Dear Alan
Many thanks for your response. I'm also sure the integration[P] in Depthmap is the pyramid relativised format but I can't find a reference to confirm it nor a formula for calculating the P-value to double check the Depthmap numbers so putting my results in front of my supervisors is problematic.
In the Social Logic of Space there is a table of P-values but I can't find any papers with a formula for calculating P-values in a stats package. I've found several formulas for calculating D-values to confirm the integration[HH] values but I'd just like to confirm integration[P] with a formula as that data appears to normalise better across my different interior environments than other integration values.
I'll keep at it and maybe see if I can get a mathematician to help out.
Thanks again
Alan
On 9 Dec 2011, at 20:52, Penn, Alan wrote:
> P-value is 'pyramid' value - it relativises against a pyramid shaped graph and is described in the Social Logic of Space - (D-value is 'diamond shaped'). I think that the depth map version is an implementation…
>
> Alan
>
> On 9 Dec 2011, at 16:29, Alan Summers wrote:
>
>> Dear Kinda
>>
>> Many thanks for that!
>>
>> I think it's safe to assume I can work out the information for the integration[Tekl] values in Depthmap from that paper and I already have the the integration[HH] corresponding with using a D-value equation.
>>
>> There is still an integration[P] in Depthmap and I rightly or wrongly assumed it was the P-value for a pyramid normalisation as discussed in Campos and Fong "A proposed methodology to normalise total depth values when applying the visibility graph analysis"
>>
>> Reproducing their tests the integration[P] worked better than the integration[HH] across closed environments of different scales though integration[Tekl] might now be my best alternative.
>>
>> I'll wrestle with the Maths some more and see where I go from here.
>>
>> Thanks for your time
>>
>> Alan
>>
|