Understanding public concerns about pesticides: an empirical examination.
Were consumers justified in their reaction to the Alar report? Are pesticides and other agricultural chemicals a serious threat to human health and to the environment? The answers to these questions, as one might expect, vary considerably depending on the source. Some argue that the risks posed by synthetic agricultural chemicals are minor compared to the risks from naturally occurring chemicals. They contend that many naturally occurring chemicals in foods pose a more serious human health threat than do synthetic chemicals (see, e.g., Ames and Gold 1990; CAST 1987). Others contend that the food system is increasingly adulterated with potent chemicals and carcinogens that pose a serious threat to human health as well as to wildlife and the environment (see, e.g., Rodale 1981; Steinman 1990). In between these relatively extreme views is a wide continuum of opinion on the risk posed by pesticides to humans and the environment. Blair (1989) provides a good review of the controversy, covering the uncertainties involved and how they have translated into confusion and a diversity of opinions about the risks and benefits of agrichemical use.
Despite continuing disagreements over the degree of risk posed by pesticides and other agrichemicals, it appears that people have become increasingly concerned about pesticide and other agrichemical use over the past 20 years (Sachs, Blair, and Richter 1987). This increase in concern corresponds to the emergence of a more general societal concern with environmental quality and technological threats to that quality, the emergence of a growing health consciousness among the public, and growing distrust of government regulations aimed at protecting both the environment and human health (see, e.g., Dunlap 1991; Dunlap and Scarce 1991). As a result of these interacting trends, there seems to have been an increase in the degree to which modern life is viewed as "risky," although the authors know of no longitudinal data specifically documenting this trend. However, when asked the following question, "Thinking about the actual amount of risk facing our society, would you say that people are subject to more risk today than they were 20 years ago, less risk today, or about the same risk today as 20 years ago?" a 1979-1980 national survey found that 78 percent of the public believed that life is more risky today than it was 20 years ago (Marsh and McLennan 1980). Especially noteworthy is the fact that 56 percent of those who said life is more risky subsequently indicated that they felt that the "food supply" was one of the sources of increased risk compared to 20 years ago.
Despite the apparently escalating public concern over pesticide use and the significant implications of this concern for food consumption and the agricultural industry, relatively little research has been devoted to understanding the public's view of pesticides. After briefly reviewing available literature on the topic, the results of a recent survey conducted in three Pacific Northwest states to examine the public's view of pesticides are reported in some detail.
REVIEW OF EXISTING RESEARCH
Recent public opinion polls have provided some insight into the public's perception of pesticide issues. For example, a 1990 national survey sponsored by the American Farm Bureau Federation (AFBF) and conducted by Porter/Novelli (1990, 2) found that although four out of five (79 percent) agreed that "America's farmers are very concerned about the safety of the food they produce," some degree of concern over the use of pesticides was expressed by nine out of ten respondents (89 percent). The report also noted that, "While over half of the public indicated that they are 'very concerned' about both spoilage (58 percent) and pesticides (55 percent), when asked which they are most concerned about, pesticides was cited approximately three times more frequently than spoilage" (Porter/Novelli 1990, 5). In addition, 62 percent agreed that "the dangers to human health posed by pesticides outweigh their benefits in protecting crops from insect pests" (Porter/Novelli 1990, 5). The report continued by summarizing responses to a number of items with the statement that "Consumers are 'chemophobic.' That is, they are fearful, confused and concerned about the use and possible misuse of farm chemicals" (Porter/Novelli 1990, 6).
Although limited to one state, another study clearly suggested that concern over pesticides had increased substantially in recent years. In a 1984 survey of Pennsylvania residents, Sachs, Blair, and Richter (1987) included a number of items used in 1965 survey in that state by Bealer and Willits (1968). This trend study found that consumer concern over the risks of pesticide use increased significantly over the 19 years between surveys. The second survey revealed large increases in concern over the effects of pesticides on wildlife, consumers, and the farmers using them, with the result that 70 to 80 percent expressed concern ("some" or "a great deal") over each of these impacts in 1984. Sachs, Blair, and Richter also reported a substantial decline for the 19-year period in the proportion indicating agreement that "farmers are careful with pesticides," that "government adequately regulates chemical use in or on food," and that "foods purchased from retail stores are adequately inspected" (with less than half agreeing with the last two statements in 1984) (1987, 104).
While covering a shorter time period, an on-going nationwide study of consumers has similarly found that concerns about pesticide use are increasing. Although covering only 1984 to the present, the Food Marketing Institute (1990) has found that consumers consistently rank pesticides as the most serious hazard in food, with a low of 73 percent in 1985 and a high of 82 percent in 1989 indicating that pesticides are a "serious hazard" (the percentage for 1990 was 80).
