Reflections on Using Participant-Generated,Digital Photo-Elicitation in Research With Young Canadians About Their First Part-Time Jobs

What is Already Known?

This manuscript in relation to what is already known: This paper confirms the value of photo elicitation in conducting research with children and youth.

What This Paper Adds?

This manuscript in relation to what it adds to current understanding: This paper adds to current understanding by outlining changes in photo elicitation practices with the popularization of on-line image options and the increasing prevalence of cell phones.

Introduction

Participant-generated photo-elicitation involves inviting participants to take photographs and then discuss these photos in subsequent one-on-one interviews or in focus groups. Some call this method participant-driven photo-elicitation because in the interviews or focus groups, participants frequently guide the conversation about the images, thus “‘driving’ the interview” (Shaw, 2013). Participant-generated photo-elicitation is attractive for adults interested in research with young people because the influence of the participants appears to offer a counterweight to both the unequal relationship between researcher and researched and the added inequalities between adults and young people (Albanese, 2009; Christensen & James, 2000). For instance, participants might bring content into interviews that is not available to, noticed by, or considered important by adult researchers (Clark-Ibáñez, 2004; Guillermin & Drew, 2010).

Following Mizen, Pole, and Bolton (2001) in their use of participant-generated photo-elicitation in their study of young people’s “light work” ¹ in Britain, the main goal of this research was to deepen our Canadian-based research into young people’s earliest paid work by both prioritizing their viewpoints and foregrounding the materiality and geography of work (Meo, 2010; Mizen, 2005; White, Bushin, Carpena-Méndez, & Ní Laoire, 2010). In this article, we discuss some of the challenges we encountered in gaining institutional ethics approval and organizational support for the use of this research method—challenges that were illuminating in terms of how outside assessors of our methodology perceived risks related to youth, workplaces, photography, and technology. We then discuss how our use of modified participant-generated photo-elicitation unfolded. As we outline these processes and findings, we concentrate especially on how rapidly changing technology shaped the possibilities and challenges of photo-elicitation for us. One of the key issues here is the unanticipated way in which young workers’ selection of existing, publicly accessible Internet images to convey aspects of their work successfully sidestepped concerns raised by the institutional ethics review board. A primary aim of this article is to reflect on the challenges and innovations of our photo-elicitation process to assist others who may opt to use photo-elicitation with young people in the future. We conclude, along with earlier studies, that overall the method seemed to produce more democratized interview dynamics and deeper insights into young people’s early work.

Uses, Strengths, and Weaknesses of Photo-Elicitation

In photo-elicitation, researchers use photographs, supplied by the researcher or the participants in an interview or focus group context, as a tool to foster discussion (Harper, 2002). For instance, Leonard and McKnight (2015) in their interviews with young participants about navigating certain spaces in Belfast introduced researcher-generated images of these urban spaces into the interview with the goal of deepening the discussion. Researchers have given various names to forms of photo-elicitation where the photos are taken by participants rather than researchers and then discussed in an interview or focus group setting, including “auto-driven photo-elicitation” (Mandleco, 2013), “participant-driven photo-elicitation” (Danker, Strnadová, & Cumming, 2017), “photo self-elicitation” (Mizen, 2005), and “participant-generated photo-elicitation (Guillemin & Drew, 2010). In our work, we favor “participant-generated photo-elicitation” as it highlights the use of participant-supplied images. “Photo-voice,” which is sometimes conflated with photo-elicitation, is an approach that uses photo-elicitation but specifically with marginalized groups with the intention of fostering self-representation, empowerment, critical consciousness, and social change in their communities (Guillemin & Drew, 2010; Power, Norman, & Dupré, 2014). Böök and Mykkänen (2014) have called their use of this approach “photonarrative” to highlight the centrality of agency among the participants.

