Roman R. Williams & Kyle Whitehouse

Originally published in June 2015 in Review of Religious Research.


Visual research methods—image-based techniques for collecting, analyzing, and explaining data—are not mainstream in the social scientific study of religion. Few sociologists employ them, fewer still to study religion. While mainstream qualitative and quantitative methods have much to offer, words and numbers alone may miss important dimensions of religion and spirituality in the contemporary world. This research note provides an overview of work on the most commonly used visual research technique in the sociology of religion: photo elicitation (PE). We frame our essay around four questions. What is visual sociology? What is photo elicitation? How has PE been used in the social scientific study of religion? And how else could this technique be utilized in religious research?


Visual research methods are not mainstream in sociology, much less in the sociology of religion. The absence of these methods is curious given the visual and material nature of religion and spirituality in the contemporary world. Houses of worship, for example, of all types dot the social landscape (Vergara 2005; Richter 2007; Krieger 2011; Day 2014). The visual permeates religious culture (Morgan and Promey 2001) and images of religion are present throughout popular culture (Nardella 2012). Markers of religious presence and culture show up in global cities as migrants settle into their new context (Krase and Shortell 2011). The presence of religion is not limited to the material and symbolic features of culture, it is likewise found in the cultural toolkits that help religious and spiritual people navigate their everyday lives (Williams 2010b, 2013; Ammerman 2013). Research methods that rely on words or numbers alone may miss visual data important to understanding religion and spirituality. In this article we discuss a visual research technique known as photo elicitation (PE), the use of images in interviews to generate data. This research note is the result of a faculty-student research project on visual research methods in the sociology of religion. We frame our essay around four questions. What is visual sociology? What is photo elicitation? How has PE been used in the social scientific study of religion? And how else could this technique be utilized in religious research?

What Is Visual Sociology?

Howard Becker observed that ‘‘[p]hotography and sociology have approximately the same birth date, if you count sociology’s birth as the publication of Comte’s work which gave it its name, and photography’s birth as the date in 1839 when Daguerre made public his method for fixing an image on a metal plate’’ (Becker 1974:3; cf. Harper 1988). And while these two new ways of seeing were slow to integrate, photographs not only ‘‘shaped Durkheim’s ethnographic depiction of the Arrernte, but also his selection of theoretical concepts and methodological procedures’’ even though photographs are absent from Elementary Forms of Religious Life (Kreinath 2012:368). Early examples of images in social research appear in 31 articles written in the American Journal of Sociology (AJS) between 1896 and 1916 (Stasz 1979). Among them 23 articles mention religion, but few offer more than a superficial treatment. McClintock’s photograph of ‘‘A Foot Washing Service of the Hardshell Baptists on the Mud Fork of Island Creek’’ (1901:19) may be the first image to depict religion published in a sociology journal.After 1916, however, photographs are absent from sociology journals until much later in the twentieth century. Schnettler and Raab suggest that photography was unpopular in sociological research because the ‘‘increasing influence of statistical methods induced an abrupt substitution of photos by formula[s], charts, and tables as the predominant form of appropriate scientific illustration’’ (Schnettler and Raab 2009:266).

Even though the use of photographs in sociological research may have been brushed aside temporarily, some sociologists of the early twentieth century recognized their potential. An important example is Middletown (1929) co-author Robert Lynd, who was influential in shaping the work of Roy Stryker and the 270,000 images made during the Great Depression by Farm Security Administration (FSA) photographers (Suchar 1997). Lynd became enthusiastic while viewing FSA photographs, Stryker recalled years later, and remarked, ‘‘‘This is a wonderful device for sociologists.’ [Lynd] then got off onto a long discourse on the need to make people really see’’ (Stryker and Wood 1973:8, their emphasis; cf. Suchar 1997). Apparently the images were a little too sociological for the taste of renowned photographer Ansel Adams, who criticized FSA photographers as ‘‘a bunch of sociologists with cameras’’ (Stryker and Wood 1973:8). Adams’ criticism was prescient, as one FSA photographer, John Collier, Jr., went on to pioneer photo elicitation (Collier 1957) and to make important contributions in visual anthropology (Collier 1967; Collier and Collier 1986). While FSA photographs have received some attention from scholars of religion (e.g., McDannell 2004), they await further analysis—or as Becker has suggested, photos potentially answer a question if one knows what that question is (Rieger 1986:7).

It was not until the 1970s that visual sociology took form around a core group of sociologists, most of whom were ‘‘photographers as well as sociologists, and predisposed to field work research’’ (Harper 1996:71). Because of these ethnographic origins, some mistake visual sociology as a subfield of qualitative sociology. While many visual techniques may seem to be located at the qualitative end of the research methods spectrum, such a view is too narrow. Others may think of visual sociology as a subdiscipline in sociology analogous to the sociology of religion or cultural sociology, but this is also an incomplete understanding. Neither subfield nor subdiscipline, visual sociology is best understood as

…a cross-cutting field of inquiry, a way of doing and thinking that influences the whole process of researching (conceptualizing, gathering, and communicating). It is not just a ‘‘sociology of the visual’’ (as subject), but also a method for sociology in general (whatever its field: law, religion, culture, etc.) and a way of thinking, conceptualizing, and presenting ideas and findings.