The above studies indicate that the public is concerned about pesticide use and that pesticides are viewed as a significant risk in modern societies. The extent and nature of the risks posed by pesticides have been further clarified by several risk perception studies which compare the perceived risks of pesticides to those of a range of technologies and activities. For example, in a recent survey of residents of three northern California communities, 63 percent of respondents were "very concerned" over "pesticide residue in food," ranking it fourth in a list of ten potential hazards (Pilisuk, Parks, and Hawkes 1987). Slovic similarly found that the risks posed by pesticides are ranked relatively high (generally in the top third) among a wide range of 30 potential risks by both citizen groups and experts (1987). Slovic's data also provided insight into the relatively high risk rating of pesticides. Pesticides were found to have moderately high ratings on two underlying risk dimensions, "unknown risk" and "dread risk" (nuclear power and nuclear war rate high on these two dimensions), that are consistently found to produce high ratings of overall riskiness.(1) Similarly, an analysis of the dimensions underlying ratings of 93 technological hazards found pesticides to be among the group categorized as "multiple extreme hazards" (along with nuclear power), rating highly on the dimensions of "mortality" and "delayed impacts" (Hohenemser, Kates, and Slovic 1983).
In short, the public tends to view pesticides as constituting a relatively major risk, which is not surprising as pesticides are viewed as a threat that is not well understood; that has delayed, longterm, and potentially fatal consequences; and is therefore "dreaded." In view of these findings, it should come as no surprise to learn that the public supports efforts to reduce pesticide use in agriculture. For example, the AFBF survey found that only 19 percent agree that farm chemicals should continue to be used at current levels, while 66 percent agree that farmers should limit the amount of chemicals they use, and 15 percent agree that the use of farm chemicals should be banned (Porter/Novelli 1990). An even larger majority (89 percent) agree that "we should reduce the use of chemicals as much as we can because the less chemicals in our food the safer we are" (Porter/Novelli 1990, 11).
While the high level of public concern over pesticides appears to translate directly into support for increased governmental efforts to limit agrichemical use (Porter/Novelli 1990), it is unclear as to how much impact such concern is having on consumer behavior per se--outside of the extreme cases such as the Alar controversy and the similar publicity given to the poisoning of Chilean fruit. The AFBF survey, for example, found that consumers' pesticide concern had minimal impact on consumption as, "Two out of three consumers (64 percent) do not avoid any foods because they might be harmful to one's health" (Porter/Novelli 1990, 6). More to the point, perhaps, the Pennsylvania study found no relationship between level of personal concern about pesticide use and behaviors such as growing one's own fruits and vegetables, buying directly from farmers, and personal use of pesticides on fruits and vegetables (Sachs, Blair, and Richter 1987). This is not surprising because many people who express concern over pesticides do not have the option of growing their own food nor easy access to direct markets for farm produce.(2)
In contrast, trend studies by Cambridge Reports/Research International found high and rising levels of agreement with the following two statements about "changes in day-to-day behavior because of ... concern about the environment": (1) "Avoiding the purchase of certain kinds of packaged food products because of the chemical additives they contain" (from 52 percent in 1987 to 69 percent in 1990), and (2) "Avoiding the purchase of certain kinds of fresh food because of the chemicals used in food production" (from 32 percent in 1987 to 50 percent in 1990) (1990, 18).
The information reviewed suggests that pesticide use in agriculture is seen as a serious risk that elicits a fairly high level of public concern. Despite this perceived risk, these and other surveys indicate that much of the public still sees a role for pesticides in modern agriculture. The AFBF survey found that "Over two-thirds (69 percent) agree that chemicals increase farm productivity and that, without them, food would be less plentiful, have more blemishes and be more costly" (Porter/Novelli 1990, 9). Surveys in Kansas and Oregon likewise reported majorities agreeing that pesticides increase the availability of food; these studies also found considerable dissent concerning whether pesticides increase or decrease the price of food, cost of food production, and the quality and safety of food (Kansas State University 1983; Mason 1980). These results appear to reflect a fair amount of ambivalence in the public's overall view of pesticides--a relatively high level of concern with pesticide impacts, but less confidence that pesticide use in agriculture can be eliminated, and a consequent desire for the government to regulate pesticide use. Such results also suggest the importance of examining these various facets of attitudes toward pesticides in more detail than has previously been the case.
