Skip to main content
SearchLoginLogin or Signup

Standing on the Shoulders of Pioneers in Experimental Criminology: The Cambridge-Somerville Youth Study at 85 Years (1935-2020) and Beyond. The 2021 Joan McCord Award Lecture

Objectives The article reflects on some pioneers in experimental criminology, including those who contributed to the development and longevity of the Cambridge-Somerville Youth Study (CSYS), and reports on a new program of research on the CSYS, extending the study to 85 years ...

Published onNov 05, 2021
Standing on the Shoulders of Pioneers in Experimental Criminology: The Cambridge-Somerville Youth Study at 85 Years (1935-2020) and Beyond. The 2021 Joan McCord Award Lecture


Objectives The article reflects on some pioneers in experimental criminology, including those who contributed to the development and longevity of the Cambridge-Somerville Youth Study (CSYS), and reports on a new program of research on the CSYS, extending the study to 85 years (1935-2020) and beyond.

Methods The key focal points are the CSYS, a randomized controlled trial of a delinquency prevention intervention, and the people who were central to its development, implementation, and follow-ups. A wide range of published sources and archival records are used.

Results The idea of ‘standing on the shoulders of giants’ or standing on the shoulders of pioneers in one field or another is a hallmark of science and scientific progress. This is at the heart of a new program of research on the CSYS, with four areas under investigation: (1) intervention effects over the full life-course; (2) development of criminal offending over the full life-course; (3) intergenerational effects (over three generations); and (4) historical understanding. Each area is discussed with reference to the wider context, new findings, and work underway or planned for the years ahead.

Conclusions Joan McCord is one of our giants in experimental criminology. She revived the CSYS, made it into the pioneering longitudinal-experimental study that we know today, and set the stage for a new program of research on the study.

Keywords: Cambridge-Somerville Youth Study; Delinquency prevention; Randomized controlled trial; Longitudinal-experimental design; Joan McCord; History of experimental criminology


Welsh, B.C. (2021). Standing on the shoulders of pioneers in experimental criminology: The Cambridge-Somerville Youth Study at 85 years (1935-2020) and beyond: The 2021 Joan McCord Award Lecture. Journal of Experimental Criminology. DOI: 10.1007/s11292-021-09491-w.


It is a tremendous honor to receive the 2021 Joan McCord Award from the American Society of Criminology’s Division of Experimental Criminology (DEC) and the Academy of Experimental Criminology (AEC). I am extremely grateful to the awards committee, the Chair of the DEC/AEC, Barak Ariel, and the three distinguished scholars (and experimental criminologists) who nominated me for the award. Receiving this award is especially meaningful because I have been fortunate to continue—in some small measure—Joan McCord’s life-work through a new program of research on the Cambridge-Somerville Youth Study (CSYS).

The story of becoming the next director of the CSYS and launching this program of research is great fun to tell, but it may test the patience of some readers. Some brief remarks are nevertheless pertinent to the focus of the current article. I only knew Joan for a few short years in the early 2000s before her death in 2004. These years were marked by any number of major accomplishments in her brilliant career, including serving as a founding member of the Campbell Collaboration’s Crime and Justice Group, the third president of the AEC (2003-2004), and co-chair of the National Research Council and Institute of Medicine’s ‘Panel on Juvenile Crime: Prevention, Treatment, and Control’ (McCord et al. 2001). Her research and writings on the CSYS around this time (e.g., Dishion et al. 1999; McCord 2002a, 2003), not to mention beforehand (e.g., McCord 1978, 1981; McCord and McCord 1959a, 1959b, 1960), greatly influenced my own research on the early prevention of delinquency and helped to shape my journey in experimental criminology.

Jump ahead about 10 years to a research project on the iatrogenic effects of crime prevention programs (Welsh and Rocque 2014), an effort to build upon Joan’s classic article ‘Cures that Harm: Unanticipated Outcomes of Crime Prevention Programs’ (McCord 2003), we were left to wonder about the future of the CSYS. A circuitous route of correspondence with several of Joan’s colleagues and some others led me to Joan’s eldest son, Geoffrey Sayre-McCord. An invitation to visit Geoff at his home university, the University of North Carolina at Chapel Hill, resulted in wonderful conversations about his Mom and the study and the chance to go through dozens of boxes of Joan’s papers. This was followed by a request for a detailed proposal on my plans for the study (written with my colleague Jack McDevitt) and some checking out of my bona fides. (I passed!)

With this award lecture I have been given the special opportunity to write about Joan McCord (née Fish; 1930-2004), the CSYS (her study) that she directed over the course of five decades (1956-1959 and 1975-2004), and her significant and lasting influence on experimental criminology. The article has two main objectives. The first is to reflect on some of the early pioneers of experimental criminology, specifically, those who contributed to the development and longevity of the CSYS. Also important is to reflect on other pioneers of experimental criminology who were influenced in one way or another by Joan McCord and the CSYS, and who are also part of the metaphorical ‘standing on the shoulders of giants.’ The second objective is to report on the new program of research on the CSYS, which has extended the study to 85 years (1935-2020) and beyond. Four sets of research questions are guiding this work, and each is discussed with reference to the wider context, new findings, and work underway or planned for the years ahead.

Pioneers in Experimental Criminology

The idea of ‘standing on the shoulders of giants’ or standing on the shoulders of pioneers in one field or another (or multiple fields) is a hallmark of science and scientific progress: One generation building on the work of past generations. This may come about by initiating replication experiments, revisiting lingering questions and ideas, examining new research questions, and exploring new frontiers in the research (see Mears and Cochran 2019). Integral to this view is a deep recognition for those who broke new ground or paved the way before us. My introduction to the personal nature of this view was through my Ph.D. supervisor and now close colleague David Farrington. Writing in the foreword to his Festschrift, Farrington (2012: xvii) provides the provenance of this famous saying and notes how he too is one of its beneficiaries:

In a letter to Robert Hooke in 1676, Isaac Newton said, “If I have seen a little further it is by standing on the shoulders of giants.” This is how I feel. I am not sure that I have seen a little further than anyone else (probably not!) but I have certainly “stood on the shoulders of giants,” metaphorically of course.