(Pauwels 2011:13)

Likewise, this approach considers the broadest possible range of visual materials including, but not limited to, photographs, film, video, print media, digital media, maps, and drawings—we use the term images as shorthand for these visual data. Visual sociology, then, is an identifiable and ever-expanding toolkit of image-based techniques for collecting, analyzing, and presenting information about society and culture.1 While visual sociology may be defined in various ways (Grady 1996), all agree that it is a way of doing sociology visually.

What Is Photo Elicitation?

When measured by number of publications, photo elicitation appears to be the most popular visual research method in the sociology of religion. Originally developed by John Collier (1957), this technique is ‘‘based on the simple idea of inserting a photograph into a research interview’’ (Harper 2002:13) which is used as a prompt to ‘‘invoke comments, memory and discussion’’ (Banks 2007:65). While the variety of materials that may be used in an interview—drawings, artwork, and objects, for example (Ganesh 2011; Ridgely 2011)—has led some to begin using the term ‘‘elicitation interviewing’’ (e.g., Wagner 2011), photographs are the mainstay and therefore the focus of this article. The procedure for photo elicitation is straightforward. First one must consider the source of the photographs. In visual research, images fall along a continuum ranging from found visual artifacts (a photo found at a church rummage sale) to researcher-generated data (a photograph taken in the field by a researcher). Between these ends lie ‘‘artifacts with known provenance’’ (e.g., a magazine advertisement), ‘‘other researcher’s data,’’ and ‘‘respondent-generated data’’ (Pauwels 2011:7). As one progresses along this spectrum from found to researcher-produced images, the researcher’s involvement in image production as well as their knowledge of why and how an image was produced increases. Collaboration is also present as researchers partner with respondents, other researchers, or professionals (e.g., a photographer) to produce images. In photo elicitation, using the same images with all interviewees—whether the photo was made by the researcher, prepared by a professional photographer, or found—makes comparisons among research participants possible. Using participant-produced images in PE, on the other hand, has the advantage of giving agency to the research participant (they decide what to photograph within the guidelines of the research) and has the potential of taking the research in directions the researcher may not have anticipated (see Ammerman and Williams 2012:123–127).

At the time of the interview, the research subject is given the photographs and invited to discuss them. As the interviewee tells the story of their photograph, the researcher may nudge the conversation in directions relevant to the aims of the research. Those who use this data-gathering technique comment on the way it changes the typical researcher-participant relationship. Where standard interview techniques tend to be face-to-face, the researcher and participant tend to be shoulder-to-shoulder focused on the image before them. Likewise, photo elicitation is known to shift the researcher-participant power dynamic. ‘‘If the PE interview goes well the person being interviewed sees himself or herself as the expert, as the researcher becomes the student. The photo becomes a bridge between people who may not even understand the extent to which they see the world differently’’ (Harper

Another merit of photo elicitation is its ability to break the frames of participants (Harper 2002) and researchers (Samuels 2004). The concept of ‘‘breaking frames’’ was first used by Harper (2002) to describe the ability of photos to help participants see their familiar and mundane everyday lives in a new light. Harper, for example, achieved this by showing farmers aerial photos of their farms and historic photographs of farm life in interviews. Samuels extends this notion to the researcher’s frame: putting the camera in participants’ hands can break the frame of the researcher, creating unanticipated results that allow the participants’ ‘‘own world and perspective to impinge on [the researcher’s] frame of reference’’ (Samuels 2004:1542).

On a practical level, several challenges arise with PE. While some of these challenges are discussed in our overview of the literature below, a few general comments are appropriate. Schwartzenberg does well to remind visual researchers that ‘‘…human memory is fraught with inaccuracy, distortion and subjectivity’’ (2005:72). While adding photos to an interview provides visual evidence from a specific moment in time, PE still relies on the memory and interpretations of research subjects. The interval between the moment a photograph is made and its use in an interview is an important consideration: using family photos from many years ago (as Schwartzenberg did) may be problematic when it comes to accuracy or objectivity if those are one’s aims. Ammerman and Williams (2012) avoid this issue by thinking about the images and stories told about them during an interview as a narrative process, meaning-making, constructing reality—perhaps we could revise the Thomas theorem (Thomas and Thomas 1928:572) to state that ‘‘if the narratives elicited by photographs are presented as real by the interviewee, they are real in their consequences.’’

When participants produce images during the course of research, a researcher is faced with purchasing and delivering cameras to participants, offering tutorials on how to use cameras and the basics of photography, retrieving and processing film, scheduling and conducting interviews, and preserving research materials. Some of these logistical challenges are mitigated by the use of participants’ own cameras (e.g., cell phone) and the ability to use digital technologies. Analog equipment may have the advantage, however, of preventing participants from previewing and redacting images. Sometimes participants are less than enthusiastic about the assignment of taking photos, forgetful, or slow to complete their photo assignment. The logistical, technical, and financial costs associated with PE are minimized when using found or researcher-produced images. Confidentiality concerns also must be addressed. Institutional review boards unfamiliar with or resistant to visual research in the social sciences may need to be persuaded. In some cases, images may not be shown in research reports due to copyright concerns or to honor a participant’s wishes. None of these challenges are insurmountable and strategies for addressing these issues are found in the growing body of visual methods literature (see Clark 2012; Mitchell 2011:15–32; Rowe 2011; Wiles et al. 2011). In addition to the literature reviewed below, we recommend that those interested in using photo elicitation read the relevant chapter in Douglas Harper’s Visual Sociology (2012:155–187). In our opinion it offers one of the best introductions currently available.

