, since they are, in fact, gneisses. The entire RATE study assumes that the Precambrian sections consisted of *only one* rock unit. Humphreys admits that if it was not only one rock unit, or if the experiment failed to identify rock types, it would be inappropriate for their modeling efforts.

When studies fail to follow the basic scientific method just so they can reach the conclusion the researchers want to reach, it is immensely frustrating. Science and the scientific method is all about questioning preexisting assumptions and displaying skepticism, and such studies are an insult to scientific credibility.

YEC can explain away any results of any experiment. If Earth looks billions of years old, creationists can just say that the creator created Earth to look a lot older than it is! Totally unfalsifiable and not science!

Wait. You aren't upset about the use of "Confirmation bias" xD
Meh!
Rukado
Posts: 527
Add as Friend
Challenge to a Debate
Send a Message
5/10/2016 4:29:40 PM
Posted: 6 months ago
At 5/10/2016 2:30:16 AM, tejretics wrote:
At 5/9/2016 7:03:14 PM, Rukado wrote:

Prove that homosexuality is a mental illness, and one that has adverse effects on parenting.

If you think you're a duck, would you be mentally ill? Why?
RuvDraba
Posts: 6,033
Add as Friend
Challenge to a Debate
Send a Message
5/11/2016 1:53:47 AM
Posted: 6 months ago
At 5/10/2016 2:33:08 AM, tejretics wrote:
At 5/9/2016 8:29:27 PM, RuvDraba wrote:
At 5/9/2016 9:04:14 AM, tejretics wrote:
There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.
While I haven't looked at these papers in detail, Tej, what you describe seems not so much bias as deliberate, systematic prejudice, which makes it pseudoscience rather than science.
Regnerus's paper (a) assumes a conclusion, and (b) finds actual correlations (read: not causation) to back the conclusion. That's "bias," not "pseudoscience."
Having looked more deeply, Tej, I believe it's academic misconduct in the form of willful neglect.

Regnerus didn't just start with his expected conclusion, he seems to have deliberately chosen misrepresentative data to support it. As I understand it, his study took two sample groups:

1) People 18 years or older who had an 'intact biological family' for the preceding years; and
2) People 18 years or older who had a parent believed to have been involved in a same-sex relationship prior to their eighteenth birthday; and

These are two incomparable groups. There's a case to compare stable heterosexual parenting with stable homosexual parenting, or unstable heterosexual parenting with unstable homosexual parenting, or stability rates themselves, but not stable heterosexual parenting with parenting whether stable or not, which includes suspected homosexual encounters. I cannot conceive a legitimate scientific rationale for comparing incomparable groups, and didn't read one in the paper. So the only conclusions I can draw is either gross incompetence or deliberate misconduct through willful neglect of normal sociological practices.

A comparison of homosexual and heterosexual parenting is a routine topic for sociological study, Regnerus is sociologically trained, housed at a reputable institution, and has a lengthy publications list on his CV [http://www.markregnerus.com...], so incompetence is not credible.

Regnerus' paper was denounced by his own faculty, and has been broadly criticised by peak sociologist societies, but it should never have passed peer review, and I see no reason that an academic with poor ethics should still have a job.
Fly
Posts: 2,044
Add as Friend
Challenge to a Debate
Send a Message
5/11/2016 5:37:42 PM
Posted: 6 months ago
At 5/9/2016 9:04:14 AM, tejretics wrote:
Confirmation Bias in Studies

There are some studies that are deliberately crafted to gain certain conclusions--studies that seriously annoy me due to the sheer level of bias the scientists have.

(1) LGB parenting

One example of such a study is Mark Regnerus's study on the effects of same-sex parenting. Here is Ana Samuel's summary of Regnerus's study [http://www.familystructurestudies.com...] and the study itself. [https://www.ncbi.nlm.nih.gov...] The study concluded that children who saw their parents engage in same-gender relationships were at greater risk of adverse outcomes (e.g. "being on public assistance, being unemployed, and having poorer educational attainment." [http://pediatrics.aappublications.org...]).

The worst part is, in multiple of Regnerus's media appearances, he used the study as proof that same-sex parenting should be made illegal. [http://www.slate.com...] That is truly pathetic, because (1) his study never established a causal link between the same-sex parents and harms to children (merely establishing a correlation) and (2) his study wasn't even about same-sex parenting--it was about children witnessing their parents in same-gender relationships, which were often broken apart. The same results would come on in opposite-gender relationships when you take into account that only two of Regnerus's subjects were actually raised by same-gender parents--and both of those subjects had positive outcomes. [http://thinkprogress.org...] Mark Oppenheimer explains about Regnerus's University of Notre Dame profile: "'Mark alluded to the fact that his academic interest in family formation trends and processes had arisen while still an evangelical, and his recent entrance into the Catholic Church has shaped his own thinking about fertility and family life,' the profile says. Dr. Regnerus 'also hinted at future contributions that his academic research could potentially make to the larger Catholic Church.'" [http://www.nytimes.com...]

(2) Young earth

Pro-creationist studies almost always carry confirmation bias in them. The RATE model of creationist physicist Russell Humphreys [https://en.wikipedia.org...] is the most frustrating. The RATE project study by Dr. Humphreys essentially argues that the presence of helium in zircon crystals prevents uranium-lead dating to function. A bit of background, uranium lead dating works by the following formula:

Nnow = Norig * e^[(-lambda)*t]

Here, "Nnow" is the number of uranium atoms currently present in the sample of zircon rock or zircon crystals. "Norig" refers to the number of uranium atoms originally in the sample, which is equal to the sum of lead and uranium atoms in the sample, since lead atoms displace some uranium atoms in the sample. Lambda is the decay rate of uranium. The age of zircon rock is t. If we have methods to find all variables other than t, then we can find the age of zircon rock.

Humphrey's study questioned the ability of researchers to find the decay rate of uranium because of the presence of helium in zircon rock. Here's why the study is complete nonsense. Dr. Humphreys' RATE model fails to fulfill Quality Assurance or Quality Control Procedures. "Essential QC/QA Concedures include properly collecting, identifying, labeling, storing and monitoring all samples. If the collection site of a specimen is unknown or if it has been improperly stored for several decades, any resulting data are often useless." [http://www.talkorigins.org...] Humphreys insists that most of the Precambrian sections of the Fenton Hill cores are "granodiorites." This is flawed, since they are, in fact, gneisses. The entire RATE study assumes that the Precambrian sections consisted of *only one* rock unit. Humphreys admits that if it was not only one rock unit, or if the experiment failed to identify rock types, it would be inappropriate for their modeling efforts.

When studies fail to follow the basic scientific method just so they can reach the conclusion the researchers want to reach, it is immensely frustrating. Science and the scientific method is all about questioning preexisting assumptions and displaying skepticism, and such studies are an insult to scientific credibility.

In cases such as this, where the "scientist" is willfully working toward supporting ONLY his preexisting conclusion which he espouses on the basis of personal morals or other bias, I don't even bother crediting that person with the title of "scientist." That person is hardly practicing science. Rather, I call that person a "researcher." And even that is being charitable on my part...
"You don't have a right to be a jerk."
--Religion Forum's hypocrite extraordinaire serving up lulz