Stories Matter: A Novel Approach to Exploring Perceptions, Discourses, and the Symbolic Social Order in Pastoral Psychology

This article locates story completion (SC), a novel and underutilised qualitative data collection tool, as a valuable adjunct to traditional qualitative methods for pastoral psychology. In contrast to traditional self-reporting qualitative interviews or surveys, SC necessitates a researcher writing a story “stem” or “cue” – or, more specifically, an opening sentence – which participants are asked to continue in their own words. Uniquely, for SC, it is the stories themselves (which may be either first- or third-person perceptions, or constructions, of a phenomenon) which are subject to data analysis. Story completion has the benefit of being able to target implicit perceptions, or social constructions, depending on the epistemological framework chosen. However, this method has seldom been used in pastoral psychology, despite its potential significance. In this article, I demarcate SC as a prospective qualitative research tool for use in pastoral psychology, distinguishing it from other qualitative methodologies. I trace its emergence from psychoanalytic thought to its current usage in qualitative psychology research. I argue that SC has profound potential, especially for those looking to examine stigmatised topics or populations with sensitivity. Nevertheless, studies which utilise SC need to be theoretically cognizant and align fully with the ontological or epistemological assumptions of the researcher. I introduce and expound on varied epistemological frameworks that can be used in conjunction with story completion, further discussing their relative merits and potential drawbacks for pastoral psychology. I propose that, methodologically, SC is beneficial for accessing sociocultural discourses and broader representations surrounding religiously and culturally complex topics. I offer a case study of one recent research study, which used SC within the context of mental health and religion, to demonstrate its merit and applicability to the field. In doing so, I provide three contrasting epistemological readings of the data to show how these might be applied in practice.


Introduction
Story completion (SC) has been described as an "innovative, exciting and flexible method for qualitative researchers" (Gravett, 2019, p. 1) yet also, paradoxically, as "the . . . qualitative data collection that you've never even heard of" (Clarke et al., 2019, p. 1). Indeed, as of yet, this method has not been widely applied within qualitative research more generally and is unheard of within pastoral psychology and related disciplines. This is surprising as the method offers vast potential for pastoral psychology. In this article, I aim to introduce SC as a new possibility for qualitative data collection in pastoral psychology. Firstly, I discuss the landscape of qualitative research, why stories and language have been considered significant, and how SC fits within this picture, including its background and distinguishing features. Subsequently, I elucidate varied theoretical epistemological backgrounds appropriate for SC and discuss how these may impact the claims studies can make about the data generated. To provide some practical pointers for interested researchers, I also examine some potential benefits and areas to consider when running these studies in pastoral psychology before guiding the reader through a case study which reflects on the design and analysis of the first study to utilise SC method within the context of pastoral psychology, spirituality, and religion. Whilst I hope that the material will spark interest in the reader, this is not intended to be a definitive nor step-by-step guide to designing an SC study. 1 Instead, I hope to offer the interested reader a taster of this new method and to bring forward new qualitative methodological possibilities for pastoral psychology and related disciplines.

Stories, experience, and qualitative research in pastoral psychology
People tell stories. Whether these are about themselves, others, or the world around them, stories are influential in conveying information and (re)attributing meaning to past, present, and future events (Ganzevoort, 1993). Some may even go as far as to contend that stories themselves construct the very reality we experience (Burr, 2015). Therefore, storytelling has been, and remains, a significant feature of human life. As O'Toole (2018) contends, "[P]eople are storied beings and to generate a more in-depth understanding of people and their experiences, researchers need to begin with their stories" (p. 175). Simultaneously, stories themselves may be represented within the symbolic social order, specifically, the shared world of linguistic communication, intersubjective relations, and ideological conventions which are in the vanguard and regulate the acceptance of social attitudes and norms (Hewitt & Shulman, 1979). Stories matter.
Meanwhile, the subject matter of religion, or spirituality, poses a particular difficulty to the psychosocial researcher. While certain topics, such as frequency of prayer or church attendance, can be directly (and appropriately) quantified, other topics, such as what prayer or church attendance means to individuals, are considerably more intricate and frequently less quantifiable. This difficulty has not stopped many psychologists and social scientists from endeavouring to quantify such meanings. However, the development of qualitative methods has shown that these meanings can be studied in ways which are thicker and richer, being capable of capturing language, idiographic accounts, and phenomenological meanings (Coyle, 2008). The 'turn to language' in the 1960s marked a large-scale transformation in the social sciences, with many academics beginning to move away from traditional hypothetico-deductive or quantitative methodologies toward research paradigms, which emphasised the importance of language, experience, and meaning (Willig, 2012). These approaches, although not monolithic, are broadly understood as qualitative in that they seek to capture experience and meaning.
Pastoral psychology, which explores aspects of religion and spirituality from physical, social, human science, and interfaith perspectives, has embraced qualitative research from its inception (Swinton & Mowat, 2016), with issues of experience, context, and meaning frequently being captured through traditional interviews, focus groups, or surveys (O'Connor et al., 2001). While pastoral psychology has historically valued and championed the significance of stories and their potential therapeutic role in pastoral counselling (Bohler, 1987), SC as a research methodology has not yet been established as a qualitative data collection method within this discipline. Furthermore, until recently, there was no specified approach for exploring storied accounts of meaning beyond narrative approaches (Andrews et al., 2008). Although stories arguably lie at the centre of all qualitative research, SC has a distinctive history and focus.

