I know this is funny, but I really need it... Please help me

[Guest]

Hi all,

I have a table as follows.
-------------------------------------------------------------------
Country Q1 Q2 Q3 Q4 Q5 Q6
Germany 84% 80% 90% 55% 76% 81%
Benelux 58% 31% 78% 25% 68% 75%
Spain 60% 56% 62% 8% 70% 87%
France 36% 35% 58% 3% 62% 71%
Italy 80% 80% 42% 25% 92% 83%
UK 35% 38% 48% 4% 62% 70%
-------------------------------------------------------------------
Q1-Q6 refer to the responses of 6 countries to 6 survey questions,
which show the different attitudes to a new accounting policy by
percentages.
Now I have to use some statistics methodology to say that France is
least favorable to this new policy in these countries. I'm very sure
this is the final result and the numbers already told that. But the
requirement of this work forces me to use a kind of math statistics
method to say this conclusion.
So, PLEASE help me and give me some suggestions.
Thank you in advance. Any help would be appreciated.

Zh.Y

 

[JulieD]

Hi

according to my calculations of average & median UK is least favourable to
the policy - not france

used
=AVERAGE(range)
=MEDIAN(range)

Cheers
JulieD

 

[Bob Phillips]

Assuming the data in A1:F7, put this in B9 and copy across

=INDEX($A2:$A7,MATCH(MIN(B2:B7),B2:B7,0))&" is least favourable to "&C$1

or just

=INDEX($A2:$A7,MATCH(MIN(B2:B7),B2:B7,0))

 

[Myrna Larson]

I think you need the help of a bona-fide statistician to tell you what
statistical test will give valid results.

A couple of issues come to mind:

What's the margin of error for each of these percentages? For a given
question, are 35% and 36% significantly different? (I doubt it). What about
35% and 38%, 3% and 4%?, 62% and 68%

Are all questions to be given equal weight?

JulieD used average and median and based on those 2 measurements, UK is lower
than France. That doesn't fit with your "eye-ball" impression. You'll have to
figure out why not.