How Has Photo Elicitation Been Used in the Social Scientific Study of Religion?

Even though Harper’s (2012) introduction may be a helpful starting point for anyone interested in photo elicitation, it has at least one shortcoming. In it Harper builds upon an earlier attempt to map the terrain of PE by identifying and summarizing ‘‘all examples of photo elicitation’’ (2002:15), recognizing that while he ‘‘searched carefully in the updated literature review [he] undoubtedly missed some this time around’’ (2012:179). His efforts resulted in a collection of 111 dissertations, articles, book chapters and books from PE’s invention by John Collier in 1956–2010. Only two entries in Harper’s bibliography, however, focus on religion (Samuels 2004; Vassenden and Andersson 2010). These results are incomplete: Harper missed sixteen of the eighteen works published in English during the period 1956–2010. This oversight amounts to missing 14% of the literature on photo elicitation. In what follows, we provide an overview of the literature, including the work inadvertently omitted by Harper and adding six pieces published 2011–2013. Like Harper we are aware that, despite our best efforts and intentions, we also may have inadvertently missed important examples of photo elicitation in the social scientific study of religion. Our overview is organized around the visual materials continuum described above (found, researcher produced, and participant-produced images) and the literature is discussed in chronological order.

Found Images

Images are ubiquitous in contemporary society and many may be useful in interview settings to create conversations about religion and spirituality. Three studies employ found images in PE to study religion and spirituality. In her investigation of the religious-like devotion of Mac computer enthusiasts, Lam included ‘‘pictures from the current Apple ad campaign… and a ‘Mac Addict versus Windows User’ picture from the MacAddict magazines’’ (2001:247) during semi-structured interviews. The advertisement portrays stereotypical Mac and PC owners, and was used to elicit responses about these differences. The image proved effective in teasing out the meaning of being a Mac enthusiast, and the role of Mac culture in identity construction and maintenance. As helpful as PE may have been in her study, Lam has little to say about PE as a method: the author does not identify the technique as PE, draw from available literature, or include any visual data in the article.

In their study of 15-to-25-year-olds in Britain, Savage and her interdisciplinary co-investigators use fifteen photos of advertisements (identifiable English brands) and cultural icons (such as Prince Charles kissing Diana, the World Trade Center attacks, and separate photos of a generic elderly couple, female athlete, and male model, among others) to explore participants’ meanings, values, and spirituality. Interviews were conducted in group settings in which the authors asked participants to select a photo to which they were particularly attached and to explain the attraction. In addition, two groups were invited to choose three or four photos and create a story using the photos. These interviews were productive in generating conversations with and among participants, thereby pursuing the researchers’ objective to observe ‘‘the kind of thinking and interacting that occurred in response to images’’ (Savage et al. 2006:98). These conversations allowed researchers to consider the role of images in creating, maintaining, and/or contesting the boundaries of cultural norms and values. Interestingly, they found that ambiguous images, ‘‘rather than traditional, or even ‘alternative,’ religious images, have the most potential for eliciting open thinking, formative spirituality, and talk about religion…’’ (Savage et al. 2006:115). While the authors are not in conversation with the broader visual methods literature, their reflections on their methodology make an important contribution to the PE literature.

A familiar sentiment in visual methods is that the standard qualitative toolkit lacks sufficient equipment for collecting data on social and cultural phenomena. Such was the experience of Notermans and Kommers who set out to ‘‘discover the meanings of Dutch Lourdes pilgrims attribute to Mary. However, Mary appeared to be so intimately and intensely involved in people’s personal sorrows that they were unable to express themselves in consistent stories about her’’ (2012:3).2 Instead of generating articulate responses, interview questions about participants’ personal relationship with Mary fell flat, producing silence.

These silent reactions to questions about the personal relationship with Mary can be clarified in three interrelated ways. One reason is that people’s personal communication with Mary often also lacks the words. It is mainly a language of thought and herewith an inner and silent communication….Another reason is that when people speak aloud with Mary, they often articulate standard prayer texts such as Hail Mary and do not look for own words [sic] to express their relationship with her. A final and most relevant reason is that to the pilgrims who have been interviewed, Marian devotion is an intensely emotional practice…. Mary mirrors the excessive pain people experience in everyday life, and that pain is hard to tell.

(Notermans and Kommers 2012:7)

Faced with this challenge, Notermans and Kommers turned to elicitation interviews, which the authors describe as ‘‘iconographic elicitation’’ due to their use of 30 cards bearing different visual representations (icons) of Mary. By re-interviewing participants in this manner, Notermans and Kommers were able to (1) ‘‘discover how pilgrims select from various icons and what reasons they have for loving particular ones ‘very much’ and others ‘not at all’’’ and (2) ‘‘overcome the problem of reticence encountered when asking about personal relationships with Mary’’ (2012:8). This variation of photo elicitation ‘‘…helped both the pilgrims and the interviewer to elicit the stories that otherwise would probably not have been told’’ (2012:8). Drawing forth the interior worlds of research participants is a concern shared by sociologists of religion and visual researchers (e.g., Hogan and Pink 2013). Notermans and Kommers’ article is an example of how visual methods may be useful in this enterprise. The burden of using found images is on the researcher who is required to identify material that is appropriate to the aims of the research, which often involves some level of trial, error, and refinement (see Harper 2012:167–179). By employing culturally significant images (Mac computer advertisement, popular culture, and Marian icons), these studies were able to draw in respondents and stimulate conversations that lend meaning to social and cultural artifacts.

