# We first take out whatever variance is explained by international differences in GDP per capita

< Previous | Next >

#### Jin akashini

##### Senior Member
Hi every one,

I'm reading World Happiness Report 2017, and I would like to ask you the meaning of this complicated phase :"take out whatever variance is explained by international differences in GDP per capita"

"When we use the resulting national averages to explain happiness, we first take out whatever variance is explained by international differences in GDP per capita"

Can you give me explanation on this phrase, especially "take out whatever variance is explained by international differences" ?

Thanks alot,

• #### wandle

##### Senior Member
In a wealthier country, people are liable to be happier (more goods, more opportunities, more free time, more entertainments).
The study makes allowance for that by reducing the national average of wealthier countries to make a level basis of comparison for other characteristics.

#### Jin akashini

##### Senior Member
Dear Wandle, thanks for your answer, but I don't get your point. "we first take out whatever variance is explained by international differences in GDP per capita" =The study makes allowance for that by reducing the national average of wealthier countries ( Actually I don't understand much about "national average..is national average score or level ?) and what about GDP ?

Maybe you mean that the study reduce GDP of wealthier countries to basic level, so that they can study Generosity between countries by equal manner ?

#### wandle

##### Senior Member
When we use the resulting national averages to explain happiness, we first take out whatever variance is explained by international differences in GDP per capita"
This means that the variations (in national averages) due to GDP per capita are removed from the national averages before the national averages are used for explaining happiness.

For the meaning of 'national averages' one needs to consult the article.

#### Andygc

##### Senior Member
The "national averages" are, I presume, the national average happiness scores - the average happiness score for the population of each nation. Some of the variance (difference between nations) could be attributed to high GDP countries being happier simply because of their higher GDP. The authors of the report reduce national happiness scores in proportion to the assumed effect of higher GDP.

(Sounds pretty arbitrary to me - it's perfectly possible to be poor and happy or rich and miserable.)

#### Egmont

##### Senior Member
... (Sounds pretty arbitrary to me - it's perfectly possible to be poor and happy or rich and miserable.)
Possible? Of course. However, when dealing with averages involving millions of people as we are here, it seems to be statistically less likely.

#### Andygc

##### Senior Member
I think the thread is about the meaning of the text, not the validity of the assumptions made by the authors. However, making arbitrary assumptions about the effects of a confounder, unless supported by good evidence, indicates unsound methodology and makes any conclusions questionable. Mind you, I do find the idea of producing a "World Happiness Report" both pointless and risible. How meaningful is a nation's average happiness? How does one research national happiness in closed societies such as North Korea now or East Germany 50 years ago?

#### Jin akashini

##### Senior Member
Thanks Andygc, Egmont and Wandle...I think I get your idea

< Previous | Next >