This is an important theme in Richard Tremblay’s (2021) new book about leading social and behavioral scientists born during World War II and their contributions to understanding and preventing violence. Indeed, in the book’s preamble, Tremblay (2021: i) has the following to say, “Knowing their stories, you can stand on the shoulders of these giants to look to the future of this subject and potentially contribute to its next steps.” Of the 12 scientists profiled, a good number may self-identify as experimental criminologists, including three past recipients of the Joan McCord Award, not to mention the Stockholm Prize in Criminology (Richard Tremblay, David Farrington, and Friedrich Lösel). They recount their mentors and other scholars who inspired them to pursue new and pressing research questions, which resulted in many brilliant discoveries and helped to advance knowledge and improve public policy. Mentioned too are the many, sometimes dozens of, young researchers who these scientists mentored and trained and were likewise inspired to follow in their mentors’ paths, forge new areas of discovery, or both.

More details emerge on this front through an historical understanding of experimental criminology. For example, in Sherman’s (2005) narrative of the beginnings and arc of experimental criminology—and its relationship to analytical criminology—we are introduced to the famous pioneers of the field (Joan McCord included), as well as some of the completely ‘forgotten’ figures. While each figure and their experiment(s) may not necessarily represent a through line to a subsequent generation, many connections are important for experimental criminologists today.

Using a more focused area of experimentation in crime and justice—randomized controlled trials in policing (from 1970 to 2011)—Braga et al. (2014) examine co-author and mentoring relationships among this professional network of researchers. Findings point to a rather small group of, but “vital few,” experimental criminologists who are responsible for (1) producing a rather large share of the growth in policing experiments and (2) mentoring the next generation of police researchers, who themselves are making major contributions to a continued growth in policing experiments. Time will tell if this second generation can replicate this experience or even expand the network.

Joan McCord and her network

Joan McCord is one of our giants in experimental criminology. As important as her research and scholarship has been to the field, her legacy is also one of inspiration for and influence on other experimental criminologists, some of whom are also pioneers in the field. At the same time, Joan was inspired by others and, in the context of the CSYS (the main focus here), she built on the work of those who were instrumental to the development and continuation of the study, as well as sought counsel from some leading scholars of her day.

Readers of Joan McCord’s voluminous works on the CSYS will be familiar with Richard Clarke Cabot (1868-1939), the founder and first director of the study. Cabot would become influential to Joan’s future work on the study and her life-long commitment to experimentation. In her writings on the study, she paid tribute to his “deep commitment to science, his long-term vision for the study, and his fierce dedication to trying to improve the life chances of underprivileged boys” (Welsh et al. 2017: 79). While Cabot was by no means a trained criminologist, and many other professions can rightly lay claim to him, he serves as an important figure in the history of experimental criminology. (See Historical Understanding section for more details on Cabot.)

Also influential to Joan’s work on the study and experimentation more generally were Edwin Powers and Helen Witmer, the researchers who undertook the first evaluation of the CSYS (Powers and Witmer 1951). As study director from 1941 to 1951, Powers was responsible for making sure that the preventive intervention continued to run despite some rather formidable obstacles during the war years, as well as ensuring fidelity to the model and maintaining detailed case records. His post-war writings were also important to articulating the study’s contributions to the prevention of delinquency (see Powers 1949, 1950). In the case of Helen Witmer, her statistical acumen would prove invaluable to the study’s first evaluation.

In preparing for the third and most ambitious follow-up of the study (1975-1979), Joan McCord seemed mindful of the benefits of tapping into influential scholars of the day. At the very least, long-term follow-ups of preventive interventions were exceedingly rare and may need some justification. Joan convened advisory conferences to discuss this and other key elements of the project. Participating in the conferences was a prominent group of scholars, including Michael Wadsworth, Jack Block, Jerry Bachman, and Glen Elder (McCord 2002b).

It was during this period of time, if not already underway, that Joan’s research was starting to influence and inspire others—the next generation of experimental criminologists. In reflecting on the “giants of longitudinal research” who had the greatest influence on him, David Farrington noted the following about Joan: “Joan McCord always bubbled over with enthusiasm about everything. She was a wonderful role model who energized younger scholars with her intellectual excitement” (Farrington 2012: xviii). This was a common refrain among criminologists (see also Farrington 2007; Tremblay and Farrington 2004).

Her research and scholarship would also have a direct impact on others’ longitudinal and experimental studies. For example, Farrington’s Cambridge Study in Delinquent Development drew upon her work for its measures of parental supervision, discipline, and attitude (Tremblay and Farrington 2004: 6). McCord’s influence was even more direct for Richard Tremblay’s Montreal Longitudinal-Experimental Study: it “was designed to prevent the iatrogenic effects of interventions she had identified” (Tremblay and Farrington 2004: 6). In fact, Tremblay was so moved by McCord’s (1978) findings of the 30-year follow-up of the CSYS that he asked her to join him in planning the Montreal experiment. She accepted, leading to a major innovation in the design of the social skills training component of the intervention and an important and lasting impact of the intervention overall:

Rather than group the at-risk boys to work on their social skills, groups were created that had the following ratio: one disruptive boy to three highly prosocial boys. This strategy was an attempt to use peers as positive role models. There is evidence from the follow-up assessments that the treatment boys had friends who were less deviant than the control boys (Tremblay et al. 1992). Grouping one or two deviant boys with a majority of highly prosocial boys appears to have created lasting prosocial friendships. Indeed, alongside the behavioral parent training, this would prove to be an important causal mechanism in the effectiveness of the preventive intervention. (Welsh and Tremblay 2021: 8)

Many collaborations on the Montreal study would follow (e.g., McCord et al. 1994), not to mention a life-long friendship between these two experimental criminologists.

The Cambridge-Somerville Youth Study: Past, Present, and Future

A full accounting of the development, characteristics, and key findings of the Cambridge-Somerville Youth Study does not seem necessary for this article. In part, this is because published lectures of some past recipients of the Joan McCord Award have provided excellent summaries of the study (see e.g., Farrington 2006; Gottfredson 2010; Tremblay 2005), not to mention that comprehensive accounts of the study have been written by Joan McCord (e.g., McCord 1984; McCord and McCord 1959b, 1960; see also Sayre-McCord 2007) and by others who directed and worked on the study before her (e.g., deQ. Cabot 1940; Powers and Witmer 1951). It is also the case that detailed accounts of the study have recently been published in this journal (Welsh et al. 2019a) and elsewhere (e.g., Tremblay et al. 2019; Welsh et al. 2017; Zane et al. 2016, 2017).

