Examining the Relationship between Household Satisfaction and Pollution

Examining the Relationship between Household Satisfaction and Pollution Debra Israel Indiana State University Arik Levinson Georgetown University Paper to be Presented at the Eastern Economics Association Meetings February 23, 2003 Draft Version: Please Do Not Cite Without Author s Permission Abstract: This paper uses survey data to examine the relationship between ambient levels of pollution and peoples stated levels of "happiness" or "satisfaction." We estimate stated happiness as a function of individual characteristics (including household income, marital status, education, employment status), and country characteristics (including national income and pollution). We combine data from the World Values Survey (WVS), with country characteristics from the World Bank World Development Indicators. The primary motivation for this research is methodological. Existing methods for valuing changes in environmental quality include hedonic models, travel-cost models, averting expenditures, and contingent valuation (Freeman, 1993). This project explores a new technique that complements existing approaches and in some cases may be a preferred substitute. In particular, this technique may be useful for making comparisons where comparable data needed for other techniques is either not applicable or not feasible to gather. If subjective well-being measures vary systematically with pollution levels across a broad range of countries, increased availability of these types of data may allow a comparison of the willingness to pay for environmental quality for a broad range of countries of differing income levels. We find evidence that for at least one pollution indicator, per capita water pollution, as pollution increases, subjective well-being decreases. The air pollution indicators of total suspended particulates (TSP), nitrogen dioxide, and sulfur dioxide do not appear to have a statistically significant impact on happiness or satisfaction. This study illustrates that data on subjective well-being, in conjunction with data 1

on income and environmental pollution, may be helpful in gaining a better understanding of people s willingness to tradeoff pollution and income. Introduction This paper uses survey data to examine the relationship between ambient levels of pollution and peoples stated levels of "happiness" or "satisfaction." We estimate stated happiness as a function of individual characteristics and country characteristics. The primary motivation for this research is methodological. Existing methods for valuing changes in environmental quality include hedonic models, travel-cost models, averting expenditures, and contingent valuation (Freeman, 1993). This project explores a new technique that complements existing approaches and in some cases may be a preferred substitute. In particular, this technique may be useful for making comparisons where comparable data needed for other techniques is either not applicable or not feasible to gather. If subjective well-being measures vary systematically with pollution levels across a broad range of countries, increased availability of these types of data may allow a comparison of the willingness to pay for environmental quality for a broad range of countries of differing income levels. Much of the popular debate surrounding trade, global pollutants, and international aid assumes that developing countries have lower demand for environmental quality. However, observed ambient pollution is the consequence of interactions between the supply of and demand for environmental quality. While recent empirical evidence has documented some forms of local pollution (airborne lead, sulfur) that have declined significantly in industrialized countries despite robust local economic growth (see for example Grossman and Krueger, 1995), little research has focused on differences among countries in the demand for environmental quality. One of the main obstacles to this research is the lack of comparable data on the willingness to pay for environmental quality available for cross-country comparisons. This issue has important implications for economic development policies: Should environmental protection begin only after a country achieves a certain level of income, or is it an important priority earlier in the development 2

process? For pollutants that have both domestic and global effects, the choices that developing countries make may be of worldwide concern. While a large literature on the measurement of demand for environmental quality in developed countries exists and in the past decade this research has also expanded to developing countries, direct comparisons of results between developed and developing countries from separate studies is difficult since methods differ from study to study. 1 Previous cross-country comparisons that have included multiple countries, both developing and industrialized, have utilized qualitative responses to questions on willingness to pay for environmental quality (Bloom, 1995; Brechin and Kempton, 1994; Dunlap and Mertig, 1996; Inglehart, 1995; Israel, 1999; Israel and Levinson, 2002). Methodological Approach and Data Interest among economists in utilizing subjective well-being measures for empirical economic research appears to be growing. For example, Di Tella, et al. (2001) examine the marginal rate of substitution between inflation and unemployment using subjective well-being data from OECD countries and the USA. In a 1997 edition of the Economic Journal, a symposium on the controversy of economics and happiness was published (Frank, 1997; Ng, 1997; Oswald, 1997). In a recent edition of the Journal of Economic Literature, Frey and Stutzer (2002) examine the potential relevance of happiness research for economics. They review the work that has been done examining the relationship between stated happiness and income, unemployment, and freedom or democracy. They also suggest a number of additional areas in economics where survey research on happiness would be useful, including examination of quality of life indicators like environmental quality. 1 See Whittington (1998) or Georgiou et al. (1997) for the current status of contingent valuation research in developing countries and Portney (1994) for a discussion of the state of the debate on using contingent valuation versus other methods to measure environmental quality demand. 3

