1 / 10

ECONOMIC & SOCIAL DATA ARCHIVE WORKSHOP

ECONOMIC & SOCIAL DATA ARCHIVE WORKSHOP Exploring the Potential of International Micro-data Sources British Academy, London, 14 March 2006 COMBINING INDIVIDUAL AND CONTEXTUAL DATA ACROSS TIME AND SPACE PROFESSOR RICHARD ROSE Director, Centre for the Study of Public Policy

metta
Download Presentation

ECONOMIC & SOCIAL DATA ARCHIVE WORKSHOP

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. ECONOMIC & SOCIAL DATA ARCHIVE WORKSHOP Exploring the Potential of International Micro-data Sources British Academy, London, 14 March 2006 COMBINING INDIVIDUAL AND CONTEXTUAL DATA ACROSS TIME AND SPACE PROFESSOR RICHARD ROSE Director, Centre for the Study of Public Policy University of Aberdeen www.abdn.ac.uk/cspp

  2. INTRODUCTION*Differences between units to which data refers and to which generalizations refer 1. Ecological fallacy: Every survey shows that within a country, individuals differ. Therefore, it is an error to focus on national mean and ignore individual variations around the mean. 2. Individualist fallacy: Generalize about individuals regardless context, e.g. ignore difference between relative income quartiles and difference in national mean income in, say, Sweden and Slovakia or ignore differences in party competition when studying party identification in Britain and China.*National surveys must examine individual differences for that is all there is. *Cross-national studies can deal with the effects on individuals of BOTH individual and contextual variations and multi-level hierarchical modelling statistics makes this straightforward. Three examples follow: i. Health: Individual and contextual effects of Soviet experience ii. Cosmopolitan communication: Europeans who use Internet and know English iii. Time as context: The effect of the month by month passage of time on support for the regime, 14 pooled New Russia Barometer surveys, 1992-2005.

  3. 2. LOW LEVEL OF MALE LIFE EXPECTANCY IN RUSSIA Life expectancy for men at birth, 2003 Years Source: The World Health Report 2005, http://www.who.int/whr/2005/annexes-en.pdf, accessed 9.3.06.

  4. 3. Q. If Russians have terrible health, why are some Russians healthier than others? A. Individual not ecological factors largely influence health Dependent variable: self-assessed health on a five point scale. 13-country mean: 3.3; standard deviation: .94. Self-assessed health scale OLS R2: 30.8% MLH R2:32.7% Bars show impact on health of change in the independent variable (unstandardized b coefficient multiplied by range) Self-control .77 .70 .27 .24 .19 .18 .15 CIS country Age Female Income Educ GDP PPP -.15 -.29 -1.32 -1.33 Source: Centre for the Study of Public Policy, New Europe Barometer 2004, for details see www.abdn.ac.uk/cspp.

  5. 4. KNOWLEDGE OF ENGLISH BY COUNTRY % having some knowledge of English *English official national language Source: Percentages based on replies to 2003 European Quality of Life Survey with national results weighted to each country's share of the total population of 28 countries.

  6. 5. INTERNET USE BY COUNTRY, 2003 % using internet Source: Percentages based on replies to 2003 European Quality of Life Survey with national results weighted to each country's share of the total population of 28 countries.

  7. 6. GDP MORE IMPORTANT THAN LIVING IN ENGLISH-SPEAKING COUNTRY FOR COSMOPOLITAN COMMUNICATION Effect on access to the European public space Access scale Bars show impact on communication of change in the independent variable (unstandardized b coefficient multiplied by range) GDP per capita .83 Education English nat’l lang. .58 Income quartile .43 .25 -.24 Manual -1.08 Age Variance accounted for (Pseudo R2): 51.1% Source: Calculated from Multi-level Hierarchical Model analysis .

  8. 7. WHEN RUSSIANS INTERVIEWED MAKES A DIFFERENCE, 1992-2005 Q. Here is a scale for ranking how our system of government works. The top, plus 100, is the best; the bottom, minus 100, the worst. Where on this scale would you put our current system of governing? Support • : Mean Vertical line: One standard deviation around the mean Approve Neutral Disapprove Source: New Russia Barometer surveys, I to XIV, 1992-2005, for details see www.abdn.ac.uk/cspp

  9. 8. DIFFERENT WAYS IN WHICH TIME AFFECTS REGIME SUPPORT Bars show impact on support of change in the independent variable (unstandardized b coefficient multiplied by range) Change in support Impact of freedom 1992 1993 1994 1995 1996a 1996b 1998 2000a 2000b 2001 2003 2004 2005 Impact of Presidential popularity 1992 1993 1994 1995 1996a 1996b 1998 2000a 2000b 2001 2003 2004 2005 Impact of current economy 1992 1993 1994 1995 1996a 1996b 1998 2000a 2000b 2001 2003 2004 2005 Source: From Rose, Mishler and Munro, FROM TRANSFORMATION TO RESIGNED ACCEPTANCE: THE BOTTOM UP VIEW OF RUSSIANS. Cambridge University Press, autumn, 2006.

  10. 9. POLITICAL INERTIA CHIEF INFLUENCE ON REGIME SUPPORT Change in support Impact on Russian support for current regime of: 1992 1993 1994 1995 1996a 1996b 1998 2000a 2000b 2001 2003 2004 2005 * Net effect of all influences in Table 9.3, excluding monthly change. Source: From Rose, Mishler and Munro, FROM TRANSFORMATION TO RESIGNED ACCEPTANCE: THE BOTTOM UP VIEW OF RUSSIANS. Cambridge University Press, autumn, 2006.

More Related