• Welcome to the new Internet Infidels Discussion Board, formerly Talk Freethought.

Liberal Bias in Social Psychology Taints Research?

NobleSavage said:
I agree. In very broad terms I think liberals tend to think most people are good. Conservatives, tend to think we are all sinners in the eyes of god. Liberal think welfare is a good idea and if it's not working there needs to be more of it. Conservative tend to think people will game the system.
Click to expand...

lpetrich said:
It's interesting to see how the liberal side is much more willing to question itself than the conservative side. For all their complaining about people considering themselves victims, conservatives are remarkably big on considering themselves victims.
Click to expand...

Loren Pechtel said:
I'll take it a bit further.

Liberals tend to think that people are simply a product of their environment. If the environment is good they'll be good, someone who behaves badly is because of a bad childhood somehow due to exploitation by business and needs help, not punishment. Ignore the fact that business is just people, if the people are good there will not be the exploitation.

Conservatives tend to think that people are generally bad unless pushed to be good by religion. Someone who ends up in a bad position is due to their own misdeeds (which may include laziness) and is never a product of their environment. Business is always good, ignore the fact that it's pretty hard for a bad people to create an ethical business.
Click to expand...

Hmm. Either I'm not being clear at all, or you guys aren't reading what I've said... or you're all falling foul of the problem that I'm pointing out.

Not one of you is questioning your assumptions. Each of you is assuming that you know the truth, and assuming that what you "know" to be the truth is the truth, and then moving on to find "confirmation" of this "truth" in what you've read. Even when that truth isn't there. All of you would make very crappy researchers in a survey environment.

I'm not making any value judgment regarding liberalism or conservatism in the context of social science. I have no opinion of whether there ought to be more conservatives in social science - I don't care. My opinion extends no further than saying that for good science to be done, the personal viewpoint and belief of the researcher should be irrelevant. The beliefs and opinions of the researcher should not affect the outcome of the survey - they should not introduce bias or skew into the design of the research itself.

I have also said that this is difficult to do.
 
What is a liberal bias?

Fifty years ago, the idea the blacks should be allowed to vote and attend integrated schools in the American South were liberal positions.

The ideas that women should vote, hold property in our names, and divorce abusive husband were at one time or another not merely liberal but radical beliefs.

Scientific racism held sway in the early social sciences and to say that race did not determine intelligence was heresy or at best hilarity.

Study and experiment lead to changes in theory and perceptions of reality. The truth does more than set you free, it changes how we think and what was radical to say yesterday, may be conventional wisdom today and an eccentric notion tomorrow.
 
AthenaAwakened said:
What is a liberal bias?

Fifty years ago, the idea the blacks should be allowed to vote and attend integrated schools in the American South were liberal positions.

The ideas that women should vote, hold property in our names, and divorce abusive husband were at one time or another not merely liberal but radical beliefs.

Scientific racism held sway in the early social sciences and to say that race did not determine intelligence was heresy or at best hilarity.

Study and experiment lead to changes in theory and perceptions of reality. The truth does more than set you free, it changes how we think and what was radical to say yesterday, may be conventional wisdom today and an eccentric notion tomorrow.
Click to expand...


It's a good question and that be answered and addressed by the guy making the charge. But the definition of liberal and conservative don't really describe a situation of political change, but just nomenclature to define groups having certain position rather than having positions that want to change.
 
This is nothing new but it has a lot of relevant to the atheist/skeptic movement.

Here some history of why I think there is a strong left leaning bias in the social sciences. I also think there is a publication bias based on the "consensus" of beliefs in sociology circles.

***

I was digging into the the divide between Social Constructivists (SJW types) vs Evolutionary Psychology in the 1970s; basically a nature v nurture debate. This is important because it is also the division between the atheism+ people and the more science oriented skeptic/atheist/secular movement.

What I discovered is that much of this split can be traced back to 1975 with the publication of Sociobiology by E.O. Wilson. In this book Wilson made some very mild claims about genetics influencing behavior in animals but at the time the world of academia, deep in the radicalism of the 60s and civil rights movement, saw any attempt to consider genetic influences on behavior as little more than a Nazi eugenics program.

Into the fray step two atheist biologists: Steven J Gould and Richard Dawkins.

Gould sides up with the FAR political left and uses his influence to discredit anybody who dares suggest that genetics plays a role in behavior. Dawkins is on the other side with his recent publication of the Selfish Gene (1976). Consider at the time our public discourse had everything from the SLA, the Weathermen, Black Panthers to Milton Friedman and Ayn Rand all wrapped up in the fear nuclear war and Armageddon.

The social justice warriors of this time imagined they were fighting against a post-apocalyptic world of eugenics and totalitarian ideologies.

Dawkins and Gould would battle over these issues for decades until Gould died and Dawkins became an icon of the atheist movement.

