Minneapolis domestic violence experiment criticism

"fl l.)cserihc fhp irnnortance of

At the end of the semester, a professor asked tf I utould be interested in doing some research on sexualharassment in the workplace for her ouer the summer. For the research,I had to read research articles and summarize them for the professor.While I was reading the articles, I would come ocross the research methods the authors used, which included multivariate analysis, OLS Regression analysis, SPSS, and ANOVA tables. I thought it u)as incredible how I came full circle back to the research methods I learned! My research methods class set me on a course that has changed my time in college and possibly inJlu- enced my future career.

Emily G. , Student

! ntimate partner violence (IP\) is a major problem in countries around the I world. In a tI.S. survey of 1 6,507 men and women sponsored by the Department ofJustice and the Centers for Disease Control and Prevention, 3 5 .6% of women and 28.5%" of men said th.y had experienced rape, physical violence, or stalking by an intimate partner at some time in their lives (Black et aI. 2011).A" interna- tional survey by the Wbrld Health Organizatton flMHO) of 24,000 women in 10 countries estimated lifetime physical or sexual abuse ranging from a low of L5% inJapan to a high of 7l% in rural Ethiopia QVHO 2005) (see Exhibit 2.1).

What can be done about this problem? In 1981, a historic experiment was funded by the Police Foundation and the Minneapolis Police Departrnent to determine whether immediately arresting accused spouse abusers on the spot would deter funrre offending incidents. For misdemeanor cases, the experimental course of action involved the random assignment of police to respond by either arresting the suspect or giving the suspect a simple warning. The experimental treatment, then, was whether the suspect was arrested, and the researchers wanted to know whether arresting the suspect was better than not arresting the suspect in reducing recidiaisvn (subsequent assaults against the same victim). The study's results, which were widely publici zed, indicated that arrest did have a deter- rent effect. Pardy as a result of these findings, the percentage of urban police deparunents that made arrest the preferred response to complaints of domestic violence rose from IA% in 198+ to 90"/" in 1988 (Sherman 1 992). Six other cities later carried out studies similar to the Minneapolis e4periment (collectively, this was called the Spouse Assault Replication Program [SARP]), but city to cit1z, the results were mixed (Buzawa and Btzawa L996; Hirschel, Hutchison, and Dean 1992; Pate and Hamilton 1992; Sherman 1992; Sherman and Berk 1984).In some

THE PROCESS AND PROBLEMS OF RESEARCH RELATEDTO CRIME AND CRIMINOLOGY

t, 60 (E +. tr oo L

t40

.$'p$$*:-t$$l.'1.-'$5.'.:.* qtQt'

iEffi Ever experienced physical violence n Ever experienced sexual violence 1 Ever experienced physical or sexual violence, or both

Source: World Health Organization 2005. Multi-country Study onWomen's Health and DomesticViolence: Summary Report.

cities (and for some people), arrest did seem to prevent future incidents of domestic assault; in other cities, it seemed only to make matters worse, contributing to additional assault; and in still other cities, arrest seemed to have no discernible effect. After these replications of the original Minneapolis experiment, peo- ple still wondered, 'Just what is the effect of arrest in reducing domestic violence cases, and how should the police respond to such cases?" The answer simply was not clear. The Minneapolis experiment, the studies modeled after it, and the related contro- versies provide many examples for a qzstematic overview of the social research process.

ln this chapter, we shift from examining the wlry of social research to an overview of the bow-the focus of the rest of the book. We will consider how to develop a question for social research and then how to review the existing literature about this question while connecting the question to social theory and, in many studies, formulating specific testable hypotheses (see Exhibit 2.2).We will then discuss different social research strategies and standards for social research as a prelude to covering the details about these stages in subsequent chapters. You will find more details in Appendixes A and B about reviewing the literature. We will use the Minneapolis experiment and the related research to illustrate the different research strat- egies and some of the related techniques.The chapter also enpands on the role of social theo- ries in developing research questions and guiding research decisions. By the chapter's end, you should be ready to formulate a research question, critique previous studies that addressed this question, and design a general strategy for answering the question.

