Fostering Sustainable Behavior

Barriers

We each have hunches about why people engage in activities such as walking to work, recycling or composting. For instance, theories regarding personal motivations for recycling abound. Recycling, it has been suggested, is popular because it serves to alleviate our guilt for not making the more difficult and inconvenient changes toward sustainable living. This hypothesis suggests that curbside recycling is simply an antidote to the guilt we feel when, for example, just after placing our recycling container at the curb, we hop into our own personal global warming factory and head off to work. Other theories suggest that individuals recycle because it is convenient, those around us recycle, it makes us feel good about ourselves, or we are simply badgered into it by our children.

Hunches regarding what motivates people to engage in sustainable behavior are important. These personal theories need to be identified for what they are, however: simply speculation. Speculation regarding what leads individuals to engage in responsible environmental behavior should never be used as the basis for a community-based social marketing plan. Prior to designing such a plan you need to set aside personal speculation and collect the information that will properly inform your efforts. To create an effective community-based social marketing strategy, you must be able to sort through the competing theories and discover the actual barriers that inhibit individuals from engaging in the activity you wish to promote. Once you have this information, you are well positioned to create an effective strategy. The purpose of this chapter, then, is to introduce methods for uncovering barriers.

Three Steps for Uncovering Barriers

Uncovering barriers involves three steps. You want to begin by reviewing relevant articles and reports. Following this review, focus groups are conducted to explore in-depth attitudes and behaviors of community residents regarding the activity. Building on the information obtained from the focus groups, a phone survey is then conducted with a random sample of residents. A phone survey can greatly enhance knowledge of the barriers to the behavior you wish to promote. If you have a consultant doing this research for you, it is wise to ask for an interim report at the end of these three steps in which information gleaned from the literature review is presented, results of the focus groups and phone survey are detailed, and promising social marketing strategies based on this research are identified. For organizations that typically have research done by consultants, this chapter is meant to provide information against which you can scrutinize proposals. If you are likely to do this work internally, this chapter will provide you with enough information to set out a clear research strategy. When combined with additional reading, this chapter will provide you with enough information to conduct your research in-house.(1)

1. Literature Review

Since the barriers to sustainable behavior are activity specific (see Chapter 1), the first step in designing a community-based social marketing strategy is to review relevant articles and reports. Prior to conducting your literature review, you should be clear on your mandate. If your position involves promoting mass transit over driving, then your literature search is already well defined. However, if you have a broad mandate, such as promoting residential energy or water conservation, to expedite your search you will need to further clarify your mandate before proceeding. Residential energy conservation, for example, can include behaviors as diverse as weather-stripping, adding additional insulation, installing clock thermostats, closing and opening windows, installing compact fluorescent bulbs, or planting trees.

There are four sources of information that you will want to tap into for your literature search.

• Thumb through trade magazines and newsletters for related articles. Often these articles are summaries of more extensive reports and can be good leads for where to search for in-depth information.
• You will want to find out what reports have been written on the topic by other communities. These reports are often difficult to obtain but are well worth the effort. Begin by contacting organizations that act as information clearinghouses for the behavior you wish to promote. For example, contacting the United States National Recycling Coalition, the Recycling Council of Ontario, or the Waste Watch Centre, can be invaluable if you are designing a waste reduction initiative.(2) If a relevant clearinghouse does not exist, call several well-connected individuals to trace down reports that have been prepared for other organizations.
• Search the databases of your local or closest university for related academic articles. Many of the articles that will be of interest to you can now be found by electronically searching databases. When you conduct these searches, pay particular attention to recent review articles that synthesize the current state of knowledge on the topic. At this site you will find a searchable database of academic articles on fostering sustainable behavior. You can search this database by behavior and/or the behavior change tools described in subsequent chapters. This website also contains a discussion forum where you can exchange ideas and ask questions of others who are involved in designing programs and/or conducting research in a particular area.
• Once you have reviewed the reports and academic articles that you have found, call the authors of studies that are of particular interest. Often these individuals will have pre-press publications that you will not be able to find elsewhere. Further, they may currently be engaged in research that can inform your efforts. Academics can be a particularly useful resource for tracking down research articles and reports that you may have missed in your previous searches. Mention the studies you have found and ask if there are other studies of which you should be aware. Often they will be willing to fax you a listing of relevant articles. Finally, ask if you can call back at a later point in your project to obtain further advice. Cultivating a good relationship with an academic who works in your area can assist you not only with keeping abreast of current literature, but also with issues related to analyzing your phone survey data and designing and evaluating your project.

