Keywords

Whereas the psychology of religion/spirituality (R/S) has historically focused both on the positive and negative outcomes of religious beliefs and institutions relevant to the divine or sacred (e.g., Hill et al., 2000), positive psychology has focused more specifically on topics related to flourishing, including well-being and resilience (Snyder et al., 2021). Despite differences in topics of study, these fields share commonalities in their research methods and methodological diversity (or lack thereof; see Table 8.1). Both fields largely emphasize quantitative, cross-sectional data, providing an efficient but incomplete picture of very complex phenomena. In this chapter, we describe various measures and research designs used in the psychological study of R/S and positive psychology, and we outline the strengths and weaknesses of each. We also discuss emerging research and data collection issues (e.g., open science practices and cross-cultural concerns), and we suggest ways the fields of positive psychology and the psychology of R/S can increase methodological rigor and diversity. Throughout this chapter, we use specific research examples in positive psychology and psychology of R/S, but in no way is our brief review meant to be exhaustive.

Table 8.1 Methodology of Empirical Articles on Positive Psychology (PP) and the Psychology of Religion and Spirituality (PoRS), 2000–2020
Full size table

What Kinds of Measures?

The measures that researchers use have important and often unexamined effects on the conclusions researchers can draw. Asking individuals if they are spiritual or forgiving, for example, has the potential to elicit very different answers compared to observing individuals’ spiritual or forgiving behavior. Below we summarize some strengths and limitations of measures that are commonly used in the fields of positive psychology and the psychology of R/S.

Self-Reports

Both fields rely heavily on self-report measures (Yaden et al., 2018; see Hill et al., Chap. 7, this volume, for a review), which enable the efficient measurement of multiple dimensions of complex constructs. However, self-reports can be subject to self-presentation bias, especially when measuring socially desirable constructs (Tourangeau & Yan, 2007) such as religiousness (Presser & Stinson, 1998) and well-being (Heintzelman et al., 2015). Several reviews have noted positive relationships between social desirability and individual differences in religiousness (e.g., Gebauer et al., 2017), leading some to question whether religious people only present themselves as more altruistic (Galen, 2012) and less prejudiced (Batson et al., 1993). Self-reports might be particularly problematic for certain concepts such as humility, given that “self-reports of high levels of humility may ironically indicate a lack of humility” (Davis et al., 2010, p. 246). Because of a heavy reliance on self-reports, these two fields may actually be studying how religious and happy people think they should be, rather than how religious and happy people truly are.

One solution is to control statistically for social desirability (e.g., Saroglou et al., 2005, Study 4). However, it is difficult to distinguish between self-presentation and actual high levels of desirable behaviors (e.g., Heintzelman et al., 2015), so partialling out social desirability may remove legitimate variance from other measures. Fortunately, another solution lies in the use of informant reports.

Informant Reports

Friends, family members, and acquaintances of religious individuals report that those individuals tend to display more empathy (Łowicki & Zajenkowski, 2019), agreeableness (McCullough et al., 2003), and prosociality (Saroglou et al., 2005) than their less-religious peers. Informant reports have also been used for constructs such as patience (Shubert et al., 2020) and humility (Davis et al., 2010). Not only might informant reports circumvent participants’ social desirability biases, but discrepancies between self- and informant-reports themselves can predict variance in well-being (Shubert et al., 2020).

Although useful, informant-reports are also limited. Informant- and self-reports are based on different information (Vazire, 2010). For instance, Meagher et al. (2020) noted that informant ratings of humility were based on observable conciliatory behaviors, whereas self-ratings of humility were likely based more on knowledge about one’s own thoughts. These types of discrepancies could lead to different behavioral correlates of peer- and self-rated humility. This limitation may also apply to other constructs with cognitive and relational components, such as religiousness, self-regulation (Hardy et al., 2020), or forgiveness. Additionally, Galen et al. (2014) argued that informant reports of relationships between religiousness and prosociality were based in part on stereotypes that religious individuals were more helpful. Vazire’s (2010) self–other knowledge asymmetry model incorporates these and other explanations for degrees of self–other rating (dis)agreement. In contrast, Helzer et al. (2014) ruled out the heuristic confound in informant reports of moral character, and they found results suggesting that informants were less subject to presentation biases.