Historical roots of story completion
Story completion initially evolved as a type of projective test for psychotherapists, psychiatrists, and psychologists to evaluate the personality and psychopathology of patients (Rabin & Zlotogorski, 1981). These projective tests involved asking individuals to react to ambiguous stimuli, such as the well-known Rorschach inkblot test (Rorschach et al., 1921(Rorschach et al., /1998. The assumption was that because the patient could not objectively demarcate the nature of the stimulus, they would be forced to rely upon their own inner or unconscious understandings in order to extrapolate and determine meaning. Through this process, it is claimed by psychodynamic theorists that patients reveal, or 'project', truths about themselves that they may not themselves be immediately conscious of or might not freely offer up due to fears of potentially breaking social norms (Rabin, 2001). Thus, projective tests were operationalised by Murray (1943Murray ( /1971 as providing "an x-ray picture of [the] inner self." Since their origin in psychoanalytic clinical practice, however, projective tests have also been used to collect 'data' for research purposes. Notably, this has included developmental psychology, particularly with the assessment of attachment styles in child and adolescent populations (Green et al., 2000). In these studies, projective tests have frequently adopted quantitative approaches, with intricate and psychometrically determined coding systems suitable for quantification. Many of these studies espouse a realist epistemology in that they view the projective test as accessing objective data about individuals or groups. It is worth noting that this epistemological foundation has been problematised by social constructionist qualitative researchers (Burr, 2015).

The (re)emergence of story completion in qualitative studies
Beyond the assessment of attachment, SC was first adapted for use in qualitative research in a 1995 study by feminist psychologists Celia Kitzinger and Debra Powell. They used SC, with a social constructionist epistemology, to analyse how male and female undergraduate students constructed infidelity within the context of a heterosexual relationship. This first use of SC involved providing participants with a story cue, or prompt, in which they were prompted to respond in their own words. More recently, a group of critical social psychologists, Braun et al. (2019), have developed formal guidelines for the use of SC in critical social psychology and qualitative research. As such, SC is currently undergoing renewed development as a methodology in qualitative psychology yet remains seldom explored in pastoral psychology, with only two publications using this method to date within the field of religion and spirituality (see Lloyd & Panagopoulos, 2022a;.

Contrasting story completion with other qualitative methods
Many traditional qualitative methodologies may assume direct access, 2 or a linear passage, from a participant's discursive account of their experience or perception of the world to meaning and experience. Following transcription, these first-person accounts or words (which are typically gathered through interviews, focus groups, or surveys) then become subject to qualitative analysis by the researcher in order to access patterns of experience and render claims about the meaning-making of groups or individuals (Willig, 2012). By contrast, SC involves a researcher scripting a story "stem" or "cue" -or rather, the beginning of a story, usually with an opening sentence or two -and requesting that participants complete or continue the story in their own words and terms. Unlike traditional modes of qualitative data collection involving the gathering of direct self-reports, SC affords the researcher a fundamentally distinct approach to data collection that allows researchers to access either social constructions or perceptions about a given topic. See Table 1 for a visual comparison between SC and traditional qualitative methods. Morrow (2005) encourages qualitative researchers to clarify the philosophical and theoretical assumptions or paradigms underpinning their research, and this is also true for SC. Story completion is a versatile method; indeed, I contend that it can be used with a range of epistemological frameworks, making it suitable for a range of research questions relevant to pastoral psychology. However, there are implications for the claims that can be made about the data and study implications depending upon the theoretical approach taken; researchers who embark on an SC study should reflect on their chosen theoretical assumptions from the start. It is also vital to ensure that the type of research question created matches the researcher's own epistemological or theoretical approach; 'perception' questions often lean more towards realist or contextualist research, whereas 'construction' and 'representation' questions are most often employed in social constructionist research.

