Digital Skills In Northern Ireland – Supplementary Document
Digital Skills In Northern Ireland – Supplementary Document
The Digital Skills in Northern Ireland report examines the digital skill level of people in Northern Ireland, sourced from the Continuous Household Survey (CHS).
The Further Information section (section 4) of the report contains an overview of key topics, and this document aims to provide additional background information, including further detail on the CHS (primarily sourced from the CHS survey documents) and the methodology.
1. The Continuous Household Survey (CHS)
1.1 Background
The Continuous Household Survey (CHS) is the longest-running continuous social survey carried out in Northern Ireland, conducted by the Central Survey Unit (CSU) of the Northern Ireland Statistics and Research Agency (NISRA). The dataset used for this publication is from April 2021 to March 2022.
The CHS is designed to provide an ongoing snapshot of social and economic conditions in Northern Ireland. It is carried out continuously to ensure that the information is always up to date and reflects everyday life and conditions in Northern Ireland today. The survey asks questions on a variety of topics, such as internet access, the environment, tourism, libraries, health, sport and education.
1.2 The questionnaire
The questionnaire has two distinct parts; a household schedule and an individual schedule. One person on behalf of the entire household answers the household schedule, while all participating adults in the household answer the individual schedule.
In a change to previous years, there is only one version of the questionnaire in the 2021/22 survey year, similar to the 2020/21 survey year. Therefore, all respondents are exposed to the same set of questions.
1.3 The fieldwork and sample size
Fieldwork began on 1 April 2021 and ended on the 31 March 2022. Addresses are split across the 12 months, with approximately a 10% reduction in allocations in July, August and December to allow for reduced interviewer availability in these months.
In 2021/22, 3,533 households took part in the survey. When ineligible addresses are discounted from the sampling frame this gives a survey response rate of 41%. Within those 3,533 households, a total of 4,103 individuals took part in the survey (1.2 individuals per household).
Table 1: Sample sizes from the Continuous Household Survey 2021/22 (by age and gender), aged 16 and over, April 2021 to March 2022
| Characteristic | n |
|---|---|
| Age | |
| 16 to 24 | 181 |
| 25 to 34 | 462 |
| 35 to 49 | 963 |
| 50 to 64 | 1,201 |
| 65 and over | 1,296 |
| Gender | |
| Males | 1,721 |
| Females | 2,382 |
| Base = 100% | 4,103 |
1.4 Impact of COVID-19
NISRA suspended all face-to-face households interviews in March 2020 due to the COVID-19 pandemic. In the 21/22 reporting period, all CHS interviews were conducted by telephone. With the lifting of some restrictions, from the month of July 2021, CSU implemented the knock to nudge method. This meant that Interviewers could once again call at sampled address to encourage people to participate in the survey while adhering to the COVID health and safety advice/restrictions (social distancing/PPE). Interviewers were not permitted to enter the property, only to collect contact information to complete the survey by telephone at an agreed time. Further detail on the impact of COVID-19 on CHS data collection is available within the CHS 21/22 Technical Report.
The CHS Things Users Need to Know report advises that there are a number of factors to consider when interpreting the results. While survey methodology changed, the impact of the Coronavirus (COVID-19) pandemic and the resultant introduction of new public health regulations, guidance and advice may have also fundamentally changed peoples’ behaviour and attitudes. It is difficult to separate out change caused by the methodological adjustments and actual behavioural change at this point in time.
- The change in data collection mode from face-to-face to telephone may have altered how people responded to the survey.
- The change in data collection mode also necessitated some streamlining of the questionnaire and changes to how some questions were asked or presented as well as the response categories associated with them. This may also have implications for how people responded to the survey.
- The achieved CHS 21/22 response rate in telephone mode was 41% and this is a slightly lower response compared to the normal achieved response rate of 55% in face-to-face mode. This has reduced the number of cases at the household and individual levels. The precision of the survey estimates in the 2021-22 year is also reduced compared to previous findings.
- The demographic profile of the achieved sample has changed in comparison with previous years in terms. There is more of an under-representation of people aged 16 to 44 compared to previous years.
2. Methodology and Quality Information
2.1 Representativeness of the sample
In any survey there is a possibility of non-response bias. Non-response bias arises if the characteristics of non-respondents differ from those of respondents in such a way that they are reflected in the responses given in the survey.
Accurate estimates of non-response bias can be obtained by comparing characteristics of the achieved sample with the distribution of the same characteristics in the population at the time of sampling. Such comparisons are usually made to the current Census of Population data.
To assess how accurately the Continuous Household Survey sample reflects the population of Northern Ireland the sample has been compared with characteristics of the Northern Ireland population from Mid-Year Population Estimates.
Table 2: Mid-Year population estimates and the Continuous Household Survey respondents
| Characteristic | Mid-Year Population estimates 2020 (%) | CHS 2021/22 (%) |
|---|---|---|
| Age | ||
| 16 to 24 | 13% | 4% |
| 25 to 34 | 16% | 11% |
| 35 to 44 | 16% | 15% |
| 45 to 54 | 17% | 17% |
| 55 to 64 | 16% | 20% |
| 65 and over | 21% | 32% |
| Gender | ||
| Male | 49% | 42% |
| Female | 51% | 58% |
| Base = 100% | 1,499,695 | 4,103 |
2.2 Weighting
In order to estimate the number of people at each digital skill level, it is necessary to gross the findings from the survey to the NI population. The respondent data is weighted to the NI population by age group and sex. This ensures that the sample is representative of the overall NI population.
The Continuous Household Survey had an under-representation of people aged 16-44 and males in 21/22. To account for this the survey is weighted against the Mid-Year Population estimates.
The sample is split into seven age groupings for both sexes resulting in 14 weighting cells. The age groups used for weighting were 16 to 24, 25 to 34, 35 to 44, 45 to 54, 55 to 64, 65 to 74, and 75 and over.
2.3 Sampling Error
No sample is likely to reflect precisely the characteristics of the population it is drawn from because of both sampling and non-sampling errors. An estimate of the amount of error due to the sampling process can be calculated. For a simple random sample design, in which every member of the sampled population has an equal and independent chance of inclusion in the sample, the sampling error of any percentage, p, can be calculated by the formula:
\[ s.e. (p) = √(p*(100 – p)/n \] where n is the number of respondents on which the percentage is based.
The sample for the Continuous Household Survey is drawn as a random sample, and thus this formula can be used to calculate the sampling error of any percentage estimate from the survey. A confidence interval for the population percentage can be calculated by the formula:
95% confidence interval \[
p+/- 1.96 * s.e. (p)
\] If 100 similar, independent samples were chosen from the same population, 95 of them would be expected to yield an estimate for the percentage, p, within this confidence interval. The absence of design effects in the survey, and therefore of the need to calculate complex standard errors, means that standard statistical tests of significance (which assume random sampling) can be applied directly to the data.
Table 1 shows the sampling variability of digital skill level of NI respondents using 95% confidence intervals. We would expect that in 95% of samples the range would contain the true value.
Table 3: Sampling variability of digital skill level in Northern Ireland, aged 16 and over, April 2021 to March 2022
| Digital skill level | Percentage | Percentage lower limit | Percentage upper limit |
|---|---|---|---|
| No skills | 17.6 | 16.4 | 18.8 |
| Basic skills | 57.3 | 55.8 | 58.9 |
| Above basic skills | 25.1 | 23.7 | 26.4 |
3. Contact
If you require further information or would like to provide feedback please contact:
Holly McAteer
Email: Holly.McAteer@nisra.gov.uk
Web: Digital skills in Northern Ireland