In this paper we utilize subjective well-being data to examine the possibilities of measuring the tradeoffs people are willing to make between pollution and income. To our knowledge, the only other paper that has examined this is Welsch (2002) where he estimates happiness regressions using aggregate averages of stated happiness by country. In this paper, since we have access to the individual responses, we are able to control for other individual and household characteristics that might affect happiness, as well as including the national characteristics of income and pollution. Theoretically, we would like to estimate a utility function where utility is a function of both market goods (represented by income) and environmental quality. Totally differentiating the estimated function, and finding the derivative of household income with respect to environmental quality (holding all else constant including happiness) would measure the marginal willingness to pay for environmental quality. In this paper we examine various measures of pollution that are available for a broad range of countries to see if a systematic relationship appears to exist between pollution and happiness, as a precursor for the type of estimation of the marginal willingness to pay for pollution just described. Our basic strategy is to estimate an ordered probit model with happiness or satisfaction as the dependent variable, including income and pollution and other individual and household characteristics as explanatory variables. Both the happiness and satisfaction questions allow respondents to choose from an ordered set of responses. By estimating an ordered probit rather than a linear regression model, we are not assuming that a change from not at all happy to not very happy has the same relationship with the explanatory variables as a change from somewhat happy to very happy. However, for purposes of comparison, for the satisfaction equation (since the responses are on a continuum from one to ten), a linear regression was also estimated. 4

The ordered probit model is based on the following latent regression where y* is the unobserved subjective well-being and y are the four observed categorical choices reflecting this underlying subjective well-being (note that happiness is measured with four choices, satisfaction with ten choices). 2 To estimate the parameters,, is assumed to be normally distributed with a mean of zero and variance of one. The probabilities of the y values, given the parameters of interest are: As in a probit model, the parameter estimates cannot directly be interpreted as the marginal effects of a change in a variable on the probability. In the ordered probit, for the first and last categories, in the case of happiness, very happy or not at all happy, the sign of the parameter estimate gives an unambiguous indication of the direction of the effect, however, the same is not the case for the middle categories. 2 The following discussion of the ordered probit model is based on Greene (1993). 5

For data we rely on the World Values Survey (WVS), maintained by the Institute for Social Research at the University of Michigan. The WVS is designed to be a representative survey carried out using consistent methodologies across numerous countries. We focus on the most recent available wave of the survey, carried out predominantly from 1995-96, with some countries surveyed in 1997 or 1998. 3 The attached appendix contains the wording of the key happiness and satisfaction questions. We combine these data with national income and environmental variables from the World Bank World Development Indicators. Thirty countries had consistent data on the happiness and satisfaction questions, national income, water pollution, and all of the relevant respondent characteristics. Table 1 contains the means of the relevant household-level variables, while the means of the national variables, both per capita GDP and environmental indicators, are in Table 2. For this study, happiness is coded with 4=very happy, 3=quite happy, 2=not very happy and 1=not at all happy. Satisfaction ranges from 1-10 with 10 the most satisfied and 1 the most dissatisfied. One difficulty in cross-national comparisons which include many countries from different regions and of differing income levels, is that comparable data are not always available to measure pollution. In this paper we show results using measures of total suspended particulates (TSP) in micrograms per cubic meter, nitrogen dioxide in micrograms per cubic meter, sulfur dioxide in micrograms per cubic meter, and emissions of organic water pollutants in grams per day per capita. 4 We focus on the last measure of 3 The first wave of the data was conducted only for European countries. The last wave is not yet available. 4 TSP, sulfur dioxide, and nitrogen dioxide are available by city in the World Development Indicators (2002), the national figure we use is a population-weighted average of the available figures. Organic water pollutants are also taken from the World Development Indicators (2002) and then divided by population and multiplied by 1000 to convert to grams per day per person. 6