Researcher-Produced Images

At the other end of the visual materials continuum are images created by the researcher. In comparison to found images, when a researcher produces the images, she is more aware of the context in which the image was made and image-making decisions (e.g., how it was framed, camera settings, visual effects) that affect its mood or composition. Three studies introduced researcher-produced images into interviews. Benetta Jules-Rosette (1975, 1980) used photographs and video footage to collect data for her ethnography of the Apostles of John Maranke, an African initiated church. She presented the photographs ‘‘to members after the ceremonies to elicit their personal impressions of ritual events’’ (1975:47). ‘‘[B]y using photographs as a starting point for my questions,’’ Jules-Rosette remarks, ‘‘elusive manners that were considered important to membership were brought to light’’ (1975:52). Not only is her 1975 book a compelling ethnography, she appears to be the first to use PE to study religion.

In a second study that employed researcher-produced images, Eleanor Nesbitt explores the perspectives and experiences of religion among 8-to-13-year-olds in Coventry, England (Jackson and Nesbitt 1993; Nesbitt 1993, 2000a, b). She notes that researching religion among children is often problematic: sensitivity, age, language, and reflexivity must be carefully navigated, especially when using traditional interviews. These issues may surface through significant age differences between children and researchers, children’s difficulty with the more academic language that is sometimes used in interviews, and the researcher being unaware of the power dynamics at play. While no research method will truly do away with these issues, Nesbitt suggests that PE may reduce them by shifting the focus away from the child to the image. Instead of being a research subject under the microscope, the child is given agency and treated as a collaborator in examining the social world. This reorientation of the interview allows children to show their knowledge of a certain area if elements of the photo are missing or otherwise out of the frame, as was the case in an overexposed photo of images in a Hindu temple.

The ‘‘Contemporary Architecture of Religious Transmission’’ study (Hintze et al. 2008) utilized a combination of researcher-produced and found images to investigate ‘‘how secularization and desecularization in Western Europe is becoming influenced by new migrant groups’’ (Vassenden and Andersson 2010:159). Images of religious buildings (mosque, Catholic church, Protestant church) helped researchers explore congregations’ relationship(s) to its surrounding community in Hamburg, Oslo, and London (Andersson 2009) and photos of artifacts such as the Koran shed light on the meaning(s) and practice of religion in participants’ daily lives (Vassenden and Andersson 2010).

Anders Vassenden and Mette Anderson note that ‘‘quite a few of [the respondents] said they had found it easier to answer oral questions than to talk about images’’ (2010:151). Interviewees’ unfamiliarity with photo elicitation may have contributed to difficulties: often times interviewees think the researcher is looking for a ‘‘right’’ answer instead of an open response to the photograph. Another reason talking about images was difficult has to do with the breadth of meaning of photos. For example, a Bible or a Qur’an can mean any one of a multitude of things,
and people must mull over the direction they wish to answer instead of simply answering a straightforward question posed in a typical qualitative interview. Even so, Vassenden and Andersson are persuaded that the results of the project would have been different if they had used a standard oral interviewing strategy.

Most likely, we would not have heard these stories, as we simply would not have known how and what to ask. It would have taken in-depth expertise of both Islam and migrant Catholics merely to formulate the proper questions in order to get these stories. This is noteworthy, as it suggests images can have the quality of tapping perspectives that researchers may not have thought about in advance.

(Vassenden and Andersson 2010:159)

Their conclusion echoes Harper’s assertion that ‘‘photo elicitation mines deeper shafts into a different part of human consciousness than do words-alone interviews’’ (2002:22–23).

Participant-Produced Images

Images produced by interviewees fall between found and researcher-generated images on the visual materials continuum. This variation of PE invites research participants to make photographs relevant to the study, but without the researcher’s direct involvement. Participants typically receive instruction on how to use photographic equipment (e.g., a disposable camera) and directions on what to photograph. Sometimes these directions are intentionally vague; in other cases participants are given a list of categories or specific topics to photograph, which is referred to as a ‘‘shooting script’’ (Suchar 1997). When interviewees produce the images, the research becomes more dynamic and collaborative—it has the effect of switching the role of researcher and interviewee, allowing the participant to drive the research (Clark 1999; Heisley and Levy 1991).

Jeffery Samuels employed a shooting script of 11 topics with children in training to become Buddhist monks in Sri Lanka (Samuels 2004, 2007, 2010). Samuels makes an important methodological observation when he discusses how much more productive this technique is than traditional interviews: the photographs, memories, and narratives helped to challenge the researcher’s preconceived notions of what it means to be/become a monk. Samuels reports that, in his experience, PE assignments without detailed guidelines (i.e., a script) result in interviews that lack the depth and thoughtfulness of those that utilized a shooting script (2007).