Finally, if you are having the literature search done by consultants, ask that they search for relevant information in each of these four areas.

2. Focus Groups

The literature review will assist in identifying issues to be explored further with residents of your community through focus groups. A focus group consists of six to eight community residents who have been paid to discuss issues that your literature review has identified as important (when focus group participants are volunteers there is a strong likelihood that they are participating because they have a greater interest in the topic than others in the community). The participants for the focus groups are usually randomly chosen from the community. To select the participants, simply choose random phone numbers from the phone book. When contacting the potential participants, be sure to let them know how their names were selected. To ensure a good rate of participation, make it convenient for people to participate. Arranging transportation for participants and childcare can significantly increase participation rates. Remember, you want your focus group participants to be as representative of the community as possible. The more barriers that you remove to participating, the more representative your focus groups will be.

Focus groups provide an opportunity to discuss in-depth the perceptions and present behaviors of community residents relevant to the activity you are planning to promote. To maximize what you can learn from the focus group, you should come to the meeting with a set of clearly defined questions that have been informed by your literature review. You will want to begin the session by informing the participants that they were chosen at random to provide your organization with information about the relevant behavior. You will also want to reassure them that there are no right or wrong answers for the questions that you will be asking them and that what you are most interested in is their perceptions. You will want to remind them that their responses are confidential. Since you will be steering the conversation through the set of questions that you have created, you will want to have a co-worker act as a note taker.

Avoid sending information packages prior to conducting focus groups. If you provide information prior to running the focus groups your participants will no longer be representative of your community.

As the facilitator for the discussion it is important that you establish a supportive but firm role with the attendees. It is not unusual to have one or two members of a focus group attempt to monopolize the discussion and in so doing make other members feel that their comments are not important. Your role is to facilitate in such a way that less assertive members, or individuals who might have differing views, feel comfortable in speaking out. Prior to conducting your first focus group you will need to be comfortable with statements such as "I have received some very informative feedback from you, now I would like to hear what others have to say," and "I understand that you feel strongly about this issue, but I also know that some people have very different views on this matter, would anyone like to share them?" These statements assure participants that even if there are some belligerent or overly talkative members, you are ensuring that views of other members will be heard.

Remember that you are interested in people's views unadulterated by any information that you might present in your subsequent program. Therefore, avoid sending information packages prior to conducting focus groups (handing them out afterward is fine). If you provide information prior to running the focus groups, your participants will no longer be representative of your community.

When the focus groups are completed, you will want to summarize the comments that have been made. One effective technique is to tabulate the number of times that a specific comment was made, or agreed with, by members of the focus group. In general, you should pay close attention to comments that are made frequently (e.g., "I would compost, if I could be assured that it would not attract rodents").

3. Phone Survey

Focus groups are an essential step in enhancing your understanding of how community residents view the behavior you wish to promote. However, by themselves focus groups do not provide sufficient information upon which to base a social marketing plan. Focus groups are limited by the small number of participants, the impact that members of the focus group have upon one another, and the qualitative nature of the answers obtained. The small number of participants makes generalizing the results to the larger community unwise and, while interviewing participants in groups is cost-effective, members of a focus group can have a substantial effect on what opinions are expressed. Further, the qualitative data obtained in focus groups places considerable limits on the types of analyses that can be performed. Despite these limitations, focus groups provide valuable in-depth information about what issues residents see as important and also how they speak about the topic. As such, focus groups will help enrich your understanding of the activity you wish to promote, and ensure that a more comprehensive survey will be well constructed and that questions contained in the survey will be readily understood by the respondents.

Several methods are available for obtaining reliable information on the current beliefs and behaviors of community residents regarding the activity you wish to promote. These methods are person-to-person interviews, a mailed survey, and a phone survey. While personal interviews are capable of providing reliable and in-depth information, they suffer from two significant limitations; they are expensive to conduct and take a considerable amount of time to complete. To conduct person-to-person interviews, a random sample of residences would first be selected. Next, each of these homes would be mailed a letter introducing the purpose of the interview to them. Each household would then be called and, if willing, a time for an interview would be arranged. Paid interviewers would then travel to each home to conduct the interview. While this detailed process is occasionally warranted, conducting person-to-person interviews usually is an inefficient use of your resources.