Theoretical assumptions
Broadly speaking, researchers utilising SC have chosen from three epistemological frameworks: essentialist or realist, contextualist, and social constructionist. It is worth noting, however, that these three theoretical approaches represent a spectrum and that some studies may combine elements, perhaps with a focus on participant perceptions and discourses. I will now consider each of these approaches in turn and sample some possible research questions used with each framework (Fig. 1). Research Claims: Data may be partly treated as reflecting participants perspectives but more attention given to how these are socially enabled and mediated through shared meanings.
Published Example: Lloyd & Panagopoulos (2022) Study Focus: What narratives, or, discourses do participants use to make sense of XXX?
Research Claims: Data treated as relative and not reflecting participants inner beliefs. Attention given to language and discourses and how these construct particular realities for participants and social groups.

Realist story completion studies
Firstly, SC can be positioned within a realist epistemology. Here, the emphasis lies on discovering particular psychological meanings imagined to be linearly connected to stories. Participant stories may be examined qualitatively as if the words are referring to real events or meanings and as giving direct insight into the participants' psychological life and perspectives on a given topic. For example, Livingston and Testa (2000) used qualitative SC within an experimental design in which the female participants were given alcohol, a placebo drink, or no drink to explore women's perceptions of their vulnerability to male aggression in a heterosexual dating scenario. Participants were asked to imagine themselves as the female character in their story cue and to write in the first person. Responses were treated by researchers as representing participants' true beliefs about this topic. This position, however, necessitates an interpretive leap in which the researcher assumes that the personal feelings and motivations of the story writer are unproblematically and linearly attached to story writing and are thus accessible through SC.

Contextualist story completion studies
Alternatively, SC can be framed somewhere between a realist and a social constructionist framework in that it is concerned not only with individual and broader social perceptions (as in social constructionist studies) but also with how these are socially mediated and embedded. This approach is designated 'contextualist'. Contextualist paradigms are approximately analogous to critical realism (Willig, 2019) in that the focus is on participants' perspectives (analogous to realist accounts, which attempt to expose inner psychologies) but also considers how these perceptions are socially embedded and enabled. Hence, social context and discursive life are both explored. This approach has the advantage of focusing on the micro level of participant perceptions about a given topic; however, it also enables researchers to explore the particular discourses participants draw upon to construct or make sense of certain phenomena through their narratives. Lloyd and Panagopoulos (2022a) used SC to explore social perceptions of self-harm within a large sample of evangelical Christians. Instead of asking participants outright about their experiences or views of self-harm, they were asked to respond to a gendered story cue in which a fictional devout Christian character, who had experienced some form of selfharm, visited their local church. Participant responses were analysed and presumed to represent both the participants' social perceptions about self-harm and their constructions of self-harm within the context of the Christian faith.

Social constructionist story completion studies
Finally, and most frequently in SC, a social constructionist approach can be adopted. Story completion studies which utilise a social constructionist framework direct attention to the discourses and language used by story writers to construct or recount their stories (Burr, 2015;Kitzinger & Powell, 1995). In a constructionist framework, the stories are not imagined as capturing any singular psychological reality per se but, instead, are presumed to socially create that reality through language (Burr, 2015). In other words, analysed stories are not assumed to access participants' own perspectives on a given topic; rather, they are assumed to reflect the particular social context and discourses available to that person to make sense of it. Frith's (2013) research on orgasmic absence is an example of a social constructionist-informed SC study. Analytic attention was given to how the stories utilised and reproduced differing gendered discourses, including women's duty to be sexually attractive to preserve men's sexual interest and the construction that men's sexual desire is uncontrolled and easy to satiate. However, using a social constructionist framework meant that claims about the data in general and the possibility of these reflecting extra-discursive or 'real' perceptions of the participants are limited. I now discuss some sample research questions for SC studies in pastoral psychology before exploring the benefits and drawbacks of SC studies as applied to pastoral psychology.

Differing research questions in pastoral psychology and story completion
The versatility and creative potential of SC are two of its fundamental benefits; consequently, it can be employed to research a broad range of topics related to religion and spirituality. Story completion is particularly well-matched for research studies which attempt to explore individual perceptions, views, and social constructions. However, research questions focusing on lived experience are not easily capturable using SC. When developing research questions in SC designs, as in any other qualitative project, there is a need to ensure that the topic is both focused (in that it explores a phenomenon) but also sufficiently broad and open-ended (see Fig. 2 for examples of topic areas amenable for SC studies in pastoral psychology and

