Gender Differences in Religious Beliefs

Recent research indicates that women in the general population are more religious than men. This study was developed to determine if this trend holds specifically for college students. The study tested the hypothesis that women at a satellite campus of a large Midwestern university display higher levels of religiosity than the men there. Forty-seven college students were given a survey asking them questions about their religious beliefs. Respondents were asked to rate their own religiosity and spirituality, to describe their level of religious conviction, and to state whether or not they believe in God, an afterlife, traditional religious paranormal phenomena such as angels and ghosts, the power of prayer, and the meaningfulness of life. The results of the survey showed that college women do appear to manifest more religiosity than college men, both overall and on specific measures of religiosity. This could be due to a predisposed tendency for women to be more religious or perhaps instead to the traditional lack of women's roles in the workplace giving them more time to devote to religiosity.

Religion is one of the most salient factors in many people's lives, and literally hundreds of distinct sets of belief have emerged over the centuries (Jensen, Jensen, & Wiederhold, 1993; Duncan, Donnelly, & Nicholson, 1992; Maltby, 1995; et al.). Culture, upbringing, age, and numerous other factors contribute heavily to which belief systems a person is likely to adopt as their own (Levitt, 1995; Schoenfeld & Mestrovic, 1991; Harley & Firebaugh, 1993). Thus it is no surprise that gender tends to be a significant contributor to a person's religiosity (Das & Harries, 1996; Johnson, George, & Saine, 1993; Levitt, 1995; et al.). Recent research seems to declare with an almost unanimous voice that women are more religious than men.

In a recent survey, significantly more women than men indicated that religion was "very important" to them (Johnson, George, & Saine, 1993). They demonstrate more religiosity, report having had more religious experiences, and tend to have a more positive attitude toward religion than men do (Davis et al., 1990, cited in Levitt, 1995; Harrison, 1983, cited in Levitt, 1995). Also, numerous studies have shown that women "are more likely to express a greater interest in religion..., have stronger personal religious commitment..., and attend church more frequently" than do men (Miller & Hoffmann, 1995, p. 63). The 1990 General Social Survey showed that more women than men tend to read the Bible and pray regularly (Miller & Hoffmann, 1995), and in Great Britain there has been a noticeable decline in the number of male churchgoers in the Anglican Church in recent years (Levitt, 1995). Additionally, according to a recent validating study of Fowler's (1981, cited in Das & Harries, 1996) Theory of Faith Development, the majority of women between the ages of 21 and 30 years are at a higher stage of faith development than the majority of men.

Numerous theories have been proposed to explain why this trend holds so universally. One theory points out that religion is "a tender-minded attitude" (Maltby, 1995, p. 529); the traits of passiveness, obedience, and caring are correlated with higher levels of religiosity (Miller & Hoffmann, 1995). These traits are also frequently associated with women, so perhaps certain aspects of the characteristically feminine personality predispose women to be more religious than men. Another theory relates to the fact that, in the past, women have comprised a smaller proportion of the workforce than men traditionally have. Because they had fewer roles in the workplace, they had more time to devote to religion and thus became more religious (Miller & Hoffmann, 1995). For some reason, though, women seem to display higher levels of religiosity than men.

Comparatively little of this prior research has been done on college students to determine if this trend holds specifically for them. For that reason, this study was developed – to determine what gender differences, if any, were present among college students in a satellite campus of a large Midwestern university. The study tested the hypothesis that women at this university would display higher levels of religiosity than the men there, thus further strengthening the idea that women are more religious than men.

The participants in this study were all college students enrolled at a satellite campus of a large Midwestern university. We set up a table in a heavily traveled area of campus on two separate weekdays and asked students who passed by to fill out a survey. A total of 47 people completed the survey, 23 men (48.9%) and 24 women (51.1%). This did not match the male-to-female ratio at this particular campus (29 males : 71 females). 23 of those students were traditional-aged college students (age 18-24), while 24 of those students were nontraditional-aged (25 or older). This did closely match the traditional-to-nontraditional age ratio at this campus (51 traditional students : 49 non-traditional students).

The survey was a written questionnaire consisting of 14 questions: ten multiple choice, two multiple choice with free response components, and two fill in the blank (see Appendix I). Four of the questions (Questions 11, 12, 13, and 14) were demographic questions to determine participant religion, doctrine (denomination), age, and gender. Questions 1 and 2 asked the participants to evaluate their own levels of religiosity and spirituality, respectively. The remaining questions were designed to measure six specific aspects of religiosity: belief in God (questions 3 and 4), belief in an afterlife (questions 5, 5a, 5b, and 6), belief in traditional religious paranormal phenomena (e.g. angels, spirits; question 7), belief in the effectiveness of prayer (question 8), belief in the meaningfulness of life (questions 10 and 10a), and strength of religious beliefs (question 9).

