4. Results
4.2. Hypothesis testing
Table 6b
In addition, to test our hypothesis we also ran OLS regressions which is based upon the following equation:
TotalAmountNewDonationssqrt = INTERCEPT + b1Genderfemales + b2Gendermales + b3Q2Age + b4Q3Education + b4Q6aAgePartner + b5Q6bEducationPartner +
b6Q7Netincome + b7Q9Charitywork + b8Treatment + b9TotalAmountFormerDonations + e
Using SPSS version 27, we firstly ran a linear regression for each type of donor which included all variables and indicated significance for the regular donor group: F = 2.986; p <
.003, for which we refer to the models 1 in table 7 below. Model 1 showed non-significance for the no donor group: F = .556; p < .807 and for the non-regular donor group: F = 1.145; p
< .369.
To arrive at a best possible fit for predicting our dependent variable, we conducted multiple regressions of which the Spearman correlations from table 5 gave us indications for the possible beta’s. We also used the independent variable ‘Treatment’ in our regressions which is used to test our hypothesis.
However, the proposed beta’s for ‘treatment’ coded with a ‘1’ for treatment and ‘0’
did not improve our model. The best fit for the non-regular donor group and the regular donor group are provided by models 2 which explain .153 of variance for the non-regular group at F
= 2.626, p < 0.053 and .151 of explained variance for the regular donor group at F = 6.172; p
< 0.001 group, which are both to be considered as weakly in terms of adjusted R2’s.
Table 7 OLS regressions by type of donor
From the t-test and from this OLS we can conclude that there is no significant effect on showing additional project performances to donors or not. Therefor we cannot find support for hypothesis H1 that donation requests with extended reporting on the performance of projects have a positive effect on receiving donations from regular private donors
Furthermore, in our regressions we find no support for our hypothesis H2 that donation requests with extended reporting on the performance of projects have a negative effect on receiving donations from incidental private donors.
Also, from our regressions we find no support for our hypothesis H3 that donation requests with extended reporting on the performance of projects have a positive effect on receiving donations from non-givers.
DonorType Gender =
females
Gender = males
Q2Age Q3Education Q6aAge Partner
Q6bEducation Partner
Q7Netincome Q9Charity-work Individual or Partner
Treatment TotalAmount Former Donations
No donor Model 1
Constant** included -0,117 -.042 .004 .020 .036 .197 .064 -.150
-Adjusted R-squared -.080 -.080
F .556
Sig. .807
N 49
Model 2
Constant*** included -.042 0.071 .159
Adjusted R-squared -.014
F .765
Sig. .519
N 49
Non-regular Model 1 .
Constant included .002 .182 -.250 .391 -.190 -.093 .217 -.151 .287*
Adjusted R-squared .036 .036
F 1.145
Sig. .369
N 36
Model 2
Constant** included -.268 .389* -.184 .370**
Adjusted R-squared .153
F 2.626
Sig. .053*
N 36
Regular Model 1
Constant*** included .043 .065 -.219** -.281** .128 .038 -0.27 -.120 .381***
Adjusted R-squared .136 .136
F 2.986
Sig. .003
N 115
Model 2
Constant*** included -.198** -.217* .142 .390***
Adjusted R-squared .151
F 6.172
Sig. <.001
N 115
* significance at the 0.10 level
** significance at the 0.05 level
*** significance at the 0.01 level Dependent variable:
TotalAmountNew Donationssqrt
Because we were interested in how well tied donors are to certain organizations, in our survey we asked the individuals that decided to donate, which considerations they take into account when entrusting ngo’s with a donation. Respondents were allowed to select multiple answers, of which the graph below represents the results for each type of donor.
From a first view, we see that the overall percentage of the non-regular donors to each category are somewhat lower than those of the regular givers. The (former) no donor group and regular donor group value the CBF-brand for national and international organizations almost equally. The no former donor group and regular donor group both score high
percentages on the personally known category, which might indicate that within the no donor group there are potentially new donors for the future who, in their present life stage, lack income to donate to charity, which we explain from the fact of having a lot of young individuals (37% < 24 years) in our sample (table 2a).
The selection of the ANBI category is quite low for the former no donor group, which is to be explained from the fact that an ANBI registration of a non-profit organization is necessary for donors to deduct gifts from income tax, which is not important to know when you are a non-giver.
The recommendation of an organization by the inner circle category appears to be less valued by the no donor group than both other groups and is also slightly lower than CBF
0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 %
CBF brand active only in NL
CBF brand NL + Devel. Countries
ANBI
Recommendation inner circle
Supported by government
Personally Known
Considerations by trusting an ngo
No donors(n=137) Non-regular donors (n=45) Regular donors(n=214)
branding for national as well as international organizations, which might indicate that the no former donor group mostly relies on the CBF brand when performing the act of giving.
Overall, the personally known category with high percentages for the no donor and regular donor appears to be of major importance to all groups which indicates that personal ties or familiarity, not meaning individuals from the inner circle, are most important to donors.
The reason why non-regular donors are scoring lesser on certain options may indicate that these people, who are giving incidentally, do not go too deeply into the considerations for donating and seem not to display certain need for bounding with an ngo.