The CSYS was founded and directed by Richard Clarke Cabot, a renowned physician and professor of clinical medicine and social ethics at Harvard University. Planning began in 1935, with the programmatic element operating from 1939 to 1945. The study involved a randomized controlled experiment of a delinquency prevention program with a prospective longitudinal survey of the development of offending embedded in it—known today as a longitudinal-experimental design (Farrington 2006). Six hundred fifty underprivileged boys of average and difficult temperament (later reduced to 506), ages 5 to 13 years (median age = 10.5 years), from Cambridge and Somerville, Massachusetts, were matched in pairs (known as “diagnostic twins”; deQ. Cabot 1940: 146) and one member of each pair was randomly allocated to the treatment group. Referred to as “directed friendship,” the preventive intervention involved individual counseling through a range of activities and home visits with the families. Treatment group boys received the intervention for a mean average of 5.5 years. Boys in the control group received no special services.

Follow-up assessments of criminal offending and other outcomes have been carried out at key stages of the participants’ life-course and for more than 70 years: transition from adolescence to adulthood (in 1948; Powers and Witmer 1951); early adulthood (in 1956; McCord and McCord 1959a, 1959b, 1960); middle age (from 1975-1979; McCord 1978, 1981); and old age (from 2017-present; Welsh et al. 2019b).

Joan McCord’s involvement with the CSYS goes as far back as the early 1950s. With her husband, William McCord, she directed the 1956 follow-up, which was a product of several consequential events in their own right. One of these events involved Joan taking a position as a research assistant at Harvard University, working with prominent child development researchers, including Harry Levin, John Whiting, and Eleanor Maccoby (McCord 2002b). Also influential was her husband’s graduate work with Sheldon and Eleanor Glueck at Harvard, the publication of Joan’s first book, Psychopathy and Delinquency (McCord and McCord 1956), and funding from the Ella Lyman Cabot Foundation, the same foundation that supported the development and operation of the CSYS through 1945 (and was established by Richard Cabot upon the death of his wife in 1935).

Two decades later, Joan McCord most famously initiated and directed the study’s next long-term follow-up (30 years post-intervention), which took place in two phases between 1975 and 1979. Readers may be interested to know that Joan McCord had plans for another long-term follow-up of the study (52 years post-intervention), when participants were between the ages of 63 and 71 years.1

As outlined briefly in the Introduction, this brings us to the study’s latest follow-up assessment, which started in 2017 and is still underway. The first phase of this follow-up represents 72 years post-intervention, with participants between the ages of 83 and 91 years. It is important to note that this latest follow-up is one component of a larger program of research on the CSYS, and the following summarizes the key research questions that are being investigated:

(1) On intervention effects over the full life-course: Have the iatrogenic effects observed in middle age persisted in old age?

(2) On the development of criminal offending over the full life-course: What are the long-term offending trajectories, patterns of desistance from offending, and duration of criminal careers?

(3) On intergenerational effects (over three generations): What have been the effects on the children of the study participants? Have the children of the treatment group men (compared to their control counterparts) also experienced undesirable outcomes?

(4) On historical significance: What is our historical understanding of the development of the study, its influences on delinquency prevention and the discipline of criminology, and what are the lessons for today?

The next four sections discuss these research questions with reference to the wider context, new findings, and work underway or planned for the years ahead.

Intervention Effects

Long-term follow-up assessments (i.e., of 10 or more years) of preventive interventions with criminological outcomes are highly important yet rare (Farrington 2021; Farrington and MacKenzie 2013). They are important because they offer insights on a number of key issues that hold implications for prevention science and public policy, including if initial effects wear off or decay over time, if initial effects improve over time, if improvement is a function of “sleeper” effects, and how different outcome measures may become more relevant with time and age of participants (Farrington and Hawkins 2019; see also Hill et al. 2016).

Despite the potential scientific and policy merits of long-term follow-ups, two key questions confronted us in carrying out a vastly longer assessment of the CSYS, one that would trace participants well into old age and cover the full life-course. The first question had to do with attribution of effects to the intervention many decades later. This is less of a concern for follow-ups in the range of five to 15 years, for example, largely owing to a relatively shorter period of time since program termination and (for early preventive interventions) the youthful age of participants. The developmental perspective makes no distinction on the plausibility of intervention effects over a short, medium, or long duration. Here, the central idea is that “earlier experience determines later behavior” (Tremblay and Craig 1995: 152). In addition, there is considerable evidence of behavioral continuity of offending: childhood antisocial behavior is linked with juvenile crime; juvenile crime is linked with offending in early adulthood; and offending in early adulthood is linked with offending in the adult years (Farrington 2013).

The second question concerned the iatrogenic effects that were observed in the study’s previous follow-up (McCord 1978, 1981), specifically: Is there any benefit from a longer follow-up of a preventive intervention that had caused harm? Several considerations were at play. One had to do with the process of ageing out of involvement in criminal offending or a life of crime. In their long-term follow-up of the Gluecks’ prospective longitudinal study of 500 delinquent boys and 500 matched non-delinquent boys from Boston (Glueck and Glueck 1950), Laub and Sampson (2003) found that the mean number of criminal offenses declined significantly when the men were in their fifties and continued to fall throughout their sixties. Was this usual trajectory maintained or lost among the treatment group men (relative to their control counterparts) in the CSYS? It is plausible that the iatrogenic effects from midlife could wear off or decay. It was also the case that we were inspired by other researchers who had conducted longer follow-ups—upwards of 25 years post-intervention—of experiments of crime prevention interventions with iatrogenic effects (O’Donnell and Williams 2013; Poulin et al. 2001; Sherman and Harris 2013). Finally, and perhaps putting to rest any equivocation about the benefits of a longer follow-up of a harm-causing preventive intervention, is the need for greater understanding of iatrogenic effects and that this knowledge could help inform evidence-based policy. Here, the thinking is that this knowledge can serve as another key input in the policymaking process (Welsh et al. 2020b).