Behavioral Measures

Behavioral measures can address social desirability by providing covert and psychologically costly measures of relevant constructs (Batson et al., 1993). Yet behavioral measures are quite rare in positive psychology and psychology of R/S. Many assessments of religious behavior are actually self-reports of behaviors, such as religious attendance (Presser & Stinson, 1998) or prayer frequency (Masters & Spielmans, 2007). In one exception, Grossoehme et al. (2011) coded written prayers for styles of religious coping. Some online archival data can also be considered “behavioral,” such as the content of online sermons (e.g., Graham et al., 2009) and online search information related to R/S and positive psychology constructs (MacInnis & Hodson, 2015). Behavioral measures can also be assessed in the laboratory, such as the use of reciprocity to measure gratitude (Tsang & Martin, 2017) and physical distance from a conflict-discussion partner to study humility (Wright et al., 2018).

Despite their strengths, behavioral measures should complement and not replace self-reports. Self-reports can provide more nuanced pictures of psychological constructs. For example, self-report measures of gratitude (McCullough et al., 2002) can assess gratitude with more complexity, compared to reciprocity behavior (Tsang & Martin, 2017). Alone, behavioral measures might have ambiguous validity; reciprocity could be the result of gratitude but also could reflect compliance to norms. Moreover, because behaviors can be particularly sensitive to context, they should not be used as sole indicators of individual differences (Dang et al., 2020).

Implicit Measures

Self-report measures are most relevant when assessing conscious, easily accessible attitudes. In contrast, implicit measures purport to assess unconscious attitudes in ways that are less affected by self-presentation. For example, an implicit measure of humility (relative to arrogance) assesses individuals’ reaction times when associating humble and arrogant words with self and other words (Rowatt et al., 2006). Religious beliefs and self-concepts can also be measured implicitly (LaBouff et al., 2010). Measures such as the Implicit Association Test (Shariff et al., 2008) and the Affect Misattribution Procedure (Ross et al., 2019) have been utilized to assess the strength of people’s associations between supernatural words and concepts of true/false and real/imaginary.

Physiological Measures

In both the psychology of R/S and positive psychology, physiological measures can uncover mechanisms and functions of constructs. Brain imaging technology has shed light on brain structures involved in R/S (Maselko, 2013) and in character virtues such as patience (van den Bos et al., 2014). Other physiological measures (e.g., blood pressure, heart rate, and galvanic skin response) have been used to study constructs such as spirituality (Newberg, 2014) and gratitude (Ginty et al., 2020). Chemical markers like high oxytocin may be relevant to R/S (Sasaki et al., 2011) and forgiveness (Worthington et al., 2015), and low cortisol may be relevant to optimism (Nicolson et al. 2020).

Physiological indicators are indirect measures that can provide insight into the embodied nature of psychological phenomena. Because the processes assessed are automatic, they are less contaminated by self-presentation biases. As with behavioral measures, physiological measures are often paired with other assessment methods in order to enhance their validity. Despite their strengths, physiological measures are costly to administer, both in terms of time and resources, especially when compared to other methods such as self-report surveys. Moreover, particular care should be taken to choose appropriate control or comparison groups (Newberg, 2014).

What Kinds of Studies?

Even the best measures cannot compensate for lack of rigor in research design. Below we summarize the strengths and weaknesses of several quantitative research designs that are utilized in R/S and positive psychology studies, as well as a few qualitative and mixed-methods designs.

Cross-Sectional, Correlational Research

The vast majority of research in these two fields are cross-sectional, correlational studies (see Table 8.1). Many key variables, such as religious affiliation or dispositional hope, are challenging to manipulate experimentally. As an example, most studies on R/S and depression utilize cross-sectional, correlational data (Koenig et al., 2012). These studies usually show an inverse relationship between R/S and depression, consistent with a protective effect. However, because of the causal ambiguity embedded in cross-sectional research, it may also be that depressed people have less motivation to attend religious services (Maselko et al., 2012). Likewise, there may be confounding variables (such as personality; Mihaljevic et al., 2016) that are causing associations between R/S and depression.

Longitudinal Research

Longitudinal studies measure variables at different points in time, removing some causal ambiguity. They can enable researchers to see (a) if earlier levels of one variable affect later levels of another, (b) if there are bidirectional temporal associations between variables, and (c) if there are between-person differences in within-person trajectories across time. As an example, Hardy et al. (2020) utilized growth curve modeling to uncover positive bidirectional relationships in within-person variation in R/S and both cognitive and emotional self-regulation over time. Studies that include several measures of predictor and outcome variables over time (and control for covariates) are better able to test for directionality and reciprocal relationships (VanderWeele et al., 2016; see also Davis et al., Chap. 18, Appendix 18.S2, this volume; Long & VanderWeele, Chap. 25, this volume, Appendix 25.S2).