Beplication:

Thc ahiliti, sl '*.{t *{f,ii* tit*{}y w e'rf}*r1ffi*fr!,trs b* dupli*al*d,

CHAPTER 2 o THE PROCESS AND PROBLEMS OF RESEARCH RELATED TO CRIME AND CRIMINOLOGY 27

Exhibit 2.1 International Prevalence of Lifetime Physical and Sexual Violence byr an Intimate ' Partner, Among Ever-Partnered Women by Site

ffi,@)@,@)@+miffi IDENTIFYING A RESEARCH QUESTION

How does a researcher interested in criminology and criminal justice-related issues decide what to study and research? A research questiou in criminology and criminal justice-related sciences is a question about some aspect of crime or criminals that you seek to answer through the collection and analysis of firsthand, verifiable, empirical data. The types of questions that can be asked are virtually limidess. For example, "Are children who are violent more likely than nonviolent children to use violence as adults?" "Does the race of a victim who is killed influence whether someone is sentenced to death rather than life imprisonment?" "Why do some kinds of neighborhoods have more crime than others? Is it due to the kinds of people who Iive there or characteristics of the neighborhood itselfl" "Does community policing reduce the crime rate?" "IIas the U.S. government's war on drugs done anything to reduce the use of illegal drugs?" So many research questions are possible in criminology that it is more of a challenge to specif, what does not qualif, as a social research question than to speci$, what does.

But that does not mean it is easy to specifi, a research question. In fact, formulating a good research question can be surprisingly difEcult. We can break the process into three stages: identifiring one or more questions for study, refining the questions, and then evaluat- ing the questions.

Where to Start? Formulating a research question is often an intensely personal process in addition to being a scientific or professional one. Curiosity about the social world may emerge from your "per- sonal toubles," as Mills (1959) put it, or your personal experiences. Examples of these trou- bles or experiences could range from an awareness you may have that crime is not randomly distributed within a city but that there seem to be "good" or safe pars oftown and "bad" or unsafe areas. Can you think of other possible research questions that flow from your own experiences in the world?

The research literature is often the best source for research questions. For example, other researchers may also pose interesting questions for you to study. Most research articles end with some suggestions for additional research that highlight unresolved issues. For exam- ple, Sherman etal. (1992) concluded an article on some of the replications of the Minneapolis experiment on police responses to spouse abuse by suggesting that "deterrence may be effec- tive for a substantial segment of the offender population. . . . However, the underlying mech- anisms remain obscure" (706). A new study could focus on the mechanisms: Why or under what conditions does the arrest of offenders who are employed deter them from future crimi- nal acts? Exacdy what occurs when someone is arrested for domestic violence that may lead him or her not to be violent against a spouse in the future? Is it the brute fear of being arrested and having to go to jail? Is it the fear that one's employer may find out and fire him or her? Is it the fear that members of the communiq may learn about the arrest and the offender may

Besearch questitln:

A qusstir:ri ahout $c,rfir)

a$p*fft of crirH*, firiffiinal$,

*r Ih,* crinrir:al justic*

$ysteffi, the ans,,ryer t* whi*h is s,ruuhi through rolixcti*r, anrJ anal tsis *i {rrsthanri, ve rifiahlr, *rililiri*'al dala"

28 SECTION I . FOUNDATIONS FOR SOCIAL RESEARCH

Exhibit 2.2 Launching a Research Project

lose his or her good standing in the neighborhood? Is it all these? Any issue of a journal in your field is likely to have comments that point toward unresolved issues.

Another source for research questions is theory. Many theoretical domains are used to inform research questions in our discipline, including sociological, psychological, and criminological theories. Some researchers spend much of their careers conducting research intended to refine an answer to one central question. For example, you may find rational choice theory to be a useful approach to understanding diverse forms of social behavior, such as crime, because you think that people seem to make decisions on the basis of personal cost-benefit calculations. So, you may ask whether rational choice theory can explain why some people commit crimes and others do not or why some people decide to quit committing crimes while others continue their criminal ways.

Finally, some research questions adopt a very pragmatic rationale concerning their research design. You may focus on a research question posed by someone else because doing so seems to be to your professional or financial advantage. For instance, some researchers conduct research on specific questions posed by a funding source in what is termed a request for proposals (RFP). (Sometimes the acronyrn RFA is used, meaning reqaest for applications.) Or you may learn that the public defenders in your ciq are curious as to whetler they are more successful in getting their clients acquitted of a criminal charge than private lawyers.

Refi ning Research Questions As you may have guessed, coming up with interesting criminological questions for research is less problematic than focusing on a problem of manageable size. We are often interested in much more than we can reasonably investigate with our limited time and resources (or the limited resources of a funding agenry). Researchers may worry about staking a research project (and thereby a grant) on a particular problem, so they commit to addressing several research questions at once, often in a jumbled fashion. It may also seem risky to focus on a research question that may lead to results discrepant with our own cherished assumptions about the social world. In addition, the prospective commitment of time and effort for some research questions may seem overwhelming, resulting in a certain degree of parallsis (not that the authors have any experience with this!).

