Robert Rosenthal Memorial Tributes

• A Tribute from Howard Friedman

  UCR Distinguished Professor Robert Rosenthal died January 5, 2024 in Riverside, Calif. at age 90. He played a key role in our department for 25 years, including being named a University Professor, the highest honor in the California-wide UC system.

  A number of us at UCR knew him first from his years at Harvard, as a mentor, colleague, research collaborator, teacher, and/or helpful scholar. I first met him at Harvard when he was 39, already world famous. (Dr. Rosenthal left his position as the Edgar Pierce Professor of Psychology at Harvard to come to UCR in 1999.) So there was a uniquely wonderful regularity as he assumed the same roles for a new generation at UCR for so many years.

  Bob—he always insisted that everyone call him “Bob”—was born in Germany just as the Nazis came to power. His sterling intellectual journeys echoed his many geographic journeys. As a young child, he fled to Rhodesia (a British African protectorate), and then to New York and then Los Angeles. He reported that the kids in Germany harassed him because he was Jewish, and the British kids in Rhodesia beat him up on the way home from school because he was German. Convincing his teachers to let him leave school a little early each day, he later liked to joke that he was the psychologist who invented the head start program. That was Bob. Always ebullient and upbeat, he only turned serious and firm when someone was being illogical or using a suboptimal research design or the wrong statistics.

  Bob received his BA and PhD from UCLA, where he also worked in the UCLA stadium hot dog stand. He studied with Bruno Klopfer, a student of Carl Jung. Klopfer was a developer of the Rorschach and related projective tests, and Bob’s dissertation derived from the Thematic Apperception Test (TAT). It compared projection in paranoid schizophrenic patients versus a normal control group after an experimental manipulation. Perhaps motivated by his teenage interest in understanding the scientific basis of parapsychology and ESP, Bob examined his dissertation pretest data and found that his groups already differed in the direction that would support his expected result. Thus was launched the career and the research field of experimenter expectancy effects—widely known as “the Rosenthal effect.”

  Always cognizant of such subtle biases and expectancies, Dr. Rosenthal pointed out that when he moved from his first academic position at North Dakota to a professorship at Harvard in 1962, his pending unpublished papers (now submitted with the Harvard letterhead) were suddenly accepted for publication. Contacted by a school principal, Lenore Jacobson, Bob then jumped in to study expectancy effects in teachers—a real-life setting—which led to the monumental Pygmalion in the Classroom (1968). The tremendously impactful Pygmalion effect describes situations where high expectations from others improve one’s behavior and performance.

  The Pygmalion studies at first brought a vicious reaction from many educators, who erroneously viewed it as an attack on the integrity of teachers. As more and more studies of expectancy effects  were conducted, there was no single mathematical way to summarize them. So, true to form, Dr. Rosenthal looked for an analytic, scientific solution. He spent many years working with colleagues, especially the Harvard statistician Donald Rubin, helping to develop and refine the techniques of “meta-analysis.” Meta-analysis became a major underpinning of psychological and policy research. Of course, his scientific findings about interpersonal relations have come to be not only understood, but also capitalized upon, as educators, coaches, therapists, and leaders learned to expect the best from their students and advisees.

  Bob’s efforts as a deep thinker and leader about research ethics are sometimes underappreciated. He persuasively argued that studies are highly unethical if they are poorly designed or analyzed—they are not only a waste of time and resources, but also can lead to unwarranted conclusions and policies. It is unethical to conduct poor quality research—ranging from participant selection biases to false inferring of causality (both of which he studied in some depth)—but also by being overly cautious and failing to do a study that needs to be done. Notably, his work helped lead the efforts for medical and policy research to do fully “double-blind” (double masked) studies.

  Shunning long-term grand planning in research, Professor Rosenthal took things step-by-step. Endeavoring to discern precisely how expectancies were subtly communicated, he turned to the study of nonverbal communication. And of course he thereby helped pioneer yet another field. Who was more sensitive to reading nonverbal cues and which channels (facial, vocal, body movement) were key in which situations? Reliable and valid measurement was needed, and so an extensive research project developed the remarkable Profile of Nonverbal Sensitivity (PONS test). Many students and colleagues joined the efforts, which helped launch the revitalized modern study of nonverbal communication, and his students—and their students—studying nonverbal communication are now a major force in social psychology.

