Fairness and Nanotechnology Concern
Abstract
Research suggests that fairness perceptions matter to people who are asked to evaluate the acceptability of risks or risk management. Two separate national random surveys (n = 305 and n = 529) addressed Americans’ concerns about and acceptance of nanotechnology risk management in the context of the degree to which they view scientists and risk managers as fair. The first survey investigated general views about scientists across four proposed dimensions of fairness (distributional, procedural, interpersonal, and informational). The results show that respondents who believe that the outcomes of scientific research tend to result in unequal benefits (distributional fairness) and that the procedures meant to protect the public from scientific research are biased (procedural fairness) were more concerned about nanotechnology. Believing scientists would treat them with respect (interpersonal fairness) and ensure access to information (informational fairness) were not significant predictors of concern. The second study also looked at these four dimensions of fairness but focused on perceptions of risk managers working for government, universities, and major companies. In addition to concern, it also examined acceptance of nanotechnology risk management. Study 2 results were similar to those of study 1 for concern; however, only perceived informational fairness consistently predicted acceptance of nanotechnology risk management. Overall, the study points to the value of considering fairness perceptions in the study of public perceptions of nanotechnology.
1. INTRODUCTION
Nanotechnology, like other emerging technologies, represents a unique subject around which to study the dynamics of risk perceptions as well as to provide practical guidance to those who wish to influence the pace of societies’ acceptance of novel technologies. Technology proponents often point to the detrimental impact that public pressure on political decisionmakers has had on useful innovation (e.g., nuclear power and agricultural biotechnology). Technology critics equally point to technology proponents’ tendency to bring new products to market without adequate health and environmental protection. In response, commentators from the social sciences, humanities, and within the science community have argued that the process of new technology development and diffusion often suffers from a dearth of democracy. This position argues that openness and respect for citizens’ concerns, manifested through activities such as upstream public engagement, can help the science community avoid public backlashes by ensuring that potential problems are addressed during technology development processes.(1–5)
This study focuses on the extent to which citizens believe that authorities involved in nanotechnology risk management, such as the scientists and risk managers, are fair when it comes to dealing with the public and allocating the benefits of the research. Moreover, it examines the proposition that these fairness perceptions will relate to citizens’ concerns about nanotechnology and acceptance of risk management. The focus on fairness reflects a position that the factors underlying fairness perceptions—which 30 years of social psychology research suggest includes dimensions related to distributive outcomes, procedures, interpersonal treatment, and information(6)—are precisely the outcomes risk communicators argue would come from meaningful efforts by decisionmakers to integrate the public's views into risk decision making.(7,8) Studying what citizens think about the fairness of decisionmakers involved in nanotechnology could therefore provide one means of assessing the degree to which current efforts at upstream engagement have an influence in the United States.
Below, we provide a brief overview of public opinion research that has specifically focused on nanotechnology as well as past research on trust and fairness. We then present the results of two U.S. national surveys that sought to assess the impact of fairness on concern about nanotechnology (studies 1 and 2) and acceptance of nanotechnology risk management (study 2). The final discussion section addresses the potential for future research on fairness perceptions in the context of emerging technologies.
2. BACKGROUND
2.1. Public Opinion about Nanotechnology
Nanotechnology is the term given to “the understanding and control of matter at the dimensions of roughly 1–100 nanometers where unique phenomena enable novel applications.”(9) Nanotechnology does not focus on a specific type of product but rather the scale of technology or process used to make that technology. Working at the nanoscale opens up opportunities for advances in key economic fields such as electronics, materials, and pharmaceuticals, as well as a range of consumer product areas such as cosmetics.(10) Government agencies responsible for science and economic development in many countries have identified nanotechnology as a key area of strategic importance and have devoted substantial resources toward basic and applied research.(11,12)
A number of surveys have investigated public opinion about nanotechnology in the United States and Europe in recent years.(13–25) This work suggests that most people know very little about nanotechnology but that there is, nevertheless, more optimism than pessimism about its likely impact. Much of this research has focused on the relationship between knowledge or awareness of nanotechnology and the relative balance of risks and benefits. A meta-analysis of this work suggests that a small such relationship exists.(26) The discussion accompanying the meta-analysis notes, however, the importance of other variables in explaining nanotechnology concern. Some of these variables include environmental concern(23,24,27) and religious conviction;(18,21,28) however, the dominant nonawareness explanation has been trust.
