Economic crisis and public attitudes toward science: A study of regional differences in Spain

Transcription

1 INTERNATIONAL POLITICAL SCIENCE ASSOCIATION (IPSA) XXII nd World Congress of Political Science Madrid, 8-12 July 2012 RC35: Technology and Development Panel: Regional R&D Policies: Models, Instruments and Players Economic crisis and public attitudes toward science: A study of regional differences in Spain Luis Sanz Menéndez Institute of Public Goods and Policies (IPP), Consejo Superior de Investigaciones Científicas (CSIC), Madrid Luis.Sanz@csic.es Gregg G. Van Ryzin School of Public Affairs and Administration Rutgers University, Newark Vanryzin@rutgers.edu July 2012 First draft, please don t quote without contacting the authors 1

2 Abstract We investigate change in public attitudes toward science and technology (S&T), as well as support of government S&T spending, in the context of the European economic crisis, which has been particularly severe in Spain. Although there is little theory on the effects of economic crisis on public attitudes towards science and technology, some empirical evidence suggests that public support for S&T in a time of economic crisis could decline in the face of more pressing priorities, such as jobs and social services. But the public may also view S&T as a strategic pathway out of an economic crisis. We test these competing hypotheses employing two national surveys from Spain, implemented before (2006) and after (2010) the onset of the economic crisis in that country. Moreover, the surveys provide equal size samples for each of Spain s 17 regions, which have varied a great deal in their post crisis experiences of unemployment and economic recession. It is thus possible to correlate changes in attitudes about S&T with the severity of the recession by regions over time. We find that, in regions hit hardest by the economic crisis (compared to less affected regions), trust in the benefits of S&T increased substantially, as did general public interest in S&T. Similarly, residents of the hardest hit regions were more likely after the crisis to choose S&T (out of a list of policy areas) as a priority for government, and somewhat more likely to express support for increases in government S&T spending. Implications of our findings for research on the public understanding of science and for public policy are discussed. 2

3 Introduction The financial downturn that started in 2007 and then accelerated with Lehman Brothers bankruptcy in 2008 has been transformed into a global economic crisis. After various government stimulus packages, in which science and technology had a relevant role (OECD 2009), and the bailout of some Euro countries (Greece, Ireland, Portugal), the dominant government policy adopted in Europe has been one of fiscal consolidation, meaning significant budget cuts in many areas of public spending. Policy priorities with respect to the areas of public spending to be preserved have varied quite a bit. This has particularly been the case with government spending on science and technology (S&T): some governments increased the priority given to research funding, while others significantly cut their science budgets (which has been clearly the case of Spain; see Science, 2012). The crisis has contributed to significant increases in unemployment levels, the insecurity in of many groups in society, and fears about the future. As a result, the crisis seems to be contributing to a significant change in the aggregate preferences of citizens regarding government priorities. For example, in the second semester of 2010, more than 52% of Europeans believe that tackling poverty and social exclusion should be given priority as a policy, with an increase of 8 points compared to a previous survey early in the same year (Eurobarometer, 2010a). There was already evidence of a general support towards welfare state spending in the 1990s (Svallfors, 1997, 1999; Taylor Gooby, 2001) and explanation of national variations (Blekesaune & Quadagno, 2003, Fraile & Ferrer, 2005). We know that the citizens preferences regarding government spending change in times of economic crisis. There have been analyses of the effects of crisis on public opinion in other areas, such as the environment, that confirm such changes (e.g. Scruggs & Benegal 2012). When confronted with more pressing priorities such as jobs and cuts in social services, citizens change their priorities in terms of government action; priority assigned to welfare policies, especially those related with 3

