Section 1. How to Conduct Research: An Overview

Learn about the need for effective advocacy research, and gain skills in conducting such research, in order to be able to back up your statements with facts and statistics.

 

There are all kinds of advocacy. Your goal may be to raise the profile of an issue, to provide for the needs of a particular segment of your community, to protect citizens from physical or economic harm, to stem a disease before it becomes epidemic in the area, or to expose and oust a corrupt official, among others. You may be concerned with human services, health care, education, the environment, economic development, political issues, human rights, or the overall physical, social, and economic health of your community. You may be trying to influence legislators and other policy makers, to hold officials and corporations accountable for their actions, to change the behavior or methods of health and human service providers, or to increase cultural awareness and competence among all members of the community.

Regardless of what your effort is aimed at, you'll need information before you can do anything else. How many working people in your community have no health insurance? What exactly is that company dumping into the river, and what effects does it have? How do you know that the grant-funded program you want to continue has been effective? Where are you going to get the material for that fact sheet you need for the public meeting on Friday?

In order to persuade legislators or other officials to change laws and policies, to provoke an official investigation, or to sway public opinion, you'll need information. To get it, you'll almost undoubtedly have to do some research. That could mean anything from combing through piles of documents in the stacks of a library to taking photos and talking to witnesses. Some of your searching may be tedious, and some may be exciting, but all of it will involve work.

What do we mean by conducting research?

The word "research" often conjures up images of someone sitting in a library squinting at a large book with very small print, or hovering over a beaker of nasty-looking liquid in a chemistry lab.

Don't let those images scare you, though - research really means doing some work to find the answer to a question. The question may be relatively simple: What percentage of people who visit the community health clinic have no other way to get medical treatment? It may also be extremely complex: What's the most effective way to provide health insurance for working people in the community who don't have it?

Different kinds of questions require different kinds of answers, and different kinds of answers require different kinds of research to ferret them out. Knowing what you need is the first step.

• If the answer you're looking for is one of known fact - What's the chemical composition of a certain substance? What effects does it have on humans? - then you're probably going to find it by looking through books or articles in a library, or by consulting an expert.
• If you're seeking reinforcement for your advocacy for a particular intervention - How effective is that health promotion program? Why use this method instead of that one? Why ban this substance in food products? - then you may have to conduct a study of some sort, or sift through existing records.
• If you're trying to find facts that will help you convince policy makers to move in a particular direction - How many people in the community are employed, but have no health insurance? - you may find yourself working in the municipal archives or studying census data.
• If you're searching for evidence of harmful and/or illegal action on the part of a corporation or government agency - Is that paper mill dumping waste directly into the river again? And why isn't the regulatory agency doing anything about it? - you may have to do some actual detective work: searching through documents, taking pictures, analyzing samples of river water, talking to employees.

Research encompasses all these methods and more. What kind of research you should do depends on what and what kind of information you need.

Why do research?

The advantages of whatever you're advocating for may seem obvious to you, but that doesn't mean they're obvious to others. You'll have to show them that your ideas will significantly benefit the community. Research can help you toward this goal in several ways:

• Research gives your advocacy substance. Your research adds facts and statistics to your belief and passion. The latter two are important, but they won't actually convince too many people who disagree with you. Hard evidence might.
• Research gives you new information to help make your case. Often, your research will turn up powerful information on your side that even you didn't know about. It will also help you determine how much funding is needed, approximately how long it will take to see results, and the likely consequences of doing nothing.
• Research can show you what's most likely to address your issue successfully. (As this item indicates, your research will probably serve multiple purposes. It may help you plan and design your effort as well as advocate for it.) As an advocate, you have to know exactly what to advocate for. What have others done that's worked? What affects the issue in an experimental situation? The answers to these and similar questions will put you in a position to choose and advocate for strategies that are apt to be effective.
• Research can provide you with anecdotes and examples to use. While statistics are most convincing in certain situations, one actual example is often more powerful than reams of data, because it makes the issue immediate and real. An anecdote doesn't prove a case, but it can make it easier for people to understand exactly what the issue is about. Hearing someone else's story makes people realize "That could be me," and gives politicians a way to explain to voters why they favor a policy or bill.
• Research can confirm what you were already sure of. You may "know" that you're right about a particular issue, but it brings a great deal more security to be able to say that all the experts in the field agree with you, or that studies have shown that what you're advocating for works really well.
• Research allows you to make cost-benefit arguments. If 85% of former inmates return to jail; and if it costs $65,000.00 a year to keep one person in jail; and if a program that keeps 85% of former inmates out of jail costs $20,000.00 a year per person - you do the math. The prevention program sounds expensive, but it will actually save the taxpayer an average of $28,500.00 a year per former inmate. Research will tell you the costs of both alternatives and the success rates of both at keeping people from returning to jail.

