Last updated: 09-2020
Definition and Sources
The definition of skills varies widely among different sources and disciplines. Standard economics, for example, treats “skills” as the set of knowledge and other worker qualities that influence one’s labour productivity (Frazis & Loewenstein, 2007). In psychology — especially industrial, organizational and educational psychology — the term “skill” is far more specific. It typically refers to the application of some innate “ability” to the execution of a given task (Nickols, 2011; Krathwohl, 2002). In this sense, skills are regarded as distinct from the related concepts of “knowledge,” “ability,” and “competencies” (see Box 1).
In 2019, Employment and Social Development Canada (ESDC) published their Skills and Competencies Taxonomy to facilitate a pan-Canadian dialogue on skills. The taxonomy is based on multiple internal products such as the Career Handbook and the Essential Skills profiles, as well as international occupational taxonomies such as the US Occupational Information Network (O*NET). ESDC’s taxonomy thus fills a long-standing gap by officially defining skills in Canada as the “developed capacities that an individual must have to be effective in a job, role, function, task, or duty.” This definition is consistent with the way skills are conceptualized in the field of psychology.
Box 1. It's an ability, it's a knowledge! No, it's a skill, tool or technology!
Generally, an “ability” is a capacity or attribute required to engage in a task. It can be thought of as the potential for carrying out an activity. For example, to communicate orally with others, an individual must possess the “ability” to access and use the part of the brain needed for speech and language processing. That is, they must have the potential for communication.
“Knowledge” pertains to information. It can be factual, such as knowledge of terminology. It can be conceptual, such as knowledge of theories, models and principles. It can be procedural, such as the knowledge of techniques and methods. Or, it can be metacognitive, such as self-knowledge.
“Skills” are at the nexus of abilities and knowledge. They involve the exercise of one’s ability and knowledge to carry out a specific task (click here to watch a short visual explanation). Consider the skill of “critical thinking.” To think critically, one must possess the cognitive capacity to engage in thought (the ability to think), as well as the necessary background information about the subject (factual knowledge). Then, applying one’s knowledge of the topic, as well as drawing on a variety of other information (e.g. cultural knowledge), the individual can proceed to engage in the task to be completed, such as analyzing which patients should receive care first or evaluating the best tool needed to complete a particular job.
It is also important to note that skills are distinct from “tools and technologies.” It is common in job postings, for example, to see SAS or Python listed as skills, but technically speaking, these should be labelled as tools and/or technologies. For instance, an individual could analyze data by building a statistical model (skill) using SAS (tool) or create a natural language processing algorithm (skill) in Python (tool). This difference is important, especially in the context of skill transferability, in which the skill one has developed is similar across different occupations but maybe the tools are different. In this case, a candidate who may lack knowledge of a specific tool, say a statistical modeller who uses SAS versus one who uses Python, can very easily acquire the knowledge to succeed in the other.
This combination of skills, knowledge and abilities are known as competencies. They often include a specified level of performance at which a skill should be executed and are commonly used in the discipline of Human Resources (HR) and in creating occupational standards, such as with the Red Seal program.
Skills are also described, and further clarified, in relationship to one another through various skill classifications, taxonomies and ontologies (see Box 2), which again vary across sources (see Table 1).
One common classification approach divides skills into two groups: those that involve conscious intellectual effort (i.e., skills that involve thinking, reasoning or remembering) versus those related to personality traits, attitudes, motivations and behaviours.1 Even within this approach to classification, differences in category labels and individual skill placement exist. For example, skills related to personality traits, attitudes, motivations and behaviours go by a variety of labels, such as “non-cognitive,” “social,” “socio-emotional” or “soft,” to name a few (see Box 3). Moreover, a skill labelled “cognitive” in one classification may be categorized as “non-cognitive” in another, as has been the case with the skill of “critical thinking.”
Dividing skills into cognitive and non-cognitive categories is only one approach to classification; additional classifications exist. The US O*NET system, for example, provides a skills taxonomy, which at the highest level, classifies skills as either basic or cross-functional. The European Commission’s European Skills/Competences, Qualifications and Occupations (ESCO) taxonomy, on the other hand, divides skills into cognitive or practical (see Table 1). And, while binary classifications are the most common, other disciplines and fields may use ternary classifications (e.g., cognitive, affective, psychomotor).
