Professional Documents
Culture Documents
The Gerontologist 2014 Bradshaw Geront Gnu020
The Gerontologist 2014 Bradshaw Geront Gnu020
*Address correspondence to Matt Bradshaw, PhD, Department of Sociology, Baylor University, 316 Draper Academic
Building, One Bear Place #97326, Waco, TX 76798-7326. E-mail: drmattbradshaw@gmail.com
Received May 16 2013; Accepted February 24 2014.
Purpose of the Study:Research has linked several aspects of religionincluding service attendance, prayer, meditation, religious coping strategies, congregational support
systems, and relations with God, among otherswith positive mental health outcomes
among older U.S. adults. This study examines a neglected dimension of religious life:
listening to religious music.
Design and Methods: Two waves of nationally representative data on older U.S.adults
were analyzed (n=1,024).
Results: Findings suggest that the frequency of listening to religious music is associated with a decrease in death anxiety and increases in life satisfaction, self-esteem, and
a sense of control across the 2 waves of data. In addition, the frequency of listening to
gospel music (a specific type of religious music) is associated with a decrease in death
anxiety and an increase in a sense of control. These associations are similar for blacks
and whites, women and men, and low- and high-socioeconomic status individuals.
Implications: Religion is an important socioemotional resource that has been linked with
desirable mental health outcomes among older U.S. adults. This study shows that listening to religious music may promote psychological well-being in later life. Given that
religious music is available to most individualseven those with health problems or
physical limitations that might preclude participation in more formal aspects of religious
lifeit might be a valuable resource for promoting mental health later in the life course.
Key Words: Religion, Psychological well-being, Aging, Coping, Resilience
There is a complex relationship between age and psychological well-being. Some older adults report better mental health than their working-age counterparts, whereas
others experience declines in psychological well-being in
later life (Wu, Schimmele, & Chappell, 2012; Yang, 2007).
These disparate outcomes are the result of individual differences in (a) biological and environmental risk factors that
undermine mental health and (b) psychosocial resources
that either directly enhance psychological well-being or
buffer against risk factors (Blazer & Hybels, 2005). This
The Author 2014. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved.
For permissions, please e-mail: journals.permissions@oup.com.
Page 1 of 12
Background
Research has shown that listening to music is associated with lower levels of anxiety (Evans & Rubio, 1994),
greater satisfaction with life and feelings of escape from the
hardships of life (Hayes & Minichiello, 2005), lower levels
of anger and antisocial orientations (Baker & Bor, 2008;
Krahe & Bieneck, 2012), and reduced depressive symptoms
and other forms of psychopathology (Chan, Wong, Onishi,
& Thayala, 2011; Erkkila etal., 2011). Research on physiological processesincluding neuroimaging studies (e.g.,
functional magnetic resonance imaging and positron emission tomography)shows that listening to music is associated with the activation of pleasure and reward circuits in
the brain (Blood & Zatorre, 2001; Koelsch, Fritz, Yves v.
Cramon, Muller, & Friederici, 2006). Other research has
Page 2 of 12
Page 3 of 12
Hypotheses
HI: Listening to religious music will be associated with
improvements in mental health over time among older
U.S.adults.
Dependent Variables
Psychological distress at both baseline and follow-up was
measured with an index composed of the mean value from
the following eight items (Cronbachs =.869 at baseline
and .880 at follow-up), which were taken from the Center
for Epidemiologic Studies Depression scale (CES-D; see
Hertzog, Van Alstine, Usala, Hultsch, & Dixon, 1990;
Krause & Ellison, 2003): (1) I felt Icould not shake off
the blues even with the help of my family and friends. (2)
I felt depressed. (3) I had crying spells. (4) I felt sad.
(5) I did not feel like eating, my appetite was poor. (6) I
felt that everything Idid was an effort. (7) My sleep was
restless. and (8) I could not get going. Each item was
coded 1=rarely or none of the time to 4=most or all of the
time. These eight items were the only ones available in the
data that could be used to measure psychological distress.