The math works this way: If you start with 100 former inmates, and expect that 85 of them will be back in jail, those 85 will cost $5,525,000.00 (85 x $65,000) annually to keep in jail. If the same 100 former inmates participate in the alternative program, it will cost $1,700,000.00 (85 x $20,000) for the 85 who are successful, and $975,000.00 (15 x $65,000) for the 15 who go back to jail, for a total of $2,675,000.00. The difference between $5,525,00.00 and $2,675,000.00 is $2,850,000.00, divided by 100 former inmates gives $28,500.00 per person, the savings from using the alternative program. And that's only in the first year, or however long the alternative program lasts: once the program is over, the 85 who are successful cost the public nothing. In fact, they're probably working and paying taxes. It's only the 15 still in jail that taxpayers are funding.

• Research gives you credibility. If you do your research well, it identifies you as a serious advocate who does the groundwork before you try to convince people of your position. It will make people more willing to listen to you, and to believe what they hear.
• Research can short-circuit the opposition. If you've done careful research not only on your own position, but on the opposition's position as well, you'll have the information to answer their charges and questions, and either to disprove their claims, or to make reasonable and logical arguments for the soundness of your position over theirs.
• Research sets you up as the expert on the issue. If you become known as the one with the right answers, people - legislators and other officials, concerned groups, the general public - will come to you with their questions and concerns. When you're recognized as the authority, your advocacy position becomes infinitely stronger.

When should you do advocacy research?

Any advocacy requires some basic research, but there are times when research is particularly valuable.

When you're trying to get legislation passed

You'll need research for two reasons:

• To show the need for the legislation. Federal funding for AIDS research increased when advocates were able to demonstrate that the disease had reached epidemic proportions, and that it affected not only gay men, but other segments of the population as well. Similarly, adult literacy advocates in Massachusetts were able to secure a commitment to adult education and an increased funding appropriation when they presented legislators with a list of more than 13,000 people on waiting lists for programs around the state.
• To show public support for proposed legislation. When legislators know that their constituents really want them to vote for a bill, or to address a particular issue, they'll generally do it, especially in an election year. If you can demonstrate - through surveys, interviews, or other means - that the public backs your effort, it puts you in a much stronger position.

When you're seeking to arouse community concern about an issue that needs attention

Often, the research needed here is statistical. What percent of teens drop out of high school in the community, and how does that compare to other, similar communities? Have there been increases in homelessness over the past year, or the past several years? Are there more homeless families than there were? Has youth violence increased in the community, and, if so, who are its victims? Numbers are what's required here, so that people can see trends in issues of importance to the community.

Do numbers lie? Mark Twain said that "there are lies, damned lies, and statistics." By that he meant that you can make the same statistics seem to mean different things by manipulating them in different ways. If there has been a five-year upward trend in homelessness, for instance, one group may state that and stop there, while another may point out that the trend in fact peaked three years ago, and has been heading downward since. It may not have gotten down to where it was five years ago, but it's headed in that direction. So two groups, using the same statistic, can come to different conclusions: that growing homelessness is a problem; or that homelessness is on the decline, and is not a problem.

An intelligent opponent - and, if you have opponents, you should always assume they're at least as smart as you are, and that they'll do their homework - could demolish your argument if you aren't careful how you use the numbers you find. Your research has to be impeccable - always make certain that information that seems to prove your point
actually does so, that you don't exaggerate to make your case stronger, and that you pay attention to anything that seems to show the opposite.

