LET US PRAY: RELIGIOUS INTERACTIONS IN LIFE SATISFACTION. Andrew Clark* (Paris School of Economics and IZA) Orsolya Lelkes (European Centre, Vienna)

Transcription

1 LET US PRAY: RELIGIOUS INTERACTIONS IN LIFE SATISFACTION Andrew Clark* (Paris School of Economics and IZA) Orsolya Lelkes (European Centre, Vienna) June 2007 (Preliminary version) Abstract We use recent survey data on individuals in twentyone European countries to consider possible spillover effects of individual religiosity on life satisfaction. We use three different measures of religiosity: denomination, churchgoing and prayer. We find, as is usual, that own religious behaviour is positively correlated with individual life satisfaction, controlling for demographic characteristics and country fixed effects. Average religious behaviour in the region also has a positive impact: people tend to be happier in regions with a higher proportion of religious individuals. Religious activities positively affect both those who are religious and those who are not. In contrast, greater number of atheists (those who say they do not belong to any religious denomination at present) has negative spillover effects for the religious and the non-religious alike. Our findings suggest that Protestants like to be in the dominant religion, while the well-being of Roman Catholics is unaffected in such case. One possible explanation for this difference may be the different institutional structure of these churches. Since Protestants do not have a Rome, the settings of their regional church may have a greater impact on them. JEL Codes: Keywords: Life Satisfaction, Religion, Externalities * Corresponding author: PSE, 48 Boulevard Jourdan, Paris, France. Tel: Andrew.Clark@ens.fr. The Norwegian Social Science Data Services (NSD) is the data archive and distributor of the ESS data. We are grateful to the ATIP programme of the CNRS for financial support. Thanks to Ed Diener for useful discussions.

2 Appendix Table 1. Measures of religiosity in 21 European countries, 2002/2003 % of total N 1) Churchgoer (Attends religious services at ,180 least once a month) 2) Prays (Prays at least once a week) ,896 3) Denomination Roman Catholic ,633 Protestant ,577 Other 6.4 1,870 No Religion ,245 Source: ESS 2002/2003, weighted frequencies Table 2. Churchgoing and prayer by religious denomination, % Churchgoer Prays Roman Catholic Protestant Other No Religion Total Source: ESS 2002/2003 (N= 29,375; Weighted frequencies). Table 3. Prayer and churchgoing Churchgoing No Yes Total No Prayer Yes Total Source: ESS 2002/2003 (N=29,375; Weighted frequencies).

3 Table 4. The distribution of life satisfaction in European countries Freq. Percent Extremely dissatisfied , , , , , , , Extremely satisfied 2, Total 29, Mean 6.8 Median 7.0 Source: ESS 2002/2003, weighted frequencies Table 5. Level and size of regions by country Number of regions Level (Eurostat classification)* Sample size of smallest region Sample size of largest region Austria 9 NUTS Belgium 3 NUTS Czech Republic 8 NUTS Denmark 15 NUTS Finland 5 NUTS France 9 NUTS Germany 13 NUTS Great-Britain 12 NUTS Greece 10 NUTS Hungary 7 NUTS Ireland 8 NUTS Italy 3 NUTS Luxembourg 1 NUTS Netherlands 12 NUTS Norway 7 NUTS Poland 11 NUTS Portugal 4 NUTS Slovenia 11 NUTS Spain 7 NUTS Sweden 8 NUTS Switzerland 6 NUTS Source: ESS 2002/2003 (N=29,375; Unweighted frequencies) * See In some cases, regions were merged, due to their small size. Figure 1. Religious denominations in European countries

4 Proportion of a specific religion PL PT IE IT ES AT LU SI HU BE FR CH CZ DE NL GB SE DKNOGR FI Roman Catholic Other Religion Protestant Source: ESS 2002/2003 (N=29,375; Weighted frequencies). Measure of religiosity: self-reported identification with a particular religion or denomination

5 Figure 2. Average regional religiosity in European countries Churchgoing Average religiosity per region: churchgoing AT BE CH CZ DE DK ES FI FR GBGRHU IE IT LU NL NO PL PT SE SI Prayer Average religiosity per region: prayer AT BE CH CZ DE DK ES FI FR GBGRHU IE IT LU NL NO PL PT SE SI Source: ESS 2002/2003 (N=29,375; weighted means). Whisker-and-box plot: horizontal line in the box: median, lower hinge: 25th percentile, upper hinge: 75th percentile, horizontal lines outside the box: lower or upper adjacent values, dots: outside values

6 Table 6. Life satisfaction and religiosity by country Life satisfaction Religious activity Denominati on % % % Roman % % Other Mean score churchgoer prayer Catholic Protestant religion Austria Belgium Switzerland Czech Republic Germany Denmark Spain Finland France Great Britain Greece Hungary Ireland Italy Luxembourg Netherlands Norway Poland Portugal Sweden Slovenia Total Source: ESS 2002/2003 (N= 29,375; Weighted frequencies).