Box 2. Classifications and taxonomies and ontologies, oh my!
Classifications, taxonomies and ontologies are three methods for describing and visualizing the relationship between items. A classification is a simple grouping of items into categories based on one or more attributes. For example, one classification categorizes skills into two categories labelled “hard” and “soft” (see Box 3).
Taxonomies show the relationship between many items as a hierarchical, tree-like structure. It can be thought of as multiple classifications connected to each other to form a hierarchy. For example, the US O*NET skill classification is in fact a taxonomy. While skills are separated into two categories at the broadest level — basic and cross-functional — skills are further grouped into several additional classifications.
Like taxonomies, ontologies also show the relationship between many items but in a non-hierarchical manner. That is, within an ontology, many relationships and cross-relationships exist and are modelled. Whereas taxonomies are graphically depicted as tree-like diagrams, ontologies tend to be depicted as webs. One example of a skills ontology is Encube, which models skills in relation to one another.
Table 1. Common top-level skill classifications
|Skill Classification Broad Categories||Example||Definitions|
|Cognitive v. Non-cognitive||Zhou (2016); Pierre et al. (2014)||Cognitive skills involve conscious intellectual effort. They refer to those skills used in understanding complex ideas and engaging in various forms of reasoning.
Non-cognitive skills are related to personality traits, attitudes and motivations, as well as behaviours that are socially determined, such as integrity and interpersonal interaction.
|Basic v. Cross-functional||US O*NET system||Basic skills facilitate learning or the acquisition of knowledge. Cross-functional skills facilitate performance of activities that occur across jobs.|
|Cognitive v. Practical||European Commission ESCO||Cognitive skills involve the use of logical, intuitive and creative thinking.
Practical skills involve manual dexterity and the use of methods, materials, tools and instruments.
Box 3. Moving away from hard and soft skills
In English, there is a long standing use of the labels “hard” and “soft” to refer to skills related to the performance of specific occupational tasks and behavioural skills related to one’s personality and interaction with others, respectively. Although the terms “hard” and “soft” are used both in common language and in the literature (see, for example, Heckman & Kautz, 2012), they are not used in Canadian policy. In addition, there has been growing social stigma around these terms; therefore, the use of more neutral language is encouraged. The terms “occupational” and “social-emotional” are preferred by ESDC.
Given the challenge of defining skills conceptually and the wide range of definitions and classifications, it is not surprising that measuring skills can be complicated. Nevertheless, skill measurement is a critical issue among policy makers, educators and labour market professionals.
Methods for assessing and measuring skills can be divided into two groups: testing and proxy/self-reporting. Different options offer trade-offs in both quality and feasibility.
The most direct approach to assessing an individual’s skills is through skills testing. Occupational skills tests are immersive (i.e. hands-on) job simulations that asses a test taker’s ability to perform the tasks of a certain job. Applicants can be asked, for example, to complete a portion of the work that they will be doing on the job. Depending on the occupation and position, these assessments can be completed in-person with a trained evaluator or online via a written test. The most well-known example of direct skills testing is the coding challenges frequently used during the recruitment of programmers and software engineers, but specialized HR companies are emerging to help employers test the skills of their applicants across a wide range of occupations. The main advantage of skills tests is that they allow for the assessment and evaluation of occupation-specific skills — that is, they test skills directly tied to the specific tasks a worker is expected to carry out.
In addition to occupational skills tests, other types of assessments, collectively known as psychometric tests, are used to assess skills, and more generally, determine job preparedness. Psychometric tests, however, do not test occupation-specific skills. Instead, they are designed to capture general aptitude and behavioural characteristics, and can include personality profiling tests; logical, numerical and verbal reasoning assessments; and situational judgement tests. Psychometric tests are sometimes used by employers to predict whether a candidate is able to perform a job or possesses the behavioural traits needed to be successful in a particular role (Tandon, 2019; Sharma, 2018).