Scholars conducting research with children and youth have been attracted to using photographs in interviews about a wide range of topics and with young people of many different ages (see Mandleco, 2013). One of the attractions of photo-elicitation in various forms is how it seems to allow for greater participant influence over the research dynamics, directions, and outcomes. It therefore appears to offer a partial response to concerns that researchers bring their own interests and agendas to research settings, pursue researcher-derived questions to the neglect of participants’ interests, and/or control the representation of participants through the analysis and presentation of data. Such concerns are of course heightened when researchers and research participants are unequally positioned within wider classed, racialized, and other often intersecting hierarchies. Attempts to partially mitigate age-based hierarchies, in particular, have led many child and youth scholars to adopt forms of research that aim to increase participant input and even control (Albanese, 2009, Christensen & James, 2000), including participant-generated photo-elicitation (Böök & Mykkänen, 2014). In Danker, Strnadová, and Cumming’s (2017) research with young people with autism, for instance, they found that using participant-generated photo-elicitation allowed participants to choose the photos they wanted to take, to choose the ordering in which the photos would be discussed, and to explain their photos as they wished.

In addition to fostering opportunities to participate in shaping our research project, our initial attraction to participant-generated photo-elicitation was inspired by Mizen, Pole, and Bolton’s (2001) research into young people’s early jobs in Britain. Mizen et al. employed a range of data collection techniques, including interviews, group discussions, and photo diaries. For the photo diaries, participants took their own pictures with basic cameras and were encouraged to write reflective comments on six of their pictures, as a form of “photo self-elicitation”; these photos and comments were then discussed in interviews (Mizen, 2005). Mizen (2005) notes that his research team was initially disappointed with the seemingly static nature of the photographs that participants took, with many photographs being of objects and places rather than people and activities. However, the research team found that using these photos as a basis for conversation in the interview setting provided rich insights into participants’ work activities and their significance (Bolton, Pole, & Mizen, 2001).

Others have been similarly drawn to participant-generated photo-elicitation because of its promise of both increasing participant influence over the research process and findings and generating richer data. Meo (2010) used participant-generated photo-elicitation in her study on the relationship between young people’s class-based positioning and their experiences at school in Buenos Aires. Participants took photos with supplied disposable cameras, and then those photos were discussed in an interview. In comparing these photo-elicitation interviews with more conventional ones that she also conducted, Meo found that the former was almost twice as long, deeper, and more participant led, especially with more economically marginalized participants. The interviewees prioritized what they considered to be important and, in this project, had a say into which images would be published in articles. Similarly, in their photo-voice research project with young people about their experiences of living in Western Newfoundland, Power, Norman, and Dupré (2014) found that introducing photos taken by the participants with digital cameras into focus group discussions led to participants sharing feelings about their community that remained unexpressed in earlier focus groups without such photos.

White, Bushin, Carpena-Méndez, and Ní Laoire (2010) similarly incorporated participant-generated photography into their research design, in an effort to allow young participants more control over their research project on migrant children in Ireland. White et al. distributed disposable cameras to their 10- to 12-year-old migrant and nonmigrant participants and then talked with them about the photos in interviews. As with Meo (2010), they found that this technique increased the participants’ ability to introduce issues into the interview exchange and to more independently share narratives about their own lives. White et al. also felt that the greater empowerment of child research participants in this way led to insights that might not otherwise have been accessed. For example, compared to the nonmigrant children in their study, the migrant children took more pictures of nuclear family members and fewer of extended family members who were more likely to be far away. Similarly, in their multimethod study on children’s physical activities, although they did not include an interview component in their use of photos, Darbyshire, Macdougall, and Schiller (2005) found that photography contributed topics such as backyard equipment and interaction with pets and illustrated the exuberance of play. Others have suggested that participant-generated photo-elicitation can also help younger research participants open up about more difficult topics of discussion (Guillermin & Drew, 2010; Mandleco, 2013).

While highlighting the apparent advantages of participant-generated photo-elicitation in terms of both researcher-researched dynamics and content, some researchers have pointed to associated challenges. For example, it has been pointed out that participant-generated photo-elicitation is not a panacea for addressing hierarchies in adult–child research relationships. While the technique can provide child participants with greater voice and influence over the process and content of a research project, it does not fundamentally alter these hierarchies (Böök & Mykkänen, 2014). Also, while participant-generated photo-elicitation may increase young participants’ influence over the content of a given project, Meo (2010) notes that this can be difficult for researchers if the initial focus of the study is derailed as a result of being reshaped by their participants.

