COLLECTIVE ACTION AT THE SHORELINE:
A COMPARISON OF LAKE MANAGEMENT ORGANIZATIONS
IN VILAS COUNTY, WISCONSIN, U.S.A.
––––––––––––––––
A Scientific Paper Presented to
The Faculty of the School of Sustainability
Arizona State University
––––––––––––––––
In Partial Fulfillment
of the Requirements for the Degree of
Master of Science
––––––––––––––––
by
Dane Whittaker
April 24, 2020

1

ABSTRACT
Community-based volunteer organizations are critical to natural resource management in
the United States. However, due to volunteer involvement, these organizations struggle with
collective action problems: coping with free riding, solving commitment problems, arranging for
the supply of new institutions, and monitoring individual compliance with sets of rules. In this
study, we explore how volunteer organizations can overcome these challenges. To explore how
they overcome these challenges, we use the Institutional Analysis and Development framework
and the Institutional Design Principles. These frameworks help us understand the impact of
natural resource conditions, community attributes, and the rules in use impact volunteer
organizations. For this research, we focused on lake organizations in Wisconsin. We collected
our data through semi-structured interviews with thirty-one lake organizations and public data.
The data were analyzed using constant comparison and linear regression, followed by qualitative
comparative analysis (QCA). We reinforce the importance of considering the system holistically
when managing a resource—the natural resource conditions, the community attributes, and the
rules in use. Our study shows the importance of graduated sanctions and low-cost conflict
resolution on social-ecological system outcomes. Volunteer-based resource management are an
effective way to tailor management strategies to the natural resource condition and the
community attributes.
TABLE OF CONTENTS
1. Introduction
1.1. Overview
1.2. Wisconsin Lakes and Lake Organizations
2. Methods & Data
2.1. Case Selection
2.2. Data Collection
2.3. The Outcomes: Seven Lake Organization Goals
2.4. The Conditions: Environmental, Social, and Institutional
2.5. Analytical Approach: Linear Regression Analysis
2.6. Analytical Approach: Crisp-Set Qualitative Comparative Analysis (csQCA)
3. Results & Discussion
3.1. Basic Regression Analysis
3.2. Necessary Conditions
3.3. Sufficient Conditions
4. Conclusion
5. Acknowledgments
6. References
7. Appendices

2

1. INTRODUCTION
1.1 Overview
The majority of natural resource management responsibilities in the United States belong
to federal and state agencies. These federal and state agencies, however, are increasingly relying
on volunteer-based resource management groups to achieve conservation and restoration goals
(Armitage, 2005; Bruyere & Rappe, 2007). Volunteer-based resource management encourages
participation of communities and resource users in decision-making through the incorporation of
local knowledge and institutions in management, regulatory, and enforcement practices for more
sustainable resource management outcomes (Armitage, 2005). Local participation and
knowledge accumulation occur in volunteer organizations comprised of people who live around
or rely on a resource for their livelihoods (Ostrom, 1990). When changes threaten a community’s
natural resource, resource users often respond by forming volunteer organizations to address the
challenge (Gabriel & Lancaster, 2004; Korth & Klessig, 1990; Ostrom, 1990). These
organizations emerge from the community with a goal for the resource in mind; they design their
rules based on the cultural and biophysical context (Ostrom, 2009). These volunteer-based
organizations play a critical role in resource management as federal and state agency budgets
shrink, and resource usage increases.
The budgets of federal resource management agencies have either remained constant or
declined for the past several decades (Bruyere & Rappe, 2007). Although budgets shrank,
outdoor recreation participation has increased, creating a gap in maintenance and services.
Agencies rely on volunteers to fill this gap (Bruyere & Rappe, 2007). In 2012, the US Fish and
Wildlife Service reported that 2.2 million volunteer hours, equivalent to 1,036 full-time
employees, supplemented their 9,000-employee workforce by over ten percent (US Fish &
Wildlife Service, 2013). The contributions from volunteers go beyond saving money for natural
resource managers. Volunteers are deeply committed to the resources they help manage. They
provide services like observing changes in natural resources and providing environmental
education. As agencies recognize that “one size fits all” solutions are a poor match to the
complexity of social-ecological systems (SES), volunteer-based organizations can help adapt
rules to the needs of the resource and the community using the resource (Ostrom, 1990).
Volunteer involvement in resource management comes with challenges. Volunteer-based
organizations struggle with collective action problems such as: “coping with free-riding, solving
commitment problems, arranging for the supply of new institutions, and monitoring individual
compliance with sets of rules” (Ostrom, 1990). With these challenges—and no formal authority
to overcome them—volunteer-based organizations struggle to meet their goals. Goals themselves
present a challenge to community-based natural resource management organizations. People’s
different uses of natural resources result in diversity in goals for resource management
(Gabriel & Lancaster, 2004; Peterson et al., 2003). Collective action problems and variety of
use are common issues faced by communities managing shared resources.
There is a reasonable understanding of how a community of resource users overcomes
the collective action problems experienced by volunteer-based organizations. Ostrom and her
colleagues identified eight institutional design principles (IDPs), through a systematic case study
3

review, that are associated with the persistence of community-based resource management
(Ostrom, 1990). The design principles are: 1) clearly defined boundaries, 2) congruence between
appropriation and provision rules and local conditions, 3) collective-choice arrangements, 4)
monitoring, 5) graduated sanctions, 6) conflict-resolution mechanisms, 7) minimal recognition of
rights to organize, and 8) nested enterprises (Ostrom, 1990). Follow up studies support the IDPs
with minor adaptations (Baggio et al., 2016; Cox, Arnold, & Villamayor Tomas, 2010). We
separate the first design principle 1) clearly defined boundaries into 1A) user boundaries—the
boundaries between who can and cannot use the resource—and 1B) resource boundaries—clear
boundaries defining the resource system as suggested by Agrawal and Cox et al. (Agrawal, 2002;
Cox et al., 2010).
Ostrom and her colleagues proposed the Institutional Analysis and Development
framework (IADF) to analyze the outcomes of repetitive interactions by diverse people. The
IADF is used to explore how institutions—the rules humans use to structure repetitive
interactions—persist, dissolve, and evolve (Ostrom, 2005). According to the IADF, people with
diverse interests interact in a social space; the context shapes the interactions and social space
(Ostrom, 2005). The external contexts that influence the social interactions include the rules in
use, the attributes of the biophysical world, and the structure of the community where the
interactions occur (Ostrom, 2005). These three types of factors affect the actions that people can
take, the benefits and costs of the actions, and the potential outcomes. We call these
environmental, social, and institutional conditions and situate the eight institutional design
principles in the IADF as the institutional conditions or rules in use.
In this study, we explore how the combinations of external factors and rules in use lead
to different outcomes in lake SESs. To do this, we collect primary data about the goals and
contextual factors—using the institutional design principles and the IADF—for volunteerbased organizations managing lakes in a small geographical area. Most studies of the
institutional design principles rely on secondary case study analysis, which presents data
completeness and variable consistency challenges (Ratajczyk, 2016); there are few examples
of studies that collect primary data (Agrawal & Chhatre, 2006; Shin et al., 2019). We collected
our data during the summer of 2019 in Vilas County in the Northern Highland Lakes District of
Wisconsin, U.S.A., which is home to more than 1,300 lakes and 115 lake organizations
(University of Wisconsin - Stevens Point, 2019; Vilas County Tourism and Publicity, 2020).
Vilas County’s rare geology and geography—a high concentration of lakes and organizations
in a small area—enable the study of institutional arrangements and their impact on various
social and environmental outcomes for lake SESs.
This study of lake organizations in Wisconsin resulted in insights about the conditions
of lake SESs that experience desired social and environmental outcomes. Like Agrawal &
Chhatre and Shin et al., we found the eight institutional design principles played a key role in
these outcomes—even as the desired outcomes varied by the organization. Further, we saw
that there are multiple causal pathways to success, which would mean that the design of
management institutions, or rules in use, should consider the environmental and social
conditions in contrast to uniform policies. In this study, we ask lake organizations about the
rules they use and combine the rules in use data with ecological and social data sets. We first

4

provide an overview of lakes in Wisconsin and the multi-level management structure before
reviewing the methods, results, and findings of our study.
1.2 Wisconsin Lakes and Lake Organizations
1.2.1 Freshwater lakes in Wisconsin
Wisconsin is home to 15,000+ freshwater lakes (Wisconsin DNR, 2019). Most of the
lakes are in the Northern Highland Lakes District, which has an area of 5330 km2 and includes
Vilas, Oneida, and portions of Forest, Iron, and Price counties. The area has been repeatedly
glaciated, leaving a flat and lake-filled landscape when the glaciers retreated 12,000 years ago.
These lakes range in area from small, temporary lakes to well over 1000 hectares and range in
depth from one to more than thirty meters (Peterson et al., 2003).
Freshwater lakes play an important role in SESs. They provide permanent and migratory
habitat for terrestrial and aquatic biodiversity. They offer recreational opportunities, including
but not limited to fishing, skiing, paddling, swimming, and wildlife watching. Lakes are also
central to economies created by lakefront property, tourism, and the freshwater recreational
fishing industry. The recreational fishing industry was estimated to be worth $29.9 billion in
2016 (U.S. Department of the Interior, U.S. Fish and Wildlife Service, U.S. Department of
Commerce, & U.S. Census Bureau, 2018).
Despite their important role, changes in recent years are threatening lake SESs. Climate
change is changing temperature and precipitation patterns, thus affecting lake productivity and
water level (Magee, McIntyre, Hanson, & Wu, 2019). Fish population and biomass changes have
been observed in northern temperate lakes as a result of warmer water (Embke et al., 2019).
Land-use change around lakes is also increasing. Lakefront property owners are removing the
natural buffers created by trees and grasses in favor of agriculture, houses, and lawns. These
landscape-scale land-use changes are affecting nutrient cycling and soil erosion, which impact
water clarity (Jennings, Emmons, Hatzenbeler, Edwards, & Bozek, 2003; Johnston & Shmagin,
2006). Boaters moving from one lake to another often carry water, animals, and plants between
lakes, increasing the risks of aquatic invasive species like zebra mussels, spiny water fleas, and
Eurasian Watermilfoil. These aquatic invasive species are hard to eradicate once they are
introduced and can permanently alter the biodiversity and community structure within lakes
(Rothlisberger, Chadderton, McNulty, & Lodge, 2010).
There are also political and cultural threats. Regulatory changes related to resource use
and zoning impact lakes. In 2015, Wisconsin updated their zoning codes so that townships and
counties could not set more restrictive zoning ordinances than those set by the state legislature.
This change in regulation impacted the ability of local government to tailor their codes to protect
shorelines and fish habitats (Wisconsin Shoreland Alliance, 2019). New technology can also
have effects on culture and resource use. Boats used for wake surfing can cause shoreline erosion
in small lakes through the wakes they create and the repetitive paths they take in the lake.
Additionally, rental property platforms like Airbnb and HomeAway attract new lake users that
use lakes intensely and are unfamiliar with the cultural norms and ecology of the lakes.

