| A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Prolific Poll (UK) — September 2026 | |||||||||||||||||||||||||
2 | Political donations: full data tables | |||||||||||||||||||||||||
3 | ||||||||||||||||||||||||||
4 | Commissioned by | None. This survey was designed, funded, conducted and published by Prolific for its own research. There is no external client or sponsor. | ||||||||||||||||||||||||
5 | Conducted by | Prolific | ||||||||||||||||||||||||
6 | Fieldwork | 21 September 2026 | ||||||||||||||||||||||||
7 | Method of interview | Online, self-administered; survey scripted in Qualtrics | ||||||||||||||||||||||||
8 | Universe represented | UK adults aged 18 and over | ||||||||||||||||||||||||
9 | Sample size | 1,500 UK adults. Each respondent evaluated six hypothetical donations, giving 9,000 vignette evaluations in total. | ||||||||||||||||||||||||
10 | Sampling | Prolific opt-in online participant panel, recruited to quotas for age, sex and region. Age and sex come from Prolific's participant-level profile data and were not re-asked in this survey. | ||||||||||||||||||||||||
11 | Weighting | Weighted to the UK adult population on age and sex, region, education by age, housing tenure, and 2024 general election vote, using regularised raking. Effective sample size approximately 912; design effect 1.64; maximum weight 4.75, with no respondent weighted above five times the average. | ||||||||||||||||||||||||
12 | Weighted or unweighted | The experimental results in this workbook are UNWEIGHTED. Treatment was randomised within each respondent, so the comparisons are identified by that randomisation rather than by the composition of the sample. Weighted equivalents are reported alongside as a sensitivity check wherever they were produced. The two policy questions are WEIGHTED, because there the aim is to describe the population rather than to estimate a randomised effect. Every table states which applies. | ||||||||||||||||||||||||
13 | Margin of error | As with any sample survey these results are subject to a margin of error. For a finding of 50% on the full weighted sample this is approximately +/-3.2 percentage points at the 95% confidence level; findings further from 50% carry a slightly smaller margin of error. As the British Polling Council notes, all polls are subject to a wide range of potential sources of error beyond sampling variation alone. | ||||||||||||||||||||||||
14 | Questions covered | The six-vignette donations experiment (DON_V1 to DON_V6), the perceived-influence question (DON_INFLUENCE), and the two policy questions (DON_CAP_SUPPORT, DON_PUBLIC_FUNDING). The full questionnaire is published separately. | ||||||||||||||||||||||||
15 | Published alongside | “Britons don't think large donations from overseas donors should be legal. Except Reform voters”, Prolific, September 2026. | ||||||||||||||||||||||||
16 | BPC membership | Prolific is a registered member of the British Polling Council and abides by its Rules of Disclosure. This workbook contains the full response distributions, weighted and unweighted bases, demographic crosstabs and model estimates for every question covered. Anything further, including a fuller description of the weighting procedure, is available on request. | ||||||||||||||||||||||||
17 | Contact | oriol.jover@prolific.com | ||||||||||||||||||||||||
18 | ||||||||||||||||||||||||||
19 | Contents | |||||||||||||||||||||||||
20 | ||||||||||||||||||||||||||
21 | Design | The three randomised elements, their levels, and the within-respondent balance | ||||||||||||||||||||||||
22 | Exp_Grid | Table 1. Acceptance by amount × donor profile, the 18-cell grid | ||||||||||||||||||||||||
23 | Exp_Amount | Table 2. Acceptance by amount, raw and model-adjusted, plus contrasts | ||||||||||||||||||||||||
24 | Exp_DonorProfile | Table 3. Acceptance by donor profile, plus residence and registration contrasts | ||||||||||||||||||||||||
25 | Exp_RecipientParty | Table 4. Acceptance by recipient party and own-party effects | ||||||||||||||||||||||||
26 | Exp_Amount_byVI | Table 5. Amount by current voting intention, with per-tenfold slopes | ||||||||||||||||||||||||
27 | Exp_Donor_byVI | Table 6. Donor profile by current voting intention, with effects | ||||||||||||||||||||||||
28 | Model_estimates | Table 7. All model contrasts and specification tests (LEGAL outcome) | ||||||||||||||||||||||||
29 | Policy_Toplines | Table 8. Headline support for a cap and for public funding | ||||||||||||||||||||||||
30 | Dist_LEGAL | Table 9. Full four-point distribution, primary outcome | ||||||||||||||||||||||||
31 | Dist_ACCEPT | Table 10. Full four-point distribution, secondary outcome | ||||||||||||||||||||||||
32 | Exp_Grid_byParty | Table 11. All 36 experimental combinations | ||||||||||||||||||||||||
33 | DON_INFLUENCE | Table 12. Perceived influence of large donors, 0-10 | ||||||||||||||||||||||||