The best way to avoid these problems is to develop the research question one bit at a time with a step-by-step strategy. Do not keep hoping that the perfect research question will sprilg forth from your pen. Instead, develop a list ofpossible research questions as you go along. At the appropriate time, you can look through this list for the research questions that appear more than once. Narrow your list to the most interesting, most workable candidates. Repeat this process as long as it helps improve your research questions. Keep in mind that the research you are currendy working on will likely generate additional research questions for you to answer.

Evaluating Research Questions In the third stage ofselecting a criminological research question, you evaluate the best can- didate against the criteria for good social research questions: feasibility (given the time and resources available), social importance, and scientific relevance (IGng, Keohane, and Verba tee+).

The research question in the Minneapolis Domestic Violence Experiment, "Does the formal sanction of police arrest versus nonarrest inhibit domestic violence?" certainly meets the criteria of social importance and scientific relevance, but it would not be a feasible ques- tion for a student proiect because it would require you to try to get tJle cooperation ofa police

CHAPTER 2 . THE PROCESS AND PROBLEMS OF RESEARCH RELATED TO CRIME AND CRIMINOLOGY 29

deparrnent. You might instead ask the question "Do people (students) think that arrest will inhibit domestic violence?" This is a question that you could study with an on-campus sur- vey. Or perhaps you could work out an arrangement with a local battered women's shelter to study the question "What leads some women to call the police when they are the victims of domestic violence, and why do they sometimes not call?" A review of the literature, however, might convince you that this and other questions may not be scientifically relevant because they have been studied enough.

Feasibility

You must be able to conduct any study within the time frame and with the resources you have. If time is limited, questions that involve long-term change may not be feasible-for example, "If a state has recendy changed its law so that it now permits capial punishment for those convicted of murder, does it eventually see a reduction in the homicide rate over time?" This is an interesting and important question but one that requires years of data collection and research. Another issue is what people or groups you can expect to gain access to. Although well-experienced researchers may be granted access to police or correctional departrnent files to do their resbarch, less seasoned and lesser-known researchers or students may not be granted such access. It is also often difficult for even the most experienced ofresearchers to be given firll access to the deliberations of a criminal jury. For someone interested in white-collar crime, recording the interactions that take place in co{porate boardrooms may also be aboo.

The Minneapolis Domestic Violence Experiment shows how ambitious social research questions can be when a team of seasoned researchers secures the backing of influential groups. The project required hundreds of thousands of dollars, the collaboration of many social scientiss and criminal justice personnel, and the volunteer efforts of 41 Minneapolis police officers. But don't v/orry; many worthwhile research questions can be investigated with much more limited resources. Of course, for this reason, the Sherman and Berk (1984) question would not be feasible for a student project. You might instead ask the question, "Do students think punishment deters spouse abuse?" Or perhaps you could work out an arrange- ment w'ith a local police departrnent to sflrdy the question, "How satisfied are police ofEcers with their treatrnent of domestic violence cases?"

Social lmportance

Criminological research is not a simple undertaking, so you must focus on a subsantive area that you feel is important and that is important to the discipline and,lor important for public poliry. You also need to feel personally motivated to carry out the study; there is litde point in tying to answer a question that does not interest you.

In addition, you should consider whether the research question is important to other people. Will an answer to the research question make a difference for society? Again, the Minneapolis Domestic Molence Experiment is an exemplary case. If that study had showed that a certain type of police response to domestic violence reduced the risk of subsequent victimization ,

^ grea;t deal of future violence could be prevented. But clearly, criminology and

criminal justice are far from lacking important research questions.

Scientific Relevance

Every research question in criminology should be grounded in the existing empirical litera- wre.By ground.ed,we mean that the research we do must be informed by what others before us have done on the topic. Whether you formulate a research question because you have been stimulated by an academic article or are motivated by questions regarding your own personal experiences, you must turn to existing criminological literature to find out what has already been learned about this question. Even ifyour research topic has already been investigated

SECTION I o FOUNDATIONS FOR SOCIAL RESEARCH30

by someone else, it would not necessarily be a bad idea for you to do research on the issue. It would be unreasonable to think of any criminological research question as being setded for all time. You can be sure that some prior study is relevant to almost any research question you can think of, and you can also think ofbetter ways to do research than have been done in the past.