As noted above, the latest follow-up of the CSYS is being carried out in two phases. Between 2016 and 2018, records were located for 96.4% of the participants (488 of 506), with a total of 446 confirmed deceased (88.1%) and 42 alive (8.3%). Mortality was the focus of this 72-year post-intervention follow-up (Welsh et al. 2019b). Matched-pairs analyses showed no significant differences for all outcomes of interest: mortality at latest follow-up; premature mortality (< 40 years); and cause of mortality (natural versus unnatural). A Cox proportional hazard regression analysis indicated no difference in time to death between the treatment and control group men. In not being able to detect iatrogenic effects on mortality, the main implication is that the iatrogenic effects on mortality experienced in middle age did not persist to old age (i.e., up to age 90).

One view of this change in intervention effects over time is that the observed iatrogenic effects on mortality in middle age “may have been a singular event, irrespective of their concordance with effects for a wide range of other outcomes” (Welsh et al. 2019b: 8). In commenting on the findings of this follow-up, Farrington and Hawkins (2019) suggest that the earlier iatrogenic effects may have been rather weak, as indicated by tests of statistical significance and the magnitude of the effect. The next phase of this follow-up, which is underway and focused on criminal offending, may also provide further insight on an explanation for the study’s latest findings on mortality.

Development of Criminal Offending

While the CSYS is well known for being the first randomized controlled experiment in criminology (Weisburd and Petrosino 2004) and one of the earliest experiments of a social intervention (Forsetlund et al. 2007), Cabot also designed the study to understand the development of criminal offending over the life-course. Indeed, it is this focus—by drawing on the prospective longitudinal component of the study—that accounts for the overwhelming share of Joan McCord’s research and publications on the CSYS (for a full list of publications, see Sayre-McCord 2007).

From her early involvement with the study, Joan was well aware of the importance of this line of research: “In fact, the causes of crime came to be our major focus of attention” (McCord and McCord 1959b: 9). Moreover, according to Farrington (2006: 130), even after the 30-year follow-up, the CSYS “would not have become so famous if it had been merely an experiment with negative results, however well the experiment had been conducted.” For Farrington (2006; see also Farrington 2013), it was the study’s longitudinal-experimental design that made it unique and important. So, with each new follow-up of the CSYS, there is also the opportunity to investigate the natural history of criminal offending. Following Joan’s analytic strategy (see McCord 1984), this focus draws on the study’s treatment group participants (n = 253).

Several events set the stage for this line of research. The first, which I will return to later in the article, had to do with Cabot’s professional relationship with Sheldon and Eleanor Glueck (see Cabot 1930). The second event concerned Joan McCord’s early involvement in the study, specifically, the coding of case history records. Over the course of 1957, she trained and supervised six researchers in the coding of the case histories of the treatment group boys and their families. She reported that high inter-rater reliability scores were achieved for the coding and, even more importantly, that the “records were not contaminated by retrospective biases” (McCord 1992: 204). This was made possible because the records were coded prior to the collection of the data for the 1956 follow-up and the researchers were blind to the outcomes of the 1948 follow-up. The importance of this event cannot be overstated, and we attempted to summarize its importance as such:

It seems fair to say that, in the absence of the coding of these records, the longitudinal survey component would have been greatly compromised, thus severely restricting the ability to investigate the natural history of offending from childhood to middle age (McCord 1984) and into old age… (Welsh et al. 2019a: 448)

Another important event, which is less an event and has more to do with the research process (and, of course, also holds implications for investigating intervention effects), was Joan’s meticulous record keeping and dogged tracing of the participants. By the end of the 1975-1979 follow-up, Joan had managed to locate records for 494 participants (T = 248; C = 246), which translates to a highly impressive 97.6% sample retention. The other good news for the current follow-up of the study is that Joan maintained detailed records (electronic and paper) for all of the participants (living and deceased), including last home address, telephone number, personal identifiers such as social security number, and next of kin.

We previously outlined several key questions that are part of this line of research: “What is the duration of a ‘criminal career’? What do long-term offending trajectories look like? How do patterns of both the reduction and cessation of offending develop? What explanations emerge for such patterns and what is the weight of evidence for them?” (Welsh and Zimmerman 2015: 333). Other questions worthy of investigation over the full life-course have to do with another important outcome, premature mortality, as well as its relationship to criminal offending.

As with our investigation of intervention effects, research on the natural history of criminal offending is also taking place in two phases. The first phase focused on the relationship between criminal offending in middle age and mortality in old age (Zane et al. 2019). The second phase, which is underway, focuses on criminal offending and mortality over the full life-course. With respect to the first phase, we used court convictions of criminal offenses collected during middle age (mean = 47 years) and death records collected during old age (up to age 89). Through the end of 2016, a total of 216 participants of the treatment group (or 85.4%) were confirmed deceased. We found that mortality was related to offending over the life-course, but only when offending was measured using group-based trajectory modeling, and only from middle age into old age. Specifically, from middle age onward, life-course persistent offenders were more likely than adolescence-limited offenders and non-offenders to die earlier and from unnatural causes (see also Skinner et al. 2021). Results also indicated that childhood risk factors for delinquency were not associated with mortality risk over the full life-course.

Intergenerational Effects

The intergenerational transmission of criminal offending, not to mention alcoholism and its relationship to offending, formed a key part of Joan McCord’s research on the CSYS. Drawing on the treatment group men in middle age (mean birth year = 1928) and their fathers (mean birth year = 1896), Joan was interested if crime tended to run in families and why. For example, she found that criminal fathers were significantly more likely to have criminal sons than were non-criminal fathers (45% vs. 28%). She also found that criminal sons were significantly more likely than non-criminal sons to have criminal fathers (32% vs. 18%) (McCord 1991). As noted by Tremblay et al. (2019: 12):

She reached the tentative conclusion that transmission of offending, as well as alcoholism, from fathers to their sons was largely mediated by parenting practices and other family risk factors rather than direct social learning of deviant behavior, biological inheritance, or other causal mechanisms.