Thomas Josephsohn used participant-produced images to investigate how objects relate to religious identity and ‘‘transientness’’ (church-hopping) in a nondenominational evangelical Christian college group. These interviews were essential in identifying (1) the ways religious identity is portrayed and achieved through icons and (2) how ‘‘the group was able to provide for the needs of its constituency without having to resort to superficial practices that may have led to dissatisfaction’’ (Josephsohn 2007:34).

Sarah Dunlop’s (2008) study of religion and spirituality among youth aged 17–25 years in five former Soviet countries combines found and participant-produced images. Her ambitious project involved three procedures. First, ‘‘living space interviews’’ presented 35 found images to participants and asked them to discuss which images they would or would not hang on their bedroom wall. Next, interviewees discussed the images on their walls and what these pictures mean and represent. A final portion of the first interview involved asking participants direct questions about their spirituality and beliefs. In a second interview, participants were shown a wide spectrum of images and asked about their future, including what they thought hope was, what they hope for in life, and their conception of life after death. A third interview invited participants to take three photos a day for a week of ‘‘objects, places, people, and moments of significance’’ (Dunlop 2008:31). ‘‘This part of the field research was fascinating,’’ Dunlop remarks, ‘‘because the camera was in the hands of the students, giving them the power to represent their world’’ (2008:31). Dunlop’s creative methodology and reflection on visual techniques make an important contribution to the literature.

Understanding the everyday religious experiences is an important direction in research in the sociology of religion (Ammerman 2007; McGuire 2008; Edgell 2012). Two studies use photo elicitation in the investigation of lived religion: Nancy Ammerman’s ‘‘Spiritual Narratives in Everyday Life’’ project (Williams 2010b; Ammerman 2013) and Roman Williams’ doctoral thesis on evangelical Christian international students from Asia studying in American colleges and universities (Williams 2010a, 2013). Ammerman and Williams’ methodological contribution centers on the way they incorporate PE and narrativity. They advocate for a narrative approach to lived religion built upon the notion that social life is storied and that these stories are important to the social construction of meaning. By incorporating PE in their studies, they offer participants an additional way of telling their stories, one that transforms the researcher-participant relationship and the data they collected.

The seemingly simple act of putting a camera in the hands of participants can redefine their relationship to the research….They are no longer passive subjects responding to a researcher’s queries; they become fieldworkers who reveal answers to questions researchers might never have asked. The data are produced by participants: they are in the driver’s seat, they make decisions about what is important (or not). As a result they can guide us down new conceptual paths, helping us see the social world in new ways. As participants step back from their taken-for-granted social lives to reflect on their own visual images, themes such as meaning, community, relationships, memory, religion, and spirituality may emerge. [Photo elicitation interviews] stimulate agency among participants and stimulate reflexivity about their social world.

(Ammerman and Williams 2012:125–126, italics in original)

The purchase of PE for seeing religion in everyday life is evident in their subsequent publications in which they use participants’ verbal and visual narratives to discuss religion in the everyday lives of Americans (Ammerman 2013) and the social construction of a religiously-infused calling to secular vocations among international students from Asia studying in the United States (Williams 2013).

Sarah Dunlop and Pete Ward blur the boundaries between photo elicitation and photovoice (Wang and Burris 1997) in their study of young Polish immigrants and second-generation youth in Plymouth, England. After an initial period of participant observation in Polish businesses and Catholic churches, Dunlop and Ward invited research participants to a photography workshop. At the conclusion of the workshop the researchers asked for volunteer participants and invited them to make pictures that ‘‘represented what was sacred to them’’ over a week (Dunlop and Ward 2012:436). In addition to one-on-one interviews in which they discussed the photographs they took, participants met for focus groups and created collages using pictures from the internet. Finally, an exhibition of the photographs of sacred images were held at two local churches. In the process of thinking about and visually representing the sacred, these images offered windows into the experience of migrants by allowing them to voice concerns about differences between religion in Poland and England, the uncertainty of their lives during a time of migration, and shifting understandings of the self and the sacred. Even more than the importance of their findings is the applied nature Dunlop and Ward’s research: their use of PE and photovoice in a religious setting is unique in the visual methods literature. Using visual techniques to help marginalized groups tell their stories or express their needs is a direction worthy of pursuit in the social scientific study of religion.

Practical theologian Hendrik Pieter de Roest adapts photo elicitation to the study of church closures in The Netherlands. While his methodology is social scientific, he is also interested in giving ‘‘a voice to the congregation and demostrat[ing] the
importance of common space, its practices and symbols for the individual’’ and to ‘‘identify some indications for how to guide processes of church closures’’ (de Roest 2013:312). Like Dunlop and Ward above, de Roest’s project falls under the heading of action research. This innovative project is a modified form of photo elicitation that involves participants supplying written or oral comments on their own images or ones made by a third party (a congregant who happened to be an amateur photographer). Instead of using the images in a semi-structured interview, he attempts ‘‘…to learn which parts of the church building or symbols have become important or precious to them and what is at stake in their emotional attachment to them’’ (de Roest 2013:301–302) by analyzing the images. And while de Roest is unclear on this point, he seems to have attended congregational meetings organized by the pastor of the closing church in which participants shared their emotions, memories, and videos. These data allow him to identify six concepts ‘‘that give a picture of the values associated with the church building’’ (de Roest 2013:303). These concepts include sacred memories, attentiveness and solidarity, cherished welcoming fellowship, a creative and unconstrained atmosphere, participatory worship, and the building as a shelter and landmark. Through their images and the narratives they evoked, de Roest’s involvement allowed a congregation that was losing its sacred space to preserve their sense of meaning, memory, and community.


