What is measurement of religiosity?
During the twentieth century, academic psychologists were concerned with developing an empirical science that would compare favorably with other natural sciences. Hence, the constructs in models of human behavior were operationalized with the goal of yielding numeric data that could be analyzed statistically. Quantitative methods (experiments, quasi-experiments, surveys, correlational studies) superseded the earlier descriptive, qualitative methods (Oedipal interpretations, clinical and literary case studies, introspective reports) and required relatively objective measurement of psychological attributes or traits. A fundamental notion in psychometric theory is that measurements taken on people should at the very least allow ordering along some continuum or dimension, such as intelligence or anxiety. However, measurement also may connote appraisal or understanding. Thus, the early explorations of religious experience (such as those of Sigmund Freud, Carl Jung, or William James) may be very broadly considered as involving, in some sense, the “measure” of religious experience.
The development of models and methods in the study of religiosity (religious feelings, beliefs, and behaviors) parallels this general timeline, although the emphasis on empiricism occurred in the latter half of the twentieth century. Since then, most psychologists of religion have been academic social psychologists who relied heavily on psychometric measures of religiosity, such as self-report questionnaires. However, since the 1980s, a resurgent interest in narrative or interpretive methods has competed with the psychometric tradition. Additionally, psychology of religion has become an international enterprise more open to qualitative methods. For example, Europeans are not averse to phenomenological interpretations of religious experience. The result of these diverging views is a broadening of the methodological base for testing models and theories of religion and an increasing sophistication with respect to measurement of religiosity. Many psychologists welcome this development, as they feel that more meaningful research is possible. For example, Kenneth I. Pargament, a renowned psychologist who studies coping behavior and religious beliefs, maintains that empirical methods must be balanced with phenomenological or interpretive methods, since religious experience is often private and symbolic and, therefore, not observable.
Mainstream social psychologists regard attitudes as learned habits for responding to social stimuli and attempt to identify the cognitive, affective (emotional), and behavioral components of the attitude. Religiosity (also known as religiousness) is generally understood as a person’s essential attitude toward religion. The word “religion” has a Latin root that implies binding and restraining. Religion is therefore a personal and social force that serves to bind people together in a community of worshipers, unite them in reverence with a spiritual dimension of existence, and restrain their inappropriate impulses via moral commandments.
The form of attitude surveying with which the average American would be most familiar is the national opinion poll as conducted by George Gallup, Louis Harris, a major newspaper or magazine, or a marketing research firm. For example, the Gallup Poll measures a person’s overall religiosity by asking, “How important would you say religion is in your own life?” Nationally, more than half of the adults surveyed in 2012 respond “very important,” where importance tends to increase with age. In addition, there are gender, ethnic, and church affiliation differences; for example, the subgroups comprising women, African Americans, or Protestants are most likely to say that religion is very important.
Pollsters usually break down religion into specific behavioral, cognitive, and affective components. An important behavioral component is affiliation with a particular denomination (a religious organization within the host culture usually referred to as a church). In 2013, according to Gallup, about 83 percent of all adults in the United States claimed a religious affiliation (such as Protestant, Catholic, or Jewish), but only about 69 percent said they were actually members of a specific denomination. The largest single denomination in the United States in 2013 was Roman Catholic (27 percent), followed by Baptists (12 percent), then Methodists (5 percent). However, taken in the aggregate, 41 percent of the adults surveyed were Protestant.
Attending worship services is another behavior that may be measured. In 2013, 39 percent of US adults attended a service during the last seven days. Since 1939, this figure has been very stable. Other measurable religious activities include reading the Bible and praying: 47 percent of U.S. adults have read the Bible in the last week and 75 percent pray daily.
Another approach would be to measure people’s level of acceptance or endorsement of specific church policies (or of government laws relating to religion). For example, 59 percent of American Catholics disagree with the official teachings of their church of not allowing divorced people to remarry in the church and 79 percent disagree with not permitting people to use artificial means of birth control, according to 2014 poll conducted by Univision. Polls have also discovered that about three-quarters of all Americans would accept teaching about world religions and the Bible (as literature and history) in the public schools.
