Ipsos poll, 22 October 2026
Ipsos, fieldwork to 22 October 2026: Labour 32.6%, 8.4 points ahead of Conservative on 24.2%, then Reform UK 19.8, Lib Dem 8.7, Green 8.4. Great Britain figures as published, and England's from the pollster's tables, from 1,109 people.
Fieldwork 15 Oct–22 Oct 2026 · sample 1,109 · GB adults
Vote shares
| Party | GB | England |
|---|---|---|
Conservative |
24.2 | 24.4 |
Labour |
32.6 | 34.1 |
Lib Dem |
8.7 | 9.7 |
Reform UK |
19.8 | 19.7 |
Restore Britain |
2.9 | 2.9 |
Green |
8.4 | 9.2 |
| Your Party | 0.0 | 0.0 |
SNP |
2.9 | 0.0 |
Plaid Cymru |
0.4 | 0.0 |
| Others | 0.1 | 0.2 |
GB figures: Pollster tables (Grade A).
How the England figure was made
England column in the pollster's own tables.
| Basis | HeadlineEquivalent |
|---|---|
| Rounding | Unrounded |
| In the average | Yes Primary (current average) |
| Table | Headline (Total voting) |
| Source | Pollster's published tables |
Checks: ok TOTAL_SUMS_100: Answers in the Total column add up to 100% (expected 100, got 100.09, tolerance 1.5); ok COUNTS_MATCH_SHARES: Printed counts / base match the printed whole-number percentages (563 cells) (expected 0, got 0, tolerance 0); ok ENGLAND_MATCHES_REGIONS: England column agrees with the English regions added up (expected 0, got 0.104, tolerance 1); ok S1_BASE_IS_SAMPLE: Weighted base of the all-respondents table equals the sample size in the footer (expected 1109, got 1109, tolerance 1); ok TOTAL_SUMS_100: Answers in the Total column add up to 100% (expected 100, got 100, tolerance 1.5); ok TOTAL_SUMS_100: Answers in the Total column add up to 100% (expected 100, got 100, tolerance 1.5); ok COUNTS_MATCH_SHARES: Printed counts / base match the printed whole-number percentages (471 cells) (expected 0, got 0, tolerance 0); ok ENGLAND_MATCHES_REGIONS: England column agrees with the English regions added up (expected 0, got 0.161, tolerance 1); ok TOTAL_SUMS_100: Answers in the Total column add up to 100% (expected 100, got 100, tolerance 1); ok COUNTS_MATCH_SHARES: Printed counts / base match the printed whole-number percentages (471 cells) (expected 0, got 0, tolerance 0); ok ENGLAND_MATCHES_REGIONS: England column agrees with the English regions added up (expected 0, got 0.17, tolerance 1); ok TOTAL_SUMS_100: Answers in the Total column add up to 100% (expected 100, got 100, tolerance 1.5); ok COUNTS_MATCH_SHARES: Printed counts / base match the printed whole-number percentages (350 cells) (expected 0, got 0, tolerance 0); ok ENGLAND_MATCHES_REGIONS: England column agrees with the English regions added up (expected 0, got 0.461, tolerance 1); ok TOTAL_SUMS_100: Answers in the Total column add up to 100% (expected 100, got 100.09, tolerance 1.5); ok COUNTS_MATCH_SHARES: Printed counts / base match the printed whole-number percentages (551 cells) (expected 0, got 0, tolerance 0); ok ENGLAND_MATCHES_REGIONS: England column agrees with the English regions added up (expected 0, got 0.209, tolerance 1)
Notes
Upgraded from the pollster's tables (IpsosReader 1.0). England: England column as printed. | Was a GB-only copy from Wikipedia.
From what people said to the headline
EnglandEach row is one step in turning people's answers into a published figure. Party shares exclude don't knows. The last row is our reading of the same raw answers. How our reading works →
| Step | LAB | CON | REF | LD | GRN | RB | Lab lead over Con | Shares |
|---|---|---|---|---|---|---|---|---|
|
What people told the pollster
Everyone asked, weighted to match the population. Don't knows and won't-votes set aside.
|
34.1 | 23.6 | 20.0 | 9.5 | 9.0 | 3.3 | +10.5 |
|
|
After the pollster's turnout step
Headline counts people who say 9 or 10 out of 10 likely to vote and that they always, usually or 'it depends' vote
|
34.1
-0.1
|
24.4
+0.8
|
19.7
-0.4
|
9.7
+0.2
|
9.2
+0.2
|
2.9
-0.5
|
+9.7 |
|
|
The pollster's published headline
Undecided and prefer-not-to-say are asked which party they are most inclined to support; those still undecided are excluded.
|
34.1 | 24.4 | 19.7 | 9.7 | 9.2 | 2.9 | +9.7 |
|
|
Our reading
The raw sample with half the average amount raw samples have missed by since 2015 taken off each party.
|
33.1
-1.0
|
25.6
+2.0
|
19.4
-0.6
|
9.4
-0.1
|
8.7
-0.3
|
3.3 | +7.5 |
|
In the raw sample, don't know 8.5%, would not vote 9.1%, refused 1.0% of those asked in England gave no party. England figures: the pollster's England column. Source: 2026_10_Ipsos Political Monitor Tables October 2026.pdf (the pollster's own tables).
Checks made when the tables were read (17 of 17 passed)
- ✓ VI all: Printed counts / base match the printed whole-number percentages (563 cells)
- ✓ VI all: England column agrees with the English regions added up
- ✓ VI all: Weighted base of the all-respondents table equals the sample size in the footer
- ✓ VI all: Answers in the Total column add up to 100%
- ✓ VI all unweighted: Answers in the Total column add up to 100%
- ✓ VI likely to vote: Printed counts / base match the printed whole-number percentages (471 cells)
- ✓ VI likely to vote: England column agrees with the English regions added up
- ✓ VI likely to vote: Answers in the Total column add up to 100%
- ✓ Headline (Total voting): Printed counts / base match the printed whole-number percentages (471 cells)
- ✓ Headline (Total voting): England column agrees with the English regions added up
- ✓ Headline (Total voting): Answers in the Total column add up to 100%
- ✓ V2 squeeze: Printed counts / base match the printed whole-number percentages (350 cells)
- ✓ V2 squeeze: England column agrees with the English regions added up
- ✓ V2 squeeze: Answers in the Total column add up to 100%
- ✓ LIKEV: Printed counts / base match the printed whole-number percentages (551 cells)
- ✓ LIKEV: England column agrees with the English regions added up
- ✓ LIKEV: Answers in the Total column add up to 100%
Figures as of . Published by Polling Forecast.