***

The liberal side tends to hold that human nature is a social construct and that with just the right policies (always dictated by them), then humanity can be perfected (hence the focus on social justice policies) They tend to reject the notion of 'human-nature'. People on the right tend to accept the idea of human nature and see the left's attempt at social engineering as dangerous and wrong headed. Because of this view, they prefer to develop laws and institutions to control or punish the worse of human nature (hence the focus on law enforcement).

I think they are both nuts and that we need to have an evidence based sociology. We need "reals over feels."
 
AdamWho said:
This is nothing new but it has a lot of relevant to the atheist/skeptic movement.

Here some history of why I think there is a strong left leaning bias in the social sciences. I also think there is a publication bias based on the "consensus" of beliefs in sociology circles.

***

I was digging into the the divide between Social Constructivists (SJW types) vs Evolutionary Psychology in the 1970s; basically a nature v nurture debate. This is important because it is also the division between the atheism+ people and the more science oriented skeptic/atheist/secular movement.

What I discovered is that much of this split can be traced back to 1975 with the publication of Sociobiology by E.O. Wilson. In this book Wilson made some very mild claims about genetics influencing behavior in animals but at the time the world of academia, deep in the radicalism of the 60s and civil rights movement, saw any attempt to consider genetic influences on behavior as little more than a Nazi eugenics program.

Into the fray step two atheist biologists: Steven J Gould and Richard Dawkins.

Gould sides up with the FAR political left and uses his influence to discredit anybody who dares suggest that genetics plays a role in behavior. Dawkins is on the other side with his recent publication of the Selfish Gene (1976). Consider at the time our public discourse had everything from the SLA, the Weathermen, Black Panthers to Milton Friedman and Ayn Rand all wrapped up in the fear nuclear war and Armageddon.

The social justice warriors of this time imagined they were fighting against a post-apocalyptic world of eugenics and totalitarian ideologies.

Dawkins and Gould would battle over these issues for decades until Gould died and Dawkins became an icon of the atheist movement.

***

The liberal side tends to hold that human nature is a social construct and that with just the right policies (always dictated by them), then humanity can be perfected (hence the focus on social justice policies) They tend to reject the notion of 'human-nature'. People on the right tend to accept the idea of human nature and see the left's attempt at social engineering as dangerous and wrong headed. Because of this view, they prefer to develop laws and institutions to control or punish the worse of human nature (hence the focus on law enforcement).

I think they are both nuts and that we need to have an evidence based sociology. We need "reals over feels."
Click to expand...

I kinda get the feeling you think one side is a little nuttier than the other.

Just an observation.
 
I think that there is a lot to the idea that certain professions tend to self-select based on the nature of the work. I don't believe that many conservatives become sociologists or many liberals become economists or engineers.
 
SimpleDon said:
I think that there is a lot to the idea that certain professions tend to self-select based on the nature of the work. I don't believe that many conservatives become sociologists or many liberals become economists or engineers.
Click to expand...

In the US I can see that, but do you think that is a worldwide trend?
 
AthenaAwakened said:
What is a liberal bias?

Fifty years ago, the idea the blacks should be allowed to vote and attend integrated schools in the American South were liberal positions.

The ideas that women should vote, hold property in our names, and divorce abusive husband were at one time or another not merely liberal but radical beliefs.

Scientific racism held sway in the early social sciences and to say that race did not determine intelligence was heresy or at best hilarity.

Study and experiment lead to changes in theory and perceptions of reality. The truth does more than set you free, it changes how we think and what was radical to say yesterday, may be conventional wisdom today and an eccentric notion tomorrow.
Click to expand...

AthenaAwakened (awesome name, btw) I think that you've confused liberal positions with liberal biases in the context of research. All of the concepts that you've outlined are positions that were considered liberal at the time they were first introduced. But that's not what the OP was addressing.

The OP is addressing the concept of introducing bias into research. For the moment, let's set aside any particular political affiliation, and approach it from a less sensitive topic. Let's say... Ice cream ;). Let's say that you like chocolate ice cream, and I like strawberry, and we disagree on which one is the best. So I decide to have a poll of all of our fellow board members. So I ask them to choose from the flowing statements:
  • Strawberry is the best ice cream, because the light and refreshing fruit flavor is reminiscent of summer and childhood.
  • Chocolate is the best ice cream, because the poop color makes me feel like eating my own feces.
I think you can agree that this is pretty shoddy research, right? What I've done, in a ridiculously blatant way here, is to introduce my own personal bias into my research. I've phrased the poll question in such a way that my beliefs are reflected - and in doing so, there's an emotional content introduced into the questions that is likely to lead the poll takers to respond in a non-neutral fashion. The results of this poll are likely to be skewed because the design of the poll is biased by my belief and my preference.