A related topic clearly deserving more attention is examination of the personal characteristics associated with differing attitudes toward pesticides. The studies have generally found only modest variation in pesticide concern across various segments of the public (e.g., Sachs, Blair, and Richter 1987), although most have found women to be somewhat more concerned than men (Blair and Sachs 1988; Mason 1980; Pilisuk, Parks, and Hawkes 1987; Porter/Novelli 1990) and younger adults to be a bit more concerned than their older counterparts (Kansas State University 1983; Mason 1980). The relationship between education and concern over pesticide use is less clear. Both Pilisuk, Parks, and Hawkes (1987) and Porter/Novelli (1990) found a slight negative relationship between education and concern over pesticides, but Blair and Sachs (1988) reported finding a slight positive relationship. Similarly, no consistent evidence could be found for the possible relationships between income or residence and concern over pesticides. The consistent gender differences are compatible with other studies showing that women tend to be more concerned about a range of technological risks than men (Gould et al. 1988; Pilisuk, Parks, and Hawkes 1987). In order to understand the sources of attitudes toward pesticides, more attention clearly needs to be given to possible variation in these attitudes among differing segments of the public.
The primary purpose of the present study was to examine public attitudes toward pesticides in some detail. A secondary purpose was to investigate the utility of major demographic characteristics--sex, age, education, income, and residence--for predicting attitudes toward pesticides.
Data for this survey were collected as part of a "Northwest Tri-State Program for Pesticide Analysis" being developed by Oregon State University, University of Idaho, and Washington State University. The sample for the states of Oregon, Idaho, and Washington was provided by Survey Sampling, Inc. of Westport, CT, a major source of samples for social and marketing research. It consisted of randomly drawn telephone numbers that represent the numbers (listed and unlisted) for the various telephone exchanges in the three states.
Data were collected via telephone interviews conducted during March 1990. Up to five calls were placed to each number, and interviews were requested with the household member 18 years of age or older who had the most recent birthday. Of the 1,200 listings in the sample, 435 complete and two partially complete interviews were conducted. The cooperation rate for this study (the ratio of the number of completed interviews to the number of completions plus refusals) was 64.9 percent or 437 out of 673. The completion rate (the ratio of the number of completed interviews to the total number of potential respondents) was 57.5 percent or 437 out of 760.(3)
The final sample of 437 consisted of 54 Idaho residents, 129 Oregon residents, and 254 Washington residents--numbers that are close to the expected proportions based on state populations. Data on the demographic characteristics of the total sample, as well as the three state subsamples, are given in Table 1. While data from the 1990 census were not available for comparisons, the sample will no doubt be found to be somewhat above average in education and income, as is true of most survey samples. It should nonetheless provide a reasonably accurate cross section of the three states and is certainly adequate for the major goals of examining the nature and social correlates of public concerns about pesticides.(4)
The Measuring Instrument
The interview questionnaire included 15 items designed to tap a wide range of pesticide issues. While the safety and necessity of pesticide use in agriculture and other contexts were the primary foci, items dealing with nonagricultural uses and the roles of government and the food industry vis-a-vis pesticides were also included. In addition, information was obtained on the five demographic variables noted.
First, the percentage distributions for responses to the 15 items organized according to the three major sections of the interview are described. Then the consistency in responses to the items is noted, indicating the appropriateness of combining all 15 into a single measurement with reasonably good internal consistency. The results of a principal components analysis, which revealed the emergence of two distinct and readily interpretable factors or dimensions and a third somewhat less clearly interpretable dimension among the attitudes, are also reported. Finally, the results of regression analyses are presented, in which both the entire set of pesticide items as well as the three factors were regressed on the five demographic variables (singly and in combination) to examine demographic correlates of pesticide attitudes.
TABLE 1 Demographic Characteristics of Total Sample and State Subsamples(a) Total Washington Oregon Idaho Sample Size 437 254 129 54 Gender Female 55.5% 55.5% 55.6% 55.6% Male 44.5 44.5 44.4 44.4 Age Under 25 10.7 9.4 12.1 15.1 24-35 27.1 28.8 23.4 26.4 36-50 34.6 35.4 39.7 30.2 51-65 18.1 18.5 18.5 15.1 Over 65 9.5 7.9 11.3 13.2 Education Less than High School 6.5 5.9 9.0 3.7 High School Graduate 30.2 33.9 23.0 29.6 Some College 38.0 36.6 39.4 40.7 College Graduate 25.3 23.6 28.7 26.0 Income $15,000 or Less 19.1 15.6 26.5 18.8 $15,001 to 25,000 18.6 16.9 23.0 16.7 $25,001 to 50,000 45.2 47.6 39.0 48.0 Over $50,000 17.1 20.0 11.5 16.7 Residence City or Suburb 40.3 43.9 42.5 18.5 Small Town/City 41.9 37.5 38.6 70.4 Farm/Country 17.7 18.6 18.9 11.1 a Percentages may not equal 100 due to rounding.
