Abstract
The purpose of the current study was to use Orsmond and Cohn’s feasibility framework to compare two methods of collecting photo elicitation interviews: (1) in-person and (2) online among low-income community college students. We described the feasibility of the recruitment and retention procedures and compared the participants’ characteristics and the type of data obtained by data collection modality. Focus group participants (n = 34) were invited to participate in photo elicitation interviews regarding barriers to food access and associated material hardships. Prior to the pandemic, photo elicitation interviews were conducted in-person. Due to pandemic-related stay-at-home policies, photo elicitation interviews shifted to a video conferencing platform. Descriptive and bivariate analyses were used to compare the two data collection methods in terms of sample characteristics, the average length of each interview, and the number and type of photos submitted. Of 34 recruited participants, 28 participated in photo elicitation interviews (82.4% response rate; n = 13 in-person; n = 15 online) with a mean age of 39.75 years (range: 19–62). No significant differences were found in socio-demographic characteristics between interview modality groups. In both online and in-person photo elicitation groups, participants were more likely to be female, non-Hispanic Black, single, and unemployed. Overall, both modalities were found to be feasible to recruit and retain participants and collect data from low-income, community college students. We found similar data in terms of photographs, and similar codes and themes were generated from the interview data across both modalities. Using both modalities provided an opportunity to promote equity in research through inclusion of hard-to-reach populations who may experience barriers to participation such as transportation or childcare. This study can inform recruitment and retention efforts for quantitative and qualitative research, and data collection efforts for photo elicitation interviews. The lessons learned are critical to generating rich data and advancing research conducted among vulnerable populations.
Keywords
Introduction
Photo elicitation is a qualitative research technique that uses photographs to evoke discussions in an individual interview (Thomas, 2009). The photos are the focus of the conversation rather than interviewer questions (Harper, 1987). The two main types of photo elicitation are based on photograph source: researcher-driven (participants evaluate and talk about photos provided by the researcher) and participant-driven (participants select the photos to be used in the interview) (Van Auken et al., 2010). The use of photos facilitates greater interactions and in-depth conversations between researcher and participants compared to interviews without photos (Bignante, 2010; Collier, 1957). Allowing participants to select photos to guide the interview is an opportunity for participants to be more involved in data generation. Contrary to traditional research methods, the participant takes the lead role (Harper, 1987), which could reduce the barriers between the researcher and the participant (Van Auken et al., 2010). This method allows researchers to give precedence to the themes that matter to participants rather than researchers imposing assumptions, frameworks, or perceptions (Holloway & Galvin, 2016). This allows the participant to describe the situation or event on his/her own terms, and not the terms that researchers want them to use (Holloway & Galvin, 2016). This can turn a vulnerable situation into an empowering situation for the participant (Holloway & Galvin, 2016; Wilson et al., 2015). Thus, photo elicitation is an appropriate methodology for studies involving vulnerable and hard-to-reach populations (Harper, 2002) and to explore sensitive research questions, such as food insecurity and related material hardships.
Photo elicitation has been adapted in multiple studies evaluating food-related issues including food insecurity, eating habits, and food choices (Daugherty et al., 2019; Green et al., 2021; Johnson et al., 2010, 2011; Maley et al., 2010; McClain et al., 2019). Daugherty et al. (2019) used photo elicitation to explore food insecurity and use of a campus food pantry among college students at a rural four-year public higher education institution. Participants in this study discussed how they prioritized their budget, use of food pantries, and having enough food to sustain themselves. Typically, photo elicitation interviews are conducted as in-person interviews (Copes et al., 2018; Green et al., 2021; Johnson et al., 2011). However, photo elicitation interviews have also been conducted online via video conferencing platforms or as telephone conversations (Craig et al., 2020; Daugherty et al., 2019). Daugherty et al. (2019) conducted the initial interview online and the follow-up photo elicitation interview either over the phone or via video conferencing platform. Craig et al. (2020) conducted photo elicitation solely using a video conferencing platform. Conducting interviews online could reduce financial, geographic, and physical mobility barriers among disadvantaged populations wanting to participate in research (Eigege et al., 2022). Craig et al. (2020) reported that rapport was built more rapidly during online photo elicitation interviews compared to in-person interviews.
However, to date there are no studies comparing the two modes of photo elicitation interviews (in-person vs. online) for accessing food-related issues among hard-to-reach populations. Feasibility testing is an important step in intervention and study development to ensure good stewardship of participant time and trust, as well as funding resources (Orsmond & Cohn, 2015). Orsmond and Cohn (2015) propose a framework for feasibility studies to ultimately evaluate if an intervention or study can be successfully conducted within a target population in a specified setting. This framework breaks feasibility down into five major components, assessing the primary aspects of: recruitment and sample characteristics; data collection procedures and outcome measures; acceptability and suitability of intervention and study procedures; resources and ability to manage study; and preliminary evaluation of participant responses. This framework is a guide, and not all aspects of the framework need to be evaluated to establish feasibility (Gadke et al., 2021). In this paper, we focused specifically on recruitment, retention, and data collection procedures.