This method also raises practical challenges around the use of cameras. Many participant-generated photo-elicitation projects have been done with disposable cameras (e.g., see, Allen, 2009; Meo, 2010; Mizen, 2005; White et al., 2010), meaning that cameras were bought and distributed to the research participants, then collected by the researchers who would get the film processed before returning to the research participants to discuss the photographs. This process is expensive, in terms of having to buy the cameras and develop the film, and time-consuming. Of course, the use of film also means that the participants are limited to a certain number of shots. Sometimes the cameras can be quite precious to the children, and the opportunity to take photographs is exciting, while other times the cameras can be forgotten (Clark, 1999). Mandleco (2013) notes how in her study, participants did not always have their cameras with them when a photo opportunity arose, or in some cases, they took a lot of pictures in one short burst rather than a series of pictures across times and places, which limited the breadth of subjects and activities represented in the photos that the children brought into subsequent interviews.

Meo (2010) and others also make the important point that researchers using participant-generated photos as part of the research data need to be attentive to the fact that photographs are not self-evident representations of a particular “reality.” Photos are created within a certain context, in this case a research context, that shapes what is photographed and how (Power et al., 2014). Some researchers have cautioned that when teaching participants how to use a camera for participant-generated photo-elicitation, the researchers may also inadvertently shape participant understandings of what might be appropriate photos (Power et al., 2014). There are often also conventions around photography, which will vary across time, place, and generation, such as dressing up for photos, taking only “scenic” shots, or creating only photographs that show people being “happy” (Guillermin & Drew, 2010). These conventions may shape what participants produce. To this end, Meo (2010) notes that sometimes it is important to be equally attentive to what is absent as to what is present in a participant-generated photo. Indeed, Böök and Mykkänen (2014) found that sometimes photos that were overexposed, or that did not capture exactly what the photographer wanted, could nonetheless lead to as fruitful discussions as more “successful” photos.

While researchers note some challenges involved with this method, the advantages for our research on children’s first jobs seemed to outweigh the disadvantages and so we decided to proceed with an ethics proposal outlining our planned methodology. As we will highlight, ethics board concerns, gate-keeping, and technological shifts offered new challenges and opportunities.

Navigating Institutional Ethics

For our study, we wanted to focus on young people’s experiences and perceptions of their very early work in Canada, from informal work such as babysitting and snow shoveling to more formal work such as newspaper delivery and fast-food preparation. We planned to use word of mouth and posters to locate young people who were 16 and under and working in their first paid jobs. ² We also proposed a participant-generated photo-elicitation component involving participants taking pictures at work. This proposal was submitted to a university ethics board. ³

Australian researcher Louisa Allen (2009) used participant-generated photo-elicitation for a project focused on how students learn about sexuality at school with the goal of increasing young people’s involvement in, and control over, the research. She asked her Australian participants to take their own pictures of the sexual culture of their schools (as they saw it), creating a photo diary (a collection of photos) over 7 days, and then she engaged them in a photo-elicitation interview about the images they collected. Allen (2009) discusses the protracted, risk-focused, governmental approach that she experienced in the course of having her proposed research reviewed by her university’s ethics review board. Her proposal to offer 16- to 19-year-olds disposable cameras, for example, was framed by the ethics review board as risky and dangerous, both due to her broad research focus on young people and sexuality and due to the construction of young people as incompetent, irresponsible, and at-risk relative to adults. The members of the board were concerned about what the young people might do with the cameras, that young research participants might make irresponsible choices in their photographic choices, and about the vulnerability of those being photographed. Participants were as a result only allowed to take pictures “in places where normal access is granted” and “with identifiable people” (Allen, 2009, p. 406)—this meant that a photograph of body parts, for example, handholding, could not be taken. Participants also had to sign a form saying that they would return the camera after 7 days, keep it safe, and only use it for the purposes of the research. Finally, the board stated that no one under 16 could be photographed without parental permission. As the school where she was doing her research felt this was an untenable option, anyone being photographed by the participants had to be 16 years old or over.