4 unemployment protection and fight against social exclusion, increases in times of crisis. In contrast, other policies, such are security, defense or environmental protection, tend to see a reduction in the priority attributed by citizens and in the assessment regarding the allocation of government resources. These patterns are also present in the Spanish situation (AEVAL 2011). However, very little is known regarding the factors influencing changes in citizens preferences for science and technology and the role of government spending in this policy area. In Europe in 2010, 14% of the citizens thought that R&D should be a priority, with a small increase of 1 point in comparison with the previous semester. At the same time, in the US, the comparison between the pre crisis and post crisis has seen a reduction in those that believe that R&D is receiving less money than is necessary (US NSF 2012), while other areas radically increase that share, like education. However, general results of the 2010 surveys in Europe (Eurobarometer 2010b) and US (NSF) when not compared with other policy areas present similar levels of interest, support and positive attitudes towards science than the previous pre crisis surveys. However, general results could hide some significant changes that we would like to explore. This is why in this paper we investigate the change in public attitudes toward science and technology (S&T), as well as support of government S&T spending, in the context of the European economic crisis, for the Spanish case, a country that has suffered a very rapid increases in unemployment levels. We would also like to get an idea if the relevance of this S&T policy is consistent with citizens views on the priorities and benefits they expect from government spending in science. Analytical Background In this paper we try to combine two streams of literature to address our research question regarding the effects of the crisis on changes in citizens general attitudes towards science and technology and on their support for government S&T spending. To begin with, there is a rich literature regarding citizen s attitudes towards science that has accumulated some evidence of about the general 4

5 determinants of positive attitudes towards science and technology (Bauer et al 2007). From this field the level of awareness, interest and substantive knowledge concerning science issues by the public have been used to account for various general attitudes towards science. The dominance of the so called public deficit model has been widely noted (Sturgis & Allum 2004). First was a focus on scientific literacy (Miller 1983a; Miller 1999), with concerns about the limited scientific knowledge of the public. Second was the emergence of the public understanding approach (Miller 2004), which acknowledges more complexity in the analysis of attitudes towards science. Additionally, when we move from general attitudes toward science to more concrete fields of science, prior studies suggest significant levels of ambivalence (Bauer 2002, Nisbet 2005). Individual beliefs vary greatly from one society to another, according to various institutional factors and social structures. Different factors such as the level of income, the different social structures, or varying expenditure levels in a policy area may affect attitudes toward public spending. A high level of development in a country (or a region) and a high income level tend to moderate the demands of citizens for public intervention and the preferences toward more materialist or postmaterialist policies (Inglehart 1990). There are significant variations in national attitudes towards science, with some studies suggesting an inverse U shaped function connecting the level of development and industrialization of societies and the level of support or acceptance of science in Europe (Bauer et al 1994; Durant et al 2000). Most developed societies have become more cautious regarding some side effects of S&T (Gaskell et al 1999) and the postmaterialist transformation (Inglehart 1990) has been used to account for the nonlinear relationship between the level of economic development (post industrialism values) and ambivalence regarding science. We also know that citizens preferences and their determinants regarding public spending vary substantially depending on the sector of public policy. For example, public preferences toward welfare state policies (health, education, pensions, or unemployment benefits) seem to respond to a more coherent structure of determinants (Jacoby 1994). Additionally, the different contexts as well 5

6 as the features, expectations or outcomes of these policies are also likely to condition citizens attitudes toward public spending (Svallfors 2003). As a result, a single structure of determinants of attitudes towards public spending can not be applied across all public policies and programs (Monroe 1979). It has been a long assumption in the area of public understanding of science that positive attitudes towards science translated to support for more public spending on S&T. Recently, some confirmation has been found that interest (Muñoz et al 2012) and the complete set of factors that appear to be relevant for explaining the positive attitudes towards S&T serve also as an explanation of the priority and the support for more government spending in this area (Sanz Menendez et al 2012). But there is no clear theory regarding the effects of an economic crisis on public support for science and government spending on S&T. A comparison of survey results before and after the onset of the crisis suggests an apparent increase in citizens support for S&T as a priority (see introduction); also more support in the selection for S&T as a priority government spending area has been identified (FECYT 2011, Pavone et al 2011). Based on diverse empirical evidence, we could have two competing hypothesis for accounting for the impact of the crisis. Some of the prior studies would seem to suggest that public support for S&T in a time of economic crisis should decline in the face of more pressing priorities, such as jobs and social services. The expectation would be that S&T support behaves like citizens support for environmental issues (Scruggs & Benegal 2012); both policy domains could be considered postmaterialist. On the other hand, we might also expect that the public would view S&T as a strategic pathway out of an economic crisis; if so, then citizens might express an increase in positive attitudes toward science and more support of S&T spending as a part of the solution, suggesting that researchers are successful in building hopes and expectations (Van Lente & Rip 1998; Sanz Menendez & Cabello 2000) regarding science and technology results. 6

