Pew's New Religious Typology

The Pew Forum has released a new report, The Religious Typology: A New Way to Categorize Americans by Religion (pdf here), that classifies Americans into three categories:  highly religious, somewhat religious, and non-religious.  Those categories are further broken down as shown in this table into the seven key groups that comprise the typology.

Broad categories	The seven groups	Description
Highly religious	Sunday Stalwarts	Traditional, actively involved in congregation.
	God-and-country Believers	Less active at church, socially and politically conservative, anti-immigrant
	Diversely Devout	Traditional plus belief in psychics, reincarnation, and spiritual energy
Somewhat religious	Relaxed Religious	Religion important personally but not necessary to be moral, not engaged in traditional practices
	Spiritually Awake	Believe in heaven, hell, and New Age belief, but do not engage in traditional practices
Non-religious	Religion Resisters	Believe organized religious more bad than good, politically liberal and Democratic
	Solidly Secular	Virtually no religious beliefs, reject New Age beliefs

Because the seven groups are constructed based on religious behaviors and beliefs, they will cut across the denominational identities often used to classify individuals.  For example, 13% of the Sunday Stalwarts are Catholic, but 25% of the Relaxed Religious are Catholic, as are 9% of the Religious Resisters.  Also, the Sunday Stalwart group includes individuals from non-Christian faiths (even though the do not technically worship on Sunday).

A primary merit of this seven-group classification is that it highlights some of the diversity in the American religious marketplace.  The challenge, of course, is to find the right number of groups.  To many and the classifications become meaningless, but too few and the classifications are also meaningless.  For this report, Pew used a statistical procedure called cluster analysis to let the data reveal what appear to be statistically coherent subgroupings in the data.  Doing this does involve making assumptions, such as what are the crucial data to include in the analysis.  Here, Pew included data from 16 particular survey questions about religious practices and broad beliefs.

Of course, this procedure ignores other forms of diversity, such as more particular theological beliefs, that are not captured in those 16 questions.  But this is the nature of social scientific inquiry:  some data are privileged over others, wisdom and care should be used in deciding on what is privileged;  and honesty about the pros and cons  and strengths and weaknesses of the analysis and results should be maintained.