A simplified overview of research methods
Posted by:
Pistachio
()
Date: March 13, 2007 05:19AM Recently there have been a number of references on the forum to published research articles supporting specific dietary and nutritional views. Sometimes it's fairly easy to understand what the author of the article is attempting to convey, while other times it may seem that only a rocket scientist could relate to what is stated. Yet still in a sense, we are all researchers, since we are learning--sometimes by personal trial and error and other times from the experience or perspective of others--about a way of nourishment that has yet to be accepted by and intergrated into mainstream society. As such, we are both the researcher and and the guinea pigs, so to speak. That said it may be interesting to take a peek at what is involved in formal research methods and how they are carried out to reach a determined conclusion.
If there is enough interest, it should be fun as some of the posters' line of work involve doing actual research and can contribute much while others enjoy investigating what has already been published. The more input from experienced as well as novice participants regarding definitions, methods of acquiring and processing information and general questions can make this a learning experience we all can benefit from. Anyone wanna play? If the answer is yes, then I'll start with some definitions along with an example. First of all, just what is research? Basically, it is a system that uses the scientific method to arrive at the answer of a question. Standard definitions and processes are used when conducting so that terms that a researcher performing an investigation let's say in Hawaii would have the same meaning to another researcher reading about it in let's say India. These include terms such as: Population: This refers to the entire group that meets the criteria that will be studied about. Sample: A portion of the population (usually a proportionally small amount) intended to represent the population being considered. For example, there is a post on the forum titled 100% Raw Foodists ... $ / Week on Your Food?. In this case the intended population are people who only eat 100% raw foods and the question inquires on how much is spent. As of this writing, 13 individuals including the originator of the post responded to the question. Obviously, not all 100% rawfoodists participated, but rather only a small number of persons. In this case those 13 would be the sample to be studied with the intention of reflecting what the rest of 100% would spend every week, had this been an actual study. In the sample, at least one subject would likely not qualify as the respondent Rawgetarian did not specify how much s/he spent, but rather how much a non-specific symbiotic rawfoodist spends, which is not what the originator of post asked. Re: A simplified overview of research methods
Posted by:
rawgosia
()
Date: March 13, 2007 06:00AM Me wanna play!
Suppose that a researcher went to a raw-vegan retreat where she/he tested 50 willing participants out of 80 there, and out of these 50, 30 had lower than accepted (by the current standards) levels of B12. Now to the questions and answers: Q1: Does this mean that 30/50=60% of raw vegans suffer from B12 deficiency? A1: No. Q2: Does this mean that 30/80=37.5% of raw vegans suffer from B12 deficiency? A2: No. Q3: Oh well, what does it mean then? A3: The result means that 30/50=60% of tested participants had lower than accepted (by the current standards) levels of B12. Q4: Does it mean that 30/50=60% of tested participants suffer from B12 deficiency? A4: No. It only means 30/50=60% of tested participants had lower than accepted (by the current standards) levels of B12. Note that these standars are based on the averages taken from the cooked population. In particular, cases of raw vegans with no clinical symptoms of B12 deficiency dispite their levels below the norm, are well known. Q5: What can we say about the whole population of raw vegans based on the result of the above mentioned experiments then? A5: I would need to introduce a few concepts from statistics here. Perhaps next time, OK? Gosia RawGosia channel RawGosia streams Edited 1 time(s). Last edit at 03/13/2007 06:01AM by rawgosia. Re: A simplified overview of research methods
Posted by:
la_veronique
()
Date: March 13, 2007 06:15AM i can think of better things to play
like frisbee or body surfing or... heck.. anything then ruining my fun tickle my bones.. raw retreat by having someone jab a needle into my arm to find out if i meet B12 "standards" doesn't sound fun to me i'd rather play tennis anyone? Re: A simplified overview of research methods
Posted by:
singinraw
()
Date: March 13, 2007 06:52AM Hi Veronique,
Sure I'll play Of course you'll win cause I ain't played tennis since high school and even then, I never won, I suck, lol!!! its ok though the winner can keep the looser for a personal juicing servant and send them out to get fresh wild greens like stinging nettles and lambsquarters and plantain and dandilions and can even hit Mrs. Wrigglesworths's garden for some parsley!!! Then we can juice it with the emerald greenstar!!! Re: A simplified overview of research methods
Posted by:
la_veronique
()
Date: March 13, 2007 12:46PM a su servicio
mi amigo Re: A simplified overview of research methods
Posted by:
Pistachio
()
Date: March 13, 2007 02:17PM Great! You two are in as the first volunteers to participate in a study. The question to be considered is: Does diet affect performance in tennis matches?.