7 If S&T and related government policy are located in a postmaterialism domain, we would expect to observe a general reduction in the public interest in science and in the support for government spending on S&T. However, there is a second alternative, that science and technology policies do not simply reflect postmaterialist values but also have a strong materialist component associated with the perception of growth, development and the promise of increasing economic wellbeing of citizens. In this case, we would expect to find an increase support for S&T and government spending in this domain. We will test these competing hypotheses employing two national surveys from Spain, implemented before (2006) and after (2010) the onset of the economic crisis. In the context of a single country, we have the opportunity to analyze the changes in public attitudes towards science and support for S&T government spending in the different regions and control for the different level of the impact of crisis. As a consequence of the empirical evidence (Bauer et al 1994; Quintanilla & Escobar 2005; Quintanilla et al 2011) regarding the level of development and the relevance of science in a region/country, the general expectations are that living in a more advanced region will influence positively the level of support for S&T polices and spending; alternatively, if the post materialist expectation and the responsive policymakers hold, the most developed regions may reduce their support for the allocation of public resources to science and technology. This is why in this paper we aim to address, through the analysis of difference in difference, two main questions regarding the changes in general attitudes towards science and technology and, also, public support for S&T government spending. We also take, as control variables, some of the variables that the literature has identified as relevant for understanding both attitudes towards science and support for S&T government spending. Some personal attributes (gender and age), characterization of the educational level, and also some ideological and religious beliefs are control variables in accounting for the explanation. 7

8 Methodology To test our hypothesis about the effect of economic crisis on attitudes toward science, we employ a difference in differences analysis of pooled cross sections from two comprehensive surveys conducted in Spain in 2006 and 2010, thus capturing snapshots of public opinion before and after the onset of the economic crisis in 2008 (*). The surveys were promoted Spanish Foundation for Science and Technology (Fundación Española para la Ciencia y la Tecnología FECYT), the first carried out by the Spanish Center for Sociological Research (Centro de Investigaciones Sociológicas [CIS] 2006/2652)and involved in person household interviews of randomly selected adult residents of the 17 regions of Spain, with 7056 completed interview in 2006 and 7744 completed interviews in Although the 17 regions of Spain vary greatly in population, from about 300 thousand (Rioja) to over 8 million (Andalucía), the surveys were designed to provide approximately equal size samples for each region, thus allowing for more precise estimates of public opinion at the regional level. Importantly, the regions of Spain vary a great deal in how deeply they were affected by the crisis. As Table 1 shows, the change in unemployment from 2006 to 2010 ranged from less than 7 percentage points in some regions (Pais Vasco, Navarra, Asturias) to about 15 percentage points or more in other regions (such as Valencia, Andalucía, Murcia, and Canarias) 1. We exploit this regional level variation in the severity of the crisis to identify a cluster of 8 lower unemployment regions growth and a second cluster of 9 higher unemployment growth regions. It is should be pointed out that the cluster of higher unemployment regions also have higher absolute levels of unemployment in 2010, as well as having experienced greater increases in unemployment. Unemployment, we would argue, is a meaningful measure of the impact of the crisis in a region and perhaps the most salient economic factor shaping public opinion. 1 INE. Spanish Labour Force Survey (Encuesta de Población Activa) 8

9 Difference in differences strategy We take advantage of the fact the two surveys provide snapshots of public attitudes toward science before (2006) and after (2010) the economic crisis, which emerged full force by the end of 2008 (especially following the collapse of the US investment bank Lehman Brothers in September, 2008). Clearly, the impact of the economic crisis on European society is complex and still ongoing (at the time of this writing). In Spain, the crisis has been especially severe but also quite varied across regions, which have diverse economies and substantially autonomous regional governments (*) 2. We thus employ a difference in differences statistical strategy, pooling the two cross sectional surveys, to compare change in science attitudes in the high unemployment regison to change over the same period in the low unemployment regions. In a regression framework, our model is as follows (based on Remler & Van Ryzin, 2011): Yˆ = a + b R + byr T + b (R T ) + b C b uk int c1 cj Cj Where Ŷ is a measure of attitudes towards science; R is a dummy variable indicating a respondent from a region with high unemployment; T is a dummy variable coded 0 for the pre crisis period (2006) and 1 for the post crisis period (2010); R T is an interaction term; and C1 through Cj are available control variables added to the model (as discussed shortly). The slope b R estimates the difference in science attitudes (Y) between the high and low unemployment regions during the precrisis period and is assumed constant over time. The slope bt estimates the difference in science attitudes (Y) between the pre and post reform period for the low unemployment regions and is thus assumed to be the trend that would have occurred in high unemployment regions, had the 2 The Spanish Regional Governments (Comunidades Autonomas) had a significant level of policy competencies, especially in areas of health provision and education; R&D is also a competence of regional government, but shared with the national government. 9