The purpose of the current study was to use the framework outlined by Orsmond and Cohn (2015) to evaluate the feasibility of using photo elicitation to explore food insecurity and related material hardships among low-income, community college students using two modes to facilitate interviews: (1) in-person and (2) online. Specifically, we described the feasibility of the recruitment and retention procedures and then we compared the sample characteristics of participants and the type of data obtained by data collection modality. From this feasibility assessment, we described lessons learned from conducting in-person and online photo elicitation interviews among low-income, community college students.
Methods
Research Team
The research team consisted of six members. The team was led by a PhD-level Principal Investigator (PI) experienced in the topic of food insecurity and related material hardships who designed the parent study and corresponding qualitative follow-up studies. The PI was also responsible for obtaining the necessary funding to support all the research activities. The PI, along with the Co-Investigator (Co-I), had qualitative expertise. The rest of the research team consisted of trainees: one postdoctoral research scholar, one doctoral student, one masters level student, and one undergraduate student. Research team participation in project activities was as follows: all photo elicitation interviews were conducted by the PI; all text messaging and communication to maintain participant engagement was conducted by the PI; the postdoctoral scholar downloaded, organized, and managed all photographic data received. The coding was led by the Co-I and included the four trainees. Thematic analysis involved the entire research team.
Recruitment
This study was conducted with a subsample of participants that were a part of a parent study which evaluated an on-campus food distribution program that occurred at a large, urban community college system in the South (Hernandez et al., 2021). Specifically, a sub-sample of participants engaged in focus groups for the purposes of understanding barriers and facilitators to participating in the program (Eigege et al., 2022). At the end of the focus groups, participants (n = 34) were then invited to participate in photo elicitation interviews (January 2020 – April 2020). Prior to leaving the focus group, all focus group participants were immediately compensated for their participation in the focus group. Participants were provided a physical gift card for in-person focus groups and an electronic gift card for online focus groups.
Retention
Several retention strategies were implemented. Establishing means of communication. All focus group participants were emailed the photo elicitation interview prompts and corresponding instructions and were required to verbally confirm receipt of the email to the research team prior to leaving the interview room/disconnecting from the online platform. Immediate scheduling. The majority of the focus group participants were scheduled for their photo elicitation interview prior to leaving the focus group. Most photo elicitation interviews were scheduled within one week of the focus group. Multiple reminders. Potential photo elicitation participants were sent text reminders three times per week about their photography task and upcoming scheduled interview. In addition, participants were sent a reminder text the morning of the interview regarding the time and location (in-person/online). Consistent facilitator of qualitative interviews. The PI conducted both the focus groups and the photo elicitation interviews. This was done to increase trust and to ensure that the photo elicitation experience was empowering. Further, all scheduling and follow up communications were conducted by the PI in an effort to promote rapport-building and maintain continuity.
Data Collection
Participants in the photo elicitation interviews were invited to submit at least six photos documenting or describing barriers to food access and associated material hardships. Participants were offered hand-held digital cameras as this is standard practice for photographic data collection. Providing hand-held digital cameras provides equal opportunities for participants who may not have access to a camera phone. However, all participants preferred to use the camera on their cell phones. Formal training on how to use the cell phone camera was not needed as everyone in the group had taken photos with their cell phone previously. To facilitate photo selection, six prompts were provided with the intention of guiding the participants and to encourage them to think creatively about the topic. The prompts were not a list of questions required to be answered nor was it expected that participants would provide a photo for each prompt. The following prompts were provided to interested potential participants: 1. What is it like to be a college student who does not have enough food to eat/difficulty accessing food? 2. What in your life or community has helped you overcome food shortages? 3. What in your life or community has slowed you down in overcoming food shortages? 4. What feelings or metaphors do you want to tell/show other people about the difficulty of accessing food? 5. What are your solutions to improve food access/the Food Scholarship Program? 6. What are your hopes for the future? And what might help you get there?
Participants were advised not to share photos of others without consent due to ethical concerns, and to avoid use of stock photos. The purpose of the photos, the prompts, and the instructions regarding the type of photos (e.g., avoid stock photos) were also emailed to the participants. Participants sent their photos to the designated research team member by email prior to the day of the interview. Specifically, participants sent one email which included all of their photos. For security and confidentiality purposes, participants were instructed not to take any photographs showing personal identifying information, or any images of people’s faces. After receipt of photos, the postdoctoral scholar confirmed the receipt of the photos through email and reminded participants of the date and time of their upcoming photo elicitation interview. The postdoctoral scholar then labeled and organized all photos onto a secure drive that only the research team had access to.
