Psychometric details are best found in various original publications cited here. Conceptual underpinnings and psychometrics are described in the original paper on the scale [2]. A list of the items with the introduction is found in Table 1. One must keep in mind that individual psychometric results depend on the assumptions made and the populations and datasets on which they are based. Chronbach's alpha's for the scale in English and in translation, for both the 16-item version and the adapted 6-item version (more later on why the 6-item version is not recommended), have been consistently high, 0.89 and above [2,9-11], for example 0.94 and 0.95 for the General Social Survey samples [12]. Item distribution and skew was assessed as adequate, showing a wide range of variability in responses [2]. Test-retest results have been reliable, with test-retest Pearson correlation of 0.85 over two days [2]. Bailly and Roussiau [4] and Ng et al. [11] also show good test-retest reliability in translations of the scale. Both versions were stable over time, internally consistent, and the forms were equivalent and valid in an all-African American sample [13].

Although there can be trait-like consistencies over time, the scale also has state-like characteristics, lending it to examinations of change over time, as variability can be documented. For example, it was used in a daily diary study of pain quite effectively [14], and in a study of effects of meditation on pain tolerance [15]. It has been used to measure change over time with various interventions in healthcare, substance abuse treatment, education, and religious settings further described in the Results section [16-18].

Each item in the scale was designed to tap a separate feature of spiritual experience, but it was hoped that there would be a cohesiveness of conceptual underpinning. The original paper on the scale's psychometrics demonstrated a single factor, with weaker loadings by certain items, especially the two compassionate love items (13 and 14, self-less caring and accepting others) [2]. These two items often have a weaker loading, but conceptually are important to the breadth of content addressed in the scale.

Translations of the scale into Chinese, Spanish, French, Brazilian Portuguese and German have shown loading on a single factor. The Chinese translation showed a Chronbach's alpha of 0.97 using the single factor model that emerged through Principal Component Analysis [11]. The Spanish translation tested in the Mexican population had a Chronbach's alpha of 0.91, with a single factor solution, with items 2, 13 and 14 loading less strongly [10]. The German translation of the 16-item scale has shown a single factor structure, and supports the unidimensional structure of the scale and provides other psychometric support for its use in that population [19]. Confirmatory Factor Analysis shows a single factor in a European French population [9]. Kalkstein and Tower [20] show a single factor, even though there was weaker loading on the two compassionate love items (13 and 14). Robinson et al. 2007 found a one factor solution in an alcoholic population in treatment, with a 0.92 Chronbach's alpha [17].

Ellison and Fan, using the 2004 General Social Survey (GSS) results, examine both a one-dimensional and two-dimensional approach [12]. They divided the scale into items containing eight items specifically mentioning ‘God’ and a separate group, they called ‘Non-Theistic’ including the remaining eight ‘non-God’ items, and did the same for the 6-item scale in the 1998 GSS. They provide some support for a global single-factor model, while also supporting a two-factor model. A California sample of a substance abuse population showed a two factor structure [21]. The division was different than Ellison and Fan, and they called their item subgroups theistic (1,3,4,5,7,8,9,10,15) and self-transcendence (2,6,11,13,14), after eliminating items 12 (thankful for blessings) and 16 (close to God). It is expected that for some, the more explicitly theistic items will be responded to differently, but this does not necessarily make it a separate factor in the scale as applied more generically. The single factor solution is the dominant one at present.

Labels assigned to subgroups can unduly influence interpretations of results. Some items which might fall into the theistic grouping on face value might be able to be responded to positively by a non-theist. An example of this might be: “I feel God's love for me through others.” Given the introductory sentences for the scale (“A number of items use the word God. If this word is not a comfortable one for you, please substitute another idea which calls to mind the divine or holy for you.”) this item has been reported with high frequency by non-theists, as discovered in the initial qualitative developmental research on the scale. Another example of an item that cannot be clearly given a theist or non-theist distinction is: “I desire to be closer to God, or in union with the divine.” From the original development interviews, this item for example reflected spiritual experiences of an agnostic scientist, who would not be classified as a theist yet felt this item captured experience. “I feel a connection to all of life” is particularly salient for non-theists, however as environmental awareness increases, this may become increasingly relevant for theists. For example, connection with the natural world is strongly emphasized in the Franciscan Christian spiritual tradition. Although divisions of the scale on these lines may prove useful for some populations, the scale was designed based on research that demonstrated that these divisions are not as clear cut as we often assume. It is important not to create boundaries or categories that may not exist in the mind of the respondent. It is expected that cultures will also differ as to if, or how, this kind of division might be useful.