You should never have to argue about whose numbers are "right." Yours should always be right, because you've checked them carefully, and analyzed them so you know what they really mean. If they don't agree with what you're advocating, you can reframe your issue - present it in a different way, or restate the problem it presents. In the case of homelessness, for instance, you may want to argue that whether it's rising or declining, it's still a serious problem that the community should tackle: "One homeless person in our community means we're not doing what we should. Even one is too many, and we have 450."

If your research in fact shows that you're dead wrong about what you're trying to prove, you may have to rethink it. No matter how much you may "know" emotionally that you're right, if the evidence says otherwise, you have to admit it and make sense of that.

When important programs or services or whole groups of people are under attack

Some examples:

• Food stamps and fuel assistance programs are often characterized by opponents as "unnecessary." Employing research to show how many people in the community need and use them helps voters understand why they are necessary.
• Immigrants may be blamed for employment shortages, when in fact those shortages are usually the result of economic trends and factors that have little to do with immigration. Furthermore, many immigrants take jobs that employers have difficulty filling - low wage, dirty, and physically demanding (migrant labor, for example). Your research can demonstrate that those who accuse immigrants of responsibility for unemployment actually have no argument.

When government officials are corrupt or otherwise guilty of wrongdoing

Research can serve to find the proof that exposes corrupt, dishonest, or unethical officials. From the small town treasurer who helps herself to a few thousand dollars from municipal funds to Richard Nixon, whose coverup of the Watergate break-in cost him the Presidency, officials who violate the law are generally caught by careful research rather than by some dramatic event. Bob Woodward and Carl Bernstein's interviews with various participants in the Watergate affair, for instance, eventually led them to the source known as "Deep Throat" who blew the whistle on the Nixon administration.

When government or another entity is lying to the public

At some times, there are things the public, even in a democratic society, should not know. If, in World War II, the press had routinely published the locations of Allied ships, for instance, the enemy would have been extremely pleased - and the ships in question would have been sunk. When the government or another entity lies to the detriment of the public good, however, or to protect itself politically, that is another matter.

Two examples:

• Cigarette manufacturers for years produced phony scientific studies that "proved" that smoking was harmless and non-addictive, while reputable scientists invariably found the opposite. The reputable research convinced the Surgeon General's office, and a large majority of the American public, that profits were more important to the cigarette companies than people's lives, and smoking in America dropped sharply.
• During the Reagan administration in the U.S., James Watt, the anti-environment Secretary of the Interior, constantly found himself contradicted in the press by scholars or journalists who had researched the accuracy of statements he had made, and found them to be false. Ultimately, Watt became such a political liability that he had to be replaced. Had no one bothered to check his accuracy, he could have continued to misrepresent environmental issues for political reasons, at the expense of the public interest.

When it's necessary to prevent harm

The source of acid rain in the eastern U.S. was a mystery until research showed that it was caused by substances released into the air by factories in the Midwest. If you're wondering why there are no more fish in the local river, or if there's some question about your drinking water, you might want to look into acid rain effects, or spend some time in the town archives and in the library finding out what substances local factories are dumping in the river, and what chemical effects those substances have.

Environmental threats to public well-being aren't the only kinds of harm that research can help prevent. Looking into the effects of proposed government policies can save citizens from unfortunate social or economic consequences... or help to correct those consequences, if the policy is one that's been shown to be successful. Research into the relationships among corporations, accounting firms, and Wall Street analysts could perhaps have saved investors from such financial disasters as the collapse of Enron, one of the largest American corporations. Many thousands of middle-income citizens lost their retirement money in that debacle, either directly or through pension fund investments.

When it's important in order to further the public interest

In these situations, you may not be looking for anything specific. Health care advocates, for instance, often conduct research to determine a community's health care needs, and what could best meet those needs, so that they'll know exactly what to advocate for.

How do you do research?

Since, as we'll discuss in more detail below, there are many different types of research, there's no one set method for doing all types. We will, however, propose some general guidelines that apply to any kind of advocacy research, and try to give you some more specific pointers on conducting each of the types of research you're likely to encounter.

General guidelines for research

Ask for help

You don't necessarily have to do everything yourself. There may be many possibilities for assistance out there.