Table 2 reports the distributions of responses to the three general questions regarding pesticides asked at the outset of the interview. The first dealt with the safety of pesticides for the human food supply. Even with the addition of the caveat that the pesticides "are used according to approved directions," 45 percent indicated that they believe such pesticide use is unsafe ("somewhat" or "very"). The second question (which contained the same caveat) focused on the safety of pesticides for "the environment," with a substantially higher proportion of the sample (65 percent) believing they are unsafe in this regard. That the sample sees pesticides as posing a relatively greater threat to the environment than to humans is consistent with the findings in Pennsylvania (Bealer and Willits 1968; Sachs, Blair, and Richter 1987). The third question, which focused specifically on possible pesticide contamination of groundwater (perhaps the most salient environmental threat posed by pesticides), elicited a still higher level of concern. Nearly four out of five (79 percent) were at least "somewhat concerned" about water contamination by pesticides. Overall, responses to these three general items indicate a moderately high level of public concern over the safety of pesticides for the food supply, a higher level of concern over their safety for their environment, and a very high level of concern over pesticide contamination of water supplies--an issue involving both environmental and human health considerations--even when pesticides "are used according to approved directions."
TABLE 2 Percentage Distributions for General Pesticides Safety Items(a) Q-1. First I'd like you to tell me how safe you believe pesticides are for our food supply when they are used according to approved directions. Do you believe they are ... 1. VERY SAFE 9% 2. SOMEWHAT SAFE 45 3. SOMEWHAT UNSAFE 30 4. VERY UNSAFE 15 5. Don't Know 2 N = 436 Q-2. Next, when pesticides are used according to approved directions, how safe do you think pesticides are for the environment. Do you think they are ... 1. VERY SAFE 5% 2. SOMEWHAT SAFE 28 3. SOMEWHAT UNSAFE 41 4. VERY UNSAFE 24 5. Don't Know 3 N = 434 Q-3. Some people feel that the residues from pesticides and other agricultural chemicals are contaminating the groundwater and affecting the quality of water supplies. Others do not. I'd like to know the degree to which you are concerned or not concerned that your community's water supply has been or will be contaminated from the residues of pesticides and other agricultural chemicals. Are you ... 1. VERY CONCERNED 41% 2. SOMEWHAT CONCERNED 38 3. SOMEWHAT UNCONCERNED 14 4. VERY UNCONCERNED 6 5. Don't Know 1 N = 437 a Percentages may not equal 100 due to rounding.
The respondents were next asked whether they agreed or disagreed with the wide range of items listed in Table 3, covering the necessity and risk of pesticide use in food production, confidence in farmers' use of pesticides, and attitudes toward government regulation of pesticides. Seventy-two percent agreed that "Pesticide residues in food are a big health risk" (Q-4), a figure comparable to that reported by Sachs, Blair, and Richter (1987). While this is much higher than the 45 percent who indicated that they believe pesticides to be unsafe for the food supply (Q-1), the difference may stem from the caveat about pesticides being "used according to approved directions" attached to the first question. An even greater majority (84 percent) agreed with Q-6 that "Food should be grown with fewer pesticides." However, 55 percent also agreed with Q-8 that "Pesticides are necessary for the growing of food." Taken together, responses to these items seem to indicate that the public sees pesticides as posing risks and would like to see their use curtailed, but nonetheless sees some degree of use as necessary--results compatible with those of the AFBF survey (Porter/Novelli 1990).
Two other items dealt with perceptions about farmers and food processors. While a sizable majority (64 percent) believed that "Farmers and food processors are starting to reduce their use of pesticides" (Q-7), a smaller majority (55 percent) nonetheless disagreed that "Farmers make sure that the food I buy is safe to eat" (Q-9). These responses suggest that while much of the public sees farmers (and food processors) as beginning to reduce pesticide use in response to mounting public concern, only a small portion have a lot TABULAR DATA OMITTED of faith that farmers are insuring food safety (note that only 12 percent "strongly agree" with Q-9).