We encountered a similar framing of young people and photography in our own encounter with a university ethics review board. First, asking participants to generate photos was considered more dangerous than asking them to participate in taped interviews, even though photos, like words, are composed, arranged, and posed (White et al., 2010). Feedback from university ethics board members suggested that they felt that photos might reveal information about young people’s work that could potentially get the young participant into trouble with their employers and put us as researchers (and perhaps our universities) into a tricky spot if any problematic workplace issues were photographically (rather than just verbally) revealed.

Mizen (2005) explains how in the research project on young people’s work that he conducted with Bolton and Pole, the team left it up to the young participants to decide whether to take photos at their workplaces or not. Some chose not to attempt this out of concern that they might jeopardize their job by doing so. Others who attempted the exercise experienced a range of responses from employers. Some found that their employers helped out and even wanted to be photographed, while others found that their employers would not allow them to proceed. In the latter case, some still proceeded to take photos surreptitiously.

In our case, this kind of judgment was not left to the young participants. Our original proposal was that our participants would be invited to take pictures at work after gaining written permission from their employer to do so. The ethics board however determined that this was too risky because divulging to an employer that they were participants in a research project on work might result in being fired. We revised our design to stipulate that participants could take pictures about their work, but not “at work.” As it turned out, while participants tended to follow this guideline, several in our project took pictures on their newspaper delivery routes and while babysitting anyway. These provided important insights into how participants understood taking a picture “at work” and their own understandings of risk. Taking photographs of streets where they delivered newspapers, their newspaper carts, and children’s toys in homes where they were babysitting, for example, clearly did not seem “risky” to them.

In our proposal, we also suggested that participants could have the option of taking pictures either with their cell phones or with disposable cameras. Similar to Allen’s research ethics review, members of the ethics board were concerned about what the young workers might do with the pictures they took, a concern that was heightened by the idea that they could easily post images online. Ethics board members were concerned that participants might post images on Facebook, for example, in ways that might undermine confidentiality and also publicly implicate the university in unanticipated ways. These concerns seemed to problematically suggest that young people are irresponsible, particularly when it comes to their use of social media. It was thus suggested that it might be better to have all participants use disposable cameras, so that the pictures would then be retained/controlled by the researchers. As the latter was only a suggestion, however, and because participants could upload pictures to social media platforms from their phones whenever they wished anyway, we kept to the original plan for the young workers to use their own cell phones if they had them. In the end, all participants took photos with either their cell phones or tablets.

Based on having consulted with one of our university ethics boards about uses of photography in research, prior to submitting our ethics application, we said that if participants wanted to take pictures of other people, they could only do so if any recognizable individuals being photographed signed a permission form. Also based on ethics board guidance, we asked participants to sign a form saying that they would follow certain protocols for taking pictures, that is, they would not take pictures at work, they would not take photos considered “inappropriate” or “offensive,” they would not upload photos to social media in order to keep their participation in the project confidential, and if they were under 14, they would obtain their parent(s) signature on this form as well.

In sum, as with Allen’s research experience, we found that our proposal to introduce photos taken by young people into the research project raised more concerns than exclusively talk-based interviews because of apprehension on the part of the ethics reviewers about what young people might photograph and what they might do with those photographs. Young people were considered particularly vulnerable (specifically to losing their jobs), and, particularly likely to misuse their photos.

Our Data Collection

We interviewed participants in three separate Canadian cities in two different provinces, Ontario and British Columbia. ⁴ One author conducted all the interviews in two of the cities; another author conducted all of the interviews in the third city. Participants were located through word of mouth and advertising on posters; all participants aged 16 and under had parental permission to be involved in the study. First, we conducted open-ended, qualitative interviews with participants about their work. Participants were then invited to take photographs related to their work but not at work and to bring their photographs to a second, follow-up interview for discussion. For the photo-elicitation component, we advised participants that we were looking for:

Anything that reflects your life as a young person who has a job. It could be getting ready for work or coming home from work. It could be of tools or things you use to do your work. It could be something you bought with the money you earned at work. Most of all, we want you to be creative and take photos that will give us lots to talk about.

