LETTER
Citizen Science: Exploring the Potential of Natural Resource Monitoring Programs to Influence Environmental Attitudes and Behaviors Sarah K. Chase1,2 & Arielle Levine2 1 2
School of Environmental and Forest Sciences, University of Washington, Seattle, WA 98195–2100, USA Department of Geography, San Diego State University, San Diego, CA 92182–4493, USA
Keywords Behavior change; citizen science; conservation; diversity; environmental attitudes; natural resource monitoring; participatory monitoring; pro-environmental behavior; public participation; social networks. Correspondence Sarah K. Chase, School of Environmental and Forest Sciences Box 352100, University of Washington Seattle, WA 98195-2100. E-mail:
[email protected] Received 22 September 2016 Accepted 30 May 2017 Editor Phillip Levin
Abstract Citizen science programs monitoring ecosystems and natural resources are promoted for their potential to foster environmental awareness and stewardship. We surveyed volunteers in natural resource monitoring programs to determine whether they perceived changes in their environmental attitudes and decision-making. The majority of participants perceived changes in their attitude toward the resource being monitored, but not in their decision-making toward the resource they monitored or toward the environment more broadly. While the resources volunteers monitored in this study were diverse, program volunteers themselves were not. Participants were largely white, older, affluent, well-educated, held strong preexisting environmental attitudes, and were involved in other conservation, research, or management efforts. While engaging this narrow range of self-selected volunteers has the potential to reinforce existing pro-environmental attitudes through strengthening social networks, citizen science programs can increase their potential to promote attitude and behavioral change by making a concerted effort to engage a more diverse “citizenry.”
doi: 10.1111/conl.12382
Introduction Citizen science is increasingly recognized as an important and cost-effective means to obtain natural resource monitoring data to inform conservation policy. But data by itself have limited ability to influence conservation policy and outcomes, which are highly dependent on public attitudes and behavior (Toomey et al. 2016). Because citizen science directly engages the public in conservation-relevant research and experiences, it is also touted as having tremendous potential to reinforce pro-environmental attitudes (Gommerman & Monroe 2012) and foster increased environmental stewardship (McKinley et al. 2015). If such claims are warranted, citizen science could be an important tool for advancing environmental protection goals. As Schultz (2011) and others have argued, “conservation means behavior” and, ultimately, conservation goals can only be met by significant, but often elusive, attitude and behavioral changes at an individual level.
Citizen science programs often operate under the assumption that increasing participant knowledge of environmental issues fosters pro-environmental attitudes, which in turn increases tendencies toward proenvironmental behaviors (Jordan et al. 2011). Yet, this information “deficit” model of public understanding and action (Sturgis & Allum 2004) is flawed; knowledge and attitudes do not necessarily translate into action. Experiments have found that pro-environmental attitudes are frequently not correlated with pro-environmental behavior (Kollmuss & Agyeman 2002; Heberlein 2012), and increased environmental knowledge is not always associated with pro-environmental attitudes (Weaver 2002; Hunter & Rinner 2004). More recent models seek to explain this “gap” between knowledge, attitudes, and behavior through the inclusion of barriers and incentives (Toomey & Domroese 2013). For example, Kollmuss & Agyeman (2002) present a framework in which demographic factors (gender, age, education, etc.) interplay with external factors (institutional, economic, social, and
C 2017 The Authors. Conservation Letters published by Conservation Letters, March/April 2018, 11(2), 1–10 Copyright and Photocopying: 1 of 10 Wiley Periodicals, Inc. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
Citizen science: attitude and behavior change
cultural) and internal factors (motivation, environmental knowledge, values, attitudes, environmental awareness, emotional involvement, locus of control, responsibilities, and priorities) to create incentives or barriers for pro-environmental behavioral change. This complex relationship between knowledge, attitudes, and behavior indicates that citizen science programs may not necessarily promote pro-environmental attitudes or behaviors among participants. Indeed, systematic studies have not detected statistically significant attitude changes (Brossard et al. 2005; Crall et al. 2013), although Toomey & Domroese (2013) suggest that volunteers do perceive changes in their own attitudes and behaviors, and a range of other potential outcomes have been documented (Stepenuck & Green 2015). Participation in citizen science may encourage changes at the individual level, such as discussing environmental concerns with friends or changes in personal decisions, while involvement in collective management efforts is less common (Overdevest et al. 2004; Jordan et al. 2012). Participation in citizen science may promote increased knowledge about and concern for the resource being monitored and a deeper connection to nature (Haywood et al. 2016). The participatory nature of citizen science may also foster community collaboration and participation in management (McKinley et al. 2015), promote increased social capital, and expand social networks for local environmental action (Overdevest et al. 2004; Haywood 2015; Stepenuck & Green 2015). While the precise drivers behind engagement in natural resource management remain unclear, citizen science is frequently viewed as a potential gateway for engagement. Here, we explore whether participation in citizen science programs monitoring natural resources results in changes in environmental attitudes or behavior. We document how participants perceive changes in their own attitudes and decision-making as a result of participating in monitoring. Ultimately, we argue that a deeper understanding of who self-selects to volunteer in citizen science programs and the ability of these programs to truly influence attitudes and behavior are critical to maximizing the potential for citizen science monitoring to advance conservation goals.