2.2. Trust and Fairness
This study questions the extent to which perceptions about scientists and risk managers influence concern about nanotechnology. It is grounded in research on trust and fairness associated with actors charged with managing risk. Trust has been an important area of risk communication research for some time and has played a central role in discussions of why people express concern and support regarding emerging technologies,(27–30) including nanotechnology.(13,16,25,26,31,32) Trust also figures centrally in organizational justice or fairness research, as the perceived trustworthiness of authorities can influence perceptions of fairness. This study draws on work from social psychology that emphasizes trust as one component of fairness.(33) The focus on fairness broadens the discussion from previous research examining trust in scientists to include perceptions of procedural fairness and allocation of benefits; it is grounded in the assumption that one may not trust the scientists but still consider the process or outcomes as fair.
As with trust, the influence of fairness has figured prominently in discussions of risk management and public participation,(34–41) as well as in public relations.(42) While the references to fairness is this work are common, none of these studies take advantage of the deep body of work on fairness from social psychology.(43) Arguably, the focus on weighing the risks versus benefits of nanotechnology(25) invokes questions of fairness regarding the allocation of outcomes. Similarly, the concept of distributional fairness relates to whether individuals believe they have received a fair outcome from a decision, whether in terms of receiving an equitable share based on input, an equal share to other affected parties, or enough of a share to meet basic needs.(44–46)
Rather than limiting our inquiry to the fairness of outcomes—given the uncertainty about potential outcomes-related nanotechnology—we suggest it may prove productive to assess the degree to which nonoutcome-oriented aspects of fairness affect people's attitudes toward nanotechnology. It is not that outcomes do not matter; rather, fairness research highlights that outcomes are but one component of a decision and that other aspects often play substantial roles in views about decisions and decisionmakers.
In particular, the fairness of the decision-making process can significantly influence the extent to which individuals subjected to the outcomes of the process are satisfied with authorities and outcomes.(47–49) The concept of procedural fairness thus focuses on the importance of unbiased and appropriately enacted procedures, including individuals’ ability to have a voice in a decision-making process that affects them.(47,49–52) Research has demonstrated that the nature of the process underlying a decision is often as, or more, important than a decision's distributional outcome,(6) although the extent to which the issues are viewed as having moral implications can mitigate the importance of procedural fairness.(53,54) Much of the research has focused on procedural fairness in the workplace, but a substantial body of work has also focused on political and risk management contexts with parallel findings.(36,40,55–61)
In addition to distributional and procedural fairness, researchers have suggested an interactional model of fairness that emphasizes the interpersonal treatment that individuals receive in the process.(62,63) Interactional fairness has been described as the “social side” of fairness and thought to be comprised of interpersonal and informational dimensions.(64) Interpersonal fairness concerns the degree to which individuals are treated with respect and dignity by those executing the procedures (here, scientists and risk managers).(48,49,65,66) Informational justice refers to the explanations conveyed along with procedures or in response to outcomes. Measurement focuses on whether individuals feel that authorities have provided appropriate explanations and disclosure during decision-making processes.(64,67,68) As Colquitt and colleagues(33) summarized in their extensive review, interpersonal fairness primarily addresses perceptions of treatment related to outcomes whereas informational fairness examines perceptions of explanations provided to help people evaluate procedures.
Recent work examined the relationship between the proposed four dimensions of fairness, using measures adapted from Colquitt,(67,69) to investigate the relationship between the perceived fairness of local scientists and community concern about nanotechnology.(18) The results found that the perceived distributional fairness of scientists’ research outcomes was related to nanotechnology concern in that respondents were less concerned when they perceived a more favorable distribution of outcomes. Informational fairness was also negatively related to nanotechnology concern, in that respondents were less concerned when they believed that local scientists would be timely and transparent about the risks associated with their research. All four dimensions of fairness were related to local residents’ belief that scientific research was appropriate for their community. Similar results have been found for perceptions of nuclear energy.(70)
Although the initial nanotechnology findings are suggestive of the role that fairness might play in nanotechnology concern, the localized setting of the study in a university community limited the ability to generalize the findings. The nuclear research was similarly grounded in a specific community. The current research sought to determine if these concepts would perform in a national setting, using a representative sample of U.S. adults. Based on previous research, we focus on two general hypotheses: (H1) the perceived fairness of authorities (scientists and risk managers) will relate to nanotechnology concern (studies 1 and 2); and (H2) the perceived fairness of risk managers will relate to acceptance for nanotechnology risk management (study 2). Acceptance is conceptualized as satisfaction with current risk management and its perceived legitimacy. The analyses also focus on a general research question (RQ1) exploring whether different dimensions of fairness have distinct impacts on concern and acceptance of risk management.