Death anxiety was measured with a mean index
(Cronbachs = .849 at baseline and .895 at follow-up)
of four items taken from scales in the literature (Krause
& Ellison, 2003; Neimeyer, 1994): (1) I find it hard to
face up to the fact that I will die. (2) Thinking about
death makes me feel uneasy. (3) I do not feel prepared to
face my own death. and (4) I am disturbed by the shortness of life. Each item was coded 1=strongly disagree to
4=strongly agree. No other items on death anxiety were
available in thedata.
Life satisfaction was tapped with a mean index
(Cronbachs =.735 at baseline and .731 at follow-up) of
four items that were available at both waves of the data:
(1) These are the best years of my life. (2) As I look
back on my life, I am fairly well satisfied. (3) I would
not change the past even if Icould. and (4) Now please
think about your life as a whole. How satisfied are you
with it. The first three items are from the Life Satisfaction
Index A(Neugarten, Havighurst, & Tobin, 1961); response
categories ranged from 1=strongly disagree to 4=strongly
agree. The fourth item assessed satisfaction with life as a
whole, and response categories ranged from 1=not satisfied at all to 5 =completely satisfied. Because these three
items did not have identical response categories, they were
converted to equivalent z-scores prior to the construction
of the composite measure.
Self-esteem was measured with a three-item mean index
using questions developed by Rosenberg (1965). Specific
items included (1) I feel Iam a person of worth, or at least
on an equal plane with others. (2) I feel Ihave a number
of good qualities. and (3) I take a positive attitude toward
myself. Response categories ranged from 1=strongly disagree to 4=strongly agree. The Cronbachs for this index
was .903 at baseline and .904 at follow-up. These three
items were the only ones available in the data that could be
used to measure self-esteem.
A sense of control was measured with a mean index
(Cronbachs = .848 at baseline and .862 at follow-up)
composed of the following four items, which are similar to
those developed by Rotter (1966) and Mirowsky and Ross
(1991): (1) I have a lot of influence over most things that
happen in my life. (2) I can do just about anything Ireally
set my mind to. (3) When Imake plans, Im almost certain to make them work. and (4) When Iencounter problems, Idont give up until Isolve them. Each item is coded
1=strongly disagree to 4=strongly agree. No other measures of personal control were available in the data.
Independent Variables
Frequency of listening to religious music at baseline was
measured with the following question: How often do you
listen to religious music outside churchlike when you
are home or driving in your car? This variable was coded
1=never to 8=several times a day. In addition, a measure of frequency of listening to gospel music at baseline
was examined using the following item: Now Ihave some
questions about gospel music only. How often do you listen
to gospel music? This variable was also coded 1=never to
8=several times a day. Religious music is a broad category
that includes, but is not limited to, gospel music. It may
also include Gregorian chanting, contemporary Christian,
and even some forms of bluegrass music, among others. For
both of these variables, a series of categories was also constructed to check for nonlinear relationships with mental
health. No clear nonlinear patterns were observed, so each
was treated as a continuous variable in the models. Both
Page 4 of 12
Page 5 of 12
Covariates
InteractionTerms
To test for race, gender, and SES differences in the relationship between listening to religious music and mental health,
interaction terms were constructed for listening to religious
(and gospel) music times: (1) black, (2) women, (3) education, and (4) financial strain. To reduce multicollinearity,
all continuous variables were zero centered prior to constructing the interaction terms (Aiken & West, 1991), and
all interaction terms were entered into the models one at a
time. Variance Inflation Factor statistics indicated that multicollinearity was not a problem in the models.
MissingData
Missing data on all variables were dealt with using multiple
imputation techniques in SAS 9.3. The number of missing
cases was less than 5% for most measures. The results presented subsequently are based on five imputed data sets.