Merits and drawbacks of story completion as applied to pastoral psychology
There are numerous merits and drawbacks to using SC in the pastoral psychology context. The merits of SC are as follows: 1. Story completion provides rapid access to a range of responses, including socially undesirable ones. Because participants are prompted to write hypothetically, SC is advantageous when exploring direct perceptions or experiences about a sensitive topic that might hinder honest or more reliable responses. Story completion can be viewed as offering an alternative to approaches that ask people directly about their views and understandings of a particular topic, instead asking them to write about the hypothetical behaviour of others (Will et al., 1996) or how they would feel in a hypothetical situation. 2. Participants who respond to SC do not have to identify with or justify their stories in the same way as if they were asked directly in an interview or qualitative survey. Instead, in SC, participants may be asked to respond to a hypothetical first-or third-person story cue with whatever comes to mind. Participants may therefore feel more relaxed and engage in the study more openly. This gives SC the compelling benefit of penetrating the "social desirability 'barrier' of self-report research", which obscures much qualitative research (Moore et al., 1997, p. 372), and thus may be particularly useful when approaching religiously or culturally sensitive topics. 3. Story completion is accessible to more participants than traditional qualitative methods as it does not necessarily require direct experience of a particular phenomenon to take part (e.g., self-harm, demons, mental illness and spiritual devotion, prayer, religious attendance, female genital mutilation). By asking participants to respond to hypothetical stories, researchers are able to elicit participants' perceptions or constructions (depending on the selected epistemological framework). 4. Story completion offers an array of easy-to-implement comparative design options. For example, SC may explore gendered or other intersectional differences (e.g., sexuality, age, disability, religion, ethnicity) by asking participants to respond to the same story stem with key characteristics altered. It supports researchers in exploring, for example, divergences in how different social groups make sense of a scenario and whether participants respond differently to variations in the story character. An example of this  and constructionist (discursive constructions)

Fig. 2 Exemplar Story Completion (SC) Topic Areas for Pastoral Psychology
Recovery and a higher power/the Divine in Alcoholics Anonymous What are the perceptions of a higher power among recovering alcoholics attending Alcoholics Anonymous? (Lloyd & Panagopoulos & Lloyd, in prep) Contextualist (perceptions and discourses) might be when a researcher is interested in how male and female members may be perceived in different religious communities and in mapping any differences in completed stories. Stories may also opt not to include comparison but may generate a story stem which draws out participants' assumptions and perspectives about such differences. A key published example of this approach is by Lloyd and Panagopoulos (2022a), who created a story stem designed to elicit responses to female self-harm in the evangelical Christian church. They found that the female characters were frequently portrayed as emotionally volatile and immature -a critical finding which helps to address stigma in this area. 5. Story completion offers potential in terms of methodological creativity. Researchers may opt for two-step qualitative approaches, where SC completion is used at the beginning of a study to elicit perceptions or constructions, followed by open-ended qualitative questions through written text. In combining story completion with later qualitative surveys, I recommend that SC precede qualitative questions. I have used this approach in several papers and find it helpful to access both perceptions and experiences. Finally, there is also the option to use visual cues, such as graphics, cartoons, or pictures. 6. Story completion has the added advantage of being economical in terms of time and resources; participants may complete the stories via an online survey link, and responses are ready for analysis immediately without transcription. 7. Story completion research often triggers fewer ethical concerns than research focused on eliciting data through personal accounts (e.g., interviews, focus groups). This is particularly the case when SC is used in online studies, which makes it easier for participants to remain anonymous and removed from interaction with the researcher. 8. Story completion is fun and novel, and many participants (and researchers) enjoy participating.
Drawbacks and areas to consider in SC are as follows:

Story completion does not readily tap into participants' experiences of phenomena.
This can be confusing for researchers who are used to using traditional qualitative data collection tools to gather experiences directly. Whilst broad perceptions can be accessed within SC studies dependent on the theoretical approach, asking for participants' own experiences in SC studies removes one of the major benefits of this approach, that of social desirability (see point 1 above under benefits). 2. Stem construction in SC studies can be surprisingly complex, and the following need to be carefully considered: length of the story stem and the amount of detail provided and the need for authentic and believable characters. Firstly, whilst there is no one size fits all in terms of the length of the story stem, too long a stem may risk losing participant interest. Suppose the study topic under consideration is expected to be familiar to potential participants. In that case, a sentence or two will suffice, but more may be needed with more intricate or unfamiliar topics to orientate participants and ensure that the stem designs are targeting the phenomenon of interest. Creating longer and more complex stems, however, risks the researchers' projecting their own meanings onto the participants, who will inevitably be influenced by the stem design. Secondly, the stem design should be easy for participants to relate to and engage with -using relatable names and characters. 3. Stem construction needs to be guided by the literature and study focus -ideally, there may be a theoretical reason for designing the study topic in the first place. For example, does the literature suggest that a particular focus is needed, such as a focus on women or men, people of a particular social group, or other religious minority groups? 4. Researchers need to think through how they operationalise perception, imagination, construction, and other key terms -this partly depends on whether participants have direct experience of the phenomena under interest and the epistemological basis of the study. For example, Lloyd and Panagopoulos (2022a) recruited participants both with and without direct experience of self-harm for their contextualist-informed study. Nevertheless, all participants were required to 'imagine' the story -they did not distinguish this from perception, assuming both were interconnected (Fettes, 2008). When submitting SC studies for peer review, reviewers may seek clarification on such terms and their uses within the studies. 5. Story completion stems can be either first-or third-person design. It is essential that researchers consider early on whether they will use first-or third-person story stems. For example, is it significant for the particular project that participants imagine themselves as a character? Or, is it better that they are more removed and adopt the role of narrator (third person)? Third-person story stems have the advantage of the participant being more removed from the story and hence improve the chances of acquiring more socially undesirable responses that participants might not typically offer during traditional qualitative interviews. This is especially significant in pastoral psychology when interrogating religiously or spiritually sensitive topics of interest. 6. There are no defined rules regarding sample size. Sample size depends on a number of variables, including (a) the intricacy of the study design -more story stems may require more participants to be able to say something meaningful about each story, particularly if you intend to make comparisons; (b) the epistemological approach adopted by the study -realist studies aiming to make claims about perceptions and the 'real world' may wish to include larger samples, whereas social constructionist studies require less because these studies are not so concerned with generalising results or making claims about an external social reality; and (c) the depth and intricacy of individual stories -acquiring more detailed stories with richer narrative detail may mean that fewer stories are required. 7. The need for balanced ambiguity. As discussed, SC is particularly beneficial when exploring underlying assumptions or perceptions of a particular topic -for example, perceptions of the demonic in relation to mental illness in fundamentalist communities or perceptions of self-harm in relation to a female gender identity versus a male. Both of these topics are considered sensitive in nature, and it depends upon the researcher to decide what they are trying to explore through their SC study.

A case study of story completion in pastoral psychology
To demonstrate the utility of SC in the context of pastoral psychology, I present here a case study that details one of the only published SC studies within the field of religion, spirituality, and mental health (Lloyd & Panagopoulos, 2022a). By describing our aims and thinking behind the design of the study, I aim to tease apart some of the core considerations researchers should be aware of when designing studies that utilise SC. In doing so, I offer some constructive criticism regarding what we could have done differently. In addition, I discuss the process of story stem design and analysis and highlight some general potential pitfalls along the way. I conclude this case study section by presenting three differing analytic readings of our SC data, offering the reader some tangible examples of realist, contextualist, and social constructionist interpretations of SC data.

Context, rationale, and participants
In this study, we were interested in exploring the phenomenon of self-harm within an evangelical Christian context. Prior research suggested that evangelical Christians may be more likely to ascribe belief in demons, sin, generational curses, or other spiritual influences as causative to mental distress (Lloyd, 2021a, b;Lloyd & Hutchinson, 2022;Lloyd & Waller, 2020). Some research suggested that this could have either positive or negative influences upon the individual and collective wellbeing (Lloyd, 2021a), with the potential for impacts on help-seeking (Lloyd et al., 2021) and level of mental illness . Most of the existing research, however, had focused on mental illness in often homogeneous terms rather than zooming in on specific types of mental illness (e.g., anxiety, depression, self-harm) and had drawn upon either traditional qualitative frameworks with semistructured interviews or surveys or had used a quantitative design to capture predictors of attitudes and beliefs. Given the particularly stigmatised nature of self-harm (Aggarwal et al., 2021) and its severely underreported status, with those who do disclose self-harm demonstrating diminished help-seeking (Nearchou et al., 2018), we considered it significant to explore wider social perceptions within this population.

Writing the story stem
One of the core challenges when designing SC-based studies is the creation of the story stem or cue. Braun et al. (2019) suggest that the stem needs to provide enough detail to orientate the participant to respond to the phenomena of interest. For our study, stem construction was steered by existing literature which suggested that Christian communities may exclusively conflate mental illness with personal sin (Lloyd, 2021a;Wilder, 2012). Hence, the stem design was written in an attempt to capture perceptions related to the coexistence of Christian faith and mental ill health through the term 'devout'. Moreover, as a secondary goal, including a female name in the story stem meant that, through our analysis, we could explore and illuminate how female identity might intersect with perceptions of self-harm and mental illness amongst evangelical Christians (Weaver, 2014). We could have opted for a comparison design at this stage to tease apart the perception of gender more fully; however, our focus for this study remained primarily on self-harm, with a secondary goal of exploring any gendered assumptions which may be attached to these perceptions. Researchers in pastoral psychology who design SC studies should carefully consider what phenomena they aim to explore and how the particular character used might provoke gendered, sexed, or other characteristic assumptions as this may intrude upon the data collected; in other words, they should think at an early stage about how the name, gender, or other words used in the story stem might imply or lay overt meaning on said stem. Sometimes this can be useful when wanting to target particular phenomena, but it can also create unhelpful noise. At this point, researchers may also want to include a two-stage qualitative or mixed-method design whereby participants respond to the SC component before completing open-ended qualitative questions. We did not include this second stage for the present study but are including it in future studies (Lloyd & Panagopoulos, 2022b).

