Title: Social networks amongst older people in Europe: a quantitative comparative analysis using the 2001 I
1Social networks amongst older people in Europe a
quantitative comparative analysisusing the 2001
International Social Survey Programme (ISSP)
- Presentation to the European Sociological
Association, Conference, Sept. 2007 - Phil Haynes
- Reader in Social and Public Policy
- HSPRC, University of Brighton
- p.haynes_at_brighton.ac.uk
- With Professor Michael Hill, and Laura Banks
- This work is supported by UK ESRC research grant
RES-000-22-2114
2International Social Survey (ISSP)
- 2001 ISSP focus on social networks
- Our project uses a sub sample
- aged gt50
- 18 OECD countries, 13 in Europe
3European countries/regions
4Background
- Importance of social networks to quality of life
in older age (Auslander Litwin,1990) - Social networks linked with social capital
(Putnam 1995) - Some suggestion that stronger social network
patterns are associated with better health
(Zunzunegui, 2004 Blumstein, Guralnik
Walter-Ginzbury, 2004) - Important differences within Europe, stronger
family networks in southern Europe (Millar and
Warman, 1996 Pickard, 2003)
5Research Questions
- What is the relationship between family contact
and other forms of social networking? - Are gender differences in the social networking
of older people similar in all European
countries? - Within the countries of Europe, is there any
relationship between attitudes towards family
based care and the social networking patterns,
and social policy expenditure, in those countries?
6Social Networking (ISSP)
- Participation in community organisations
- Quantity of close friendships
- Contacts with close family members other than
immediate partners - Brothers and sisters
- Adult children
7Community Participation
- Seven ordinal scores from questionnaire
- Measuring participation in unions, politics,
sport, religion, community, and other local
groups - Strong single factor when combined using
Principal Components Analysis (PCA) - Separate measures developed for work and non work
based participation
8Community Participation
9Predictors of higher community participation
(European sample)
NB Age is not a significant predictor
10Region - explaining membership of High Community
Participation Score Group
11Community Participation and Gender
- Across the European sample men have higher
community participation scores than women
(t3.253 df6946 p0.001), but this varies
between countries. - In a few countries, womens average score is
slightly higher than mens, but this not
substantive or statistically significant (GB,
France, Finland and Spain) - Regional differences for women in West and East
(next slide) - The only individual country sub sample that has a
statistically significant difference between
women and mens scores is the Czech Republic,
with a noticeable lower level of community
participation for women (t 2.963 df458 p
0.003).
12Odds ratios for being a member of higher
community participation group, computed within
regional sub samples
13Close Friends Score
14Close Friends Score and Gender
- In Europe, womens close friend scores are on
average higher than mens. But the difference is
small and not statistically significant. - In general, within country differences, in mean
scores between men and differences are also small
and not statistically significant - In Norway, womens mean scores are significantly
higher than mens (t -4.401. df586 p0.000)
15Odds ratios for being a member of higher close
friends group, computed within regional sub
samples
Being an older age does have some decreased
effect on friendship
16Family Contact Scores
17Family Contact Score and Gender
- Womens family contact mean scores are higher
than men in all countries - Except in Hungary and Denmark, (where the small
opposite differences are not statistically
significant).
18Predictors of Higher Family Contact Scores.
European sample, ages 50 and over, with reference
to Southern Europe as compared with residence in
other regions
19Attitudinal, ordinal scale measures
- Adult children have a duty to look after elderly
parents - Q. provides some interesting country differences
- It is right to develop friendships just to be of
use to you - Q. about reciprocal caring relationships
- doesnt provide much evidence of country
differences (but Spain and Poland are outliers) - The government should provide a decent standard
of living for the old - Q. doesnt provide much evidence of country
differences, respondents tend to support the
statement strongly in all countries (but USA and
Japan have significantly lower scores than
European countries)
20Association between family contact and attitudes
to care
21Conclusion
- Regional patterns are strong in Europe
- Relationship with traditional higher public
spending, N Eur - W Eur., mid table, group policy developments to
blend gov. policy outputs with social network
outputscash for care - Social class has an impact on community
participation across Europe, and on friendships
in Eastern Europe - No evidence that early retirement amongst all
social class groups leads to increased community
participation - Evidence of social divide middle class, more
participation - Gender differences differences between north
and south. - South women reporting more family contact
- North women reporting more closer friendships