Using theory – To predict outcomes,describe,analyse or interpret? A framework for analysing the use of theories

Abstract

This article directs attention to theory in academic writing. The empirical material analysed consists of 52 Swedish doctoral dissertations on social work in healthcare, published between 2008 and 2019. The aim of the study is to describe and suggest a framework for analysing the use of theories in research work. The results suggest that the field is characterised by an intersection of sciences, mirrored by two overarching ways of understanding theory. One approach, less frequently drawn upon, is analysed as influenced by the health sciences, understanding theory as a means of predicting outcomes by using different variables, without defining them within a broader theoretical framework. Theorising is here analysed as the enabling of predicting outcomes, by generalisations. The more common approach to theory is analysed as influenced by the social sciences, understanding theory as a way of understanding society and social relations, explicitly defining and referencing the concepts used. Within this second approach, theorising is done differently and, drawing on Wolcott´s descriptions of ways of doing analysis in ethnography, we propose three ways of understanding this theorising, conceptualised as Description, Analyses and Interpretation. In the article we outline how different understandings of theory and theorising allow for different kinds of knowledge to be gained. The framework presented can be useful for students and researchers when planning research and writing academic texts, clarifying different ways of understanding theory and the potential use of variables and theoretical concepts: from aim, to analysis and implications.

Introduction

Theory can be understood as the core of social sciences, essential for understanding and analysing relationships between the individual and the structural (e.g. Mills, 1959). Over the past decades, there has been a vivid discussion about the roles of theory in social science research and education. It has been argued that students ought to learn to use theory in the same way they learn to use research methods: not as finished concepts to draw upon, but as practical processes and skills that can be developed, given knowledge of theory (Swedberg, 2016). This shift to theorising has been appraised but also criticised for emphasising questions of how, rather than focusing on central questions of theory—what and why—as well as for focusing on individual researchers’ thinking rather than the collective dimensions of research (Bertilsson, 2016). It has further been argued that there is a need for a stronger focus on conceptualisation, highlighting the relations between the theoretical and the empirical, that escapes both theoretical apriorism and naïve empiricism (Carleheden, 2016). A need to develop ways of demonstrating how these conceptualisations can contribute, for theorising to “go somewhere”, has also been highlighted (Krause, 2016:28).

There is, however, no consensus among social scientists about what theory is (e.g., Abend, 2008), and if you step outside the social sciences, particularly into the healthcare field, the debate about theory becomes even more pronounced. Discussing potential benefits of or need for the use of theory can, in these contexts, be a central argument of science (e.g., Chin-Yee, 2014; Green, 2000). Theory can also be used differently. In the social sciences, theory is usually associated to ways of understanding society, social interactions, and social change (Abend, 2008), and its functions can be described as giving “direction to research work in explaining, organizing, analyzing, and predicting phenomena and showing their relationships in order to enhance understanding” (Chijioke et al., 2021:156). In the health sciences, this explicit use of theory to interpret empirical material has been recommended but is not a common practice (Wensing and Ullrich, 2023). When theory is used in these contexts, it is often synonymous with a hypothesis: a way of specifying relationships among variables to provide explanations for predicting outcomes in the future (e.g., Portney and Watkins, 2015).

This article takes these discussions on what theory is and what it can do as its starting point, focusing on theory in doctoral dissertations concerning social work in healthcare. The aim of the study is to describe and suggest a framework for analysing the use of theories in research work. The analytical interest is directed at the potential knowledge gained from different ways of theorising.

The study addresses both “theoretical concepts” and “theory”, and the two are used interchangeably, referring to the use of specific theoretical concepts, and more general theories. The theories themselves are not the focus of the analysis, nor is the question whether the individual dissertations can be analysed as consistent, or “successful” in their use of theories. The dissertations that make up the empirical material are analysed as examples, and the analysis directs attention to how theories can be used in research work.

Previous research

Although the use of theories is fundamental in many academic fields, there is a lack of research focusing on different ways of using theory. Chijioke et al. (2021) have, however, explored theorising in textbooks and articles about research methodology in public administration and found that scholars as well as students sometimes fail to use theory in academic writing. Zambo (2014:514) analyses theories used in action research in education doctorate programs in the United States and highlights the importance of theory-practice transference, further emphasising how researchers “need a broad array of theories that are understandable and can help them understand their problem, discover new possibilities, and gather data in strategic, reasoned, and ethical ways”. The relationship between practice and research has also been discussed within social gerontology, and Alley et al. (2010) highlight how this field is characterised by an intersection of the social and health sciences, often aiming at applicability. Social gerontology has a tradition of being influenced by nearby fields, such as sociology, psychology, social psychology, and economics, which Alley et al. (2010) suggest influences the use and understanding of theory in research.