S.K. Chase & A. Levine
monitored water quality in different regions of the United States (San Diego County, CA, Charleston, SC, and south-central WI). The size of volunteer pools ranged from 19 trained volunteers to more than 750, monitoring narrow to broad geographic areas (Table 1). This diversity in geography, size, scope, and type of resource monitored was purposefully intended to represent a wide range of citizen science programs.
Data collection and analysis Surveys were administered to the full available volunteer pool for each program, excluding participants under the age of 18 or whose participation had lapsed more than 5 years1 . The survey instrument included openended, 5-point Likert-type, and multiple choice questions focusing on motivation for participation, perceptions of change in attitudes and decision-making, engagement in natural resource research and management efforts, and demographic information. Both a mail and web-based survey instrument were developed and distributed following Dillman’s Tailored Design Method as closely as possible (Dillman et al. 2014), with care to minimize measurement error between the two formats. Survey distribution varied between programs based primarily on availability of volunteer contact information2 . A small incentive was provided to increase response rates. Nonparametric statistical tests were used to analyze the quantitative survey data, which were ordinal or nominal in nature and not normally distributed. We used the Mann–Whitney U test to determine statistical significance for ordinal variables and chi-squared test for nominal variables. Analyses were conducted using SPSS version 24 (IBM, Chicago, IL, USA). Open-ended responses were analyzed using an iterative, inductive approach (Thomas 2006), whereby themes of interest emerged from a systematic reading of the data. All codes were grouped into primary thematic categories and subcategories, representing the key themes that are presented here.
Results Volunteer demographics and background
Methods Case study programs We surveyed participants in eight citizen science programs that monitor natural resources (Table 1), using a purposeful sampling approach (Patton 2002). Six of these programs, monitoring diverse resources, were located in southern California (USA). Three of the programs
We received a total of 306 surveys, with a response rate of 31.2%3 . Across all programs, the demographic profile of volunteers was predominantly white (90%), evenly split between males and females, and tended to be older than average for the corresponding geographic region (Table 2). Half of respondents had an annual household income of $100,000 or more, and 40% had a graduate degree. Seventy-three percent reported volunteering with
C 2017 The Authors. Conservation Letters published by Conservation Letters, March/April 2018, 11(2), 1–10 Copyright and Photocopying: Wiley Periodicals, Inc.
2 of 10
Conservation Letters, March/April 2018, 11(2), 1–10 Wiley Periodicals, Inc.
Water quality
2008c
Rock River Basin (10 counties in south central WI) 15 sites in the Charleston area, SC
2002
1.6 km stretch of San Gabriel River, Long Beach, CA
2011
Coastal CA
San Diego County, CA
1999
Wildlife, focusing on selected key species Pacific green sea turtle (Chelonia mydas) 2005
San Diego County, CA
2003b
Flora biodiversity
Nearshore rocky reefs Water quality
San Diego County, CA
2000
San Diego County, CA
2014
Least tern (Sternula antillarum) Water quality
Geographic extent
Year established
x
x
140
19 (5)
x
x
x
x
x
x
Collect long-term datasets
745
75 (20)
340
144 (14)
250 (30)
29
Trained volunteers (core group of regularly active members)
x
x
x
x
x
x
x
Improve management and/or conservation efforts
b
x
x
x
x
x
x
Engage Support advocacy community or public efforts
Key program goalsa
Key program goals were determined based on an examination of program mission statements, as well as conversations with program managers. Formal project funding ended in 2008. c Volunteers were not included until 2013.