Before being given out to the participants, the survey was first pilot-tested among a small group of students at the university who were currently enrolled in a psychology research methods course. Minor changes were made in the wording of questions 1, 5a, 5b, and 8, while a definition of "effective" was added to question 9 and two new answer options were added to question 10.

Students who passed by the table we set up on the two days we were present were asked to fill out a written survey on "religious beliefs." The students were not told that we were looking for gender differences, though they were all asked to report their gender in the survey itself. Informed consent documents were given out for students to sign at the same time as the surveys themselves were distributed, and students completed the consent documents and the surveys at or near the table where we were set up. Immediately upon their completion, we collected both the surveys and the consent documents from the students and placed them in separate piles. To maintain confidentiality, the students were only asked to place identification (their names) on the consent documents; no identification was requested on the surveys themselves.

Using the survey data, a numerical value representing level of religiosity was assigned to each participant. These values were derived from assigning numerical values to each answer response on the survey, for all questions excluding numbers 5b, 11, 12, and 14. On questions 1 and 2, responses of "very religious" or "very spiritual" were assigned 3 points each, while responses of "moderately religious" or "moderately spiritual" were assigned 2 points each. No other responses to those questions were awarded any points. On questions 3, 6, 7, and 13, no points were awarded for either a response of "none" or a response of "other" if other was marked as "don't know" or "undecided," but for any other response, one point was awarded, with a maximum of one point awarded for each question, regardless of how many responses were marked. For questions 4 and 8, one point was awarded for an answer of "Yes," while no points were awarded for responses of "No" or "Don't know." For the question sequence of 5, 5a, and 5b, one point was awarded for an answer of "Yes" to both 5 and 5a and any response for 5b, while any other responses in 5 and 5a earned no points. In question 9, one point was awarded for a response of "I hold fairly strongly to my religious beliefs...," two points for "I hold very strongly to my religious beliefs...," and three points for "I would never change my mind" – unless the respondent indicated by his or her responses to the other questions that he or she was not religious, in which case no points were awarded. No other responses for question 9 earned any points. For question 10, one point was awarded for either a response of "Yes, and I know what it is" or one of "Yes, but I don't know what it is," while no points were assigned for any other responses. Finally, in question 10a, if a response made a clear reference to a religious term found elsewhere in the survey (e.g. "God" or "Christ"), it was awarded one point; all other responses received no points. These points were then totaled to arrive at each participant's overall religiosity score. These scores ranged from 0 to 18, with 0 being not religious and 18 being very religious.

The results show that women overall scored higher than men on this general religiosity scale (see Table 1). The mean score for men was 10.53 (SD = 5.26), while the mean score for women was 14.43 (SD = 2.44). A one-tailed t-test showed this difference to be significant (t = -2.98, α = 0.01, df = 38).

The results showed similar trends on a number of individual measures of religiosity. When asked to self-report their own religiosity, men's scores varied widely from "very religious" to "not at all religious," while all the women but one rated themselves as being either "moderately religious" or "very religious" (see Table 2). These differences were significant (t = -3.58 at α = 0.01, df = 44). So were the differences on the individual measure of self-reported spirituality (t = -3.63 at α = 0.01, df = 45), in which the majority of women (62.5%) marked themselves as being "very spiritual," while the plurality of men (43.5%) fell into the category of being "moderately spiritual" (see Table 3). There was not a significant difference in the self-assessed strength of religious beliefs (t = -1.43 at α = 0.01, df = 41; see Table 4) or in belief in the meaningfulness of life (χ² = 4.58 at α = 0.01, df = 3; see Table 5). In religious affiliation, there was a just slightly significant gender difference (χ² = 12.74 at α = 0.05, df = 4), with 91.7% of women labeling themselves as belonging to a religion as opposed to only 68.2% of men (see Table 6). However, there was no significant gender difference overall in other beliefs commonly associated with religion, such as God, the afterlife, and in traditional religious paranormal phenomena like angels and ghosts (at α = 0.01: χ² = 6.43, df = 5, for belief in God; χ² = 6.45, df = 5, for belief in the afterlife; and χ² = 9.98, df = 9, for belief in traditional religious paranormal phenomena; see Table 7).