For the current program of research on the CSYS, the immediate interest has to do with intergenerational effects related to the experimental test of the preventive intervention, more specifically: What have been the effects on the children of the study participants? Have the children of the treatment group men (compared to their control counterparts) also experienced undesirable outcomes? Investigating these questions requires collecting data on the third generation (G3) of the study. With G3 including both males and females, there will also be the opportunity to investigate the interaction of gender with transmission of offending: from father to son and father to daughter. It will also be important to investigate key individual, family, and community factors associated with intergenerational effects on criminal offending.

Few intervention studies have attempted to investigate the intergenerational effects on antisocial behavior using three generations. Perhaps the most important of these studies is the Seattle Social Development Project, a nonrandomized controlled trial of a multi-modal preventive intervention of about 500 first-grade children and their parents designed to increase parental attachment and school bonding (Kosterman et al. 2019). With many of the study participants now parents themselves, it was possible to examine if the long-term desirable effects of the intervention carried over to G3 (mean age = 7 years; n = 182). Hill et al. (2020) found that children in the treatment group, compared to the controls, showed significant improvements across a range of outcomes, including higher academic skills and performance and lower behavioral problems.

Planning for this line of research on the CSYS is underway (see Welsh et al. 2018). As with our longer follow-up of the study participants (the second generation or G2), the identification of G3 begins with the wealth of personal information that Joan meticulously collected and recorded on G2 (see above). Both traditional and online/social media search methods will be used in the tracing of G3. In the case of traditional methods, Laub and Sampson (2003) reported some success in locating participants by using the Boston telephone directory, even though 35 years had elapsed since last contact. In their long-term study of the Boston Special Youth Program, Moule and Decker (2013) found that, for example, Google searches were useful in obtaining information and for providing links to other sites, such as web pages and Facebook profiles of participants. In fact, we relied on a wide range of both of these methods to trace participants for the 72-year post-intervention follow-up to assess effects on mortality.

Historical Understanding

The CSYS is often referred to as a formative study in the annals of criminology, and the broad parameters of its design, development, and main actors are well known for the most part. Indeed, Joan often introduced papers by drawing attention to the role of Richard Cabot, the influence of leading criminologists and other academics of the day (e.g., Sheldon and Eleanor Glueck, William Healy, Augusta Bronner), and the treatment of juvenile delinquents during the early 1900s. Even so, it stands to reason that there may be some interest in how such a study came to be (including its novel research design), the context in which it was developed and implemented, and how the study and the lives of the participants were shaped by major historical events, including the Great Depression and America’s involvement in World War II.

The new program of research offers an opportunity to develop a deeper and richer historical understanding of the CSYS. It is important to note that this research focus was inspired by one of our own former historians and a dear colleague at Northeastern, Nicole (Nicky) Rafter (e.g., Rafter 2004, 2010); leading works on the beginnings of experimentation of crime and justice (e.g., Farrington 1983; Sherman 2005); and Joan McCord herself who, had it not been for her prolific research and writing on the fundamentals of the study, surely would have turned some of her attention to research on the study’s history. It is also noteworthy that this research focus allows for an examination of the study’s influences on delinquency prevention, early developmental crime prevention, and experimental criminology.

So far, research on the history of the study has examined three key developments. The first has to do with Richard Cabot, his storied life and career that culminated in the development and implementation of the CSYS (which would turn out to be the last project of his life), and the influences (personal, professional, and institutional) that motivated him to design the intervention with the aim of preventing delinquency in the first instance and trying to bring about individual-level change (see Welsh et al. 2017). While it has been written that Cabot had a profound influence on the Gluecks’ research on criminal careers, originating in a 1925 seminar on social ethics taught by Cabot and attended by Sheldon Glueck (Laub and Sampson 1991: 1407; see also Allport 1951: vi), the Gluecks themselves were a major influence on Cabot developing the CSYS, let alone his plans for a long-term follow-up and to investigate the natural history of offending. Cabot’s motivation to prevent delinquency in the first instance was partly owing to his revulsion at the high rate of recidivism (80% after 5-15 years) reported in the Gluecks’ study of male offenders in the Massachusetts Reformatory (Glueck and Glueck 1930). He made this clear in the book’s foreword (Cabot 1930) and in his presidential address to the 58th annual meeting of the National Conference of Social Work:

How splendidly ineffective are our foolish pea-shooters, our ‘reformatory’ attempts to change habits of delinquency! … I have an idea that the treatment of juvenile delinquency is now bad, wasteful, and ineffective. This idea comes mostly from the Gluecks’ studies, published and not yet published. (Cabot 1931: 440, 452)

The second notable development is the origins of the study’s novel and highly rigorous evaluation design: pair-matching in combination with random allocation (see Welsh et al. 2019a). By the beginning of the planning of the study (in 1935), there had been only four published randomized controlled experiments of social interventions (Forsetlund et al. 2007) and, outside of medicine (Amberson et al. 1931), not one example of researchers employing matching with random allocation. It is noteworthy that matching was the predominant method used to generate control groups in intervention studies before the first published use of random allocation in 1928 (Forsetlund et al. 2007). This prompted a key question: What motivated Cabot to use pair-matching in combination with random allocation when other rigorous designs were available and in use? Our research initially narrowed in on two plausible explanations: The design “represented a natural carry-over from medicine” (something that Joan put forth; see McCord 1992), and/or “it appealed to Cabot on the grounds of using an even more rigorous method of experimentation than what was the standard of the day” (Welsh et al. 2019a: 446). Subsequent archival research has offered additional support for the latter (rather than the former), with one point concerning the need to overcome additional uncertainties about testing intervention effects on social behavior (Claghorn 1927). Powers and Witmer (1951: 82), in documenting the rigors of the pair-matching process, echoed this view:

This account of the matching process, unavoidably complex, brings to light the difficulties of achieving adequate experimental controls in investigations of therapeutic methods; indeed, in any investigation that concerns itself with personalities or social behavior.

The third development that has been the focus of our research uses the CSYS as the focal point to understand some of the tensions that undergird the pursuit of experimental evaluation methods to achieve like-with-like comparisons (e.g., pre-allocation stratification and matching vs. alternate allocation and random allocation) and the comparative histories of this tension in medicine and the social sciences, with a particular emphasis on criminology (see Podolsky et al. 2021; Welsh et al. 2020a, 2021). It is noteworthy that the early history of random allocation within matched pairs intersects with some major figures in experimentation, including Ronald Fisher (1926), who advocated for random allocation among agriculture plots, and William Sealy Gosset, who published under the pseudonym ‘Student’ and favored ‘balanced randomization’ (Student 1931).


