"Diet research built on a 'house of cards'?" (problem: relying on self-reports of what people eat)

[Dolphin]

(Just noticed this is from February)

Some of us are interested in the problems with relying on self-report questionnaires so thought this might be of interest to somebody.
It could also be said to be on the problem of relying on self-reports of compliance.

ANALYSIS
Diet research built on a 'house of cards'?
Nutrition studies depend on people telling the truth. But they don't.
By Kelly Crowe, CBC News
Posted: Feb 24, 2015 1:07 PM ET

http://www.cbc.ca/news/health/diet-research-built-on-a-house-of-cards-1.2968704

There's also a tendency for research subjects to tilt their dietary recall towards healthier behaviour.

 

[jimells]

"And as you might expect, some individuals who have based much of their research career and program on the use of these methods say we can't abandon them because in part we don't have anything better."

If these "scientists" insist on using flawed methods (and claiming their conclusions are valid) because they can't think of a better method, then maybe they are actually in the food marketing business instead of scientific research.

Every time I use my food stamp card the government knows exactly what I bought, when, and where. If any of us Useless Eaters buy a candy bar or use the card out of state then those purchases are noted by the Governor, who never misses a chance to castigate the Lazy Welfare Bums.

My point is that collecting food consumption data is easy: give the study subjects a debit card for their groceries and restaurant meals. It wouldn't be a hundred percent of course, but it would certainly be more accurate than asking someone to remember what they ate a month ago.

 

[user9876]

If these "scientists" insist on using flawed methods (and claiming their conclusions are valid) because they can't think of a better method, then maybe they are actually in the food marketing business instead of scientific research.

Every time I use my food stamp card the government knows exactly what I bought, when, and where. If any of us Useless Eaters buy a candy bar or use the card out of state then those purchases are noted by the Governor, who never misses a chance to castigate the Lazy Welfare Bums.

My point is that collecting food consumption data is easy: give the study subjects a debit card for their groceries and restaurant meals. It wouldn't be a hundred percent of course, but it would certainly be more accurate than asking someone to remember what they ate a month ago.

There was a dieting tv programme in the UK where they talked a family who were over weight about their diet. The then installed cameras and followed them for a few days and compared what people said (and I think genuinely thought) they were eating with what they actually ate. I only saw a couple of episodes but the families seemed quite shocked about what they were actually eating.

So I'm not sure that the real issue is people not telling the truth but peoples perception of routine stuff is just not very accurate.

When looking at interventions such as GET there is perhaps an important message in terms of people perceiving they are following rules when they aren't or have just substituted one activity for a different one. With no careful monitoring no safe conclusions can be drawn.

 

[Valentijn]

There was a dieting tv programme in the UK where they talked a family who were over weight about their diet. The then installed cameras and followed them for a few days and compared what people said (and I think genuinely thought) they were eating with what they actually ate. I only saw a couple of episodes but the families seemed quite shocked about what they were actually eating.

Secret Eaters? I agree - they are genuinely completely clueless about their basic calories. And there's issues with not understanding the value of calorie reduction versus exercise, and how the binge generated by a 30 minute workout results in gaining weight, not losing it.

 

[Snow Leopard]

My point is that collecting food consumption data is easy: give the study subjects a debit card for their groceries and restaurant meals. It wouldn't be a hundred percent of course, but it would certainly be more accurate than asking someone to remember what they ate a month ago.

I don't think it is that simple. Who is eating this food? Their family? Their friends?

 

[jimells]

There was a dieting tv programme in the UK where they talked a family who were over weight about their diet. The then installed cameras and followed them for a few days and compared what people said (and I think genuinely thought) they were eating with what they actually ate. I only saw a couple of episodes but the families seemed quite shocked about what they were actually eating.

So I'm not sure that the real issue is people not telling the truth but peoples perception of routine stuff is just not very accurate.

I might have seen the same program - I remember a table piled with all the food eaten by the contestants during the course of the program. It was quite an impressive shot!

I didn't mean to suggest that study participants are dishonest but rather that recollections of memory is very poor evidence. I suspect the main reason for using questionnaires is they are quick & dirty and relatively cheap.