• Other activists. If you're concerned about an issue, the chances are others are, too. If you can find them, you may also find that much of the information you need is already available, or that they're willing to help you get it.
• Academics or other experts. Depending on what you need to find out, help may be as close as the relevant department of a local university. If, for instance, you need to understand the chemistry of a pollutant, a chemistry grad student or professor may be more than willing to help. You may be able to find someone who's an expert on a particular area of law or business or public health who can save you countless hours of searching on your own.
• Librarians. Libraries are obviously great sources of information, and librarians are usually both eager and able to help. That's their job, and many librarians view difficult research requests as interesting and enjoyable challenges. They, too, may be able to save you enormous amounts of time.
• Journalists. If the issue you're addressing, or your area of advocacy, involves an interesting story or something the public clearly needs to know (Is the community's drinking water safe?), an investigative journalist may be more than willing to work with you. Many of the biggest news stories of past decades - the questions about American intelligence gathering before the Sept. 11 terrorist attacks, for instance - have resulted from investigative journalism. Most reporters would love the chance for a big story, and they have vast information resources available.
• People inside an entity you're researching. If there's something illegal or unethical going on, there may be someone inside the entity whose discomfort with the situation will make him willing to talk to you.

Try to find out if someone's already done the research you're planning

There may already be studies relating to the issue you're concerned with, or someone else in the community may already have spent weeks finding the information you're looking for. Don't make life any more difficult than it has to be: look around before you begin.

Learn the basics about your issue

Your research will do you no good if it isn't accurate and to the point. If there are laws involved, make sure you have read and understood them before you look for evidence that they've been broken, or need to be changed. If you suspect that someone might be violating professional ethics, find out whether that profession has a formal code of ethics, and study it carefully. If there's an environmental issue, learn the science behind it, at least well enough so that you can explain it, and so that you can understand and counter arguments against your position.

If you read something you don't understand, find help. Consult an attorney or legislator, a science teacher, your brother-in-law - whomever you can find that actually knows about the area you need to understand. If your information is wrong, or if you don't fully understand what you're talking about, you'll undermine your position, and your advocacy will fall on deaf ears.

Know what you're going to use the information for

Knowing the reason for gathering the information will help you decide exactly what to look for, how much you need, and what form it needs to be in. Some common reasons for advocacy research:

• To call public attention to an issue. In this case, you might look for information (statistics or anecdotes) that will help people understand how the issue relates to their community.
• To demonstrate public support for an issue. A survey could show that the community overwhelmingly supports funding for a new school, for instance.
• To guide your advocacy efforts. Survey results that tell you what the public, or particular segments of the public, actually think about an issue will help you decide how to approach it.
• To bolster your argument with facts, theories, respected opinions, or others' experience.
• To counter opponents' false or mistaken claims or counter-arguments.

This is a situation in which your research must be, as we mentioned before, impeccable. Don't fudge anything - someone will be waiting for the opportunity to pounce on the least inaccuracy or error, and to use it to try to discredit your whole argument. Use only real, reliable figures, and don't over- or underestimate their meaning. Don't stretch a point - assuming that all high school dropouts are illiterate, for instance - or it will come back to haunt you. Don't mistake anecdote for fact: one story doesn't prove a point. Perhaps most important, don't give in to the natural tendency to overestimate the value of a piece of information that supports your point of view. If you can't make a strong case with the information as it is, you may need to reexamine your assumptions.

• To demonstrate that addressing the issue would have benefits for the community or society as a whole, as well as for those directly affected. You might want to show the benefits to business resulting from adult education programs, for instance, or the increase in local real estate values that increased spending on education could bring.
• To prove wrongdoing or questionable practice on the part of an entity or official. This might mean doing research to show that the entity was actually committing the wrong or harmful act - bribery, illegal polluting, stealing public funds - that you said it was; or it could mean finding evidence that what the entity freely admits doing - using a given substance in its products, engaging in particular practices in the Third World - is in fact wrong or harmful.

Be skeptical about your sources, and check all your facts twice

All sources of information are not equally reliable. Reputable news organs like the New York Times, the Washington Post, and Newsweek, for instance, demand that all stories be carefully checked before they're published. That doesn't mean that they never get anything wrong, but it does mean that they're less likely to than, say, the National Enquirer.