Being tied to occupational competencies, skills tests are the most direct method for assessing and measuring individual skills. Psychometric tests, on the other hand, which assess broad level traits, abilities and/or knowledge provide a more indirect assessment. For example, personality tests, which frequently assess the Big Five personality traits (i.e., openness to experience, conscientiousness, extraversion, agreeableness and neuroticism), can help employers infer how a candidate might fit into a particular role within their organization, as well as to judge how a candidate might react or behave. Similarly, aptitude tests, which measure cognitive abilities such as numeracy and literacy (and thought of as foundational skills), can provide evidence of a candidate’s potential but are not tied to specific occupational tasks. Table 2 summarizes the different testing options and provides advantages and limitations for each.
Table 2. Types of tests
|Type of Assessment||Examples||Advantages||Limitations|
|Occupational skills tests||Programming test for a software engineer
Soufflé-making test for a chef
|Identifies individual skills in the context of specific work-related tasks||
|Psychometric tests: Personality profiling tests||Caliper Profile||Assesses traits such as curiosity, self-discipline and assertiveness; can be used to gauge how a worker “fits” into the organization and the position||
|Psychometric tests: Aptitude/cognitive tests||PIAAC||Measures general skills such as verbal aptitude, literacy and numeracy; allows for objective cross-comparison||
Proxy and/or self-reporting
An alternative approach to measuring skills via testing is to use proxies and/or self-reporting, data for which is collected through surveys of individuals and employers. Common proxies include occupation, educational attainment or qualification, and field of study. Unlike skills or psychometric testing, proxies do not identify or measure skills. Rather, they provide an indication of the general skill level of the population.
Alternatively, individuals can be asked to self-identify their skills and skill-level (i.e., self-reporting). While having the advantage of being able to identify individual skills, there are legitimate concerns about the accuracy and credibility of this information. For example, studies suggest that individuals struggle to reliably rate abstract occupational characteristics, such as those relating to skills, and tend toward overestimation.
- Proxying skills with occupationsAn occupation is a collection of jobs with similar work requirements. In many countries, national statistical agencies use standardized systems to name and code occupations. The standardized system in Canada is the National Occupational Classification (NOC), maintained by ESDC and Statistics Canada. NOC codes are categorized according to the type of work performed (called skill type) and the education or other criteria required (called skill level). If it is assumed that all workers within an occupation have the same sets of skills, then the share of employed workers in each occupation can provide an indication of the skills in the workforce.Proxying skills by occupation is commonly used in the estimation of skills shortages (i.e. a lack of job applicants with the required skills for the job). Labour market signals, such as wages, vacancy rates, hours worked and/or employment growth, can provide evidence of changes in the supply of and demand for a given occupation. If, for example, it is determined that there is a shortage of, say, registered nurses, it is concluded that there is a shortage of the skills required for this occupation.This approach is limited, however, in that the skill requirements of an occupation are never fixed, and the task content of occupations changes over time in response to technological and organizational change, as well as the demands of customers and the evolution of the supply of labour.
- Proxying skills with education/qualificationsEducational attainment, including professional qualifications and credentials, is another indicator that is used as a skills measure. As with occupations, if it is assumed that all individuals with, say, a bachelor’s degree, have the same set of skills, then the share of individuals with a certain level of education can be used as a measure for skills. Related to educational attainment is the duration of education. The assumption in using duration as a measure of skill is that there is a positive relation between an individual’s time spent in education and their skill levels. In Canada, for example, which frequently ranks among the top three OECD countries in terms of population with tertiary education, 55.9% of women and 54.5% of men from the working-age population hold a post-secondary credential.Education is most used to proxy skills when estimating the incidence of skills mismatches (i.e., when workers possess greater or fewer skills compared to the specific skills requirements of their current jobs). Since skills themselves are not directly measured, researchers estimate skills mismatches by measuring the extent to which workers have higher or lower qualifications relative to the requirement for their jobs.Education has been shown, however, to map poorly to the skills one uses on the job. Moreover, there can be high skill variability among individuals with the same educational attainment.
- Proxying skills with field of studyField of study (i.e., subject or content area studied in school) can provide more information on skills than educational attainment alone. Rather than assuming all bachelor’s degree holders, for example, have the same set of skills, one assumes that all Chemistry majors have the same set of skills. The share of individuals in the workforce with certification related to a particular field of study can then be used as an estimate of the skill supply.As with education/qualifications, field of study is most used to estimate skills mismatches. In this case, researchers measure the extent to which workers, trained in one field, work in another (e.g. an English major working as a statistician).
