Klik op een omslag om naar Google Boeken te gaan.
Bezig met laden... Data Mining for Fund Raisersdoor Peter B. Wylie
Geen Bezig met laden...
Meld je aan bij LibraryThing om erachter te komen of je dit boek goed zult vinden. Op dit moment geen Discussie gesprekken over dit boek. geen besprekingen | voeg een bespreking toe
Geen bibliotheekbeschrijvingen gevonden. |
Actuele discussiesGeen
Google Books — Bezig met laden... GenresGeen genres LC-classificatieWaarderingGemiddelde:
Ben jij dit?Word een LibraryThing Auteur. |
Below is the synopsis I wrote at the time I read it:
Purpose of Book
* To provide a general outline of a statistically-oriented method to improve funding activities by mining your current donor database
* To provide general techniques for analyzing data, as well as provide cautions against bad techniques
How the Process Can Improve Endowment Activities
* Allows the organization to more accurately target quality prospects, either to increase participation rates, or to find major givers more inclined to donate
* Allows the organization to reduce costs, or more effectively use limited resources, i.e., phone smaller sets of people, limit the size of mailings, while increasing donations
Outline of Method (Non-Technical)
1. Export sample of donor database
2. Split sample into smaller components
3. Find relationships between donor features and giving
4. Select the significant variables
5. Develop scoring system
6. Validate findings
7. Test finding on limited appeals and compare results
Assumptions
* Assumes the donor data is extractable and randomized
* Requires export from donor database, or access via SQL
* Assumes additional software for statistics (DataDesk, SAS, SPSS)
Limitations
* Requires IT staff, analytical staff, donor contacts, and management to coordinate efforts
* Requires IT and analytical staff have adequate skills to implement
* Judges variables of data by both its intrinsic value and based upon its inclusion in database ( )