As it turned out, most interview participants did not participate in the follow-up portion. Challenges with recruitment for the follow-up portion included the fact that the research project in British Columbia ran into March Break, when many were away with family, and the fact that in one of the cities in Ontario, participants recruited through a child and youth service organization were difficult to recontact. Indeed, our relationship with this child and youth service organization led to another challenge related to gate keeping. While this organization generously agreed to notify their young members about the research project, to provide interview space, and to set up interviews, out of concern for exposing the young people to risk, they did not allow us to contact the participants directly at any point in our research process (despite us meeting the requirement of a police check). As a result, after the first interview, we were not able to directly communicate with the participants to give them detailed information about the photography component. We also had to wait for the organization to contact the participants a second time and to set up the follow-up interview; later, when some of the participants had photographs on their phones but did not send them to us, we were not able to follow-up with the participants directly to request the pictures.

In what follows, we focus on the 12 of the total 32 participants that participated in the follow-up, photo-elicitation component. Of these, there was a range of apparent engagement, with some appearing very keen and others more neutral about their involvement. Some also seemed perplexed and challenged by the rule that they could not take pictures while at work itself. The number of photos they each brought to their follow-up interview, on their cell phones, tablets, or computers (to which cell phone pictures had been transferred), ranged from 3 to over 12. Of the 80 photos that were collected for the project, 56 were participant generated, and 24 were participant selected, publicly accessible Internet images (discussed further below). Overall, we were pleased with the way photo-elicitation seemed to both democratize the interview dynamics and deepen the content of the interviews. However, we also encountered some unanticipated challenges related to using digital rather than film photography.

Digital Photos

As we have discussed, many accounts of participant-generated photo-elicitation come from an era when the new technology of cheap digital and disposable cameras offered an accessible way to facilitate participants taking photos that could then be shared with researchers. In the current research context, however, cell phones with high-quality cameras are ubiquitous among young people in the Global North, making disposable cameras almost obsolete. This reality offers both opportunities and challenges. In terms of opportunities, young people are now more likely to be familiar with how to take photographs (e.g., Danker et al., 2017). Further, the use of cell phones means that researchers do not have to worry about supplying and then retrieving the cameras and developing the film. Young people also tend to have their cell phones with them most of the time, making it easy to spontaneously take a picture of anything they feel is relevant to the research. In terms of the subsequent interview dynamics, the use of participant controlled cell phones, tablets, and computers appeared to allow greater freedom for participants to decide which photos they would like to share and the order in which they would be viewed. For instance, photos could be easily retaken and deleted, allowing young people to control self-directed interviews in new ways not previously reported upon or experienced by other researchers. For those who then transferred their photos from their phones to a tablet or computer, they were then able to review, select, and reorder them.

Yet we also faced some challenges with the use of cell phones and tablets for photography. As all participants used their cell phone or tablet cameras, we have wondered whether some of the participants might have devoted more time to the task and generated more pictures if given a disposable camera which would have been considered more unusual or special, and with which they would had only a limited number of shots. Also, because participants brought their pictures to the interview on their cell phones, tablets, or laptops, we faced a new challenge of getting the images from the participants after the follow-up interview had occurred. While most eventually e-mailed their pictures to us, there are some that we never received, which was disappointing. This technical issue was partly linked to gate keeping on the part of the child and youth service organization discussed above, which prevented us from fully explaining the photo-elicitation process. More broadly, this issue is a reminder that researchers need to provide participants with direct guidance, from the start, about how best to copy images to the researchers. For instance, Danker, Strnadová and Cumming (2017) avoided these challenges by providing participants with detailed instructions on how to transfer their images to a USB key that was then given to the researchers. Finally, without handheld photos, we were not able to engage in processes described by Clark (1999) who asked participants to organize their photos, as “tangible props” into piles, for example, of good and bad experiences and also suggested that certain pictures could be put aside to be talked about later. These strategies would be more cumbersome if pursued with photos on a phone, tablet, or computer screen.