a
Charleston Waterkeeper
San Gabriel River Sea Turtle Monitoring Project Reef Check California Rock River Coalition
San Diego Audubon Ternwatchers San Diego Coastkeeper San Diego Plant Atlas Project San Diego Tracking Team
Program title
Resource of interest
Table 1 Summary of eight case-study citizen science programs
x
x
Increase education or awareness
x
x
Collect and store specimen samples
S.K. Chase & A. Levine Citizen science: attitude and behavior change
C 2017 The Authors. Conservation Letters published by Copyright and Photocopying:
3 of 10
Citizen science: attitude and behavior change
S.K. Chase & A. Levine
Table 2 Demographic characteristics for survey respondents. Results are presented for all programs, as well as broken down by geographic location (San Diego County, California, and relevant program counties in Wisconsin), compared to 2014 census estimates for the United States and the corresponding geographic areas (in parentheses)a Demographic factor SexA Male Female AgeB 18–34 years old 35–54 years old 55–74 years old 75 years old and above Highest level of education attainedC Less than a bachelor’s degree Bachelor’s degree Graduate degree Annual household incomeD $100,000 RaceE White alone Non-whitec
San Diego programs
California programs
Wisconsin programb
All programs
n = 96 43.8 (50.2) 56.2 (50.4) n = 98∗∗∗ 14.3 (35.4) 26.5 (34.8) 52.1 (22.7) 7.1 (7.1) n = 98∗∗∗ 12.2 (68.7) 44.9 (20.0) 42.9 (11.3) n = 54∗∗ 16.7 (18.9) 5.6 (21.0) 25.9 (29.9) 51.8 (30.2) n = 90∗∗∗ 87.8 (71.0) 12.2 (29.0)
n = 191 48.7 (49.7) 51.3 (50.3) n = 194∗∗∗ 22.7 (33.0) 33.5 (36.1) 40.2 (23.8) 3.6 (7.1) n = 193∗∗∗ 10.4 (72.1) 46.6 (18.1) 43.0 (9.8) n = 106∗∗∗ 16.0 (20.4) 4.7 (21.1) 20.8 (29.0) 58.5 (29.5) n = 174∗∗∗ 85.6 (62.1) 14.4 (37.9)
n = 77 44.2 (49.6) 55.8 (50.4) n = 77∗∗ 15.6 (29.2) 39.0 (36.0) 40.2 (26.6) 5.2 (8.2) n = 77∗∗∗ 13.0 (68.3) 51.9 (20.9) 35.1 (10.8) n = 41∗ 2.4 (17.2) 19.5 (23.0) 48.8 (33.9) 29.3 (25.8) n = 77∗ 97.4 (90.9) 2.6 (9.1)
n = 277 48.0 (49.2) 52.0 (50.8) n = 280∗∗∗ 21.4 (30.6) 35.0 (35.4) 39.6 (26.0) 4.0 (8.0) n = 279∗∗∗ 11.1 (73.3) 48.8 (17.2) 40.1 (9.5) n = 154∗∗∗ 13.0 (23.2) 9.7 (23.7) 28.0 (30.0) 49.3 (23.1) n = 260∗∗∗ 89.6 (73.8) 10.4 (26.2)
∗
P < 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001. 2014 estimates from the United States Census Bureau Current Population Survey. See Supplemental Material for chi-square results. b See Supplemental Material for a list of Wisconsin counties included. c Includes Asian or Pacific Islander, Black or African American, Latino, Native American or American Indian, and two or more races. No survey respondents identified as either Black or Native American. a
Table 3 Key volunteer characteristics for survey respondents, presented as percentages aggregated across all programs Volunteer characteristic Volunteered with conservation organization in the past year (not including the program) No other volunteer activities beyond data collection for the program Involved in formal research or management efforts (beyond data collection for the program) Involved in a professional capacity Involved in a volunteer capacity Volunteered with the program for at least an hour in the past month Volunteered with the program for at least an hour in the past year
All programs 65.7 14.5 39.6 21.9 19.1 49.3 72.9
the program in the past year, and 66% reported volunteering with other conservation organizations, as well (Table 3). Thirty-nine percent of respondents indicated that they were also currently engaged, either professionally or as volunteers, in formal research or management efforts beyond their role collecting data for the citizen science program.
Volunteer attitudes and decisions Analysis of the Likert-type data indicated that the majority (57%) of respondents agreed that their attitude toward the specific resource they monitored had changed since starting to volunteer with the program, and 43% agreed that their attitude toward the environment more generally had changed (Figure 1). There was a strong positive association between these two variables; those who perceived a change in their attitude toward the resource were likely to also perceive a change in their attitude toward the environment more broadly (X2 = 80.545, df = 1, P
