I only came into this thread to say 74% of statistics are made up on the spot ;)

But, it's a good discussion, I enjoyed reading the thread.

Originally Posted by ronni62:
I agree totally with both of you! The most important thing I gained from my college statistics class is that all statistics can be skewed-not only by the built-in margin of error, but by the way the researcher asks the questions, whether it's worded positively or negatively, and many other methods that the 'researchers' can actually use to obtain the results they want to see for the point they are trying to make! In some cases, the margin of error is so large that the stat becomes meaningless (think about political polls:p).

Hey this is exactly what I got out of Stats I and II as well. There are so many variables involved in coming up w/these numbers and you never know what those variables are. Statistics are malleable bits of information that can be presented for whatever result you intend.

Statistics? Bah, who needs 'em! You never fail unless you fail to try.

As a researcher, I hate to see statistics get a bad rap. On the whole, I think the posters above are correct, but in the defense of statistics, they really are very useful tools -- for example, in understanding how likely events are to occur. I really do think they are beneficial.

However, they are misunderstood and misused. For example, if you did research with a certain sample or using certain questions, you can't apply the statistics beyond the sample or questions, and yet people (including researchers) do that all the time. In other words, it's not always the statistics that are the problem, but how people use them.

Originally Posted by :
Statistics as a learning tool. A tool for change.

Great point! :)

I think it's statistics like that that are responsible for much of the success here. It's those statistics that tell us that the common attitudes and methods don't work and we need something better. It's statistics like those that made many of us sit down and think and figure out how it was going to be different this time.

Statistics are a tool. To someone who understands them, they're a very useful tool. To someone who doesn't (or someone who uses them intentionally to mislead) they become meaningless (or outright lies).

I think it's the interpretation that gets people into trouble.

Consider:

Everyone who eats, dies, therefore eating causes death and you should not eat.

Now that is an easy one, because it's easy for anyone to see the fault in that logic. But the same type of logic is often used in situations that make it a lot more difficult to see the fault in the logic. If a statement "makes sense" it is often assumed to be true without considering that the logic could be faulty.

Statistics when used correctly, can give us alot of information we otherwise wouldn't have. Research is important in many situations (I wouldn't take a drug that hadn't been tested), and the only way to evaluate research test results are with statistics.

The statistics that suggest that support groups tend to increase the success of behavior change (not just weight loss, but smoking, drug addiction, compulsive behaviors...) certainly influenced my decision to join TOPS. Does that mean that I think I couldn't lose weight without a support group? NO - but it did reinforce my decision that it was worth a shot and I could find it helpful (and I have).

I think the reason weight loss research is so sketchy, is that we're still treating weight loss as if it were a single behavior instead of as the dozens it probably truly is.

I know countless people who say 'Oh I'm on a diet' but dont take it at all seriously. I mean who doesnt know those people who always say theyre on diets and are on different diets every week?

If you looked at statistics of people who SERIOUSLY want to lose weight in the long run, like on this website for example then the statistics are going to be much more acurate.

I would encourage everyone here to take part in the national weight control registry. The requirements are: 1.lose at least 30 pounds 2.keep it off for at least 1 year or more.
Even if you are not to your goal weight, and are still trying to lose more, you can still join if you meet those 2 requirements.
Lets help those stats!

I still have 6 more pounds to lose and then keep at least that amount off for a year before I can join. I will sign my husband up as soon as his year is up. He reached 30 pounds lost back in May.

Originally Posted by SkinnyDogMom:
I would encourage everyone here to take part in the national weight control registry. The requirements are: 1.lose at least 30 pounds 2.keep it off for at least 1 year or more.
Even if you are not to your goal weight, and are still trying to lose more, you can still join if you meet those 2 requirements.
Lets help those stats!

I still have 6 more pounds to lose and then keep at least that amount off for a year before I can join. I will sign my husband up as soon as his year is up. He reached 30 pounds lost back in May.

But couldn't you just fake your numbers and your maintenance to skew the results? Sorry, just being a bug today. ;) :)

:frypan:

squashing a bug!!!

Of course anyone can lie. I don't see why someone would take the time to go to the website, fill out the information just to skew results...but then again, I am an optimist and believe in the good in others.

Originally Posted by Hat Trick:
You never fail unless you fail to try.

That's a great way to look at it. Hurray for the positive approach that is so universal here at 3FC.

Originally Posted by JulieJ08:
I think it's statistics like that that are responsible for much of the success here. It's those statistics that tell us that the common attitudes and methods don't work and we need something better. It's statistics like those that made many of us sit down and think and figure out how it was going to be different this time.

