Fad Diets: What Really Works for Weight Loss
Friday, April 24th, 2009 at
4:00 am
Explore the concept of healthy eating scientifically and shed light on the molecular basis of energy balance. On this edition, UCSF nutritionist Andrea Garber takes a look at
Related posts:
- Are there any diets for pregnant women to prevent unnecessary weight gain? I am obese and I am wondering if anyone knows...
- Bodybuilding Weight Gain Diets to Bulk up Any Bodybuilder Most people think that if they are going to...
- HealthWatch: Hollywood Cleansing And Detox Diets HealthWatch: Hollywood Cleansing And Detox Diets Detox and cleansing diets...
Tagged with: dieting • diets • fad • nutrition
Filed under: Weight Gain Diets
Like this post? Subscribe to my RSS feed and get loads more!
I am shocked. That first graph showing CHD against saturated fats is completely useless. It shows 16 points randomly distributed up and down the vertical axis. Only two, the Japanese and Italian points, are significantly lower fat and CHD. Two points don’t make a trend!
lower. For example, she picked out the US point on the graph, but she might equally well have chosen the French or Swiss data, which show very low CHD but the same fat intake as the US. The graph is statistically meaningless.
I mean the second graph.
Come to that, the first graph shows no trend at all – just three isolated points at top right. How does she get the trend line from that graph? It is totally arbitrary.
Good . But I lost over 30 lb in one month using weight loss plan from LSWEIGHT(.)INFO
great job p.s.? never give up
The correlation was 0.71, a lot more reliable than your guesstimations of the correlation.
I just do not think that the figure of 0.71 is significant in this case. The sample is too small and too bunched. Just for comparison, what would be the correlation rate with the best-fit horizontal line? Can you tell us please? I ‘guesstimate’ that it would be just slightly less than 0.71.
I agree with Auriganus here. The raw correlation of a scatter of points with a line is not enough. You have to compare it with other possible lines. For example, if you have a single cluster of points, you can get the same correlation drawing a line through it in any orientation. It doesn’t tell you anything at all.
I dont think the woman in the video knows what a graph is, never mind how to analyse the data!
I love her critique of the “low-fat” diet. I cringe when I see “diet” products, period. Switching from “whole” foods to processed crap wasn’t a healthy idea, reallly??? I would have never guessed. Hey, I found the cause of obesity in the USA: Lazy, gluttonous pigs don’t want to just eat less and move more.