Given these responses, it is not surprising that a large majority (85 percent) agreed with Q-5 that "Our government should be doing more to regulate the use of pesticides," while two-thirds (66 percent) disagreed with Q-10 that "There are too many government regulations on the use of pesticides." Clearly a majority of the public sees government as having the responsibility of protecting them from pesticides, a position that is consistent with their fear of pesticides and their skepticism about farmers insuring food safety.
The items in Table 4 take a different tack, focusing on the perceived necessity of pesticides for a variety of purposes in addition to food production. The first two, Q-11 and Q-12, focused on food production. The first reveals a sizable majority (74 percent) indicating that pesticides are necessary ("very" or "somewhat") to protect food from insects and disease, while a smaller majority (53 percent) believes that they are necessary to control weeds that reduce crop production. The second most necessary use for pesticides in the eyes of the public, behind controlling insects and diseases in food production, is to control pests in institutions (Q-14), as 69 percent indicate they are necessary for this purpose. Controlling pests in the home (Q-15) is seen as necessary use by 65 percent of the public, while their use for protecting ornamental plants (Q-13) is seen as necessary by only 37 percent of the public. Thus, while controlling insects and diseases in food production is seen as the most necessary use of pesticides, the related use of weed control in crop production comes in fourth behind institutional and home uses. Overall these results indicate that a majority of the public does see the necessity of some pesticide use in food production, thus complementing the results in Q-8.
TABULAR DATA OMITTED
Internal Consistency and Dimensionality Among the Items
Analyses were conducted to determine the appropriateness of creating one or more multi-item measures from the 15 individual items. First, the corrected item-total correlations for each item and a measure of internal consistency (coefficient alpha, the mean of all possible split-half reliabilities) for all 15 items were obtained (SPSS 1988). All of the items were found to have significant item-total correlations, ranging from a low of .21 for Q-7 to a high of .63 for Q-2, with an average of .50. As a result, it is not surprising that a summated rating (or Likert) scale comprised of scores on all 15 items exhibits acceptable internal consistency as the coefficient alpha of .86 indicates. Consequently, it appears appropriate to combine all items into an "Attitude Toward Pesticides Scale," hereafter termed the "Total Scale," where high scores represent an "anti-pesticide use" orientation and low scores a "pro-pesticide use" orientation.(5)
Because the items covered such an array of issues and due to the exploratory nature of the study, a principal components analysis was also conducted in order to determine if distinct factors or dimensions emerged (SPSS 1988). The results of this analysis, employing a varimax rotation, are reported in Table 5. They indicate the presence of two clearly distinct and readily interpretable dimensions and a third somewhat less clear dimension.
All of the items from Table 4, which dealt with the perceived necessity of pesticide use in various settings, along with Q-8, which focused on the necessity of pesticides in food production, load heavily on the first factor and generally do not have high cross-loadings on the other two factors. These items all appear to reflect attitudes toward the "necessity" of pesticide use. Six of the other items, including the first three general items from Table 2 and the first three from Table 3, load most heavily on the second dimension. They appear to reflect a concern with the "safety" of pesticide use. The remaining three items load most heavily on the third dimension, TABULAR DATA OMITTED although Q-9 loads almost identically on the first and third dimensions and heavily on the second dimension as well. Given the nature of these items, it seems plausible to term this a "trust" dimension (even though the most obvious indicator of "trust," Q-9, loads heavily on both of the other dimensions).
The three dimensions yielded by the principal components analysis--particularly the first two--make substantive sense and appear to have obvious policy relevance. Three additional summated-rating scales were constructed: (1) the "Necessity of Pesticide Use Scale" (or "Necessity Scale") consisting of the items in factor one; (2) the "Safety of Pesticide Use Scale" (or "Safety Scale") consisting of the items in factor two; and (3) the "Trust in Food Industry Scale" (or "Trust Scale") consisting of the two items clearly constituting factor three. Because the coding for the items is the same as the Total Scale (all 15 items), high scores on these subscales represent an "anti-pesticide" orientation--expressing concern over the safety of pesticides, not seeing pesticide use as necessary, and not trusting the food industry's use of pesticides. Examining the interrelations among these three dimensions of pesticide attitudes and especially examining possible variations across the dimensions in terms of demographic correlates should provide needed insight into the nature and sources of public attitudes toward pesticides.