This study, as with most previous studies of emerging technologies, does not focus on the degree to which specific decision processes lead to acceptance of a specific decision (e.g., a siting decision). Rather, the focus is on the types of attitudes we expect will be important to people’s willingness to accept the continued societal use and development of an emerging technology. In this regard, much of the work on nanotechnology draws on concerns that failure to attend to developing concerns about agricultural biotechnology, in particular, has limited the ability of decisionmakers to take advantage of such technologies to solve real-world problems.(3) There is substantial conceptual overlap between discussion of local decision making and broader concerns about public engagement, but the difference between these two levels of analysis needs to be recognized. Although local-level behavior by risk managers will likely contribute to societal views, it should also be expected that social forces—particularly mediated exposure to risk and risk managers—will also be important.
3. STUDY 1 METHODS
We contracted with Knowledge Networks to conduct the online survey, which was fielded October 19–25, 2007. Knowledge Networks uses a probability-based design to obtain a panel of individuals that are representative of the U.S. population and has conducted a wide range of professional and academic surveys.(71) Once selected for a survey, individuals are sent a notification e-mail that a survey is available for them to take. Overall, 460 individuals were sampled, and 305 completed the survey, yielding a 66% completion rate.
An analysis of respondents’ demographic characteristics shows that 48% of respondents were male, and 73% were white. The median household income range was between $40,000–49,999, and the median age was 49 years old. In terms of the highest level of education, 29% had completed a bachelor's degree or higher, 29% had some college, and another 29% had received a high school diploma.
The survey asked respondents to evaluate the perceived fairness of scientists, in general, and the outcomes of scientific research. We chose scientists because they represent a key actor working with nanotechnology. Table I lists the items used to represent the four dimensions of fairness, as well as the item means, standard deviations, scale reliabilities, and correlations. These statements were included on a single page of the survey and presented in a random order to each respondent. Although the survey used similar measures to the earlier study,(18) we were unable to use the full number of items in the current analysis due to low reliability scores for the combined scales. In the case of procedural fairness, we only used one item to represent the dimension. We return to this limitation in Section 3.2.
| Mean | (SD) | |||
|---|---|---|---|---|
| Distributional | ||||
| I receive a fair share of the benefits of scientific research. | 3.49 | (0.89) | ||
| Scientific research helps more people than it hurts. | 3.65 | (0.92) | ||
| Scale mean (n = 253) | 7.13 | (1.50) | ||
| Cronbach's α | 0.54 | |||
| Procedural | ||||
| The procedures that protect the public from potential risks of scientific research are biased (n = 255) (reversed). | 3.03 | (.83) | ||
| Interpersonal | ||||
| If I were to speak with a scientist, he or she would treat me politely. | 3.50 | (0.79) | ||
| If I were to speak with a scientist, he or she would treat me with dignity. | 3.32 | (0.83) | ||
| Scale mean (n = 255) | ||||
| Cronbach's α | 0.77 | |||
| Informational | ||||
| Scientists are usually candid when they speak about potential impacts of their research. | 3.22 | (0.90) | ||
| There are procedures in place to ensure that people like me have accurate information about scientific research that might affect them. | 3.04 | (0.94) | ||
| Scale mean (n = 255) | ||||
| Cronbach's α | 0.47 | |||
| Correlations | (2) | (3) | (4) | |
| Distributive (1) | 0.15* | 0.39* | 0.44* | |
| Procedural (2) | 0.07** | 0.34* | ||
| Interpersonal (3) | 0.31* | |||
| Informational (4) | ||||
- Responses ranged from 1 = strongly disagree to 5 = strongly agree, reversed where noted.
- # p < 0.01 *p < 0.01 (one-tailed).
We measured the dependent variable, concern, with a scale that assessed citizens’ views about the risks and benefits of nanotechnology, consistent with much of the previous research.(26) The scale was comprised of three statements, randomly ordered, measured with 5-point responses: The potential risks of nanotechnology outweigh the potential benefits (M = 2.72, SD = 0.70); The use of nanotechnology poses a serious environmental hazard (M = 2.76, SD = 0.64); The use of nanotechnology poses a serious public health risk (M = 2.72, SD = 0.67) (Cronbach's α= 0.84).
For media attention, respondents assessed how much attention they paid to news about politics (M = 5.48, SD = 2.43), science (M = 6.07, SD = 2.11), and the environment (M = 6.08, SD = 2.37) on a scale ranging from 1 for “very little attention” to 10 for “very close attention.” We combined attention to news about science and the environment into one measure (M = 12.15, SD = 4.12, Cronbach's α= 0.81). Media attention is included given its explanatory role in previous research on emerging technologies as a measure of interest in science and technology.(29,72) Influential political science research sometimes treats such attention as a proxy for knowledge.(73) Media attention is also commonly addressed as a potential amplifier of risk in the social amplification of risk framework(74) but, in the current context, it is expected that the more likely role is to identify those respondents who have affinity toward technology.