While the above relates to the processes involved in participants taking their own pictures, we found that there was another way in which participants “generated” images that they brought to the interview setting—through accessing publicly accessible Internet images. One of our early participants cleverly decided that as an alternative to taking a photo within her actual workplace, she would instead include pictures of her workplace that she found online. This strategy prevented participants from having to tell their employers that they were participating in research and thus opened up a compelling new avenue for our research. There is now a proliferation of images online, and they are easy to download, print, and/or send. Many of our participants worked in businesses that had a number of related images online, including both large chains and small, local businesses. Inspired by this unanticipated innovation in terms of how photos could be “participant generated,” we subsequently suggested this possibility as another way in which other participants could supply work-related images for the follow-up interview if they wished. Various participants did this, presenting us with a combination of their own pictures and publicly accessible Internet images of their own workplaces or those that were similar to their own. The images that they brought reflected their thinking about what they wanted to share and talk about. With the photos as prompts, interviewees introduced topics such as workplace machinery and tools (e.g., fryers and drive-through headsets), workplace layouts (e.g., where employees entered a building, details of staff rooms), and employers’ marketing campaigns. While this photo-elicitation strategy raises some new issues (like copyright in displaying these pictures, and participant confidentiality ⁵ ) and moves away from the unique, specific, and artistic photos that participants might take, the Internet-derived photos seemed thoughtfully chosen and were exciting and fruitful in terms of allowing participants to share images of places that the ethics review had determined they could not photograph.

Participant Control and Depth of Discussion

Overall, we found that the photo-elicitation component of our research added significantly to our project, reflecting others’ work with photo-elicitation. In our interviews, the use of photo-elicitation (1) generally helped to democratize the interview dynamics and (2) often opened up room for unanticipated topics and more detailed descriptions and stories.

First, as others have found using photo-elicitation with young people, the dynamics of the interviews sometimes changed significantly, and positively, with the follow-up interview. This change that we noticed may in part be simply about doing a second interview, when a relationship has already been established. However, there were moments when the sharing of photos clearly had a role in positively changing the dynamics. It was their pictures that participants were bringing in, and they knew what they wanted to say about them, which seemed to bring the participants confidence and increased engagement. For instance, for their follow-up interviews, sisters Taylor and Charlotte had set up a different room in their home from where we had interviewed the first time, with two chairs side by side, facing a propped up tablet. Their pictures were already loaded onto the tablet in the order that they wanted them, and they each determined the pace at which we went through their pictures. As well, as we viewed and discussed the photos, we as interviewers sometimes found ourselves to be more relaxed and curious, as we were responding to new images that we saw rather than following an interview guide prepared in advance.

Second, the follow-up interviews opened up room for more detailed descriptions of young people’s work experiences, leading to deeper and broader conversations. Many of the participants went into much more detail than before about exactly how they did their jobs, pointing out parts of the pictures that added to this detail. The photos prompted more questions about their jobs from us as well. For example, Amanda brought in pictures illustrating incentives used at her workplace, such as stickers and swag used as rewards, which led to a rich conversation about how effective such incentives are, how rigorously the reward system was followed, and whether any employees tended to get favored, all topics that had not emerged in the initial interview.

In another interview, the photo-elicitation component included Zach’s deeper discussion of a subject that he had raised more briefly in the first interview. Zach had worked for a number of months at a fast-food restaurant and while he had appreciated the resulting income, he had recently quit because it was interfering with school. In the first interview, Zach had talked about how earning money allowed him to spend some of it on friends. He explained:

my friend at the time, he didn’t have a job […] he had, like, he didn’t have money to do stuff and he’s like “well let’s go see Star Wars on opening night and I’m like, “okay” and so I bought the tickets for both of us (hmmm) ⁶ and then now he paid me back once he has a job and I don’t (right [laughs] so it worked out) kind of stuff like that (yeah) but [it was] not really that, like I, “I need all this money so I can go out and impress my friends” [because] I don’t.

At the second interview, among his pictures were photographs of items that he had purchased with his wages, including computer equipment and a Star Wars movie ticket (Figure 1). In his discussion of the latter, he elaborated on his earlier reference to buying the Star Wars movie ticket for his friend: Zach:

This was just my Star Wars ticket [interviewer laughs] which is like, but it was like a pretty big moment for me (that’s awesome) cuz it was the first time I bought movie tickets and I bought two, one for me and one for my friend Gavin who didn’t have a job at the time […]

I:

…did the movie feel different knowing that you’d sort of earned the money to (kind of, yeah) watch it? Like how?