Cognitive dimensions of religiousness include beliefs about God or a spiritual reality, as well as what people believe about religion. According to a 2014 Gallup poll, 86 percent of US adults believe in God or in a higher power. As of 2011, 85 percent believe in heaven and 75 percent believe in hell. In 2013, 56 percent believed that religion, rather than being out of date, could answer most of the day's problems; 21 percent believed that religion is increasing its influence on American life, while 76 percent believed it is losing influence. According to Gallup, in 2014, about 45 percent of adults surveyed had a "Great deal" or "Quite a lot" of confidence in the church and organized religion.
The affective components of religiosity deal with emotions, priorities, values, and evaluations. For example, in evaluating the overall priority they give to religion in their lives, 56 percent of Americans state that religion is “very important,” while only 22 percent say it is “not very important.”
In the 1960s, a dominant theme in research on religiosity emerged in response to the work of Gordon Allport, who had developed a model of religious orientation characterized by intrinsic and extrinsic dimensions. The former type of religiosity is interiorized, private, devotional, and based on individual commitment. Some items purporting to measure intrinsic religiosity have dealt with personal piety, church attendance, and the importance of religion. Extrinsic religiosity is more institutional, public, and pragmatic. Some of the items on scales designed to measure the extrinsic dimension include seeing religion as a vehicle for social relationships, consolation of grief, maintenance of order, and adherence to tradition. Many psychometric scales have been developed to assess orientation, an aspect of religiosity. Some measure orientation with a single intrinsic/extrinsic, bipolar scale, others with two distinct subscales. In contrast, in 1991, C. Daniel Batson and P. Schoenrade developed the Quest Scale to measure a third type. A quest orientation recognizes the positive value of doubt in the face of complex, existential questions regarding the meaning of life. What is particularly interesting is that Quest Scale scores tend to correlate negatively with prejudice measures.
Peter C. Hill and Ralph W. Hood, Jr., edited a much-needed compendium of scales titled Measures of Religiosity (1999). There are 126 scales grouped in seventeen chapters, thematically arranged. For example, there are measures of beliefs, attitudes, orientation, development, commitment, experience, values, and coping. Scales are presented along with information about the measured variable or dimension; scoring; psychometric properties, including evidence for reliability and validity; characteristics of people studied during the test development phase (norming samples); documentation regarding where the measure appeared in the research literature; and references for follow-up study. By identifying measures of similar constructs with different names or, conversely, measures of dissimilar constructs with the same name, Hill and Hood hope to promote a better understanding of the constructs measured. Richard Gorsuch also has suggested that construct and convergent validity could be improved if different researchers would use the same measures in well-developed, theoretically driven programs of research.
It is rare for scales used in the measurement of religiosity to meet well-known criteria for psychological tests. For example, few researchers publish standardized norms or even basic descriptive statistics for the samples used, such as measures of central tendency (means or medians) and variability ( standard deviations or ranges). While reliability may be good, information regarding validity is often inadequate. Validity is further compromised by the inherent sampling bias of many American religiosity scales, since most were developed using samples of convenience comprising US Protestants. Therefore, scales may be invalid for groups other than US Protestants. Traditional criteria and standards for scales may be found in texts such as Psychological Testing (1997) by Anne Anastasi and Susana Urbina or in the Standards for Educational and Psychological Testing (1999).
Religiosity measures may be classified as substantive scales or functional scales. If the former, the focus is on content; and if the latter, the focus is on process. For example, measures of religious beliefs tend to be substantive and are about what is believed, whereas measures of religious orientation tend to be functional and are variants or departures from Allport and Ross’s Religious Orientation Scale (1967).