The Minneapolis experiment was built on a subsantial body of contradictory theories about the impact of punishment on criminality (Sherman and Berk 1984). Deterrence theory predicted that because it was a more severe penalql arresting individuals would better deter them from repeat offenses than not arresting them. Labeling theory on the other hand, pre- dicted that arrest would make repeat offenses more likely because it would stigmatize offend- ers. Clearly, the Minneapolis researchers had good reason to perform another study. Prior research and theory also helped them develop the most effective research design.

SOCIAL RESEARCH FOUNDATIONS

How do we find prior research on questions of interest? You may already ftnow some of the relevant material from prior coursework or your independent reading, but that won't be enough. When you are about to launch an investigation of a new research question, you must apply a very different standard than when you are studying for a test or merely seek- ing to learn about domestic violence or any other topic. You need to find reports ofprevious investigations that sought to answer the same research question that you wish to an$tre! not only those that were about a similar topic. If there have been no prior studies of exacdy the same research question, you should find reports from investigations of very similar research questions. Once you have located reports from prior research similar to the research that you

31CHAPTER2. THEPROCESSANDPROBLEMSOFRESEARCHRELATEDTOCRIMEANDCRIMINOLOGY

Patrick r. Garr, PhD, Director, Program in Criminal fustice

Patrick J. Carr is the program director of the Program in Criminal Iustice, as well as associate professor of soci- ology at Rutgers

and violence that consumes so much of their daily lives, to ask questions of themselves and the people charged with running [Philadelphia], and to have a serious conversation with teachers, parents, city offi- cials, comrnunity leaders, state legislators, reporters, politicians, and anyone else who wants to know what is going on in the city to move forward on solutions inspired by the youth perspective." The Philadelphia Youth Solutions Project is a venue for Philadelphia's young people to offer their own expert advice on how to transform the city based on their experiences and perspectives.

Carr and Kefalas are ethnographic research- ers who seek to understand people's experiences through participating in their lives and interviewing them in depth. In anothel project, they investigated the experiences of young adults growing up in a small midwestern town by living in the town and sharing in community experiences. Their subsequent book was Hollouri ng Outthe Middle:The RuralBrainDrain and,What It IVIeans for America (2009).

It

more, he is an associate membe r of the MacArthur Foundation's Research Network on Transitions to Adulthood. He earned his PhD in sociology from the University of Chicago in 1998 and his master's degree in sociology from University College Dublin in 1990. His research interests include communities and crime, informal social control, youth violence, and the transition to adulthood.

Carr and his wife, Maria Kefalas (Saint Joseph's University), are founders of the Philadelphia Youth Solutions Proj ect (http :/ /www.pysp.org) , which "offers a safe space for Philadelphia's young people to explain their views and emotions about the danger

l'l

Refereed journals:

J*urn'al* that s*lect rnscarth p;ifiors fcr publi*aii*n has*d i:n thr pe*r r*vi*rrys *f rsth*r r;rs*ial s*i*ntisis.

wish to conduct, you may expand your search to include investigations about related topics or studies that used similar methods. You want to be able to explain what your proposed study adds to prior research as well as how it takes into account what has already been learned about your research question.

Although it's most important when you're starting out, reviewing the literature is also important at later sages of the research process. Throughout a research project, you will uncover new issues and encounter unexpected problems; at each of these times, you should search the literature to locate prior research on these issues and to learn how others responded to similar problems. Published research that you ignored when you were seeking to find other research on domestic violence might become very relevant when you have to decide which questions to ask people about their attitudes toward police and other authorities.

Searchi ng the Literature Conducting a thorough search of the research literature and then reviewing critically what you have found lays an essential foundation for any research project. Fortunately, much ofthis information can be identified online, without leaving your desk, and an increasing number of published journal articles can be downloaded direcdy onto your own computer (depending on your particular access privileges). But just because thereb a lot available online doesnt mean that you need to fir,d it all. Keep in mind that your goal is to find reports of prior research investigations; this means that you should focus on scholarly journals that choose articles for publication after they have been peer reviewed by other social scientists-refereed journals. Newspaper and magazine articles wont do, although you may find some that raise important issues or that summarize social science research investigations.

