We are seeking sponsorship for a programme of research to measure the effects of the Covid-19 pandemic on enterprise in the UK. We consider that a new approach is required because of the inaccuracies associated with the BICS research programme, as provided by the Office of National Statistics. Details of the problems with BICS are described elsewhere on this website (see reflections and also the comments on the Purple Research report as published earlier this year) but the principal difficulties are due to the exclusive use of on-line data collection, which ensures that those communities where Covid-19 is most prevalent are under-represented in the data produced by the ONS. Thus the official data on the effects of the pandemic consistently under reports the true position.
This under-reporting of the true effects of the pandemic is further exacerbated by the fact that the sample for BICS is drawn exclusively from the IDBR business sample frame, as provided to the ONS by HMRC. Use of that frame for the sample excludes many small businesses, primarily those operated by the self-employed with a turnover below the £85,000 VAT threshold.
To ensure better coverage of the whole UK economy we propose to use telephone data collection and to conduct two separate but linked surveys, one aimed at the population of all persons aged 16 or more and the other targeted at businesses. Use of the telephone for the main elements of our work has the following advantages:
The respondent to be a company director or senior manager and the topic is concerned with the effects of the Covid-19 pandemic, the associated government advice and their plans for dealing with the consequences. As with the population survey we intend to build upon trial work completed during the previous lockdown, the results for which can be reviewed here; we would also expect to replicate some key questions from the BICS programme as provided by the ONS
Further details of the proposed methods are:
Sample: The sample design is for a sample of a minimum of 1,200 responses, stratified by UK region, broad activity group and employee size. This would be an ‘ideal’ sample for reasons of the analytical advantages to be gained, in particular it will enable a snapshot of the position within each of the UK regions. Further information is outlined in the section on results below.
We would select the sample from our in-house copy of the Dun & Bradstreet Business File for the UK, coding each selection to identify which of the quota cells they fall into. Details of this design are available on request but It should be noted that this sample source does not rectify the difficulties associated with the BICS series as noted above, it is for that reason that we propose also to conduct the population survey. It is that survey which will provide the data on the self-employed and other small businesses, not covered by the BICS. Meanwhile a careful record will be maintained of no contacts and other non-effective responses by region and other sample characteristics and any response bias will be adjusted for.
The Interview: As stated in the introduction the majority of the work will be conducted by telephone and, whenever we have the respondents’ agreement, responses will be recorded. We will therefore begin the interview with a straightforward open ended question to hear what each respondent says, when told of the topic. Subsequently these ‘open ends’ will be analysed using the latest software and the results will not only have the value of unvarnished reactions but also used to frame questions in subsequent weeks, if appropriate.
We would then say “I would now like to ask you some specific questions that we are asking all businesses …”.
With the following proposed topics:
Additional classification questions to be asked are:
Finally, at the end of the interview we will offer them a feedback report and seek an email address to send the report to and invite them to participate in a follow up after a further week. All respondents will also be informed of the list of survey sponsors.
Results: The proposed sample structure will enable useful comparisons to be drawn across each of our 12 regions, main activity groups and sizes of business as well as identifying any differences between foreign owned and UK owned companies. Additionally we will be able to determine any effects by age of the company and the main bank they use.
Our proposal is to cover much the same ground as that explored in our original Monitor trial study but modifying the core questionnaire to measure the specific effects resulting from Covid-19, including the chancellor’s initiatives. We will therefore include these specific questions as used in the trial:
together with the appropriate follow up questions in each case.
We will also include questions relating to redundancy, furlough status and even business start-ups and closures.
We also intend that the results shall be compared with the work of the ONS and we therefore propose to replicate the relevant questions as used in the current Opinions and Lifestyle Survey (OPN), these include:
In the past seven days, have you avoided contact with older people or other vulnerable people because of the coronavirus (COVID-19) outbreak?
Sample: The proposed sample will be a mixed mode RDD (random digit dial) sample to ensure no possibility of selection bias, other that any bias introduced by differential response. The landline part of the sample will be selected to be representative of each region and, indeed, city versus country. The Mobile part of the sample will be stratified by mobile network. Response characteristics will be carefully monitored and the data weighted to account for any differential response rates. We intend a minimum sample of 2,400 but would hope for support to achieve significantly more.
We are seeking sponsors for either or both surveys and we also welcome any ideas for specific questions. A special question service is also available at a minimum cost of £3,000 for ether survey.
Contact: For information on sponsorship please email [email protected]. If you require further information on the technical details of the surveys please email [email protected] or call Tony Dent on 07885088521.