Experience sampling and daily diary methodologies are two longitudinal methods that help examine within-person processes and have been gaining traction in the fields of positive psychology and the psychology of R/S. For example, Balkaya-Ince et al. (2020) used an experience sampling approach when investigating the daily and momentary fluctuations in Muslim American adolescents’ self-reported religious/spiritual identity in relation to civic engagement and maternal religious/spiritual socialization. They collected data at multiple (random) time points within a day by sending a signal to the participants’ cell phones. Benefits to the experience sampling approach include increased ecological validity (Hektner et al., 2006) and limiting recall bias that may occur in cross-sectional self-report studies (Shiffman et al., 2008), but self-report bias may still persist with this methodological approach (Hofmann & Patel, 2015).

Experimental Research

Although true experiments were initially rare in R/S, the use of religious/spiritual primes in experiments has become more common. For example, experiments have identified causal relationships between religious/spiritual salience and prosociality (Shariff et al., 2016). Positive psychology has utilized experiments since its inception, yet the proportion of experimental studies is still incredibly small relative to other nonexperimental designs (see Table 8.1). Much of experimental positive psychology has focused on interventions to promote outcomes like happiness and well-being (Carr et al., 2020).

Despite its advantages for causal inference, experimental research has its own challenges. Behavioral priming results have been notoriously difficult to replicate (Cesario, 2014), and questions remain about the strength of religious/spiritual priming effects (van Elk et al., 2015), especially among those who do not identify as religious and/or spiritual. The careful consideration of control groups is also of paramount importance (Galen, 2012).

Qualitative and Mixed-Methods Research

Although quantitative research is useful for theory testing, it may lose much of the richness and complexity inherent in religious/spiritual and positive psychological constructs. Qualitative and mixed-methods research can provide the depth that quantitative research might miss. Qualitative research involves studying phenomena in their natural setting and trying to understand their subjective meanings. These studies collect observational, interview, textual, or audiovisual data and then analyze it using approaches such as grounded theory, phenomenology, narrative analysis, ethnography, case study (Creswell & Creswell, 2018), thematic analysis (Braun & Clark, 2006), and consensual qualitative research (Hill et al., 1997). Mixed-methods research combines elements of both qualitative and quantitative research to capitalize on the strengths of each (Johnson et al., 2007).

Quantitative research has generally dominated in the psychology of R/S (Hood, 2012) and positive psychology fields (Snyder et al., 2021). Yet since the 1990s, the number and rigor of qualitative psychological studies of R/S and positive psychology has increased substantially (Davis et al., 2016; Snyder et al., 2021). Qualitative methods hold promise for the empirical study of R/S and positive psychology because many of these phenomena are multidimensional, multilayered, context-sensitive, and culturally embedded (Loewenthal, 2013; Pedrotti & Edwards, 2017), and qualitative methods often can capture these aspects better than quantitative methods alone (Davis et al., 2016).

Similarly, mixed-methods research is becoming more common in the psychology of R/S and positive psychology (Davis et al., 2016; Snyder et al., 2021). Mixed-methods research tends to be costly in terms of resources, but it draws on strengths of qualitative and quantitative methods to develop a more complete and contextualized understanding (Creswell & Creswell, 2018). For example, in a large mixed-methods study, Delle Fave et al.’s (2011) qualitative data revealed evidence that across six countries in Europe, Africa, and Australia, laypeople defined happiness as “a condition of psychological balance and harmony” (p. 185), and they mainly associated their happiness and meaning with the quality of their family and social relationships (but not with their religiousness or spirituality). Likewise, the study’s quantitative data revealed that participants’ R/S did not contribute significantly to their subjective sense of happiness, meaning, or satisfaction with life.

What Kind of Setting? Field Research

Research setting is crucial when considering methodological diversity. In both fields, most research is conducted in the laboratory or via online surveys; very little field research exists either in psychology of R/S (Wright, 2018) or positive psychology (see Table 8.1). Field experiments provide causal tests with higher ecological validity compared to lab experiments. Although fieldwork provides rich data and information about a specific context, it may be limited in its generalizability (Anczyk et al., 2019; cf. Wright, 2018). Field research can be useful in reaching diverse populations and in understanding phenomena not well-captured with student samples or in the lab (e.g., religiously motivated violence; Ginges et al., 2011).

One example of fieldwork in R/S is the study of religious rituals (see Hobson et al., 2018, for a review). Rituals often lead to positive outcomes, including enhanced regulation of emotions, goals, and social connections (Hobson et al., 2018). In one cross-cultural field experiment, participants who engaged in a “high-ordeal” (i.e., physically taxing/painful) religious ritual were more generous and held more inclusive social identities than those who engaged in a “low-ordeal” ritual (Xygalatas et al., 2013).

