Abstract
Quotations serve multiple purposes in qualitative research, such as supporting claims, illustrating findings, explaining, impressing, and offering readers vicarious experiences. Despite their widespread use, there is limited guidance on selecting and reporting quotations, leading to potential issues of “cherry-picking” quotes that may not accurately represent the findings. This study aimed to fill this gap by establishing guiding principles for qualitative researchers. Through an in-depth synthesis of literature focusing on papers that have discussed standards and guidelines for selecting and reporting quotations (
Background
Quotations are one of the tools used by qualitative researchers to support their claims (Sandelowski, 1994). Quotations can allow researchers to document scientific truth while adding artistic value to their work (Sandelowski, 1994). They also give voice to those who have previously gone unheard (Corden & Sainsbury, 2006a). Quotations, or the voices of participants, are often used to illustrate findings through examples or to highlight features of research (Eldh et al., 2020; Taylor, 2012). They are frequently used in qualitative research to explain, impress, or represent (Eldh et al., 2020). They also evoke emotion, providing readers with vicarious experiences (Corden & Sainsbury, 2006a; Sandelowski, 1994). Incorporating quotations can potentially enhance the readability of a text by infusing it with additional hues, vitality, and vividness while also captivating the audience (Corden & Sainsbury, 2006a).
Despite the widespread use of quotations and their significant roles in qualitative research, there is a lack of guidance to inform decisions to select and report quotations (Eldh et al., 2020). Several scholars indicated that the conceptual and theoretical foundations for using quotations in social research were not sufficiently established. Frequently, minimal explanations regarding the selection process for quotations were provided in the methodology section or elsewhere in the studies (Corden, 2007; Corden & Sainsbury, 2006b). Although rigorous and transparent qualitative analysis is essential for ensuring trustworthiness in qualitative research (Lincoln & Guba, 1985), few qualitative studies provide justification or rationale for selecting and including quotations in their research (Eldh et al., 2020). This lack of transparency may lead to the practice of “cherry-picking” quotations to support claims, even if the selected quotations may not necessarily represent the findings (Brennan, 2022). This study was undertaken to address the gap by synthesizing existing literature on standards and guidelines for selecting and reporting quotations and examining current practices in qualitative studies to distill guiding principles and recommendations that could inform the practices of qualitative researchers.
Methods
This paper consisted of two components. Initially, we conducted a thorough literature review on papers that focus on standards and guiding principles for the selection and reporting of quotations in qualitative research. The results derived from reviewing fourteen articles were synthesized to identify key factors to consider when selecting and reporting quotations. Furthermore, a systematic review was carried out, following predefined inclusion criteria delineated below. The purpose of this review was to assess current practices regarding the utilization of quotations in qualitative health research. To ground the review in a coherent and meaningful context, we selected maternal health in refugee populations as the focal topic. This decision was based on the lead author’s primary area of research and practical expertise, allowing for a more informed and nuanced interpretation of the literature. Papers were included if they met the inclusion and exclusion criteria below. The authors independently reviewed the papers and conducted screening using Rayyan (Ouzzani et al., 2016), a web-based systematic review application, employing the blinded screening functionality.
Inclusion Criteria
• Searched PubMed database • Empirical studies on refugee maternal health (the search terms combined three key concepts: “refugee,” “qualitative study,” and “maternal health”) • Any original qualitative studies that utilized quotations in their studies • Even if the study participants belonged to groups other than refugee women, such as healthcare providers, if the research focused on maternal health among refugee women, the papers were considered for inclusion.
Exclusion Criteria
• A systematic review, scoping review, narrative synthesis, meta-ethnography, meta-synthesis, or evidence synthesis • Quantitative studies or mixed methods studies even though they include some quotations • Research protocols • Studies without any quotations • Studies exclusively focus on infants or children without addressing maternal health, maternal healthcare accessibility, or utilization (e.g., papers on child feeding or breastfeeding among refugee communities) • Studies not directly relevant to pregnancy or maternal health, despite involving pregnant or postpartum refugee women, such as those concerning HIV/AIDS or hepatitis B among pregnant refugee women, were omitted.
After applying the predetermined criteria to screen the papers, the selected qualitative studies were systematically reviewed to examine how quotations were chosen and reported in the studies. The following elements derived from the literature were specifically assessed (Brennan, 2022). • Whether the authors documented their approach or rationale for selecting quotations • The guiding principles or criteria employed for selection • If the authors indicated any editing of the quotations • The number of quotations utilized in each study • How and where quotations were incorporated within the study • If authors drew quotes from diverse participants to offer layered evidence • Any language or translation considerations, if applicable
Results
In this section, we outline five key considerations identified through the examination of papers discussing the guiding principles and standards for selecting quotations in qualitative research. Subsequently, we present the results from the systematic literature review, elucidating current practices of quotation use in qualitative health research focused on refugee maternal health. Finally, drawing from these insights, we provide recommended practices for researchers seeking guidance on the selecting and reporting of quotations in qualitative health research.