5

The Wisconsin legislature has encouraged a collaborative natural resource management
strategy to address these numerous changes. The collaboration includes the Wisconsin
Department of Natural Resources, the University of Wisconsin system, local and county
governments, non-profits, and volunteer-based organizations.
1.2.2 Wisconsin’s public waters & their stewards
The public trust doctrine informs Wisconsin’s navigable waters management; Article IX
of the Wisconsin Constitution ensures public access and free use of the waters. In Wisconsin’s
interpretation of the public trust doctrine, Wisconsin’s navigable waters must be maintained for
particular types of uses. Wisconsin Statute 33 charges the Wisconsin Department of Natural
Resources (WI DNR) with protecting and maintaining navigable waters (Wisconsin State
Legislature, 1974).
The statute requires the WI DNR to form a multi-level collaboration that includes
resource users, the WI DNR, the University of Wisconsin System (UW), other government
agencies, and public and private organizations (Wisconsin State Legislature, 1974). This
collaboration is known as the Wisconsin Lakes Partnership and includes the over 600 lake
organizations, WI DNR, UW-Extension service, and many state, county, and local partners
(Wisconsin DNR, 2019). The UW-Extension Lakes Program facilitates the Wisconsin Lakes
Partnership through knowledge dissemination and by serving as an interface between residents
(i.e., lake organizations), government agencies (i.e., WI DNR), and non-profits (Mandal &
Lawrence, 2017). In this way, Statute 33 encourages lake users—those most invested in the
resource—to form organizations, a necessary condition for successful common-pool resource
governance (Mandal & Lawrence, 2017; Ostrom, 1990).
Lake users form lake organizations, which have a variety of goals, including preventing
or treating aquatic invasive species, maintaining or enhancing their fishery, protecting water
quality, and member education (Gabriel & Lancaster, 2004). As they mature, their goals may
evolve; one lake organization had a goal to establish leadership continuity as the executive board
ages. Lake organizations are one of two types: lake associations or lake districts. Lake
associations are voluntary organizations made of citizens that range from social organizations to
incorporated non-profit organizations. Lake associations have no regulatory power over lake or
land use activities; they use informal influence and volunteer time and money to contribute to the
management of lakes (Gabriel & Lancaster, 2004; Wisconsin Lakes Partnership, 2018). Lake
associations undertake significant management activities in Wisconsin in partnership with the
WI DNR and other formal management groups (Gabriel & Lancaster, 2004; Korth & Klessig,
1990). “Protecting environmental and water quality” is the primary goal and motivation for the
formation of lake associations (Gabriel & Lancaster, 2004). Lake districts are specialized units of
government designed to manage a lake or group of lakes. They can tax property in the district to
levy funds for lake protection and rehabilitation. Lake districts have statutory responsibilities to
the resource, local citizens, and taxpayers (Wisconsin Lakes Partnership, 2018). Lake districts
manage projects that require a larger budget and can own public infrastructure or expensive
equipment like weed harvesters and lake aerators (Gabriel & Lancaster, 2004).

6

The popularity of lake organizations has changed over time, as seen in Figure 1. The
most notable event was the establishment of the Wisconsin Lakes Partnership in the 1970s. After
the partnership was established, more than four hundred lake organizations were formed.

Figure 1. The number of lake organizations founded by decade through 2010.
Source: UW-Extension Lakes Program.

To achieve their goals, lake organizations use a combination of grants, partnerships,
donations, and volunteer hours (Gabriel & Lancaster, 2004). The WI DNR has several types of
grants for lake studies, treatments, enhancements, and monitoring (Wisconsin DNR, 2019).
There are specific criteria that define a qualified lake association or district—including public
access to the lake—to receive a grant (Wisconsin Lakes Partnership, 2018). The grants also
require specific permits and partnerships. The funding, permits, and partnerships are how the WI
DNR fulfills the administrative requirements of Wisconsin Statute 33. In addition to the
resources available from the state, lake organization members contribute in the form of donations
and volunteer hours. These contributions are required for lake organizations to receive grants
from the WI DNR. The grants require a portion of the costs in-kind from the organization. Some
grants require a volunteer hour match. Lake organizations meet their goals through a
combination of state oversight and participation of those most affected.
Though it seems straightforward, garnering participation by those most affected is
challenging. Collective action problems are common in community-based natural resource
management groups like lake organizations. Like many volunteer-based organizations, a small
number of highly committed individuals do most of the work. Lake organizations must design
7

institutions that overcome these challenges and support the enforcement of rules without being
perceived negatively by their neighbors. These challenges are exacerbated in settings where
people only spend a portion of their time like Wisconsin’s inland lakes where over half of all
Vilas County lakefront housing is used “for seasonal, recreational, or occasional use” (Stedman,
2006). When these part-time residents are visiting their lake house, they want to relax. They do
not want to contribute to management (Stedman, 2006).
This study explores the pathways to desired outcomes in lake SESs and how they
compare. The remainder of the paper explains the unique dataset of thirty-one lake organizations
collected in Vilas County during the summer of 2019, the pathways to different outcomes, and
the implications and further research opportunities for volunteer-based resource management
organizations.
2. METHODS & DATA
We conducted semi-structured interviews to collect data about thirty-one lake
organizations and thirty-nine lakes in Vilas County during the summer of 2019. We
supplemented the data we collected with data published by multiple sources. These sources
included the WI DNR, UW-Extension lakes program, United States Geological Survey (USGS),
the North Temperate Lakes US Long-Term Ecological Research Network (NTL LTER), and the
Jones Lab at the University of Notre Dame. We analyzed the data using linear regression and
crisp-set qualitative comparative analysis to identify the conditions that led to the outcomes for
the lake SESs. We used constant comparison to analyze the goals mentioned in the summer of
2019 interviews. The conditions we used are the result of summer 2019 interviews and a review
of articles about lake SESs.
2.1 Case Selection
The lakes and organizations in this study are in Vilas County, Wisconsin (Figure 2A).
Vilas County is part of the territory that the Chippewa (Ojibwe) ceded to the United States in
1837 and 1842 (Riera et al., 2001). The southwest part of Vilas County is a reservation whose
lakes and land are governed by the Lac du Flambeau band of Lake Superior Chippewa Indians.
We chose Vilas County for our study because of its geography and geology.
Geographically it is very far North making the growing season short. Additionally, the soil is
poor. These factors combined make the area uninteresting to the commercial agriculture industry.
The absence of agriculture and the predominance of tourism results in support for conserving the
lakes without much consideration of competing uses. It is the site of many studies like research
by the NTL LTER, the National Ecological Observatory Network (NEON), the University of
Notre Dame Environmental Research Center, and the University of Wisconsin Center for
Limnology Trout Lake Station.
An interesting geological feature is the number of lakes. Vilas County is home to more
lakes than any other county; it has 1,320 of Wisconsin’s 15,000 lakes (Stedman, 2006). In
addition to the 1,300+ lakes in Vilas County, there are roughly 115 lake organizations. With a

8

high concentration of lakes and organizations, Vilas County was an attractive study site for the
research questions we pursued.
A)

B)

Figure 2. Our sample lakes in Vilas County, Wisconsin. A) Vilas County is in the Northern part of
Wisconsin on the border of Michigan. B) The sixty-two Vilas County lakes outlined in blue were eligible
for our study. The lakes filled in blue are the thirty-nine lakes managed by the thirty-one organizations we
interviewed. Source: County Boundaries 24K and 24K Hydro Waterbodies (Open Water) published on
dnrmaps.wi.gov.

We selected lakes and lake organizations (Figure 2B) using three criteria. First, we
selected lakes with public access. Lakes with public access have a boat ramp or landing where
non-residents can access the lake for recreation, fishing, and other uses. Lakes that have public
access are faced with greater collective action problems because there is potential for over-use by
non-residents who are less susceptible to resulting negative effects. Second, we included lakes
with lake organizations that manage three or fewer lakes to study organizations that manage
similar system complexity. For example, we exclude the Eagle River Chain of Lakes
Association, which manages ten lakes and has 829 members (ERCOLA, 2020). Finally, we
selected lakes that are managed by the WI DNR. After applying these filters, there were fifty-two
organizations and sixty-two lakes eligible for our study.
2.2 Data Collection
We interviewed thirty-one of the fifty-two eligible organizations. The thirty-one
organizations manage a total of thirty-nine lakes. Most organizations, in our study, manage one
lake; five organizations manage more than one lake. To schedule the interviews, we contacted
the president or primary contact listed on the UW-Extension Lakes Program website, listed on
lake organization websites, or provided by partners at the Vilas County Land & Water
Conservation Department. We received a response from forty-one of the organizations through
email, online contact forms, phone calls, and in person. Once in contact with the president or
primary contact, we asked them to invite one to four other members to the interview for a
maximum of five interviewees. The interviews lasted one to two hours and were conducted in
community centers, lake organization member homes, and once on a boat.
We used a semi-structured interview methodology. The interviews started with each
attendee signing the letter of consent approved by the Arizona State University IRB. Next,
9