  Bob always wanted to know the probability of a chance finding, but, saying that God also loved p<.06, he helped lead the way against the ridiculous but longstanding practice whereby psychology journals would reject manuscripts where significance testing did not reach the magical .05 level, regardless of the quality and importance of the research. Noting that two p’s of .06 are stronger evidence against the null than one p<.05, Bob helped pioneer the turn in social science analysis to effect sizes and meta-analysis, and away from omnibus F-tests. His longstanding campaigns, with distinguished colleagues, to emphasize effect sizes resound throughout today’s research in psychology and in medicine. Along the way, one of his points was that studies with null results are less likely to be published, and so again, true to form, Dr. Rosenthal looked for biases in meta-analysis and wrote extensively about solving the “file drawer problem.”

  Ever concerned with statistical power and precision, Bob spent much effort developing and teaching Contrast Analysis. His legions of brilliant and essential collaborators and students on these various topics are too numerous to name here but are a Who’s Who of psychologists and methodologists.

  We loved having this prize-winning teacher at UCR. His many teaching awards include the Association for Psychological Science (APS) Mentor Award for Life Achievement. Generations of students remember Bob  telling the story of Clever Hans, the savvy horse studied by Oskar Pfungst in 1911, who could seemingly solve math problems and answer by moving his hoof. (Hans was actually reading the nonverbal cues of his questioners about when to stop pawing). In lectures, Bob would paw the ground with his leg, gleefully imitating the horse. When asked how he successfully mentored so many superstar students, Professor Rosenthal answered, “High expectations.”

  Here at UCR, there is an eponymous ongoing weekly methods proseminar called R-Squared. His many awards include the APA’s Gold Medal Award for Life Achievement, election to the American Academy of Arts & Sciences, the James McKeen Cattell career award from the APS, and an Honorary Doctorate from the University of Giessen—the city where he was born and fled.

  With hundreds of thousands of citations, his work revolutionized the fields of psychology and education, and fundamentally improved social science research methods across disciplines. At UCR, at Harvard, and around the world, Professor Rosenthal is survived by the countless students and colleagues from whom he always expected the best.

   

  Howard S. Friedman
  University of California, Riverside
  June 2024

• Remarks Delivered by David Funder

  Remarks delivered by David Funder at Robert Rosenthal Memorial
  Forest Lawn, February 24, 2024

  It is a deep honor to be asked to speak today in memory of Robert Rosenthal.

  Although it occurs to me, I don’t recall anybody, ever even once, calling him “Robert.”; he’s always been either “Bob” or – to most students – “Dr. Rosenthal.”

  Bob is one of my heroes; many and probably all of you, I’m sure, feel the same way. My goal today is talk about why we feel this way about him.

  When Roberta and I were briefly discussing how today might go, I warned her that “I could go all day.” I could, but I won’t; still there is so much to say.

  I first met Bob in 1982, which in the way time compresses over the decades, doesn’t really seem all that long ago to me. I was a brand-new, young, and very naïve Assistant Professor newly arrived at Harvard, and afflicted with a very well-deserved case of imposter syndrome. Professor Rosenthal was a world-famous senior professor with an intimidating reputation whose office in William James Hall, as it happened, was two doors away from mine.

  I quickly learned what others also have discovered – you might be justifiably intimidated by the reputation, but you would not be intimidated by the person once you met him and experienced first-hand his generosity, kindness, and patience. At Harvard he did for me what he did for many others, which was to open doors, make me feel welcome and even as if I might belong, at a place that doesn’t always have that effect on people. He made sure I became a Fellow of Kirkland House, for example – even if I was never sure quite what that entailed – and made sure I was included in various opportunities across the campus.

  And, even better, Patti and I were hosted at Bob and Marylou’s beautiful home in Lexington for memorable evenings that included, one at least on occasion, generous quantities of Ouzo. (If you don’t know what that is, and I didn’t, it’s a delicious but dangerously potent Greek liqueur).