The Internet is a great source of information, but there's no guarantee that any of it is accurate. Anyone can establish a website and put on it practically anything she wants to, without regard to whether it's accurate or not. If you find information on the Internet, unless it's from the website of a generally reliable source - the New York Times, the Encyclopedia Britannica, your state university, the U.S. Treasury Department - it's best to be cautious about using it without first checking it elsewhere.

As mentioned several times in this section, if your facts - even facts that aren't necessarily an important part of your argument - are wrong, your opposition will make you pay for it. Even something as apparently reliable as a chemistry textbook might not be trustworthy. If it was published 30 years ago, it might contain "facts" that have since been reinterpreted. Always check your findings if your source is even the slightest bit questionable.

Be persistent

Information may be difficult and tedious to find. An agency or organization - not necessarily one you're researching - may throw up roadblocks. You may have to cut through mountains of red tape just to gain access to a paper you want to see, or to get some small piece of information. It's common for citizens' groups to have to sue under the Freedom of Information Act (see below) to get the government to release public documents. Keeping at it is perhaps the most important research tool there is.

Types of research, and some suggestions for each

"Academic" research

This term refers here to seeking facts, general information on a topic, historical background, study results, etc., that have been published, or exist in public documents. There are a number of ways to get this kind of information:

• Use libraries. Both public libraries and more specialized ones - university libraries, law libraries, genealogical libraries, Presidential libraries, etc. - are open to the public for research on some basis (sometimes free, sometimes for a fee or membership charge, for instance). Librarians at all types of libraries tend to be both knowledgeable and helpful. If you have easy access to Washington, D.C., New York City, or Boston, the Library of Congress, the New York Public Library, the Boston Public Library, and the Harvard University library system are four of the largest and most comprehensive in the world.
• Take advantage of newspaper archives. Newspapers keep archives - generally consisting of a complete collection of past copies of the newspaper - in many cases going back over a hundred years. Many are now on microfilm or electronic media, and are often available through library information services or on the Internet (sometimes for a fee).
• Consult government and other archives. Federal, state, and local governments in the U.S. keep archives of government documents. They are generally available for research to any citizen, but, in some cases, you may have to travel to a particular place (Washington, DC, or a regional archive branch, e.g.) to find the documents you're looking for. Some organizations, corporations, and institutions also keep archives that may be helpful.

Use the Freedom of Information Act to gain access to public documents. The Freedom of Information Act (FOIA) provides for public access to all executive branch documents and records of the U.S. federal government, except those whose release might jeopardize national security or compromise the safety or privacy of individuals, or which are otherwise exempt under the provisions of the law. FOIA applies only to federal government documents, however, and only to those of government agencies and other parts of the executive branch. It doesn't apply, for instance, to the records of Congress, a corporation, a private organization or institution - for-profit or nonprofit - that receives no public money, or a private citizen (although some of these may be public documents as well).

However, most states do have their own versions of the FOIA (often called Public Records Acts) that cover state and local agencies and officials. For links and more information about individual states, the Freedom of Information Center at the University of Missouri, which also has information on FOIA.

Other government and corporate documents

Use other government and corporate documents. There are a number of websites of watchdog organizations that try to hold governments or corporations accountable for what they do. They keep track of the activities of these entities, and try to publicize them when they are not in the public interest.

• Use census data. Demographic (population, age, race, education, income, etc.) and a surprising amount of other information is available by census tract, town, county, state, etc. from the U.S. Census.
• Surf the Internet. The Internet, with literally millions of websites from all over the world, is probably the single largest source of information on the planet. It may also be the least reliable, in that, as explained above, there is no control on what information may be posted. You have to be cautious, unless the website is one that is known to be reliable and accurate. Given that caution, however, there is almost no limit to what you can find if you search persistently enough.
• Ask sympathetic experts. People who are experts in the issue or field you're involved with, and who are sympathetic to your cause (or simply interested in what you're doing) are often willing to provide you with information, or even conduct some of your research pro bono (as a public service).