everyone filled out Worksheet 1: Changes to the Lake (Appendix 1). Most of the interview was a
guided discussion about the lake organization structure and rules (Appendix 2). Each interview
had the same facilitator and two notetakers. The notetakers took independent notes on the
discussion.
Following each interview, the notetakers immediately coded the institutional design
principles as present or absent based on their notes. Each notetaker coded independently and then
compared their decisions. When the notetakers disagreed, the facilitator made the final decision.
The two design principles that had a high level of disagreement at the beginning of the data
collection period were: monitoring and low-cost conflict resolution. The disagreements came
from unclear definitions. Once we resolved to include informal monitoring, like lake
organization members noting rule violations from their boats, and programmatic monitoring, like
boat ramp inspections, in our definitions of monitoring, we did not have further issues with
disagreement. For conflict resolution, the challenge was deciding what is considered rapid and
low-cost. We included the use of lake organization meetings and science-based conflict
resolution but excluded lawsuits and court-based conflict resolution. Once we reached an
agreement, the notetakers entered the data into our database.
2.3 The Outcomes: Seven Lake Organization Goals
The people living around the lakes use the lakes in different ways. As a result, lake
organizations have multiple social and environmental goals. Figure 3 shows the goals mentioned
by the lake organizations during our interviews. We used constant comparison, a process in
which each finding is compared with present findings to determine whether it is the same or
different, to identify eleven themes in the lake organization goals recorded in our field notes
(Glaser & Strauss, 1967). Lake stewardship, education, and aquatic invasive species
management were most common; organizations managed 56 % of the thirty-nine lakes in the
study with these goals. The next three goals, named by 20% or more of lakes, were focused on
community building, aquatic invasive species prevention, and water clarity. These findings are
consistent with Gabriel and Lancaster’s survey results (Gabriel & Lancaster, 2004). The least
common goals were a) transition to a lake district and b) to enhance property values, which
included 5% or less of lakes. We were surprised to find that the lake organizations that we
interviewed did not mention fishery protection and zoning issues as often as lake organizations in
the 2004 Gabriel and Lancaster study.

10

Figure 3. The eleven lake organization goals and the proportion of the thirty-nine lakes managed with
each goal. Source: 2019 Interview Dataset.

In Table 1, we map the goals to outcomes for each lake. We use data available via the WI
DNR, UW-Extension Lakes Program, and our 2019 Interview Dataset. We focus on the common
goals; we mapped seven of the eleven goals in Figure 3 to outcomes in Table 1. We do not
include habitat restoration, zoning protection, a transition to a lake district, and property values
in this study.

11

Table 1. The mapped outcomes and dichotomization of seven of the goals mentioned by lake
organizations during the 2019 interviews. The data for the outcomes come from several public sources.
Appendix 4 shows the distribution of continuous variables.
Goal
Lake
Stewardship
Education
AIS
Management
Community
Building
AIS Prevention
Water Clarity
Fishery
Management

Outcome
Lake Management Grant

Present (1)
Received

Absent (0)
Not Received

Source
WI DNR

Clean Boats, Clean
Waters (2019)
AIS Treatment Grant

Participated
Received

Did Not
Participate
Not Received

UW-Extension
Lakes
WI DN R

Participation in
Organization
Eurasian Watermilfoil
(2019)
Very High Water Clarity
Adult Walleye per Acre

≥ 0.65

< 0.65

Present

Absent

2019 Interview
Dataset
WI DNR

Very High
≥ 1.42

Moderate, Low
< 1.42

WI DNR
WI DNR

We mapped goals to outcomes based on lake organizations’ description of their goals.
When talking about lake stewardship organizations mentioned general lake management and
shoreline protection, lake organizations apply for lake management grants to understand and
make improvements to the lake. Without a grant, they have no authority to make changes. AIS
treatment grants are specific to AIS management; they allow lake organizations to apply
chemical and manual treatments to the lake. Education of members and lake users happens in
many different ways; however, Clean Boats, Clean Waters (CBCW) is the most widely adopted
and recorded approach. Through CBCW, volunteers educate lake users about the risks of AIS.
When lake organizations talked about community building, they mentioned increasing
membership and neighborhood connections. Organization participation is a function of
membership that controls for variations in the number of houses around a lake. Lake
organizations are very concerned about EWM. When they talked about AIS prevention, it was
most often about EWM. Water clarity is the only goal that is the same as its outcome. Fishery
management like AIS prevention could be general, but lake organizations mentioned walleye
most often as the fish they care most about; this is also a fish that the WI DNR manages through
habitat improvement and by stocking—adding juvenile fish to a lake. While lake organizations
stated general goals, the way they described the steps they take to meet them made mapping a
measured outcome straightforward.
Table 2 compares the outcomes for lake SESs with a goal to those without a goal related
to the outcome. On the one hand, lakes with AIS management and AIS prevention goals receive
more plant management grants and have a lower incidence of Eurasian Watermilfoil than lakes
that do not state these goals. On the other hand, lakes with lake stewardship and community
building goals receive fewer general grants and have lower participation than organizations that
do not state these goals. We used the Mann-Whitney U test (Mann & Whitney, 1947) for
significance and found all these significant to 0.1. For water clarity and fishery management,
there was no significant difference in outcome between the two groups.

12

Table 2. Impact of Stated Goals on Success. This table shows the percentage of the lakes in our study
managed under each goal. We then show the percent successful as measured by the metrics in Table 1.
compare success between lakes that have stated the goal to lakes that have not stated the goal. Using the
Mann-Whitney U test (Mann & Whitney, 1947), all group differences have at least a 0.10 significance
level except water clarity and fishery management. Source: 2019 Interview Dataset, WI DNR.

Organizational Goal
Lake Stewardship
AIS Management
Education
Community Building
AIS Prevention
Water Clarity
Fishery Management
1
Significance < 0.1

% Lakes
56%
56%
56%
41%
26%
21%
18%

Metric
GRNT
APM
CBCW
PART
~EWM
CLAR
ABUN

Stated
73%
55%
31%
31%
90%
25%
47%

Not Stated
100%
24%
65%
61%
62%
19%
71%

z-Score
-1.431
1.631
-1.731
1.541
1.291
-0.23
-1.15

The outcomes in Table 2 are used in our analysis. In the next section we explore the
combinations of conditions that lead to the outcomes. Although we cannot conclude anything
about outcomes from goal setting alone, we include goal setting as a condition in our analysis.
2.4 The Conditions: Environmental, Social, and Institutional
The environmental, social, and institutional conditions listed in Table 3 are the product of
a comprehensive, iterative selection process. First, we selected the theoretically relevant
variables from the IADF and IDPs (Ostrom, 1990, 2005). We then included theoretically and
empirically derived conditions from a literature review and the summer 2019 interviews,
respectively. Through an iterative process of analyzing different outcomes in dialogue with our
cases, we identified the following variables as most useful to understand our outcomes.

13

Table 3. The dichotomized environmental, social, and institutional conditions and their data sources. The
dichotomization of continuous variables uses the median value, for plots see Appendix 4.
Condition
Environmental
Eurasian Watermilfoil (2019)
Lake Type
Lake Size (ac)
Lake Depth (ft)
Distance from Road (ln(m))
Conductance (uS/cm)
Total Phosphorous (ug/L)

Present (1)

Absent (0)

Source

Present
Seepage, Spring
≥ 377
≥ 32
≥ 6.58
≥ 69
≥ 12.4

Absent
Drainage
< 377
< 32
< 6.58
< 69
< 12.4

Stock Walleye (since 2000)
Social
Participation in Organization
Building Density
Lake Organization Type
Institutional
Graduated Sanctions
Accessible Conflict Resolution
Exclusion
Work with Consultant
Town Lakes Committee
Outcome as a goal

Yes

No

WI DNR
WI DNR
WI DNR
WI DNR
USGS
NTL LTER
Jones Lab, NTL LTER, WI
DNR
WI DNR

≥ 0.65
≥ 16.58
Lake District

< 0.65
< 16.58
Lake Assoc.

2019 Interview Dataset
USGS
2019 Interview Dataset

Present
Present
Present
Yes
Member
Yes

Absent
Absent
Absent
No
Not Member
No

2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset

The data we used for the conditions come from several sources, including the WI DNR,
USGS, NTL LTER, Jones Lab, and our 2019 Interview Dataset. Ten of the environmental,
social, and institutional conditions we used are categorical. For the remaining seven conditions,
we evaluated the distribution (Appendix 4). We used the median to convert them into
dichotomous variables, which is essential for the analysis method we used. The condition
“outcome as a goal” is drawn from the goals in Table 2.
2.4.1 Environmental Conditions
Several environmental conditions affect lake SES outcomes. For our study, we
considered Eurasian Watermilfoil, lake type, lake size, lake depth, distance from road,
conductance, total phosphorous, and stock walleye.
Lake type, lake size, and lake depth are all morphological characteristics of the lake that
affect the processes within the lake. Deeper lakes and spring-fed lakes are both clearer, which is
tied to the phosphorous cycling in lakes and groundwater fluxes (Johnston & Shmagin, 2006).
Drainage lakes, lakes that have an inlet or outlet, are more susceptible to aquatic invasive species
(Smith, Smith, Barko, & Barko, 1990).