3.1. Study 1 Results
Given the relatively normal distribution of our key variables and the linear relationship between these variables, we used ordinary least-squares (OLS) regression to test the hypothesized relationship between fairness and nanotechnology concern. The control variables were entered in an initial wave followed by the fairness variables to allow for the reporting of the relative impact of the fairness variables on the overall explained variance. The regression results (see Table II) include a significant, negative relationship between distributional fairness and concern, suggesting that respondents factored the perceived fairness of benefits into their opinions. The results also show a negative relationship between procedural fairness and concern. When individuals perceived that the procedures that protect them from potential risks of scientific research were biased, they were also more concerned about nanotechnology. Informational and interpersonal fairness are not significantly related to concern in the regression analysis, though informational fairness has a significant, zero-order correlation with concern. H1 is therefore supported and RQ1 points to the specific importance of distributive and procedural fairness.
| r | B | B | |
|---|---|---|---|
| Male | 0.07# | −0.07 | −0.06 |
| White | 0.12* | 0.07 | 0.10 |
| Education | −0.17** | −0.17** | −0.11* |
| Age | 0.07# | −0.01 | −0.00 |
| Income | −0.07# | 0.01 | 0.02 |
| Political news attention | 0.04** | 0.28** | 0.28** |
| Science news attention | −0.20** | −0.32** | −0.29** |
| Adjusted-r2(F-test) | 0.09** | ||
| Part-r | |||
| Distributive Fairness of Science | −0.21** | −0.15** | −0.17** |
| Procedural Fairness of Science | −0.15** | −0.14 | −0.14** |
| Scientists’ Interpersonal Fairness | 0.01 | 0.02 | −0.07 |
| Scientists’ Informational Fairness | −0.10* | −0.07 | 0.02 |
| Cum. Adjusted-r2 (F-test) | 0.12** | ||
- # p < 0.10; *p < 0.05; **p < 0.01 (one-tailed).
- n = 303 for concern; pairwise deletion.
With regard to the other measures, the results show that attention to science news had a significant, negative relationship with concern. That is, the more science news respondents reported consuming, the less concerned they were. This result differs from the earlier, community study, where no significant relationship between science news attention and nanotechnology concern was found. The results also showed a significant, positive relationship between attention to political news and concern, suggesting that the more political news respondents consumed, the more concerned they were about nanotechnology. This relationship occurred even though political news attention did not have a significant bivariate correlation with the concern, suggesting suppression.
The control variables explained about 9% of the variance in concern. The addition of the fairness measures increased the explained variance to 12%. The introduction of the fairness variables did not lead any of the control variables to lose their significance.
3.2. Study 1 Discussion
Following on previous research,(18) the results of study 1 provide further evidence that individuals base their concerns about nanotechnology not only on the allocation of benefits (distributional fairness) but also the perceived fairness of procedures regulating scientists’ behaviors (procedural fairness). In contrast with the results from the previous community-specific survey, informational fairness did not have a significant relationship with concern. In neither study did interpersonal fairness show a significant relationship with concern.
It may be that the measures, which asked respondents to assess how scientists, in general, would behave toward them, were simply too abstract. Although it is not uncommon for surveys to ask respondents to provide opinions on scientists in general,(75) an examination of the responses to the fairness questions shows that the most common response to the informational and interpersonal fairness items was neutral, ranging from 49% who would neither agree nor disagree that there are likely “procedures in place to ensure” access to information, to a high of 54% who would not agree or disagree that scientists would treat them with dignity if they were to speak to them. It may be that further efforts to contextualize the questionnaire might find different results.
Indeed, when viewing the results, it is important to consider the primary limitation, which is the low reliability for the fairness variables. Due to the low reliability, we were only able to use one item to measure procedural fairness and two variables for the other measures (the survey itself had three measures for each fairness dimension). The fact that the earlier community study used similar questions yet received higher reliabilities may be because the community study was conducted in two counties with large academic research presences.(18) In contrast, this study used a national sample; thus participants may have been less likely to have concrete views about what specific information they expected to receive from scientists. In this setting, it might therefore make sense for respondents to default to more general views about scientific bias. Different measures and a focus on a broader range of nonlocal actors may have shown different relationships. We examine this possibility in study 2.