Although social psychologists still dominate the field, developmental, cognitive, and evolutionary psychologists are contributing measures that correspond to their models of religious experience. While measures across these domains are primarily paper-and-pencil questionnaires, some are structured or semistructured interviews and some, as in the evolutionary approach, measure neurophysiological variables as indicators of religious or spiritual states.
Numerous applications of the various measures of religiosity are possible. Researchers can correlate any of these to other attitudes, personality traits, or demographic variables. Questions such as whether religious people are more superstitious, how religiosity differs between Democrats and Republicans, and whether religion helps people cope with marital problems can be addressed. For example, depending on how one decides to measure religion (and how one measures superstition), there is a slight tendency for more religious people to be a little less superstitious, but there are many people who are neither very religious nor very superstitious, and there are some who are both.
Using data from political polls, it can be verified that Jews, Catholics, and black Baptists tend to vote for Democrats, while most mainline white Protestant groups tend to vote for Republicans. Much of this correlation can be explained by historical and social-class features, however, in addition to the religious positions of the denominations.
Whether (and how) religion helps people cope with marital or other real-life problems is a difficult question to resolve. True experimentation, with random assignment of people to experimental and control conditions, would be necessary to confirm a cause-and-effect relationship. What seems to be apparent, however, is that religious people have a lower incidence of divorce and report slightly higher levels of marital satisfaction. This could be attributable to the fact that religious people feel more obligated to report that they have better marital relations, or it could be attributable to the fact that people who have problems in staying with a spouse also have problems in staying with their religion. When parents are asked whether religion has helped strengthen their family relationships, nearly four in five report that it has.
Clinical applications of the measurement of religiosity are also numerous. Very religiously committed individuals may have a problem with entering purely secular psychotherapy. The therapist may be seen as a nonbeliever who will challenge the patient’s worldview. Depending on the denomination, patients’ motivation for change may be tempered by the belief that their sufferings are a punishment inflicted by God.
From a more positive perspective, a patient’s religion can both serve as a source of impulse control (for example, as a check on suicidal tendencies) and provide a wide range of formal and informal social supports. For all these reasons, it is necessary for clinicians to assess the religiosity of their patients. Tolerant therapists can then use the patient’s worldview as a reference point. Therapists who cannot tolerate a given patient’s religiosity can make an appropriate referral (to another therapist who can) before the therapist is frustrated and the psychotherapeutic relationship has been damaged.
Other applications are possible in social and applied psychology. By understanding the religiosity of their “target segment,” for example, advertisers can tailor commercial and political messages to synchronize them with the values and worldviews of potential customers or voters.
In the late nineteenth century, psychologists turned to the field of religion and speculated about its origins and importance. William James was one of the foremost scholars of this period. His approach was chiefly that of the case study. The strength of his qualitative and narrative approach was that religion was embedded within the broader context of human life. The weakness was that two different investigators could look at the same religious person or phenomenon and come to very different conclusions, for there were no quantitative data.
The psychoanalyst Erik H. Erikson also employed the case study approach. For example, he wrote about the lives of Martin Luther and Mahatma Gandhi in terms of his eight-stage, epigenetic theory. Other theorists with a positive view of religion included neo-Freudians such as Carl Jung and Erich Fromm, and humanistic “third force” representatives such as Abraham Maslow. However, for these theorists, religion was secondary to their studies of personality, and their work was, therefore, seen by academic psychologists of religion as more philosophical than scientific.
For these reasons, the quantitative approach of the pollsters and psychologists dominates academic psychology. Yet doubt may be cast on the scientific status of this approach as well. A serious problem has to do with social desirability—people often respond in a way they think they should respond. Even though responses may be anonymous, people may (unconsciously or not) seek to portray themselves in a favorable light (in other words, as more religious than they really are). Another concern is the use of ambiguous terms in questions. Each denomination tends to define terms in its own way. What one denomination calls “services” might be called “worship” or “mass” by another. The Lord’s Supper is also known as Holy Communion and the Eucharist. Terms such as “God” and “personal commitment” may be so vague as to preclude construct validity.