- Self-reportingIt is also possible to collect information on skills by having individuals or employers self-report the skills they possess or the skills needed in their day-to-day jobs. Self-assessment approaches can be advantageous, because they can cover a wide range of skills, are relatively easy to distribute and can provide results more quickly than testing. However, the reliability and validity of the way people self-assess their skills is a concern. This data can also be collected from online sources, particularly job posting websites where job seekers and employers post the skills they possess and the skills they are looking for in applicants, respectively.
Sources of skills data by assessment type: testing
Skills data collected through psychometric testing is available from the Programme for the International Assessment of Adult Competencies (PIAAC), which conducts an assessment and analysis through its Survey of Adult Skills. The survey is conducted every ten years in over 40 countries, including Canada, and measures the cognitive and workplace skills needed for individuals to participate. Five thousand individuals between the ages of 16 and 65 participate in each country. Respondents complete two surveys. First, the background questionnaire collects information on a range of factors that influence the development and maintenance of skills (e.g., education, social background, language). Included are questions to ascertain the generic skills that respondents use in the workplace and their social and emotional skills. Second, the direct assessment evaluates skills in three domains: numeracy, literacy and problem solving with technology.
Sources of skills data by assessment type: proxying
Several surveys are produced and distributed at the pan-Canadian level that capture proxied skills information. Data is collected on both individuals (i.e., the Labour Force Survey, Census, the National Graduates Survey, and the Survey of Labour and Income Dynamics) and employers (e.g., Job Vacancy Wage Survey), as well as through administrative data (i.e., the Longitudinal Immigration Database and the Education and Labour Market Longitudinal Platform).
The Labour Force Survey is a monthly household survey with a sample size of approximately 56,000 households. It provides estimates on the labour force status, wages and demographic characteristics of the civilian non-institutional population 15 years of age and older. The LFS collects data on educational attainment and occupation. Educational attainment is captured according to the classification of highest educational attainment at four levels: less than secondary school graduation; secondary school diploma or equivalent; some postsecondary education; and postsecondary certificate, diploma or degree. At a more aggregate level, data can be collected according to the binary classification of highest certificate, diploma or degree.
The Census of Population (Census) is a mandatory survey of all households in Canada conducted every five years. The Census collects data on educational attainment, occupation and the field of study of the highest certificate, diploma or degree, among other socio-economic characteristics.
The National Graduates Survey (NGS) collects data on the experiences and employment outcomes of graduates of postsecondary programs. It captures information on educational attainment and fields of study, as well as employment and unemployment.
The Survey of Labour and Income Dynamics (SLID) contains questions on labour market participation over time. It follows a panel of individuals over a six-year period, collecting detailed information on occupation and educational attainment, among other socio-economic characteristics. SLID was replaced by the Canadian Income Survey (CIS) in 2012.
The Job Vacancy and Wage Survey (JVWS) is Canada’s primary data source for measuring job vacancy levels and rates by detailed (4-digit) NOC. In addition to data on occupations, the JVWS also collects data on education. Unlike the LFS and Census, however, which collect the highest educational attainment of an individual, the JVWS captures the minimum level of education required for the job.
The Longitudinal Immigration Database (IMDB) provides data on the socio-economic outcomes of immigrant tax filers through the combination of administrative immigration and tax data files. The IMDB contains information on educational attainment and occupation, among other socio-economic characteristics.
The Education and Labour Market Longitudinal Platform (ELMLP) tracks the earnings of graduates by educational attainment and field of study by linking together three core administrative data sets: the Post-secondary Student Information System (PSIS), the Registered Apprenticeship Information System (RAIS) and the T1 Family tax records (T1FF).
Sources of skills data by assessment type: self-reporting
Web scraping is a process by which information is gathered from public websites for retrieval and analysis. One application of web scraping is collecting data — specifically skills and other work requirements information — from online job boards and corporate websites. Currently, several data analytics firms (e.g., Vicinity Jobs, Burning Glass Technologies and TalentNeuron) do this type of work in Canada. While the results of such data collection and analysis are not usually available to the public, the Labour Market Information Council (LMIC) has partnered with Vicinity Jobs to create the Canadian Job Trends Dashboard. This free public tool allows users to explore the skills and other work requirements from jobs across Canada, including detailed information on how the information should be used, what information is available, caveats and limitations, and methodology.