Analytic Strategy
Data analysis was conducted in three steps. First, descriptive statistics were calculated. Second, bivariate correlations between key dependent and independent variables
Results
Descriptive Statistics
Table 1 presents descriptive statistics. On scales of 14,
mean levels of the dependent variables for Waves 1 and 2,
respectively, were 1.565 and 1.465 for psychological distress, 2.042 and 1.994 for death anxiety, 2.891 and 2.830
for life satisfaction, 3.444 and 3.495 for self-esteem, and
3.020 and 2.988 for a sense of control. The differences
in means across the two waves of data were small for all
measures but were statistically significant at p < .05 or less
for each measure except for death anxiety. The sample was
equally split between blacks and whites. Nearly 62% of
the respondents were women, 47.3% were married, and
43.1% reported that their health was fair or poor (compared with good or excellent). The average respondent was
around 75 years old, with roughly 11.3 years of formal
education. Levels of financial strain, as measured at baseline, were relatively low (M=1.769 on a 14 scale). The
average respondent reported attending religious services
roughly two to three times a month and praying approximately daily. Of particular importance to this study, the
average respondent reported listening to religious music
outside of church about once a week (M=4.947). This variable had a range of 1 (never) to 8 (several times a day), and
its distribution had a slight negative skew. The frequency
of listening to gospel music was slightly lower, averaging a
few times a month (M=4.277), and it had a similar distribution to religiousmusic.
Ancillary analyses (not shown) revealed that listening to religious music was more common among blacks
than whites (M = 6.201 and 3.920, respectively), women
compared with men (M=5.215 and 4.643, respectively),
individuals with a below average level of education compared with their better-educated counterparts (M=5.663
and 4.580, respectively), and individuals who experienced
above average levels of financial hardship compared with
those who experienced below average levels (M = 5.631
and 4.547, respectively). All differences were statistically
significant at p < .05 or less. Differences in the frequency
of listening to gospel music by race, gender, and SES were
similar to the patterns reported for listening to religious
music in general.
Page 6 of 12
SD
Range
1.565/1.465b
2.042/1.994
2.891/2.830b
3.444/3.495b
3.020/2.988b
49
62
75.139
11.281
1.769
47
43
5.729
6.774
4.947
4.277
0.619/0.598
0.595/0.670
0.592/0.570
0.494/0.515
0.509/0.559
6.670
3.484
0.993
2.723
1.877
2.468
2.582
14
14
14
14
14
65101
125
14
19
18
18
18
Bivariate Correlations
Table2 shows bivariate correlations for key dependent and
independent variables. The correlations between baseline
and follow-up levels of the dependent variables were .265
for psychological distress, .256 for death anxiety, .290 for
life satisfaction, .193 for self-esteem, and .197 for a sense
of control (all are statistically significant at p < .001 or
less). The correlations between the frequency of listening
to religious music and four of the five mental health variables (death anxiety, life satisfaction, self-esteem, and sense
of control) were salutary and statistically significant, but
relatively weak, at both waves; correlations ranged from
.076 to .171 in magnitude. The correlations between listening to religious music and the remaining mental health outcomepsychological distressmeasured at baseline and at
follow-up were negligible and not statistically significant.
The findings for gospel music were similar with one exception: this variable was associated with psychological distress as well, especially at Wave 2.Overall, these bivariate
patterns suggested that listening to religious and/or gospel
music had a statistically significant, yet weak, relationship
with several different aspects of mental health in latelife.
Regression Analyses
Table 3 shows OLS parameter estimates from the regression of follow-up (Wave 2)mental health on baseline (Wave
1)mental health, covariates, and the frequency of listening to
religious music. These models tested H1, which stated that listening to religious music would be associated with decreases
Variable
1.000***
0.265***
0.131***
0.085***
0.298***
0.106***
0.186***
0.132***
0.348***
0.115***
0.015 ns
0.025+
1.000***
0.099***
0.185***
0.141***
0.276***
0.103***
0.205***
0.134***
0.202***
0.022 ns
0.042**
1.000***
0.256***
0.158***
0.127***
0.296***
0.113***
0.137***
0.027+
0.117***
0.071***
1.000***
0.096***
0.249***
0.077***
0.257***
0.034**
0.155***
0.024+
0.043**
1.000***
0.290***
0.343***
0.140***
0.437***
0.157***
0.137***
0.091***
1.000***
0.124***
0.377***
0.122***
0.377***
0.171***
0.184***
1.000***
0.193***
0.355***
0.160***
0.091***
0.066***
1.000***
0.137***
0.406***
0.119***
0.159***
10
1.000***
0.119***
0.139***
1.000***
0.197***
0.076***
0.073***
1.000***
0.742***
11
1.000***
12
Notes: 1=psychological distress T2; 2=psychological distress T1; 3=death anxiety T2; 4=death anxiety T1; 5=life satisfaction T2; 6=life satisfaction T1; 7=self-esteem T2; 8=self-esteem T1; 9=sense of control T2;
10=sense of control T1; 11=listening to religious music; 12=listening to Gospel music. n=1,024; ns = not statistically significant at p < .05.