7 Table 7. Life satisfaction and individual religiosity. Ordered logit regressions (1) (2) (3) Individual religiosity: Roman Catholic 0.181** (0.030) Protestant 0.150** (0.032) Other Religion (0.055) Attends religious service at least once a month 0.213** (0.027) Prays at least once a week 0.154** (0.024) Other personal characteristics: Second Income Quartile Group 0.246** 0.251** 0.248** (0.032) (0.032) (0.032) Third Income Quartile Group 0.342** 0.348** 0.346** (0.032) (0.032) (0.032) Highest Income Quartile Group 0.570** 0.580** 0.575** (0.034) (0.034) (0.034) Inactive 0.109** 0.106** 0.107** (0.028) (0.028) (0.028) Unemployed-active ** ** ** (0.065) (0.066) (0.065) Unemployed-inactive ** ** ** (0.090) (0.090) (0.090) Education: lower secondary (0.038) (0.038) (0.038) Education: upper secondary (0.038) (0.038) (0.038) Education: post secondary, non-tertiary 0.130* 0.143** 0.141** (0.051) (0.051) (0.051) Education: tertiary * 0.082* (0.041) (0.041) (0.041) Health-Fair ** ** ** (0.029) (0.029) (0.029) Health-Bad ** ** ** (0.053) (0.053) (0.053) Health Hampers a Lot ** ** ** (0.057) (0.057) (0.057) Health Hampers a Little ** ** ** (0.031) (0.031) (0.031) Male ** ** ** (0.022) (0.022) (0.022) Age ** ** ** (0.005) (0.005) (0.005) Age-squared/ ** 0.831** 0.835** (0.049) (0.049) (0.049) Married 0.423** 0.424** 0.426** (0.034) (0.034) (0.034) Separated ** ** ** (0.087) (0.087) (0.087) Divorced * (0.049) (0.049) (0.049)

8 Widowed * * * (0.054) (0.054) (0.054) Children living at home or not (0.027) (0.027) (0.027) Country fixed effects Yes Yes Yes Observations Log likelihood at zero Reference categories: no religion, lowest income quartile group, paid work, primary education or below, healthgood, health does not hamper, never married Standard errors in parentheses. * significant at 5%; ** significant at 1%

9 Table 8. Spillover effects of specific religious denominations: Life satisfaction regressions Roman Catholics Protestants % Roman Catholics in Region 0.610** 0.715** (0.220) (0.269) (0.563) (0.472) % Protestants in Region ** 1.705** (0.423) (0.495) (0.322) (0.345) % No Religion in Region ** ** (0.300) (0.343) Observations Other Religion No Religion % Roman Catholics in Region ** 0.979** (0.678) (0.733) (0.212) (0.218) % Protestants in Region ** 1.135** (0.614) (0.766) (0.235) (0.251) % No Religion in Region ** (0.779) (0.212) Observations Ordered logit estimates. The models include personal controls and country fixed effects. Robust standard errors in parentheses. * significant at 5%; ** significant at 1% Table 9. Spillover effects of religious activities: Life satisfaction regressions Not Churchgoer Churchgoer Not Prayer Prayer % Churchgoers in Region 0.989** 0.823** (0.268) (0.316) % Prays in Region 1.275** 1.031* (0.270) (0.415) Observations Ordered logit estimates. The models include personal controls and country fixed effects. Robust standard errors in parentheses. * significant at 5%; ** significant at 1%

10 Table 10. Spillover effects of religious activities by gender Men Not Churchgoer Churchgoer Not Prayer Prayer % Churchgoers in Region 0.886** (0.315) (0.410) % Prays in Region 1.309** (0.308) (0.534) Observations Women Not Churchgoer Churchgoer Not Prayer Prayer % Churchgoers in Region 1.080** 1.317** (0.338) (0.348) % Prays in Region 1.229** 1.434** (0.342) (0.437) Observations Ordered logit estimates. The models include personal controls and country fixed effects. Robust standard errors in parentheses. * significant at 5%; ** significant at 1% Table 11. Spillover effects of belonging to the dominant denomination All Protestants Roman Catholics Belongs to dominant denomination 0.145** (0.024) (0.104) (0.095) Belongs to dominant denomination (incl. atheists) ** (0.023) (0.076) (0.064) Belongs to absolute majority denomination (incl.atheists) ** (0.024) (0.089) (0.075) Observations Note: dominant denomination defined at the regional level. Ordered logit estimates. The models include personal controls and country fixed effects. Standard errors in parentheses. * significant at 5%; ** significant at 1% Table 12. Spillover effects of the dominant denomination Roman Other Protestants Catholics Religion No Religion Dominant Religion in Region is Catholic * (0.066) (0.140) (0.216) (0.066) Dominant Religion in Region is Protestant ** ** (0.134) (0.078) (0.226) (0.070) Observations Note: dominant denomination defined at the regional level. Ordered logit estimates. The models include personal controls and country fixed effects. Standard errors in parentheses. * significant at 5%; ** significant at 1%

11 Table 13. Spillover effects of religion, controlling for social contacts, voluntary work, donations and help to others (2) Social (3) Help to (4) Participated in (5) (6) Voluntary work in (1) contacts others organizations Donation organizations Average religiosity per region: churchgoing 0.979** 1.048** 0.994** 1.014** 1.025** 0.891** (0.232) (0.242) (0.232) (0.229) (0.230) (0.241) Attends religious service at least once a month 0.189** 0.179** 0.179** 0.180** 0.179** 0.179** (0.030) (0.029) (0.030) (0.030) (0.030) (0.030) % in the region who meets with friends or relatives less than once a month (0.612) Meets friends, relatives or colleagues at least once a month 0.451** (0.046) % in the region who does voluntary work (0.346) Voluntary work in the last year 0.107** (0.032) % in the region who donates money to vol org (0.254) Donated money to voluntary organizations in the last year 0.102** (0.028) % in the region who participates in voluntary organizations (0.245) Participated in voluntary organization activities in the last year 0.092** (0.029) % in the region who helps others (0.318) Help other more than once a month (not counting vol work) 0.101** (0.026) Observations Log likelihood