[Bsharp]

Harvard undergrads from the 1930's are not a representative sample of any group other than that specific group. The sample is in no way random or representative of the general population in the least bit. I can't believe that anyone thinks that this study has any meaning towards the population in general.

I'm sorry if that's overly cynical, but why is it so hard for Harvard academics to understand that the factors affecting people from the Mid-Western middle class, Southern urban poor, New England suburban wealthy, etc. are all very different? Not to mention the factors affecting various cultures internationally.

This is a study of similar people with relatively similar genetics, backgrounds, upbringing, values, wealth, intelligence, and overall lifestyle. How is this of any value to people who aren't like them?

EDIT: I'm realizing that the study's conclusions might be getting exaggerated in the media. I haven't read the actual study, so I'm not sure if the authors are claiming anything as significant as the media is.

[AndrewKemendo]

In general I agree with you that this is unrepresentative for a nation as a whole - see my other comment for that.

However one advantage of a sample like this is that significant independent variables like the ones you mentioned are widely held constant, which means that the evaluated variables will have a stronger power. So it is informative relative to others.

It would be best to take this study into context with other studies which compare educational achievement and poverty level.

[Bsharp]

Agree that there are other things this study may be beneficial for. I edited my initial comment to acknowledge that this study and its conclusions are probably being hyped by the media, and it's actual value and the limitations of the study may be addressed by the authors in the study.

Unfortunately, the authors' words will be replaced by journalists to get attention.

[kristopolous]

http://www.theonion.com/articles/ah-to-be-young-rich-white-m...

[avenger123]

+1, Love it.

[smtddr]

If you liked that, you may also like: http://whatever.scalzi.com/2012/05/15/straight-white-male-th...

[001sky]

The sample is in no way random or representative of the general population in the least bit.

This is overstating it, and you are focused on culture. DNA testing could confirm a few points that would be 'representative' of the general population. If you avoid the overtechnical/literal interpretation, you are not saying so much. The implication this is otherwise 'anecdotal' data is probaly a bit too simplistics as well.

[Bsharp]

So you think "warm relationships" and IQ - two metrics used for some of the study's key conclusions - are represented similarly among this group as the general population?

I agree that it may be valuable for things such as health metrics, since it holds socioeconomic factors relatively constant, like another commenter on this thread mentions. But at the very least these people certainly had access to better health care than the average person, or at least higher-than-average intelligence to be able to handle adverse situations during their life. That alone should be enough to eliminate any meaning this may have towards people in general.

[001sky]

As other commeters have mentioned, you're ignoring what is possibly good about the data. For example, if you are trying to study a perforance envelope you may want to omit samples with known flaws. Such a sample might not be representative, but it may provide insight into potential | obvious flaws or weaknessed not obeserved. So, from a bayesian perspective it might be interesting. Also, there is the other tack of abstracting out the culture and focusing on the science (biology). Thirdly, the strategy of using this data as a single composite to be compared to similar comoposite data (eg, stratified sampling) down the road. This is all without even evaluating the data present, but it is just some reasons why its premature to "jump the gun" in the way you suggest. Better to drop the pre-conceived biases and keep an open mind until something better comes along. You have to remember that a 75 year data set is a rare thing as a historical fact, and thus dismissing it due to the limitations of the historical period (which are inseperable from the data) is throwing the baby out with the bath. {etc}

[mkhalil]

Who are you to say?

Okay, it maybe not be a one-size fit's all study, not sure if that study is even possible...but it's still really useful. How do you plan they do a 75 year study on todays generation? Time machine?

This is how stuff works. You do a study on one set, and do a study on another sample, and compare then.

And you obviously didn't read it because they didn't have similar lifestyles. The whole study was based on outcomes from different lifestyles/upbringings.

"How is this of any value to people who aren't like them?", even if this was only good to study on specific group, this question is completely useless. Why study any people of anytime? Romans? Greeks? Who cares we're not like them.

Also, stop the spreading the common notion that everyone at Harvard is the same. If you have a chance to actually visit the campus, I'd recommend it. It's fairly diverse. Don't worry...they won't bite.

[wybo]

There could still be reasons for picking such a sample.

1) In the far future, when most direct hardships are a thing of the past our lives probably resemble those of Harvard students more, than those of a random population sample in the 1930's

2) It is relatively easy to think of things that may make poor, disadvantaged, discriminated against, etc. people unhappy. And there is some data on this. While it was/is less clear for the advantaged.

3) Mentioned elsewhere, a study can gain discerning power if variables that vary a lot are kept constant. Similarly to how when doing a scientific experiment, one wants to control most variables besides the one one wants to test (simply makes the study more narrow, but deeper).