A Literature Review of Guiding Principles and Standards for Selecting Quotations in Qualitative Research
The Five Considerations for Selecting and Presenting Quotations
•
•
•
•
•
Current practices of selecting and reporting in the literature
After searches in PubMed, 129 articles were found. Following the initial screening of titles and abstracts, 61 papers were considered for review. After a thorough examination of the full papers according to the inclusion and exclusion criteria, thirteen papers were ultimately included in the final review (the list of the reviewed papers can be found in Appendix 2).
Current Practices in Qualitative Health Research (
On average, the papers reviewed in this systematic review included 20.5 quotations (with a minimum of 9 and a maximum of 40; Standard Deviation = 8.1). Remarkably, most studies aimed to capture diverse participant voices or perspectives by extracting statements from multiple interviewees. For example, in one study, all quotations were extracted from different respondents (Agbemenu et al., 2021). However, one study reported six quotations from the same interviewees out of 11 (Scott & Wallis, 2021). In three other studies, it was challenging to determine whether statements from various participants were used as demographic details or interviewee identifiers were not included with the quotations (Balaam et al., 2016; Henry et al., 2020; Riggs et al., 2020).
All studies, except for two, provided detailed descriptions of the language-translation process from the interviewee’s native language to English. Most studies also outlined the interview procedures, including methods used for translation. For instance, Riggs et al. (2020) stated, “interviews conducted in an Afghan language were transcribed into English,” and Yelland et al. reported, “all interviews and focus groups were audio-recorded and transcribed verbatim; interviews conducted in an Afghan language were transcribed into English” (Yelland et al., 2016). Although juxtaposing native and English language quotes could enhance “transparency, rigor, and cross-cultural understanding of research findings” (Younas et al., 2022), none of the studies presented native quotes alongside their English translations.
All selected articles presented quotations in the results section except four articles, including Kabakian-Khasholian et al. (2022), featured quotations not only in the results section but also in the title, such as
The methods of presenting quotations varied across articles. Several articles presented quotations on separate lines from the main text by indenting the quotations using the same or smaller font size as the main text, while others used quotation marks, incorporating quotes in the main text. Alternatively, some utilized italic fonts along with quotation marks for the quotations. Five studies appeared to edit or refine quotations for clarity, using indicators such as abbreviations or ellipses to denote modifications. However, few studies clarified whether editing had occurred or explained how it was carried out.
Both exemplary and suboptimal practices were observed. First, commendable practices are outlined as follows: numerous articles endeavored to demonstrate that certain quotations reflected the experiences or perspectives of the majority of participants by using phrases like “Participants often thought…”, “As described by them…” or “Participants were generally…” (Mohammadi et al., 2017). Considering the space constraints of a paper, highlighting the prevalence of themes among participants could serve as an effective and concise approach to convey the level of agreement in their experiences and perspectives without including all quotations.
Some studies included demographic information with each quotation to enable a deeper contextual and nuanced understanding. For example, Russo et al. provided specific details about the interviewees, such as “Zamira, one child born in Australia, currently pregnant with the second child. Residing in Australia for 5 years (Russo et al., 2015).” Similarly, Kabakian-Khasholian et al. (2022) provided basic demographic details for each quotation, such as “Iraqi woman; 2nd child” and “Syrian woman married to a Palestinian; 3rd child.” Willey et al. (2020) also included specific demographic information for each participant, such as duration of residence in the host country, country of origin, and parity. The addition of demographic information for each quotation adds nuances, placing the statement in broader contexts.
Several articles incorporated diverse voices of participants, rather than relying on a small subset of participants, to ensure a balanced representation of interviewee experiences. Yelland et al. (2016) conducted interviews with 59 participants (16 Afghan women, 14 Afghan men, 10 midwives, and 19 professionals such as doctors, nurses, bicultural workers, and counselors). All 15 quotations were derived from different respondents, except for one Afghan woman who was quoted twice.
However, several non-exemplary cases were also identified. In some studies, interviewee IDs were either not provided alongside each quotation or were presented in a confusing manner, making it difficult to trace the origin of statements. This lack of clarity undermines the reliability of quotations, which serve as evidence supporting qualitative research findings, as their authenticity cannot be readily verified.
For example, Riggs et al. reported only general demographic characteristics or occupational groups of the interviewees, such as “Afghan women,” “Afghan men,” “community-based health professionals,” or “midwives,” without assigning unique identifiers (Riggs et al., 2020). As a result, readers could not determine whether quotations attributed to these groups belong to the same or different individuals. Similarly, Riggs, Elisha et al. did not provide any identifying information alongside participant quotes (Riggs et al., 2017). In another study, both individual and group interviews were conducted (Scott & Wallis, 2021), but it was not clear whether the numbers assigned to each interview referred to groups or individuals, further complicating the interpretation of data. Furthermore, many selected articles omitted participant demographic information entirely, using only generic labels such as “Focus Group 1” or “Focus Group 2” (Balaam et al., 2016).
Overall, the findings suggest that the criteria or rationale for selecting quotations were rarely reported in the reviewed qualitative studies. This highlights a gap in transparent reporting, despite its recognized importance in qualitative research methodology. While clear documentation of research procedures is considered essential, the criteria for selecting and reporting quotations as evidence to support research findings remain overlooked and underreported.