14

Conductance and total phosphorous are conditions that describe a lake’s productivity or
ability to support plant and animal life. Higher total phosphorous and conductivity indicate
higher productivity (Johnston & Shmagin, 2006; Smith et al., 1990).
Eurasian Watermilfoil and the distance from the road are both environmental factors that
may impact how people use the lake and the amount of investment from a lake organization.
Eurasian Watermilfoil is an aquatic invasive species that is easily spread through fragments and
can make lakes unnavigable (Smith et al., 1990). Distance from the road is a measure of the
accessibility of a lake.
Stock walleye indicates where the lake has been stocked with juvenile walleye since
2000. The WI DNR stocks walleye to maintain a healthy fish population in the lake when fish
mortality outpaces natural reproduction.
2.4.2 Attributes of the Community
There are three attributes of the community that we identified through interviews and
literature review. These are building density, participation in organization, and lake organization
type.
Building density is calculated as the number of buildings per unit of the perimeter.
Building density reflects the lakefront property development, which affects the shoreline habitat.
The shoreline development can affect runoff into the lake and fish reproduction. Additionally, it
indicates the concentration of individuals living around the lake, which can affect cooperation.
Participation in organization is the percent of households who are members of the lake
organization. Percent participation is different from volunteering. Many of the lake organization
participants just pay dues or pay dues and attend the annual meeting. It is a much smaller group
that contributes time to lake management.
Finally, we include the lake organization type (LDST). We only have a small sample of
lake districts in our study, so we do not see this variable in our later causal combinations. The
lake organization type, however, can play a role in solving collective action problems given a
lake districts’ mandatory participation and ability to levy taxes.
2.4.3 Rules-in-Use
Neither the biophysical conditions nor the attributes of the community—explained in the
previous two sections—alone explain the benefits of a lake organization for a lake nor how
differences between organizations impact their performance. We include the institutional design
principles or rules-in-use, which Ostrom describes as “an essential element or condition that
helps to account for the success of these institutions in sustaining the [common pool resources]
and gaining the compliance of generation after generation of appropriators to the rules in use”
(Ostrom, 1990). Figure 4 shows the proportion of the thirty-one, interviewed organizations that
employ each of the eight design principles. We see that some design principles are present for
most or all lake organizations and others that vary by organization.
15

Figure 4. The proportion of Lake Organizations using each Institutional Design Principle (IDP).
Source: (Ostrom, 1990), 2019 Interview Dataset

All lake organizations have 1A) clearly defined boundaries and 8) nested enterprises.
Lake organizations are formed around lakes “for the purpose of undertaking a program of lake
protection and rehabilitation” (Wisconsin State Legislature, 1974). It is inherent to this type of
organization that the physical boundaries of the resource are clear to the organization.
Additionally, Wisconsin Statutes Chapter 33 establishes that the lakes will be managed in
a partnership between the Department of Natural Resources, the University of Wisconsin system,
other government agencies, and public and private organizations known as the Wisconsin Lakes
Partnership (Wisconsin State Legislature, 1974). As a result, design principle eight—nested
enterprises—is required by Wisconsin State law. Even though nested enterprises are present for
all lake organizations, there are different types of partnerships formed by lake organizations.
These partnerships include partnering with other lake organizations, a town lakes
committee, the DNR, Vilas County Land and Water Committee, or a consultant. We found that
almost all organizations partner with the WI DNR or Vilas County Land and Water Conservation
Department, most partner with a consultant or town lakes committee, and few organizations
partner with other lake organizations. Lake organizations rely on funding, permits, and training
from the WI DNR and Vilas County Land and Water Conservation Department, so they must
partner with them if they take any management actions on the lakes. The lake management
planning and lake study grants require lake organizations to partner with an approved consultant
to be funded, so it is typical for lakes that are applying for grants to partner with a consultant.
16

Lake organizations do not have to work with a consultant; they can join a town lakes committee
that applies for grants and commissions a consultant on behalf of the lakes in their township.
These town lakes committees serve as a network of the lakes in their area. Finally, it is rare for
lake organizations occasionally to collaborate with the lakes upstream or downstream from them.
Over 90% of the organizations had rules in practice that facilitate 2) congruence between
appropriation and provision rules and local conditions, 3) collective-choice arrangements, 4)
monitoring, and 7) minimal recognition of rights to organize. Qualified lake associations and
lake districts are the two types of lake organizations that can apply for grants to carry out
projects to protect or rehabilitate the lakes they are formed around. To be a qualified lake
association, the organization must “allow any individual to be a member if they reside within one
mile of the lake for at least one month each year, allow any individual to be a member if they
own real estate within one mile of the lake, not have articles of incorporation or bylaws which
limit or deny the right of any member or class of members to vote” (Wisconsin Lakes
Partnership, 2018). The people who live around the lake have a voice in the proceedings of the
lake association; people who visit the lake for day-use may not have a voice. Lake districts
require any landowners within the border of the district to be a member and landowners; resident
and non-resident, have the right to vote. Therefore, we were not surprised to see that 94% of the
organizations interviewed had collective-choice arrangements.
Monitoring can be formal or informal. 97% of the lake organizations in Vilas County had
one or the other. Often, they had both. Formal monitoring occurs through two programs
facilitated by the University of Wisconsin Extension Lakes Program: Citizen Lake Monitoring
Network and Clean Boats Clean Waters. Through the Citizen Lake Monitoring Network,
“volunteers measure water clarity, using the Secchi Disk method, as an indicator of water
quality. Volunteers may also collect chemistry, temperature, and dissolved oxygen data, as well
as identify and map plants or watch for the first appearance of Eurasian Water Milfoil near boat
landings” (Department of Natural Resources, 2019). Clean Boats Clean Waters “inspectors
perform boat and trailer checks for invasive speciess, distribute informational brochures and
collect and report any new water body infestations” (UW-Extension Lakes, 2019). Most lakes
participate in Citizen Lake Monitoring Network; however, fewer participate in the Clean Boats
Clean Waters (CBCW) program. CBCW requires a significant number of volunteer hours.
In addition to formal monitoring, the lake organization members also serve as informal
monitors. As users of the lake, people will notice when people are complying with or breaking
the rules. Many of the people we interviewed, for example, own pontoon boats. When they were
cruising the lake in the evenings, they noticed if people were water skiing after the no-wake
hours started. There is no formal program that assigns monitoring time to different pontoon
owners; this informal monitoring is common in resource systems when the systems are small,
and the users live or work nearby.
Finally, Wisconsin Statute Chapter 33 provides lake organizations the right to organize.
Lake organizations are legitimized by Wisconsin law and supported through the Wisconsin
Lakes Partnership.

17

The three design principles that were employed by lake organizations with more variation
were 1B) the exclusion of non-permitted users, 5) graduated sanctions, and 6) conflict-resolution
mechanisms. These are the three rules-in-use that we employ for qualitative comparative
analysis. In our interviews, we found that exclusion was carried out through physical structures,
like narrow or shallow entry to a lake, or through social pressure. An example of social pressure
is encouraging homebuyers who intend to rent out their cabin to consider buying a property on
another lake. Sanctions—verbal warnings and written notices—were employed by fewer than
fifty percent of the organizations. The organizations that did use graduated sanctions decided
against verbally warning their neighbors because members did not think it was their job—they
are not the police. Finally, conflict resolution was rarely needed by organizations. For the
organizations that had experienced conflict, most found low-cost ways of resolving the issue,
such as inviting a DNR scientist to mediate. However, a couple cited going to court. These rules
in use derived from the Ostrom institutional design principles complement the biophysical
conditions and attributes of the community following the contextual variables outlined in the
IADF.
The institutional conditions that we include in our analysis are the exclusion of nonpermitted users, graduated sanction, and conflict resolution. We also included partnerships with a
consultant and with town lakes committees. Lastly, we included organizational goals as a
condition to understand the lake SES outcomes.
2.5 Analytical Approach: Linear Regression Analysis
Environmental and social conditions could explain the goals and outcomes defined by
lake organizations. We used linear regression to explore the effect of social and environmental
variables on water clarity and adult walleye per acre. We used a logit model to do the same for
the presence of Eurasian Watermilfoil (EWM). We explored the three outcomes where data was
available for many lakes across Vilas County; we were not able to test all outcomes.
2.6 Analytical Approach: Crisp-Set Qualitative Comparative Analysis (csQCA)
Charles C. Ragin developed qualitative comparative analysis (QCA) as a “synthetic
strategy” to “integrate the best features of the case-oriented approach with the best features of the
variable-oriented approach” (Ragin, 1987). According to Ragin, a case-oriented approach
(qualitative) assesses a case holistically, while a variable-oriented approach (quantitative)
separates the case into its parts. QCA combines features of both approaches, but it is more
clearly a case-oriented, qualitative method used to compare cases systematically.
In QCA, each case is a combination of conditions that as a whole comprises the case
(Rihoux et al., 2012). QCA draws from set theory and Boolean algebra, the rules of logic—
creating logical sets using “and” and “or” operators—established by these approaches make it an
easily replicable method of case comparison (Ragin, 1987). The replicability of QCA is a
significant asset of this approach when compared to qualitative techniques without formalized
rules of logic (Rihoux et al., 2012). Additionally, QCA is transparent about the choice of cases,
variables, and the tools for analysis.

18

There are three types of QCA analyses: crisp set, fuzzy set, and multi-variate. Crisp set
QCA (csQCA), the method we employ, uses dichotomized variables; all continuous and
categorical variables are coded as present or absent. csQCA is most appropriate for our analysis
because eight of the twelve factors we employ are dichotomous. Based on the sensitivity analysis
we conducted, we do not have cause to assume that varying degrees of the remaining four factors
would have a significant impact on the outcomes.
csQCA analysis has two steps: identification of necessary conditions and generation of
sufficient causal conditions for an outcome. In QCA, a condition is necessary for an outcome if it
is always present when the outcome occurs; a condition is sufficient for an outcome if the
outcome always occurs when the condition is present (Rihoux & Ragin, 2009). We used the
fsQCA 3.0 software—developed by Charles Ragin and Sean Davey—for both of these steps
(Ragin & Davey, 2016).
Using the software, we analyze the consistency of an outcome for a given set of
conditions; we follow the convention where combinations with a consistency score equal to or
greater than 0.80 are kept (Ragin & Davey, 2016). We then use the Quine-McCluskey algorithm,
also called the tabulation method, to simplify the combinations to their minimal number of
conditions (McCluskey, 1956).
When using QCA, the researcher views causation as context and conjecture specific; the
researcher does not specify a single causal model that best fits the data. The researcher, instead,
identifies the multiple models that exist within comparable cases (Ragin, 1987). QCA, therefore,
allows for “multi conjectural causation” across cases; there can be more than one combination of
factors that lead to a given outcome called equifinality (Rihoux et al., 2012).
The results from a systematic comparison of cases, QCA, may be used in “modest
generalization.” We can form propositions to apply to similar cases with due caution (Rihoux et
al., 2012). QCA, as suggested before, does not assume some of the key ideas in mainstream
statistics. These ideas include permanent causality, uniformity of causal effects, unit
homogeneity, additivity, and causal symmetry (Rihoux et al., 2012).
We take an unconventional approach in this study, repeating csQCA’s identification of
necessary and sufficient conditions for multiple outcomes. Most studies identify necessary and
sufficient conditions for a single outcome. In the following sections, we will explain the
outcomes and conditions used in our analysis.
3. RESULTS & DISCUSSION
We use linear regression analysis to identify environmental and social conditions that
influence success on water clarity and Eurasian Watermilfoil prevention. We then use crisp-set
qualitative comparative analysis (csQCA) to compare the thirty-nine lakes whose lake
organizations we interviewed. Following the standards in the csQCA methodology, we used a
two-step analysis. First, we identified the necessary conditions for each outcome. Second, we
identified sufficient conditions using the Quine-McCluskey algorithm for solution simplification.

