AGENDA Occupation Coding using ISCO-08 1. Occupations and occupational status 2. ISCO-08 Training session for PIAAC – General principles – Comparison with ISCO-88 and ISCO-68 Harry B.G. Ganzeboom – Problems with ISCO-08 Break: Please submit questions! Free University Amsterdam Bologna, January 20 th 2010 3. Questions and Answers 4. Do’s and Dont’s of Occupation Coding Occupations 2 Why are occupations important? • OD Duncan: “the single best indicator of social status”. • EO Wright: “Sociology’s core variable”. Occupations and • RM Hauser: “Occupational status is the better version of the economist’s ‘permanent income’ concept”. Occupational Status • Occupations are important as dependent variables (occupational attainment studies) and independent variables (occupation stratification studies) in educational (and occupational!) status attainment, health, voting, consumption, marriage etc. Occupations 3 Occupations 4 Occupations: what are they? Complicated and multi-faceted • Common descriptions of occupation refer to • Combination of work tasks and duties that is transferable across work establishments. multiple elements like: • Occupation is related but NOT identical to: – Set of required skills and competencies – Job – Responsibility, authority – Firm / work organization – Autonomy – Industry – Status in employment. – Education / Qualification • And respondents tend to talk about quite a – Salary grade but more... – Employment contract (e.g. indefinite-fixed term, self- employed-salaried). Occupations 5 Occupations 6 1
Question format -- open Occupations in PIAAC • Because occupations are complicated, it is often advised to • Resp’s current / last occupation: collect the information in an open format. – Job title • Underlying assumption is that no set of closed questions – Most important responsibilities can sufficiently measure the required details. – Kind of business, industry or service • Questions usually have two elements: – Job title – What does your firm/organization make or do? – Describe major duties and tasks – Sector of the economy (private..public) • This information is recorded verbatim and then post- – Employee / self-employed processed (coded in the office). This implies that measuring occupations is a lot of work and that many – Firm/organization size things can go wrong. – Number of subordinates Occupations 7 Occupations 8 Father’s / Mother’s occupation A first look at ISCO-08 • Hierarchical digit system: • Father’s / mother’s occupation: • Major groups – Job title • Sub-major groups – Kind of work • Minor groups • Unit groups • NOT: • ISEI-08 (as estimate on ISSP 2002-2007) – Self-employment – This is a provisional scale that was generated using a straight conversion from ISCO-88 data. – Supervisory status – Some imputation using ISEI-68 data. Occupations 9 Occupations 10 Occupational status Occupational status measures (1) • Occupations / occupation codes are not directly • Continuous indicators: used in statistical analysis – Prestige (subjective) : popular evaluation of attractiveness of occupations. • They enter statistical analysis as occupational – Socio-economic index [SEI] (objective) : scaling of status measures: occupations by criterion variables in the status – Sociologically meaningful indicators of hierarchical attainment process. position in society. • Discrete indicators: – There exist several (internationally comparative) – Socio-economic classes (apriori) : categories organized occupational status measures, each with different by common work and contract situation. theoretical backgrounds and applications. Occupations 11 Occupations 12 2
Comparative occupational status ISEI – a MIMIC moel measures EDUCATION � OCCUPATION � INCOME • SIOPS: Standardized International Occupation Prestige Scale (Treiman): harmonization of some 60 national prestige scales. • ISEI: International Socio-Economic Index of ISEI is defined as the scaling of detailed occupations that maximizes the indirect effect of education on occupational status (Ganzeboom & Treiman): income and minimizes the direct effect. constructed on large harmonized international data files that contains comparable indicators of occupations, education, earnings. � ISEI measures how occupations transfer education into earnings. Occupations 13 Occupations 14 Socio-economic classes Why socio-economic classes? • EGP (Erikson-Goldthorpe-Portocarero) organizes • Some researchers like discrete measures better occupational positions into 11 discrete categories. than continuous ones… • For EU: ESEC (European Socio-Economic • Some theories see occupational stratification as an Classes) distinguishes 9 discrete categories. intrinsically discrete structure, that is more defined by boundaries than by distance . • Socio-economic classes draw as much on (A) occupations, as on (B) status in employment (self- • There is an enormous amount of evidence that employment, supervising status). intergenerational occupational reproduction / mobility is a process that cannot be sufficiently EGP/ESEC cannot be defined on father and captured by a single parameter (such as an OLS mother for PISAAC data, because status-in- regression coefficient). employment is not collected. Occupations 15 Occupations 16 A problem … The solution … • While there is a new occupational classification • I have created a new (but provisional) ISEI-08, (ISCO-08) available, we do not have (yet) using data from the ISSP 2002-2007 (N=198000, occupational status measures for it. 45 countries) and a straight recode of ISCO-88 � ISCO-08, • Not easy to solve for ISEI: we need data coded in ISCO-08 to construct ISEI and it is not attractive • Also, an ISEC-08 (socio-economic classes) and to use this classification because there no ISEI yet. SIOPS-08 (prestige) will become available soon. • Other measures (SIOPS, EGP/ESEC) can be • Note that in PISAAC ISEC-08 is useful for constructed without ISCO-08 data, but need such respondents, but not for fathers and mothers. data for validation. Occupations 17 Occupations 18 3
ISCO in the past ISCO-08: • The International Standard Classification of Occupation [ISCO] has been revised recently into a fourth edition: 1958, 1968, 1988 and now 2008. • Stratification researchers (and survey researchers in Structure and pitfalls general) have used the 1968 version extensively (e.g. Political Action 1974 study, the Treiman SIOPS scale, various national classifications are derived from it). • ISCO-88 has become the de-facto standard for classifying occupations in international survey projects (ESS, ISSP, SHARE, PISA, IALS etc.) Occupations 20 General principle 1: four Stated goals of ISCO-08 hierarchical digits • Bring occupational classification in line with • ISCO is organized in four nested hierarchical levels: changed technologies and division of labor (e.g. – Major groups 2 2000 ICT/IT). – Sub-major groups 21 2100 – Minor groups 217 2170 • Make ISCO applicable in a wider range of – Unit groups 2174 2174 countries and economies. • Note that I use the convention to add trailing zeroes (ISCO • To mend often noted problems with the itself does not do this). This is particularly important when application of ISCO-88. you allow coding at a variable level of precision. • To produce a minor revision, not a totally • The aim is to group occupations within digit level by different classification. degree of similarity. Occupations 21 Occupations 22 General principle 2: skill levels State of revision • Occupational classifications generally are a mix of • ISCO-08 has been decided upon in December 6 th three dimensions: 2007. – Skill level • What is on the ILO website as of January 2010: – Product, industry – No manual!! – Status in employment (self-employment, supervising). – Draft definitions of all groups • ISCO chooses skill levels as the primary – Many-to-many conversions between ISCO-88 and dimension and seeks to avoid status-in- ISCO-08 (both directions), listing all mergers and employment. splits. • (The argument is that status-in-employent should be separately measured from occupation.) Occupations 23 Occupations 24 4
Recommend
More recommend