The survey results indicate that at the satellite campus of the large Midwestern university being tested, women are more religious than men. Overall and on two self-report measures of religiosity, though not on other individual measures, the women scored significantly higher than the men did. This adds tentative support to the original project hypothesis, that the female college students would display higher levels of religiosity than the male students. It also agrees with the prior research showing that women are the more religious sex (e.g. Das & Harries, 1996).

These results could possibly shed new light on some of the theories which have been proposed to explain the observed gender differences in religiosity. The study seems to shed doubt on one of these proposed theories while supporting another. One theory, that women are more religious because they have historically spent little time in the workforce and thus have had more time to devote to religious activities (Miller & Hoffmann, 1995), becomes more dubious when the results of the study are examined. Working women presumably have no more time to devote to religious activities than do working men, so if the theory is true, then working women should not be any more religious than the men. The same is likely to be true for college students. Although there are numerous extraneous variables not specifically dealt with in this study which could confound this, college women are not likely to have any more time to devote to religious activities than college men; thus college women should not be any more religious than college men. The college women examined in this sample are more religious than the examined college men, however. Thus this social learning theory, if applicable to college students as well as to the workforce, is not supported by the evidence.

Since social roles such as enrollment in college may be insufficient to explain these gender differences in religiosity, this lends support to the theory that these gender differences are caused instead by the different character traits of men and women (Miller & Hoffmann, 1995). Women simply may be predisposed by certain aspects of the characteristically feminine persona to be more religious people. Such a theory predicts that no matter what the environment, women will be more religious than men. The results of this study support this idea.

These conclusions are tentative, however; some caution must be used in generalizing the sample data gathered in this study to the entire population of this satellite campus. Because this was a pilot study containing only 49 respondents, it is a nonrepresentative sample which may not be indicative of the entire population at this or any other college campus. Also, a total of nine different responses to various questions had to be thrown out because the participants either did not complete the items or wrote in answers of their own, indicating that some of the questions may not have been clear to all of the respondents. This was controlled for as much as possible in the construction of the survey; however, due to the subjective, metaphysical nature of religiosity, the results of the survey may not be an accurate representation of the respondents' religious beliefs. Additionally, extreme caution must be used when applying the results to support or refute one of the aforementioned theories. Since this study was designed simply to evaluate whether or not gender differences exist among college students rather than to evaluate one of these theories, significant extraneous variables which might have influenced the outcome of the study have not been accounted for, and these variables may confound the results when they are applied to one of the theoretical models.

Therefore, more research should be conducted to determine the validity of this evidence. Further studies using representative samples of student populations should be conducted at other universities to determine if indeed college women are more religious than college men.

The author of this study would like to acknowledge Kevin Howell, Danielle Sanders, and Beth Winer for their assistance in writing and administering the survey used in this project.

Das, A. & Harries, B. (1996). Validating Fowler's theory of faith development with college students. Psychological Reports, 78, 675-79.

Duncan, D. F., Donnelly, J. W., & Nicholson, T. (1992). Belief in the paranormal and religious belief among American college students. Psychological Reports, 70, 15-18.

Harley, B. & Firebaugh, G. (1993). Americans' belief in an afterlife: Trends over the past two decades. Journal for the Scientific Study of Religion, 32, 269-78.

Jensen, L. C., Jensen, J., & Wiederhold, T. (1993). Religiosity, denomination, and mental health among young men and women. Psychological Reports, 72, 1157-58.

Johnson, R. W., George, D. T., & Saine, K. C. (1993). The Christian Orthodoxy Scale: A validity study. Psychological Reports, 72, 537-38.

Levitt, M. (1995). Sexual identity and religious socialization. British Journal of Sociology, 46, 529-36.

Maltby, J. (1995). Personality, prayer, and church attendance among U.S. female adults. The Journal of Social Psychology, 135, 529-31.

Miller, A. S. & Hoffmann, J. P. (1995). Risk and religion: An explanation of gender differences in religiosity. Journal for the Scientific Study of Religion, 34, 63-75.

Schoenfeld, E. & Mestrovic, S. G. (1991). With justice and mercy: Instrumental-masculine and expressive-feminine elements in religion. Journal for the Scientific Study of Religion, 30, 363-80.

OVERALL RELIGIOSITY SCORE	Number of Males	Number of Females
17 – 18	2	7
15 – 16	1	3
13 – 14	5	7
11 – 12	5	2
9 –10	2	2
7 – 8	0	0
5 – 6	0	0
3 – 4	1	0
1 – 2	2	0
0	1	0

* Excludes the scores of seven participants – six for not following directions and one for not taking the survey seriously (for the response to question 4, which asked the participants to name something they considered to be holy, this respondent wrote in "House of Bob").