34 | Policy_Distributions | Table 13. Policy questions in full, and the question-order effect | ||||||||||||||||||||||||
35 | Policy_Demographics | Table 14. Policy questions by demographic group | ||||||||||||||||||||||||
36 | Exp_Demographics | Table 15. Experiment results by demographic group | ||||||||||||||||||||||||
37 | Model_estimates_ACCEPT | Table 16. Model estimates for the secondary outcome | ||||||||||||||||||||||||
38 | ||||||||||||||||||||||||||
39 | How to read this workbook | |||||||||||||||||||||||||
40 | ||||||||||||||||||||||||||
41 | The experiment | Every respondent judged six hypothetical donations. Each one is a single sentence built from three independently randomised elements: the amount, the donor's connection to the UK, and the recipient party. The six were balanced within each respondent, so everyone saw each amount once, met each donor profile twice, and judged three donations to Labour and three to Reform UK. See the Design sheet. | ||||||||||||||||||||||||
42 | Acceptance rates | Unless a table says otherwise, the figure shown is the share saying a party should be “legally allowed” to accept the donation, that is the two categories “Definitely should allow” and “Probably should allow” combined. The underlying question used a four-point scale. | ||||||||||||||||||||||||
43 | Shares and effects | Two different quantities appear in this workbook, and they are labelled separately. A share is a percentage of people or of answers. An effect is a difference between two shares and is reported in percentage points, with a 95% confidence interval. | ||||||||||||||||||||||||
44 | Bases | Tables built on the grid of amount by donor profile carry at most one answer per respondent per cell, so the respondent base and the answer base are the same. Tables pooled across amounts carry two answers per respondent per donor profile; those state an answer base. Both are given. | ||||||||||||||||||||||||
45 | Model estimates | Within-person estimates come from linear probability models with respondent fixed effects and standard errors clustered by respondent. See the Model estimates sheet. | ||||||||||||||||||||||||
46 | Rounding | Percentages are rounded to one decimal place. Rounding can produce totals that do not sum to exactly 100%. | ||||||||||||||||||||||||
47 | Caution | This is an opt-in online panel, not a random probability sample. The strongest claims in this workbook are the randomised within-person comparisons. Absolute levels, and generalisation to the wider population, warrant more caution. Donations described in the experiment were hypothetical. | ||||||||||||||||||||||||
48 | Subgroup scope | Two deliberate choices. The policy-question demographic breakdowns (Table 14) pool both question orders, because splitting by order as well would leave several cells too thin to report. The experiment demographic crosstabs (Table 15) exclude current voting intention, which has its own fuller treatment in Tables 5 and 6. | ||||||||||||||||||||||||
49 | Empty cells | Where no respondent in a group selected a given answer the figure is shown as 0.0% and no confidence interval is reported, because none can be estimated. Four cells on the DON_INFLUENCE sheet are affected and are listed there. | ||||||||||||||||||||||||
50 | ||||||||||||||||||||||||||
51 | ||||||||||||||||||||||||||
52 | ||||||||||||||||||||||||||
53 | ||||||||||||||||||||||||||
54 | ||||||||||||||||||||||||||
55 | ||||||||||||||||||||||||||
56 | ||||||||||||||||||||||||||
57 | ||||||||||||||||||||||||||
58 | ||||||||||||||||||||||||||
59 | ||||||||||||||||||||||||||
60 | ||||||||||||||||||||||||||
61 | ||||||||||||||||||||||||||
62 | ||||||||||||||||||||||||||
63 | ||||||||||||||||||||||||||
64 | ||||||||||||||||||||||||||
65 | ||||||||||||||||||||||||||
66 | ||||||||||||||||||||||||||
67 | ||||||||||||||||||||||||||
68 | ||||||||||||||||||||||||||
69 | ||||||||||||||||||||||||||
70 | ||||||||||||||||||||||||||
71 | ||||||||||||||||||||||||||
72 | ||||||||||||||||||||||||||
73 | ||||||||||||||||||||||||||
74 | ||||||||||||||||||||||||||
75 | ||||||||||||||||||||||||||
76 | ||||||||||||||||||||||||||
77 | ||||||||||||||||||||||||||
78 | ||||||||||||||||||||||||||
79 | ||||||||||||||||||||||||||
80 | ||||||||||||||||||||||||||
81 | ||||||||||||||||||||||||||
82 | ||||||||||||||||||||||||||
83 | ||||||||||||||||||||||||||
84 | ||||||||||||||||||||||||||
85 | ||||||||||||||||||||||||||
86 | ||||||||||||||||||||||||||
87 | ||||||||||||||||||||||||||
88 | ||||||||||||||||||||||||||
89 | ||||||||||||||||||||||||||
90 | ||||||||||||||||||||||||||
91 | ||||||||||||||||||||||||||
92 | ||||||||||||||||||||||||||
93 | ||||||||||||||||||||||||||
94 | ||||||||||||||||||||||||||
95 | ||||||||||||||||||||||||||
96 | ||||||||||||||||||||||||||
97 | ||||||||||||||||||||||||||
98 | ||||||||||||||||||||||||||
99 | ||||||||||||||||||||||||||
100 |