19

3.1 Linear Regression Analysis
We start with linear regression analysis (Table 4) of certain environmental and social
conditions on three lake SES outcomes. We explore water clarity, adult walleyes per acre, and
the presence of Eurasian Watermilfoil. WI DNR data is widely available for Vilas County lakes
for these outcomes. We examined whether having a lake organization, having a public landing,
the size, the depth, the lake type, the distance to a secondary road, and building density have a
significant effect on these outcomes.
Table 4. The WI DNR has data available on the water clarity, walleye abundance, and Eurasian
Watermilfoil (EWM) presence for 1,136 lakes in Vilas County. We examined whether having a lake
organization, having a public landing, the size, the depth, the lake type, the distance to a secondary
road, and building density has a significant effect on these goals through regression analysis.
Source: WI DNR, USGS, NTL LTER

Goal Metric
Model Type
constant
Lake organization
Public landing
Size (acres)
Max depth
Drainage
Drained
Seepage
Spring
Distance secondary road
Building density

Water Clarity
Linear Regression
+++

Adult Walleyes
per Acre
Linear Regression
+++

EWM
Logit
+

–
––

+++
–––

–

N
342
2
R
0.4767
Significance: +++/––– p<0.001 | ++/–– p< 0.01 | +/– p< 0.1

–

++

82
0.1492

148
0.2708

As expected, we see that water clarity is explained by the depth of the lake (Johnston &
Shmagin, 2006). Drainage lakes have lower water clarity. A drainage lake has an inlet and outlet
and often have higher nutrient loads than other lakes, leading to water with lower clarity. Lakes
with a public landing and spring-fed lakes have a modest negative effect on water clarity. Lakes
with a public landing give access to boaters and swimmers, which can stir up sediment that has
settled on the bottom of the lake. Groundwater flows into spring lakes from the bottom of the
lake.
Adult walleye abundance is modestly impacted by building density around the lake.
Walleye require certain conditions in shallow water to spawn; lakefront property development
may impact this (Jennings et al., 2003; Sass et al., 2006). Additionally, there may be higher
fishing pressure on lakes with more development.

20

The presence of Eurasian Watermilfoil (EWM) is lower for deeper lakes and higher for
lakes with high building density and lakes with lake organizations. EWM grows on the bottom of
lakes and is often transported by boat into a lake. Nutrients may be limited in deeper lakes,
negatively impacting EWM growth (Johnston & Shmagin, 2006). Building density around the
lake probably means more lake traffic providing more opportunities to transport EWM. It may
seem counter-intuitive that a lake having a lake organization is slightly positively correlated with
EWM. It is common for lake organizations to form in response to a threat—aquatic invasive
species appearance is a common cause (Gabriel & Lancaster, 2004).
Our basic regression analysis suggests that lake organizations have an insignificant role
in these outcomes and that there may be multiple factors influencing the outcomes. The
complexity of SESs makes them a good fit for csQCA, which allows multiple pathways to the
same outcome through combinations of environmental, social, and institutional conditions.
3.2 Necessary Conditions
A necessary condition is always present when the outcome occurs (Rihoux et al., 2012).
We evaluated whether each condition is necessarily present, necessarily absent, or not necessary
for each outcome. For a condition to be considered necessary, it should have a consistency score
of greater than or equal to 0.90 (Cebotari & Vink, 2013). This score means that the condition is
present or absent in 90% of the cases with that outcome.
We evaluated the necessity of the causal conditions in Table 3 for the seven outcomes
and found lake depth is a necessary condition for very high water clarity. During the analysis of
necessity, one considers the presence and absence of the conditions (Cebotari & Vink, 2013). We
only find necessary conditions for very high water clarity, and lake depth only explains 36% of
the outcome. These findings support our hypothesis that lake SES outcomes depend on
combinations of factors and that there are multiple pathways to an outcome.
Table 5. Necessary conditions by outcome. UPPERCASE means the variable is present; lowercase
means the variable is absent. Conditions are considered necessary if they have a consistency value of
0.90 or higher.

Outcome
Very high water clarity

Necessary Conditions1
DEEP

Consistency
1.00

Coverage
0.36

1

For abbreviations see Appendix 5

Lake depth (DEEP) is necessary for very high water clarity. This finding is consistent
with a study by Johnston and Shmagin, where they found lake depth to be the single best
predictor of water clarity. Lake depth is tied to phosphorous cycling in the lakes and groundwater
fluxes (Johnston & Shmagin, 2006). These processes both affect the prevalence of algae and
vegetation in the lake and thus the water clarity. Because the necessary conditions only start to
explain lake SES outcomes, we next explore the sufficient conditions whose combinations lead
to success in our sample.
3.3 Sufficient Conditions

21

The analysis of sufficiency identifies combinations of the environmental, social, and
institutional conditions that lead to the seven outcomes (Table 6). In this analysis, the conditions
sufficient to explain an outcome vary by the outcome assessed. For example, the conditions that
explain receiving a lake management grant differ from the conditions that explain very high
water clarity. For each of the outcomes, there are multiple combinations of factors that lead to
success. Each line in Table 6 represents a combination of variables that lead to the outcome.
Table 6. The combinations of environmental, social, and institutions conditions that lead to the seven
outcomes studied. Following the conventions of Boolean algebra, UPPERCASE letters mean the
condition is present and the value is “1”; lowercase letters represent absence and the value is “0”. The
operators used are the logical “AND” represented by the multiplication symbol “*” and the logical “OR”
represented by the addition symbol “+” (Rihoux et al., 2009). Each line represents a combination of
variables that lead to the outcome.
Combinations1

Lake
Management
Grant
Received

[CONS] +
[TLC*SANC*(stewg+dens)] +
[tlc*STEWg*dens]

1, 0.97

AIS
Treatment
Grant
Received

[DENS*road]*[(cons*AISMg)+CLAR] +
[DENS*ROAD*AISMg*clar] +
[EWM*road*clar*AISMg] +
[EWM*CONS]

1, 0.88

Clean Boats,
Clean
Waters
Participation

[EWM*SANC*ROAD]*[DENS+(SIZE*CONF)] +
[ewm*sanc*SIZE*dens] +
[road*SANC*CONF*SIZE]*[ewm+DENS]

1, 0.72

Participation
in Org
≥ 0.65

[CONS*commg]*[(SANC*road)+(SIZE*EWM)] +
[CONS*ROAD]*[(sanc*commg)+(sanc*SIZE)+(size*EWM)] +
[cons*road*COMMg*SIZE] +
[cons*commg*ROAD*SANC]

1, 0.86

Eurasian
Watermilfoil
Absence

[clar*dens]*[AISPg+(SANC*cond)+(TP*DEEP)] +
[clar*tp*deep*cond*aispg] +
[clar*DENS*SANC*COND] +
[clar*sanc*AISPg] +
[CLAR*tp*DEEP]*[SANC+cond] +
[dens*tp]*[(cond*DEEP)+(clar*deep)]

1, 0.96

[DEEP*SEEP*(ROAD+CLARg)]

1, 0.88

[clar*DEEP]*[(sanc*dens)+(cond*SANC)+(COND*sanc*STOCK)] +
[clar*cond*dens*stock] +
[CLAR*DEEP*COND*SANC]

1, 0.75

Very High
Water
Clarity
Adult
Walleye/acre
≥ 1.42
1

Consistency,
Coverage

Outcome

Abbreviations used are available in Appendix 5.

22

The combinations that lead to the seven outcomes range in complexity and number. For
example, very high water clarity has one pathway comprised of four conditions. High
participation in the lake organization has four pathways with six conditions. All of the pathways
have a consistency of 1. A consistency score of 1 means the cases that exhibit the conditions in
that combination have the same outcome. The coverage ranges from 0.72 to 0.97, which means
the pathways explain 72-97% of the cases with the outcome. The outcomes are sensitive to the
way the variables have been dichotomized. When the conditions are dichotomized on the mean,
rather than the median, the same conditions explain 63-94% of the outcomes (Appendix 7).
Outcomes can also be conditions in SESs. Very high water clarity is an outcome that lake
organizations care about; it influences the appearance of EWM and adult walleye abundance.
The interconnected nature of SESs systems blurs the line between cause and effect.
There are three pathways by which lake organizations receive a lake management grant.
These pathways explained 97% of the cases when lake organizations received grants. The first
pathway is working with a consultant (CONS); consultants are paid through grants to conduct
lake studies or prepare lake management plans for lake organizations. They provide scientific
knowledge and have streamlined processes based on years of experience. The second pathway
includes being a member of a Town Lakes Committee (TLC) and employing graduated sanctions
(SANC) when there is no stewardship goal (stewg), or the building density is low (dens). Town
lake committees can apply for grants on behalf of lake organizations and are forums for sharing
information between organizations. Graduated sanctions (SANC) mean that organizations are
sophisticated enough to enforce their rules and do it on a sliding scale, which promotes learning.
The third pathway includes organizations that have a stewardship goal (STEWg), are not town
lakes committee members (tlc), and have low building density (dens) around the lake. These
organizations are focused on stewardship; lake management grants provided by the WI DNR are
the best method to make changes to the lake. Receiving a lake management grant was achieved
in three ways, which involve working with information aggregators—consultants and town lakes
committees—and organizational sophistication shown through graduated sanctions and goal
setting.
Lake organizations received aquatic invasive species (AIS) treatment grants when they fit
one of four pathways. These pathways described 88% of the cases when an AIS treatment grant
was received. The four pathways fall into two groups, lakes with high building density (DENS)
and lakes with Eurasian Watermilfoil (EWM). The first high building density pathway is lakes
that are close to a secondary road (road). These lakes are accessible, which may increase the nonresident traffic on the lake. Higher non-resident traffic would lead to a greater risk of the
introduction of AIS during boat launching. The second high building density pathway includes
lake organizations with aquatic invasive species management goals (AISMg) and that manage
moderate to low clarity lakes (clar) that are not close to a secondary road (ROAD). These
organizations need AIS treatment grants to reach their goals. For lake organizations with EWM,
a rapidly spreading AIS that chokes out other plant life, one pathway includes organizations with
aquatic invasive species management goals (AISMg) managing lakes moderate to low clarity
lakes (clar) near secondary roads (road). These accessible, EWM-plagued lakes need AIS
treatment grants to meet their goals and prevent the spread of EWM. The fourth pathway
includes organizations who work with consultants to manage EWM-plagued lakes. Consultants
23