***p < .001; **p < .01; *p < .05; +p < .10.
1
2
3
4
5
6
7
8
9
10
11
12
Page 7 of 12
The Gerontologist, 2014, Vol. 00, No. 00
Page 8 of 12
Table3. OLS Parameter Estimates (Standard Errors in Parentheses) From the Regression of Follow-Up (Wave 2)Mental
Health Variables on Baseline (Wave 1)Mental Health Variables, Covariates, and Frequency of Listening to Religious Music
Psychological distress
1.072 (0.078)***
0.209*** (0.033)
0.069 (0.044)
0.090* (0.040)
0.005+ (0.003)
0.013* (0.006)
0.041* (0.021)
0.034 (0.040)
0.114*** (0.039)
0.014+ (0.008)
0.011 (0.012)
0.000 (0.009)
0.101
Life satisfaction
0.070+ (0.041)
0.042 (0.038)
0.004 (0.003)
0.008 (0.006)
0.042* (0.019)
0.092** (0.037)
0.111*** (0.037)
0.008 (0.007)
0.008 (0.011)
0.019* (0.008)
0.116
Self-esteem
Sense of control
2.822*** (0.124)
0.168*** (0.033)
2.345*** (0.123)
0.186*** (0.036)
0.007 (0.039)
0.023 (0.035)
0.004+ (0.003)
0.008 (0.005)
0.041* (0.018)
0.032 (0.035)
0.051 (0.034)
0.000 (0.007)
0.019+ (0.011)
0.015* (0.007)
0.047 (0.042)
0.035 (0.038)
0.007** (0.003)
0.004 (0.006)
0.040* (0.019)
0.039 (0.037)
0.043 (0.037)
0.020** (0.007)
0.011 (0.012)
0.021** (0.008)
0.053
0.057
Notes: n=1,024.
***p < .001; **p < .01; *p < .05; +p < .10.
Table4. OLS Parameter Estimates (Standard Errors in Parentheses) From the Regression of Follow-Up (Wave 2)Mental
Health Variables on Baseline (Wave 1)Mental Health Variables, Covariates, and Frequency of Listening to Gospel Music
Psychological distress
Intercept
Mental health
variable T1
Black
Female
Age
Education
Financial strain
Married
Fair or poor health
Religious attendance
Prayer
Listening to
Gospel music
Adjusted R2
Death anxiety
Life satisfaction
Self-esteem
Sense of control
1.063*** (0.074)
0.209*** (0.032)
1.556*** (0.094)
0.258*** (0.034)
2.057*** (0.102)
0.231*** (0.031)
2.863*** (0.121)
0.167*** (0.033)
2.362*** (0.121)
0.186*** (0.036)
0.080 (0.046)
0.090* (0.040)
0.005+ (0.003)
0.015** (0.006)
0.036+ (0.020)
0.032 (0.039)
0.112*** (0.039)
0.015* (0.008)
0.011 (0.011)
0.004 (0.009)
0.001 (0.052)
0.005 (0.045)
0.011*** (0.003)
0.022*** (0.007)
0.019 (0.023)
0.014 (0.044)
0.057 (0.043)
0.004 (0.009)
0.029* (0.013)
0.022* (0.010)
0.093+ (0.044)
0.042 (0.038)
0.004 (0.003)
0.007 (0.006)
0.042* (0.019)
0.095** (0.037)
0.108*** (0.037)
0.008 (0.007)
0.015 (0.011)
0.003 (0.008)
0.016 (0.040)
0.022 (0.035)
0.004+ (0.003)
0.008 (0.005)
0.042* (0.018)
0.033 (0.035)
0.050 (0.034)
0.000 (0.007)
0.024+ (0.010)
0.008 (0.007)
0.058 (0.044)
0.034 (0.038)
0.007** (0.003)
0.005 (0.006)
0.043* (0.019)
0.039 (0.037)
0.043 (0.037)
0.019** (0.007)
0.011 (0.011)
0.021** (0.009)
0.096
0.089
0.110
0.053
0.068
Notes: n=1,024.
