The final campaign opinion polls from June 2017 are listed here
| Pollster | Sample dates | Sample size | CON% | LAB% | LIB% | UKIP% | Green% |
|---|---|---|---|---|---|---|---|
| ELECTION 2015 | 7 May 2015 | 37.8 | 31.2 | 8.1 | 12.9 | 3.8 | |
| Opinium | 04 Jun 2017 - 06 Jun 2017 | 3,002 | 43 | 36 | 8 | 5 | 2 |
| Kantar Public | 01 Jun 2017 - 07 Jun 2017 | 2,159 | 43 | 38 | 7 | 4 | 2 |
| Panelbase | 02 Jun 2017 - 07 Jun 2017 | 3,018 | 44 | 36 | 7 | 5 | 2 |
| ComRes/Independent | 05 Jun 2017 - 07 Jun 2017 | 2,051 | 44 | 34 | 9 | 5 | 2 |
| YouGov/The Times | 05 Jun 2017 - 07 Jun 2017 | 2,130 | 42 | 35 | 10 | 5 | 2 |
| ICM/The Guardian | 06 Jun 2017 - 07 Jun 2017 | 1,532 | 46 | 34 | 7 | 5 | 2 |
| Ipsos-MORI/Evening Standard | 06 Jun 2017 - 07 Jun 2017 | 1,291 | 44 | 36 | 7 | 4 | 2 |
| Survation | 06 Jun 2017 - 07 Jun 2017 | 2,798 | 41 | 40 | 8 | 2 | 2 |
| AVERAGE | 04 Jun 2017 - 07 Jun 2017 | 17,981 | 43.2 | 36.4 | 7.9 | 4.3 | 2.0 |
| Actual Result | 08 Jun 2015 | 43.5 | 41.0 | 7.6 | 1.9 | 1.7 |
The pollsters had an error in the gap between the Conservatives and Labour. They saw an average gap of 6.8%, but the actual gap was only 2.5%.
| Pollster | Sample dates | Sample size | CON% | LAB% | LIB% | UKIP% | Green% | SNP% |
|---|---|---|---|---|---|---|---|---|
| ELECTION 2015 | 7 May 2015 | 14.9 | 24.3 | 7.5 | 1.6 | 1.3 | 50.0 | |
| Ipsos-MORI/STV | 22 May 2017 - 27 May 2017 | 1,016 | 25 | 25 | 5 | 1 | 1 | 43 |
| Panelbase/Sunday Times | 26 May 2017 - 31 May 2017 | 1,021 | 30 | 20 | 5 | 2 | 1 | 42 |
| Survation/Sunday Post | 31 May 2017 - 02 Jun 2017 | 1,024 | 27 | 25 | 6 | 1 | 1 | 40 |
| YouGov/Times | 01 Jun 2017 - 05 Jun 2017 | 1,093 | 26 | 25 | 6 | 1 | 1 | 41 |
| Poll Average | 20 Apr 2015 - 01 May 2015 | 4,154 | 27.0 | 23.8 | 5.4 | 1.3 | 1.0 | 41.5 |
| Actual Result | 08 Jun 2017 | 28.6 | 27.1 | 6.8 | 0.2 | 0.2 | 36.9 |
There was some noticeable error in the polls for Scottish voting intention as well. Labour's vote was underestimated, and the SNP vote was overestimated. The SNP lead over Labour was estimated at 18pc, but was actually only 10pc. So the loss of SNP seats was more than was expected. The broad outline of the result was unaffected — the polls correctly suggested that the SNP would lose seats, but still win most seats, and that the Conservatives would gain seats.