The Weatherization Experiences Project: A Social Network Study

A version of this article appears in the November/December 2014 issue of Home Energy Magazine.

SHARE

Click here to read more articles about Weatherization

Individuals belong to complex webs of social networks comprised of many types of relationships, interactions, and functions. As part of a national evaluation of DOE’s Weatherization Assistance Program (WAP), researchers from Oak Ridge National Laboratory (ORNL) did an analysis to explore the links among individuals in weatherized homes, weatherization staff, and WAP agencies within these multirelational social systems. In this article we refer to these links as nodes, and we refer to the ORNL study as the Weatherization Experiences (WE) project.

Analyzing these social networks helps us to understand

1. if weatherization information is being shared;
   
   
2. what type of weatherization information is being shared;
   
   
3. what core values are currently in place that might support or hinder adoption of new energy use behaviors; and
   
   
4. the factors that motivate individuals and households to take action after they receive information on energy.

We expound on the results of the study by exploring the potential use of social networks for three purposes. The first purpose is to diffuse information on saving energy. The second purpose is to educate the public concerning indoor environmental quality (IEQ) and health. The third purpose is to determine if social networks can be used for climate change mitigation and to promulgate adaptation strategies designed to benefit vulnerable populations, such as those targeted by WAP.

 

Conversation Topics—Shared and Remembered

Figure 1. The number one topic shared by weatherization agency interviewees was the measures that are installed in homes.

Motivations for Contacting WAP Providers and Respondents

Figure 2. Almost half of the respondents r eported money and home comfort values as the top motivating factors for contacting a WAP provider.

Network Typology

Figure 3a. The size of the node is a function of the number of edges that lead out from the node, referred to as out-degree.

Network Clustered by Agency

Figure 3b. The network clustered by agency gives a better view of agency activity with DOE in the center. The size of nodes is based on the level of out-degree, and the color of the nodes is based on the type of agency.

The WE Project

For the WE project, we selected ten local agencies that had previously been selected for the WAP national evaluation’s case study report. We asked these ten agencies to recruit interested staff and clients for the WE project. This technique, known as participatory research, allowed us to collect information, starting with the initial client or staff member as a node of communication. These individuals—both staff members and clients—were trained to conduct the research. They were trained in a workshop facilitated on-site at the agency by research professionals through ORNL and by other social scientists subcontracted for this task. Once trained, the new researchers administered questionnaires to members of their own networks to find out whether their experience with weatherization had encouraged others to save more energy, and to understand what had motivated them to do so.

The questionnaires were designed to answer five questions:

1. Whom did you tell?
2. What did you say?
3. What did they hear?
4. What did they do?
5. Why did they do it?

In addition, we used a “snowball” sampling method. The researchers asked the members of their social network if they had discussed weatherization with anyone within their social network (referred to here as the second round). If so, they were asked to provide names and contact information so that researchers could interview these people as well.

Impacts of Social Network Interaction

We measured the impact of an interaction using four criteria. These criteria were (1) the number of reported applications for WAP services; (2) the number of people who invested in energy efficiency measures; (3) the effectiveness of communication, based on information shared versus information remembered; and (4) in what ways interviewees changed their home energy-consuming behavior. Dyadic ties (relationships between two people) fall into four categories: similarities, social relations, interactions, and flows. In the WE project we looked at the communication flow of weatherization experiences between two individuals with social relations and at the outcomes of those interactions.

We found that weatherization information diffused into social networks does in fact influence related actions and behaviors. We defined taking action as contacting a WAP agency or a private contractor or completing any do-it-yourself (DIY) project (see Table 1). Other questions we sought to answer were:

Which aspects of the weatherization experience did the interviewer share most often with the respondents?

What information did the respondents remember hearing?

The topic most often discussed in the first round of interviews was the weatherization measures installed in a home. Energy or cost savings were a close second (see Figure 1). Next in order were ways to save energy, satisfaction with the program, comfort, and health and safety topics. These results provide evidence that messages concerning these topics are in fact being shared and remembered to at least the second degree of separation. The results also support the theory that the shorter the distance from the first node (the staff member or the actual client) to other nodes in the network, the quicker the diffusion of the information.

We conducted analyses in order to determine the characteristics of the initial node. These characteristics included the type of interviewer, whether staff member or client, male or female. Our purpose was to explore relationships between these attributes and level of impact of the first and second rounds of interviews. We found that respondents are more likely to do DIY projects if the initial node was male or was a staff member. We also found that weatherization staff members do encourage their social networks to contact weatherization agencies for more information. Finally, we found that although information shared from the initial node reaches beyond these networks, that information may not strongly influence those who receive it. More than half of the respondents reported that this communication with a trusted source inspired them to change their energy usage behavior. The top five changes reported were (1) turning off lights, (2) unplugging appliances, (3) reading utility bills, (4) adjusting the thermostat, and (5) changing the air filter.