10 crisis not affected attitudes. Finally, the slope on the interaction bint estimates the difference in differences, or the net difference (from the trend) in the dependent variable (Y) for the high unemployment regions and is thus the presumed causal effect of the crisis on attitudes toward science and technology. Measures The survey questions in both years asked about interest in, knowledge of, and attitudes toward S&T, as well as basic social, economic, and political characteristics of respondents. However, because different data collection institutions were used in 2006 and 2010, some of the questions were not asked in the same way and thus were not directly comparable over time. Still, there were a sufficient number of directly replicated questions to allow for a pooled analysis of both surveys, including questions about general attitudes toward science, support of science policy, and background factors to serve as control variables. The summary statistics for the available analytical variables are shown in Table 2. Beginning with the dependent variables, we selected two consistently asked measures of the public s general attitude toward science and technology. The first is a direct self reported level of interest in science and technology on a 1 5 scale (from 1=very uninterested to 5=very interested) asked a part of a list of current topics presented to respondents (including, for example, economics and business, politics, culture and the arts, sports, and so on). To make this measure consistent with our other dependent variables, which are all dummy variables, we recoded those scoring at least 3 (the midpoint) as 1, indicating those interested in science, and recoded the rest (those scoring less than 3) as 0. The second measure of general attitudes is a dummy variable indicating respondents who believe that, on balance, the benefits of science and technology outweigh the risks. It comes from a question in which there were two other choices (both coded zero in our 10

11 analysis), specifically that the benefits and risks of S&T were about equal, or that the risks of S&T outweighed the benefits. We have three measures of support for S&T policy that are available as dependent variables for a pooled analysis. The first comes from a question in the survey that asked respondents to imagine that they could decide where to spend public funds and then showed respondents a card with the following sectors listed: public works, public safety, transportation, science and technology, environmental protection, defense, justice, culture, and sports. Respondent could choose up to three of these sectors, and we created a dummy variable that indicates respondents who chose science and technology as one of their three choices 3. The other two measures of support for S&T policy come from a more direct set of questions about support for S&T spending by the Spanish central government and by the regional government. These questions asked whether respondents believed the government spends too much, the right amount, or too little on science and technology. We coded those who selected too little as an indicator of the desire for more public spending on science and technology. The available control variables, shown in Table, include sex, age, education, practicing Catholic, and political ideology. Education was operationalized as two dummy variables, one indicating a low level of education, meaning less than a full secondary education, and another indicating a high level of 3 The list of policies offered in this question deliberately excluded health, education, employment and pensions because these programs enjoy a wide popularity among citizens and are known to be selected as the first choices for public spending. It was a way of magnifying the visibility of science and technology policy for citizens. In general survey (AEVAL 2011; CIS 2840, 2010) in which all policy fields (14 different ones) were included together in the same question (in favor or against budget cuts), science and technology policy resulted the third policy field having less opposition to budget cuts, just before defense and public works, but after culture, transportation, justice, environment and to a big distance of the set of policies related to the welfare state. 11

12 education, meaning the completion of at least some post secondary education. Practicing Catholic was included as a proxy for social and religiously conservative attitudes, which may influence interest in and support of science. And political ideology, measured on 1 10 left right scale, was included to hold constant differences in political values and preference across regions and over time. Analysis and Results Because our dependent variables are all dichotomous, we ran the difference in differences analysis using probit and report the results as marginal effects 4. Table 3 shows the results for the general attitudes toward science, including interest in science and trust that the benefits of S&T outweigh the risks. For each outcome, two models are shown: a basic difference in differences model, without control variables, and another model with adjustment for the six available control variables. This was done to provide some sense of the sensitivity of the results to potential confounding from these control variables. The main coefficient of interest is the interaction term, which represents the net change in the probability of the outcome in higher unemployment regions relative to the change over time that occurred in the lower unemployment regions. As Table 3 shows, interest in science significantly increased in higher unemployment regions, relative to the change that occurred in lower unemployment regions. The coefficient (focusing on the full model with control variables), indicates a net gain of nearly 6 percentage points in the level of interest in science in higher unemployment regions of Spain. The net gain in trust that the benefits of S&T outweigh the risks is even larger, nearly 14 percentage points. Figure 1 illustrates these results graphically, showing the clear change in the higher unemployment regions, relative to the change experienced in the lower unemployment regions, in the level of trust in the benefits of science and technology. 4 We used Stata 12 and the dprobit command. 12