help lake organizations carry out the AIS treatment activities funded by the grants. Lake
organizations dealing with EWM that set AIS management goals or partner with consultants
receive AIS treatment grants to manage lakes that have high building density or are close to
secondary roads.
Clean Boats, Clean Waters (CBCW) is an AIS education program carried out by
volunteers who inspect boats at launch ramps across the state of Wisconsin. Three pathways
explain 72% of the cases where lake organizations participated in CBCW during the summer of
2019. The first pathway includes lake organizations employ graduated sanctions (SANC) to
manage lakes with EWM (EWM) that are not close to secondary roads (ROAD). These
conditions indicate that they already have an AIS, but they are committed to educating people
about its spread through boat ramp monitoring and rule enforcement. The second pathway
includes organizations that employ graduated sanctions (SANC), but do not have Eurasian
Watermilfoil (ewm). These lakes are large and have a low building density. CBCW is a
volunteer-based program; lakes with graduated sanctions have stronger rule enforcement and
perhaps stronger organizations. The third pathway is large lakes (SIZE) near secondary roads
(road) managed by organizations with graduated sanctions and conflict resolution. The size and
accessibility of these lakes may put them at risk, so they participate in CBCW and have a
sophisticated institutional structure. The lake organizations that participate in CBCW vary in
structure as do the lakes they manage. Some organizations participate as a preventative measure;
others have EWM and still participate. Some organizations supplement CBCW with graduated
sanctions, and others do not.
High lake organization participation, ≥ 65%, is explained by four pathways. These
pathways explain 86% of the cases where organization participation is high.
First, lake organizations that partner with consultants (CONS) and do not have a
community-building goal (commg). Members participate in surveys and workshops, like aquatic
plant identification, during lake management studies by consultants. The resulting products are
exciting and serve as strategy documents for the organization. These organizations, which
manage large (SIZE) or accessible (road) lakes, might not have a community-building goal
because they have high participation. The second pathway includes lake organizations that work
with consultants (CONS) and are not close to a secondary road (ROAD). The third pathway is
large, accessible lakes that have community building goals (COMMg). Finally, organizations
that are not close to a secondary road (ROAD) and employ graduated sanctions (SANC) have
high participation. The pathways that lead to high participation differ by lake size and
accessibility. Common strategies like sophisticated organizational practices, partnering with a
consultant, and goal setting, lead to high participation.
The absence of Eurasian Watermilfoil is the result of six pathways, which explain 96% of
the cases where EWM was absent. The first pathway includes lakes that have moderate to low
water clarity (clar) and low building density (dens). Less light penetrates water with lower
clarity, which inhibits EWM growth (Smith et al., 1990). Additionally, some of these lakes are
deep (DEEP), which inhibits EWM growth for the same reason. The next pathway is shallow
(deeps) lakes with moderate to low water clarity (clar). These lakes have low conductivity (cond)
and total phosphorous (tp). Conductivity and total phosphorous are different measures of lake
24

productivity; low conductivity and low phosphorous indicate low lake productivity. The third
pathway also includes moderate to low water clarity (clar) lakes managed by organizations with
graduated sanctions (SANC) in place. These lakes also have high conductivity (COND) and high
building density (DENS). Though the lake productivity and building density seem to be
favorable to EWM, the graduated sanctions play a role in preventing EWM. The fourth and final
pathway with moderate to low water clarity includes organizations that set AIS prevention goals
(AISPg). The fifth pathway is very high water clarity (CLAR), low total phosphorous (tp), deep
(DEEP) lakes that either have low conductivity (cond) or graduated sanctions (SANC).
Phosphorous is a nutrient that promotes EWM growth (Smith et al., 1990), so low levels of
phosphorous in combination with the other factors prevent EWM presence. The final pathway
includes lakes with poor growing conditions for EWM that have low building density (dens).
Eurasian Watermilfoil is prevented by unfavorable environmental conditions like low lake
productivity and water clarity; graduated sanctions and goal setting also play a key role in
preventing this aquatic invasive species.
Very high water clarity is the result of one pathway, which explains 88% of the cases
where water clarity is very high. The lakes in this group are deep (DEEP) and either seepage or
spring lakes (SEEP). Both of these conditions are associated with phosphorous cycling in the
lakes; deep, seepage or spring lakes have less phosphorous and, therefore, slower algae and plant
growth (Johnston & Shmagin, 2006). These lakes were also far from a secondary road (ROAD)
or the organization had a water clarity goal (CLARg). The lakes far from a secondary road may
have less traffic, churning less sediment, or have more natural watershed leading to fewer runoff
nutrients. Very high water clarity is a function of the hydrology in the lake; very clear lakes are
deep, seepage or spring lakes.
The proportion of adult walleye per acre is higher in three scenarios. These scenarios
explain 75% of the cases where the number of adult walleye per acre was equal to or higher than
1.42. In two of the pathways, the water clarity is low to moderate (clar).
The first pathway is deep (DEEP), moderate to low clarity lakes. The low water clarity
and depth make these good walleye lakes. Additionally, the walleye populations benefit from
low building density (dens), graduated sanctions (SANC), high conductance (COND), and
stocking (STOCK) in various cases. The second pathway is low conductance (cond) lakes with
low building density (dens) and organizations that do not stock (stock). These lakes have low
productivity and are not deep; this goes against the understanding of what makes a good walleye
lake. The low density and lack of stocking may mean these lakes are out of the way, without
much fishing pressure. The third pathway is clear (CLAR), deep (DEEP), high conductance
(COND) lakes that employ graduated sanctions (SANC). The natural conditions in the lake are
favorable to walleye, and the graduated sanctions mean that the harvest limits are probably
enforced. The lakes with more adult walleye/acre tend to be environmentally favorable and either
less developed or with graduated sanctions in place.
Trends across the seven outcomes show that some outcomes have a greater social and
institutional impact, others have a greater environmental influence, and some are a blend.
Receiving a lake management grant and having high participation are both heavily influenced by
working with a consultant, graduated sanctions, and goal setting; these are social and
25

institutional conditions. Adult walleye abundance and very high water clarity are influenced by
environmental factors like lake productivity and lake depth, respectively. Finally, receiving an
AIS treatment grant, CBCW participation, and EWM prevention are a cross over. These three
cross-over outcomes are management activities carried out by lake organizations to prevent and
manage AIS. Positive outcomes are a result of the natural conditions of the lake such as water
clarity, social conditions like the building density, and institutional conditions like graduated
sanctions.
4. CONCLUSION
We asked how the combinations of environmental, social, and institutional conditions
lead to different outcomes for lake SESs with volunteer-based organizations. We found that
multiple combinations can lead to an outcome; these combinations vary in number and
complexity by the outcome. Social-ecological systems acknowledge the interactions between
environmental, social, and institutional factors. Some lake SES outcomes were influenced by
environmental factors, others by social and institutional, and others by a combination. Outcomes
that were also conditions, like water clarity, blurred the relationship between conditions and
outcomes. Social-ecological research, which focuses on systems as a whole, must acknowledge
the complex interactions between different types of conditions and the outcomes in the system.
A case-based, systematic comparison presents challenges in preparing data, but it
facilitates understanding the complex relationships in social-ecological systems. The approach
we used to collect primary data for a moderate number of cases during one field season was
effective for overcoming the challenges described by Ratajczyk with institutional design
principle data completeness. With a sample of 31 organizations and 39 lakes, gathering
secondary data on the environmental conditions of the lakes required compiling multiple data
sources. The reward was seeing the similarities emerge. By examining a sample of similar
organizations and geographically similar lakes, we identified the contextual nuances in the
organizational structures and lakes. In a traditional case study, these would not have been
identifiable as trends. In a traditional statistical approach, the combinations would not have
emerged and the context would have been lost. Now that the pathways have been identified,
other methods like case studies or statistical analysis can explore the nature and prevalence of
these pathways.
We chose Vilas County for our study because of its geography and geology; these same
factors make it different from many other US counties. There are only a few counties in
Wisconsin, Minnesota, and New York that have the number of lakes found in Vilas County and
the reduced competition between user groups. Vilas County, as a result, has unique networking
opportunities between management organizations. The models developed in Vilas County, like
town lakes committees and volunteer-based organizations, could inform policy in other parts of
the US. We believe that an interesting next step is to explore the ways that lake organizations
emerge and collaborate in the face of changes—like regulatory changes.
Our exploration of different goals focused on lake organizations. Lake organizations have
numerous, diverse goals. Lake organizations are one of several groups that use lakes, and goals
differ by user-group. Managing lakes for different outcomes is challenging. One approach is to
26

identify the underlying drivers that could maintain system resilience, no matter how success is
defined. For example, one person may want to continue catching walleye, another may want to
swim in clear water, another may dream about having a nesting pair of loons on the lake, and a
fourth may want to maintain the value of their lakefront property. Though goals are diverse and
motivated by different uses, there are underlying processes like shoreline development that could
impact all the outcomes.
Natural resource management is an essential and resource-intensive function. Innovative
models of management are needed to adapt to increasing threats from climate change, land-use
change, and invasive species to social-ecological systems. The lakes region in Vilas County,
Wisconsin provides a collaborative model that relies on volunteer-based resource management;
these solutions are not one size fits all. Conditions for success depend on the desired outcomes
and the conditions present in the social-ecological system. Better understanding the dynamic
nature of the environmental, social, and institutional context on outcomes is critical for designing
social-ecological systems that remain resilient to the increasing challenges of the twenty-first
century.