Correlations Among the Dimensions
The correlations among the Total Scale and the three subscales created via the principal components analysis are shown n Table 6. As expected, the Total Scale scores generally correlate highly with scores on the subscales: .83 with the Necessity Scale and .87 with the Safety Scale, but .51 with the short Trust Scale. In contrast, the Necessity Scale and Safety Scale are only moderately correlated (.51), not surprising given that they were produced by a varimax rotation which attempted to create orthogonal dimensions. Finally, the Trust Scale is only modestly correlated with the Safety Scale (.34) and the Necessity Scale (.25). In short, the principal components analysis appears to have produced three reasonably distinct dimensions, as the subscales used to measure them are at most only moderately correlated with one another.
The final analyses involved regressing scores on the Total Scale and three subscales on the five demographic variables included in the survey. Table 7 reports the pertinent results, showing both the bivariate correlations between each scale and each demographic variable (the top number in each cell) as well as the results of multiple regression analyses in which scores on each scale were regressed on all five variables simultaneously. The standardized regression coefficients or betas (the bottom, italicized number in each cell) indicate the effect of each demographic variable while controlling for the effects of the other four. In general, the two coefficients tend to be similar in magnitude and significance, indicating that the effects of the individual demographic variables are not altered when the effects of the other four are taken into account.
TABLE 6 Correlation Matrix for the Pesticide Scales(a) Total Necessity Safety Trust Scale Scale Scale Scale Total Scale 1.00 .83 .87 .51 Necessity Scale 1.00 .51 .25 Safety Scale 1.00 .34 Trust Scale 1.00 a All scales are coded so that high scores represent "anti-pesticide" orientations.
Comparing the coefficients for each demographic variable across the four scales yields some interesting results. First, the weak but significant TABULAR DATA OMITTED relationship between sex and pesticide attitudes, as measured by the Total Scale, appears to be due almost entirely to women's substantially higher scores on the "safety" dimension, as no relationship is found between sex and either the "necessity" or "trust" dimensions. Second, the modest relationship between age and overall attitudes, with the younger being more anti-pesticide in orientation, is heightened on the trust dimension, diminished slightly on the safety dimension, and reduced to insignificance on the necessity dimension. The modest positive relationship for education, indicating that an anti-pesticide orientation increases with education, is due to somewhat stronger relationships between education and both necessity and trust, and disappears for the safety dimension. In the case of income, there is no relationship with overall attitudes, but significant relationships do emerge for safety and trust. Whereas those with higher income levels have lower levels of concern about pesticide safety, they also express more distrust in the food industry. Finally, residence is not significantly related to any of the pesticide attitude dimensions.
To summarize, sex, age, and education are modestly but significantly correlated (at both the zero-order and partial levels) to overall pesticide attitudes, with women, younger adults, and the well-educated being a bit more anti-pesticide than their counterparts. When the safety of pesticides is examined, however, women are significantly more likely to be concerned about safety, younger adults are slightly more concerned about safety, and income rather than education is related to safety concerns (with those in the lower income levels being slightly more concerned than their higher income counterparts). A very different pattern is found for the perceived necessity of pesticide use, as only education is found to be significantly related, with the more educated being more likely to see pesticide use as necessary. Finally, sex is not related to level of trust in the food industry to protect consumers from pesticides; however, younger respondents, the well-educated, and those in higher income levels are more likely than their counterparts to express distrust.
What these results indicate is that the demographic correlates vary somewhat depending upon which aspect of attitudes toward pesticides is being examined. While women are significantly more concerned than men about the safety of pesticides, they are not necessarily any less likely to view pesticide use as unnecessary nor to have less trust in the food industry. In other words, a search for the social roots of attitudes toward pesticides must take into consideration the various facets of such attitudes. Two very distinct facets, safety and necessity of pesticide use, and a possible third facet, trust in the food industry, are suggested by this study.
Of course, it should be acknowledged that, as indicated by the results of the multiple regressions, the examination of the demographic correlates of pesticide attitudes is limited in its success. As can be seen in the bottom two rows of Table 7, the five demographic variables combined produce very modest multiple correlation coefficients, ranging from a low of .20 for the Necessity Scale to a high of .28 for the Trust Scale. Consequently, the five variables together explain very little variation in pesticide attitudes, regardless of the dimension examined. These results are consistent with those of prior studies (e.g., Sachs, Blair, and Richter 1987) and emphasize that attitudes toward pesticides do not vary substantially across the major demographic sectors of society. They also suggest the necessity of looking elsewhere in order to fully understand the roots of differing perceptions of pesticides.