Our review of the literature suggests a disconnect between the sociology of religion and visual sociology. First, visual sociologists are relatively blind to important visual research conducted in the sociology of religion. Harper missed 16 of the 18 publications that use PE to examine religion and these articles are not cited widely within the broader visual methods literature. Blind spots such as this deprive visual sociologists of methodological and theoretical developments in the sociology of religion. Second, while a growing number of sociologists of religion continue to adopt visual research techniques, many researchers do not demonstrate an awareness of the visual sociology literature. On average, PE studies of religion include four references to visual research per article, whereas all other PE studies tend to include three times as many citations in each piece. It seems that sociologists of religion are disconnected from the visual methods literature; and their work is not as informed by the practical, methodological, and theoretical insights of visual sociologists as it should be. This article supplements Harper’s overview of the PE literature as a step toward addressing these concerns.

How Else Could Photo Elicitation be Utilized in Religious Research?

The range of applications of photo elicitation represented by the research outlined above offers a starting point for thinking about how PE may be incorporated into a study. Against the backdrop of the wider literature, this work on religion only scratches the surface. Instead of reproducing what has already been discussed elsewhere by Harper (2012), it seems more beneficial to offer some ideas about how else photo elicitation could be used in religious research with particular attention to applied approaches that may be of interest to the journal’s readership. Briefly we focus on two promising applications of PE: needs assessment and evaluation research.

First, taking a cue from the work of Dunlop and Ward (2012) photo elicitation may be employed as a tool for needs assessment—de Roest’s (2013) work is another example that should be considered even though we focus on Dunlop and Ward. The photographs taken during Dunlop and Ward’s study of the sacred among Polish migrants to England not only reveal participants’ concerns about shifting meanings of religion between cultures, unsettled feelings due to the migration process, and shifts in self-understanding, the images and the stories behind them also indicate needs. For example, these migrants expressed their needs for relationships, a sense of community, and practical tools to process existential challenges to their plausibility structures.

While Dunlop and Ward combine photo elicitation and photovoice, PE by itself is just as effective in needs assessment. One strategy is to invite participants to take contrasting photographs of personal likes and dislikes, a congregation’s opportunities and challenges, a religious groups strength’s and weakness, or where an interviewee feels close to or distant from the sacred. Prompts such as these that invite participants to think about the rage of their experiences and offer researchers points of comparison. And while our preference is for participant-produced images—while this is partly due to our experience in research, our preference rests on the ability of participant-produced images to engage research participants—a researcher may be successful with her own or found images.

Second, PE also shows promise as a means of program evaluation. A participant or researcher could chronicle an event visually or document different aspects of a program and use those images in an interview with participants to discuss their experiences, invite reflections on effectiveness, and solicit suggestions for improvement. A shooting script (a list of topics to photograph) could be used to collect visual data that focus on specific program outcomes. Photos taken at the beginning, midpoint, and end of a program also may be helpful in isolating instances of continuity and change (Hurworth 2004:167). When used in an interview, employing the same researcher-produced photographs with all program participants has the advantage of being able to make comparisons among participants, whereas photos unique to an interviewee (i.e., participant-produced) may reveal aspects of the program under evaluation that may not otherwise have been explored. And if aspects of the program are particularly abstract and therefore difficult to articulate or measure, images offer an additional means of communication (see Hurworth


Choosing the right methodological tool for the research task is important. PE is not an ‘‘infallible technique’’ (Collier 1957:858; cf. Samuels 2007), ‘‘magic formula’’ (Vassenden and Andersson 2010:151), or guarantee of useful interviews (Harper 2002:20). Even so, photo elicitation—and visual sociology more broadly—offers an additional tool for studying religion, one that compliments words-and-numbers approaches to research. Like Harper, we are persuaded that ‘‘collaborating with study participants to understand the meaning of an image, no matter who made it or for what purpose, is a dynamic, exciting proposition’’ (2012:206).


This essay is the result of a faculty-student research project funded by a Calvin College Alumni Association Faculty Grant and a Calvin College Department of Sociology and Social Work Deur Award.


1 Helpful introductions to visual research in the social sciences include Harper (2012), Mitchell (2011), and Pink (2013)—Williams (2014) offers a review of these books.

2 We are working from the OnlineFirst version of this article, which was available in September 2012.


Ammerman, Nancy T. (ed.). 2007. Everyday religion: Observing modern religious lives. New York, NY: Oxford University Press.

Ammerman, Nancy T. (ed.). 2013. Sacred stories, spiritual tribes: Religion and spirituality in everyday life. New York: Oxford University Press.

Ammerman, Nancy T., and Roman R. Williams. 2012. Speaking of methods: Eliciting religious narratives through interviews, photos, and oral diaries. In Annual review of the sociology of religion, volume 3: New methods in the sociology of religion, ed. L. Berzano, and O.P. Riis, 117–134. Leiden, The Netherlands: Brill.

Andersson, Mette. 2009. Religion in Inner City Oslo. Street Signs Spring: 5–7.