An analysis of the entire set of long form items included in the NIA/Fetzer Multidimensional Measurement booklet, included the 16-item DSES, and found that the DSES items form a useful and psychometrically valid single factor in that larger group of items [22]. Idler et al. found that the six DSE items constituted a single dimension when the entire BMMRS on the GSS was analyzed [5].

Continuing to consider the scale as uni-dimensional seems warranted at this time, taking these various factor analysis investigations into account. Since each item was designed to tap a dimension of spiritual experience, it is expected that items will perform differentially in various studies. The inclusion of the Introduction is important to enable effective factor analysis of the scale to take place. Premature separation of items into subgroups can actually get in the way of understanding.

It is tempting to tweak a scale. One reason is our own biases. We read an item and it does not seem to fit how we and those we know personally describe things, and then we make what seems like a sensible change. Our biases in the religious/spiritual realm can be quite strong and we can often be unaware of them. A few authors have changed the wording of the scale items substantively and still call their versions a scale of Daily Spiritual Experiences. This is strongly not advised. The wording of this scale is critical to its effectiveness and ability to tap the construct of interest. Changes in wording can strongly affect performance, as well as the capacity to compare with the many studies that have used the scale. If the wording is changed, the capacity to use the available norms is adversely affected, as is the ability to compare results between studies. The DSES does not meet all needs, and other scales should be used to address other needs.

The word “God” was consciously used in the DSES. For more detailed rationale see Underwood [1]. A recent study of 8–12 year-olds in Canada substituted “higher power” for God in the God items [23]. Although the study did show predictive value in the DSES items used, interviews by Underwood while developing the scale showed a lack of clarity with the phrase “higher power,” and this substitution is not recommended. Not all view God as either higher or a power. Some view God as a collaborator in life's decisions [24], or someone who “has your back” [25]. ‘God’ may be a contentious word for many, but the Introduction to the scale enables people to make their own substitutions as needed rather than the investigator deciding to edit it for them, perhaps erroneously. The scale's Introduction states: “A number of items use the word God. If this word is not a comfortable one for you, please substitute another idea which calls to mind the divine or holy for you.” Additionally, in the scale there are many non-God items that provide additional scope for responses. The DSES affirms the value of not watering down the concept of spirituality by discarding religious terms, but leaving space for spirituality that is religiously grounded and framed, as well as that which may not be.

Response categories for the scale are also important. The scale is a frequency scale for a reason. It encourages users to search their memory for specific experiences and attempt to measure frequency. A study using Rasch analysis and item response theory on the Portuguese version in Brazil found that overall the categories were appropriate and discriminating [26].

Sometimes adaptation to a particular population merits a specific change. Examples of appropriate adaptations are rare but include substituting “Allah” for “God” in the items for a study of Afghan refugees [27], and substituting “Creator” for “God” for use in a study of Zuni Pueblos an indigenous American population in an ongoing study by Kamilla Venner at the University of New Mexico. Because of the homogeneity of these groups and early testing of these forms to identify equilibration with the original concepts, the adaptation was appropriate in these cases. The scale has been republished in a number of venues, and sometimes the wording is not accurate. Sometimes tables used in various publications are inaccurate regarding item wording. It is important to refer to the Underwood peer-reviewed publications in this regard.

Details on scoring and analysis are found in the Underwood [1] paper and Underwood and Teresi [2]. Often more frequent daily spiritual experience is represented by lower crude totals or mean scores. An algorithm is used to create a more intuitively understood higher score for more experiences, which is helpful in describing correlations. Using “higher DSE's” rather than “DSES score” is a way around this problem in the text, and using the transformed scores in tables can also help here. Occasionally the term “higher DSES” is incorrectly used to describe more spiritual experiences, just assuming the transform. This can be somewhat confusing, so it is important to read the texts clearly.

Mean scores are usually the most useful for analysis, rather than total score. This also allows for examination of individual items or subgroups. Individual items that bear theoretical connection to the outcome of interest might a priori be hypothesized to have a connection in a particular study and could be examined in this way.

Another way of doing analysis is to dichotomize the responses. Where significant break points can best be drawn may differ from one study to another. It is possible that grouping “many times a day” and “every day” could provide a significant break point that has implications. As well, dividing into never or almost never, and all other frequencies of experience, may also yield significant results, depending on the research question of interest.