SUMMARY AND CONCLUSIONS
In sum, despite its exploratory nature, this study reveals several interesting findings that deserve further investigation. In particular, the results indicate that it clearly makes a difference how one measures public concern about pesticides. Not only is concern over pesticide safety rather distinct from perceptions of the necessity of pesticide use (and, less clearly, both of these from trust in the food industry's use of pesticides), but these differing dimensions are related to major demographic characteristics in varying (and admittedly limited) degrees. These results suggest that future research on the public's view of pesticides should carefully formulate sets of items designed to tap the dimensions of safety, necessity, and especially trust in the food industry and views of government regulations (the latter two might emerge as separate dimensions if measured more fully than in the present study). In addition, distinguishing more adequately between the safety (or conversely the risk) of pesticides for humans versus the natural environment (especially but not exclusively wildlife) might be fruitful. The growing attention given to such notions as "environmental ethics" and "animal rights" suggests that perception of pesticides may be based primarily on perceived threat to human well-being for some people, whereas for others it may stem more from perceived threat to wildlife and the environment (see, e.g., Nash 1989).
It would also be fruitful to examine pesticides from a more explicit "risk perception" perspective, whereby pesticides are evaluated on a number of dimensions known to influence perceived "riskiness" relative to a range of other modern technologies (see, e.g., Gould et al. 1988; Hohenemser, Kates, and Slovic 1983; Pilisuk, Parks, and Hawkes 1987; Slovic 1987). Research examining public concern about pesticides relative to concern over other technologies could provide a better sense of the degree of public unease about pesticides, and would be particularly valuable from a policy standpoint.
In addition, the limited amount of risk perception research that has been conducted on reasonably representative samples of the public has found, as this and other studies have found for pesticides, that basic demographic characteristics are poor predictors of the perceived riskiness of a wide range of technologies. Such research suggests that in order to understand the public's attitudes toward various technologies, especially the degree to which government regulation of the technologies is favored (perhaps the most crucial policy issues), one must examine phenomena such as the perceived risks relative to the perceived benefits of each technology (see especially Gould et al. 1989, chapter 5). The distinction found in the present study between the perceived safety of pesticides and the perceived necessity of pesticide use may well reflect an inherent weighing of costs and benefits that underlie public attitudes toward pesticides.
In conclusion, while the present study provides insights into public attitudes toward pesticides, more research needs to be conducted if a good understanding of how the public views this widespread--and widely debated--agricultural practice is to be developed. Developing a better understanding of the public's view of pesticides may enable both government and the agricultural industry to implement policies and practices which are more compatible with public preferences and thus avoid future debacles of the magnitude surrounding consumer rejection of apples due to the Alar scare.
Riley E. Dunlap is Professor, Departments of Sociology and Rural Sociology, and Curtis E. Beus is Research Assistant, Department of Rural Sociology, Washington State University, Pullman.
1 Characteristics of "unknown risk" include unobservable, new, and delayed consequences; those associated with "dread risk" include fatal consequences, catastrophic potential, and lack of control (Slovic 1987).
2 Such lack of readily available opportunities to take actions consistent with one's attitudes are among the many factors that typically dampen attitude-behavior relationships. For a good review of this issue, with specific reference to environmental attitudes and behaviors, see Weigel (1985).
3 The other 440 numbers were distributed as follows: nonworking or wrong numbers, 281; ineligible (e.g., businesses), 85; rang but not answered, 52; and other, 22. For more detailed information on response rates, see Social and Economic Sciences Research Center (1990).
4 While the results of univariate analyses, such as frequency distributions, are easily influenced by sampling error (and especially question wording), the relationships found between variables are generally much more robust (see, e.g., Schuman and Scott 1987).
5 All of the items were coded on a five-point scale, where "1" equals the most "pro-pesticide use" attitude and "5" the most "anti-pesticide use" attitude. For Q-1 and Q-2, "Very Safe" = 1 and "Very Unsafe" = 5, with "Don't Know" = 3, while Q-3 was coded in the reverse direction. For Q-4, Q-5, Q-6, and Q-10, "Strongly Agree" = 5, "Strongly Disagree" = 1, and "Don't Know" = 3, while Q-7, Q-8, and Q-9 were coded the opposite. Finally, for Q-11 through Q-15, "Very Necessary" = 1, "Very Unnecessary" = 5, and "Don't Know" = 3. Missing data were coded the same as "Don't Know."
Ames, B. N. and L. S. (1990), "Too Many Rodent Carcinogens: Mitogenesis Increases Mutagenesis," Science, 249 (August 31): 970-971.
Bealer, R. C. and F. K. Willits (1968), "Worriers and Non-Worriers Among Consumers About Farmers' Use of Pesticides," The Journal of Consumer Affairs, 2: 189-204.
Blair, D. (1989), "Uncertainties in Pesticide Risk Estimation and Consumer Concern," Nutrition Today (November/December): 13-19.