***p < .001; **p < .01; *p < .05; +p < .10.
Discussion
There is growing interest in the connection between religion and mental health (Smith et al., 2003). Scholars
Intercept
Mental health
variable T1
Black
Female
Age
Education
Financial strain
Married
Fair or poor health
Religious attendance
Prayer
Listening to religious
music
Adjusted R2
Death anxiety
Page 9 of 12
Acknowledgments
A previous version of this paper was presented at the 2012 Meeting
of the Association for the Sociology of Religion, Denver, CO. The lead
author would like to thank Baylor University for providing start-up
funds that contributed to this research. The authors would like to
thank Neal Krause and Interuniversity Consortium for Political and
Social Research (http://www.icpsr.umich.edu) for providing access to
the data used in this study.
References
Aiken, L., & West, S. G. (1991). Multiple regression: Testing and
interpreting interactions. Thousand Oaks, CA: Sage.
Aldridge, D. (1995). Spirituality, hope and music therapy in palliative care. The Arts in Psychotherapy, 22, 103109.
Bailey, L. M. (1984). The use of songs in music therapy with cancer
patients and their families. Music Therapy, 4, 517. doi:10.1093/
mt/4.1.5
Baker, F., & Bor, W. (2008). Can music preference indicate mental
health status in young people? Australasian Psychiatry, 16, 284
288. doi:10.1080/10398560701879589
Blazer, D. G., & Hybels, C. F. (2005). Origins of depression in
later life. Psychological Medicine, 35, 112. doi:10.1017/
S0033291705004411
Blood, A. J., & Zatorre, R. J. (2001). Intensely pleasurable responses
to music correlate with activity in brain regions implicated in
reward and emotion. Proceedings of the National Academy of
Sciences of the United States of America, 98, 1181811823.
doi:10.1073/pnas.191355898
which could have contributed to the improvements in psychological well-being observed in thedata.
To build on this study, future research should address the
limitations outlined earlier. In addition, investigators might
profitably examine the effects of listening to diverse types of
religious music. It is not clear whether certain specific genres
of religious music are more beneficial for mental health than
other types, due to tones, rhythms, or lyrical content. For
example, Gregorian chanting could promote serenity and
calm, set a somber spiritual mood, evoke a sense of awe and
majesty, or trigger other religious cognitions or emotional
responses that may influence the mental health outcomes
studied here. Other types of religious music, such as contemporary Christian, bluegrass, and upbeat Black gospel, tend
to embody certain characteristics and emphasize particular
lyrical themes that may strengthen faith commitments, reassure believers, and uplift listeners emotionally.
One recent study of thematic content of religious music
based on interviews with older African Americans provides a useful start on this topic. In this work, Hamilton
and colleagues (2013) identified several distinct types of
religious songs. Among the most common of these were
songs that focused on thanksgiving and praise, connection
with God, strength and resilience, and a sense of control.
Some religious songs were also instructive, emphasizing
a sense of obligation to help others. Hamilton and colleagues and their study participants also pointed to songs
that promoted prayerful communication with God, as well
as those that reminded the faithful of life everlasting. This
latter theme is especially noteworthy because previous
studies have associated belief in life after death with desirable mental health outcomes such as reduced anxiety and
depressed affect (Ellison, Burdette, & Hill, 2009; Flannelly,
Koenig, Ellison, Galek, & Krause, 2006). Although the
work of Hamilton and colleagues focused specifically on
older African Americans, many of these themes could resonate with other groups as well, and their possible distinctive benefits merit future investigation.
Future research should also seek to identify the mechanisms by which listening to religious music affects mental
health. For example, to the extent that listening to religious
music facilitates a secure attachment to a loving and supportive God, this could be an important mechanism by
which religious music affects mental health (Bradshaw
etal., 2010; Kirkpatrick, 2005). In addition, research suggests that music actually alters physiological processes in
health-promoting ways (Blood & Zatorre, 2001; Koelsch
et al., 2006). In particular, future research should certainly attempt to examine the role of neurotransmitters
such as dopamine and serotonin in mediating the relationship between listening to religious music and mental
health. Given the growth of non-Christian populations in
Page 10 of 12
Page 11 of 12
Bradshaw, M., & Ellison, C. G. (2010). Financial hardship and psychological distress: Exploring the buffering effects of religion.