Zach:

Like, I, I don’t, I don’t even know if it did or not honestly but it was just such a fun experience (right) that I don’t really care who bought them, maybe. It was more like “I earned this” like (right) I, I can, I don’t know, yeah maybe it was more of something like that where I was kind of just like “I earned this, I, I’m able to do this.”

OPEN IN VIEWER

Figure 1.

Star Wars movie ticket (Zach).

In this commentary, Zach speaks more about the emotional aspects of earning and spending money from his first job, which offers us as researchers deeper insights into the meanings of early paid work for young people.

Another powerful example of how photographs altered the interview dynamics and produced more in-depth, detailed interview content was evident in the follow-up interview with Alexander. The first time Rebecca interviewed Alexander in a noisy Tim Hortons, they had a good interview, although the brevity of many of Alexander’s answers meant that Rebecca was frequently encouraging him to go into more detail. He did not seem comfortable in the interview and did not provide many examples of the points he was trying to make.

Rebecca learned that he had worked seasonally at a pumpkin farm and also delivered papers regularly. In terms of the latter, a few years earlier he had switched from delivering a weekly, free paper, to the more secure and somewhat more prestigious job of delivering a daily paper that required a subscription. In the first interview, Rebecca had asked if he had developed any relationships with the customers of the latter daily paper and he had replied positively saying,

…like one of the things that happens when you’re delivering papers is that customers tend to start talking with you (hmmm). You tend to be one person that they can strike a conversation up with (mhmm) so that you definitely form relationships.

This was an interesting and valuable response, but in the context of discussing photos during the second interview, a conversation about newspaper delivery was more comfortable and led to much more information being shared.

When Rebecca and Alexander met the second time, Alexander was a tiny bit late. He apologized for the delay and sat down across from Rebecca, his phone already out so he could show her pictures that he had gathered. He leaned in toward her in a way that he had not done before, inviting her to look at the pictures too. The photos of the pumpkin farm, including various family activities that could be pursued there before buying a pumpkin, were all from the Internet, and he also had snapped two pictures from his paper route.

Alexander had specific things that he wanted to say about each photograph, and he guided Rebecca through them, providing many details about his work as he went. In response to the images, Rebecca asked spontaneous, unscripted questions that seemed to animate Alexander in a way that the more formal predetermined questions of the first interview had not.

Eventually, they moved from the Internet images of the pumpkin patch to the two he had taken from his paper route. The first was an image of a newspaper and mailbox (Figure 2).

So this is basically putting a newspaper in the (oh nice), there, so it’s not like you can just, not [like] back in the movies where you can just ride your bike and throw it up (right) and try and hit the porch and hit the window, [and say] “oh whatever.” No, you have to walk up to every mailbox and there’s sometimes people [who] will come out and talk to you. I didn’t get anybody on my route [today] so no pictures (right) of them but when I’m putting it in there they’ll come out sometimes and I’ll talk to them and we’ll have good chats.

OPEN IN VIEWER

Figure 2.

Delivering newspapers (Alexander).

The style of talking shifted here toward both Alexander and Rebecca being friendlier and more casual with one another and with Alexander sharing the kind of description that more fully conveyed aspects of the work of newspaper delivery. Rather than the adult researcher going through preestablished questions in sequence, as happened in Interview #1, the image led to a more free flowing and deeper conversation. In this case, the conversation about this single photo moved on to cover information about whether he received monetary tips from these customers, how often these were offered, and how much people gave, as well as details about the challenges of locating mailboxes at rich people’s houses and how these mailboxes were often different from the ones in his previous, less wealthy, neighborhood; how he felt about the safety of his route; and a very detailed story about how one time he was worried that a pizza delivery man was following him. A single photograph thus led to more spontaneous interviewer–interviewee interactions as well as much deeper detail about work activities and their associated emotions.

While participant-generated photo-elicitation was very successful with Alexander and many others, in some cases, the responses in the follow-up interview were actually shorter, particularly in the data gathered through the child and youth service organization that strictly limited communication between ourselves and the participants. As we were prevented from fully describing the photo-elicitation component, the participants may not have completely understood the reasons for the researcher’s interest in their photographs. If so, this points to the importance of researchers communicating the purpose of participant-generated photo-elicitation as clearly as possible in order for it to be effective. It may also be the case, however, that some participants simply felt that they did not have a lot more to say and as a result, the photos did not elicit many additional details. Another possibility is that for some research participants who produce photos, it may be that simply producing the photo feels like sufficient engagement, as the photos may be thought to “speak for themselves,” a challenge that some other researchers have found in using photo-elicitation with young people (see Böök & Mykkänen, 2014; Meo, 2010).