All photo elicitation interviews were conducted by the PI. To facilitate discussions over the photos, interviews were conducted according to the mnemonic PHOTO (Hergenrather et al., 2009). The PHOTO approach eased discussion through five questions: 1. Describe your
Prior to the pandemic, photo elicitation interviews were conducted in-person, in a private conference room on a university campus (January 2020 to mid-March 2020). Due to pandemic-related stay-at-home policies, photo elicitation interviews were conducted using a video conferencing platform (mid-March through April 2020). All photo elicitation interviews were conducted in the same modality as the focus groups in which the participants had participated (i.e. either in-person or online). This provided familiarity with the data collection process and helped build rapport. The data collection was approved by the Institutional Review Board at the University of Houston; data analysis was approved by the Institutional Review Board at the University of Texas Health Science Center - Houston. All participants completed informed written (in-person) and verbal (online) consent, and all research was in accordance with the Declaration of Helsinki. Participants were compensated with a $30 physical (in-person) or electronic (online) gift card to a national discounted department and grocery store for their time and effort. A parking voucher was also made available for the in-person interviews.
Measures
Recruitment and Retention Procedures
The number of individuals that participated in the photo elicitation interviews compared to the number of individuals invited to participate in the interviews was recorded to calculate the response rate. We also recorded any complaints regarding the retention strategies that were implemented.
Sample Characteristics
Socio-demographic characteristics of the individuals that participated in the photo elicitation interviews were extracted from the parent study (Hernandez et al., 2021). As part of the parent study, sociodemographic data were collected through administrative review of community college records. Sociodemographic data collected include: age, gender (male, female), race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, other), marital status (married, unmarried: divorced/separated/single), and academic level (freshman, sophomore, junior, senior, graduate degree, unclassified), and employment status (full-time, part-time, not employed). Surveys were also used to collect self-reported information on the number of adults and children in the household.
Data Collection
To evaluate data collection procedures, the average number of photo elicitation interviews per week, the length of each interview based on the audio recordings, and the number of photos submitted per participants were documented per data collection modality (in-person vs. online). We also recorded participants’ perceived burdens and challenges in data collection.
Analysis
The sample characteristics, the average length of each interview based on audio recordings, and the number of photos submitted per participant were evaluated between modality of data collection using independent t-tests for continuous variable and proportion test for categorical variables. Coding and thematic analysis of photos was conducted during regular data analysis meetings. No software was used for coding and analysis of photos.
Results
Recruitment and Retention Procedures
Of 34 focus group participants, 28 participants agree to participate, and all 28 who agreed to participate in photo elicitation interviews took and submitted photos and subsequently completed the interview (82.4% response rate; n = 13 in-person; n = 15 online). Due to the high response rate and lack of reports indicating the retention strategies to be burdensome, it was concluded that the retention procedures were acceptable.
Sample Characteristics
Descriptive Statistics of the Socio-Demographic Characteristics of the Low-Income, Community College Students Who Completed the Photo Elicitation Interviews, by the Total Sample and by Interview Method (n = 28), Mean (SD) [Range] or Number (%).
Note. No significant differences on socio-demographic characteristics by interview modality. SD, Standard Deviation
Data Collection
Example Photos From Each Modality.
Average Audio Minutes of Photo Elicitation Interview and Average Number of Photos for the Total Sample and by Data Collection Method, Mean (SD) [Range].
Note. SD, Standard Deviation.
aNo significant differences on length of interview by interview modality [t (26) = 1.38, p = .18].
bSignificant differences on the total number of digital and stock photos by interview modality [t (26) = 1.72, p < .05].
cNo significant differences on number of digital photos by interview modality. [t (26) = 1.56, p = .07].
dNo significant differences on number of stock photos by interview modality. [t (26) = −.31, p = .62].
Conclusions
Lessons Learned From Conducting Photo Elicitation Interviews In-Person and Online Based on the Orsmond and Cohn (2015) Feasibility Framework.