water pollution due to its wider availability for countries in our data. Ideally, the pollution indicators would be more closely matched with the geographic area of the household responding, but given the limitations of the data we are implicitly examining the importance of pollution in the respondent s country on satisfaction or happiness. Again, ideally, to examine the trade-off between income and environmental quality, we would prefer to include a direct measure of absolute household income. However, in the WVS, the income categories are relative rankings of income, and differ by country. Respondents were asked to place themselves in one of 10 income brackets, defined by specific monetary ranges (e.g. $0-$10000, $10000- $20000, etc.). These brackets were different for each country, and were meant to correspond to the income deciles for each country. However, it is clear from the survey responses that the categories rarely matched income deciles. As a consequence, we have only a rough idea of the relative income of households in each country. Therefore, per capita gross domestic product (GDP) measured using purchasing power parity from the World Development Indicators (2002) is included as an indicator of average income in the country. Since income has been shown to have diminishing marginal utility (see for example Inglehart (1996)) a quadratic term is also included for per capita GDP. The other respondent characteristics that are included as explanatory variables are indicator variables for employment characteristics, education level, marital status, age, gender, number of children, religious (self-described), and poor health (self-described). Apart from the last two characteristics, this is replicating the type of explanatory variables included in DiTella et al. (2001). Results Table 3. shows the raw correlations between the environmental indicators and happiness or satisfaction. Note that the sample size differs because the environmental indicators are not all available 7

for all the countries. In all cases, with the exception of nitrogen dioxide and satisfaction, as pollution increases, stated happiness or satisfaction decreases. Table 4. shows estimated ordered probit equations with happiness or satisfaction as the dependent variable for each of the pollution indicators, also controlling for relative household income and per capita GDP. Generally, subjective well-being is increasing with relative household income, with the exception of the 8 th category which is somewhat smaller in magnitude than the 7 th income category (and the 6 th income category is somewhat lower than the 5 th in the satisfaction equation with per capita water pollution). Subjective well-being is also increasing with per capita GDP. The pollution variables overall do not exhibit a statistically significant relationship with satisfaction or happiness, except for per capita water pollution. An increase in per capita water pollution is associated with a decrease in subjective well-being. In Table 5, focusing on per capita water pollution, the estimated equations include additional control variables. The first two columns contain the results of the ordered probit estimation, while the last column shows regression results for the satisfaction equation. The positive relationship of per capita GDP to happiness continues to hold, as does the negative impact of water pollution. The negative parameter estimate on the quadratic term for per capita GDP, suggests that per capita GDP has a diminishing marginal utility (statistically significant only in the satisfaction equations). Controlling for additional respondent characteristics, the parameter estimates on the relative income categories are no longer all statistically different from zero, however, in both the satisfaction and happiness equations those in the highest relative income category are more likely to be very happy or satisfied than those with lower relative income. Utilizing the parameter estimates on per capita GDP and per capita water pollution from the regression model, an estimate for the marginal willingness to pay for pollution reduction may be calculated. This estimate will vary by country because of the quadratic term included for per capita GDP. 8

For the median per capita GDP level of $6,207, the marginal willingness to pay is calculated to be $660 per gram per capita per day per year (obtained by dividing the parameter estimate on water pollution by the marginal effect of per capita GDP from both the linear and quadratic terms). Note that the mean is 7.43 grams, so a reduction of one gram per capita per day per year is a reduction of more than 10 percent. The parameter estimates on education are not statistically significant in the happiness equation, however, in the satisfaction equation, those with some elementary or completed elementary education appear more satisfied than those with secondary or university education, while those with at least some university education are more satisfied than those with some or completed secondary education (and all are less satisfied than those in the omitted category of no education). The employment categories (employed is the omitted category) suggest that the unemployed are less happy or satisfied than the employed, as are retired persons, although by a smaller magnitude (and not statistically significant in the happiness equation). On the other hand, students and housewives (wording used in the survey) are more happy or satisfied than the employed (although not statistically significant in both equations). Happiness or satisfaction appears to decline with age, although at a decreasing rate, given the positive quadratic term. The parameter estimate on female suggests that they may be less happy or satisfied than men, however, this difference is not statistically significant. Examining marital status (single is the omitted category) suggests that those who are married and living together as married are happier or more satisfied than single respondents. Of the categories of divorced, separated and widowed, separated respondents were less happy than single respondents, while the other parameter estimates were not significantly different from zero. The parameter estimates on number of children indicator variables were not statistically significant in either the satisfaction or happiness equations, although the parameter 9

