Your article (Momentum tipped to back Long-Bailey as poll puts her in front, 16 January) quotes a poll published by LabourList, carried out by Survation. This poll is simply not an accurate view of Labour members’ current voting intentions.
As the article states, this is a poll of LabourList readers only, who will be more politically engaged than the average Labour member. Choosing such an unrepresentative sample invalidates the results.
One of the main reasons for the polling disaster at the 2015 general election was that polling companies included too many politically engaged people in their samples.
The LabourList sample only consists of politically active and informed members, when we know from Prof Tim Bale’s work that most Labour members do not engage with the party beyond occasionally liking items on social media. Survation weighted the survey respondents to reflect the demographics of the Labour membership, but since they do not deal with the imbalance on political engagement, the sample is still unrepresentative.
The results of this survey were very different to those of the earlier YouGov poll, based on a sample of all members, which put Keir Starmer firmly ahead in the leadership race. This should ring alarm bells.
Cllr Christabel Cooper
Labour councillor and data analyst, London
• Brian Wilson (Letters, 11 January) points out that Labour might do better to think about the wishes of the electorate rather than its membership. That put me in mind of Archbishop William Temple’s supposed quote that the church is the only society that exists for the benefit of those who are not its members; if Labour is a “broad church”, it still needs to be looking to the needs of those it serves.
Rev Canon Cllr Steve Parish
• Join the debate – email email@example.com
• Read more Guardian letters – click here to visit gu.com/letters
• Do you have a photo you’d like to share with Guardian readers? Click here to upload it and we’ll publish the best submissions in the letters spread of our print edition