Don't worry Singin, you'll be given accelerated tennis lessons for a few weeks at no charge to you at the Merena Lilliams tennis academy so that you will be at a comparable level to La V. During the study, one of you will be fed 100% raw organic sun-ripened vegan meals, supplented with B12. The other will receive 100% non-raw non-vegan meals without B12. Both meals however will appear to be identical because the non-raw meal will be texturized to feel vegan, flavored with a synthetic taste of luscious tropical fruit and garden fresh vegetables, and vibrantly colored with FD&Z approved petroleum based colors. Neither participant will know which meal is real nor will the person who serves the food to the participants. This brings us to 2 other concepts: Double-blind experiment: Neither the subjects in the study (in this case La V, Singin and hopefully at least one more person--to be in the control group) nor the person administering the test (in this case the person serving the meal) knows who is in the experimental group or in the control group. This is done to minimize the participant deliberately or unconsciously changing how they'd perform in the experiment by being aware of which role they are playing. Control group: Subjects in the experiment who do not receive the experimental treatment. In the above example it will be a tennis player whose diet has not been modified for this test and so s/he continues to eat as usual including the usual intake or lack of B12 supplementation. This way the researcher can note the outcome of someone who during the same time period did not make any changes nor have any interventions. Note that Singin mentioned that s/he isn't currently in shape to perform successfully in a tennis match. Such background information is relevant and should be taken into consideration, whether or not tennis lessons were given prior to the start of the experiment and also clearly noted in the study. Why? Because it could affect the outcome. Let's say Singin was a vegetarian couch potato with poor cooked food choices who switched to a raw vegan diet only at the beginning of the experiment, while the person in the control group was an athletic meat eater who played tennis since childhood. While playing Singin becomes exhausted in the first match and the prominent publication New Lark Ties report on the study. If the participant's background prior to the study were not presented, then a headline such as "Studies at the prominent Perky University in Kadiforny demonstrate that vegan raw foodists perform worse than meat-eating player in tennis match" could give a distorted impression of the outcome. Re: A simplified overview of research methods
Posted by:
starfruit
()
Date: March 13, 2007 11:01PM I want to play and not just tennis. Keep us posted on your next experiment plans. xoxo, Starfruit Re: A simplified overview of research methods
Posted by:
sunshine79
()
Date: March 14, 2007 01:14AM I wanna play! I wanna play!
have to read the rest of the thread first hold on...... Re: A simplified overview of research methods
Posted by:
starfruit
()
Date: March 14, 2007 01:44AM Hey Let's play in New York. I see your a fellow nycer and i go to school near where you go to work (look at other post) xoxo, Starfruit Re: A simplified overview of research methods
Posted by:
sunshine79
()
Date: March 14, 2007 01:58AM Ok experiment idea- tackling the problem of a condition known as vulnerable plaque in heart disease. Researchers are perplexed by this unstable form of heart disease. And since nutritional therapy is already widely accepted in heart disease, I say set up side-by-side nutritional studies involving these most unstable patients. 1 would be the low-fat diet, 1 would be atkins, 1 the south beach diet, and 1 of course the raw diet. Take 30 volunteers for each diet, house them in an inpatient research facility to ensure full adherence to each diet, and make it a 30-day study (or less I suppose depending on how long such volunteers are actually willing to commit to such a study).