Yes - we know it's not going to be easy and we'll have to work at it and keep keep keep trying.

Originally Posted by Heather:
For example, if you did research with a certain sample or using certain questions, you can't apply the statistics beyond the sample or questions, and yet people (including researchers) do that all the time.

Just wanted to comment - and you can't use the response values in ways they weren't collected to support. I've seen too many reports that "averaged" non-mathmatical choices like A,B,C,D,E for example. On the whole though, I think statistics are beneficial.

Ok, I'm new and have no idea where the statistics started from. Can someone please point me in the right direction?

Just taking the people in my life that I know are dieters...me, my sister, my mom, my best friend, her friend, and my dad...the statistics are that 2/7 have taken weight off and kept it off.

My best friend's friend took off over 100 pounds and kept it off for something like 7 years. My sister lost something like 25 pounds and has kept it off but has also gotten way too thin at times.

My mom has been one of those that will jump on something, do it for a couple days, then claim that she can't lose weight because she's older and in menopause. She has had so many different pieces of exercise equipment that generally just gather dust. I have a treadmill they got, then gave me because it wasn't uses...they have a new one. She gave me "the bean" because she didn't use it. My sister has something they gave her, my best friend has something they gave her (and again...got a new one). They have an eliptical machine that has been sitting in their garage for well over a year...more like 3 I think. She just got a new piece and they have an exercise bike that doesn't get any use except when my kids get over there. Is she the sort that is going to get the weight off and keep it off? No, not till she finally decides that she will work at it and do what she says. She wants the magic pill.

My best friend tried and was doing really well for a while, but gave up because she didn't want to take the time to make the meals that the diet recommended. She and her mom are both going to be starting it in another week or two...supposedly. We'll see. She's been overweight her whole life, but it truly is in her genes...heck, she and her twin were over 8 pounds each at birth which is NOT normal for twins! She just is comfortable with herself really, is part of a family that is almost all overweight except for those that don't take after the Inuit part of the heritage (ie., the blonde kids). I would love for her to be able to get to where she is more thin, and I hope she can do it if her mom is doing the same diet with her. It remains to be seen though. I have more faith in her doing it than I do in my mom.

My dad doesn't really care. He'll try to eat healthy, and generally does...steams everything and eats low fat soup low sodium salt. Still...he doesn't care. He is diabetic, and doesn't even really care about that. Tests rarely and doesn't always eat regularly. He's a coma waiting to happen.

Me, I got fat not by overeating but because of medication. I've tried before to lose the weight, but I panic because I have anorexic tendencies and don't want to let them flare up. For instance, I started making sure I didn't eat after 7:00PM. Then, I bumped it up to 5:00PM. Then I bumped it up to noon, and then 8:00AM. For several months back in my 20's I was eating UP TO 200 calories a day and 0-1 grams of fat. Even doing weight watchers I started seeing how many points I could have leftover. If I have to count anything, I'll start getting obsessive about it. I KNOW I have these tendencies and have to fight them all the time. We'll see how I do. I make myself stop if I find I'm getting to obsessive.

Originally Posted by WebRover:
Just wanted to comment - and you can't use the response values in ways they weren't collected to support. I've seen too many reports that "averaged" non-mathmatical choices like A,B,C,D,E for example. On the whole though, I think statistics are beneficial.

I'm a psychologist, and we actually do that all the time (take responses on an ABCDE scale and use as continuous data). Drives some statisticians crazy, but others say it doesn't do that much harm (I'm overly simplifying).

Needless to say, there are huge academic slug-fests about these issues -- which I guess we should avoid here! :)

I've been checking out the archives from the blog Half of Me (a fabulous read, BTW) and found this link to a UCLA study

But I still think we've got something special here at 3FC. :cb:

Originally Posted by SkinnyDogMom:
:frypan:

squashing a bug!!!

LOL :)

I mentioned the 'faking the numbers' bit in reference to the discussion about stats. Just saying that, for the most part, I believe stats are largely open to debate because they are too easily manipulated. Whether it's in the way the questions are posed to how the data is collected to how it's presented, etc.

I was thinking that if someone wanted to use the stats from the national weight control registry they could easily mislead people into thinking one thing or another. But their results w/be so skewed as to be dismissable because of the way the data was collected (people posting on a website). But that wouldn't stop someone from releasing a 'stat' saying that 'xx% of people who registered w/the national weight loss registry blah, blah, blah' and that stat w/come off as being credible when in fact, it wasn't.

Just a round-about way of saying don't put too much into those 'stat' numbers, that's all. :)

Please, don't get out the Raid! :eek: :D