Every year, the web offers more and more usefirl material, including indexes of the published research literature. You may find copies of particular rating scales, reports from research in progress, papers that have been presented at professional conferences, and online discussions of related topics. Web search engines wilf-also find academic journal articles that you can access direcdy online (although usually for a fee). Most of the published research literature will be available to you online only if you go through the website of your college or university library. The library pays a fee to companies that provide online journals so that you can retrieve this information without paying anything extra yourself. Of course, no Iibrary can afford to pay for every journal, so ifyou cant find a particular issue of a particular journal that you need online, you will have to order the article that you need through interlibrary loan oq ifthe hard copy ofthe journal is available, walk over to your library to read it.

As with any part of the research process, your method for searching the literature vrill affect the quality of your results. Your search method should include the following steps:

Specify Your Research Question. Your research question should be neither so broad that hundreds of articles are judged relevant nor so narrow that yorr miss important literature. "Is informal social control effective?" is probably too broad. "Does informal social control reduce rates of burglary in my town?" is probably too narrow. "Is informal social control more effective in reducing crime rates than policing?" provides about the right level of specificity.

ldentify Appropriate Bibliographic Databases to Search. Criminal Justice Abstracts is a great place to start, however, because the field is interdisciplinary in nature, Sociological Abstracts or SocINDEX may also meet many of your needs. If your focus is on psychological questions, you'll also want to include a search in the online Psychological Abstracts database, PsycINFO, or the version that also contains the firll text of articles, PsycARTICLES. Of course, many journals are listed in several databases, so it is best to begin a search in Criminal Justice Abstracts, and then move on to the others. It will save you a lot of time if you ask a

32 SECTION I . FOUNDATIONS FOR SOCIAL RESEARCH

librarian to teach you the best techniques for retrieving the most relevant articles to answer your questions.

To find articles that refer to a previous publication, such as Sherman and Berk's study of the police response to domestic violence, the Social Science Citation Index (SSCI) will be helpfirl. SSCI is an extremely useful tool for tracing the cumulative research in an area across the social sciences. SSCI has a unique "citation searching" feature that allows you to look up articles or bool,o, see who else has cited them in their worlg and find out which articles and boo[s have had the biggest impact in a field.

Create a Tentative List of Search Terms. List the parts and subparts of your research question and any related issues that you think are impoftant: "informal social control," "policing," "influences on crime rates," and perhaps "community cohesion and crime." List the authors of relevant studies. Specifi, the most important journals that deal with your topic.

Narrow Your Search. The sheer number of references you find can be a problem. For example, searching for "social capital" in July 2017 resulted in 7 ,67 5 citattorrs in SocINDEX. Depending on the database you are working with and the purposes of your search, you may want to limit your search to English-language publications, to journal articles rather than conference papers or dissertations (both of which are more difficult to acquire), and to materials published in recent years. If your search yields too many citations, try specifying the search terms more precisely (e.g., "neighborhood social capital"). If you have not found much literature, try using more general or multiple terms (e.g., "social relations" OR "social ties").

Whatever terms you search first, don't consider your search complete until you have tried several different approaches and have seen how many articles you find. A search for "domestic violence" in SocINDEX onJuly 30,2017,yielded 12,358 hits; by adding "effects" or "influences" as required search terms and limiting the search to peer-reviewed articles published since 2010, the number of hits dropped to 566. But focusing even more by add- ing "police response" resulted in two articles. So, if you are focusing on issues such as tlose in the Sherman and Burk study, you probably need to use a strateg'y a bit narrower than the next-to-last one.

lJse Boolean Search Logic. It\often a good idea to narrow your search by requiring that abstracts contain combinations of words or phrases that include more of the specific deails of your research question. Using the Boolean connector anil allows you to do this, while using the connector or allows you to find abstracts containing different words that mean the same thirg.

Use Appropriate Subject Descriptors. Once you have found an article that you consider appropriate, look at the "descriptors" field in the citation. You can then redo your search after requiring that the articles be classified with some or all of these descriptor terms.

Check the Resu/fs. Read the tides and abstracts you have found and identif, the articles that appear to be most relevant. If possible, click on these article tides and generate a list of their references. See ifyou find more articles that are relevant to your research question but that you have missed so far. You will be surprised at how many imporant articles your initial online search missed.

