Thank you Klaus! Your calculator and explanations have helped me with applying these principles to workplace diversity.
Question. Where can I find what the relative difference is between q=1, q=2, q=3, q=4 in terms of evenness, abundance, and richness? Also what is the definition of each q=1, q=2, q=3, q=4. Thank you!
Hi Jeff,
thanks for your donation! Regarding your questions: to my knowledge there is no simple relation regarding the different orders of hill numbers and evenness, abundance or richness. Just that for the extreme cases the Hill number of order q=0 represents the richness (number of species), and q = inf represents the 1/max(p) (Berger-Parker index). Special case q = 1 Shannon and q = 2 Simpson.
As far as I recall, the order q is defined as integer and has to do with the generalized mean: q=0 harmonic, q=1 geometric, q=2 average, q =3 rms (root mean square). So the Hill numbers are the general concept to describe diversity, and many known diversity indices are derived as a special case from there. Sorry, I don’t recall much more, it has been a while…
Regards,
Klaus
Hi Klaus: thanks so much for making your Excel template for diversity available! I was curious about why it reports/calculates the Gini-Simpson index as a percentage rather than an index value between 0 and 1. Just a different method of conceptualizing the results? Thanks much.
Hi Klaus,
Great program! I was wondering, the CR’s of my individual respondents are not so good (with a mean of 0.3). However, the group results shows a consistency below 0.1. How is that possible? Can you say something about the calculation of the group result?
Kind regards,
Linda
Hi Linda,
the group result is calculated using the geometric mean as aggregation of individual judgments (AIJ). Please see my write-up about AHP-OS software implementation section 5.
AIJ lowers CR for the consolidated matrix depending on the number of participants. Please see: Aull-Hyde, R, Erdogan, Duke, J. M. (2006). An experiment on the consistency of aggregated comparison matrices in AHP. European Journal of Operational Research 171 (2006) 290–295.
Regards, Klaus
Thank you for putting all of the effort into your AHP software and especially for making it freely available. I just wanted to let you know that I will be teaching an energy management class in KL for Universiti Teknologi Petronas in the next few weeks and intend to discuss the AHP method for project decision making using your software.
I am very grateful to you for making it available to the public. It’s a really wonderful tool however there are few things that I would like to know. I am working to assess sustainability of MHP projects. As sustainability is very broad term and can be defined in many ways in different context, I need to clearly present my theoretical framework. Project description allows 400 char but only three lines are visible to participants. Similarly in one of my criteria(Environmental dimension) there is only one sub-criteria, as comparison is not needed this tool doesn’t allow me to include the sub criteria while defining hierarchy however the final hierarchy what participants see is not the correct one as there is one more sub-criteria under it. So please could you see if something could be done or suggest me some idea to deal with it.
Thanks for your feedback.
A) 400 char equals 5 lines à 80 chars or 3+ lines à 130 chars. The full length (400 chars) of project description is visible to participants. I just tested it again. If you need more explanation for your participants, include it in an email with the session link to start their inputs.
B) Pairwise comparisons require at least 2 criteria. If you don’t have 2 sub-criteria below a main criterion, the weight of the main criterion is evaluated and corresponds to the weight of your missing (single) sub-criterion. Revise your hierarchy or consider renaming your category/sub-criterion to make it clear to your participants.
Wonderful tool! one quick question, what is the difference between
AHP group consensus and rel. homogeneity. Which measure is best to use to report how similar the answer of group participants were?
hi
its very good system. But one problem, i was filling the data, took much time, session out error. I lost all the efforts and thoughts put on.
can u do something about it?
It’s a 4×4 matrix, i.e. 4 criteria, let them call C1, C2, C3, C4.
1st line compares C1 with C2 (C1 is 7-times more important than C2 => A7, C1 is 3-times more important than C3 => A3, C1 is 3-times more important than C4 => A3)
2nd line compares C2 with C3 and C4 (C2 is 2-times more important then C3 => A2, C2 is 3-times less important than C4 => B3)
3rd line compares C3 with C4 (C3 is 2-times less important than C4 => B2)
Hi
Do this software aggregates the responses in AHP process for a combined decision?
Yes, it does.
Thank you Klaus! Your calculator and explanations have helped me with applying these principles to workplace diversity.
Question. Where can I find what the relative difference is between q=1, q=2, q=3, q=4 in terms of evenness, abundance, and richness? Also what is the definition of each q=1, q=2, q=3, q=4. Thank you!
