Monday, December 3, 2012

The Power of Longitudinal Research

Some of the oldest and most contentious debates on human beings centre around the relative influence of heredity (genetics), environment and individual voluntary action on growth and development. These include whether mental illness has genetic origins, what factors determine “success” in life, and whether adults continue to “develop” as they grow older (or whether all development happens before a certain age). These questions cross disciplinary boundaries as they involve concepts from psychology, psychiatry, sociology, and genetics.

Great thinkers like Freud and Erikson made significant contributions to these debates, but many of their contributions were based on intuitive theorizing rather than rigorous empirical evidence. With time and careful research, some of their theories have been upheld, and others disproved! The studies that have made the most impact are longitudinal studies in which a carefully chosen cohort of respondents was tracked periodically over an extended period of time.
The Harvard Grant Study
One of the most well known of these studies is the Havard Grant study which commenced in the late 1930s and early 1940s and continues till this day. The survivors of the cohort (who were Harvard sophomores when they were recruited) have now entered their 90s, and the data collected therefore allows several inferences to be drawn on adult development.

George F. Vaillant was the director of the Harvard Grant Study for over two decades. His latest book, The Triumphs of Experience, presents the latest findings. I found it a fascinating read as it not only uncovers new insights, but also questions some of the conclusions reached at earlier stages of the study. The Harvard Grant Study draws its conclusions from rigorous multivariate analysis, but Vaillant presents the findings with a distinctive and rare combination of statistical rigour and empathy for his subjects – in addition to tables containing the statistical results, there are profiles (disguised, of course) of different respondents of the study, and these give the reader a sense of being part of the study team.
The original design and subsequent evolution of the study show how much our models of adult development have changed over time. At the time the study started, physical constitution and mental health indicators were expected to be important predictors of subsequent progress of the study. Parental/family relationships and childhood upbringing were thought to be unimportant. Yet, the latest Harvard Grant Study findings show that loving relationships during childhood are important for longevity and success in life.
Findings of the Harvard Grant Study

Some of the important findings of this study reported in The Triumphs of Experience:

  • Individuals develop through their adult lives as well, not only upto the stage of adolescence.

  • The impact of childhood trauma decreases over time; more importantly, the positive experiences of a loving childhood have enduring impact.

  • Being well integrated and self-driving while young helps people live longer.

  • Divorce led to happier marriages than the bottom third of sustaining marriages.

  • Alcoholism had bigger negative impacts than measured by most previous studies. It accounted for more than half of the divorces in the Grant Study. The study shows that it is unlikely that alcoholics can return safely to social drinking, thereby upholding the methodologies followed by organizations like Alcoholics Anonymous.

  • The involuntary coping styles predicted by Freud exist, and they are important for human effectiveness

Phases of Adult Maturation

I particularly liked the chapter on maturation where Vaillant extends Erikson’s work to identify six phases of adult maturation (or what he prefers to call “development tasks”). These are shown in the graphic below. These phases form an excellent blueprint for adult evolution. What struck me was how many of us get bogged down in the tasks of identity and career consolidation when there are more fulfilling tasks ahead of us. While it’s true that many people in India do get involved in generative and guardian-like tasks, these are often restricted to the family context. Though we often discuss India’s demographic dividend, remember that India has a large and growing number of older people as well, and it would be a shame if we don’t take advantage of their wisdom.

 Important Lessons for Management of Long-term Research Programs
The Harvard Grant Study is interesting from a research management perspective as well. Over its 70+ year lifespan so far, the study has transcended several research directors and team members, but the integrity of the study has not been compromised. George Vaillant estimates that about $ 20 million has been spent on the study over time, with an average cost of $10,000 per research paper published. The study has had different sponsors at different times, and while the study had to adapt itself to the priorities of these sponsors (such as a major retailer, cigarette company and a program against alcoholism), it still managed to sustain the collection of data related to its core research questions.
With its emphasis on the choice of appropriate control variables and other related issues of study design, this book is a great primer on how to design and adapt longitudinal research studies for maximum research impact. In India, we need to build a tradition of undertaking sustained, long-term research programs, and this book that describes the approach taken by the Grant study will be invaluable in this endeavour.
[See David Brooks’ nice piece on this study in the New York Times for another perspective]

1 comment:

  1. Provides hope for a better world and validates the role of a nurturing environment - within both family, in the larger society, at the work-place and in an inclusive world.