27

5. ACKNOWLEDGMENTS
We are exceptionally grateful to the Vilas County lake organizations who participated in this
study and all the community-based organizations that contribute to resource management. We
thank the Wisconsin DNR, UW-Extension Lakes Program, and the wonderful team at the Vilas
County Land and Water Conservation Department for sharing data and stories.
I thank Marco Janssen, Chris Solomon, and Bryan Leonard for their guidance and their patience
while I developed these ideas. This research is the product of a collaboration including the field
team—Alise Crippen, Corinne Johnson, and our truck Peggy—the FishScapes group, and my
colleagues in Fun Club. This is one milestone on a lifelong pursuit fueled by curiosity and
supported by my parents, wonderful sister, and my mentors Brooke Grimes and Paige Lindstrom.
This work was supported by the National Science Foundation under Grant No. 1716066 to C.T.
Solomon and M.A. Janssen. If there are any typos, you can thank my cat, Louis.

28

6. REFERENCES
Agrawal, A. (2002). Common Resources and Institutional Sustainability. In E. Ostrom, T. Dietz,
N. Dolsak, P. C. Stern, S. Stonich, & E. U. Weber (Eds.), The Drama of the Commons (pp.
41–86). Washington, DC: National Academy Press.
Agrawal, A., & Chhatre, A. (2006). Explaining success on the commons: Community forest
governance in the Indian Himalaya. World Development, 34(1), 149–166.
https://doi.org/10.1016/j.worlddev.2005.07.013
Armitage, D. (2005). Adaptive capacity and community-based natural resource management.
Environmental Management, 35(6), 703–715. https://doi.org/10.1007/s00267-004-0076-z
Baggio, J. A., Barnett, A. J., Perez-Ibara, I., Brady, U., Ratajczyk, E., Rollins, N., … Janssen, M.
A. (2016). Explaining success and failure in the commons: The configural nature of
Ostrom’s institutional design principles. International Journal of the Commons, 10(2), 417–
439. https://doi.org/10.18352/ijc.634
Bruyere, B., & Rappe, S. (2007). Identifying the motivations of environmental volunteers.
Journal of Environmental Planning and Management, 50(4), 503–516.
https://doi.org/10.1080/09640560701402034
Cebotari, V., & Vink, M. P. (2013). A configurational analysis of ethnic protest in Europe.
International Journal of Comparative Sociology, 54(4), 298–324.
https://doi.org/10.1177/0020715213508567
Cox, M., Arnold, G., & Villamayor Tomas, S. (2010). A Review of Design Principles for
Community-based Natural Resource Management. Ecology and Society, 15(4). Retrieved
from http://www.ecologyandsociety.org/vol15/iss4/art38/
Department of Natural Resources. (2019). Citizen Lake Monitoring Network. Retrieved January
9, 2020, from https://dnr.wi.gov/lakes/clmn/
Embke, H. S., Rypel, A. L., Carpenter, S. R., Sass, G. G., Ogle, D., Cichosz, T., … Vander, M. J.
(2019). Production dynamics reveal hidden overharvest of inland recreational fisheries. 1–
6. https://doi.org/10.1073/pnas.1913196116
ERCOLA. (2020). Eagle River Chain of Lakes Association. Retrieved February 24, 2020, from
https://www.ercla.org/
Gabriel, A. O., & Lancaster, C. (2004). Management issues, characteristics and effectiveness of
lake associations and lake districts in wisconsin. Lake and Reservoir Management, 20(1),
27–38. https://doi.org/10.1080/07438140409354098
Glaser, B. G., & Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for
Qualitative Research.
Jennings, M. J., Emmons, E. E., Hatzenbeler, G. R., Edwards, C., & Bozek, M. A. (2003). Is
littoral habitat affected by residential development and land use in watersheds of wisconsin
lakes? Lake and Reservoir Management, 19(3), 272–279.
https://doi.org/10.1080/07438140309354092
Johnston, C. a, & Shmagin, B. a. (2006). SCALE ISSUES IN LAKE-WATERSHED
INTERACTIONS: ASSESSING SHORELINE DEVELOPMENT IMPACTS ON WATER
CLARITY. In J. Wu, K. B. Jones, H. Li, & O. L. Loucks (Eds.), SCALING AND
UNCERTAINTY ANALYSIS IN ECOLOGY (pp. 297–313). https://doi.org/10.1007/1-40204663-4_16

29

Korth, R. M., & Klessig, L. L. (1990). Overcoming the tragedy of the commons: Alternative lake
management institutions at the community level. Lake and Reservoir Management, 6(2),
219–225. https://doi.org/10.1080/07438149009354712
Magee, M. R., McIntyre, P. B., Hanson, P. C., & Wu, C. H. (2019). Drivers and Management
Implications of Long-Term Cisco Oxythermal Habitat Decline in Lake Mendota, WI.
Environmental Management, 63(3), 396–407. https://doi.org/10.1007/s00267-018-01134-7
Mandal, B., & Lawrence, T. (2017). Managing the Commons: How Extension Facilitates Local
Participation to Manage Natural Resources. Applied Economic Perspectives and Policy,
39(3), 499–515. https://doi.org/10.1093/aepp/ppw031
Mann, H. B., & Whitney, D. R. (1947). On a Test of Whether One of Two Random Variables is
Stochastically Larger than the Other. The Annals of Mathematical Statistics, 18(1), 50–60.
Retrieved from www.jstor.org/stable/2236101
McCluskey, E. J. (1956). Minimization of Boolean Functions*. Bell System Technical Journal,
35(6), 1417–1444. https://doi.org/10.1002/j.1538-7305.1956.tb03835.x
Ostrom, E. (1990). Governing the commons: the evolution of institutions for collective action.
Retrieved from
https://books.google.com/books?id=4xg6oUobMz4C&dq=governing+the+commons+ostro
m&lr=&source=gbs_navlinks_s
Ostrom, E. (2005). Understanding Institutional Diversity. Princeton University Press.
Ostrom, E. (2009). A General Framework for Analyzing Sustainability of Social-Ecological
Systems. Science, 325, 419–422. https://doi.org/10.1126/science.1172133
Peterson, G. D., Douglas Beard, T., Beisner, B. E., Bennett, E. M., Carpenter, S. R., Cumming,
G. S., … Havlicek, T. D. (2003). Assessing future ecosystem services: A case study of the
Northern Highlands Lake District, Wisconsin. Ecology and Society, 7(3).
https://doi.org/10.5751/ES-00557-070301
Ragin, C. C. (1987). The comparative method: moving beyond qualitative and quantitative
strategies. Berkeley: University of California Press.
Ragin, C. C., & Davey, S. (2016). Fuzzy-Set/Qualitative Comparative Analysis 3.0. Irvine,
California: Department of Sociology, University of California.
Ratajczyk, E. (2016). Challenges and Opportunities in Coding the Commons : Problems ,
Procedures , and Potential Solutions in Large-N Comparative Case Studies. 10(JULY),
440–466.
Riera, J., Voss, P. R., Carpenter, S. R., Kratz, T. K., Lillesand, T. M., Schnaiberg, J. A., …
Wegener, M. W. (2001). Nature, society and history in two contrasting landscapes in
Wisconsin, USA Interactions between lakes and humans during the twentieth century. Land
Use Policy, 18(1), 41–51. https://doi.org/10.1016/S0264-8377(00)00044-2
Rihoux, B., & Ragin, C. (2009). Configurational Comparative Methods: Qualitative
Comparative Analysis (QCA) and Related Techniques.
https://doi.org/10.4135/9781452226569
Rihoux, B., Ragin, C., Berg-Schlosser, D., Meur, G. de, Rihoux, B., & Ragin, C. C. (2012).
Qualitative Comparative Analysis (QCA) as an Approach. Configurational Comparative
Methods: Qualitative Comparative Analysis (QCA) and Related Techniques, 1–18.
https://doi.org/10.4135/9781452226569.n1
Rihoux, B., Ragin, C. C., Yamasaki, S., Bol, D., Berg-Schlosser, D., de Meur, G., … Yamasaki,
S. (2009). Configurational Comparative Methods: Qualitative Comparative Analysis (QCA)
and Related Techniques. https://doi.org/10.4135/9781452226569
30

Rothlisberger, J. D., Chadderton, W. L., McNulty, J., & Lodge, D. M. (2010). Aquatic Invasive
Species Transport via Trailered Boats: What is Being Moved, Who is Moving it, and What
Can Be Done. Fisheries, 35(3), 121–132. https://doi.org/10.1577/1548-8446-35.3.121
Sass, G. G., Kitchell, J. F., Carpenter, S. R., Hrabik, T. R., Marburg, A. E., & Turner, M. G.
(2006). Fish Community and Food Web Responses to a Whole-lake Removal of Coarse
Woody Habitat. Fisheries, 31(7), 321–330.
Shin, H. C., Yu, D. J., Park, S., Anderies, J. M., Abbott, J. K., Janssen, M. A., & Ahn, T. K.
(2019). How do resource mobility and group size affect institutional arrangements for rule
enforcement? Empirical hypotheses about institutional fit.
Smith, C. S., Smith, C. S., Barko, J. W., & Barko, J. W. (1990). Ecology of Eurasian
watermilfoil. Journal of Aquatic Plant Management, Vol. 28, pp. 55–64.
Stedman, R. C. (2006). Understanding Place Attachment Among Second Home
Owners\n10.1177/0002764206290633 . American Behavioral Scientist , 50(2), 187–205.
Retrieved from http://abs.sagepub.com/content/50/2/187.abstract
University of Wisconsin - Stevens Point. (2019). UW-Extension Lakes. Retrieved April 28,
2019, from https://www.uwsp.edu/cnr-ap/UWEXLakes/Pages/default.aspx
U.S. Department of the Interior, U.S. Fish and Wildlife Service, U.S. Department of Commerce,
& U.S. Census Bureau. (2018). 2016 National Survey of Fishing, Hunting, and WildlifeAssociated Recreation. 144. Retrieved from
https://wsfrprograms.fws.gov/subpages/nationalsurvey/nat_survey2016.pdf
US Fish & Wildlife Service. (2013). Volunteer, Friends and Partnerships: Annual Update FY
2012 People Making a Difference. Retrieved from
https://www.fws.gov/volunteers/pdfs/FINALupdated2012Volunteers.pdf
UW-Extension Lakes. (2019). Clean Boats Clean Water. Retrieved January 9, 2020, from
https://www.uwsp.edu/cnr-ap/UWEXLakes/Pages/programs/cbcw/default.aspx
Vilas County Tourism and Publicity. (2020). Lakes & Waterways - Vilas County, Wisconsin.
Retrieved February 24, 2020, from https://www.vilaswi.com/lakes-rivers/
Wisconsin DNR. (2019). Lakes. Retrieved April 28, 2019, from https://dnr.wi.gov/lakes/
Wisconsin Lakes Partnership. (2018). People of the Lakes: A Guide for Wisconsin Lake
Organizations.
Wisconsin Shoreland Alliance. (2019). Wisconsin Shoreland Alliance: Advocating for
Wisconsin Lakes. Retrieved February 24, 2020, from https://wisconsinshoreland.org/
Wisconsin State Legislature. Chapter 33 Public Inland Waters. , (1974).