Study recruitment
For study inclusion, participants needed to be at least 18 years old and identify as an evangelical Christian living in the United Kingdom. None of the participants was a pastor or member of the clergy. Following ethical approval, an advertisement was placed on online Christian social media groups with the following text: This study aims to explore perceptions of self-harm in the church community. Please consider taking part in this short, online creative study in which you will complete a short story. Anyone who is an evangelical Christian and lives in the UK is eligible to take part.
For this study, we drafted a stem which was piloted with several individuals. This was then revised so that the stem had sufficient detail to orientate participants without imposing too much meaning and allowed us to examine whether the final stem was focused enough to capture perceptions of self-harm. The stem was: "Summer, a devout Christian, would sometimes self-harm. She . . ." To ensure sufficient data quality, all participants were asked to spend a minimum of 10 minutes writing their story or to produce roughly 200 words or 10 lines. Participants were asked to create a story with whatever came to mind, thereby increasing the chances of accessing less socially desirable responses and perceptions.

Data analysis
Commensurate with other qualitative studies, SC is often analysed using thematic analysis or a variant (Braun & Clarke, 2006). For our study, we downloaded the completed stories and began by reading through each. At this stage, we discarded 19 stories due to non-story completion and an additional four containing fewer than 10 words. Our final sample included 101 completed stories on which we conducted thematic analysis (Braun & Clarke, 2006). Our analytic approach was guided both by our contextualist epistemology and our research focus. This meant that we were able to make claims about the stories by partly accessing participants' own assumptions and perceptions about selfharm in the evangelical Christian context. To provide a tangible illustration of the varied storied data that can be collected during SC studies, refer to Table 3. These stories highlight the depth and breadth of possible storied accounts when running SC studies.

Analytic challenges in story completion design
As can be seen from the sample stories collected in our study and the varied epistemological frameworks available, a particular challenge with SC studies can be at the level of analysis; this is due to the varied and diverse stories that participants construct, which can create a challenge when researchers begin coding and mapping themes. Clarke et al. (2017) suggest that researchers can opt for either a 'horizontal' or 'vertical' level of analysis. Horizontal analyses focus on identifying patterns within the stories as a whole (e.g., looking at how the cause of self-harm is depicted across the stories), often guided by the research focus. Vertical analysis, however, explores how stories unfold and progress overall (e.g., looking at how the storied accounts document the process of self-harm for the fictional character, including the beginning, middle, and end of the story and mapping any similarities or divergences). For this particular study, our analytic approach was also guided by our research question and focus, and we emphasised analyzing perceptions of the cause and cure of self-harm within evangelical communities. This meant that in reading each story we were looking for perceptions of the possible cause and cure or treatment for self-harm (horizontal analysis), as well as how the stories themselves were narrated (vertical analysis), with the character struggling > questioning faith > and reaching a conclusion or climax. Researchers who conduct SC