The DSES has been included in the General Social Survey (GSS). This allows for examination in relationship to a variety of other variables in this random sample public domain database [32,12]. Underwood [1] gives individual item breakdown of DSES item responses from this survey. For example, for “I feel God's presence,” 37% experience this many times a day to most days, with 22.4% never or almost never experiencing this. It is more common in the general U.S. population to feel God's love through others (40.7% many times a day to most days) than directly (35%), with 20.8% reporting never feeling God's love directly, and 15.0% reporting never feeling God's love though others. In response to “I am spiritually touched by the beauty of creation,” 25.7% reported this many times a day to most days, and 20.3% never or almost never. One of the least frequently reported items was “I feel thankful for my blessings,” with only 11.2% of the population reporting this many times a day to most days and 30.9% not at all. 17% express that they feel as close as possible to the divine, and 10.2% report feeling not at all close. This open-access dataset also provides the opportunity in the future for comparison with data from other countries.

The emerging data from studies of the DSES in translation provide opportunities to compare inter-culturally. For example individuals from Mexico show more frequent DSE's than in U.S. general population samples [10]. A study comparing responses from Mexico and Basque Europeans found some significant differences in responses on the DSES, and more DSE was correlated with better well-being and other variables in both populations [33]. Self-identified atheists in the Basque sample in this study reported quite high DSE's. It is important when doing cultural comparisons to eliminate confounding variables when possible. The DSES has been placed on a longitudinal study of African American health, and it has been validated in that population [34]. African Americans generally report higher DSE's. It has also been shown to be useful in various ongoing studies of Hispanic Americans, Hawaian Americans and Native Americans.

Women generally report higher DSE's than men in U.S. samples; however as studies using the translations of the scale emerge, the U.S. appears to be the exception regarding differences between men and women regarding DSES scores. The study of the DSES in Hong Kong showed no gender difference as well as no differences by age or educational level [11]. A study of European Basques and Mexicans showed no gender differences in either group [33]. The study of the DSES in a French population showed no gender differences in distribution of responses [9]. There is an emotional quality to a number of the items. It may be more acceptable for men to report emotions in some cultures than others. It is worth continuing to explore whether this also encourages males to be open to spiritual experiences more in some cultures than in others. It also is possible that some cultural differences relate to the predominant religious influences in a culture, regardless of affiliation.

There also may be some different ways in which DSES interacts with other parts of life for males and females. In one study of U.S. male college students, higher levels of DSE correlated with valuing influence and service in their career choices, whereas for females no significant relations existed among these constructs [35]. The GSS gender data showed differential outcomes for men and women regarding the effects of various religious variables and well being and health, finding that for women, higher DSE's were more predictive than religious attendance of well-being and health [36].

In a cross-sectional survey of more than 800 consecutively admitted patients at Duke University Medical Center, Koenig and colleagues found higher DSE's linked to fewer days of long-term care required [37]. They also found that spiritual experiences, prevalent in older hospitalized patients, are associated with greater social support, better psychological health, and better physical health [38].

In a longitudinal community-based study of 6,534 older adults, DSE's measured using five of the more explicitly theistic items were associated with better self-reported health [39]. Positive associations of DSE's with health were independent of public religious activity in the GSS [36]. Kalkstein and Tower [20] also showed positive correlation of DSE's with self-rated health.

DSE's, physical activity and healthy diet all contributed to higher physical well-being in a stress deterrent model, as counterbalancing stress in a study of healthy lifestyle behaviors in adults [40]. In arthritic older adults, DSE's were more common in those with arthritis, but not with other chronic diseases such as high blood pressure [41]. In a few studies no links between DSE's and specific health outcomes were found. This is not surprising. For some conditions, DSE's might form a part of dealing with the challenges of daily life, while not exerting a major influence overall on progression of or development of disease across the board. DSE's may weigh in on both sides of the equation for a variety of disease processes.

At least three studies of pain have used the DSES. A daily diary study showed DSES changes over time in rheumatoid arthritis patients, where frequent daily spiritual experiences linked with higher levels of positive mood, lower levels of daily negative mood, and higher levels of each of the social support domains [14]. It was not linked to experience of pain; however attention to pain has also been connected with diminished suffering [42]. Rippentrop [43], in 157 pain patients showed that DSE's predicted mental health status but not pain outcomes, and suggested that pain encourages engagement with DSE's. Wachholtz and Pargament used the DSES in a study of pain outcomes, where spiritual meditation groups, with higher post-intervention DSE scores, had increased pain tolerance [15].

Examining all the BMMRS subscales in a study of 168 assorted medical patients, it was found that only the DSES items significantly predicted SF-36 General Health Perception, after taking into account demographic and other health status variables. DSE items accounted for an additional 6% of the variance [44].