Joan McCord was one hell of an experimental criminologist! This could be an appropriate way to end the lecture. Of course, today’s experimental criminologists already know this. Less well known is how Joan revived the Cambridge-Somerville Youth Study, made it into the pioneering longitudinal-experimental study that we know today, and set the stage for a new program of research on the study, including a longer follow-up covering the participants’ full life-course. Impressively, this is only part of Joan McCord’s legacy; she made many other important contributions to criminology and aligned disciplines (see Sayre-McCord 2007; Tremblay et al. 2019). Crucial to Joan’s early and later work on the CSYS is how she was influenced by and drew inspiration from those who developed and worked on the study in its formative years. Also noteworthy (and not surprising) is how her formidable work on the study has influenced other experimental criminologists. As impressive as all of this may be, Joan might well have remarked that this is exactly how science is supposed to work.


Shortly after starting our research on the study, I received a letter from my colleague Linda Williams. Inside was a one-page note titled “Kevin’s ‘Top 15 Lessons’ from Joan McCord.” With the generous permission of Kevin Conway, who was Joan’s first Ph.D. student and a close colleague and collaborator, these lessons are reproduced here. I have done so because it seems quite fitting to end an award lecture in Joan McCord’s name by recognizing the full richness of her character: a deep connection to humanity, love of life, and a relentless pursuit of the truth. Here are the lessons:

(15) Take breaks to play tennis and ping-pong.

(14) When raising kids, try to see the world through their eyes.

(13) Think through a research question from beginning to end—articulate the idea, the research design, and the analyses. “OK, now how would you test that?”

(12) When travelling, always make time for viewing art.

(11) Never ask a yes-or-no question. It is better to plead ignorance and ask for forgiveness, than to be told “no.”

(10) Drink strong, freshly ground coffee.

(9) Evaluate, evaluate, evaluate. Even the most well intended attempts to help can in fact harm. As a corollary, randomize, randomize, randomize.

(8) Criticism is good. Joan once said that she never really had a good teacher—at Harvard or Stanford—because she never thought that she received the criticism that she deserved.

(7) Punishment is bad. It doesn’t work and may cause harm.

(6) Frequent used bookstores. And, if your book collection causes the foundation of your home to sink, hire somebody to install support beams.

(5) Work hard. There is no substitute for hard work.

(4) Go to the opera. Let no weather stand in your way.

(3) I would be terribly remiss if I did not mention a lesson from her late husband, Professor Carl Silver, the self-proclaimed Sage of Narberth. The lesson: Ideas are cheap. You should be able to come up with a dozen decent ideas over a pitcher of beer; data are expensive.

(1) Live life with enthusiasm. Anybody who knew Joan knew that she had a real zest for life, for travel, adventure, and trying new things. (Conway 2004, emphasis in original)


This is a team award, and I wish to thank current members of the Cambridge-Somerville Youth Study research team (Jack McDevitt, Steven Zane, Alexis Yohros, Heather Paterson, and Jillian Reeves) and several other colleagues who continue to be wonderful collaborators and/or advisors on research on the study: Scott Podolsky, Richard Tremblay, and David Farrington. I am also extremely grateful to Kevin Conway for allowing me to reproduce key lessons he learned from Joan McCord.


Allport, G. W. (1951). Foreword. In E. Powers & H. L. Witmer, An experiment in the prevention of delinquency: The Cambridge-Somerville Youth Study (pp. v-xxx). New York: Columbia University Press.

Amberson, J. B., McMahon, B. T., & Pinner, M. (1931). A clinical trial of sanocrysin in pulmonary tuberculosis. American Review of Tuberculosis, 24, 401-435.

Braga, A. A., Welsh, B. C., Papachristos, A. V., Schnell, C., & Grossman, L. (2014). The growth of randomized experiments in policing: The vital few and the salience of mentoring. Journal of Experimental Criminology, 10, 1-28.

Cabot, R. C. (1930). Foreword. In S. Glueck & E. T. Glueck, 500 criminal careers (pp. vii-xiii). New York: Knopf.

Cabot, R. C. (1931). Treatment in social case work and the need of criteria and of tests of its success and failure. Hospital Social Services, 24, 435-453.

Claghorn, K. H. (1927). The problem of measuring social treatment. Social Service Review, 1, 181-193.

Conway, K. P. (2004). Kevin’s ‘Top 15 Lessons’ from Joan McCord. Paper presented at the Annual Meeting of the American Society of Criminology, Nashville, TN, November 17-20, 2004.

deQ. Cabot, P. S. (1940). A long-term study of children: The Cambridge-Somerville Youth Study. Child Development, 11, 143-151.

Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: Peer groups and problem behavior. American Psychologist, 54, 755-764.

Farrington, D. P. (1983). Randomized experiments on crime and justice. Crime and Justice, 4, 257-308.

Farrington, D. P. (2006). Key longitudinal-experimental studies in criminology. Journal of Experimental Criminology, 2, 121-141.

Farrington, D. P. (2007). Introduction. In G. Sayre-McCord (Ed.), Crime and family: Selected essays of Joan McCord (pp. 1-10). Philadelphia, PA: Temple University Press.

Farrington, D. P. (2012). Foreword: Looking back and forward. In R. Loeber & B. C. Welsh (Eds.), The future of criminology (pp. xvii-xxiv). New York: Oxford University Press.

Farrington, D. P. (2013). Longitudinal and experimental research in criminology. In M. Tonry (Ed.), Crime and justice: 1975-2025 (pp. 453-527). Chicago: University of Chicago Press.

Farrington, D. P. (2021). Long-term follow-ups of criminological experiments. Victims and Offenders, 16, 996-1010.

Farrington, D. P., & Hawkins, J. D. (2019). The need for long-term follow-ups of delinquency prevention experiments. JAMA Network Open, 2(3), 1-3 (e190780).