13 Table 4 presents the probit analysis of the support S&T policy. Again relative to the change in the lower unemployment regions, the higher unemployment regions experienced an increase of nearly 6 percentage points in the proportion of people selected science and technology as a priority for government investment. This result is illustrated graphically in Figure 2, which shows a distinct gain in viewing S&T as a priority in the higher unemployment regrions of Spain. But the probit analysis in Table 4 does not indicate much of a difference between higher and lower unemployment regressions with respect to the view that central government is spending too little on S&T. Similarly, there is no difference in differences regarding the view that the regional government spends too little on S&T. Discussion The aim of our paper was to analyze the impact of the economic crisis in the citizens attitudes towards science and technology and their support for government S&T spending. The 17 Spanish regions were taken as a case for analysis. We measured the changes in unemployment levels between 2006 and 2010 and we developed a difference in difference analysis to test the effect of the changes as a result of the crisis. Regarding the effect of crisis on general attitudes towards S&T, the results are conclusive. Interest in S&T has significantly increased, and with the control variables included there is an expected 5.6 additional points in regions hardest hit by unemployment, relative to regions in which unemployment has increased less. Most of the control variables included in the model are also significant. The direction of the coefficients is also consistent with results of previous research. The biggest effect comes from having a low educational level, which reduces by 17.7 points the interest in S&T; having a postsecondary education degree, in contrast, increases by 16.0 points the expected interest in S&T. Results regarding gender show a reduction of 8.4 points for females, relative to males. Age and ideology have statistically significant effects, but the size of the effect is marginal. 13

14 With respect to trust that the benefits of S&T outweigh the risks, our result show a net increase of 13.8 points in regions hit by high unemployment growth in comparison with regions affected by softer increases of unemployment. Again, most of the control variables have good significance; the educational level having the biggest effect. Being females and practicing catholic reduces trust by 4 points. Again, age is significant, but the effects are very small. In summary, interest and positive assessment of global benefits regarding science appears to have increased with the impact of the crisis. However, going to the details, we witness that in regions less hit by the crisis there are quite stable results, but in regions with higher gains in unemployment the interest in S&T and the belief that science has more benefits increases. The effect of the control variables on the interest and appreciation of S&T go in the direction reported by previous research: High levels of education predict higher levels of interest and more positive attitudes towards science. Being a female produces a negative effect on interest in and perceived benefits of science. This is consistent with previous research that has sometimes presented women as less interested and more skeptical of the benefits of science, while at the same time the general literature of public spending finds women to be more supportive of social welfare policies and less with postmaterialist policies. Globally, positive attitudes towards science have been linked with scientific literacy of young, male and educated people (Miller 1983b), although gender and age are sometimes not significant predictors in multivariate analyses (Hayes & Tariq 2002; Nisbet et al 2002; Sturges & Allum 2004). Education level has been shown to be a strong predictor of support for science (Miller et al 1997). However, more educated people sometimes appear more ambivalent regarding S&T (Lujan & Todt 2000, Torres 2005) or with stronger positions (Pardo & Calvo 2002). Ideology and religious belief, when significant, also produce effects consistent with what previous research has found. Left oriented and non Catholics have been shown to be somewhat more supportive of science developments (Miller et al 1997) and religious beliefs play also moderate the impact of the awareness in support of science (Nisbet 2005). 14