Banks, Marcus. 2007. Using visual data in qualitative research. Thousand Oaks, CA: SAGE Publications.

Becker, Howard. 1974. Photography and sociology. Studies in the Anthropology of Visual Communication 1(1): 3–26.

Clark, Andrew. 2012. Visual ethics in a contemporary landscape. In Advances in visual methodologies, ed. Sarah Pink, 17–36. Thousand Oaks, CA: Sage.

Clark, Cindy Dell. 1999. The autodriven interview: A photographic viewfinder into children’s experience. Visual Sociology 14:39–50.

Collier, John. 1957. Photography in anthopology: A report on two experiments. American Anthropologist 59(5):843–859.

Collier, John. 1967. Visual anthropology: Photography as a research method. New York: Holt, Rinehart, and Winston.

Collier, John, and Malcolm Collier. 1986. Visual anthropology: Photography as a research method. Albuquerque, NM: University of New Mexico Press.

Day, Katie. 2014. Faith on the avenue: religion on a city street. New York, NY: Oxford University Press.

Dunlop, Sarah. 2008. Visualising hope: Exploring the spirituality of young people in central and Eastern Europe. Cambridge: YTC Press.

Dunlop, Sarah, and Pete Ward. 2012. From obligation to consumption in two-and-a-half hours: A visual exploration of the sacred with young Polish migrants. Journal of Contemporary Religion 27(3):433–451.

de Roest, Hendrik Pieter. 2013. ‘Losing a Common Space to Connect’: An inquiry into inside perspectives on church closure using visual methods. International Journal of Practical Theology 17(2):292–313.

Edgell, Penny. 2012. A cultural sociology of religion: New directions. Annual Review of Sociology 38:247–265.

Ganesh, Tirupalavanam G. 2011. Children-produced drawings: An interpretive and analytical tool for researchers. In The SAGE handbook of visual research methods, ed. E. Margolis, and L. Pauwels, 214–240. SAGE Publications: Thousand Oaks, CA.

Grady, John. 1996. The scope of visual sociology. Visual Sociology 11(2):10–24.

Harper, Douglas. 1988. Visual sociology: Expanding sociological vision. The American Sociologist 19(1):54–70.

Harper, Douglas. 1996. Seeing sociology. The American Sociologist 27(3):69–78.

Harper, Douglas. 2002. Talking about pictures: A case for photo elicitation. Visual Studies 17(1): 13–26.

Harper, Douglas. 2012. Visual sociology. New York, NY: Routledge.

Heisley, Deborah D., and Sidney J. Levy. 1991. Autodriving: A photoelicitation technique. The Journal of Consumer Research 18(3): 257–272.

Hintze, Benjamin, Anders Vassenden, Roger Hewitt, and Vicky Skiftou. 2008. Migration and traces of religious architecture in European areas: Perceptions of youths. Diskurs Kindheitsund Jungendforschung Heft 4:385–400.

Hogan, Susan, and Sarah Pink. 2013. Visualizing interior worlds: Interdisciplinary routes to knowing. In Advances in visual methodology, ed. S. Pink, 230–247. Thousand Oaks, CA: SAGE Publications.

Hurworth, Rosalind. 2004. The use of the visual medium for program evaluation. Studies in Qualitative Methodology 7: 163–181.

Jackson, Robert, and Eleanor Nesbitt. 1993. Hindu children in Britain. Stoke-on-Trent, Staffordshire, England: Trentham Books.

Josephsohn, Tomas. 2007. Branding faith: Objects of consumerism in religious identity construction. Unpublished master’s thesis, University of Missouri-Columbia.

Jules-Rosette, Bennetta. 1975. African apostles: Ritual and conversion in the church of John Maranke. Ithica, NY: Cornell University Press.

Jules-Rosette, Bennetta. 1980. Ceremonial trance behavior in an African church: Private experience and public expression. Journal for the Scientific Study of Religion 19:1–16.

Krase, Jerome, and Timothy Shortell. 2011. Seeing difference: Spatial semiotics of ethnic and class identity in global cities. Visual Communication 10(3):367–400.

Kreinath, Jens. 2012. Discursive formation, ethnographic encounter, photographic evidence: The Centenary of Durkheim’s basic forms of religious life and the anthropological study of Australian aboriginal religion in his time. Visual Anthropology 25:367–420.

Krieger, Martin H. 2011. Urban tomographies. Philadelphia, PA: University of Pennsylvania Press.

Lam, Pui-Yan. 2001. May the force of the operating system be with you: Macintosh devotion as implicit religion. Sociology of Religion 62:243–262.

Lynd, Robert S., and Helen M. Lynd. 1929. Middletown, a study in contemporary American culture. New York, NY: Harcourt, Brace.

McClintock, S.S. 1901. The Kentucky Mountains and their feuds I: The people and their country. The American Journal of Sociology 7(1):1–28.

McDannell, Coleen. 2004. Picturing faith: Photography and the great depression. New Haven, CT: Yale University Press.

McGuire, Meredith. 2008. Lived religion: Faith and practice in everyday life. New York, NY: Oxford University Press.

Mitchell, Claudia. 2011. Doing visual research. Thousand Oaks, CA: SAGE Publications.

Morgan, David, and Sally M. Promey. 2001. Introduction. In The visual culture of American religions, ed. D. Morgan, and S. Promey, 1–26. Berkeley and Los Angeles, CA: University of California Press.

