This commentary explores how combining datasets and making use of the Occupational Information Network (O*Net) can help researchers advance understanding of key questions about the effects of digital work.
O*Net provides measurement of the extent of technology skills and digitalisation required within occupations, and can be useful to researchers looking to study how digitalisation affects experiences of work and employment.
This includes the potential to explore: demographic differences about who works in jobs with digital skills; the extent to which job quality is being affected by jobs that are more or less digital now; and how job movement is related to digitalisation.
O*Net can therefore help provide insight into key issues of current concern to policymakers, labour standards agencies, and trade unions on the consequences of digitalisation for workers.
Surveys like Understanding Society and the Labour Force Survey (which are available to researchers via the UK Data Service), contain a wealth of data that can be used to study workers’ experiences of work and employment, including data on wages, working time, occupational group, industry sector, health and subjective wellbeing. The Understanding Society survey also provides data on workers’ perceptions of how much control they have over the way they do their jobs, how much effort they make at work and their job security and job satisfaction.
However, these datasets do not include measures of the extent to which a job involves digital skills and tasks. It is therefore difficult to use these datasets to explore workers’ experiences of digital work, or to understand how digitalisation is impacting experiences of work and employment.
One way to overcome this problem is to combine data from UK Social Surveys with data on tasks and skills at the occupational level from the Occupational Information Network (O*Net). O*Net is a free online database detailing the characteristics of over 900 different occupations, based on analysis of job roles in the USA.
The database has been developed following detailed job analyses, where the composite activities of each job role are broken down into a range of different task-related, person-related and context related characteristics. For each occupation, O*Net lists the tasks and generalised work activities that comprise the role, along with the knowledge, education and training required to fulfil it. Likely work styles and work contexts associated with each job title are also set out.
What is especially useful for scholars of digitalisation and digital futures at work is that O*Net regularly updates the composition of a very wide range of job roles, and includes a detailed assessment of the use of technology within each role, and the technological skills and knowledge required to fulfil such roles. O*Net therefore provides some of the technological and digital detail currently missing from the UK Social Surveys mentioned above, providing an important resource for researchers.
2 What data does O*Net provide?
O*Net is based on two key elements. First, the content model, which provides a framework for analysing occupations based on six categories:
Worker characteristics, including abilities, occupational interests, work values and work styles.
Occupational requirements including general, intermediate and detailed work activities, organizational and work contexts.
Worker requirements in terms of skills, knowledge and occupation.
Experience requirements in terms of length of training and experience in lower-level jobs and whether there are occupational licensing requirements.
Workforce characteristics which are measures of economic conditions and labour force characteristics of occupations along with future projections for economic conditions and labour force characteristics.
Occupation specific information which includes job title along with alternate titles used to describe the occupation, a description of duties performed, occupation specific tasks, technology skills (information systems and software) that are essential for the role, and details of any tools or machinery used.
Secondly, the database is built on the O*Net taxonomy, which was developed in 1998 and has been revised on a number of occasions since. The O*Net taxonomy maps on to the latest version of the United States Standard Occupational Classification (SOC) system. The latest version of the O*Net taxonomy was developed in 2019 to map on to the 2018 SOC.
3 How does O*Net classify and analyse occupations?
To facilitate the process of classifying occupations according to the content model, the team behind O*Net conduct an extensive and ongoing programme of data collection and analysis, details of which can be found on the O*Net website.
In the most recent cycle of data analysis, for example, 80 SOC occupations were studied and classified. Workplaces are sampled, and managers within a workplace are asked to distribute a questionnaire to workers in the occupations under investigation. Questionnaires ask workers about workers characteristics, occupational requirements and occupation specific information. Additionally, expert occupational analysts provide information on skill and experience requirements of different occupations.
This division between the information gathered via workers and experts is based on the principle that workers are best placed to describe what they do and how they do it, while experts will be better able to classify these activities into higher level skills and abilities. For example, abilities are categorised along four dimensions: cognitive, psychomotor, physical and sensory-perceptual.
This means that for each of the occupational titles categorised by O*Net, researchers can generate a report on the extent to which that occupation involves digital work.
So, if we are interested in warehouse workers and the extent of their digital skills, we would look at the O*Net occupation of stockers and order fillers. O*Net reveals that the technology skills involved in this job typically include using scheduling software, databases, standard computer software like MS Windows and Office, inventory management and enterprise resource planning software. However, although the job involves the use of technology skills, the underlying knowledge, skill and ability requirements are limited, without any need for higher level analytical skills or training. Physical abilities in this occupation are identified as more important. By contrast, O*Net explains that web developers need more specific technology skills, for example the ability to use object orientated development software and web platform development software, combined with higher level critical-thinking and complex problem-solving skills.