Among other interesting findings was the opposite relationship that media attention had on nanotechnology concern depending on whether respondents reported paying attention to science news or political news. On the one hand, the relationship between science news attention and concern is not surprising, given research that shows that people who consume science media also tend to be more supportive of technology.(29) This is not to claim that science news contains more content that is pro-nanotechnology than the political pages. Indeed, research shows that media pays very little attention to nanotechnology and even less to the potential risks.(76,77) Given the lack of attention to nanotechnology in the news, the negative relationship between political news attention and concern may reflect less an opinion derived from negative content about nanotechnology than a negative overall view about scientific authority once controlling for the positive impact for those who pay attention to science news. This relationship is not, however, consistent with general research on media use and science suggesting that general public affairs media use is weakly associated with positive views about science(72) or nanotechnology-specific work that found no relationship.(13)
Finally, although the results did show more optimism than pessimism regarding nanotechnology's risks and benefits, which is consistent with previous research,(26) a closer look at the breakdown of responses shows that only about 30% of the respondents gave a directional response whereas a majority chose a neutral response. This finding likely reflects the general lack of awareness and knowledge about nanotechnology among the general population that previous research has identified.(78,79) Providing respondents with a short description could have resulted in more willingness to respond, with the caveat that information may influence responses in a particular direction.(78)
4. STUDY 2 METHODS
To address some of the limitations in study 1 and expand our inquiry to include acceptance of nanotechnology risk management, we again contracted with Knowledge Networks to conduct an online survey with a nationally representative U.S. sample. Data collection took place April 2–18, 2010. In total, 529 individuals responded, representing a 56% completion rate. Similar to the 2007 demographic characteristics, 48% of our respondents were male, 74% were white, and the median age was 49 years old. For education, about 27% had completed a bachelor's degree or higher, 31% had some college, and 30% had received a high school diploma. The median household income had increased slightly to $50,000–$59,999.
Although we modeled the data using hierarchical OLS procedures in a way that was similar to study 1, study 2 included several changes. First, prior to asking any questions, we included a brief description of nanotechnology for respondents to read. This description was adapted from a description provided in the 2008 General Social Survey(80) and read:
Nanotechnology works at the molecular level to build new structures, materials, and machines. Even relative to other new technologies, nanotechnology is still in an early stage of development. People have frequently noted that new technologies have produced both benefits and harmful results. Today, the United States is the recognized leader in nanotechnology research and development and is seeking to expand the use of nanotechnology in society. The questions below seek to obtain your views about risk management on this topic.
Second, we chose not to include measures for attention to political and science news in this study but rather a measure for familiarity with, and attention to, nanotechnology. Responses were measured using a 6-point, semantic differential (or “bipolar”) scale with higher scores indicating greater attention. Specifically, we asked respondents to estimate how familiar they were with nanotechnology (M = 2.74, SD = 1.45), how much they had talked to others about nanotechnology (M = 2.56, SD = 1.50), and how much they followed new developments on nanotechnology in the news (M = 2.92, SD = 1.46). The scale mean was 8.22 (SD = 3.98), with a Cronbach's α of 0.88.
Third, we expanded our measurement of nanotechnology concern. Thus, while asking respondents whether they believed (1) the risks outweigh the benefits (M = 3.04, SD = 1.21), (2) the environment will be endangered (M = 3.14, SD = 1.11), and (3) public health will be endangered (M = 3.07, SD = 1.17), we also asked whether they believed (4) the potential risks can be controlled (M = 3.16, SD = 1.18). Responses were measured on a 6-point, semantic differential scale with higher scores indicating greater concern. The statements were randomly ordered on a single page. The scale mean was 12.41 (SD = 4.05), with a Cronbach's α of 0.89.
Fourth, we included a measure representing acceptance of nanotechnology risk management to use as a dependent variable in a model that includes nanotechnology concern as an independent variable. Consistent with a previous study on nuclear power,(70) two questions were asked using 6-point semantic differential scales, with higher scores indicating greater acceptance. These asked whether respondents were satisfied with current risk management efforts for nanotechnology (M = 3.52, SD = 1.26), and whether they accepted the legitimacy of current risk management efforts (M = 3.58, SD = 1.22). These statements were randomly ordered. The scale mean was 7.12 (SD = 2.39), with a Cronbach's α of 0.93. It should be noted that the questions do not seek endorsement of nanotechnology risk management policies but, rather, the degree to which they view them as acceptable.
Finally, we revised the questions used to measure the four dimensions of fairness and also used semantic differential questions instead of Likert-type scales. To contextualize these questions, we asked about the perceived behavior of risk managers who worked for industry, universities, or the federal government. We chose these three entities as representing key actors in the research, development, and policy arenas for nanotechnology. We also reworked the measurement of distributional and procedural fairness so that it addressed the perceived behaviors of risk managers and therefore something over which they, as authorities, could exercise some control. Both study 1 and the previous study on which it is grounded(18) addressed these dimensions of fairness without specifically mentioning scientists.