Farrington, D. P., & MacKenzie, D. L. (2013). Long-term follow-ups of experimental interventions. Journal of Experimental Criminology, 9, 385-388.

Fisher, R. A. (1926). The arrangement of field experiments. Journal of the Ministry of Agriculture, 33, 503-513.

Forsetlund, L., Chalmers, I., & Bjørndal, A. (2007). When was random allocation first used to generate comparison groups in experiments to assess the effect of social interventions? Economics of Innovation and New Technology, 16, 371-384.

Glueck, S., & Glueck, E. T. (1930). 500 criminal careers. New York: Knopf.

Glueck, S., & Glueck, E. T. (1950). Unraveling juvenile delinquency. New York: The Commonwealth Fund.

Gottfredson, D. C. (2010). Deviancy training: Understanding how preventive interventions harm. Journal of Experimental Criminology, 6, 229-243.

Hill, K. G., Bailey, J. A., Steeger, C. M., Hawkins, J. D., Catalano, R. F., Kosterman, R., Epstein, M., & Abbott, R. D. (2020). Outcomes of childhood preventive intervention across 2 generations: A nonrandomized controlled trial. JAMA Pediatrics, 174, 764-771.

Hill, K. G., Woodward, D., Woelfel, T., Hawkins, J. D., & Green, S. (2016). Planning for long-term follow-up: Strategies learned from longitudinal studies. Prevention Science, 17, 806-818.

Kosterman, R., Hawkins, J. D., Hill, K. G., Bailey, J. A., Catalano, R. F., & Abbott, R. D. (2019). Effects of social development intervention in childhood on adult life at ages 30 to 39. Prevention Science, 20, 986-995.

Laub, J. H., & Sampson, R. J. (1991). The Sutherland-Glueck debate: On the sociology of criminological knowledge. American Journal of Sociology, 96, 1402-1440.

Laub, J. H., & Sampson, R. J. (2003). Shared beginnings, divergent lives: Delinquent boys to age 70. Cambridge, MA: Harvard University Press.

McCord, J. (1978). A thirty-year follow-up of treatment effects. American Psychologist, 33, 284-289.

McCord, J. (1981). Consideration of some effects of a counseling program. In S. E. Martin, L. B. Sechrest, & R. Redner (Eds.), New directions in the rehabilitation of criminal offenders (pp. 394-405). Washington, DC: National Academy Press.

McCord, J. (1984). A longitudinal study of personality development. In S. A. Mednick, M. Harway, & K. M. Finello (Eds.), Handbook of longitudinal research, vol. 2 (pp. 522-531). New York: Praeger.

McCord, J. (1991). The cycle of crime and socialization practices. Journal of Criminal Law & Criminology, 82, 211-228.

McCord, J. (1992). The Cambridge-Somerville Study: A pioneering longitudinal experimental study of delinquency prevention. In J. McCord & R. E. Tremblay (Eds.), Preventing antisocial behavior: Interventions from birth through adolescence (pp. 196-206). New York: Guilford Press.

McCord, J. (2002a). Counterproductive juvenile justice. Australian and New Zealand Journal of Criminology, 35, 230-237.

McCord, J. (2002b). Learning how to learn and its sequelae. In G. Geis & M. Dodge (Eds.), Lessons of criminology (pp. 95-108). Cincinnati, OH: Anderson.

McCord, J. (2003). Cures that harm: Unanticipated outcomes of crime prevention programs. Annals of the American Academy of Political and Social Science, 587, 16-30.

McCord, J., & McCord, W. (1959a). A follow-up report on the Cambridge-Somerville Youth Study. Annals of the American Academy of Political and Social Science, 322, 89-96.

McCord, J., Spatz Widom, C., & Crowell, N. A. (Eds.) (2001). Juvenile crime, juvenile justice. Washington, DC: National Academy Press.

McCord, W., & McCord, J. (1956). Psychopathy and delinquency. New York: Grune and Stratton.

McCord, W., & McCord, J. (1959b). Origins of crime: A new evaluation of the Cambridge-Somerville Youth Study. New York: Columbia University Press.

McCord, W., & McCord, J. (1960). Origins of alcoholism. Stanford, CA: Stanford University Press.

McCord J., Tremblay, R. E., Vitaro, F., & Desmarais-Gervais, L. (1994). Boys’ disruptive behaviour, school adjustment, and delinquency: The Montreal Prevention Experiment. International Journal of Behavioral Development, 17, 739-752.

Mears, D. P., & Cochran, J. C. (2019). Fundamentals of criminological and criminal justice inquiry: The science and art of conducting, evaluating, and using research. New York: Cambridge University Press.

Moule, R. K., & Decker, S. H. (2013). ‘Hidden in plain sight:’ Locating the men and women of the 1954 Boston Special Youth Program. Journal of Qualitative Criminal Justice and Criminology, 1, 267-291.

O’Donnell, C. R., & Williams, I. L. (2013). The Buddy System: A 35-year follow-up of criminal offenses. Clinical Psychological Science1, 54-66.

Podolsky, S. H., Welsh, B.C., & Zane, S. N. (2021). Richard Cabot, pair-matched random allocation, and the attempt to compare like with like in the social sciences and medicine. Part II: The context of medicine and public health. Journal of the Royal Society of Medicine, 114, 264-270.

Poulin, F., Dishion, T. J., & Burraston, B. (2001). 3-year iatrogenic effects associated with aggregating high-risk adolescents in cognitive-behavioral preventive interventions. Applied Developmental Science, 5, 214-224.

Powers, E. (1949). An experiment in prevention of delinquency. Annals of the American Academy of Political and Social Science, 261, 77-88.

Powers, E. (1950). Some reflections on juvenile delinquency. Federal Probation, 14, 21-26.

Powers, E., & Witmer, H. L. (1951). An experiment in the prevention of delinquency: The Cambridge-Somerville Youth Study. New York: Columbia University Press.

Rafter, N. (2004). Earnest A. Hooton and the biological tradition in American criminology. Criminology, 42, 735-771.

Rafter, N. (2010). Silence and memory in criminology: The American Society of Criminology 2009 Sutherland address. Criminology, 48, 339-356.

Sayre-McCord, G. (Ed.) (2007). Crime and family: Selected essays of Joan McCord. Philadelphia, PA: Temple University Press.