In a study of 157 Congestive Heart Failure patients, DSE's were not linked with depression; however they were linked with increased self-efficacy in dealing with one's illness [45]. Since the DSES has been included in a number of ongoing longitudinal databases there is future opportunity for secondary data analysis over time, not just with the health outcomes, but also with the other variables embedded in those databases. It has been included in various on-going health studies that often do not include spiritual or religious variables for a variety of reasons.

The Chinese translation was used in a study of 245 professional and support staff in a rehabilitation services complex in Hong Kong, where DSE's were negatively correlated with anxiety, depression and total burnout [11].

A study of secondary school teachers in Nigeria showed that higher DSE was independently correlated with less work family conflict [46]. DSE's in teachers in England were correlated with having increased inner resources and finding deeper meaning in their work [47].

Daily spiritual experiences mitigated physical, cognitive, and emotional forms of burnout for medical and mental health practitioners working in palliative care [48]. Physical fatigue, cognitive weariness and emotional exhaustion were the elements of burnout assessed in this study of professionals.

In a study of correlations with being employed in a public service occupation, the combined DSES item, “I feel God's love directly or through others,” was a predictor of working in public service compared with other occupations, even after controlling for religion, whereas a non-DSES general spirituality item was not [49]. High DSE's strongly predicted attitude towards and utilization of religion in social work practice, whereas organized religion measures were not predictive [50]. A study of Protestant ministers in Germany showed more DSE's correlated with diminished burnout [51].

In a study of how spouses help one another following trauma, partners who reported more emotional regulation and higher levels of daily spiritual experience provided more help to the victim of trauma. Although emotional regulation was initially a predictor, once other variables were controlled for, only DSES remained a significant predictor [52]. Examining the entire set of BMMRS dimensions (e.g., meaning, private practices, affiliation, religious coping), and reported marital happiness on the GSS, Bell found that of the BMMRS dimensions, a married man's daily spiritual experiences was the domain most strongly correlated with his self-reported marital happiness, followed by his organizational religiousness, and beliefs concerning forgiveness. Forgiveness was most strongly correlated with a married woman's self-reported marital happiness, followed by her daily spiritual experiences [53].

Brelsford et al. found that spiritual disclosure in 378 adults correlated with DSE's, which was also correlated with measures of generativity [54]. The two compassionate love items on the full DSES have been used in a number of studies [28,55]. They have been useful in studies of altruism and self-giving behaviors and attitudes, particularly when motivation is of interest. Lower levels of daily spiritual experiences correlated with having fewer close friends among a mixed sample of 409 adults and 84 older adults [20]. In a longitudinal study of older adults, positive association was found between social network integration and spiritual experiences in older adults [39].

In 85 individuals grieving the death of a significant person in their lives, those evidencing high DSE's showed lower levels of problematic grief affect. Church attendance influenced affect only to the extent that it was positively correlated with DSE's [56].

Adolescents are often neglected in religiousness/spirituality research. The DSES was originally developed by including interviews with adolescents. The scale has been validated in adolescents, where in an ethnically diverse sample of 12–18 year olds in New York City it has shown good internal consistency and good test-retest performance [57].

Van Dyke et al. found more frequent DSE's to be a significant predictor of higher psychological adjustment and lower psychological distress in a group of urban lower socioeconomic status adolescents [58]. They also found in this group that daily spiritual experiences are associated with positive affect for males and females, and a predictor of life satisfaction for adolescent males. A regression analysis of 615 adolescents showed that daily spiritual experiences, forgiveness and positive religious coping were associated with less depressive symptomatology exclusively in girls [59].

Park et al. studied 167 young adults following diagnosis with a variety of cancers to better understand what motivates the kinds of self care health behaviors particularly required in these young people. Results showed that religious attendance had little impact on health behaviors, while daily spiritual experiences were related to greater performance of health behaviors [60]. Self-assurance only partially mediated these effects.

Robinson et al. followed 123 alcoholics for six months following entry into treatment. Analysis showed that increases in DSE's (but not positive religious coping or forgiveness) were associated with increased odds of no heavy drinking at six months, even controlling for AA involvement [17].

In an analysis of 198 alcoholics using two subsets of DSES items, both subsets were correlated with longer sobriety and helping behaviors [21]. Lower alcohol intake has been associated with higher DSE's [2]. The DSES currently forms part of an ongoing study of adolescent alcohol abuse and helping behaviors. Shorkey and Uebel reviewed 10 measures for use in addiction research, and concluded regarding the DSES that “the usefulness of this scale for assessing the spiritual experience of a person in treatment and recovery may be profound” [61].