The same item may have a different meaning for different people. For example, a person answering “rarely or never” to the question “How often do you ask God to forgive your sins?” may be an atheist who sees no purpose to the confession or a pious individual who rarely sins. Someone who disagrees with the statement “The word of God is revealed only in the Scriptures” may be an atheist or someone who believes in the possibility of present-day revelation. Certainly, national polls and measures of religiosity have ignored the importance of the context of the respondent’s denomination. Religiosity, as measured by the same scale, may mean one thing for an Orthodox Jew and another for a Jehovah’s Witness.
Many social scientists predicted the demise of religion during the twentieth century. Karl Marx believed that religion was the “opiate” of the people: a social institution used by the ruling classes to control and placate the exploited masses. After a proletarian revolution and the establishment of a just (Communist) social order, reasoned Marx, there would be no need for religion or, for that matter, the other instruments of state repression. Freud contended that, as psychoanalysis became more prevalent, people would turn away from religion; society would be composed of self-restrained individuals in control of their sexual and aggressive drives. The behaviorist B. F. Skinner regarded religious behavior as the result of accidental reinforcement, a superstitious approach to life that would diminish as humanity developed better technology for controlling the contingencies of its own reinforcement.
Instead, religion remains in myriad forms. For example, many intellectuals have moved away from institutionalized religion toward secular humanism. This requires those who study religion to rethink certain definitions, such as whether secular humanism can be defined as a religion and whether its religiosity can be measured. A more relevant question is not whether religion will continue to exist, but whether qualitative methods can attain the precision that science demands and whether quantitative methods can ever adequately measure the richness of human religious experience.
Allport, Gordon W. The Individual and His Religion: A Psychological Interpretation. 1950. Reprint. New York: Macmillan, 1970. Print.
American Educational Research Association, American Psychological Association, and National Council on Measurement in Education. Standards for Educational and Psychological Testing. Washington: American Psychological Association, 1999. Print.
Anastasi, Anne, and Susan Urbina. Psychological Testing. 7th ed. Upper Saddle River: Prentice Hall, 1997. Print.
Cook, Kaye V., et al. "The Complexity of Quest in Emerging Adults' Religiosity, Well-Being, and Identity." Journ. of the Scientific Study of Religion 53.1 (2014): 73–89. Print.
DeBono, Kenneth, and Anja Kuschpel. "Gender Differences in Religiosity: The Role of Self-Monitoring." North American Journ. of Psychology 16.2 (2014): 415–26. Print.
Erikson, Erik H. Gandhi’s Truth: On the Origin of Militant Nonviolence. New York: Norton, 1993. Print.
Erikson, Erik H. Young Man Luther: A Study in Psychoanalysis and History. New York: Norton, 1993. Print.
Gallup, George, Jr., and D. Michael Lindsay. Surveying the Religious Landscape: Trends in U.S. Beliefs. Harrisburg: Morehouse, 1999. Print.
Hill, Peter C., and Ralph W. Hood, Jr., eds. Measures of Religiosity. Birmingham,: Religious Education, 1999. Print.
Hood, Ralph W., Jr., Bernard Spilka, Bruce Hunsberger, and Richard Gorsuch. The Psychology of Religion: An Empirical Approach. 4th ed. New York: Guilford, 2009. Print.
Meadow, Mary Jo, and R. D. Kahoe. Psychology of Religion: Religion in Individual Lives. New York: Harper, 1984. Print.
Van Slyke, James A. The Cognitive Science of Religion. Burlington: Ashgate, 2011. Print.
Whitehouse, Harvey. Modes of Religiosity: A Cognitive Theory of Religious Transmission. Walnut Creek, Calif.: AltaMira, 2004. Print.
Wulff, David M. The Psychology of Religion: Classic and Contemporary. 2d ed. New York: Wiley, 1997. Print.