O*Net also marks out which technology skills are “hot”, i.e. mentioned frequently in job posts for multiple occupations, and “in demand” within that specific occupation. Overall, O*Net provides regularly updated comprehensive descriptions of what workers do, including how and whether they interact with different technologies.
4 How to use O*Net in conjunction with the UK Social Surveys
It is possible to match information of occupational tasks and skills from O*Net with the measures of occupation contained within the Understanding Society and the Labour Force Survey dataset, to identify the extent to which a worker is doing a “digital job”—using digital skills to carry out digital tasks. Dr Rachel Forshaw (pages 4-8) provides an overview of how this matching process can be carried out using Cascot software.
It is important to note that to undertake this matching, researchers will need to apply, via the UK Data Service, for access to data files containing detailed occupational codes within Understanding Society and the Labour Force Survey. They will also need to provide a rationale for why they need access to these.
O*Net is not a perfect solution to the problem of measuring the extent to which a job is “digital”. This is because there are a number of potential sources of measurement error that might arise from transferring O*Net’s taxonomy from the USA to the UK. Specifically, there are differences in the ways in which occupations are classified between the US and UK, which means that UK occupations do not always map neatly onto the O*Net taxonomy.
While O*Net has over 900 occupational categories, reflecting the US SOC2018, the UK Standard Occupational Classification has only around 300 categories. The incorporation of digital tasks within similarly titled occupations will also vary, depending on the technology adoption decisions of the firm or organisation. Further, there may be systematic differences in technology adoption between (and within) firms in the US and UK, which means that aspects of the O*Net taxonomy may apply to occupations in a different way in the UK.
It is also important to note that O*Net is regularly updated, and with each update provides an analysis of the occurrence of digital skills and tasks within occupations at that point in time. Therefore, whilst a particular release of O*Net is useful for comparing how jobs differ in their digitalisation, it does not illuminate changes in digital and technology skills over time. However, all changes, releases and updates in the O*Net database are archived and available including specific guidance on using O*Net to conduct longitudinal analysis and combining information in O*Net with other datasets.
5 Using O*Net to understand how digitalisation affects experiences of work and employment
Despite these limitations, O*Net provides a reasonable measure of the extent of technology skills and digitalisation required within jobs. It has the potential to be useful to researchers looking to study how digitalisation affects experiences of work and employment.
Combining O*Net with UK Social Survey data could also be insightful for a range of issues and questions such as:
describing demographic differences about who works in jobs with digital skills at the time the data were collected (i.e. age, gender, ethnicity, education, previous work experience);
describing differences in job quality between jobs that are more or less digital now;
looking at how movement in and out of digital jobs is associated with different experiences of work over a recent short-run period.
To illustrate how O*Net has been used alongside UK datasets by researchers, Forshaw (2020) matched O*Net data with the UK Labour Force Survey and used the resulting dataset to find that workers who changed jobs into more skilled roles enjoyed greater pay increases than other workers. By matching O*Net with Understanding Society, it would be possible to undertake similar analysis for non-wage aspects of job quality, for example perceptions of autonomy, effort, job security and job satisfaction. Longitudinal data from Understanding Society could also be used to investigate how different social backgrounds result in access to or exclusion from high quality digital work.
Measures of the extent of technology skills and digitalisation required within jobs remain rare in current social surveys. O*Net provides such measures and can help researchers explore the important question of how exactly how digitalisation is affecting experiences of work and employment. The use of O*Net data in this way could certainly help provide insight into key issues of current concern to policymakers, labour standards agencies and trades unions on the consequences of digitalisation for workers.
Professor Andy Charlwood, Chair in Human Resources Management, University of Leeds and Co-Investigator at the Digital Futures at Work Research Centre.
All-Party Parliamentary Group on the Future of Work (2021). The New Frontier: Artificial intelligence at work. Bethlehem, J. (2010). Selection Bias in Web Surveys. International Statistical Review, 78(2): 161-188.
Capital Economics (2022). AI Activity in UK Businesses: An assessment of the scale of AI activity in UK businesses and scenarios for growth over the next twenty years. January 2020.
Department for Digital, Culture, Media & Sport (2022). UK Digital Strategy. Policy Paper updated 4 October 2022.
Ernst and Young (2021). Data foundations and AI adoption in the UK private and third sector. July 2021.
HM Government (2021). National AI Strategy. Command Paper 525.
Podsakoff. P.M., MacKenzie, S.B., Lee, J.-Y. and Podsakoff, N.P. (2003). Common Method Biases in Behavioral Research: A Critical Review of Literature and Recommended Remedies. Journal of Applied Psychology, 88(5): 879-903.