estimates suggest that those with no children (omitted category) and those with four or more children may be happier or more satisfied than those with one to three children (note that this is for total number of children, not children living at home). The last two characteristics included are self-categorized assessments indicating if the respondent considers himself or herself religious and if health is considered to be either poor or very poor. Respondents considering themselves to be religious are more likely to be happy or satisfied and those with poor health are less likely to be happy or satisfied. Summary and Conclusions Subjective well-being data may provide the possibility of important insights into economic behavior. In this paper we examine the potential of using these data to estimate the marginal willingness to pay for environmental quality. We find that as per capita water pollution increases, satisfaction or happiness decreases. For the other pollution indicators, the relationship is not statistically significant. We also find that as per capita GDP increases, satisfaction or happiness increases. In general, at least when statistically significant, happiness and satisfaction are also found to increase with relative household income. This paper suggests that as better, comparable, data on environmental quality become available for a broad range of countries, this type of survey information may prove to be an important source for cross-country comparisons. Due to the quite serious data limitations of the currently available environmental data for many countries, and the lack of absolute household level income in the World Values Survey, results of this paper should be seen as an early exploration of the possibilities of this methodology, rather than a source of firm estimates of willingness to pay. 10

11

References Bloom, David E. 1995. "International Public Opinion on the Environment." Science 269:354-358. Brechin, Steven R., and Willett Kempton. 1994. "Global Environmentalism: A Challenge to the Postmaterialism Thesis?" Social Science Quarterly 75:245-269. Di Tella, Rafael, Robert J. MacCulloch, and Andrew J. Oswald. 2001. "Preferences over Inflation and Unemployment: Evidence from Surveys of Happiness." American Economic Review 91:335-41. Dunlap, Riley E., and Angela G. Mertig. 1996. "Global Environmental Concern: a Challenge to the Postmaterialism Thesis." in Social Dimensions of Contemporary Environmental Issues: International Perspectives. Ester, Peter and Wolfgang Schluchter, eds. Tilburg: Tilburg University Press, 133-164. Frank, Robert H. 1997. "The Frame of Reference asa Public Good." The Economic Journal 107:1832-1847. Freeman, A. Myrick III. 1993. The Measurement of Environmental and Resource Values. Resources for the Future: Washington, DC. Frey, Bruno S. and Alois Stutzer. 2002. "What Can Economists Learn from Happiness Research." Journal of Economic Literature 40:402-435. Georgiou, Stavros, Dale Whittington, David Pearce, and Dominic Moran. 1997. Economic Values and the Environment in the Developing World. Cheltenham, UK: Edward Elgar. Greene, William H. 1993. Econometric Analysis. 2nd Edition. New York: Macmillan Publishing Compa ny. Grossman, Gene M. and Alan B. Krueger. 1995. "Economic Growth and the Environment." Quarterly Journal of Economics 110: 353-377. Inglehart, Ronald. 1996. "The Diminishing Utility of Economic Growth: From Maximizing Security toward Maximizing Subjective Well-Being." Critical Review 10:509-31. ----------------------1995. "Public Support for Environmental Protection: Objective Problems and Subjective Values in 43 Societies." PS: Political Science & Politics 28:57-72. Israel, Debra K. 1999. Essays on Energy, Equity, and the Environment in Developing Countries. PhD Thesis. Madison, WI: University of Wisconsin-Madison. Israel, Debra K., and Arik Levinson. 2002. "Willingness to Pay for Environmental Quality: Testable Implications of the Growth and Environment Literature." Georgetown University Working Paper. 12

Ng, Yew-Kwang. 1997. "A Case for Happiness, Cardinalism, and Interpersonal Comparability." The Economic Journal 107:1848-1858. Oswald, Andrew J. 1997. "Happiness and Economic Performance." The Economic Journal 107:1815-1831. Portney, Paul R. 1994. The Contingent Valuation Debate: Why Economists Should Care. Journal of Economic Perspectives 8:3-17. Welsch, Heinz. 2002. "Preferences over Prosperity and Pollution: Environmental Valuation Based on Happiness Surveys," KYKLOS, 55:473-494. Whittington, Dale. 1998. Administering Contingent Valuation Surveys in Developing Countries. World Development 26:21-30. 13