Locate the Articles. Whatever database you use, the next step after finding your references is to obtain the articles themselves. You will probably find the firll text of many articles available online, but this will be determined by what journals your library subscribes to and the period

33CHAPTER 2 o THE PROCESS AND PROBLEMS OF RESEARCH RELATED TO CRIME AND CRIMINOLOGY

for which it pays for online access. The most recent issues of some journals may not be available online. Keep in mind that your library will not have anywhere near all the journals (and books) that you run across in your literature search, so you will have to add another step to your search: checking the "holdings" information. If an article that appears to be important for your topic isnt available from your own library or online, you may be able to request a copy online through your library site or by asking a member of the library saff. You can also check http://worldcat.org to see which other librar- ies have the journal.

Take Notes on Each Article Your Read, Organizing Your Notes lnto Standard Sections: Theory, Methods, Findings, Conclusions. L *y case, write your review of the literature so that it contributes to your study in some concrete way; don't feel compelled to discuss an article merely because you have read it. Be judicious. You are conducting only one study of one issue, and it will only obscure the value of your study if you try to relate it to every tangential point in related research.

Dont think of searching the literature as a one-time-only venture---+omething that you leave behind as you move on to your real research. You may encounter new questions or unanticipated problems as you conduct your research or as you burrow deeper into the literature. Searching the literature again to determine what others have found in response to these questions or what steps they have aken to resolve these problems can yield substantial improvements in your own research, There is so much literature on so many topics that it often is not possible to figure out in advance every subject for which you should search the literature or what type of search will be most beneficial.

Another reason to make searching the literature an ongoing project is that the literature is always growing. During the course of one research study, whether it takes only one semes- ter or several years, new findings will be published and relevant questions will be debated. Staying attuned to the literature and checking it at least when you are writing up your find- ings may save your study from being outdated as soon as it is finished.

Critically Review Research Your literature review will suggest specific research questions for further investigation and research methods with which to study those questions. Sherman and Berk (1984) learned from their literature review that there had been litde empirical research about the impact of arrest policies in domestic violence cases. What prior research had been conducted did not use very rigorous research designs. There was thus potential value in conducting new research using a rigorous design. Subsequent researchers questioned whether Sherman and Berk's results would be replicated in other cities and whether some of their methods could be improved. When the original results did not replicate, researchers designed more investiga- tions to find outwhy. In this way, reviewing the literature identifies unanswered questions and contradictory evidence.

Effective review of the prior research is an essential step in building the foundation for new research. You must assess carefirlly the quality of each research study, consider the implications of each article for your own plans, and expand your thinking about your research question to take account of new perspectives and altemative arguments. It is through review- ing the literature and using it to extend and sharpen your own ideas and methods that you become a part of the social science community. Instead of being only one individual studying an issue, you are building on an ever-growing body ofknowledge that is being constructed by the entire community of scholars.

34 SECTION I o FOUNDATIONS FOR SOCIAL RESEARCH

Sometimes you'll find that someone else has akeady searched the literature on your research question and discussed what he or she found in a special review article or book chapter. For example, Chalk and Garner (2001) published an excellent review of the research on arrest and domestic violence in the journal Nar Direxions for Eaaluation. Most of the research articles that you find will include a short literature review on the specific focus of the research. These reviews can help a lot, but they are no substitute for searching the literature yourself, select- ing the articles and other sources that are most pertinent to your research question, and then reviewing what you have found. No one but you can decide what is relevant for your research question and the research circumstances you will be facing-the setting you will study, the timing of your study, the new issues that you want to include in your study, and your specific methods. And you cant depend on any published research review for information on the most recent works. Results from new research about many questions appear continually in scholarly journals and boola, in research reports from government agencies and other organizations, and on websites all over the world; you'll need to check for new research like this yourself.

This section concentrates on the procedures you should use for reviewing the articles you find in a search ofthe scholarly literature. These procedures can also be applied to reviews of research monographs-books that provide more information from a research project than that which can be contained in a journal article.

Reviewing the literature is really a two-stage process. In the first stage, you must assess each article separately. This assessment should follow a standard format such as that repre- sented in Appendix A, "Questions to Ask About a Research Article." However, you should keep in mind that you cant adequately understand a research study ifyou treat it as a series of discrete steps, involving a marriage of convenience among separate techniques. Any research

CHAPTER 2 o THE PROCESS AND PROBLEMS OF RESEARCH RELATED TO CRIME AND CRIMINOLOGY

project is an integrated whole, so you must be concerned with how each component of the research design influenced the others-for example, how the measurement approach might have affected the causal validity of the researcher's conclusions and how the sampling strategy might have altered the quality of measures.

