Viewing the message boards in:

Guide to making claims based on research

1810121314

Replies

  • Posts: 4,926 Member
    ana3067 wrote: »

    I just tried searching in my university's database, I either don't know what search term to use or there just isn't muhc available in my school's database. I found one, which I cannot find an access link to unfortunately
    NUTRITION NOTES. Source:
    RN; Jan90, Vol. 53 Issue 1, p80-101, 1/3p
    Abstract:
    Presents updates on nutrition as of January 1990. Dangers of a hypo-allergenic diet in children; Risk faced by obese patients on a very low-calorie diet.

    Otherwise, I can't find any articles that specifically talk about long-term adherence to VLCDs.
    I stopped reading there because I'm not claiming any thing at all about VLCDs except that 1200 isn't a VLCD. Do they say 1200 calories is dangerous for overweight adults? If so, I'll read them.

    I believe you're right that there isn't much available in your university's database because there is overwhelming evidence that it's safe so studying it would be a waste of time and effort. They might refer to it as an LCD or as hypocaloric, though, if you want to keep looking.

  • Posts: 14,121 Member
    lemurcat12 wrote: »

    I think part of this is a disagreement about what kind of discussion some of these discussions are.

    For example, when someone says: "sugar is evil and makes people gain weight, regardless of how many calories they eat," I read that as a factual claim and think it's fair game to challenge it. Indeed, I think it should be challenged. Some others (not in this discussion, perhaps) seem to think that that's just an expression of personal feeling and thus that it's mean to challenge it and people should be supportive (yes! I agree! sugar sucks!) or stay quiet. I personally find it frustrating that people confuse personal feelings (X worked for me and I felt good doing it) and more general claims.

    On the other hand, when someone says "why would anyone do something so stupid as IF" (exaggeration for effect) and someone else answers that "because there are some studies that it has some good effects, etc.," I don't read that as a scientific claim about the merits of the studies in that context (although I'd be interested in such as a separate discussion, perhaps). I read it as an answer to the question "why would someone do X." And, to be totally open here, to what seems to me the implied suggestion that anyone who would do X rather than the preferred approach of the majority (which happens to be my preferred approach too) must be a moron. Maybe that latter wasn't meant, but there does sometimes seem to be a burden of proof to show that something you do that is different than the majority is worthwhile even when no claim is made of the generally applicable type.

    For example, I don't do IF, but if someone asked me why I (hypothetically) did, I could see saying that I read about it in a book and there were some studies saying it had good effects and it seemed possibly an easy way to achieve a deficit given my own eating quirks, so I figured why not try it. That seems like how lots of people live their lives, and I don't think someone who says that is necessarily interested in defending the scientific merits of those studies or claiming that anyone else in the world should do IF if they don't want to.

    Under those circumstances, I think it's totally reasonable to say "you know, getting into a debate isn't something I have the interest in doing now, but for my personal decision to try something I thought the study seemed reputable enough. If you want to analyze it on your own go for it, but at this point this is working for me so I don't really care."

    That's kind of the reverse of how I feel about meal timing. Even if there are studies that show it matters, I know there are others that show it doesn't and--more important to me--it doesn't seem to matter that much for me and eating the way I do now is a lot more sustainable than mini meals would be. For me.

    That part in bold, would be 100% ok with accepting that answer and moving on. It's when they keep insisting that "no really, I've read all about this and know it's better, but...." That's when I start eyerolling. HARD.
  • Posts: 4,926 Member

    That part in bold, would be 100% ok with accepting that answer and moving on. It's when they keep insisting that "no really, I've read all about this and know it's better, but...." That's when I start eyerolling. HARD.

    I hope you're not implying I said anything like that. If that's what you read into anything I wrote, you read very wrong.

  • Posts: 14,121 Member

    I hope you're not implying I said anything like that. If that's what you read into anything I wrote, you read very wrong.

    I've never even looked at the other thread.
  • Posts: 5,623 Member
    I stopped reading there because I'm not claiming any thing at all about VLCDs except that 1200 isn't a VLCD. Do they say 1200 calories is dangerous for overweight adults? If so, I'll read them.

    I believe you're right that there isn't much available in your university's database because there is overwhelming evidence that it's safe so studying it would be a waste of time and effort. They might refer to it as an LCD or as hypocaloric, though, if you want to keep looking.

    ..... Seriously, HOW do you have a doctorate? Researchers do not only study negative phenomena. They will even study plenty of things that appear to be common sense
  • Posts: 22,281 Member
    :|
  • Posts: 12,950 Member
    ana3067 wrote: »

    ..... Seriously, HOW do you have a doctorate? Researchers do not only study negative phenomena. They will even study plenty of things that appear to be common sense

    A wise person once told me, there is a medical school for everyone. I think we can infer further from there.
  • Posts: 3,177 Member
    edited November 2014
  • Posts: 3,177 Member
    edited November 2014
    Also, correlation can imply causation, especially if there are multiple studies showing correlation. It doesn't prove causation, but it certainly may imply it.

    It actually goes the other way around--if you have causation, you usually have correlation. Statistically, correlation does not imply causation. Multiple studies showing correlation does not add up to causation.

    Think of correlation as another way of saying there's a relationship of some kind between two things. The relationship does not necessarily mean one causes the other. Causation can only be illustrated through specific types of research that allow for specific controls for the control group(s) and treatment conditions, with specific types of data analysis/processing. Not all research that people set out to do can lend itself to a result of causation. But almost all research can reveal if there is a correlation between two groups. I use this link a lot with my students. http://www.stats.org/faq_vs.htm

    This one is just freakin' funny! :smile: http://www.latimes.com/business/hiltzik/la-fi-mh-see-correlation-is-not-causation-20140512-column.html
  • Posts: 319 Member
    edited November 2014
    When I try to read this.. I go:

    tumblr_m301feBjlr1r5kyvmo3_250.gif

    then i was like.....

    tumblr_mg2pxp4drh1s2pi0fo1_250.gif
  • Posts: 159 Member
    ^^love the cat gifs :)
  • Posts: 9,532 Member
    ana3067 wrote: »
    ...snip from hell....

