Data collection and methodology
A total of 3,497 interviews were conducted by telephone using computer-assisted telephone interviewing (CATI) systems by IFF Research.
Establishments were not pre-notified that they would be called for the survey, partly due to financial considerations and partly because it was felt that this could lead to a reduction in response rates if head offices potentially opted all the establishments in their organisation out of the survey.
Large multisite organisations and banks were pre-identified during sample processing. This enabled contacts for multisite organisations to be split across a number sample batches and released sequentially over the course of fieldwork to ensure that the various sites were not contacted within too short a time window.
Interviews were conducted with the most senior person at the site with responsibility for recruitment, human resources and workplace skills. Reassurances were provided to respondents prior to the survey, including confirmation that data would be reported in a way that would not allow them or their organisation to be identifiable. If after the first contact the respondent or gatekeeper wanted more information about the survey a reassurance email was sent (see Appendix E for a copy of the reassurance email). This reassurance email included a link to the dedicated survey website which was created and hosted by IFF Research. This website provided further background information on the research, links to the 2017 results, and a list of frequently asked questions.
All interviewers received around a two-hour briefing via video conference. Prior to the briefing, interviewers were provided with a set of briefing notes, advising them on the background and objectives of the study, how sample information had been sourced and guidance on specific questions. The briefings included a detailed run-through of all survey questions by a senior researcher at IFF Research, to help guide interviewers through more complex questions.
Fieldwork took place from 20th October 2020 to 22nd December 2020. Weekly progress updates and feedback ensured the quotas progressed as evenly as possible. However, conducting fieldwork during the COVID-19 pandemic created various challenges in terms of administering the survey. Survey fieldwork coincided with a second spike in COVID-19 cases in Autumn 2020 and the development regional restrictions. This included closure of bars and restaurants in Scotland's central belt from 9th October 2020 until the 2nd November 2020, prior to the introduction of the protection levels. Furthermore, from 20th November 2020, eleven local authorities entered Level 4 restrictions, covering the remainder of the fieldwork period. Under these restrictions, hospitality businesses and close contact services, such as hairdressers and beauty salons, visitor attractions, leisure and entertainment setting and indoor gyms, were forced to close.
IFF Research worked closely with Scottish Government and Skills Development Scotland to monitor and anticipate where restrictions may be introduced and took mitigating actions where possible. Some sectors such as Hotels and Restaurants, Wholesale and Retail, and Arts and Other Services were prioritised early on during the fieldwork period to try and maximise the response rate in these challenging circumstances.
As a result of the COVID-19 pandemic, many employers also moved their employees to homeworking, creating further practical challenges around administering the survey. With many switchboards unable to transfer interviewers to the correct respondents, processes were introduced by IFF Research to collect email addresses for homeworking respondents and email invites were sent offering respondents the opportunity to book an appointment online.
High response rates are central to the success of the Employer Skills Survey. Maximising coverage is especially important in some of the harder to reach sectors and regions that may run the risk of being underrepresented (this was even more of a challenge in the context of COVID-19, as noted above).
The overall response rate for the survey was 40%, calculated as 'achieved interviews' as a proportion of all 'complete contacts'. This compares to a 43% response rate in ESS 2017 and 44% in Scottish EPS 2019. Table 5 provides a detailed breakdown of survey outcomes. Appendix F shows how response rate varied by size and sector.
|Outcome||Number of contacts||% of all sample||% of complete contacts|
|Ineligible establishments (e.g. just 1 working proprietor at site)||1,145||4%||-|
|'Live' / out of quota||15,163||53%||-|
|Unobtainable / invalid numbers||3,656||13%||-|
|Total complete contacts||8,661||30%||100%|
|Quits during interview||164||1%||2%|
The response rate is defined as the number of achieved interviews as a proportion of the total complete contacts.
Regular adjustments were made to the balance of establishments contacted to ensure an even distribution of interviews with employers from different sectors and size bands throughout the fieldwork period. Sample was loaded into fieldwork in proportion to quota targets so that quota progression was as even as possible, and to ensure employers were called and re-called at suitable points without being over-contacted.
During fieldwork, when it became evident that a target quota within a particular cell had become unachievable (i.e. when the number of interviews required to reach the quota target was more than the remaining sample), targets were increased in other cells to compensate, using the following approach:
- In the first instance, fieldwork contractors were to increase the target within the same sector in an adjacent size band.
- If adjacent size bands had also become unachievable then any remaining achievable size bands within the sector were used to compensate.
- In some cases it was preferable to make up the shortfall in a cell within the same sizeband from a different sector. This would be the case when there was a desire to maximise the number of interviews in that sizeband irrespective of other characteristics. This was most common in the larger size bands.
There is a problem
Thanks for your feedback