That's what the OP was addressing. It's not addressing the positions being measured by the research, or the response of the populace to those positions. It's addressing the framing and the assumptions that go into the development of the surveys - The bias in the research itself.

Now, the example that I gave was of course blatant and easy. In reality, it can often be very difficult to keep bias out of survey design. Assumptions and beliefs are very difficult to reign in, especially the kind that underlie how people think about the world, because most of the time we don't even realize those are assumptions. That's just "how it is". But I do agree that in social science, it would behoove the researchers to put a bit more effort into maintaining a neutral approach to their research. Just for the sake of doing better science :D
 
It's entirely possible. If someone can give more examples of it then we can determine whether or not it's a pervasive problem.
 
Emily Lake said:
AthenaAwakened said:
What is a liberal bias?

Fifty years ago, the idea the blacks should be allowed to vote and attend integrated schools in the American South were liberal positions.

The ideas that women should vote, hold property in our names, and divorce abusive husband were at one time or another not merely liberal but radical beliefs.

Scientific racism held sway in the early social sciences and to say that race did not determine intelligence was heresy or at best hilarity.

Study and experiment lead to changes in theory and perceptions of reality. The truth does more than set you free, it changes how we think and what was radical to say yesterday, may be conventional wisdom today and an eccentric notion tomorrow.
Click to expand...

AthenaAwakened (awesome name, btw) I think that you've confused liberal positions with liberal biases in the context of research. All of the concepts that you've outlined are positions that were considered liberal at the time they were first introduced. But that's not what the OP was addressing.

The OP is addressing the concept of introducing bias into research. For the moment, let's set aside any particular political affiliation, and approach it from a less sensitive topic. Let's say... Ice cream ;). Let's say that you like chocolate ice cream, and I like strawberry, and we disagree on which one is the best. So I decide to have a poll of all of our fellow board members. So I ask them to choose from the flowing statements:
  • Strawberry is the best ice cream, because the light and refreshing fruit flavor is reminiscent of summer and childhood.
  • Chocolate is the best ice cream, because the poop color makes me feel like eating my own feces.
I think you can agree that this is pretty shoddy research, right? What I've done, in a ridiculously blatant way here, is to introduce my own personal bias into my research. I've phrased the poll question in such a way that my beliefs are reflected - and in doing so, there's an emotional content introduced into the questions that is likely to lead the poll takers to respond in a non-neutral fashion. The results of this poll are likely to be skewed because the design of the poll is biased by my belief and my preference.

That's what the OP was addressing. It's not addressing the positions being measured by the research, or the response of the populace to those positions. It's addressing the framing and the assumptions that go into the development of the surveys - The bias in the research itself.

Now, the example that I gave was of course blatant and easy. In reality, it can often be very difficult to keep bias out of survey design. Assumptions and beliefs are very difficult to reign in, especially the kind that underlie how people think about the world, because most of the time we don't even realize those are assumptions. That's just "how it is". But I do agree that in social science, it would behoove the researchers to put a bit more effort into maintaining a neutral approach to their research. Just for the sake of doing better science :D
Click to expand...

in the Brown v Board supreme court case, the studies using the black and white dolls was introduced into evidence. The attorneys for the board of education objected to evidence on the basis that the findings, and very experiment itself was biased. The study since has been proved time and time again to not be biased but sound research that disproved a conventional wisdom.

I understand what you are saying and what the OP is saying and I understood the two classes I took in sociological methodology. can studies be biased? Yes. But I don't think that's the problem here.

I have asked at least twice so far, is there a section in the paper mentioned in the op that addresses why the shift in social psychological thought and so far no one has answered that question. I did not see any explanation of the move in social and psychological theory away from structuralism, prevalent in the 1950s and 1960s when the political divide within the disciplines was fifty-fifty, toward the post modern and deconstructionist theory prevelent today.
 
The Paul said:
Togo said:
It's specifically social psychology.
Click to expand...

Oh.

That's like pondering why there aren't more creationists in the crowd when a bunch of biologists get together, then.
Click to expand...

I'm gonna give Togo some credit here. Bias is bias. Since social psychology requires analytic forms of the kind used for aerodynamics and meteorological analysis, numeric expansion and analysis, its gonna be subject to what one thinks going in. If the numbers can be supported by empirical means and the basis for the empirical analysis rests on physical principles then that is probably the best that can be expected from the field.

My critique is that a lot of the assertions (theories, schools, etc.) in Social Psychology aren't empirically supported. For instance does B = ƒ(P, E) - behavior is a function of the persona and the environment hold? What is meant by person? how is environment delimited? How does one define behavior.

I'm getting the feeling we're listening to the activity in a alien spaceship of the activity surrounding the aliens having dropped a microphone into Times Square in an alternative universe.
 