    TL;DR.

    Seriously, it's a chat forum, dude....
  • Posts: 9,532 Member
    ea15792 wrote: »
    Yes, many studies are flawed, but I would argue that the majority of posters on this (or any other forum) are not qualified to make that sort of determination.

    How do I know you're qualified to make that determination?

  • Posts: 9,532 Member
    edited November 2014
    Psychgrrl wrote: »
    Think of correlation as another way of saying there's a relationship of some kind between two things.

    Correlation is an attribute of the way we do math, it has no basis in physical reality. It's just a fancy way of saying "I don't understand the mechanism, so I'm going to guess these things are related".
  • Posts: 5,623 Member
    Mr_Knight wrote: »

    Correlation is an attribute of the way we do math, it has no basis in physical reality. It's just a fancy way of saying "I don't understand the mechanism, so I'm going to guess these things are related".

    Except correlations are determined by statistical means, not by guesswork. Correlations thus demonstrate that there is a relationship between the two variables, although further work can be done to see if something is mediating this relationship.
  • Posts: 5,623 Member
    Mr_Knight wrote: »

    TL;DR.

    Seriously, it's a chat forum, dude....

    Then move on to another thread if this is not one you care to read? It's a chat forum as you said, and there are no restrictions for how long posts need to be. I frequently come across posts that I don't care to read because htey are too long, yet I do not go in telling everyone "HEY SORRY, TL;DR AND I THUS HAVE NO CONTRIBUTION TO YOUR THREAD"
  • Posts: 9,532 Member
    ana3067 wrote: »

    Except correlations are determined by statistical means, not by guesswork.

    Statistics IS guesswork.

    The fact we use numbers to generate the guesses doesn't change the underlying reality.

  • Posts: 9,532 Member
    ana3067 wrote: »

    Then move on to another thread if this is not one you care to read? It's a chat forum as you said, and there are no restrictions for how long posts need to be. I frequently come across posts that I don't care to read because htey are too long, yet I do not go in telling everyone "HEY SORRY, TL;DR AND I THUS HAVE NO CONTRIBUTION TO YOUR THREAD"

    butt-hurt-lawyer-ad-on-park-bench.jpg
  • Posts: 12,950 Member
    Mr_Knight wrote: »

    Statistics IS guesswork.

    The fact we use numbers to generate the guesses doesn't change the underlying reality.
    Your stats might be guesswork.

    Mine are the result of an RNG.
  • Posts: 4,926 Member
    ana3067 wrote: »

    ..... Seriously, HOW do you have a doctorate? Researchers do not only study negative phenomena. They will even study plenty of things that appear to be common sense
    I don't. Most college instructors have only masters degrees.

    I'm not going to argue this field with you, I'm sorry. Good luck in your classes and your quest for better cited forum posts.
  • ana3067 wrote: »

    ..... Seriously, HOW do you have a doctorate? Researchers do not only study negative phenomena. They will even study plenty of things that appear to be common sense

    Wow, how did it come to this?
  • This content has been removed.
  • This content has been removed.
  • Posts: 368 Member
    I think Mr Knight is guessing what the underlying reality is...



    and
    don't. Most college instructors have only masters degrees.

    lol.
  • Posts: 4,926 Member
    I think Mr Knight is guessing what the underlying reality is...



    and

    lol.
    I don't know what country you're in but in the US over 50% of college faculty members aren't even full time. They're adjuncts who work in the field of study, have a lot of experience and a masters or two and teach on the side.

    And no, I'm not going to cite my sources, sorry. Believe it or don't, I don't care.
  • Posts: 368 Member
    double lolz.

    NESCAC graduate, and no, not Bates.

    and those teachers churn out people like you. it's quite the cycle!
  • Posts: 230 Member
    Bump
  • Posts: 13,575 Member
    Psychgrrl wrote: »

    It actually goes the other way around--if you have causation, you usually have correlation. Statistically, correlation does not imply causation. Multiple studies showing correlation does not add up to causation.

    Think of correlation as another way of saying there's a relationship of some kind between two things. The relationship does not necessarily mean one causes the other. Causation can only be illustrated through specific types of research that allow for specific controls for the control group(s) and treatment conditions, with specific types of data analysis/processing. Not all research that people set out to do can lend itself to a result of causation. But almost all research can reveal if there is a correlation between two groups. I use this link a lot with my students. http://www.stats.org/faq_vs.htm

    This one is just freakin' funny! :smile: http://www.latimes.com/business/hiltzik/la-fi-mh-see-correlation-is-not-causation-20140512-column.html

    It seems like you are saying the same thing as I, yet also disagreeing with me. Imply and prove are not synonyms. Just as correlation and causation are not.
  • Posts: 12,950 Member
    I think Mr Knight is guessing what the underlying reality is...



    and

    lol.

    Not in the sciences. A masters might get you a staff slot though. If there aren't and PhD holders that want it.

    In fields where a masters is a terminus degree, then yes, you'll have people with Masters degrees teaching. You'll also find people with a masters teaching at a level below their degree, I have a friend who is a stats teacher with a masters, but can't teach higher than undergrad level at his uni until he receives his phd.
  • Posts: 368 Member
    like a residency position!
This discussion has been closed.