In contrast, a mailed survey is much less expensive to conduct and the entire survey can be completed in a reasonable amount of time. However, mailed surveys have a major drawback: the number of people who will complete and return the survey, or what is referred to as the response rate, is often between 20% and 40%. Such a low response rate brings into serious question the representative nature or generalizability of the findings. Given the inconvenience of completing and mailing the survey, individuals who participate are likely more interested in your topic than those who elect not to participate. As a result, participants in a mailed survey provide an unrealistic picture of community attitudes and behavior.

Phone surveys have several advantages over mailed surveys and person-to-person interviews. First, compared with a mailed survey, it is possible to obtain a much higher response rate, providing a more accurate assessment of current community attitudes and behavior. While it is possible to obtain a much higher response rate, clearly not everyone will agree to participate. However, those individuals who choose not to participate can be asked to complete a brief refusal survey. A refusal survey consists of three to four questions that are also found in the complete survey (e.g., does your household compost). Further, the refusal survey normally takes no longer than half a minute to complete. Because the refusal survey is so brief, individuals who wish not to participate in the full survey frequently agree to complete the briefer refusal survey. By comparing responses of refusal survey participants with those of full survey participants, potential differences between participants and non-participants can be explored. If no differences exist between the two sets of responses, the results of the full survey can be more reliably generalized back to your community. If differences do appear, greater caution is warranted in generalizing the results.

In addition to providing a higher response rate than a mailed survey and the opportunity to conduct a refusal survey, phone surveys are less expensive to conduct and can be completed in a much shorter amount of time than can person-to-person interviews.

Additional advantages of phone surveys include:

• Random-digit dialling of community residents is possible (ensures a random sample of community residents);
• Phone access to otherwise difficult-to-reach populations is possible (e.g., high rise apartments, rural households).

Phone surveys are relatively easy to staff and manage. Compared with personal interviews, fewer staff are needed, the staff need not be near the sample geographically, and supervision and quality control are easier.

Seven Steps: Phone Survey

Items to include in your phone survey will be guided by your literature review and the focus groups. But how do you begin to write the survey? Writing a well-constructed survey takes time and patience. Here are some guidelines to make that process easier.

Step One: Clarify your Objective

Begin by writing a simple paragraph that describes what the survey is meant to accomplish. This paragraph has two purposes. First, it will force you to be clear on what the survey is to measure. Second, once you have it completed, you can show it to others involved in the project. You will be spending considerable time writing, conducting and analyzing the data from the phone survey. You want to make sure, before you begin this process, that those who have a stake in the results are all onboard regarding what the survey is to accomplish.

Following the example that I have used throughout this chapter, imagine that you are designing a community-based social marketing strategy for composting. You have two purposes: 1) To encourage people who are presently not composting to begin, and 2) To encourage seasonal composters to compost throughout the year. Given this background, your objective statement might read something like this:

Note that the objective paragraph for the survey indicates that there are two purposes, one of which is more important than the other. Giving priorities to different objectives of a survey can assist you later in deciding how many questions to devote to each task that the survey is to perform. Also note that comparisons between three groups are called for. In other words, your sample will need to contain three groups: year-round composters, non-composters and seasonal composters

Step Two: List Items to Be Measured

Once you are happy with your "survey objective statement," the next step is to create a list of items that "might" be included in the survey. Note that at this time you are not concerning yourself with writing questions, only with determining the "themes" that will be covered in the questionnaire. Most of the items on your list should come from what you have learned from your literature review and from your focus groups. Once you have created a comprehensive list, organize it into logical groupings. Place items related to behavior together, group attitude items together and similarly group demographic topics. Finally, once you have grouped the items on your list, you are ready to check each item against your "survey objective paragraph." You want to determine for each item on your list if it furthers the purpose of your survey. In other words, does it help to determine any of the goals laid out in your objective statement? If it doesn't, it should be eliminated. When you have your list finalized, you are ready to begin writing the survey.

Step Three: Write the Survey

In writing the survey, you will want most, if not all, of your questions to be closed-ended. Open-ended items are difficult to analyze and greatly extend the length of your survey. Keep in mind that you will want to be able to complete your whole survey in 10 minutes or less. To be able to ask as many questions as possible in a short amount of time, you will want to use only a few types of scales in your survey.