4) Practically speaking, a lot of psychological research is done on students, simply because they are within easy reach for scholars. Why Harvard students? Because that's a place where they would embark on something long-term like this 70 years ago.

5) More cynically, the scholars behind the study, and even after 70 years still the most powerful social demographic group, and probably a large portion of the most well-read, are in the sample they picked. So they sort of wanted to scratch their own itch first...

Finally, see the lovely series 7UP for a more random sample (allbeit much smaller, and British...)

http://www.imdb.com/title/tt0058578/

[cnorgate]

For what it's worth, I believe they took a sample of close to 600 people - 237 from Harvard, and another 332 who were less 'advantaged'

http://en.wikipedia.org/wiki/Grant_Study

Yes, they were all white males. Not a great study if they were trying to definitively answer the question for everyone in America. But to suggest the findings have no broader applicability would also be short sighted.

[dschiptsov]

but why is it so hard for Harvard academics to understand that the factors affecting people from the Mid-Western middle class, Southern urban poor, New England suburban wealthy, etc. are all very different?

In the East it is understood and named millenias ago. Roughly speaking, it is what they call [inner] "peace" or "balance" (of mind) which comes from realizations of one's own nature.

This is, of course, not mere abstract concepts, at least to realized people, and fundamental changes that a different state of mind propagates to the whole system (body) are easily observable and scientifically testable (chronic stresses, anxiety/personality disorders, etc.)

So, all these middle-class blah-blah people, loosely speaking, need to know thyself and [then] transform themseves.)

[Bsharp]

I'm sorry if this is ignorance, but they can scientifically test the effect of your state of mind on various psychological disorders? That seems like it's another topic of immense subjectivity. Isn't the objective measurement of one's state of mind, and even certain psychological conditions, in practice difficult/impossible to the point where studies are non-reproducible or highly debatable?

I really don't know, so if you have examples of such studies I'd be more than happy to become informed.

[dschiptsov]

There are studies of brain waves of Tibetan monks (compared to mere mortals,) but that is not the point.

The point is, that in a broad sense, Buddha's teaching (Four Noble Truths + Eight-fold Path) is an ultimate CBT, or meta-CBT if you wish, upon which any particular CBT could be easily made.

The notion that [only] a CBT provides "stable" changes in one's mental states, (de-conditioning, re-training) compared to "mere lifting of symptoms" by constant medication, for many people is not even debatable. It just works.

[ekianjo]

> for many people is not even debatable. It just works.

Not even debatable, huh?

Wikipedia: A major criticism has been that clinical studies of CBT efficacy (or any psychotherapy) are not double-blind (i.e., neither subjects nor therapists in psychotherapy studies are blind to the type of treatment). They may be single-blinded, the rater may not know the treatment the patient received, but neither the patients nor the therapists are blinded to the type of therapy given (two out of three of the persons involved in the trial, i.e., all of the persons involved in the treatment, are unblinded). The patient is an active participant in correcting negative distorted thoughts, thus quite aware of the treatment group they are in.

[...]

The element of hope and expectation on the part of the patients to get better in non-blinded trials will bias the results in favor of CBT. The informed consent procedure required to enter a psychotherapy trial biases the subjects who enter to those that are favorably inclined to the psychotherapy. Taken together, trails using psychotherapy do not meet the qualifications of high quality evidence.[96]

---

Not double-blind, then there's not much science in it, then.

[dschiptsov]

Come on, it is just words. There are millions of people who, say, quit smocking/drinking without any therapy or medication, just because they have realized they doesn't need it anymore. Is this a part of science or not?)

OK, just once.) There is a distinct process of memeization of science is going on, it is not even pop-science it is all about deriving memes from memes.

There is no secret that vast majority of studies are performed to "keep the funding going" with conclusions that would please the investors - "they cannot caught us after all". Most of these recent cancer/blood papers are not even reproducible, leave alone verifiable.

Doing non-meme science is much harder, much less profitable comparing to publishing fiction/sensationalist papers. It is a duty, not a business.

There are millions of so-called studies. "Mediterranean Diet is the most healthy one." Really? Have they factored out the water quality and air quality? How exactly? What accounts to air quality and what to the quality of water sources? In what extent? In the long run?

This is most obvious and silly example, but others are even more messy. How so-called state of stress is related to production of this or that hormones or other chemicals? How exactly it affects so-called metabolism? How they resolved sleeping issues, noise levels in the environment, air quality, diet?

The systems they are trying to study is vastly complex and correlations, even if such were discovered, not mere assumed, are not causation, and definitely not single ones.

So, to blindly follow what other people label "modern science" is, to some extent, is exactly the same as to follow a religion - just accept on faith because they said so, without asking any questions. This is meme-based, pop "science" which cannot be trusted.