For study 2, the inclusion of multiple actors (i.e., university, industry, federal government) made it necessary to focus each question on the specific actor's role in fostering fairness. Each actor was assessed on a separate page, and the order of these pages was randomly rotated to minimize order effects. As before, the order of the individual questions was also randomized. Table III provides the items, means, and standard deviations for the fairness items for each actor, as well as the correlations among the variables. Although our intent was to examine the four dimensions of fairness, the items were highly correlated for each risk manager. Although creating an overall composite measure of fairness for each actor (i.e., government decisionmaker fairness, etc.) would have been possible,1 we chose instead to look at models involving single items for each fairness dimension and create individual models for each unique actor. This approach reduced colinearity while highlighting potentially meaningful relationships. Because each fairness dimension was initially measured with two questions, we were able to run two models with unique fairness questions for each actor. The models presented later represent one of these potential models. We also, however, ran a series of additional models and indicate those places where the use of a different fairness question would lead to a different pattern of results. We will return to this issue in Section 4.2.
| I believe people who work for __ and manage risk associated with nanotechnology would … | Federal Government | Universities | Major Companies | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Mean | (SD) | Mean | (SD) | Mean | (SD) | ||||
| Distributive fairness: Allow/not allow nanotechnology to unfairly put some people at greater risk than others | 3.16 | (1.34) | 3.74 | (1.29) | 3.26 | 1.32 | |||
| Procedural fairness: Be unwilling/willing to listen to concerns | 3.31 | (1.41) | 3.74 | (1.31) | 3.30 | (1.35) | |||
| Interpersonal fairness: Lack respect/respect for respect people like me | 3.27 | (1.36) | 3.77 | (1.25) | 3.38 | (1.28) | |||
| Informational fairness: Be slow/quick to disclose potential risks | 2.98 | (1.35) | 3.55 | (1.34) | 3.04 | (1.35) | |||
| Correlations (All sig. p < 0.01) | (2) | (3) | (4) | (2) | (3) | (4) | (2) | (3) | (4) |
| Distributive (1) | 0.75 | 0.78 | 0.79 | 0.79 | 0.76 | 0.75 | 0.77 | 0.73 | 0.74 |
| Procedural (2) | 0.79 | 0.75 | 0.77 | 0.77 | 0.78 | 0.72 | |||
| Interpersonal (3) | 0.76 | 0.74 | 0.75 | ||||||
| Informational (4) | |||||||||
- Responses measured on a 6-point semantic differential scale with higher means indicating greater perceived fairness.
- N = 495.
4.1. Study 2 Results
The initial set of models for concern are similar to what study 1 found (Table IV). As before, both distributive and procedural fairness predict concern about nanotechnology. This was true whether looking at the fairness of government, university, or major company risk managers. Unlike the previous study, however, the attention variable was not significant after controlling for fairness. That attention lost significance once the fairness variables were entered into the model suggests that fairness may mediate the impact between attention to nanotechnology issues and concern about nanotechnology. Additional analyses (not shown) using Haye's SPSS macro for testing multiple mediation(81) suggest that the primary pathway for this mediation goes through both distributive and procedural fairness.(81) Also, unlike study 1, age and gender were both significant in the models for government and university actors, and income was significant for all actors. Being male and having a higher income was associated with lower concern. As before, more education was associated with less nanotechnology concern. Informational fairness was marginally significant in the model focused on university-level decisionmakers when a distributive fairness question asking about the fair distribution of benefits (rather than risks) was substituted into the model. For a second time, H1 is supported, and RQ1 points to the specific importance of distributive and procedural fairness.
| Block 1 | B | Government | University | Major Companies | |||
|---|---|---|---|---|---|---|---|
| r | B a | r | B a | r | B a | ||
| Male | −0.11** | −0.15** | −0.13** | −0.15** | −0.15** | −0.13** | −0.14** |
| White | 0.00 | −0.03 | −0.02 | −0.03 | −0.03 | −0.02 | −0.01 |
| Education | −0.10** | −0.17** | −0.09* | −0.17** | −0.17** | −0.09* | −0.11* |
| Age | −0.07* | −0.03 | −0.07* | −0.03 | −0.03 | −0.07* | −0.07# |
| Income | −0.18** | −0.23** | −0.16** | −0.23** | −0.23** | −0.16** | −0.16** |
| Attention | −0.10** | −0.12** | −0.02 | −0.12** | −0.12** | −0.02 | −0.02 |
| Adjusted-r2 (F-test) | 0.08** | ||||||
| Part.-r | Part.-r | Part.-r | |||||
| Distributive fairness (Un/fair risk distribution) | −0.28** | −0.13* | −0.37** | −0.19** | −0.26** | −0.11# | |
| Procedural fairness (Un/willing to listen)v2 | −0.28** | −0.14** | −0.38** | −0.21** | −0.28** | −0.18** | |
| Interpersonal fairness (Lack/respect for people) | −0.28** | −0.08 | −0.32** | 0.01 | −0.24** | −0.02 | |
| Informational fairness (Slow/quick to discl. risks) | −0.24** | 0.02 | −0.33** | −0.03b | −0.23** | −0.01 | |
| Cumulative adjusted-r2(F-test): | 0.18** | 0.22** | 0.15** | ||||
- # p < 0.10; *p < 0.05; **p < 0.01 (one-tailed).