My hypothesis is that the raw diet group would fare best, to the point of producing startling results which would then be extrapolated to heart disease altogether and revolutionize its treatment and help change the world and ease suffering and pain. Yes! and then the South Beach diet group would fare 2nd best because I believe heart disease is caused largely by sugar, and the SB diet emphasizes veggies which stabilize blood sugar , thus, inflammation which is what causes heart disease in the first place. Atkins & low fat would do the worst but each still better than a standard american diet because even though I think sugar is such a huge factor in heart disease it's really the lethal combination of starch + meat + fat which is the most serious contributor to heart disease. That's why people still get results on the low-fat diet as well as Atkins - not because they're ideal diets, but at least they each remove one component of this deadly triage. Ok so that's my idea and I wish I could set up this experiment myself. Let's cure heart disease! Or at least vulnerable plaque. It is such a horrible form of heart disease - currently classified incurable. Re: A simplified overview of research methods
Posted by:
sunshine79
()
Date: March 14, 2007 02:03AM Starfruit, that sounds great! I'll check the other thread... Re: A simplified overview of research methods
Posted by:
Pistachio
()
Date: March 14, 2007 02:59AM Welcome Starfruit and Sunshine. The more participants, the more creative ideas we'll come up with.
sunshine79 Wrote: ------------------------------------------------------- > Ok experiment idea- tackling the problem of a condition known as vulnerable plaque in heart disease.... I say set up side-by-side nutritional studies involving these most unstable patients. Sunshine, if you were the person conducting this research, can you think of exactly what is it you'd like to find out? In other words, if at the conclusion of the study, it were determined to be a success or if the outcome failed to prove what you were looking for, what would you have learned from it? If you are not specific in what you are looking for, then it may turn out to be just a generalized study of the condition. > My hypothesis is that the raw diet group would fare best... What is your definition of 'best' in this instance? Elimination of plaque alltogether or by at least a specific percentage? A decrease in lab values reflecting cardiovascular disease? Decreased mortality (death rate) among persons with this condition? Is your intended goal to determine: > inflammation which is what causes heart disease in the first place. or > I think sugar is such a huge factor in heart disease it's really the lethal combination of starch + meat + fat which is the most serious contributor to heart disease. >Take 30 volunteers for each diet Where would you want to get your volunteers from? A tv ad saying in effect, "if you are 18 yrs or older and have vulnerable plaque disease, please call.... Enlisting the assistance of cardiologists to inform their patients with this diagnosis of your study... A medical journal looking for subjects between let's say 30 and 60 yrs old who recently experienced a heart attack? >It is a form of heart disease - currently classified incurable. Are there any studies that addressed to what extent if any there has been success in seeing improvements? Is this condition also viewed as incurable in other parts of the world? Is it considered as incurable in other forms of healing besides allopathic medicine? Re: A simplified overview of research methods
Posted by:
sunshine79
()
Date: March 14, 2007 03:38AM Problem- vulnerable plaque.... goal- reduction of disease markers.... hypothesis- diet is key... question- which one? method- side by side comparative analysis of 4 popular diets, all of which (except raw) have already been extensively (though perhaps not comparatively- nor specifically for vulnerable plaque) studied in the treatment of heart disease.
Vulnerable plaque is an unstable form of heart disease which does not respond particularly well to any current treatment, which for heart disease is a combination of diet/lifestyle modification + drug therapy. I chose this because if I'm not mistaken I believe diet modification is already always included in any study involving heart disease. I think they no longer do drug-only heart disease studies, because the importance of diet is so well-documented for heart disease. So... the groundwork has already been laid in the medical research community to enable seamless inclusion of a new diet into heart disease research. It's open to that. Whereas say, allergies- you want to show that diet is the answer? ooh good luck with that one. Edited 1 time(s). Last edit at 03/14/2007 03:50AM by sunshine79. Re: A simplified overview of research methods
Posted by:
Pistachio
()
Date: March 14, 2007 07:18PM Do any of these questions accurately reflect what it is that you'd research?