Six or seven point scales are preferable to three, four or five point scales, in that they provide for a broader range of answers. Having a broader range is important, when most people are likely to be clustered at one end of the scale or the other. It is likely, for example, that on a four point scale most people would respond with a "3" or "4" regarding how frequently they recycle glass and food cans. However, when the scale is expanded to six items, answers will be more dispersed. Whether you use a six or seven point scale will depend upon whether you wish to provide respondents with a midpoint. Using an odd-numbered scale provides a midpoint that allows respondents who are divided in how to respond to select this option. However, the midpoint may also be selected by respondents who are unsure of how to answer. Whichever option you select, stay with it throughout the survey, to avoid confusion for respondents.

Note also that only the endpoints are spelled out for each scale (e.g., in question #1 "1-never" and "6-all the time"). Providing just the endpoints lessens the length of time that it takes to read the survey to the participants. Further, it allows you to assume that the distance between each of the items on the scale (e.g., 4 to 5) is equal. If you provide labels for each of the items on the scale, the respondent can no longer infer that the distance between each of the items is equivalent. For example, we understand that the distance between 5 and 6 is equal to the distance between 4 and 5. However, we can't assume equivalence with labels (e.g., Is the distance between "6-strongly agree" and "5-moderately agree" the same as the distance between "5-moderately agree" and "4-mildly agree" ?). Because the distance between the scale items is no longer equivalent when you apply labels, there are more limitations placed on how you can subsequently analyze the data.

Finally, note that instructions to the surveyor are typed in capital letters to distinguish them from what is to be read to the respondent.

You should not have to write the whole survey yourself.You may wish to include questions that were part of other surveys (just seek permission before doing so).

Further, you can use the demographics items in other surveys as guides for your demographic section (see sample demographic questions). Finally, as you write your survey, you will want to ask four questions of each question in your survey.

Step Four: Pilot the Survey

Once the survey has been written, pilot it with 10 to 15 residents. During the pilot, the wording and order of questions in the survey can be scrutinized. Questions that respondents find confusing or difficult to answer can be rewritten before the full survey is conducted. Further, the pilot ensures that each survey can be conducted in under 10 minutes.

Miscalculations regarding the length of time that it takes to contact respondents or complete the survey can be very costly when it comes time to conduct the survey. Your pilot will help you to ascertain that your budget is realistic. Do not include the data you obtain from the pilot with the data you obtain from the actual survey.

Step Five: Select the Sample

Once you have completed the pilot and made whatever revisions are necessary, you are ready to obtain your sample. At this point you have two options. First, you may decide to have the survey completed by a survey research firm. Prices vary significantly, so shop around, but you can expect to pay at least $20 U.S. for each survey completed (in 1997 dollars). This price will include all charges, including conducting the survey, the refusal survey, and entering the data into a spreadsheet for data analysis.

If you decide to conduct the survey yourself, you may wish to have a firm provide you with a list of randomly derived residential phone numbers for your community. How many people should you sample? There is no easy answer to this question and here is where cultivating a good relationship with an academic working in the field can be of assistance. The size of the sample and how it is obtained will determine how confident you can be in your results. However, there is one other issue that will determine the sample size needed. Certain types of statistical analyses require a minimum number of participants for each barrier investigated (usually 10 to 12). Therefore, if you are designing a survey to look at composting, and you have 20 different barriers that you wish to explore simultaneously, you will need to complete roughly 200 surveys (20 X 10).

Step Six: Conduct the Survey

If you are doing the survey "in-house," you will need to train the people who will be making the calls. Click here for a set of instructions that you can use with your callers survey instuctions.

Step Seven: Analyze the Data

Many of the current statistical packages, such as the Statistical Package for the Social Sciences (SPSS) make analyzing data much easier than it was even a few years ago. Obtaining descriptive statistics, frequencies, and comparing means is now as simple as pulling down a menu and selecting the variables and analysis that you want. Gone are the days in which you had to write complex computer instructions to analyze data. The result is that basic statistics are now within reach of virtually everyone. However, you will want to go beyond obtaining the means and frequencies to lay the groundwork for your community-based social marketing campaign.