Participant Story (Perceptions of Self-Harm in Evangelical Christian Communities)
Story Stem: "Summer, a devout Christian, would sometimes self-harm. She . . ." "[w]as racked with guilt and shame. She would feel that she had scarred the body that God had created.
But she also knew that God loves her so much and that when she hurts, He hurts too. She remembers that she's never alone. Ever. It is not sinful. God loves her whatever she does and when she cries, God will cry whilst sitting alongside her. It is not her fault she does what she does. There is always a root cause". "[h]ad prayed over and over again for God to take away the desire to self-harm but this behaviour continued, sometimes frequently, sometimes less so. Summer felt deeply ashamed about what she saw as sinful and out of control. She wondered why she couldn't stop and didn't want people at church to know about it, so kept the self-harm secret. She was receiving therapy with psychological services, but so far it didn't seem to be making a lot of difference. She often questioned her faith and wondered if God had given up on her, if she was too damaged to really help. There was a terrible sense of isolation as Summer couldn't even speak to her family about her problems, though she had an idea that her mother suspected she had problems she wasn't admitting to. At church, people seemed really happy all the time and Summer didn't feel like she was being real, but that her true self would be unacceptable." "[w]as desperate. Everything in her world; her family, church, friends, colleagues, the Bible itself told her it was wrong. Summer knew too, but it seemed the only way to find a way through the numbness was to cut until there was blood. Her blood. His blood. The cross. The abuse. The complete and utter destruction of any semblance of humanity -done to Jesus on that day. Innocence. Does He understand her? She doesn't understand herself or why she cuts. Sometimes she has no idea how the marks were made. But there they are, on her wrist. Undeniable Wounds. Self-made. "Stupid girl" she hears one of 'them' say loud and clear. The voices. They are the problem. If only she could be 'normal' and switch them off. Stop them. It's all inside her head. But it's not. As she blots the stain with her sleeve she senses another 'part' forcing its way through her brain. The dissociation and denial of the pain are strong enough to force 'Little Summer' (her trauma therapist named her) to the front. Suddenly, her body collapses and she is in the foetal position on the floor, unaware of the screaming, the tears and the pleas for it to stop. This can't be real. There are other parts too. Summer comes back to consciousness. There's blood and mess and the strange sensation she's missed something important. The amnesia plagues her life. How can she tell anyone? They will think she's mad, bad, or possessed to think she has dissociated parts of her personality inside her. Questions. The diagnosis of DID [dissociative identity disorder] was a shock at first. They will think it's demonic and disown her. Jesus. He has appeared to some of her parts before. He is always gentle. He believes her. Maybe that is enough for now." "[s]he was a crucial part of the Church. Born and brought up in it. A star of the youth group, a gifted speaker. People said she was so blessed. Life was busy with college, environmental activism, worship meetings, serving others. All the things God and her fellow church members would expect her to do. Nothing was ever enough for God. She believed He demanded this level of commitment, of service. So why were there cuts on her arms? A modern equivalent of self-flagellation? Or a cry for help? Or a piece of her life that she could control, that did not rely on obedience to the church elders, a way into things in her life that could not be papered over by a noble verse or prophecy? Perhaps all those things. When someone saw the signs and reported it as a safeguarding issue the fallout was horrific. Some refused to speak to her. Others begged her forgiveness for having taken her for granted, the leadership team was split over blame, guilt and what next. Summer confided in a college tutor who managed to get her counselling support. It took time but Summer and her faith came through the experience. She goes to a different church now". studies should try to keep in mind their particular research question and focus and let this guide the analytic process.

Comparing realist, contextualist, and social constructionist interpretations of story completion data
As discussed, SC data is commonly interpreted through one of three epistemological positions: realist, contextualist, or social constructionist. In order to offer further clarification around these three positions, I present three comparison readings of SC, each using a different epistemological interpretation. The aim is not to prioritize, or privilege, one framework of data interpretation over another but, rather, to support understanding of how these differing readings may diverge from each other, how they might be applied in practice, and where analytic attention may be directed, with consequences for the particular claims these studies can make about the storied data. Depending on the researchers' own interests and particular research questions, all epistemological positions may offer value and interest.

Realist
Realist readings in SC place emphasis upon the participants stories as revealing implicit attitudes or perceptions. Here, then, participants' stories are assumed to reflect the inner psychologies and beliefs of the story writer. In realist readings of SC data, qualitative data may also frequently be coded through quantification, a process in which dominant themes are numerated according to participant representation (e.g., how many times particular themes are identified in the coded data).
In an exemplar realist reading of the extract below, three perceptions of self-harm emerge. Firstly, there is the perception that self-harm may be connected to living in sin. Secondly, there is the perception that Summer is unable to share her experiences and must keep these "secret", and thirdly, that her experiences of self-harm led to a weakening of faith. It is worth noting that, in realist readings, interpretive language may frequently be more fixed (e.g., "the fact that", "this shows", "this reveals") and less tentative than in the other types of readings. This is because realist epistemologists often assume that there exists one discernible, or generalizable reality which may be readily captured through qualitative analysis: Summer felt deeply ashamed about what she saw as sinful and out of control. She wondered why she couldn't stop and didn't want people at church to know about it, so kept the self-harm secret. . . . She often questioned her faith and wondered if God had given up on her.

Realist exemplar interpretation
Many of the participants (68%) depicted Summer as experiencing shame as a result of her self-harm. The fact that participants represented Summer as feeling shame reveals the perception that self-harm may be perceived as connected to sin. In nearly all cases in which self-harm was described (90%), participants indicated that Summer had a desire to keep her self-harm secret, which was frequently depicted as leading her to isolate herself from others (65%). Finally, the participants' frequently depicting Summer as "questioning her faith" due to her self-harm (80%) supports evidence that self-harm may be perceived as breaching the boundaries of Christian faith due to living in sin, with negative consequences for faith in God.

Contextualist
Contextualist readings of SC data often combine realist ideas that storied data permits the researcher access to participants' perceptions or thoughts about a given topic, but frequently this is held within the larger context of the participants' social milieu. Hence, there is more of an appreciation of how participants' perceptions may have been shaped by their particular social milieu and how this may influence their perceptions of a given topic or phenomena (bi-directional relationship). Furthermore, the analytic narrative may be more tentative than realist readings given that there exists an acknowledgement of more localised and socially contingent meanings (e.g., that perceptions of self-harm may differ depending on population, time, place, etc.). Unlike realist studies, quantification of themes may not always be present, but this is often dependent upon the objectives of the particular research study and the organisational preferences of the researchers.