Can a raw food diet reduce disease markers associated with vulnerable plaque? or Is dietary modification the most effective factor in reducing disease markers associated with vulnerable plaque? or Which dietary style is most effective in reducing disease markers associated with dietary plaque? In any event, which markers would you be working on? Would you compare dietary effectiveness to other forms of currently treating this disease? What about a control group for your study? Re: A simplified overview of research methods
Posted by:
Pistachio
()
Date: March 15, 2007 12:09AM Several other people have noted their weekly food spending since the original entry mentioned in the first post on this thread. This is good as it provides more numbers to work with.
To recap, in that first entry, there were 13 responses, 1 of which did not address what the originator of the post asked. Since then, 9 more responded, bringing the total to 22. Of these 9 responders, there were 3 whose answers weren't what the originator asked either and their figures weren't included in the calculations. Two of those three did not provide a $ figure and the third did provide a figure, but only noted that the amount being less than $50. There was 1 person not among the 4 just mentioned who stated that in their family of 6 their cost was a little under 150. Given that the statement wasn't just '<150', then that figure was used. To the best of my knowledge, all except Peisinoe are in the U.S. which uses the USD as its currency. If there are other currencies that I overlooked, please feel free to correct that. Peisinoe has indicated that she in Sweden, with the SEK as its currency. On the other hand, that figure could also have been in euros (EUR) and although unlikely, even in USD. Had this been a formal study, this would have been clarified for the sake of consistency and accuracy, because it would make a huge difference in computations down the line. The amount she stated she spent was 600. If that were in SEK, then as of today that would be about 86 USD, but about 65 EUR. For the sake of this illustration, I'll use the USD conversion from the SEK, which is $86. If she is following this post and care to clarify, it would be welcomed. What is the importance of these details? Let's say someone was quoting spending habits of people whose diet is all or mostly raw foods. Citing let's say Peisinoe's figure, it could be misleading if that currency was different from that of the intended audience unless they were made aware of it. In this instance, it would seem that she spends about 10 times more than let's say David Mason, even though it serves only twice as many people--her and her husband, a non-vegetarian. Arranged in order of the lowest to the highest figures provided so far these amounts are: rawgetarian 0 taylor 0 la_veronique 0 llulu 0 sachelle 20 sharlla 40 mason 60 khale 60 flex4life 65 arugula 80 peisinoe 86 smr05k 100 kwan 100 morrisson66 100 tadey 100 oregonisaac 110 pam 150 luvmyveggies 150 rawrrr 150 minuo 200 coconutcream 210 stockholmsyndrome 250 Factors that would normally be considered in a bona fide study would be how many persons this weekly amount serves in a family, etc, but for the intent and purposes of this scenario these details were not included. Again, had this been an actual study quoted elsewhere, these aspects would be important as they could shed a different light on the subject and omitting them could distort what the original author intended to convey. Those amounts from above were placed on a scatter diagram with the x or horizontal axis representing sachelle to stockholm in the order shown above and the y or vertical axis showing the amount spent: That said, here are some more definitions: mean, median and mode. Mean refers to average by adding all the scores and dividing these by the number of subjects involved. Median refers to the middle score (not the average score) of the series of figures arranged in a sequential manner. Mode refers to the value that repeats itself the most in the series being considered. So let's say the numbers we're working with are 2, 2, 5, 7, 8, 15, 60, 64 and 70. The mean would be 25.89, the median would be 8 and the mode would be 2. In regards to the weekly spending of the posters, the mean or average amount spent would be 112.83, the median 100 and the mode 100. One more definition is outlier: This refers to a figure that falls at an extreme beyond the general cluster of other numbers. There could be a valid reason or some extenuating circumstance for this, and hopefully it would be noted somewhere in the study. Had the figure of 600 that Pesinoe mentioned been used, this could have been considered an outlier. The problem is the need to exercise caution and not throw out or omit the use of figures that would otherwise meet the criteria of study, in order to skew the results in favor of a desired outcome. Edited 1 time(s). Last edit at 03/15/2007 12:13AM by Pistachio. Sorry, only registered users may post in this forum.
|
|