RESPONDENT CONSIDERS SELF TO BE:	Percentage of Males	Percentage of Females
Very Religious	9.1%	33.3%
Moderately Religious	40.9%	62.5%
Not Very Religious	31.8%	0.0%
Not At All Religious	18.2%	4.2%

* Excludes 1 respondent, who did not respond to the survey question evaluating this criterion.

RESPONDENT CONSIDERS SELF TO BE:	Percentage of Males	Percentage of Females
Very Spiritual	21.7%	62.5%
Moderately Spiritual	43.5%	33.3%
Not Very Spiritual	26.1%	4.2%
Not At All Spiritual	8.7%	0.0%

RESPONDENT CONSIDERS OWN RELIGIOUS CONVICTIONS TO BE:	Percentage of Males	Percentage of Females
Not Very Strong	4.6%	0.0%
Fairly Strong	45.5%	33.3%
Very Strong	22.7%	41.7%
Extremely Strong	13.6%	25.0%

* Excludes 1 respondent, who did not respond to the survey question evaluating this criterion.

RESPONDENT'S BELIEF IN THE ULTIMATE MEANINGFULNESS OF LIFE	Percentage of Males	Percentage of Females
Believes Life Has a Meaning, and Knows What it Is	43.5%	50.0%
Believes Life Has a Meaning, but does Not Know What it Is	39.1%	50.0%
Believes Life Has No Meaning	4.4%	0.0%
Does Not Know if Life Has a Meaning	13.0%	0.0%

RELIGIOUS AFFILIATION	Percentage of Males***	Percentage of Females
Buddhism	4.6%	0.0%
Christianity (Protestantism) (Catholicism) (Other)**	63.6% (57.1%) (21.4%) (21.5%)	91.7% (81.8%) (4.6%) (13.6%)
Taoism	4.6%	0.0%
Agnosticism, Atheism, Confucianism, Hinduism, Islam, Judaism, and Shintoism	0.0%	0.0%
None	31.8%	0.0%
Other	0.0%	8.3%

* Excludes 1 respondent, who did not respond to the survey question evaluating this criterion.

*** Totals more than 100% because 1 male respondent marked belonging to more than one religion.

N = 23 males, 24 females; χ² (God) = 6.43, χ² (afterlife) = 6.45, χ² (traditional religious paranormal phenomena) = 9.98

CONCEPTS THAT RESPONDENTS INDICATED BELIEVING IN	Percentage of Males Believing (Does Not Total 100%)	Percentage of Females Believing (Does Not Total 100%)
God	65.2%	95.8%
Om	4.4%	0.0%
Tao	8.7%	0.0%
Gods	4.4%	0.0%
Higher Power that Influences Life	8.7%	12.5%
Universal Energy/Force	17.4%	8.3%
Heaven	65.2%	95.8%
Hell	43.5%	79.2%
Purgatory	0.0%	16.7%
Reincarnation	8.7%	12.5%
Nirvana	4.4%	4.2%
Moksha	0.0%	0.0%
Enlightenment	17.4%	4.2%
Angels	60.9%	91.7%

Demons	39.1%	54.2%
Ghosts	34.8%	37.5%
Spirits	47.8%	54.2%
Prophets sent by God	39.1%	50.0%
Jesus Christ	60.9%	95.8%
Gotama Buddha	13.0%	0.0%
Mohammed	17.4%	0.0%
Mary	47.8%	54.2%
The Devil	47.8%	62.5%

Thank you for your participation in this survey. Your results will remain completely anonymous, but they will give researchers valuable information about the religious beliefs that are present in the IUK student population.

1. “Religious” is generally considered to be involvement with religious institutions such as churches and temples and/or believing in their ways of practice. Do you consider yourself to be:

2. “Spiritual” is generally considered to be having a personal, intimate relationship with the divine. Do you consider yourself to be:

3. Which of the following divine concepts, if any, do you believe in? (Check all that apply)

6. Which of the following perceptions of the afterlife do you believe in? (Check all that apply)

8. Prayer is generally considered to be an appeal to or communication with the divine. Do you believe that prayer is effective (effective means you get what you want/pray for)?

__ I do not hold strongly to my religious beliefs, I am very likely to change my mind.

__ I hold fairly strongly to my religious beliefs, I am likely to change my mind if a good reason arises.

10. a.) If yes, what is it? ____________________________________________________ _____________________________________________________________________________

14. What doctrine, if any, do you belong to (e.g. Baptist, Catholic, Orthodox Jew, etc.)?