Path to No10: UK Regional & Demographic Polling

The map shows the leading party in each region based on the latest weighted 7-poll moving average of regional polling data. All constituencies within a region are coloured according to that region's polling leader.

• Regional polling data is aggregated from multiple pollsters (YouGov, More in Common, Find Out Now, Survation, and others) using regional cross-tab data from national polls.
• Weighting methodology: Averages are calculated using sample-size weighting within a 7-poll moving window. Polls with larger sample sizes carry greater weight.
• 7-poll window: Only the most recent 7 polls are included. As new polls arrive, older polls drop out of the window, ensuring the average reflects current trends.
• Important note on cross-tab data: Regional polling data from cross-tabs can be less reliable than dedicated regional polls due to smaller sample sizes. Results should be interpreted with caution, especially for smaller regions.
• Selecting a region displays a detailed polling trends chart with 7-poll moving average, plus the underlying polling data.

For seat projections, please use our Demographic Swingometer, which uses MRP modelling for constituency-level predictions.

Data Standardization:

• Age bins from different pollsters are standardized to five categories: 18-24, 25-34, 35-49, 50-64, and 65+.
• Non-standard age ranges (e.g., 18-29, 30-39, 40-54) are mapped to the closest standard bin based on the midpoint of the range. This ensures consistent comparison across pollsters despite varying age categorizations, though some precision may be lost when mapping wider ranges to narrower standard bins.

Aggregation Method:

• 7-poll moving average: Each data point represents a weighted average of the last 7 polls up to and including the current poll (trailing window).
• Sample-size weighting: Polls with larger sample sizes are given greater weight, as they have lower margins of error and are statistically more reliable.
• Poll-based calculation: Each poll is treated as its own entity, regardless of date. Multiple polls on the same day are each counted separately in the moving average window.

Data Sources: Age demographic data is extracted from multiple pollsters including YouGov, Survation, Find Out Now, More in Common, Opinium, Focaldata, and Freshwater Strategy.

Age-based results come from weighted crosstabs rather than primary samples, which means estimates can be more volatile, particularly for younger and older age groups. These figures should be interpreted as indicative trends rather than precise measurements.

Data Standardization:

• Gender labels from different pollsters are standardized to two categories: Female and Male.
• Variations such as "Women" and "Men" are mapped to "Female" and "Male" respectively.
• Only pollsters that provide Male/Female breakdowns are included in the analysis.

Aggregation Method:

• 7-poll moving average: Each data point represents a weighted average of the last 7 polls up to and including the current poll (trailing window).
• Sample-size weighting: Polls with larger sample sizes are given greater weight, as they have lower margins of error and are statistically more reliable.
• Poll-based calculation: Each poll is treated as its own entity, regardless of date. Multiple polls on the same day are each counted separately in the moving average window.

Data Sources: Gender demographic data is extracted from multiple pollsters including YouGov, Survation, Find Out Now, More in Common, Opinium, Focaldata, and Freshwater Strategy.

Gender-based results come from weighted crosstabs rather than primary samples, which means estimates can be more volatile and should be interpreted as indicative trends rather than precise measurements.

Data Standardization:

• Social class data uses the standard NRS social grade classification: ABC1 (middle and upper class) and C2DE (working class).
• ABC1 includes: A (upper middle class), B (middle class), and C1 (lower middle class).
• C2DE includes: C2 (skilled working class), D (working class), and E (non-working).

Aggregation Method:

• 7-poll moving average: Each data point represents a weighted average of the last 7 polls up to and including the current poll (trailing window).
• Sample-size weighting: Polls with larger sample sizes are given greater weight, as they have lower margins of error and are statistically more reliable.
• Poll-based calculation: Each poll is treated as its own entity, regardless of date. Multiple polls on the same day are each counted separately in the moving average window.

Data Sources: Social class demographic data is currently extracted from YouGov and Survation, as these are the primary pollsters that collect and publish social grade breakdowns.

Limitations: Social class data has more limited coverage compared to age and gender demographics, as not all pollsters collect or publish this breakdown.

Social class results come from weighted crosstabs rather than primary samples, which means estimates can be more volatile and should be interpreted as indicative trends rather than precise measurements.

• What This Shows: Where each party's current voters came from in 2024. For example, if Green shows 40% from Labour, it means 40% of today's Green voters voted Labour in 2024.
• Data Source: Past vote data from YouGov, Find Out Now, Opinium, Freshwater Strategy, Survation, and Verian.
• Moving Average: Uses the last 7 polls, with larger polls weighted more heavily.
• How It Works: We count actual voters moving between parties, not just percentages. This matters because 10% of a big group (like Labour) is more people than 80% of a small group (like Greens).
• Bar Chart/Sankey: Shows where current voters came from in 2024.
• Party Retention: Shows what percentage of 2024 voters stuck with the same party. If Labour shows 70%, it means 70% of 2024 Labour voters still vote Labour.