Interviews and conversations

When the information you're looking for concerns the community's past experience with an issue, the relationships between key people, what's happening inside that factory, or someone's personal experience, the best way to find it is usually by talking to individuals. You may use informal conversations, structured interviews, or something in between, but the purpose in all cases is to get as much information as possible. How to find and talk to the right people:

• Start with people you know. Even if they themselves may not have the information you're looking for, the folks you know - who trust you and can vouch for you - can lead you to those who do. Furthermore, your friends and acquaintances may have helpful information you haven't even thought of asking for.
• Network like crazy. Ask everyone you interview to recommend others who might be helpful. Put the word out to networks you're already a member of (friends, clubs, e-mail lists, professional associations, church groups, etc.) You never know just who might have that one crucial piece of information, or know the person who has it.
• Decide whether you'll get better results from informal conversations or structured interviews. In some cases, people are more likely to open up in a relaxed and informal atmosphere. In others, you'll get better information if you use a formal interview. Where people are likely to be mistrustful of strangers, it can be a toss-up: a formal structure may help them focus on the content of what they're saying, but it may also increase the distance - and the mistrust - between them and the interviewer. Often, you have to make a gut decision about what will work best.

Structured interviews

A structured interview is constructed in a particular way in order to yield particular information. Formal studies that depend upon interview information from participants usually use structured interviews in order to assure to the extent possible that all participants respond to the same questions. In informal research, you might not be quite as careful, but the structured format really does assure that various people's answers to the questions are comparable.

In constructing such an interview format, you should:

• Think about and carefully define the information you need.
• Ask directly for the information you're seeking.
• Make questions clear and unambiguous. Think out what you're asking, so that questions aren't vague or subject to too much interpretation (unless interpretation is what you're looking for).
• Ask open-ended questions. These are questions that require an "essay" answer, rather than a yes-no response. For example, instead of asking "Did you enjoy being in the program?" you might ask "What did you like and not like about being in the program?" or "What was participating in the program like?"
• Probe. Ask follow-up questions to get at what people are really saying, or to keep them talking about a topic. ("Why did you like it when the teacher asked your opinion?") Don't be afraid to pursue what may seem to be a sidetrack. Sometimes the best or most important information lies off the beaten path.
• Don't cut people off too quickly. Their stories, or what you can read between the lines, may give you information as important as what they tell you directly.
• Confirm what you're told by checking with others to the extent that you can. Remember that you're getting people's perceptions, which aren't always the same as objective reality. Rashomon, a film by the great Japanese director Akira Kurosawa, examines the nature of reality. In it, an incident is described from the perspectives of four participants, each of whom sees it totally differently. In fact, the phenomenon of Rashomon lurks everywhere: get everyone's side of the story

Surveys

If you want to know what most people in the community think about something, or how many people would take advantage of a service if it were available, a survey is a way to reach a lot of people quickly. A survey usually consists of a list of simple questions on a topic, and may include as well some chance for respondents to express a broader opinion or comment on the issue. You can conduct surveys by mail, by phone, in person, by e-mail, on a website, or by making them available in public places (leaving fliers in doctors' offices, for instance, or publishing them in the newspaper) and providing drop boxes for returning them. Each method has its advantages and disadvantages.

Here are a few key points to remember in constructing surveys.

• Make questions or items clear and easy to respond to. If a question is vague, the answers to it will be vague, too, and will vary so much as to be all but useless. In addition, vague questions are harder to answer, and therefore make people less likely to complete the survey.
• Ask for the information you need. Think carefully about what you're actually trying to find out, then design questions that will give you that information.
• Don't ask for information you don't need, just because it might be interesting, or because you haven't considered your purposes carefully. If you're not going to use the information in a substantive way, don't ask for it.
• Keep it as short and simple as possible. The less time it takes to complete a survey, the more likely people are to do it.
• Translate your survey into other languages, if you want to reach other-language populations.

Randomization and self-selection

Try to ensure randomization of your sample. If you want to get an accurate idea of the community opinion - or, more likely, the diversity of community opinion - about anything, you have to get a cross-section of the community to respond to your survey. That means males and females from all walks of life, all ages, races, cultures and ethnicities, etc. Professional pollsters - who generally use only a small percentage of a large population to stand in for the whole group - make sure they get a random sample of people in a community by using such strategies as picking every tenth name in the phone book, and/or making sure that calls or mailed surveys go to all parts of the community proportionally.)

