Let’s consider a practical example. Since the question of exercise and weight has turned up, let’s revisit it. First, let’s collect some raw data (I can’t use internet usage, since this poll is extremely biased on that axis).
For the purposes of this poll, “overweight” means a body mass index over 25. “Exercise” means at least 30 minutes a week, working hard at it, on a regular basis. “Diet” means that you actually think about the nutritional value of the food you eat, and consciously base your choice of food on that information in some significant way.
Select one of the following:
[pollid:183]
Once we have some data, we can then practice this skill on the results of the poll, and see whether (and if so, how) these variables are causally linked among poll respondants.
Then one of the two “I diet, I am overweight” options seems appropriate, depending on whether you exercise or not. Whether you have lost or gained weight recently doesn’t seem part of the poll.
A question not being very “well formulated” implies to me that it incorporates confusions, ambiguities, false dilemmas, etc.
That a different question might be more relevant to the purpose of the post, seems a different issue.
I was very careful to formulate the question to avoid confusions, particularly in the definition of ‘overweight’ (my thinking was, Obelix would claim he was not overweight, by defining ‘overweight’ I at least ensure that different definitions of ‘overweight’ do not blur the line). In the process, I did not consider the case of a person who had relatively recently started a diet (or an exercise regimen) and whose weight had changed as a result, but not sufficiently to move past the arbitrary 25 BMI line.
This was therefore probably not the best way to phrase the question, and for that I apologise (if I were to go back in time and rewrite the question, I would take that case into account). Nonetheless, the question stands as is; I think that it is more important at this point to be consistent, and thus one of the “I diet, I am overweight” options are appropriate.
Okay, sixteen people are not enough to say much from. There will be large error bars in the following statements, due to small sample size. Nonetheless.
Taking E for exercise, D for diet, O for overweight:
p(E)=0.625
p(D)=0.1875
p(O)=0.1875
p(ED)=0.1875
p(EO)=0.125
p(DO)=0.0625
Exercise and dieting seem to be pretty well correlated; either dieting causes exercise (with 100% certainty over this small data set) or exercise causes diet (about one-third of the time), or, more likely, a third factor (a desire to lose weight, perhaps) causes both dieting and exercise. Strangely, being overweight doesn’t seem to be correlated with either exercise or diet… my first instinct here is to be suspicious of the survey’s small sample size. (At the very least, I’d expect being overweight to cause dieting).
It also seems, from this survey, that the best way to not be overweight is to exercise but not diet—though a mere one vote can very easily change that conclusion, so this survey should be considered to have very little weight at sixteen responses.
Let’s consider a practical example. Since the question of exercise and weight has turned up, let’s revisit it. First, let’s collect some raw data (I can’t use internet usage, since this poll is extremely biased on that axis).
For the purposes of this poll, “overweight” means a body mass index over 25. “Exercise” means at least 30 minutes a week, working hard at it, on a regular basis. “Diet” means that you actually think about the nutritional value of the food you eat, and consciously base your choice of food on that information in some significant way.
Select one of the following:
[pollid:183]
Once we have some data, we can then practice this skill on the results of the poll, and see whether (and if so, how) these variables are causally linked among poll respondants.
This question is not very well formulated. I diet, and have lost 30 pounds or so since last december but am still overweight
Then one of the two “I diet, I am overweight” options seems appropriate, depending on whether you exercise or not. Whether you have lost or gained weight recently doesn’t seem part of the poll.
I’m saying that losing 30 pounds appears to be exactly the sort of thing we’re actually trying to find out about but the poll doesn’t check for it.
A question not being very “well formulated” implies to me that it incorporates confusions, ambiguities, false dilemmas, etc. That a different question might be more relevant to the purpose of the post, seems a different issue.
I was very careful to formulate the question to avoid confusions, particularly in the definition of ‘overweight’ (my thinking was, Obelix would claim he was not overweight, by defining ‘overweight’ I at least ensure that different definitions of ‘overweight’ do not blur the line). In the process, I did not consider the case of a person who had relatively recently started a diet (or an exercise regimen) and whose weight had changed as a result, but not sufficiently to move past the arbitrary 25 BMI line.
This was therefore probably not the best way to phrase the question, and for that I apologise (if I were to go back in time and rewrite the question, I would take that case into account). Nonetheless, the question stands as is; I think that it is more important at this point to be consistent, and thus one of the “I diet, I am overweight” options are appropriate.
Okay, sixteen people are not enough to say much from. There will be large error bars in the following statements, due to small sample size. Nonetheless.
Taking E for exercise, D for diet, O for overweight:
p(E)=0.625
p(D)=0.1875
p(O)=0.1875
p(ED)=0.1875
p(EO)=0.125
p(DO)=0.0625
Exercise and dieting seem to be pretty well correlated; either dieting causes exercise (with 100% certainty over this small data set) or exercise causes diet (about one-third of the time), or, more likely, a third factor (a desire to lose weight, perhaps) causes both dieting and exercise. Strangely, being overweight doesn’t seem to be correlated with either exercise or diet… my first instinct here is to be suspicious of the survey’s small sample size. (At the very least, I’d expect being overweight to cause dieting).
It also seems, from this survey, that the best way to not be overweight is to exercise but not diet—though a mere one vote can very easily change that conclusion, so this survey should be considered to have very little weight at sixteen responses.