Contextualist exemplar interpretation
The explicit reference to Summer's feeling of shame and her fear she had committed sin in the stories is telling. It was also not only Summer's affective experience (feelings of shame and being out of control) that was recognised but also her wider fear of negative reactions from her church community that led Summer to keep her self-harm secret. The descriptions seemed to imply that self-harm indicated a lack of emotional control and faith in God.

Social constructionist
A social constructionist reading of SC data stands in contrast to realist readings and challenges taken-for-granted notions and categories of the 'individual and the 'self' -instead drawing attention to the constructed nature of mental health, behaviour, and emotions (Lloyd & Finn, 2017). From a social constructionist perspective, it is not that participants' stories of self-harm reveal their underlying beliefs or perceptions nor some combination of individual psychologies and social contexts. Rather, storied data is interpreted as a discursive achievement. In other words, analytic attention is towards the discourses or interpretive repertories participants draw upon in making sense of self-harm (or other phenomena). These discourses are not considered part of the participants' own psychologies but are viewed as socially created and continually (de)constructed, depending upon the social context. This reading of SC data involves rejecting the belief that researchers have unfettered access into participant's 'real' perceptions or feelings in favour of acknowledgment that "we have no access either to our own emotions or to those of others, independent of or unmediated by the discourse of our culture" (Jaggar, 1989, p. 148). Interpretive narrative in social constructionist interpretations of SC studies tends to assume a more interrogative stance, exploring the particular discourses used and hypothesizing about their historical origins and function for the participant and social group surveyed. Furthermore, particular attention may be given to power dynamics in terms of how discourses function to position storied characters in terms of responsibilities and social roles. Finally, quantification of themes is seldom used in social constructionist studies -as the assumption that there exists an 'extra-discursive' or fixed reality (as in realist studies) is problematized -and little value is seen in attaching quantification to constructed themes. There is frequently also a greater recognition of the researcher's own material and how their social position may necessarily influence data analysis (with themes often recognised as co-constructed).
In an exemplar social constructionist reading of the extract below, three discourses of self-harm might be constructed. Specifically, a discourse of self-harm as sinful is seen to involve the marking out of an identificatory truth that is situated in religiously (un)acceptable acts (harm against self), while an overlapping discourse of secrecy (from others) invites a removal of Summer from other people and doubts about her relationship with God (questioning her relationship with the Divine): Summer felt deeply ashamed about what she saw as sinful and out of control. She wondered why she couldn't stop and didn't want people at church to know about it, so kept the self-harm secret. . . . She often questioned her faith and wondered if God had given up on her.

Social constructionist exemplar interpretation
The discursive construct that Summer is 'ashamed' is perhaps exacerbated and maintained by the religious discourse that self-harm is sinful. Specifically, Summer's experience of self-harm is discursively located as outside the bounds of Christianity. In other words, that experiencing self-harm is situated as violating faith seems to function to construct Summer as culpable for her self-harm and subsequent feeling of desertion by God. The consequence of this as discourse is that if Summer were to stop self-harming and regain "control", her position and faith in God would improve. The participant's phrase "out of control" seems to imply emotional volatility and may coalesce with gendered and religious discourses of women as unstable and unable to control their emotional experience.
Story completion has vast potential to complement other qualitative approaches, through its unique focus on social perceptions and discourses, particularly with stigmatised groups or when exploring sensitive topics in the field of religion and spirituality. Furthermore, SC has unique advantages compared to other qualitative methods, such as the ability to gather data at speed and being generally resource-light in terms of ethical approval, data collection, and transcription. Story completion is also uniquely flexible in its approach and is amenable to a range of theoretical and comparison designs, depending upon the researchers' own theoretical interests. There is also space for researchers to be creative in their SC approach, and SC may be easily adapted to focus on characteristics of interest (e.g., gender, sex, sexuality, ethnicity, religion, health status). However, researchers must ensure that SC studies are robustly designed and match their epistemological assumptions. Whilst SC is not suited to explore direct experiences of phenomena, SC may be utilised as one part of a battery of broader qualitative methods, such as open-ended qualitative surveys or interviews (Lloyd, 2023), and in that sense may be considered complementary to other qualitative approaches and as having the potential to "reach the parts that other methods cannot reach" (Pope & Mays, 1995). I hope that qualitative researchers in the field of pastoral psychology, religion, spirituality, and culture will consider adopting this valuable qualitative research tool in their current or future research.

Declarations
Ethical approval The data reported in this study was previously collected with full university ethical approval and participant consent for publication.

Conflict of interests The author declares no conflicts of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.