15 As in previous research (Sanz Menendez et al 2012) we have made a dual approach regarding the assessment of citizens support of S&T Government spending. We have asked citizens to choose the priority areas for public spending, excluding welfare policies; then introducing an opportunity cost of choosing. The results are significant and clear. Regions with higher increases in unemployment levels have also increase (of 5.5 points) their level of priority of government spending in S&T, when excluded welfare policies. Then it looks like the perception of this policy area for the citizens is associated to the way out of the economic crisis and the change of productive model with a much relevant role of knowledge. S&T policy appears as a form of government spending that provides hope and raising expectations (in comparison with other policies, like environment). When we include the control variables, again having a secondary diploma increases support by 15.3 point s. While being a female, practicing catholic, having a conservative ideology produces negative effects. We see that the control variables produce the same effects in interest, positive assessment and choice of S&T as priority area of spending. When we analyze the assessment of the citizens with respect the actual level of spending in S&T by the Spanish Central and Regional Governments, we found that the effect is significant only in the basic model; once introduced the control variables the significant differences between regions with higher increases in the levels of unemployment and the rests, become non significant at 10% level. This is consistent with the general finding from AEVAL, when included all policy fields. However the predicted effect of all control variables is significant and goes in the expected direction predicted in previous research. For example, women and younger people have been shown to be more favorable toward government spending, but that is not the case for women in support for S&T government spending. Positive attitudes towards science have been linked with scientific literacy of young, male and educated people (Miller 1983b); the education level has been shown to be a strong predictor of support to science (Miller et al 1997) and it appears positively related with support of 15

16 many types of government spending, except perhaps military issues. Concerning ideology and religious beliefs, attitudes to public spending are affected by people s moral values, specifically those related to economic redistribution and social justice. Left oriented and non Catholics have been shown to be somewhat more supportive of science developments (Miller et al 1997). Religious beliefs play a role also in accounting for diverse attitudes towards governments spending, with less religious people more in favor of a stronger government role (Sanders 1988); religious beliefs play also have a moderate impact on awareness of and support for science (Nisbet 2005). Based on a pre crisis data, recent research (Sanz Menendez et al 2012) find that interest in science and technology as the most important factor explaining the choice of S&T as priority government spending area, once excluded welfare areas. From our analysis of the impact of crisis we found that the hypothesis that people reduce their interest in and support for S&T by governments is not confirmed. The fact is the opposite, the interest in S&T and the choice of more government spending in S&T is positively associated with higher impacts of the crisis in their regions. It is clear that citizens believe that S&T public spending could be a positive element that could contribute to coping with the crisis. Off course the assessment could be just the result of the early stages of the crisis. A process of deepening of the crisis could produce the consequence of changing the assessment of the sort term contribution of science and technology as the way out of the crisis. Then, with the comparative analysis, pre post crisis, we can confirm an increase in the interest and trust in science, as a hoped for way out of the crisis, and also an increase in the support for to S&T as a priority for government spending. In terms of policy consequences, S&T policy makers could find that, despite the fact that S&T could be considered a postmodern policy domain, the citizens have increased their trust in S&T, even in the context of crisis, and there is a higher interest, priority and support for public spending in S&T, 16

17 that is not the case with other postmodern policy domains like the environmental policy, that could be perceived as subordinated to the needs of growth and jobs. References Agencia de Evaluación y Calidad (2011) La Administración Pública a juicio de los ciudadanos: satisfacción con los servicios, valoración del gasto, confianza en los empleados públicos y actitudes hacia la e administración. Madrid: AEVAL. Bauer, M.W. (2002) Controversial medical and agri food biotechnology: A cultivation analysis. Public Understanding of Science 11(2) Bauer, M.W. (2009) The Evolution of Public Understanding of Science Discourse and Comparative Evidence. Science, Technology & Society, 14(2) Bauer, M.W.; Allum, N.; Miller, S. (2007) What can we learn from 25 years of PUS survey research? Liberating and expanding the agenda. Public Understanding of Science 16 (1): Bauer, M.; Durant, J.; Evans, G. (1994). European Public Perceptions of Science. International Journal of Public Opinion Research, 6(2), Blekesaune, M.; Quadagno, J. (2003): Public Attitudes toward Welfare State Policies: A Comparative Analysis of 24 Nations, European Sociological Review 19 (5): Centro de Investigaciones Sociológicas (CIS) (2006) Percepción social de la Ciencia y la Tecnología Accessed on XXXX Fundación Española de Ciencia y Tecnología (2010) Percepción social de la Ciencia y la Tecnología. Accessed on XXXX Durant, J.; Bauer, M.; Gaskell, G.; Midden, C.; Liakopoulos, M.; Scholten, L. (2000). Two Cultures of Public Understanding of Science and Technology in Europe. Dierkes, Meinolf; von Grote, Claudia (eds) Between Understanding and Trust. The Public, Science and Technology. London: Routledge (reprint 2003), pp Eurobarometer (2010a) Europeans and the crisis. Eurobarometer Brussels: TNS Oppinion and Social Eurobarometer (2010b). Science and technology. Special Surobarometer 340. Wave Brussels: European Commission. Evans, G.; Durant, J. (1994). The relationships between knowledge and attitudes in the public understanding of science in Britain. Public Understanding of Science, 4 (1), Fraile, M.; Ferrer, M. (2005) Explaining the Determinants of Public Support for Cuts in Unemployment Benefits Spending across OECD Countries. International Sociology 20(4): Gaskell, G.; Bauer, M.W., Durant, M.; Allum, N.C. (1999) World apart? The reception of genetically modified foods in Europe and in U.S. Science, 285, Hayes, B.C.; Tariq, V.N. (2002) Gender differences in scientific knowledge and attitudes towards science: a comparatives study of four Anglo American nations. Public Understanding of Science 9, Inglehart, R. (1990) Culture shift in advanced industrial society, Princeton N. J.: Princeton University Press. Jacoby, W. G. (1994) Public Attitudes toward Government Spending, American Journal of Political Science, 38, 2, Lujan, J.L.; Todt, O (2000). Perceptions, Attitudes and ethical Valuations: the Ambivalence of the Public Image of Biotechnology in Spain. Public Understanding of Science 9: Miller, J.D. (1983a) Scientific Literacy: A Conceptual and Empirical Review. Daedalus 112(2) Miller. J.D. (1983b) The American people and science policy. The role of public attitudes in the policy process. New York: Pergamon Press 17