I kinda get the feeling you think one side is a little nuttier than the other.

Just an observation.
Click to expand...

A year ago I would have been on the opposite side but I have recently spent some time teaching at the high school level and the pseudoscience and politics in public education is repulsive. I have had it with these strict social constructivists who believe all human behavior is socially constructed. In academia these beliefs are self fulfilling because people who believe them are the ones going into the field and creating more search to confirm their beliefs. Unlike in science, there are no motivations to destroy orthodoxy in the humanities.

I think that pseudoscience in the humanities (specifically sociology, psychology, education, gender and ethnic studies) is an extremely important topic for skeptics to start debunking BECAUSE they often represent blind-spots on the politics left... which many atheist/skeptics identify with.

***

We should be more skeptical of ideas we like because we are sufficiently skeptical of ideas we do not like.
 
Tom Sawyer said:
It means that right wingers tend to be more stupid and not understand how research works due to their stupidity, so when research results say something negative about them they figure the problem is the research and not them ... because they're stupid.

One cannot have intelligent discussions with stupid people.
Click to expand...

sez yoo.

On the other hand, we have confrontational bias, where people love to read research which agrees with what the already believe. For many years, conservatives pointed to research which showed no connection between smoking and lung cancer. It helped that all that research was funded by the tobacco industry.

I am wondering what kind of "research question requires that one assume that a particular ideology or value system is factually true"? How does this actually work in real life?
 
Emily Lake said:
I don't have a definitive answer, but I wouldn't be even a little bit surprised to find that certain types of fields attract different predispositions. At heart, the divide between ideologies I think has more to do with how people think, how they view the world, and how they approach problem-solving than anything else. And that would naturally lend itself to differing interests in different fields.
Click to expand...

Every single field of natural and social science attracts many more liberals than conservatives. The skew varies in degree, but this survey of AAAS members who are mostly from the less liberal "hard" sciences show that only 9% of them identify as "conservative" and only 6% are registered Republicans, which makes sense since almost all of them agree with positions that are opposed by most conservatives and by the Republican party in general (e.g., Evolution, Climate Change, Stem Cell research, more gov funding for science in general, biological basis of homosexuality, trust in government, distrust in the free market to serve the public interests, etc..). You can find a few areas where scientists generally favor a "conservative" position but they are fewer and less central to political policy.

It is highly implausible that this is due to every field being intellectually biased, and far more plausible that the rational evidence-based thinking are valued by scientists in all fields moreso than by the general public, and especially by conservatives and Republicans who push claims that are blatantly irrational and anti-scientific more often than liberals do, especially on the topics that most scientist care about.

That being said, I think they still have a duty to ensure that the research they are doing is unbiased and isn't introducing skew into the questions that they are asking. They should be testing a hypothesis, not confirming an assumption ;)
Click to expand...

What is an example of them asking a biased question? The question "Are conservatives or liberals more prejudiced?" is not a biased question. Prejudice is a very real and objectively definable feature of human thinking. It is perfectly legit scientific question to ask whether people who identify with different political parties or policies display more prejudice, either against particular groups or in general against all dissimilar people to themselves. The methods used to answer the question could be biased and skew the answer, but the question itself is perfectly legit. Also, liberals and conservatives differ on many questions of fact in way that requires at least one (or both) of them to be objectively wrong. Since these errors are not random accidents but usually clearly the product of motivated bias, then it is unlikely that they make the same number of errors or equally severe and obvious errors. So, it is perfectly scientific to ask which makes more errors of fact, and thus to ask related questions, such as "Do they differ in general intelligence, in cognitive development due to schooling, or in a willingness to disregard their own knowledge and reasoning in favor of preferred conclusions?"

The research paper referred to in the OP "Bright Minds and Dark Attitudes: Lower Cognitive Ability Predicts Greater Prejudice Through Right-Wing Ideology and Low Intergroup Contact." is also asking a perfectly legit scientific question. Prejudice is an outcome of cognitive processes and it usually reflects simplistic categorical thinking, failure to take case specific information into account, failure to update assumptions with counter-factual experiences, etc.. Thus there is every reason to think it would be impacted by general cognitive abilities, and plenty of reason to think that it would relate to various defining features of right-wing ideology.

There are particular cases of ideological bias in science, and I would argue that research on racism has a severe problem in this regard. For example, you can find research defining "racism" as being in opposition to busing policies, or that defines sexism and gender bias as merely being aware of the fact that more men are in science than women (this is how the highly invalid tests of "implicit" sexism and racism tend to operationalize the terms. However, the notion that research that happens to make particular groups of people "look bad" is inherently biased and unscientific is absurd and equal to saying that research showing that one type of wild cat is more aggressive and likely to attack than another is biased and unscientific.
 
You must log in or register to reply here.
Back
Top Bottom