Researching U.S. Urban Murder Rates and Group Practice of the Transcendental Meditation and TM-Sidhi Program: Evaluation of a Prospective Quasi-Experiment

A quasi-experiment tested whether a large group practicing the Transcendental Meditation and TM-Sidhi program could reduce homicide rates in U.S. urban areas. Time series analysis of monthly data from 2002 to 2010 showed a significant reduction in the homicide rate trend during the intervention period from 2007 to 2010. Controlling for pre-existing trends, seasonality, and autocorrelation, the estimated total reduction in homicide rate was 28.4%, or 7.1% annually. This corresponds to an estimated 4,136 murders averted. The authors suggest the results are unlikely due to spurious regression and discuss the mechanism in terms of reduced collective stress.

Study at a glance

Abstract

Creation of a sustainable society ideally should include promotion of an enhanced overall quality of life, including freedom from crime, violence, and other key indicators of social stress. This study is part of a comprehensive empirical evaluation of the results of a prospective four-year quasi-experiment that sought to reduce rates of homicide and violent crime as well as to improve other measures of the U.S. quality of life and public health. The current research tests the hypothesis that group practice of the Transcendental Meditation and TM-Sidhi program by a group of theoretically predicted size would be sufficient to reduce collective stress in the larger population, as reflected in decreased rates of homicide in a sample of 206 large U.S. urban areas. Time series regression analysis of monthly data for 2002–2010 using a broken trend intervention model found significant reductions in trend for the urban homicide rate during the 2007–2010 intervention period. Controlling for pre-intervention trends, seasonality, and autocorrelation, the estimated total reduction in homicide rate was 28.4% (7.1% annually). The practical significance of these findings is also indicated by an estimated 4,136 murders averted by the reduced trend in murder rate during the intervention. Diagnostic tests are satisfactory and indicate that the results are unlikely due to “spurious regression.” The mechanism for these macro-social effects is discussed in the light of possible alternative hypotheses.