Understanding motivations and values among respondents helps weatherization agencies to improve the use of existing social networks to diffuse information and to promote socially desirable behavior within a culture. Our findings show that the topics that most respondents remembered were linked to the top motivating factors for calling a WAP provider (see Figure 2). The top two motivating factors for contacting a WAP provider were being cold in the winter and having difficulty paying bills. Therefore, it seems reasonable that more respondents would remember hearing about the installation of measures that are likely to reduce energy bills and improve the comfort of their home in winter. (Interestingly, the top motivating factor for contacting a private contractor was environmental conservation.)

Social Network Typology

To further analyze the social networks in the WE project, we used an interactive visualization software program (Gephi) to produce images that capture in graphic form the structure or typology of the network (see Figures 3a and 3b, on p. 36). This type of visualization analysis is useful to identify central nodes and patterns in relationships or interactions. In both Figure 3a and Figure 3b the nodes, or sources of information are represented by circles; the edges or linkages, which in this case represent the flow of information, are represented by lines. The size of the node is a function of the number of edges that lead out from the node. This number of edges is referred to as out-degree. This type of visualization is useful in assessing which nodes are central to the spreading of information, and their potential degree of influence on the network. More specifically, the larger the node, the more people that node shared information with. Edges are directional; they also capture information flowing back to the WAP agency if the respondent reported contacting an agency for more information.

Agencies become central hubs for identifying and receiving best-practice knowledge for weatherization. Agencies that provide weatherization services report back to the state and sometimes to federal offices concerning their experience with best-practice guidelines. Based on this feedback, improved guidelines flow back down to the agency level. Agencies then deliver these new best practices to members within their WAP-eligible service territory. Note that some WAP agencies (typically Community Action Agencies, or CAAs) have additional connections beyond this network. These connections include utility companies and other local, state, and federal agencies that develop, fund, and deliver other energy programs under the CAA services umbrella. An example is LIHEAP, the U.S. Department of Health and Human Services Low Income Home Energy Assistance Program.

The Potential of Social Networks

We have learned through this study that experiences with weatherization and energy conservation are being shared within social networks, and that this information influences the decision-making and energy-saving actions of home occupants in the areas of energy affordability, comfort, health, and environmental conservation. This is consistent with other research suggesting that the sharing of information within networks is the single most effective tactic for propelling the diffusion of innovations. A recent study looking at the social context of weatherization behavior found that social exchange of energy-related information is a predicting factor for weatherization “over and above the influence of one’s energy related knowledge” (Southwell and Murphy, 2014, p. 65.)

Utilizing network analysis such as that used in the WE project could provide interesting information on how best to communicate critical messages related to and reliant on human interest and behavior. For example, weatherization contributes to both mitigation of, and adaptation to, climate change. ORNL energy impact analysis and analysis of health benefits done separate from the WE project both show reductions in home energy consumption following weatherization. (For more on health benefits of weatherization, see “The Health Benefits of Weatherization,” HE Mar/Apr ’14, p. 36.) These reductions, in turn, lead to reductions in greenhouse gas (GHG) emissions. While multilevel climate change action plans are being developed, weatherization provides an immediate strategy; it has been cited as the top-performing home-related method for reducing GHG emissions (Dietz et al., 2009).

Further research is needed to identify other potential opportunities at the household level. How do social networks and the key members of those networks contribute to the adaptive capacity of a community? How can we maximize the potential of key members of those networks to contribute to the adaptive capacity of a community? How can we maximize their potential for improving human life on the largest scale by contributing to efforts to eradicate preventable disease?
Information that promotes energy efficiency and public health could be effectively packaged and diffused into the social networks of vulnerable populations.

Key Findings

We have found through visits to weatherized homes, occupant surveys, and interviews with researchers hired for this project that WAP can have a profound effect on the lives of the people it serves. From a parent’s newfound ability to pay utility bills to reduced emergency department visits for asthma attacks, the impact of weatherization on a household has the power to stimulate action.

Erin Rose is a researcher in the Environmental Sciences Division at ORNL. Beth Hawkins is a researcher in the Environmental Sciences Division at ORNL.

Bruce Tonn, researcher in the Environmental Sciences Division at ORLN, also contributed to this article.

We would like to thank the many people and organizations who contributed their time to the research described in this article. These include the pilot agency, the ten participating agencies, and the 74 trained interviewers. We’d also like to personally thank the social scientists who assisted with research design and implementation. They are Inga Treitler of Anthropology Imagination, Claire Cowan of the Energy Center of Wisconsin, and Brian Conlon with ORNL.

• 1
• FIRST PAGE
• PREVIOUS PAGE
• NEXT
• LAST

© Home Energy Magazine 2023, all rights reserved. For permission to reprint, please send an e-mail to contact@homeenergy.org.

Discuss this article in the Weatherization group on Home Energy Pros!