31

7. APPENDICIES
Appendix 1. Lake Changes Worksheet
1. For which lake(s) are you answering the questions below?
2. Please summarize how your lake changed over the past 10 years in 3-4 sentences.
In the following questions please check the box that most accurately describes the current state
of the lake attribute listed and indicate whether it has increased (+), stayed the same (=), or
decreased (-) over the past 10 years.
Very
Very
Change
Attribute
Low
Moderate
High
Low
High
(+, =, -)
Water Clarity
Amount of fish
Diversity of fish
Invasive plant
prevalence
Invasive animal
prevalence
Wildlife diversity
Pollution levels
Natural shoreline
Property values
Watershed quality
Personal watercraft
presence
Fishermen
presence
Local visitors
Wisconsin visitors
Out-of-state
visitors
Volunteer turnout
Annual meeting
turnout
Social event
turnout
Lake organization
membership
Housing density
Amount of
stocking
3. Please list and indicate the state and change of any other attributes that you find important.
4. How does your lake organization compare on the attributes above to the other lake
organizations in Vilas County? Please include the names of the organizations.

32

Appendix 2. Semi-Structured Interview Questions
SECTION1
What were the biggest changes you noticed in the past decade?
What do you think has caused the changes? Has your organization influenced the changes?
SECTION2
When did your lake organization form? Why did it form?
Have you considered being a lake district?
How many people are on your board?
Who lives around the lake? How many homes? What % in the lake organization?
Is there other development around the lake besides homes?
Are there other organizations you work with to manage the lake? County? DNR? (polycentricity)
How do people use the lake? Residents vs. non-residents?
What do you consider the lake?
What is your public landing like? Do you manage it? Improve it? (exclusion)
Are there rules about who can or cannot use the lake? (exclusion)
Do you participate in CBCW? AIS monitoring? Stocking? Shoreline improvement? (provision)
Are lake association members involved in rule making? Non-members? (collective choice)
Are there no wake times, special zoning requirements or other ordinances on your lake?
Has the organization suggested new ordinances or requested different catch limits? (collective
choice)
What happens when someone doesn’t follow the rules of the lake? (monitoring, graduated sanc)
What happens when there is a conflict between lake users? DNR or township? (conflict)
What are the goals of the organization? How do you meet them?
Have you had any challenges carrying out your goals? (self-determination)
Are there ordinances or regulations that you’d like to change but haven’t been able to?
Have you been asked to perform certain activities by the DNR or your township?
Why do/don’t you stock fish in your lake? Do you’d stock/not stock in the future?

33

Appendix 3. Code Definitions for Organizational Goals
Goal
Lake
Stewardship
(STEW)

Definition
General lake, shoreline,
and watershed protection,
monitoring, and
management.

Typical Exemplars

Atypical Exemplars

stewards of the
environment, protect the
natural shoreline

keep the lake healthy, keep
management plan updated,
prevent runoff

AIS
Management
(AISM)

Managing or controlling
existing AIS populations.

AIS Management

contain milfoil with
available resources,
control EWM, adequate
funds for management

Education
(EDU)

Education and outreach
goals for lake organization
members and lake users.

Education, outreach

communication on lake,
update website with info

Community
Building
(COMM)

Goals focused on building
the community, promoting
connection between
neighbors, and goodwill.

increase membership,
community building,
neighborhood connections

keep volunteers, good life,
increase membership

AIS
Prevention
(AISP)

Goal specifically mentions
preventing AIS or
protection the lake from
AIS. General lake
protection is considered
STEW.

AIS prevention, be alert
for AIS

future camera installation

Water Clarity
(CLAR)

Maintain, improve, or
monitor lake water clarity.

preserve and maintain
water quality and clarity,
water clarity

water

Fishery
Management
(FISH)

Fishery improvement,
monitoring, and
management.

fishery management,
fishery protection

good fishing

habitat restoration, habitat
improvement

helping the loons

zoning preservation

enforcing the deed
restrictions

Habitat
Restoration
(HAB)
Zoning
Protection
(ZONE)

Habitat restoration or
improvement. This can
refer to wildlife or
vegetation. Protection does
not qualify.
Goals to prevent changes
to zoning and land use
activities.

Transition to
LD (T2LD)

Transition organization
type from a lake
association to lake district.

transition org from LA

Property
Values
(PROP)

Maintain or improve
property values around the
lake.

property values

34

Appendix 4. Continuous Variable Dichotomization

Appendix 4. Continuous variables were dichotomized on the mean (solid line) and median (broken line).

35

Appendix 5. Condition and Outcome Abbreviations
The conditions used to understand the combinations that lead to outcomes for lake SESs. The
condition and outcomes, values for which they are present, abbreviation used in Table 6 and
Appendix 6, and data source.
Environmental Conditions
Eurasian Watermilfoil (2019)
Lake Type
Lake Size (ac)
Lake Depth (ft)
Distance from Road (ln(m))
Conductance (uS/cm)
Total Phosphorous (ug/L)
Stock Walleye (since 2000)
Social Conditions
Participation in Organization
Building Density
Lake Organization Type
Institutional Conditions
Graduated Sanctions
Accessible Conflict Resolution
Exclusion
Work with Consultant
Town Lakes Committee
Outcome as a goal
Outcomes
Lake Management Grant
Clean Boats, Clean Waters
AIS Treatment Grant
Participation in Organization
Eurasian Watermilfoil
Very High Water Clarity
Adult Walleye per Acre

Present (1)

Abbreviation

Source

Present
Seepage, Spring
≥ 377
≥ 32
≥ 6.58
≥ 69
≥ 12.4

EWM
SEEP
SIZE
DEEP
ROAD
COND
TP

Yes

STOCK

WI DNR
WI DNR
WI DNR
WI DNR
USGS
NTL LTER
Jones Lab, NTL LTER, WI
DNR
WI DNR

≥ 0.65
≥ 16.58
Lake District

PART
DENS
LDST

2019 Interview Dataset
USGS
2019 Interview Dataset

Present
Present
Present
Yes
Member
Yes

SANC
CONF
EXCL
CONS
TLC
*g

2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset
2019 Interview Dataset

Received
Participated
Received
≥ 0.65
Present
Very High
≥ 1.42

GRNT
CBCW
APM
PART
EWM
CLAR
ABUN

WI DNR
UW-Extension Lakes
WI DNR
2019 Interview Dataset
WI DNR
WI DNR
WI DNR

36

Appendix 6. QCA Models and Assumptions used in Sufficiency Analysis
Model: GRNT = f(CONS, TLC, SANC, STEWg, DENS)
Assumptions:
CONS (present)
TLC (present)
SANC (present)
STEWg (present)
DENS (present)
Model: APM = f(DENS, ROAD, CLAR, AISMg, CONS, EWM)
Assumptions:
DENS (present)
~ROAD (absent)
~CLAR (absent)
AISMg (present)
CONS (present)
EWM (present)
Model: CBCW = f(ROAD, EWM, SANC, CONF, SIZE, DENS)
Assumptions:
~ROAD (absent)
~EWM (absent)
SANC (present)
CONF (present)
SIZE (present)
DENS (present)
Model: PART = f(CONS, SANC, SIZE, COMMg, ROAD, EWM)
Assumptions:
CONS (present)
SANC (present)
EWM (present)
Model: ~EWM = f(CLAR, DENS, TP, SANC, DEEP, COND, AISPg)
Assumptions:
~CLAR (absent)
~DENS (absent)
~TP (absent)
SANC (present)
DEEP (present)
~COND (absent)
AISPg (present)
Model: CLAR = f(DEEP, SEEP, ROAD, CLARg)
Assumptions:
DEEP (present)
SEEP (present)
ROAD (present)
CLARg (present)
Model: ABUN = f(CLAR, DEEP, COND, SANC, DENS, STOCK)
Assumptions:
37

~CLAR (absent)
DEEP (present)
COND (present)
SANC (present)
~DENS (absent)
STOCK (present)
Appendix 7. Sensitivity analysis of the sufficient condition combinations.
Consistency,
Coverage

Outcome

Combinations1

Lake
Management
Grant
Received

[CONS] +
[TLC*SANC*(stewg+DENS)]

1, 0.94

AIS
Treatment
Grant
Received

[EWM*CONS] +
[EWM*clar*AISMg] +
[DENS*road*(cons+CLAR)] +
[DENS*ROAD*AISMg*clar]

1, 0.88

Clean Boats,
Clean
Waters
Participation

[ewm*sanc]*[(ROAD*SIZE)+(road*dens*CONF)] +
[EWM*SANC]*[DENS+(ROAD*CONF)] +
[SANC*conf*SIZE*DENS]

1, 0.73

Participation
in Org
≥ 0.67

[CONS*COMMg]*[(SIZE*ROAD)+(size*road*EWM)] +
[CONS*commg*road*EWM]*[SIZE+SANC] +
[size*commg]*[(CONS*road*ewm)+(SANC*ROAD)] +
[cons*COMMg*SIZE*road]

1, 0.84

Eurasian
Watermilfoil
Absence

[dens*tp]*[(cond)+(SANC*DEEP)] +
[CLAR*tp*DEEP]*[(SANC*AISPg)+(cond)] +
[clar*dens]*[(sanc*COND)+(SANC*cond)+(SANC*DEEP)] +
[clar*sanc*cond*AISPg] +
[clar*DENS*TP*SANC]

1, 0.93

[DEPTH*SEEP*(ROAD+CLARg)]

1, 0.75

[clar*cond*stock]*[dens+SANC]
[clar*DEEP*DENS*STOCK]*[COND+SANC]
[clar*DEEP*dens*stock]

1, 0.63

Very High
Water
Clarity
Adult
Walleye/acre
≥3

38