18 Miller, J.D. (1998). The measurement of civic scientific literacy. Public Understanding of Science, 7(3), Miller, J.D. (2004). Public Understanding of, and Attitudes toward, Scientific Research: What We Know and What We Need to Know. Public Understanding of Science, 13(3), Miller, J.D., Pardo, R. ; Niwa, F (1997) Public perceptions of science and technology. A comparative study of the European Union, the United States, Japan and Canada. Madrid: Fundación BBV Monroe, A.D. (1979) Consistency between Public Preferences and National Policy Decisions American Politics Research, 7, 1, Muñoz, A., Moreno, C.; Luján, J.L. (2012) Who is willing to pay for science? On the relationship public perceptions of science and the attitudes to public funding of science. Public Understanding of Science 21 (2): Nisbet M.C (2005) The Competition for Worldviews: Values, Information, and Public Support for Stem Cell Research. International Journal of Public Oppinion Research 17 (1): Nisbet, M. C.; Scheufele, D. A.; Shanahan, J.; Moy, P.; Brossard, D.; Lewenstein, B.V. (2002). Knowledge, Reservations, or Promise?: A Media Effects Model for Public Perceptions of Science and Technology. Communication Research, 29 (5), OECD (2009) Policy Responses to the Economic Crisis: Investing in Innovation for Long Term Growth. Paris: OECD, June 2009 Pardo, R.; Calvo, F. (2002). Attitudes toward science among the European public: a methodological analysis. Public Understanding of Science, 11(2), Pavone, Vincenzo; Osuna, Carmen; Degli Espositi, Sara (2011). Inverstir en ciencia y tecnologia en tiempos de austeridad economica: Qué opinan los ciudadanos. FECYT (2011) Percepción Social de la Ciencia y la Tecnología Madrid: FECYT, pp Quintanilla, M.A.; Escobar, M. (2005) Un indicador de cultura científica para las comunidades autónomas. FECYT (2005), Percepción Social de la Ciencia y la Tecnología en España Madrid: FECYT, pp Quintanilla, M.A.; Escobar, M.; Quiroz, K. (2011) La actitud global hacia la ciencia en las comunidades autónomas. FECYT (2011) Percepción Social de la Ciencia y la Tecnología Madrid: FECYT, pp Remler, D. K.; Van Ryzin, G. G. (2011). Research Methods in Practice. Los Angeles: SAGE Publications. Sanders, A. (1988) Rationality, self inters, and Public Attitudes on Public Spending. Social Sciences Quarterly 69(2) Sanz Menendez, L.; Cabello, C. (2000) "Expectations and learning as principles of shaping the future". Brown, N.; Rappert, B.; Webster, A. eds.(2000) Contested Futures: A Sociology of Prospective Science and Technology, Aldershot: Ashgate Press, pp Sanz Menéndez, Luis; Van Ryzin, Gregg G; del Pino, Eloisa (2012). Predictors of public support for government spending on science and technology. Submitted to Research Policy. Science (2012) Research cuts will cause Exodus from Spain. Science vol 336, 13 april 2012, pp Scruggs, L.; Benegal, S. (2012) Declining public concern about climate change: Can we blame the great recession? Global Environmental Change 22 (2): Sturgis, P.; Allum, N. (2004). Science in Society: Re Evaluating the Deficit Model of Public Attitudes. Public Understanding of Science, 13(1), Svallfors, S. (1997) Worlds of Welfare and Attitudes to Redistribution: A Comparison of Eight Western Nations, European Sociological Review 13: Svallfors, S. (1999) The Middle Class and Welfare State Retrenchment: Attitudes to Swedish Welfare Policies, Svallfors, S.; Taylor Gooby, P. (eds) The End of Welfare State? Responses to State Retrenchment, pp London: Routledge. Svallfors, S. (2003) Welfare Regimes and Welfare Opinions: A Comparison of Eight Western Countries, Sociol Indicators Research, 64,