If you glance back at the survey objective statement, you will notice that the survey had two purposes: distinguishing between composters and non-composters; and distinguishing between year-round composters and those who compost seasonally. To answer these two questions requires multivariate statistics; such as multiple regression, discriminant analysis or logistic regression. Multivariate statistics allow you to determine the factors that distinguish householders who compost from those who do not, and also enables you to analyze the relative importance of these factors. For example, a recent study that I conducted with a former student, Laurie Beers, utilized discriminant analysis and revealed the following five factors were most important in distinguishing year-round composters from non-composters. (3) Note that these factors are presented in order of importance:

• Those who composted reported a greater desire to reduce the amount of waste they produced than did non-composters.
• Non-composters perceived composting to be a more unpleasant activity than those who composted (e.g., they associated it with unpleasant odors, flies, rodents).
• Composters perceived the activity to be more convenient than did those who did not compost.
• Those who did not compost believed that they did not have the time to compost.
• Composting households reported recycling glass and cans more frequently.

Knowing which factors are most important in distinguishing individuals who have adopted a sustainable behavior from those who have not is an essential first step in developing a community-based social marketing strategy. The above results provide a clear indication of some of the barriers that would need to be surmounted to encourage more people to compost. For example, perceptions that composting is unpleasant, inconvenient and involves a significant investment of time are important issues that a community-based social marketing strategy would need to address.

Analyzing the data using multivariate statistical techniques is an essential aspect in the development of a sound marketing strategy. Less sophisticated statistical approaches such as calculating means or correlations are limited in their ability to provide information on the relative importance of the factors that lead individuals to engage in the behaviors of interest to you. Unless you or someone else in your organization has a background in statistics, you will want to obtain assistance at this point. Many graduate students are trained in multivariate statistics and with a few phone calls you should be able to find someone who will do your analyses for you. Don't be daunted at this point. While the statistical techniques that are needed require someone who is statistically sophisticated, as can be seen above, the results of these analyses can be presented in a straight-forward, understandable format.

If you are having a consultant do this work for you, you should ask for a report at this point that details the results of the focus groups and the phone survey. Further, based upon these results, request that the report detail promising social marketing approaches.

Some Closing Thoughts

Identifying barriers is an essential first step in designing a successful program. While significant pressures exist to skip this step, the simple truth is that it is impossible to design an effective strategy without identifying barriers. In our experience, the four most common reasons for skipping barrier identification include:

• belief that the barriers to the activity are already known,
• time pressures,
• financial constraints, and
• managerial staff who do not support conducting preliminary research.

Believing that the barriers to an activity are already known is very difficult to guard against. By our very nature we develop theories about why people behave as they do. If we didn't, we would find it very difficult to understand and interact with others. This tendency to develop theories about the behavior of others, can lead to a strong sense of self assurance that the barriers to an activity are already well understood. Research in social psychology convincingly demonstrates, however, that once we have developed a "hunch" we tend to pay attention to information that supports our view, and discount or disregard information that would contradict it. As a consequence, we can come to believe very strongly in our own personal theories even though they may have no factual basis. To be an effective community-based social marketer requires a healthy dose of skepticism about your own and others' personal theories.

Conducting preliminary research to identify barriers takes time. In a well organized project you can expect the identification of barriers to add two to four weeks to the development of a strategy. However, the length of time required to identify barriers pales when compared to the time and effort involved in having to design and deliver a new program if the first is unsuccessful. Similarly, while identifying barriers adds to the expense of delivering a program, there is a high return on investment given the much greater likelihood of delivering a successful program.

Building support among managerial staff will often involve dealing directly with the above three concerns. Time and cost concerns can often be dealt with by noting, as discussed above, that identifying barriers will usually save both time and money by lessening the likelihood of having to mount multiple campaigns. Managers, like everyone else, develop theories about behavior and are just as prone to believe that they already know the barriers to the activity you are to promote. There is a strong likelihood that they may ascribe to either the attitude-behavior or economic self-interest approaches discussed in the previous chapter since these perspectives are widely accepted. Finally, arrange, if possible, for managerial staff to read this guide or attend a workshop on community-based social marketing. In Canada, where the first edition of this book has been widely read, and workshops on community-based social marketing have been attended by a large number of managers, community-based social marketing is increasingly being specified by management as the method by which programs must be delivered.

Once you have identified the barriers to the activity you wish to promote, you will want to consider what behavior change tools you can use to overcome these barriers. Chapters 3 through 8 introduce a variety of tools that you can incorporate into the programs you design.