The data discussed in the section above can be accessed through the different channels presented below.
|Data Tables||Customized Products||Public Use Microdata File (PUMF)||Real Time Remote Access (RTRA)||Research Data Centre (RDC)||Canadian Centre for Data Development and Economic Research (CDER)|
|The Labour Force Survey (LFS)||Several tabulations available for educational attainment and occupation (the 1- and 2-digit NOC levels only)||Several tabulations available in accordance with confidentiality release criteria||Available||Available||Available||NA|
|Census||Several tabulations available for educational attainment, occupation (4-digit NOC) and field of study||Several tabulations available in accordance with confidentiality release criteria||Available||NA||Available||NA|
|National Graduates Survey (NGS)||Several tabulations available for level and field of study||Several tabulations available in accordance with confidentiality||Available||Available||Available||NA|
|Survey of Labour and Income Dynamics (SLID)||NA||Several tabulations available in accordance with confidentiality release criteria||Available||Available||Available||NA|
|Job Vacancy and Wage Survey (JVWS)||Several tabulations available for minimum level of education and occupation||Several tabulations available in accordance with confidentiality release criteria||NA||NA||NA||Available|
|Longitudinal Immigration Database (IMDB)||Several tabulations available||Several tabulations available in accordance with confidentiality release criteria||NA||NA||Available||NA|
|Education and Labour Market Longitudinal Platform (ELMLP)||NA||Several tabulations available in accordance with confidentiality release criteria||NA||NA||Available||NA|
|PIAAC||PIAAC in Canada public use files|
|Web Scraping||LMIC’s Canadian Job Trends Dashboard|
Information on the skills of individuals and the skills needs of employers is vital for the creation and maintenance of a globally competitive workforce and contributes to a nation’s long-term economic well-being and growth since skills improvements are linked to gains in GDP per capita. Skills data, therefore, are used primarily to inform the skills training and development of a population. This includes a variety of efforts relevant to a diverse group of stakeholders, from career practitioners and educators to job seekers and policy makers.
At the federal and provincial levels in Canada, various funding programs exist to address skills training and development (e.g., Youth Employment and Skills Strategy Program, Newfoundland and Labrador’s Skills Development Program, British Columbia’s Employer Training Grant). These programs and the amount of funding involved are based on the estimates and predictions of current and future skills shortages. Recently, the Government of Canada established the Future Skills Centre with the mandate of informing and supporting local approaches to skills training and development to ensure that every Canadian has life-long access to high-quality career advice and learning opportunities.
Information on the skills needed by employers is also used to help career professionals and job seekers make decisions about training and education opportunities. When making career decisions, whether related to an individual’s first career or a career change, job seekers and incumbents need to know which skills are in demand, as well as training opportunities and costs. Occupational profiles — detailed descriptions that include, among other information, skills requirements — are particularly valuable. While a variety of resources currently exists, such as Job Bank and ESDC’s Career Handbook, ESDC and Statistics Canada are currently working to improve the availability and granularity of occupational profiles through various approaches that will link ESDC’s new Skills and Competencies Taxonomy to the Canadian National Occupation Classification (NOC). One example is a new concordance table that will link Canadian occupations to the US O*NET system in order to leverage the skills data therein.
Finally, skills data are important to skills trainers and developers. To prepare workers to succeed in the workforce, training programs need to be relevant, providing job seekers with skills that will make them competitive and that meet the needs of the job market. This requires that programs be developed with the most current and granular skills data possible.
1. Skills are learnable. Defining non-cognitive/social skills in this manner, therefore, raises questions as to if and which personality traits, attitudes, motivations and behaviours can be changed, learned or developed. Within the psychology literature, there seem to be mixed views on this subject. For example, twin studies have been used by some researchers to conclude that personality (assessed according to the Big 5) is stable from childhood through adulthood. Other studies, however, provide evidence that stability varies depending on the personality trait, and others still conclude that not only can personality change, these changes can affect economic variables as well. This discussion, however, is beyond the scope of this entry.
2. Although LinkedIn refers to this set of work requirements as “technical skills,” it is more accurate to label these as “tools.”