Table 1. Happiness, Satisfaction, and other Relevant Variables Variable Names Means Happiness 2.97 Satisfaction 6.24 Unemployed 0.08 Self-employed 0.09 Female 0.51 Age 41.46 No Formal Education 0.04 Some Elementary Education 0.07 Complete Elementary Education 0.14 Some Secondary Education 0.17 Complete Secondary Education 0.35 Some University Education 0.07 Complete University Education 0.16 Married 0.62 Living together as married 0.06 Divorced 0.04 Separated 0.02 Widowed 0.08 Single 0.19 No children 0.25 One Child 0.17 Two Children 0.29 Three Children 0.15 Four or More Children 0.14 Household Income Category 1 0.13 Household Income Category 2 0.15 Household Income Category 3 0.14 Household Income Category 4 0.12 Household Income Category 5 0.11 Household Income Category 6 0.09 Household Income Category 7 0.08 Household Income Category 8 0.07 Household Income Category 9 0.06 Household Income Category 10 0.05 Retired 0.15 Student 0.06 Housewife 0.12 Religious 0.72 Poor health 0.09 Source: Author s calculations from World Values Survey Wave 3 N=32,961 14

15

Table 2. National Variables: Means and Standard Errors Per Capita GDP (1000 purchasing power parity $) Number of Countries Mean (s.e.) 35 9.68 (1.42) Per capita Organic Water Pollutants (grams per day) 35 7.43 (0.71) Total suspended particulates (TSP) (micrograms per cubic meter) 21 106.02 (19.96) Nitrogen dioxide (micrograms per cubic meter) 22 58.42 (6.28) Sulfur dioxide (micrograms per cubic meter) 22 35.55 (7.05) Source: Author s calculations from World Development Indicators (2002) 16

Table 3. Correlations between Pollution and Subjective Well-being Measures Per capita Organic Water Pollutants (grams per day) Happiness Satisfaction N -0.09-0.05 54,743 Total suspended particulates (TSP) (micrograms per cubic meter) -0.07-0.03 31,718 Nitrogen dioxide (micrograms per cubic meter) -0.04 0.04 32,824 Sulfur dioxide (micrograms per cubic meter) -0.06-0.10 34,138 Source: Author s calculations from World Development Indicators (2002) 17

Table 4. Ordered Probit Models with Different Measures of Pollution Household Income Category 2 Household Income Category 3 Household Income Category 4 Household Income Category 5 Household Income Category 6 Household Income Category 7 Household Income Category 8 Household Income Category 9 Happy Satisfied Happy Satisfied Happy Satisfied Happy Satisfied 0.20* 0.31* (0.10) 0.33* 0.36* 0.40* 0.44* 0.35* 0.46* Household Income 0.53* Category 10 1 Per Capita GDP (1000 purchasing power parity adjusted dollars) Total suspended particulates (TSP) (micrograms per cubic meter) Nitrogen dioxide (micrograms per cubic meter) Sulfur dioxide (micrograms per cubic meter) Per capita Organic Water Pollutants (grams per day) 0.01 (0.020-0.0005 (0.0014) 0.09 (0.10) 0.16 (0.12) 0.16 0.21 0.25 (0.14) 0.30* (0.14) 0.26 0.33* (0.14) 0.47* (0.17) 0.03* (0.01) 0.0015 (0.0010) 0.20* 0.31* 0.32* 0.37* 0.42* 0.47* 0.40* 0.49* 0.53* 0.02 (0.01) -0.0013 (0.0029) 0.18 (0.10) 0.27* (0.11) 0.28* (0.12) 0.38* 0.40* 0.43* (0.14) 0.40* 0.41* (0.14) 0.46* 0.03* (0.01) 0.0024 (0.0038) 0.19* (0.10) 0.31* (0.10) 0.33* 0.37* 0.42* 0.47* 0.39* 0.46* 0.53* 0.02 (0.01) 0.0014 (0.0040) 0.18 (0.11) 0.28* (0.12) 0.29* 0.39* (0.14) 0.41* 0.46* 0.43* 0.46* 0.53* (0.16) 0.03* (0.01) 0.0012 (0.0038) Sample Size 25,630 28,796 29,103 44,039 * significantly different from zero at 5% level Standard Errors Adjusted for Clustering by Country 1 Category 1 is the lowest income category and category 10 is the highest 0.09 0.24* 0.30* 0.32* 0.33* 0.36* (0.10) 0.30* (0.12) 0.32* 0.47* (0.11) 0.05* (0.01) -0.08* (0.02) 0.11 0.23* 0.28* (0.11) 0.35* (0.11) 0.33* 0.36* 0.34* (0.16) 0.36* (0.16) 0.54* 0.05* (0.01) -0.07* (0.02) 18

