Cracking email earlier today from Limehouse resident Alicia Joseph to the Council calling into question the validity of the latest Tower Hamlets Residents Survey.
The Residents Survey was carried out by the oddly named (for a communications company) Westco Trading Ltd – until you realise ‘Westco’ is short for Westminster Council – because Westco is the trading arm of Westminster Council.
Take it away Alicia!
I am West Indian and so are my family, my neighbours are Bangladeshi, Indian from Pakistan, White British others are from Ethiopia, Eritrea, Guyana and Cote d’Ivoire and Ireland (south and north).
We have not been asked, just our group alone would have provided a very broad sample of individuals and YET we have not been included in the sample and at no point where the company investigating these areas.
So WHERE were they investigating?
We know the borough is separated into gentrified and council estates, predominantly one ethnic grouping in one estate and not in another. So, although data analysis would suggest that the council have created a fair methodology they have not, they are not familiar with the local area and their only aim is to provide 1104 surveys for the council.
Are you saying that you have contracted and then subcontracted the work, therefore are you spending the public money twice?
Because of this same subcontracting, are you able to verify that the door to door work has been done adequately.
What are the audits and quality controls in place by the council for this work to be done correctly?
You have sampled 0.02% of the population – do you believe this is an adequate representation of the total in Tower Hamlets of approximately 308,000 residents (based on the ONS mid-year population estimate June 2017)?
Tower Hamlets is a densely populated, diverse expanding borough. A survey of 1,104 residents maximum is not sufficient to glean the appropriate information.
Point taken about being representative according to the census data. But when was this census data last updated?
If it’s from the last MORI poll 2011, then it is out of date. Seeing as the borough is extremely fluid and ever changing, with an increasingly younger population as well as much forgotten about cohort of older people 65+, surely this cannot be the right methodology of appropriating data.
When did the face to face interview take place and over what period of time? Are you targeting certain households across all postcodes or are you visiting certain postcodes on certain days and then working out the data from there?
What were the following:
- The ethnicity of those chosen and why
- The locality of the door to door knocking
- A breakdown of the age groups targeted
- When was this survey carried out
- On what grounds was it decided to single out those chosen to participate in the survey
I reject all of the information that has been provided in your survey. Although it may sound professional and substantial, the key fact remains that only 1,104 people maximum have been targeted and consulted. 0.02% of the population is not a representation of residents in Tower Hamlets nor is it the voice of the people in Tower Hamlets.
I look forward to your email.
1,104 or 2,383?
The Love Wapping Statistics Team did some very brief research into how many people would be needed to get a representative sample of Tower Hamlets opinion and it seems that for our population of 308,000 a sample of 2,383 residents would be needed (with a margin of error of 3%, and a confidence level of 95%) and 11,915 people would have to be invited to take part to get the 2,383 required.
LW has already asked the Council for some details on how the survey was conducted but fear we may have to submit a Freedom of Information (FoI) request. And we all know what happens to those, don’t we? (allegedly)
Related Internet Links
- Sample size determination – Wikipedia
- How to determine population and survey sample size? – Checkmarket
- Calculate representative sample size – Checkmarket
Huge thanks to Alicia for her unsolicited but very welcome contribution to LW. Why not sign her petition to reclaim the Limehouse Triangle for the local community to say thanks?
Moley had a fit of the giggles when he read this in the Trust and Transparency section of the Residents Survey report (page eight):
“Nearly seven in ten residents said they trust the council a great deal or a fair amount (69 per cent).”
Maybe by ‘seven in ten residents’ the survey people actually meant ‘seven of the ten senior Council officers sitting in the Town Hall canteen when we happened to be there’?
Our busy editorial office received the official Council press release about the Residents Survey on 30th August, read the first few lines then ignored the rest as is our normal practise. [This is a lie, we rarely read more than the email subject, Ed.]
We don’t print press releases from anyone, especially the Council, as we do not reproduce public relations nonsense unless we challenge every word (other publications covering East London please note).
This was the first paragraph of the release:
“Nearly eight out of ten Tower Hamlets residents (78 per cent) continue to agree that people from different backgrounds get on well together, are positive about most local services, and 69 per cent say that the council is doing a good job.”
Moley laughed so much at this gem he did a wee wee in his mole trousers (no moleskin jokes please).
Third time that has happened this week and it’s only Wednesday.