An ambitious theory of humility, compassion, and divine gratitude

Religiosity is correlated with many different psychological states and behaviors, such as empathy and generosity, or exclusivity and violence. However, the causal connections between belief, experience, and behavior are presently unclear. Psychologists Neal Krause and David Hayward recently completed a seven-year longitudinal study that seeks to tease apart the mental components of a religious life and establish causation. They follow a line of cognitive dominoes falling into each other: religious commitment tips into humility, into compassion, into support for others, into a sense of meaning. The last piece falls on gratitude to God. In the researchers’ view, this gratitude is actually indirect thankfulness for the people in the churchgoer’s life. Their theoretical scheme is an interesting attempt at integrating the abundant, but disparate research on the cognitive mechanisms underpinning religion.

Krause and Hayward are both professors at the Department of Health Behavior and Health Education at the University of Michigan School of Public Health. Both work on the correlations between religion and health. Krause’s primary research focus is on stress and coping among elderly people, while Hayward works on statistical modeling. The two combined their respective skill sets to undertake a massive nationwide survey, spanning from 2001 to 2008. For their paper, the team tested six hypotheses using data from the larger study, and found statistically significant correlations. However, the authors also readily acknowledge that their sprawling conclusions are not without problems.

The population under study consists entirely of elderly Christians, drawn from a database maintained by the Centers for Medicare and Medicaid Services. All participants were at least 66 years old, living independently, and of either African-American or European-American ethnicity (64.3% and 35.7%, respectively). 1,500 people were interviewed in their homes in the first round, although this number dropped off over time due to advancing age or death. To give a sense of the change, the second wave included 1,024 reinterviews. The fifth and final wave of interviews included only 718 of the original sample, but this group was supplemented with an additional population of younger interviewees.

One of Krause and Hayward’s main contentions is that religious value is largely derived from social bonds. Typically, one finds Durkheim’s work at the source of social models of religion, but the authors of this study cite sociologist Edward Alsworth Ross, who defined religion as “the conviction of an ideal bond between members of society and the feelings that arise in consequence of that conviction.” From this standpoint, Krause and Hayward seek to identify the specific characteristics and social actions which forge and maintain this “ideal bond.” Six hypotheses were tested against the longitudinal interviews, with each one presumably leading to the next:

1. People who frequently attend religious services tend to have a greater overall commitment to their belief system.
2. Higher commitment is correlated with greater humility.
3. Greater humility is correlated with higher levels of compassion.
4. Those with high levels of compassion are most likely to support family and friends.
5. Supporting family and friends yields a deeper sense of meaning.
6. Those with a deeper sense of meaning tend to express more gratitude toward God.

Krause and Hayward’s data analysis show a high positive correlation between the psychological states and/or behaviors within each individual hypothesis, but there’s little evidence for overarching connections. One interesting aside from their fifth hypothesis is that respondents only claimed to have a deeper sense of religious meaning if their altruism was primarily being directed toward family and friends. Those who spent most of their charitable energies on strangers did not tend to report a greater sense of meaning. This supports the notion that religion tends to stimulate parochial altruism – or kindness toward those within one’s own group – rather than altruistic universalism.

However interesting Krause and Hayward’s proposed chain of causation may be, they don’t provide much in the way of supporting evidence for directionality. Only two clear causal links were actually established in the research. The researchers compared the first wave of interviews with the second, and found that frequent church attendance did predict greater religious commitment later on, and not the reverse. In the second instance, a similar link was found between religious meaning and gratitude to God, with the former being more likely to precede the latter. With only these two pieces of evidence, the direction of causality implied by the ordering of the six hypotheses remains speculative.

Ultimately, Krause and Hayward have constructed an ambitious theory of religion and well-being, but they don’t claim that it is airtight. In fact, their paper concludes by highlighting four major problems with their study. They remind the reader again and again that the any of the mental states and activities tested for could have causal links to any other, and in multiple directions. The clear arrow from church attendance to gratitude toward God is merely the strongest causal series that the authors could establish. Secondly, the data is entirely self-reported. This is not a deal-breaker, but problems abound with this methodology, both because the interviewees may be trying to convey certain messages about themselves to the interviewer, and because people tend to have skewed views of themselves. Third, because Krause and Hayward intentionally chose a sample in the latter years of life, their research cannot be extrapolated to apply to young people. In other words, their causal model only includes the elderly. And finally, because their sample is limited to Christians, the model has no confirmed applications to any other religious tradition. In fact, without sorting their data to account for the broad religious diversity within Christianity, there’s no way to know if any of these six hypotheses would hold true from one denomination to the next.

Krause and Hayward’s study is an interesting attempt to synthesize a few of the correlations that come up again and again in the scientific study of religion. It’s also a painful reminder of how difficult it is to establish multiple causal links between religious psychology and observable behavior, let alone to create a sound theoretical scheme that can make reliable predictions. By taking this single faltering step, the authors remind us of how far the field has to go.

For more, read “Humility, compassion, and gratitude to God: Assessing the relationships among key religious virtues” in Psychology of Religion and Spirituality.