Table 5. Happiness and Satisfaction Estimated Equations with Per capita Water Pollution (continued on next page) Happiness Ordered Probit parameter estimate (s.e.) Household Income Category 2 0.05 Household Income Category 3 0.14* Household Income Category 4 0.17* Household Income Category 5 0.16* Household Income Category 6 0.15 Household Income Category 7 0.19* Household Income Category 8 0.13 (0.10) Household Income Category 9 0.17 (0.12) Household Income Category 10 1 0.35* (0.12) Per Capita GDP (1000 purchasing power parity adjusted dollars) 0.07* (0.03) Per Capita GDP squared -0.0007 (0.0008) Per Capita Organic Water Pollutants (grams per day) -0.06* (0.02) Some Elementary Education -0.05 Complete Elementary Education -0.04 Some Secondary Education -0.04 Complete Secondary Education -0.12 Some University Education 0.00 Complete University Education -0.05 SatisfactionOrdered Probit parameter estimate (s.e.) -0.03 0.07 0.10 0.14 0.13 (0.11) 0.16 0.17 (0.14) 0.21 (0.14) 0.39* (0.14) 0.12* (0.03) -0.0023* (0.0010) -0.06* (0.02) -0.07-0.18* -0.23* (0.10) -0.26* -0.20* -0.21* Satisfaction Regression parameter estimate (s.e.) -0.06 0.16 (0.17) 0.25 (0.21) 0.33 (0.21) 0.32 (0.25) 0.38 (0.29) 0.42 (0.32) 0.50 (0.32) 0.91* (0.31) 0.29* -0.01* (0.00) -0.15* (0.04) -0.19 (0.19) -0.47* (0.19) -0.59* (0.23) -0.64* (0.22) -0.50* (0.21) -0.51* (0.20) 19

* significantly different from zero at 5% level Standard Errors Adjusted for Clustering by Country Sample Size: N=32,961 1 Category 1 is the lowest income category and category 10 is the highest Table 5. Happiness and Satisfaction Ordered Probit Equations with Per capita Water Pollution (continued from previous page) HappinessOrdered Probit parameter estimate (s.e.) Unemployed -0.22* Self-employed 0.05 (0.04) Retired -0.06 Student 0.12 Housewife 0.09* Female -0.03 (0.02) Age -0.03* (0.01) Age squared 0.0002* (0.0001) Married 0.28* Living together as married 0.20* Divorced -0.06 Separated -0.19* Widowed -0.14 One Child -0.08 Two Children -0.06 Three Children -0.03 Four or More Children 0.03 SatisfactionOrdered Probit parameter estimate (s.e.) -0.26* 0.04-0.08* (0.04) 0.15* 0.04 (0.04) -0.03 (0.02) -0.03* (0.01) 0.0003* (0.0001) 0.20* 0.15* 0.01-0.06-0.01-0.05 (0.04) -0.06 (0.03) -0.01 (0.04) 0.07 (0.04) Satisfaction Regression -0.61* 0.09 (0.11) -0.19* 0.35* (0.12) 0.08 (0.10) -0.08-0.06* (0.01) 0.00* (0.00) 0.46* (0.11) 0.35* 0.05-0.12 (0.14) 0.01-0.13-0.15-0.05 0.12 20

Consider self to be religious 0.19* (0.03) Consider self to be in poor health -0.79* 0.15* (0.04) -0.64* 0.32* -1.46* (0.12) Constant 6.88* (.35) * significantly different from zero at 5% level Standard Errors Adjusted for Clustering by Country Sample Size: N=32,961 Appendix: Happiness or satisfaction questions from the World Values Survey V10 Taking all things together, would you say you are: 1. Very happy 2. Quite happy 3. Not very happy 4. Not at all happy 9. Don t know [DO NOT READ OUT] (Categories one through four were reversed for ordered probit model in this paper) V65 All things considered, how satisfied are you with your life as a whole these days? 1 2 3 4 5 6 7 8 9 10 Dissatisfied Satisfied DK=99 21