Social Science & Medicine, 71, 196204. Retrieved from http://
dx.doi.org/10.1016/j.socscimed.2010.03.015
Bradshaw, M., Ellison, C. G., & Marcum, J. P. (2010). Attachment
to God, images of God, and psychological distress in a
nationwide sample of Presbyterians. The International
Journal for the Psychology of Religion, 20, 130147.
doi:10.1080/10508611003608049
Chafin, S., Roy, M., Gerin, W., & Christenfeld, N. (2004). Music can
facilitate blood pressure recovery from stress. British Journal of
Health Psychology, 9, 393403.
Chan, M. F., Wong, Z. Y., Onishi, H., & Thayala, N. V. (2011).
Effects of music on depression in older people: A randomised
controlled trial. Journal of Clinical Nursing, 21, 776783.
doi:10.1111/j.1365-2702.2011.03954.x
Chlan, L. (1998). Effectiveness of a music therapy intervention on
relaxation and anxiety for patients receiving ventilatory assistance. Heart & Lung, 27, 169176. Retrieved from http://
dx.doi.org/10.1016/S0147-9563(98)90004-8
Dillon, M., & Wink, P. (2007). In the course of a lifetime: Tracing
religious belief, practice, and change. Oakland, CA: University
of California Press.
Ellison, C. G. (1991). Religious involvement and subjective wellbeing. Journal of Health and Social Behavior, 32, 8099.
Ellison, C. G., Bradshaw, M., & Roberts, C. A. (2012). Spiritual
and religious identities predict the use of complementary and
alternative medicine among US adults. Preventive Medicine, 54,
912. doi:10.1016/j.ypmed.2011.08.029
Ellison, C. G., Burdette, A. M., & Hill, T. D. (2009). Blessed assurance:
Religion, anxiety, and tranquility among US adults. Social Science
Research, 38, 656667. doi:10.1016/j.ssresearch.2009.02.002
Erkkila, J., Punkanen, M., Fachner, J., Ala-Ruona, E., Pontio, I.,
Tervaniemi, M., , Gold, C. (2011). Individual music therapy
for depression: Randomised controlled trial. British Journal of
Psychiatry, 199, 132139. doi:10.1192/bjp.bp.110.085431
Evans, M. M., & Rubio, P. A. (1994). Music: Adiversionary therapy.
Todays OR Nurse, 16, 1722.
Flannelly, K. J., Koenig, H. G., Ellison, C. G., Galek, K., & Krause, N.
(2006). Belief in life after death and mental health: Findings from
a national survey. The Journal of Nervous and Mental Disease,
194, 524529. doi:10.1097/01.nmd.0000224876.63035.23
Frame, M. W., & Williams, C. B. (1996). Counseling African
Americans: Integrating spirituality in therapy. Counseling and
Values, 41, 1628. doi:10.1002/j.2161-007X.1996.tb00859.x
George, L. K. (1993). Sociological perspectives on life transitions.
Annual Review of Sociology, 19, 353373. doi:10.1146/
annurev.so.19.080193.002033
Hamilton, J. B., Sandelowski, M., Moore, A. D., Agarwal, M., &
Koenig, H. G. (2013). You need a song to bring you through:
The use of religious songs to manage stressful life events. The
Gerontologist, 53, 2638. doi:10.1093/geront/gns064
Hayes, T., & Minichiello, V. (2005). The contribution of music to quality of life in older people: An Australian qualitative study. Aging
& Society, 25, 261278. doi:10.1017/S0144686X04002946
Hertzog, C., Van Alstine, J., Usala, P. D., Hultsch, D. F., & Dixon, R.
(1990). Measurement properties of the Center for Epidemiologic
Studies Depression Scale (CES-D) in older populations. Psychological
Assessment, 2, 6472. doi.apa.org/journals/pas/2/1/64
Page 12 of 12