  The most regular encounter we had at Harvard, though, was the weekly “Quantitative Lunch,” as it was called, where Bob hosted an open forum to talk about data analytic issues, that attracted participants from across the university, notably including Donald Rubin, who Bob modestly described as his “statistical tutor.”).

  I also vividly remember an encounter in the hallway of William James Hall right after I had given a talk at the social psychology proseminar. I had described some of my research on how well two acquaintances tend to agree with each other in their descriptions of a third person, and I used the Pearson correlation coefficient to quantify the degree of agreement. Bob stopped me in in the hallway afterwards and told me something I really should have known already, that a related (but more difficult to compute in those days) number called the intraclass correlation was what I should have used. In a few minutes, standing in the hall, he taught me how to compute it, a lesson I used in all my subsequent research. And, it didn’t escape my notice, that he pointed out my error privately, not during my talk, taking me aside afterward. In that way he saved me from public embarrassment – which was absolutely typical of Bob’s kindness and consideration.

  Still, I couldn’t resist pointing out to Bob, after a few months of experience, that in almost all real cases the Pearson and intraclass correlations turn out to be almost exactly the same number. “Ah,” Bob replied gently, “but the intraclass correlation is the right number.” It is, of course.

  After just four years at Harvard, I was offered an opportunity at the University of Illinois. I felt torn about leaving Harvard, and I vividly remember Bob’s advice when he heard about my opportunity: “Go.”  So I went, and shortly after that move returned to my California roots by moving to UC Riverside.

  You might imagine my astonishment when, just a few years later, Robin DiMatteo, a fellow faculty member at UCR and a former student of Bob’s at Harvard, said, “you know, Bob Rosenthal might be interested in moving here.” To try to capture how that felt, imagine a community theater being told that Meryl Streep had shown up for an audition! I literally could not believe our luck.

  But there were connections already – Robin DiMatteo and Howard Friedman had both been students with Bob at Harvard in the years before I was there, and had stayed in contact. There were other attractions for Bob and Marylou. As part of their recruitment, Patti and I took them out to dinner and I recall Marylou was excited by the possibility of playing tennis outdoors in January. And when I mentioned to Bob that we already had in place a weekly quantitative seminar (called MAMA for reasons too complicated to explain), very similar to the one he had hosted for years at Harvard – his eyes lit up and I felt the deal was virtually done. 

  When Bob arrived at UCR, he acted anything but retired. He mentored graduate students, his own and others’, supervised undergraduate theses, taught our basic statistics course and introduced a new undergraduate course in advanced research methods. He was characteristically generous with his time and attention. Any graduate student in the Department who – in those early pre-google doc days – left a draft of a paper in his mailbox found it returned a few days later, covered in (friendly but thorough) comments written in red ink.

   The last course he officially taught was a seminar on his own work, which I was honored to co-teach with him. The course became known, to at least some people, as the “Bob Show.” It was a hit.

  But that weekly open forum, in which colleagues and students could and did talk about anything having to do with data analysis, was and remained the centerpiece of his teaching. I suspect it was one of the high points of Bob’s week, and he continued to participate even after his second retirement. He attended every single Tuesday at noon, via Zoom (which he learned to use!) through the end of Fall quarter of last year.

  I feel I should say a few things about Bob’s career and the accomplishments that made him – and this is no exaggeration at all – world famous.

  A warning: this summary is going to be appallingly oversimplified, much less than it deserves, and we really could go all day, or all semester, talking about his work.

  Bob’s research career was organized around three inter-locking themes: Expectancy effects, non-verbal behavior, and research methods, notably effect size and meta-analysis.

  First, expectancy effects: the – again, no exaggeration – world famous “Pygmalion in the classroom” studies and their many, many offshoots over the years generated attention, controversy, and one powerful take-home lesson that is undoubtedly true: The accomplishments – and other behaviors – of the people you know are profoundly affected by how you treat them. And – a little less obviously but perhaps even more profound – the accomplishments and other behaviors of yourself are profoundly affected by how you are treated.