Sherman, L. W. (2005). The use and usefulness of criminology, 1751-2005: Enlightened justice and its failures. Annals of the American Academy of Political and Social Science, 600, 115-135.

Sherman, L. W., & Harris, H. M. (2013). Increased homicide victimization of suspects arrested for domestic assault: A 23-year follow-up of the Milwaukee Domestic Violence Experiment (MilDVE). Journal of Experimental Criminology9, 491-514.

Skinner, G., Farrington, D. P., & Jolliffe, D. (2021). Criminal careers and early death: Relationships in the Cambridge Study in Delinquent Development. British Journal of Criminology, in press.

Student [Gosset, W. S.] (1931). The Lanarkshire milk experiment. Biometrika, 23, 398-406.

Tremblay, R. E. (2005). Towards an epigenetic approach to experimental criminology. Journal of Experimental Criminology, 1, 397-415.

Tremblay, R. E. (Ed.) (2021). The science of violent behavior development and prevention: Contributions of the Second World War generation. New York: Cambridge University Press.

Tremblay, R. E., & Craig, W. M. (1995). Developmental crime prevention. In M. Tonry & D. P. Farrington (Eds.), Building a safer society: Strategic approaches to crime prevention (pp. 151-236). Chicago: University of Chicago Press.

Tremblay, R. E., & Farrington, D. P. (2004). In memoriam: Joan McCord, 1930-2004. The Criminologist, 29(2), 6-7.

Tremblay, R. E., Vitaro, F., Bertrand, L., LeBlanc, M., Beauchesne, H., Boileau, H., & David, L. (1992). Parent and child training to prevent early onset of delinquency: The Montréal Longitudinal-Experimental Study. In J. McCord & R. E. Tremblay (Eds.), Preventing antisocial behavior: Interventions from birth through adolescence (pp. 117-138). New York: Guilford Press.

Tremblay, R. E., Welsh, B. C., & Sayre-McCord, G. (2019). Crime and the life-course, prevention, experiments, and truth seeking: Joan McCord’s pioneering contributions to criminology. Annual Review of Criminology, 2, 1-20.

Weisburd, D., & Petrosino, A. (2004). Experiments, criminology. In K. Kempf-Leonard (Ed.), Encyclopedia of social measurement (pp. 877-884). San Diego, CA: Academic Press.

Welsh, B. C., Dill, N. E., & Zane, S. N. (2019a). The first delinquency prevention experiment: A socio-historical review of the origins of the Cambridge-Somerville Youth Study’s research design. Journal of Experimental Criminology, 15, 441-451.

Welsh, B. C., Podolsky, S. H., & Zane, S. N. (2020a). Between medicine and criminology: Richard Cabot’s contribution to the design of experimental evaluations of social interventions in the late 1930s. James Lind Library Bulletin. Available at:

Welsh, B. C., Podolsky, S. H., & Zane, S. N. (2021). Richard Cabot, pair-matched random allocation, and the attempt to compare like with like in the social sciences and medicine. Part I: The context of the social sciences. Journal of the Royal Society of Medicine, 114, 212-217.

Welsh, B. C., & Rocque, M. (2014). When crime prevention harms: A review of systematic reviews. Journal of Experimental Criminology, 10, 245-266.

Welsh, B. C., & Tremblay, R. E. (2021). Early developmental crime prevention forged through knowledge translation: A window into a century of prevention experiments. Journal of Developmental and Life-Course Criminology, 7, 1-16.

Welsh, B. C., Yohros, A., & Zane, S. N. (2020b). Understanding iatrogenic effects for evidence-based policy: A review of crime and violence prevention programs. Aggression and Violent Behavior, 55, 1-7 (e101511).

Welsh, B. C., Zane, S. N., & Rocque, M. (2017). Delinquency prevention for individual change: Richard Clarke Cabot and the making of the Cambridge-Somerville Youth Study. Journal of Criminal Justice, 52, 79-89.

Welsh, B. C., Zane, S. N., & Wexler, A. B. (2018). The Cambridge-Somerville Youth Study and intergenerational transmission of criminal offending: Key findings and planning for the next generation. In V. I. Eichelsheim & S. G. A. van de Weijer (Eds.), Intergenerational continuity of criminal and antisocial behaviour: An international overview of studies (pp. 260-275). London: Routledge.

Welsh, B. C., Zane, S. N., Zimmerman, G. M., & Yohros, A. (2019b). Association of a crime prevention program for boys with mortality 72 years after the intervention: Follow-up of a randomized clinical trial. JAMA Network Open, 2(3), 1-11 (e190782).

Welsh, B. C., & Zimmerman, G. M. (2015). Who cares about a delinquency prevention experiment of Boston boys born in the 1920s and 1930s? The need for long-term follow-ups in criminology. Criminal Behaviour and Mental Health, 25, 331-337.

Zane, S. N., Welsh, B. C., & Zimmerman, G. M. (2016). Examining the iatrogenic effects of the Cambridge-Somerville Youth Study: Existing explanations and new appraisals. British Journal of Criminology, 56, 141-160.

Zane, S. N., Welsh, B. C., & Zimmerman, G. M. (2017). Examining the historical developments and contemporary relevance of the longitudinal-experimental design of the Cambridge-Somerville Youth Study: Utility for research on intergenerational transmission of offending. Adolescent Research Review, 2, 99-111.

Zane, S. N., Welsh, B. C., & Zimmerman, G. M. (2019). Criminal offending and mortality over the full life-course: A 70-year follow-up of the Cambridge-Somerville Youth Study. Journal of Quantitative Criminology, 35, 691-713.

Brandon C. Welsh, Ph.D., is a Professor in the School of Criminology and Criminal Justice at Northeastern University and the Director of the Cambridge-Somerville Youth Study. His research focuses on the prevention of delinquency, crime, and violence and evidence-based social policy. He is the author or editor of 11 books, including Experimental Criminology: Prospects for Advancing Science and Public Policy (Cambridge University Press) and, most recently with Eric Piza, The Globalization of Evidence-Based Policing: Innovations in Bridging the Research-Practice Divide (Routledge). He received his Ph.D. in criminology from Cambridge University.

No comments here
Why not start the discussion?