Discussion and Conclusion
This study addresses the lack of clear guiding principles for selecting and reporting quotations, despite their significant role in supporting qualitative research findings. This study, based on the literature review and systematic review of qualitative research, was conducted to determine whether quotations were selected and reported using appropriate criteria and to identify relevant principles and recommended practices for selecting and using quotations in qualitative research.
Recommended Practices Based on the Systematic Review and Reporting Guidelines
First, as mentioned earlier, researchers can use quotations to explain and elucidate participants’ experiences or perceptions of certain phenomena. In doing so, they need to be explicit about the functions that the quotations serve. In addition, it is crucial to use only one or two of the most relevant quotes to represent a single idea and to ensure they are directly relevant to the topic, avoiding any that are irrelevant. If study participants are not homogeneous, researchers should provide demographic information to offer a more nuanced understanding of the contexts (e.g., White women vs. Black women, age, or years of experience), as long as there is no perceived risk of identifying the participants.
Second, qualitative research should consider the guidelines of targeted journals, such as word limits and readability. Researchers need to adhere to these guidelines and ensure quotations are not too long, as lengthy quotations can distract readers from the main point.
Third, ethical considerations include understanding participants’ intentions and protecting interviewees’ personal information. Although some studies fell into danger of choosing quotations based on researchers’ preferences rather than participants’ intentions, quotations should be chosen to explain the main themes of the research, reflecting what participants want to express about their experiences or perceptions.
Fourth, aesthetic considerations involve reporting whether quotations are used verbatim or cleaned up to ensure transparency. In this study, although most selected articles tidied up the quotations using ellipses due to word limitations or to more effectively present participants’ discourses, they did not report in advance the criteria or reasons by which they shortened or refined the quotations. If it is impossible to include participants’ discourses verbatim due to aesthetic considerations, it must be reported where the modifications were made and by what criteria since quotations are evidence for supporting research results.
Finally, inclusive consideration involves using a mix of quotations from different participants rather than just a few. Although many studies tried to include participants’ multiple perspectives as quotations (Yelland et al., 2016), it is unclear whether the consideration of participants’ diversity in choosing quotations was a common understanding in qualitative research. To ensure that the selection and usage of quotations are not based solely on researchers’ intention and more effectively support qualitative research findings through triangulation, researchers should strive to reflect the voices of various participants.
Although this study has several strengths, several limitations should be addressed. Most notably, quotations from other fields were unexplored in this study, as the review focused specifically on qualitative papers on a narrowly defined topic. Consequently, this study may not fully capture the breadth of current practices or the range of issues related to the selection and reporting of quotations in qualitative research.
Furthermore, the five considerations are not mutually exclusive; rather, they are interdependent and sometimes conflicting. For example, researchers might try to include voices from as many participants as possible to adhere to inclusive considerations but then face practical constraints such as journal word limits. This complexity requires researchers to exercise their own discretion in determining their priorities and developing strategies to address them effectively. For example, to balance inclusive and practical considerations, researchers may choose to present quotations from multiple participants in a table, organized by demographic group. This approach allows for broader representation while staying within word limits, especially in journals that do not count tables toward the manuscript’s word count.
Another important point to note is the variation in methodological approaches, although this is beyond the scope of our study. The recommended practices for selecting and reporting quotations may require adjustment depending on the specific qualitative methodology employed in each study. For instance, ethnographic research often encourages more extensive use of quotations and contextual information for “thick description” to capture the richness of cultural contexts and convey participants’ voices and perspectives in depth. In contrast, case study research typically involves multiple sources of evidence (Crowe et al., 2011), with interview data and accompanying quotations representing just one component to provide insights about a case presented. In such cases, inclusion of extensive quotes may not be necessary. Although this study did not explore methodological differences in depth, examining and comparing how quotations are used and should be used across various qualitative research traditions remains an important area for future research.
Selecting and presenting quotations is a complex task. Researchers must navigate a large volume of data, make sense of it, and select and curate quotes based on numerous, and at times, competing consideration. They strive to present data that are truthful, trustworthy, representative, inclusive, and even artistically compelling, all while working within constraints such as journal-imposed word limits. Despite these challenges, quotations are a powerful tool in qualitative research. They not only support and substantiate findings but also serve as foundational elements that bring participants’ voices to the forefront, creating a meaningful connection between the participants and the readers. By thoughtfully maneuvering among these complex and often competing demands, researchers can meaningfully contribute to the body of knowledge, using participants’ own words as a vehicle for deeper insight and understanding.
Footnotes
Ethical Approval
Our research did not necessitate ethical board approval as it focused on literature review and synthesis without the direct involvement of humans or animals.
Author Contributions
SY formulated the research question and design, performed the literature review, and screened and synthesized papers. SH contributed to the study design, conducted the literature review, and screened and synthesized papers. SY and SH collaborated on writing the manuscript, which all authors subsequently reviewed and approved.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: SY is supported in part by the National Institutes of Health under Grant (T32 CA272303).
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.
Data Availability Statement
All the articles reviewed are publicly available.