The second sage of the review process is to assess the implications of the entire set of articles (and other materials) for the relevant aspects ofyour research question and proce- dures and then to write an integrated review that highlights these implications. Although you can find literature reviews that consist simply of assessments of one published article after another-that never get beyond the first stage in the review process-your understanding of the literature and the quality of your own workwill be much improved if you make the effort to write an integrated review.

In the next two sections, we will show how you might answer many of the questions in Appendix A as we review a research article about domestic violence. We will then show how the review of a single article can be used vdthin an integrated review of the body of prior research on this research question. Because at this early point in the textyou wont be familiar with all the terminology used in the article review, you might want to put offreading t}rough the more elaborate article review in Appendix B until later in the course.

A Single-Article Review: Formal and !nforma! Deterrents to Domestic Violence

Antony Pate and Edwin Hamilton at the National Police Foundation designed one of the studies funded by the U.S. Departnent ofJustice to replicate the Minneapolis DomesticVio- lence Eriperiment. In this section, we will examine the article that resulted from that replica- tion, which was published inrhe Arnerican Sociological Reaiew @ate and Hamilton 1992).The numbers in square brackets refer to the article review questions inAppendix B.

The Research Question. Like Sherman and Berk's (198a) original Minneapolis study, Pate and Hamilton's (1992) Metro-Dade replication spouse assault experiment sought to test the deterrent effect of arrest in domestic violence cases but with an additional focus on the role of informal social control [1]. The purpose of the study was explanatory because the goal was to explain variation in the propensity to commit spouse abuse [2]. Deterrence theory provided the theoretical framework for the study, but this study was broadened to include informal sanctions such as stigma and the loss of valued relationships as well as formal sanctions such as arrest [3]. Pate and Hamilton's (1992) literature review referred, appropriately, to the original Sherman and Berk (1984) research, to the other studies that attempted to replicate the original findings, and to research on informal social control [4].

There is no explicit discussion of ethical guidelines in the article, although reference is made to a more complete unpublished report [6]. Clearly, important ethical issues had to be considered, given the experimenal intervention in the police response to serious assaults, but the adherence to standard criminal justice procedures suggests attention to the welfare of victims as well as to the rights of suspects. W'e will consider these issues in more detail later in this chapte

The Research Design. Developed as a follow-up to the original Minneapolis experiment, theMetro-Dade experiment exemplifies the guidelines for scientific research. Itwas designed systematically,with carefirl attention to specification of terms and clarification of assumptions, and focused on the possibility of different outcomes rather than certainty about one preferred outcome. The major concepts in the study, formal and informal deterrence, were defined clearly [7] and then measured with straightforward indicators-alTest or nonarrest for formal deterrence, and marital status and employment status for informal deterrence. However, the

36 SECTION I . FOUNDATIONS FOR SOCIAL RESEARCH

specific measurement procedures for marital and employment status were not discussed, and no attempt was made to determine whether they captured adequately the concept of informal social control [9, l0].

Three hypotheses were sated and related to the larger theoretical framework and prior research [8]. The study design focused on t]re behavior of individuals [1 1] and collected data over time, including records indicating subsequent assault up to six months after the initial arrest. The project's experimental design was used appropriately to test for the causal effect of arrest on recidivism 115,17). The research project involved all eligible cases, rather than a sample of cases, but there were a number of eligibility criteria that narrowed down the ability to generalize these results to the entire population of domestic assault cases in the Metro-Dade area or elsewhere [11]. There is a brief discussion of the 92 eligible cases that were not given the treatrnent to which they were assigned, but it does not clari$, the reasons for the misassignment [15].

The Research Findings and Conclusion. Pate and Hamilton's (1992) analysis of the Metro-Dade experiment was motivated by concern with the effect of social context, because the replications in other cities of the original Minneapolis Domestic Violence Experiment had not had consistent results [19]. Their analysis gave strong support to the expectation that informal social control processes are important As they had hypothesized, arrest had a deterrent effect on suspects who were employed but not on those who were unemployed @xhibit 2.3). However, marital status had no such effect [20]. The subsequent discussion of these findings gives no attention to the implications of the lack of support for the effect of marital status [21], but the study represents an imporant improvement over earlier research that had not examined informal sanctions [22]. The need for additional research is highlighted, and the importance of the findings for social poliry is discussed: Pate and Hamilton suggest that their finding that arrest deters only those who have something to lose (e.9., a job) must be taken into account when policing policies are established.