Blair, D. and C. Sachs (1988), "Public Concerns About Pesticides and the Safety of the Food Supply," in Global Perspectives on Agroecology and Sustainable Agricultural Systems: Proceedings of the Sixth International Conference of the International Federation of Organic Agriculture Movements, Volume 1, P. Allen and D. Van Dusen (eds.), Santa Cruz, CA: Agroecology Program, University of California, Santa Cruz: 187-191.
Cambridge Reports/Research International (1990), The Green Revolution and the Changing American Consumer, Cambridge, MA: Cambridge Reports/Research International.
CAST (1987), "Health Issues Related to Chemicals in the Environment: A Scientific Perspective," Comments from CAST, 1987-1, Ames, IA: Council for Agricultural Science and Technology.
Dunlap, R. E. (1991), "Trends in Public Opinion Toward Environmental Issues: 1965-1990," Society and Natural Resources, 4: 285-312.
Dunlap, R. E. and R. Scarce (1991), "The Polls--A Report: Environmental Problems and Protection," Public Opinion Quarterly, 55: 713-734.
Food Marketing Institute (1990), Trends: Consumer Attitudes & the Supermarket, Washington, DC: Research Department of the Food Marketing Institute.
Gould, L. C., G. T. Gardner, D. R. DeLuca, A. R. Tiemann, L. W. Doob, and J. A. J. Stolwijk (1988), Perceptions of Technological Risks and Benefits, New York: Russell Sage Foundation.
Hohenemser, C., R. W. Kates, and P. Slovic (1983), "The Nature of Technological Hazards," Science, 220 (April 20): 378-384.
Kansas State University (1983), "Consumer Food Safety Perceptions," a report by the Cooperative Extension Service, Kansas State University, Manhattan.
Kelley, B. and S. Yates (1989), "Alar Under Attack ... Growers Mount Damage Control," Capitol Press (March 17): 1-2.
Marsh and McLennan (1980), Risk in a Complex Society: A Marsh and McLennan Public Opinion Survey, New York: Marsh and McLennan.
Marshall, E. (1991), "A Is for Apple, Alar, and ... Alarmist," Science, 254 (October 4): 20-22.
Mason, R. (1980), "Public Opinion in Oregon About the Use of Chemicals on Food Crops," Special Report 588, Agricultural Experiment Station, Oregon State University, Corvallis.
Nash, Roderick Frazier (1989), The Rights of Nature: A History of Environmental Ethics, Madison: University of Wisconsin Press.
Pilisuk, M., S. H. Parks, and G. Hawkes (1987), "Public Perception of Technological Risk," The Social Science Journal, 24: 403-413.
Porter/Novelli (1990), A National Survey of Americans' Attitudes Toward Farmers, Farm Practices and Food Safety, Washington, DC: Porter/Novelli.
Rodale, R. (1981), Our Next Frontier, Emmaus, PA: Rodale Press.
Sachs, C., D. Blair, and C. Richter (1987), "Consumer Pesticide Concerns: A 1965 and 1984 Comparison," The Journal of Consumer Affairs, 21: 97-107.
Schuman, H. and J. Scott (1987), "Problems in the Use of Survey Questions to Measure Public Opinion," Science, 236: 957-959.
Slovic, P. (1987), "Perception of Risk," Science, 236 (April 17): 280-285.
Social and Economic Sciences Research Center (1990), "Public Opinion on the Use of Pesticides: Results from a Telephone Survey of Residents of Idaho, Oregon, and Washington," SESRC Data Report 90-29, Pullman: Washington State University.
SPSS (1988), SPSS-X User's Guide, 3rd Edition, Chicago: SPSS Inc.
Steinman, D. (1990), Diet for a Poisoned Planet: How to Choose Safe Foods for You and Your Family, New York: Harmony Books.
Weigel, R. H. (1985), "Ecological Attitudes and Behavior," in Ecological Beliefs and Behavior, D. B. Gray, Westport, CT: Greenwood Press: 57-85.
Yates, S. (1989), "20 Minutes on '60 Minutes' Cost Growers $100 Million," Capitol Press (May 12): 1-2.
|Printer friendly Cite/link Email Feedback|
|Author:||Dunlap, Riley E.; Beus, Curtis E.|
|Publication:||Journal of Consumer Affairs|
|Date:||Dec 22, 1992|
|Previous Article:||Personal factors related to consumer product disposal tendencies.|
|Next Article:||Consumer views of the need for government intervention in the airline market.|