The GSS data shows a robust positive association between DSES and happiness, excitement with life, self-esteem, and optimism. By contrast, DSES demonstrated little relationship to negative affect in this dataset. DSES effects are present no matter what the gender, age, socioeconomic status, race [12].

Bailly and Roussiau, using a DSES translation in 165 French adults age 65 and older, showed higher scores associated with higher life satisfaction and self-reported good health, but no relationship with depressive symptoms [9].

In a study of male prison inmates, increased DSE was linked to better emotional health [62]. Ciarrochi and Deneke showed relationships with positive emotion and cognitive well-being above and beyond predictions from personality variables, public religious factors, gender and age [63,64]. Data from a variety of sources showed DSE was negatively associated with state-trait anxiety, and perceived stress and positively correlated with optimism, perceived social support and quality of life [2]. Moderate links have been shown between higher DSE's and greater well-being in widows and widowers [65]. In inner city elders, having better mental health was correlated with higher DSE [66].

In groups of European Basques and Mexicans, Mayoral et al. found that DSE was positively correlated with life satisfaction, positive affect, and psychological well-being, and negatively correlated with indicators of anxiety and depression. They found no correlations with the Big Five personality factors [33].

DSE's are not uniformly associated with depression across the board. The scale was designed such that one could respond positively to the items even when depressed or experiencing negative affect. However, an inverse relationship of DSE's with depression was found in some groups. In adolescents DSE predicted less depression [59]. The DSES was put on a National Institutes of Health longitudinal study of post heart attack outcomes. No link was found to survival or re-infarction, however higher DSE correlated with less depression [67]. In a group of arthritis patients, higher DSE was correlated with more energy and less depression [41]. Mofidi et al. found a relationship of more DSE's to less depression in 630 adults, an association moderated by age and stress [68]. Suggested mediators were optimism, volunteering and social support. In a cross-sectional survey of more than 800 consecutively admitted patients at Duke University Medical Center, Koenig and colleagues found significant positive associations between the DSES and social support and levels of cooperativeness, and inverse associations between the DSES and depressive symptoms [37]. Skarupski et al. found an inverse association between frequency of spiritual experiences and depressive symptoms in older adults [39].

In 497 adults, lower levels of spiritual experience were associated with higher levels of depression, anxiety, loneliness, and fewer close friendships, controlling for other religion, age, gender and education [20]. In survivors of domestic violence, higher DSES predicted fewer depressive symptoms and higher levels of social support [69]. The positive relationship between loneliness and depression, significantly associated among homebound older adults, was weaker among the older persons who reported higher scores levels of DSE [70].

When people have been interested in examining change in spirituality over time, they have often used the DSES. It has been used to measure change following various interventions in healthcare, substance abuse treatment, education, and religious settings [16-18].

A control group study of the effects of two 1-day spiritual retreats on nurses' spirituality in a group of 199 critical care nurses, showed higher DSE's at the end of the retreat and six months post retreat [71]. In another study, the investigation of the efficacy of a self study program designed to teach nurses how to talk with patients about spirituality also showed changes in DSES scores using pre and post-test evaluations. In comparing nurses in the field with student nurses, the DSE scores of the students increased more than the practicing nurses over the course of the intervention [16].

A study looking at the effects of face-to-face prayer for patients with depression and anxiety showed that DSES scores in patients remained higher a month following the intervention [72]. A Canadian holistic medicine outcomes database has listed the DSES as one possible outcome measure [73].

A controlled trial of spiritual direction in a substance abuse population did not show an effect of the intervention on substance intake, as both groups improved over time. However the DSE's increased somewhat over the course of the intervention, and remained so six months later, whereas psychological variables showed less improvement over the course of the intervention compared with treatment as usual [74]. An online program intended to cultivate sacred moments used the DSES as the outcome measure for the 83 participants [18]. The control group wrote daily about their lives. Both groups showed decreases in perceived stress and increases in DSES over time and at six week follow-up.

The DSES is often used when others are developing scales of religious/spiritual variables to be used in specific contexts. For example, Cole et al. developed a spiritual transformation scale to be used in cancer patients, and found their scale highly correlated with the DSES [75]. Spiritual growth correlated positively with DSES and spiritual decline varied inversely. Hall, when developing a scale of spiritual maturity, in the Episcopal faith tradition, found higher scores on certain DSES items in “well-formed” Episcopalians [76].

Wachholtz and Pargament used the DSES as an outcome measure in their study of secular versus spiritual meditation. Spiritual experiences increased over time, more so in the spiritual than secular meditation group, although both groups showed some increase in DSE's [15].