Scottish Employer Perspectives Survey 2019: technical report

Technical report accompanying the Scottish EPS 2019 research report providing background information on the methodology used.


Using the data

Statistical accuracy of survey data

The table below shows sampling error for the survey results overall and for key sub-groups by which analysis is presented in the report. Figures have been based on a survey result of 50% (the 'worst' case in terms of statistical reliability), and have used a 95% confidence level. Where the table indicates that a survey result based on all respondents has a sampling error of +/- 0.7%, this should be interpreted as follows: 'for a question asked of all respondents in this group where the survey result is 50%, we are 95% confident that the true figure lies within the range 49.3% to 50.7%'.

The calculation of sampling error has taken into account the finite population correction factor to account for cases where we are measuring a significant portion of the population universe (i.e. even if two sample sizes are the same, the sampling error will be lower if in one case a far higher proportion of the population was covered).

Table 8: Sampling error (at a 95% confidence level) associated with findings of 50%, by size and sector

Subgroup Number of interviews achieved Population (Maximum) sampling error (+/-)
Overall
Scotland 2,652 151,200 1.9
Size
2 to 4 643 60,000 3.8
5 to 9 608 31,200 3.9
10 to 24 619 23,900 3.9
25 to 49 364 8,400 5.1
50 to 99 161 4,200 7.6
100 to 249 172 2,300 7.2
250+ 85 1,000 10.2
Sector
Primary Sector and Utilities 140 6,300 8.2
Manufacturing 180 6,400 7.2
Construction 200 10,800 6.9
Wholesale and Retail 540 25,800 4.2
Hotels and Restaurants 310 14,800 5.5
Transport, Storage and Communication 169 8,600 7.5
Financial Services 97 2,600 9.8
Business Services 420 26,500 4.7
Public Administration 83 2,700 10.6
Education 132 4,400 8.4
Health and Social Work 272 10,700 5.9
Arts and Other Services 209 11,400 6.7

Explaining variation between sub-groups in the report

Table 8 highlights the number of interviews achieved across the entire sample. However, a number of measures captured in the survey are based only on a subsection of the survey. For instances, questions regarding the provision of external training are only asked of those establishments who made use of external training providers to train their staff (2,167 unweighted establishments). Table 9 presents common base sizes across the survey, split by size band. Lower numbers of interviews achieved serve to increase the maximum sampling error as shown in Table 8.

Table 9: Interviews achieved across key measures and by size band

Base measure Total 2 to 4 5 to 9 10 to 24 25 to 49 50 to 99 100 to 249 250+
All employers 2,652 643 608 619 364 161 172 85
All employers with vacancies 1,729 153 351 490 331 152 167 85
All employers offering any training 2,167 340 505 566 352 155 165 84
All employers offering external training 1,635 209 334 427 301 137 146 81
All employers with apprentices 467 25 53 114 109 45 69 52

Using the data at a regional level

Table 10 shows the number of interviews achieved across each region, and the associated sampling error.

Table 10: Sampling error (at a 95% confidence level) associated with findings of 50%, by region

Subgroup Number of interviews achieved Population (Maximum) sampling error (+/-)
Overall
Scotland 2,652 151,200 1.9
Aberdeen and Aberdeenshire 313 17,200 5.5
Ayrshire 167 9,200 7.5
Borders 103 4,300 9.5
Dumfries and Galloway 111 5,900 9.2
Edinburgh and Lothians 334 19,800 5.3
Fife 150 8,700 7.9
Forth Valley 137 7,500 8.3
Glasgow 275 21,000 5.9
Highlands and Islands 432 19,300 4.7
Lanarkshire 245 16,800 6.2
Tayside 233 12,100 6.4
West 143 9,400 8.1
West Lothian 61 4,100 12.5

Contact

Email: peter.phillips@gov.scot

Back to top