  Some teachers and environments bring out your best – others, not so much. If you find yourself doing better work or learning more in the presence of a particular teacher or in a particular environment, you’ve made a valuable discovery. And, even better, with the best friends or the most valuable life partners – you will find – almost mysteriously in some cases – that they bring out your best attributes and for that reason you like yourself better when you are around them. A friend or partner like that is, in the word, a “keeper.”

  Second, I mentioned that how this happens – how expecting the best of someone tends to bring out their best – is rather mysterious. In the research of Bob and his students, it turned out to be easier to demonstrate that the effect existed, than to show exactly how it happened. The answer emerged slowly, and is still emerging, but an important basis has turned out to be what is called “nonverbal behavior.” This includes the body movements, facial expressions, and tones of voice that implicitly, rather than explicitly, and often unintentionally, communicate how we feel about each other. These effects are subtle but real. The study of nonverbal behavior has become an important area of psychology in its own right, and Bob is one of its pioneers and founders.

  Back in the early 1980’s, at Harvard, Bob’s lab included video equipment that was sophisticated for its time along with equipment to “content filter” speech – to record its tone but not content. Bob’s favorite example was a snippet of sound where an adult tells a child: “bbboontdye” (Aw, don’t cry). The sympathy is in the tone as much or more as in the words. Nonverbal behavior.

  As I said, this work went many directions, including the study of doctor-patient communication, where in collaboration with Robin DiMatteo and others research showed that when a doctor expresses engagement and interest in their patient through non-verbal behavior, the patient is more likely to take their medications when they should!

  Third, the early work on expectancy effects generated controversy and even opposition expressed through methodological critiques. In response, Bob turned himself into a statistical and methodological expert in his own right. He made fundamental contributions to the understanding of how to interpret the magnitude of a research effect, in collaboration with his friend and colleague Don Rubin, and explored the complex dynamics of recruiting and perhaps unintentionally influencing research participants with another longtime friend and colleague, Ralph Rosnow. I spoke on the phone with Dr. Rosnow a few days after Bob passed, and he mentioned that through the many years of their friendship and collaboration, they enjoyed a long phone call every Saturday up to and including the weekend before he passed.

  Another methodological contribution, which I believe was Bob’s most passionate interest in the latter part of his career, was meta-analysis. In part this grew out of the controversy over expectancy effects, but grew into an early awareness of and advocacy for an important lesson for all of science, which is this: One study, by itself, does not tell you very much. Findings come and go as different studies get done – this is particularly true, and worrisome, in the field of medicine – and the truth only emerges slowly as different studies get performed, integrated and summarized. Bob was a pioneer in developing basic tools of meta-analysis, in which the results of several and ideally many studies get summarized into a conclusion more powerful than any one study can support. Published reviews of the literature, in many fields beyond psychology – notably again, medicine – have come to rely heavily on meta-analytic methods to convey the lessons learned.

  And all of this, of course, is just a bare scraping of the surface of Bob’s research career, but I hope it gives a general sense of why he became so eminent, influential, and admired.

  I will conclude with a couple of observations about Bob Rosenthal, the person. One thing I think I have learned in my own research career is that inconsistent behavior – being one person at one time and other person at another – is a psychological danger sign. And consistent behavior -- -being the same person wherever you go and whoever you are with – is a hallmark of psychological health. Bob Rosenthal was the most consistent person I ever met, particularly in how he treated people. If an undergraduate student, a colleague, or the President of the University were to pop their head into Bob’s office, they would find themselves treated exactly the same, with kindness, respect, patience, and genuine interest.

  Are there many people you can describe as universally beloved? They are rare, and Bob is one of those, for good reason.

  A cardinal attribute of how Bob treated other people is related to his work on expectancy effects. I don’t know if he became this way as a result of his research discoveries; I didn’t know him that long ago. But I strongly suspect he was this way all long. If you talked to him, he was so interested that you found yourself becoming more interesting. Anybody who teaches large classes, or does any sort of public speaking, knows that you seek out one or two people in the audience who seem to get what you are saying and look repeatedly to them for their reaction. If Bob was in the audience, he was always that person for me and I bet for many others. If you looked his direction, you would see that was smiling gently, nodding understandingly, and paying close attention to every word. In that way he made you better. I wish he was here today.

  He made us all better, and he is and will be missed.