19 Taylor Gooby, P. (2001) Sustaining State Welfare in Hard Times: Who Will Pay for the Bill?, Journal of European Social Policy 11: Torres Albero, C. (2005) La ambivalencia ante la ciencia y la tecnología. Revista Internacional de Sociología (RIS) 42: US NSF (2012). Science and Engineering Report 2012, Chapter 7: Science and Technology: Public Attitudes and Understanding. Washingtonh DC: US NSF van Lente, H.; Rip, A. (1998) The rise of membrane technology: from rhetorics to social reality, Social Studies of Science, 28 (2): Acknowledgements This research has been funded by the Spanish National R&D Plan (Ministry of Science and Innovation grant CS and Ministry of Economic Affairs and Competitiveness CS CSO ) ) and the 2011 IPP Short Term Visiting Fellows Program (IPP 2011 VF 03) that partially funded a short term visit of Gregg G. van Ryzin to Madrid. 19

20 Table 1. Unemployment by year and region Change Lower unemployment País Vasco Navarra Asturias Galicia Cantabria Castilla y León Rioja Aragón Higher unemployment Extremadura Madrid Cataluña Castilla La Mancha Balears Valenciana Andalucía Murcia Canarias National total

21 Table 2. Summary statistics Obs Min Max Mean SD Dependent variables General attitudes toward science Interest in S&T (1=interested, 0=not interested) Trust that benefits of S&T outweigh risks (1=yes, 0=no) Suppor for S&T policy S&T seen as a priority for public spending (1=yes, 0=no) Central gov't spends too little S&T (1=yes, 0=no) Regional gov't spend too little S&T (1=yes, 0=no) Independent (diff in diffs) variables Time (1=2010, 0=2006) Region (1=more unemployment, 0=less unemployment) Time x Region (interaction term) Control variables Female (1=yes/female, 0=no/male) Age (in years) Low education (less than secondary) High education (some post secondary) Practicing Catholic (1=yes, 0=no) Ideology (1=far left to 10=far right)

22 Table 3. Probit analysis of general attitudes toward S&T (and sports) Interest in S&T Trust that benefits of S&T outweigh risks Time *** *** * ** Region *** *** Time x Region *** *** *** *** Female *** *** Age *** *** Low education *** *** High education *** *** Practicing Catholic *** Ideology *** Pseudo R square Observed P Predicted P (at X bar) Obs (n) Note: Coefficients shown are marginal effects. *** p <.01 ** P <.05 * p <.10 22

23 Table 3. Probit analysis of support for public spending on S&T S&T chosen as a priority for public spending Central gov't spends too little on S&T Regional gov't spends too little on S&T Time * *** *** ** ** Region Time x Region *** *** *** * Female *** *** *** Age *** *** * Low education *** *** *** High education *** *** *** Practicing Catholic * *** *** Ideology *** *** *** Pseudo R square Observed P Predicted P (at X bar) Obs (n) Note: Coefficients shown are marginal effects. *** p <.01 ** P <.05 * p <.10 23

24 Figure 1. Trust that benefits of S&T outweigh risks (regression adjusted) Proportion trusting in benefits of S&T Lower unemployment regions Higher unemployment regions survey 2010 survey 24

25 Figure 2. Proportion choosing S&T as a government priority (regression adjusted) Proportion choosing S&T as priority Lower unemployment regions Higher unemployment regions survey 2010 survey 25