Nardella, Carlo. 2012. Religious symbols in Italian advertising: Symbolic appropriation and the management of consent. Journal of Contemporary Religion 27(2):217–240.

Nesbitt, Eleanor. 1993. Photographing worship: Ethnographic study of children’s participation in acts of worship. Visual Anthropology 5:285–306.

Nesbitt, Eleanor. 2000a. The religious lives of sikh children: A coventry based study. Leeds: University of Leeds.

Nesbitt, Eleanor. 2000b. Researching 8 to 13-year-olds’ perspectives on their experience of religion. In Researching children’s perspectives, ed. A. Lewis, and G. Lindsay, 135–149. Buckingham: Open University Press.

Notermans, Catrien, and Heleen Kommers. 2012. Researching religion: The iconographic elicitation method. Qualitative Research. doi:10.1177/1468794112459672.

Pauwels, Luc. 2011. An integrated conceptual framework for visual social research. In The SAGE handbook of visual research methods, ed. E. Margolis, and L. Pauwels, 3–23. Thousand Oaks, CA: SAGE Publications.

Rieger, Jon. 1986. Howard Becker’s summer workshop in visual sociology: A personal view. Visual Studies 1(1):5–8.

Pink, Sarah. 2013. Doing visual ethnography. Thousand Oaks, CA: SAGE.

Richter, Philip. 2007. Sunday: A photo essay. Salisbury: Lucidity.

Ridgely, Susan B. 2011. Maybe the picture will tell you: Methods for hearing children’s perspectives on religion. In The study of children in religions: A methods handbook, ed. S. Ridgely, 80–94. New York, London: New York University Press.

Rowe, Jeremy. 2011. Legal issues of using images in research. In The SAGE handbook of visual research methods, ed. Eric Margolis, and Luc Pauwels, 707–722. Thousand Oaks, CA: SAGE Publications.

Samuels, Jeffery. 2004. Breaking the ethnographer’s frames: Reflections on the use of photo elicitation in understanding Sri Lankan monastic culture. American Behavioral Scientist 47(12):1528–1550.

Samuels, Jeffery. 2007. When words are not enough: Eliciting children’s experiences of Buddhist monastic life through photographs. In Visual research methods: Image, society, and representation, ed. G. Stanczak, 197–224. Thousand Oaks, CA: SAGE Publications.

Samuels, Jeffery. 2010. Attracting the heart: Social relations and the aesthetics of emotion in Sri Lankan monastic culture. Honolulu, HI: University of Hawaii Press.

Savage, Sara, Sylvia Collins-Mayo, Bob Mayo, and Graham Cray. 2006. Making sense of generation Y: The world view of 15- to 25-year-olds. London: Church House Publishing.

Schnettler, Bernt, and Ju ̈rgen Raab. 2009. Interpretative visual analysis developments: State of the art and pending problems. Historical Social Research 34(2):265–295.

Schwartzenberg, Susan. 2005. The personal archive as historical record. Visual Studies 20(1):70–82.

Stasz, Clarice. 1979. The early history of visual sociology. In Images of information: Still photography in the social sciences, ed. J. Wagner, 119–136. Beverly Hills, CA: SAGE Publications.

Stryker, Roy Emerson, and Nancy Wood. 1973. In this proud land: America 1935–1943 as seen in the FSA photographs. Greenwich, CT: New York Graphic Society.

Suchar, Charles S. 1997. Grounding visual sociology research in shooting scripts. Qualitative Sociology 20(1):33–53.

Thomas, William I., and Dorothy Swaine Thomas. 1928. The child in America: Behavior problems and programs. New York: Alfred A. Knopf.

Vassenden, Anders, and Mette Andersson. 2010. When an image becomes sacred: Photo-elicitation with images of holy books. Visual Studies 25: 149–161.

Vergara, Camilo Jose. 2005. How the other half worships. New Brunswick, NJ: Rutgers University Press.

Wagner, Jon. 2011. Seeing things: Visual research and material culture. In The SAGE handbook of visual research methods, ed. E. Margolis, and L. Pauwels, 72–95. Thousand Oaks, CA: SAGE.

Wang, Caroline, and Mary Ann Burris. 1997. Photovoice: Concept, methodology, and use for participatory needs assessment. Education Health and Behavior 24(3):369–387.

Wiles, Rose, Andrew Clark, and John Prosser. 2011. Visual research ethics at the crossroads. In The SAGE handbook of visual research methods, ed. Eric Margolis, and Luc Pauwels, 685–706. Thousand Oaks, CA: SAGE.

Williams, Roman R. 2010a. God’s global professionals: International students, evangelical Christianity, and the idea of a calling. Unpublished doctoral thesis, Department of Sociology, Boston University, Boston, MA.

Williams, Roman R. 2010b. Space for God: Lived religion at work, home, and play. Sociology of Religion 71:257–279.

Williams, Roman R. 2013. Constructing a calling: The case of evangelical Christian international students in the United States. Sociology of Religion 74(2):254–280.

Williams, Roman R. 2014. Visual Sociology, by Douglas Harper; Doing Visual Research, by Claudia Mitchell, and Advances in Visual Methodology, by Sarah Pink. Sociology of Religion 75(1):167–169.

Leave a Reply