- aBased on Block 2—Version 1.
- bInformational fairness reaches marginal significance (p < 0.10) if an alternative distributive fairness is used in the model; n = 495 for concern. Pairwise deletion.
The initial block of variables explained 8% of the variance, and the additional fairness variables added significantly to the model. The fairness variables explained the most in the university model and the least in the model for major companies. The significant partial-r for all of the fairness variables means that, had these been entered into the model without the other fairness variables, they would have been significant.
Whereas the key variables in the concern model were consistent across key actors (and between study 1 and study 2), there was more variability in the models focused on acceptance (legitimacy and satisfaction) of current nanotechnology risk management efforts (Table V). Both procedural and informational fairness were significant predictors of acceptance prior to entering the variable for concern into the model. Concern, however, appeared to mediate out procedural fairness in the model involving views about government and major companies, although the major company fairness relationship was somewhat unstable inasmuch as the use of alternative variables changed the pattern of results. Additional analyses(81) (not shown) suggest that procedural fairness is completely mediated when looking at government and major company risk managers and partially mediated when looking at university risk managers. The distributive fairness of government risk managers was the only case in which the expected fairness of outcome distribution impacted nanotechnology risk management policy acceptance. The interpersonal fairness of major companies emerged as marginally significant. Increased concern was associated with lower acceptance. H2 is therefore supported, and RQ1 points to the specific importance of both procedural and informational fairness.
| Block 1 | B | Government | University | Major Companies | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| r | B | B | r | B | B | r | B | B | ||
| Male | 0.01* | 0.09** | 0.06 | 0.00 | 0.09* | 0.05 | 0.00 | 0.09* | 0.06 | 0.01 |
| White | 0.07# | 0.07* | 0.11** | 0.10** | 0.07* | 0.06 | 0.06* | 0.07* | 0.10** | 0.10** |
| Education | 0.09* | 0.16** | 0.08* | 0.04 | 0.16** | 0.07* | 0.04 | 0.16** | 0.11** | 0.06* |
| Age | −0.02 | −0.07* | −0.02 | −0.05 | −0.07* | 0.04 | −0.00 | −0.07* | −0.01 | −0.04 |
| Income | 0.16** | 0.22** | 0.13** | 0.07 | 0.22** | 0.10** | 0.05 | 0.22** | 0.12** | 0.06# |
| Attention | 0.33** | 0.33** | 0.18** | 0.17** | 0.33** | 0.21** | 0.21** | 0.33** | 0.18** | 0.17** |
| Adjusted-r2(F-test) | 0.16** | |||||||||
| part.-ra | part.-ra | part.-ra | ||||||||
| Distributive fairness (Un/fair risk distribution) | 0.47** | 0.24** | 0.22** | 0.39** | −0.01 | 0.04 | 0.38** | 0.07 | 0.05b | |
| Procedural fairness (Un/willing to listen) | 0.41** | 0.11* | 0.05 | 0.48** | 0.30** | 0.18** | 0.41** | 0.14* | 0.05b | |
| Interpersonal fairness (Lack/respect for people) | 0.40** | −0.03 | −0.08 | 0.40** | −0.02 | −0.05 | 0.41** | 0.11# | 0.10#b | |
| Informational fairness (Slow/quick to discl. risks) | 0.46** | 0.21** | 0.21** | 0.47** | 0.27** | 0.23** | 0.43** | 0.21** | 0.19** | |
| Cum. adjusted-R2 (F-test) | 0.37** | 0.39** | 0.35** | |||||||
| Nanotechnology concern | −0.50** | −0.40** | −0.50** | −0.36** | −0.50** | −0.40** | ||||
| Cum. adjusted-R2 (F-test) | 0.51** | 0.49** | 0.50** | |||||||
- # p < 0.10; *p < 0.05; **p < 0.01 (one-tailed).
- aPartial-r controls for previous block(s).
- bIf alternate fairness variables are used, distributive and procedural fairness become significant while interpersonal fairness loses significance; n = 488 for acceptance (pairwise deletion).
With regard to control variables, being male, having a higher income, and paying more attention to nanotechnology news were all consistently associated with greater acceptance of nanotechnology risk management efforts. The initial block of control variables explained 16% of the variance while the addition of the fairness variable increased the explained variance up to between 35% and 37%. Adding overall concern increased explained variance to between 49% and 51%.
