Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Evaluating face-to-face fundraisers

Wendsjö, Albert LU (2019) STAH11 20192
Department of Statistics
Abstract
Face-to-face fundraising is a method used for raising money by having fundraisers working on public spaces. Fundraisers either ask for one time donations, or sign up people to give on a monthly basis, in this thesis the focus is on the latter. For this method to be useful all fundraisers must meet their target goal, meaning that they must get a certain amount of sign ups when they work. By using data on fundraiser results for Sweden for UNHCR we study how fundraisers develop. Using a Poisson regression model with mixed effects we examine whether fundraisers become better over time, whether the results differ between cities and if helpful predictions can be made. On a population level it is found that fundraisers become 0.9% better for each... (More)
Face-to-face fundraising is a method used for raising money by having fundraisers working on public spaces. Fundraisers either ask for one time donations, or sign up people to give on a monthly basis, in this thesis the focus is on the latter. For this method to be useful all fundraisers must meet their target goal, meaning that they must get a certain amount of sign ups when they work. By using data on fundraiser results for Sweden for UNHCR we study how fundraisers develop. Using a Poisson regression model with mixed effects we examine whether fundraisers become better over time, whether the results differ between cities and if helpful predictions can be made. On a population level it is found that fundraisers become 0.9% better for each week worked. This differs for all fundraisers, and at the extremes some fundraisers become 7% better/worse over time. It is also found that the results in most cities are rather equal, with the exception that the results are 30% higher in Stockholm. Lastly we used the model for predicting some results using the data of the first 5 weeks of two fundraisers. In both cases the prediction intervals are too broad to determine whether the fundraisers will reach the goal or not. (Less)
Please use this url to cite or link to this publication:
author
Wendsjö, Albert LU
supervisor
organization
course
STAH11 20192
year
type
M2 - Bachelor Degree
subject
keywords
fundraising, poisson regression, mixed effects, GLMM, bootstrap
language
English
id
9004698
date added to LUP
2020-03-10 09:29:03
date last changed
2020-03-10 09:29:03
@misc{9004698,
  abstract     = {{Face-to-face fundraising is a method used for raising money by having fundraisers working on public spaces. Fundraisers either ask for one time donations, or sign up people to give on a monthly basis, in this thesis the focus is on the latter. For this method to be useful all fundraisers must meet their target goal, meaning that they must get a certain amount of sign ups when they work. By using data on fundraiser results for Sweden for UNHCR we study how fundraisers develop. Using a Poisson regression model with mixed effects we examine whether fundraisers become better over time, whether the results differ between cities and if helpful predictions can be made. On a population level it is found that fundraisers become 0.9% better for each week worked. This differs for all fundraisers, and at the extremes some fundraisers become 7% better/worse over time. It is also found that the results in most cities are rather equal, with the exception that the results are 30% higher in Stockholm. Lastly we used the model for predicting some results using the data of the first 5 weeks of two fundraisers. In both cases the prediction intervals are too broad to determine whether the fundraisers will reach the goal or not.}},
  author       = {{Wendsjö, Albert}},
  language     = {{eng}},
  note         = {{Student Paper}},
  title        = {{Evaluating face-to-face fundraisers}},
  year         = {{2019}},
}