What’s New? Computational Methods, Cross-Cultural Data, and Open Science

Computational Methods

Through the use of new computational methods, researchers can utilize large amounts of data readily available online (Yaden et al., 2018). For instance, language data derived from social media can be analyzed with natural language processing tools such as Linguistic Inquiry and Word Count (LIWC; Pennebaker et al., 2015). Ritter et al. (2014) analyzed data from 16,000 Twitter users and found that Christians used more positive and fewer negative words compared to atheists. Computational methods can help researchers make efficient use of large amounts of preexisting data in natural settings.

Cross-Cultural Data

Researchers in R/S and positive psychology are also making use of online data repositories, such as the Association of Religion Data Archives (ARDA: www.theARDA.com), International Social Survey Programme (http://issp.org/), and Gallup World Poll (https://www.gallup.com/analytics/318875/global-research.aspx). These provide national and international data, enabling cross-religious and cross-country comparisons. For instance, Diener et al. (2011) utilized the Gallup World Poll to investigate the moderating effect of societal circumstances on the relationship between R/S and subjective well-being. These data repositories provide researchers invaluable opportunities to examine questions cross-culturally.

Open Science

Openness and transparency are emerging concerns in scientific research (Aalbersberg et al., 2018). Open science practices encourage researchers to share materials and data, preregister hypotheses and methods, attempt replication in independent labs, and verify that peer-reviewed publications have been appropriately vetted. Open science practices can be used in quantitative (Charles et al., 2019) and qualitative research (Haven et al., 2020). Preregistering hypotheses is becoming the norm (Nosek & Lindsay, 2018). Many journals are adopting open-science policies that include making materials and data available (Lindsay, 2017; see https://topfactor.org/), and several journals (including the Journal of Positive Psychology, International Journal for the Psychology of Religion, and Archive for Psychology of Religion) support open science badges and/or registered reports. We hope more positive psychology and R/S journals follow suit.

Researchers can benefit from teaming together to conduct and replicate studies. In an innovative collaboration, Landy et al. (2020) led 15 teams testing five research questions (about moral judgment, negotiation, and implicit cognition) using two large samples of participants. Effect sizes varied dramatically (and in opposite directions) when different labs tested the same research questions. Thankfully, the days of publishing single studies based on unrepresentative convenience samples without replication seem to be nearly over.

What’s Next?

In what methodological directions should psychology of R/S and positive psychology move? We present four suggestions. First, both fields should increase their use of qualitative data and mixed-method studies in order to capture the complexity and culturally embedded nature of important constructs. Qualitative and mixed methods are particularly important when researchers are exploring understudied topics (e.g., religious/spiritual resilience), complex phenomena (e.g., suffering), cultural nuances, and cross-cultural comparisons (Davis et al., 2016).

In the quantitative realm, researchers should increase their use of theory-driven experiments and longitudinal studies, thereby permitting better investigation of causal mechanisms and directionality. Much experimental work in both fields focuses on the effects of religious/spiritual priming or positive interventions. Additional basic research is needed to investigate variables that cause or change R/S and positive psychological constructs. Furthermore, researchers need to utilize more implicit, behavioral, and physiological measures, especially when studying constructs that may be highly susceptible to social desirability bias.

Researchers also need to increase their use of open science practices, and journals in psychology of R/S and positive psychology need to continue to promote (or require) these practices. Especially in these fields, where topics such as R/S may still be highly controversial (Charles et al., 2019) and interventions related to positive psychology may be broadly applied, it becomes important to adopt practices such as preregistration and open science badges. Doing so will greatly improve the quality, credibility, and replicability of research.

Last, researchers must attend to sample diversity in addition to methodological diversity. The majority of psychology studies, including those in positive psychology and the psychology of R/S, utilize unrepresentative samples (i.e., WEIRD—Western, Educated, Industrialized, Rich, Democratic samples), which limits generalizations we can make about human nature or possible cross-cultural universals (Henrich et al., 2010). The increased use of online data and computational methods can facilitate increased sample diversity. More attention to this issue is urgently needed in order to move beyond merely the psychology of Western R/S and virtues.

Conclusions

As the study of religiousness, spirituality, and positive psychology increase in prominence, the time is ripe to reevaluate these two fields’ methodological diversity. New technology is increasing researchers’ ability to look inside the brain, collect large amounts of online data from diverse samples, conduct research with the utmost integrity and generosity, and network with other researchers across the world. Researchers have more opportunities than ever to expand these fields to explore a richer, more complex psychology of religiousness, spirituality, virtue, and well-being—across a wider variety of religions, spiritualities, and cultures.