4.2. Study 2 Discussion
The results for the model focused on nanotechnology concern (Table IV) echo those of study 1 and show that both explicit outcomes, such as a distribution of risks, and procedural factors, such as perceived willingness of risk managers to listen, influence concern. The second model (Table V) focusing on the legitimacy and satisfaction with perceived efforts to manage nanotechnology risks further demonstrated that outcomes are not necessarily central to how people think about emerging technologies. Rather, for this dependent variable, respondents’ answers were primarily driven by how they perceived risk managers’ likely behavior in terms informational fairness (i.e., speed of disclosing risks) and, to a lesser extent, risk managers' willingness to listen. The significance of perceived distributive fairness of government risk managers may, however, reflect a belief that such actors have an added responsibility of ensuring fairness of outcomes. The high amount of intercorrelation between the fairness variables makes it inadvisable to make too much of these relationships, but it certainly seems plausible that the public has different expectations of different actors.
5. GENERAL DISCUSSION
Taken together, the two studies indicate that fairness can be a useful construct for understanding views about emerging technologies such as nanotechnology. This study does not seek to refute or challenge other models that put an emphasis on trust, credibility, values, or some related construct. Instead, it seeks to take advantage of three decades of social psychological research that argues that the need for fair treatment represents a fundamental human concern.(33,43,82) Further, even when trust between parties is lacking, a fair process or outcome may still lead to greater acceptance of decisions. While past measurement research has focused primarily on organizational settings, continued effort is needed to ensure high-quality measurement when studying risk decision making in community or national contexts. One of the limitations of this study is the challenges associated with translating previous measures into items that can reliably capture the four dimensions of perceived fairness.
Whereas the first study is consistent with past fairness research that emphasizes differences between distributive and nondistributive outcomes, the second study highlights differences in the perceived fairness of different risk managers and the influence of fairness on nanotechnology concern. It also brings the initial focus on nanotechnology concern into the policy arena by demonstrating that the perceived fairness of risk managers and the degree of concern have consequences for people's acceptance of nanotechnology risk management.
That interpersonal fairness failed to explain either concern or acceptance of nanotechnology risk management is interesting inasmuch as this concept emphasizes interpersonal treatment such as the degree to which risk managers respect and would be polite to respondents. It may be that nanotechnology risk managers seemed so remote from respondents’ experiences that such measures were less meaningful to them.
The order of the variables and the focus on concern as a dependent variable suggest a specific causal relationship that moves from views about risk managers toward concern and, in the end, acceptance of nanotechnology risk management. This ordering is based on the literature; however, it also makes sense that someone who is concerned about nanotechnology might come to view those who manage attendant risks as suspect. Such questions cannot be addressed with the current data. From a pragmatic standpoint, the current ordering also makes sense inasmuch as the factors underlying fairness perceptions reflect beliefs that risk managers can address through changes to behavior or effective communication such as by creating public engagement mechanisms that involve meaningful, widespread public discussion.(7,83)
In sum, the research provides further evidence that attention to nanotechnology or general science (a proxy here for potential knowledge) may only be a minor driver of nanotechnology concern(26) in comparison to perceptions about how risk managers are expected to interact with the public. The results could potentially reinforce arguments about the need to focus outreach efforts on meaningful public engagement and dialogue that builds relationships among risk managers and the public, rather than one-sided efforts designed to educate and inform the public about nanotechnology. The importance of procedural and informational fairness in these results also suggests the need to emphasize citizens’ potential to have a voice in societal discussion about nanotechnology and the need for experts to demonstrate their willingness to speak candidly about potential risks and benefits. Such results do not mean, however, that communication aimed at increasing awareness about nanotechnology is inappropriate. Greater awareness could lead to greater support, as the current data suggest and building such awareness could be part of outreach mechanism.(84,85) However, given the currently low levels of awareness of and familiarity with nanotechnology among members of the general public, it seems unlikely that this awareness will rapidly increase, except in the instance of a high-profile, positive, or negative, event that generates substantial media coverage. Should such an event or events occur, rather than becoming less important, the perceived fairness of risk managers charged with protecting public and environmental health from the potential risks of nanotechnology could become more salient to the public.
Footnotes
Acknowledgments
This research was supported by funds from the National Science Foundation under Grants 0551047, 0531160, and ECS-0335765; the Joint Institute for Food Safety and Applied Nutrition; and Cornell University Agricultural Experiment Station federal formula funds, Project No. NYC-131457, received from Cooperative State Research, Education and Extension Service, U.S. Department of Agriculture. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the funding organizations.