• Be aware of the implications of self-selection. People who are willing to respond to a survey are almost always self-selected, meaning that they are the ones who are interested enough to care about the issue the survey concerns. The responders probably know more about the issue to begin with, may have strong feelings about it, and don't necessarily represent the opinions of the majority of the community. If your respondents are all or mostly self-selected, your survey won't really give you the information you're looking for. Phone surveys are somewhat more likely than most other methods to get responses from people contacted, thus at least partially dealing with the issue of self-selection, but they require a lot of people-hours (and often money).

A special case of survey technique, one that shares some elements of a survey and some of a structured interview, is a focus group. This is a group, recruited either from a random sample or from a particular population, that then engages in a structured discussion with a facilitator who tries to gain specific information from the group members without telling them directly what that information is. (The idea is to get information that's more accurate because group members won't edit what they say in relation to the topic.) Focus groups can be a useful research tool in some situations.

Studies

A study is an investigation, based on the principles of the natural or social sciences, of a phenomenon or issue. Studies are used for such purposes as looking at the effectiveness of an intervention, finding ways to prevent or treat a medical condition, or examining the economic and other effects of a social condition (homelessness, for instance) on a community and its members.

Studies can be conducted in different ways. For instance, they can be quantitative - with results dependent on numbers and statistical procedures - or qualitative - based on observations of behavior, participants' reports of how they or their lives have changed, etc. Some studies seek to understand cause and effect - what causes something else to happen. Others look for correlations - connections - between two factors. Still others are concerned only with the answers to very specific questions: How quickly can participants learn a new skill using method A as opposed to method B?

If you're planning to conduct a study, or to have someone else - a university researcher, a watchdog organization - conduct one for you, consider these important points:

• Determine what your resources will allow you to do. If you plan a study that's going to take lots of time and involve a number of people as researchers - interviewers, data enterers, etc. - then you'll need a way to pay them, or a source of volunteers who have the background to do those jobs well. Make sure you have the money, time, and skills available to conduct the study you want.
• Clearly define what information you're looking for. A study is usually focused on answering a particular question. The question you choose is extremely important in determining how useful the study results will be. If, for instance, you're trying to show the necessity for addressing a particular issue, what about that issue does it make the most sense to study? Its economic impact? Its impact on children? Its connection to some other factor, such as unemployment or lung cancer rates? The first step in conducting a useful study is defining what you're looking at.
• If you're using statistics, use them properly. There are a great many statistical procedures that can be used for analyzing information that comes in the form of numbers. It's important to choose the procedures that will give you the information you're looking for. It's equally important not to try to make your results mean anything more or less than what they actually mean.
• Design the study so that it in fact answers the question you're asking. The issue of study design is far too complex to discuss in detail here - a university library will have several shelves of books that deal with that topic - but be aware that a careless study design can yield results that either mean nothing, or that answer a different question from the one you meant to ask.

For example, in studying a group of people, you may find that people who eat a low-fat diet live longer... but people who eat that diet may also exercise more, or take vitamins, or do something else that influences their health. A better-designed study would have at least two groups: one that ate a low-fat diet, and another with a very similar lifestyle (same amount of exercise, vitamins, etc.) that ate differently. A comparison of those groups would give more accurate information than looking at the target group and a random group. Even then, you have to be aware of the difference between correlation and cause and effect. If two factors go together - e.g., low fat diet and longer life - that doesn't necessarily mean that the one causes the other. The cause may in fact be genetic: certain genes might predispose people toward a low fat diet and longer life. Understanding that issues like these can be involved in research will help you prepare for countering opponents and bolstering your own argument.

A badly conceived or badly run study not only will yield no useful information, but can discredit your argument, or even your whole effort. If you're doing a study, and don't have the expertise yourself, then confer with experts, get professional researchers to help you, attach yourself to the coattails of a larger organization - do whatever it takes to make sure that the results of the study are accurate and will be respected.

