IAOS 2018 - Teaching official statistics in universities: some recommendations from a direct experience, L. Porciani, T. Rondinella
- 1. 16th Conference of IAOS OECD Headquarters, Paris, France, September 21st 2018 Teaching Official Statistics in Universities: some recommendations from a direct experience Linda Porciani and Tommaso Rondinella National Institute of Statistics (Istat) |Italy
- 2. Outline o Background concepts • Statistical literacy • Official Statistical literacy (OSL) • Data Science o OSL at University • Teaching experience o Final remarks and recommendations
- 3. Background concepts Statistical Literacy: “the ability to understand and critically evaluate statistical results that permeate our daily lives - coupled with the ability to appreciate the contribution that statistical thinking can make in public and private, professional and personal decisions” (Wallman, 1993) According to UNECE (2012) is made of: • Numeracy • Ability to read and communicate the meaning of data + production context. • How data can be used in decision making.
- 4. Background concepts Data science: • Research Design and Application for Data and Analysis; • Exploring and Analysing Data; • Storing and Retrieving Data; • Applied Machine Learning; • Data Visualization and Communication (Donoho, 2017)
- 5. Official Statistics Literacy (OSL): Statistical literacy + • analysis of the sources, • meaning of data and indicators, • quality criteria of data production processes, • ownership of official statistics production processes, • importance of metadata, • legislative framework. Often lacking even in specialist tertiary curricula! Background concepts
- 6. Why Official Statistics Literacy is important? “trustworthy statistics sit at the very heart of the democratic contract between the public and its elected government” (Holt, 2008, p.4) Half of Europeans do not trust their countries official statistics Source: Eurobarometer, 2015
- 7. Trust Aspects of trust in OS: • Being trusted to explain; • Trusted to be the best available; • Trusted to be the most relevant. (Alldritt and Wilcock, 2011)
- 8. Teaching experience • European Master on Official Statistics (Emos) • Seminars for 1st year students • Vocational training in public administration
- 9. The European Master in Official Statistics (EMOS) is a joint project of universities and data producers, providing post-graduate education in the area of official statistics at the European level. The network comprises 23 programmes in 14 countries. Objectives: • Ensure that the Official Statistics become part of the academic curricula • Reinforce the collaboration between universities and NSIs • Prepare graduates with specific skills • Improve the official statistics system in Europe EMOS project
- 10. EMOS project Learning outcomes: understand and to be able to analyse European official data at different levels: • Quality, • Production process, • Dissemination, • Analysis in a national, European and international context. Istat’s role: • In the classroom: Lessons and seminars on a variety of topics • At Istat: 8 weeks internships (necessary to complete the master course) • Support to Master’s thesis
- 11. EMOS classroom Students with a fair degree of autonomy and class interaction Class not numerous, (6 to 25 students) Heterogeneity of the academic, social and cultural backgrounds of origin Plurality of teachers The classroom Exchange and comparison Common thread: Recall the way contents relate to OS
- 12. EMOS Internships Microdata access protocols Formal definition of role and functions of intern Work experience Being part of production processes Applied research Limits on access to data Limits in carrying out institutional activities Strength Weaknesses
- 13. SOME GENERAL REMARKS 1. The students’ (and teachers’) OS knowledge gap 2. The emergence of different sentiments about OS 3. The need to have structured and updated materials about OS (i.e. official Handbook) 4. The inclusion of formal OS curricula inside the university courses 5. The need of an evaluation system regarding OS courses (teachers, contents, materials)
- 14. FINAL RECOMMENDATIONS Four basic pillars of OS related knowledge: • definitions, • quality criteria and practices, • sources, and • autonomy in finding official data Be spread beyond statistical authorities’ dissemination. Dedicated section in manuals Teachers’ training