A high response rate was achieved in this study. It is believed that the comprehensive, intentional recruitment and retention procedures contributed to the 82% response rate. Specifically, participants were provided immediate compensation at the completion of the focus group that consequently served as a recruitment method for the upcoming photo elicitation interview. We also took advantage of having a captive audience, whether in-person or online, to do two things. First, while the participants were still present, we set up the next interview, rather than waiting for participants to call or email the research team. Having individuals schedule their upcoming photo elicitation interview in the presence of others motivated and encouraged others to schedule their appointment to ensure they received the time slot that was best for them. Second, we ensured the research team had correct, active email accounts for the participants by sending an email with photo elicitation interview prompts and corresponding instructions while the participants were still present in the room/on screen. Further, participants had to confirm receipt of the email and this verbal confirmation prevented future miscommunication. To retain participants in the research study, we believe it was important to continue engagement with them every other day up until the morning of the photo elicitation interview. In addition to the frequency of communication, the use of multiple methods of communication (i.e. Email and text) assisted with the photo elicitation interview being a top priority for the participants.
The sample characteristics of the participants that participated in-person compared to those that participated online did not differ in terms of age, race/ethnicity, academic level, employment status, marital status, number of adults in household, or number of children in household. Further, the length of the audio recordings corresponding to the online photo elicitation interviews was longer, yet not statistically significant. The lack of differences in participants’ sample characteristics and the length of the interview could be related to homogeneity of the sample from which recruitment occurred. All the participants were engaged students who were interested in discussing their barriers to food access and other related material hardships, regardless of the data collection mode. This may have differed if the project recruited from a larger sample of community college students.
The task of taking photographs of their hardship-related experiences was found to not be burdensome for the participants; however, we did find that we needed to help direct the attention of the participants in both modality groups. Although participants were provided six prompts, we found that it has helpful to some participants if we selected 1–3 prompts for them to focus on. The selection of the prompts was based on what they had discussed in the focus group, and this made the task more manageable for some participants. Related, for some participants, taking and then sharing a photo of their hardship experiences was overwhelming. To make the concept concrete, participants found it helpful to draw upon what they had previously stated in the focus groups. The lead researcher of the photo elicitation interviews would remind participants of the examples that they themselves had just provided in the focus group. This helped participants understand the task and made the concept of photography more tangible as a mechanism to capture their experiences. Participants who were most excited about this task could immediately think of photographs that they wanted to share in their upcoming photo elicitation interview. While participants were asked to not submit stock photos, a limited number of stock photos were received. Although stock photos are not able to be used for publication purposes, they were helpful in collecting interview data from participants. The use of stock photos could potentially be an indicator that participants may not understand the meaning of “stock photos”. For some participants, there may have been concern that they would not be able to meet the photograph quota of six photos without using stock photos. For other participants, the topic of sharing photos of their living environment could have felt embarrassing, and the use of stock photos might have felt safer or less vulnerable. Thus, stock photographs could have provided participants anonymity. The information gathered from the stock photos did contribute to the richness of the data, and the use of stock photos did not differ significantly between groups.
Although the in-person group submitted on average 1.5 more photos than the online group, submission of more photos did not result in more of the given prompts being addressed. Nor did the submission of more photos by the in-person group result in additional topics being discussed and consequently generating more codes and themes. The additional photos were instead related to some participants submitting near duplicates of the same photo (i.e., two angles of the same person or activity used to discuss a single topic). In reviewing the data analysis, near duplicate photos were coded the same and resulted in the same theme. Ultimately, in our study, more photos did not generate additional or richer data.
In the end, conducting in-person and online photo elicitation interviews provided the opportunity for more individuals to participate, resulting in richer and more equitable data collection through inclusion in research. The lessons learned from the procedures that were implemented can be applied to a wide range of research studies focused on vulnerable populations to obtain a high response rate and rich data. Inclusion of the perspectives of hard-to-reach populations in research is important to inform appropriate, culturally-relevant intervention programming and policies. In trauma-informed care for vulnerable populations, it is important for people to feel a sense of choice and control in sharing their stories (Wilson et al., 2015). This agency and degree of involvement of hard-to-reach and otherwise vulnerable populations in all aspects of research is critical for equity. The participant-driven nature of photo elicitation interviews is especially well-suited to involving hard-to-reach populations in research, resulting in rich firsthand insight that is not available through quantitative research. Qualitative research that includes hard-to-reach populations should integrate both in-person and online modalities to promote equity and inclusion and reduce barrier to participation.
Footnotes
Author contributions
DCH conceived the design of the study and supervised all aspects of the study reported in this manuscript. DCH conducted the photo elicitation interviews. DCH and ANM wrote the initial draft of the paper. SSD managed the data. QLW, SSD, and CYE analyzed the qualitative data. All authors were involved in data interpretation and revising the paper. All authors had final approval of the submitted version.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the William T. Grant Foundation [grant # 187656], and the Research and Extension Experiential Learning for Undergraduate (REEU) Program of the National Institute of Food and Agriculture, USDA, [grant # 2017- 67032-26021 and #2022-68018-36607].