The investigation of facts, events, etc., where finding information requires some detective work

In an interview, Don Kent, a legendary TV weatherman in Boston, told how he differed from others in the early TV days. They would check all their devices, call the US Weather Service, and determine that it was sunny, but would rain later that day. Kent would do all that, but then, a minute before the broadcast, he'd stick his head out the window, come on TV, and say, "It's raining hard in Boston at the moment, and it should continue into the evening." As a result, viewers trusted his weather forecasts.

For the most part, that's the kind of detective work we're talking about here. Do you want to know what the physical conditions in that housing project are really like? Go down there and look for yourself. Talk to tenants, walk around, and take notes or pictures, or both.

Don't rely on your media "experience" alone. We've become used to thinking we know what other neighborhoods or other places are like, but what TV - even TV news - or movies show us may be far from reality. Especially if you're gathering information to use as a base for advocacy, the best way to understand conditions is to explore them yourself: go there physically and observe first-hand.

Detective work may also involve sifting through documents of some sort. The real Erin Brockovich, whose story was told in the movie of the same name, went through water department records and other documents, looking for water quality ratings and for the chemical substances found in the water that people in Hinkley, CA, were drinking. Her detective work called for enormous attention to detail and amazing patience; but it paid off in the evidence that led to a huge settlement for people who'd been sickened and otherwise injured by the tainted water.

In movies, TV, and books sleuthing always sounds romantic, and always makes everything fit together neatly at the end. (One of the good things about the movie Erin Brockovich is that it didn't romanticize the detailed and painstaking nature of the work Brockovich did.) In reality, searching for evidence is usually tedious, seems to take forever, and often yields no information at all. It may mean combing through thousands of pages of public records, or sitting in one place for days watching an outflow pipe. That happy coincidence that so often leads to evidence on-screen or in the pages of the latest crime novel rarely occurs in real life.

Also contrary to TV and movie images, it's unlikely that anyone will try to harm you physically, or even to discredit you, if you investigate something. The greater hazards are that your search may take far longer than you thought, that no one will answer your questions, and that most of the process will be boring and frustrating. It takes persistence and concentration on your goal to carry out a successful investigation.

The search for stories and anecdotes to illustrate your issue

An important part of an advocacy effort can be the first-person stories of people actually involved in or affected by the issue at hand. Legislators, juries, and the public are often more easily swayed by stories they can identify with than by statistics or warnings, no matter how telling. Thus, finding people with compelling stories who are willing to tell them (and who can do so simply and powerfully) may be a crucial piece of advocacy research.

If you don't work with people who might be likely candidates, then it's best to confer with those who do - line staff, agency directors, volunteers, and others who have direct contact with those affected. They'll know whose stories are gripping, and who can pull an audience into understanding.

If the issue is not one where health or human service staff are likely to have contact with those affected, you have to use other networks to find appropriate stories. Personal acquaintances, key people in the target community, and other community activists may all be good sources.

In Summary

Effective advocacy requires research. Learning all you can about your issue allows you to back up your statements with facts and statistics, and arms you against the arguments of your opponents. It gives you credibility, and establishes you as the expert on the subject. When your advocacy is adversarial - aiming to prove harmful action on the part of a polluter and stop it, for instance - careful research helps you to prove your case, and to know how to change the situation once you've done so. In short, research is indispensable to successful advocacy.

Research is especially helpful in certain situations:

• When you're trying to get legislation passed, or put new policies in place
• When you want to make the community aware of an issue
• When necessary services or innocent groups of people are under attack
• When officials are corrupt
• When the government or others are lying to the public
• When you're trying to prevent harm to the public
• When you want to further the public interest

Research takes a number of different forms, from "academic" - involving finding information from published or original-source material available in libraries and archives, on the Internet, or in other, publicly-accessible places - to talking to people, to conducting surveys and studies, to engaging in real detective work.

Regardless of the type or types of research you do, however, there are some general guidelines:

• Ask for help
• Find out if someone else has already done the research, so you won't have to
• Learn the basic information about your issue, and know it cold
• Make a clear decision about what you want to use the information for
• Check everything twice: there are very few absolutely reliable sources
• Be persistent: there's no substitute for keeping at it

If you follow these guidelines, and do the work, your research will stand you in good stead when it comes time to reinforce your passion with evidence. You'll have it at your fingertips.