Overall, the Pate and Hamilton (1992) study represents an important contribution to understanding how informal social control processes influence the effectiveness of formal sanctions such as arrest. Although the use of a population of actual spouse assault cases pre- cluded the use of very sophisticated measures of informal social conuol, the experimental design ofthe study and the researchers' ability to interpret the results in the context ofseveral other comparable experiments distinguish this research as exceptionally worthwhile. It is not hard to understand why these studies continue to stimulate further research and ongoing policy discussions.

An lntegrated Literature Review: When Does Arrest Matter?

The goal of the second stage of the literature review process is to integrate the results of your separate article reviews and develop an overall assessment of the implications of prior research.The integrated literature review should accomplish tfuee goals: (1) summarize prior research, (2) critique prior research, and (3) present pertinent conclusions (tIart 1998). We will discuss each of these goals in turn.

I . Sunmtarize prior researcb.Your summary of prior research must focus on the particular research questions that you will address, but you may also need to provide some more general background. Carolyn Hoyle andAndrew Sanders (2000) begin rhen British Jouru.al of Criminolog research article about mandatory arrest policies in domestic violence cases with what they term a "provocative" question: What is the point of making it a crime for men to assault their female partners and ex-partners? They then review the different theories and supporting research that has justified different police policies: the "victim choice" position, the "pro-arrest" position,

CHAPTER 2 . THE PROCESS AND PROBLEMS OF RESEARCH RELATED TO CRIME AND CRIMINOLOGY 37

E20 (E oo +t t?tr lc o-J C'oo €10a G ?-+. =5*,co C) Lto

38

Unemployed Employed

Source: Anthony M. Pate and Edwin E. Hamilton.1992. "Formal and Informal Deterents to DomesticViolence." American Sociological Reuieu, 57(5),691-697. Reprinted with permission from the American Sociological Association and the authors.

and the "victim empowerment" position. Finally, they review the research on the "controlling behaviors" of men that frames the specific research question on which they focus: How do victims view the value of criminal justice interventions in their own cases (tloyle and Sanders 2000)?

fukyourself three questions aboutyour summary of the literature:

a. flave you been selective? Ifthere have been more than a few prior investigations ofyour research question, you will need to narrow your focus to the most relevant and highest-quality studies. Dont cite a large number of prior articles "because they are there."

b. Is the research up to date? Be sure to include the most recent research, not only the "classic" studies.

c. Have you used direct quotes sparingly? To focus your literature review, you need to express the key points from prior research in your own words. Use direct quotes only when they are essential for making an imporant point @yrczak 2005).

2. Critique prior research. Evaluate the strengths and weaknesses of the prior research. In addition to all the points that you develop as you an$trer the article review questions in Appendix B, you should also select articles for review that reflect work published in peer-reviewed journals and written by credible authors who have been funded by reputable sources. Consider the following questions as you decide how much weight to give each article:

a. Howwas the report reviewed prior to its publication or release? Articles published in academic journals go through a rigorous review process, usually

SECTION I . FOUNDATIONS FOR SOCIAL RESEARCH

Unemployed

Exhibit 2.3 Percentage of Suspects With a Subsequent Assault by Employment Status and Arrest Status

involving carefirl criticism and revision. Top refereed journals may accept only 10% of the submitted articles, so they can be very selective. Disserations go through a lengthy process of criticism and revision by a few members of the dissertation writer's home institution. A report released direcdy by a research organization is likely to have had only a limited review, although some research organizations maintain a rigorous internal review process. Papers presented at

professional meetings may have had litde prior revievr. Needless to say, more confidence can be placed in research results that have been subject to a more rigorous review.

b. What is the author's reputation? Reports by an author or a team of authors who have published other work on the research question should be given somewhat greater credibility at the outset.

c. Who funded and sponsored the research? Major federal funding agencies and private foundations fund only research proposals that have been evaluated

carefirlly and ranked highly by a panel ofexperts. They also often closely monitor the progress of the research. This does not guarantee that every such project report is good, but it goes a long way toward ensuring some worthwhile products. On the other hand, research that is funded by organizations that have

a preference for a particular outcome should be given particularly close scrutiny (Locke, Silverman, and Spirduso 1998).

3. Present pertinent conclwionl Don't leave the reader guessing about the implications of the prior research for your own investigation. Present the conclusions you draw from the research you have reviewed. fu you do so, follow several simple guidelines:

a. Distinguish clearlyyour own opinion of prior research from the conclusions of the authors of the articles you have reviewed.

b. Make it clear when your own approach is based on the theoretical framework that you use and not on the results ofprior research.