Hi Jeff,
thanks for your donation! Regarding your questions: to my knowledge there is no simple relation regarding the different orders of hill numbers and evenness, abundance or richness. Just that for the extreme cases the Hill number of order q=0 represents the richness (number of species), and q = inf represents the 1/max(p) (Berger-Parker index). Special case q = 1 Shannon and q = 2 Simpson.
As far as I recall, the order q is defined as integer and has to do with the generalized mean: q=0 harmonic, q=1 geometric, q=2 average, q =3 rms (root mean square). So the Hill numbers are the general concept to describe diversity, and many known diversity indices are derived as a special case from there. Sorry, I don’t recall much more, it has been a while…
Regards,
Klaus
Hi Klaus: thanks so much for making your Excel template for diversity available! I was curious about why it reports/calculates the Gini-Simpson index as a percentage rather than an index value between 0 and 1. Just a different method of conceptualizing the results? Thanks much.
Hi John,
just a question of number presentation, values are identical. There is no deeper meaning behind.
Klaus
Hi Klaus,
Great program! I was wondering, the CR’s of my individual respondents are not so good (with a mean of 0.3). However, the group results shows a consistency below 0.1. How is that possible? Can you say something about the calculation of the group result?
Kind regards,
Linda
Hi Linda,
the group result is calculated using the geometric mean as aggregation of individual judgments (AIJ). Please see my write-up about AHP-OS software implementation section 5.
AIJ lowers CR for the consolidated matrix depending on the number of participants. Please see: Aull-Hyde, R, Erdogan, Duke, J. M. (2006). An experiment on the consistency of aggregated comparison matrices in AHP. European Journal of Operational Research 171 (2006) 290–295.
Regards, Klaus
Dr. Goepel,
Thank you for putting all of the effort into your AHP software and especially for making it freely available. I just wanted to let you know that I will be teaching an energy management class in KL for Universiti Teknologi Petronas in the next few weeks and intend to discuss the AHP method for project decision making using your software.
Regards,
Rich Mignogna
Rich, thanks for the comment. Let me know, in case you have any feedback on the program.
Klaus
I am very grateful to you for making it available to the public. It’s a really wonderful tool however there are few things that I would like to know. I am working to assess sustainability of MHP projects. As sustainability is very broad term and can be defined in many ways in different context, I need to clearly present my theoretical framework. Project description allows 400 char but only three lines are visible to participants. Similarly in one of my criteria(Environmental dimension) there is only one sub-criteria, as comparison is not needed this tool doesn’t allow me to include the sub criteria while defining hierarchy however the final hierarchy what participants see is not the correct one as there is one more sub-criteria under it. So please could you see if something could be done or suggest me some idea to deal with it.
Thanks for your feedback.
A) 400 char equals 5 lines à 80 chars or 3+ lines à 130 chars. The full length (400 chars) of project description is visible to participants. I just tested it again. If you need more explanation for your participants, include it in an email with the session link to start their inputs.
B) Pairwise comparisons require at least 2 criteria. If you don’t have 2 sub-criteria below a main criterion, the weight of the main criterion is evaluated and corresponds to the weight of your missing (single) sub-criterion. Revise your hierarchy or consider renaming your category/sub-criterion to make it clear to your participants.
Thank-you for replying. will definitely consider it 🙂
.
Can i know why the summary worksheet don’t show the consolidate result even i have already put in all inputs and adjusted the consistency.
Did you select the correct number of participants and selected participant “0”?
Wonderful tool! one quick question, what is the difference between
AHP group consensus and rel. homogeneity. Which measure is best to use to report how similar the answer of group participants were?
Kind regards,
Marieke
In principle you can use both. AHP group consensus is adjusted for AHP to give a range from 0% to 100%. See my paper eq. 10 and 14 here
hi
its very good system. But one problem, i was filling the data, took much time, session out error. I lost all the efforts and thoughts put on.
can u do something about it?
I’m sorry. Will take out the session time limit.
how to input the matrix like this :
1 7 3 3
1/7 1 2 1/3
1/3 1/2 1 1/2
1/3 3 2 1
It’s a 4×4 matrix, i.e. 4 criteria, let them call C1, C2, C3, C4.
1st line compares C1 with C2 (C1 is 7-times more important than C2 => A7, C1 is 3-times more important than C3 => A3, C1 is 3-times more important than C4 => A3)
2nd line compares C2 with C3 and C4 (C2 is 2-times more important then C3 => A2, C2 is 3-times less important than C4 => B3)
3rd line compares C3 with C4 (C3 is 2-times less important than C4 => B2)