Finally, the photographs uniquely provided us with insight into our project on children’s earliest work by highlighting material culture and mobility. As also observed by Mizen (2005), many of the images our participants brought to the follow-up interviews, both that they took and that came from the Internet, were of work-related objects (e.g. a lawn mower, a backpack used for walking to work, or a set of story books for babysitting). These images became the basis for detailed accounts of work activities. At first, these kinds of photos seemed less interesting to us, with little indication of activity or relationships, but in the interviews, they evoked details about exactly what the participants did when they were working. For example, a picture of a shovel (Figure 3) led to a detailed conversation with Bob about his snow-shoveling work, for example, that he had been shoveling his neighbor’s deck and driveway in response to requests for several years, that he would go over to shovel when his neighbor came over and asked him to, that he used a particular shovel from his own house to do this work because it had a bigger scoop, and that the work was sometimes quite tiring.

OPEN IN VIEWER

Figure 3.

Snow shovel (Bob).

In addition, perhaps because the participants were proscribed from bringing photos taken at their actual workplaces, there were an unexpected number of photos that were linked to getting to and from work (e.g., pictures of the steps into a bus, a sidewalk, a city map, and in one case, a picture taken while driving past a workplace at night to illustrate how isolated and hard to access the workplace was; Figure 4). These images led to unanticipated interview discussions about the challenges of work-related transportation when you are a young person who cannot drive in a suburban or semirural environment.

OPEN IN VIEWER

Figure 4.

Workplace at night (Amanda).

Other important issues raised and/or developed more deeply as a result of the particular images brought by the participants included the topic of workplace uniforms. When participants supplied photos of their uniforms, this led to conversations about workplace rules surrounding the use and care of this clothing. Conversations about rules, regulations, and politics at work more broadly were also precipitated. For example, when discussing an Internet-derived picture of the front of her actual workplace, Amanda raised the issue of only being allowed to enter through certain doors at work (different ones than the customers), and Charlotte’s own photograph of her management training duo-tang led to a significant discussion of workplace stations, hierarchies, and training patterns. Internet images of fryers and other equipment brought by others led to more detailed discussions about workplace health and safety issues. Finally, Internet images of work-related objects similar to those used at their workplaces, such as drive-through windows, headsets, cash registers, and amusement park activities from their actual or similar workplaces, led to many more stories about encounters with clients and customers than had occurred in our first interviews.

The introduction of pictures, whether taken by the participants or found by them on the Internet, thus frequently altered the power dynamics in ways that seemed to empower the participants, deepened the details of our discussions, and brought forward into the interviews the relevance of mobility, as well as other topics that we had not at first asked, or heard significant details about.

Conclusion

Our research experience is consistent with the literature on participant-generated photo-elicitation: for the most part, this approach added important democratizing dimensions to our interview dynamics with young workers and led to a valuable depth of detail, and a breadth of topics specific to our study of young workers’ jobs. Like other researchers who have used this method, however, we encountered requests to rework our research design to reduce perceived risks associated with our proposed combination of young people and photography. This is not a reason to abandon photo-elicitation but indicates the need to carefully attend to and sometimes challenge perceived risks associated with young people’s generation of images.

One of the newer features of our research project relative to those that have been previously reported on was that the ubiquity of cell phone ownership and usage among our participants opened up opportunities around participant ease of taking photos, suggesting the current accessibility of photo-elicitation to many. We have also realized, however, that it is necessary to ensure research procedures include consistent retrieval of participants’ digital images. The use of participant initiated provision of publicly accessible photos from the Internet was another innovation that we were able to quickly adopt in ways that were beneficial to the project and would likely be useful to others. We conclude that participant-generated photo-elicitation continues to have value as a method for researchers even as the ways in which it is carried out reflect changing technological realities and possibilities.