Järkestig Berggren et al. (2021:291) discuss how social work has been criticised for not being theory-driven and argue that “the field is preoccupied with trying to understand, define and find answers to social problems and improving guidance for practice”. They further discuss many researchers having an eclectic approach to theory and use theory in different ways: some use well-established theories and concepts, some combine theories from different fields, and some perform new elaborations of an existing theory. Järkestig Berggren et al. (2021:292) do not recommend or criticise these different ways of using theory but call for “a more in-depth discussion about the theoretical approaches used”.

Hicks (2016) explores how theory is conceptualised and discussed in social work research published between 1968 and 2016, and links the question of what counts as theory to practical, moral, and epistemological dimensions that have consequences for social work in general. Hicks finds that most discussions about theory involve definitional ambiguity, or questions about the utility of theories. Less frequently, theory is discussed related to empirical research, which Hicks (2016:402) argues makes “most of the work on theory and social work … speculative or applied”. Hicks (2016:400) further concludes that theoretical definitions sometimes are being “provided without reflexivity, so that the version of theory presented may appear factual rather than based on any kind of explicit epistemological, ethical or ontological stance.” This, combined with a prioritisation of applicability, is described as a downplay of theory´s potential of both challenging tacit knowledge and engaging in critical thinking. While Hicks (2016:406) focuses on how theory and the use of theory in social work research are discussed, he also concludes how “relatively little empirical investigation of how theory is conceptualised and used within social work also indicates the need for further research”. Chijioke et al. (2021) also conclude that – though theory can be used to explain, describe, analyse, and predict phenomenon, aid understanding, give clarity, and bridge concepts, variables, and hypotheses – there is a need for further research on the use of theory. In this article, we decided to take a closer look at precisely this: how theory is conceptualised and used in research work.

Material and method

This study includes Swedish doctoral dissertations on social work in healthcare, written between 2008 and 2019, by authors having a social work degree. We decided to start in 2008 due to an earlier study mapping Swedish dissertations in this field until this year (Flink et al., 2008). The year 2019 further marks when board certification was introduced for healthcare counsellors (health social workers) in Sweden, requiring a new master’s program for those already having a bachelor’s degree in social work. A total of 52 dissertations were found via Social Work departments at universities and networks (for more details, see Svärd et al., 2024).

Grounded theory

The process of analysis had three stages. In the first stage, all parts of the dissertations were read, with particular attention to research aims, theory, analysis, discussion, and implications. We divided the dissertations among ourselves and conducted a preliminary analysis that was crosschecked by all authors. In cases of uncertainty, all authors read the dissertation in question, and through discussions, we jointly agreed on an analysis.

In a previous publication (Svärd et al., 2024), we showed how a clear majority of the dissertations, 90.4%, were found to explicitly mention at least one theory or theoretical concept, and in total, more than 140 theoretical concepts were identified. These concepts were classified into four groups, depending on the primary perspective taken: I) Psychological and psychosocial theories and models, II) Organisation and profession theories, III) Interactionist and critical theories, and IV) Ecological and system theories. Many dissertations combined different theories, such as profession theories and critical theories. Among compilation theses (comprising 3–7 scientific articles), different articles within one dissertation often used different theories. The results of this first stage of analysis have previously been published (Svärd et al., 2024).

During the analysis we developed an interest in how the texts not only used different theories, but how they also seemed to use theoretical concepts in different ways. The analysis thus evoked new questions and guided by what Charmaz (2012) calls a comparative, iterative, and interactive method, a second stage of the analysis began with us systematically describing the inductive process of asking more questions about the empirical material.

In this second stage of the analysis, the first author read the dissertations explicitly mentioning theory in-depth, looking specifically for references to theoretical and/or methodological concepts, trying to understand their function. Inspired by Bertilsson (2016), and with the intention to focus on the collective dimensions of research rather than the individual researcher, the analytical gaze was directed at the texts. We did not aim to, or try to, analyse the theoretical intentions of the authors, or elaborate on whether these had been met.

In line with Charmaz (2012), the ambition was to develop tentative interpretations through coding and to systematically check and refine the categories. The reading was thus inductive but guided by an overarching interest in understanding the functions of theory, aiming to test and successively raise the analytical level of this inquiry.

In coding, inspiration was taken from Charmaz (2012), invoking analytic questions from the start: looking for processes, actions, and meanings, rather than themes, asking what the theories were doing and what functions the concepts had. In this analysis, patterns emerged related to what functions the theoretical concepts filled in the texts. These patterns spanned from: (A) Dissertations where concepts had an assorting function for the analysis of the empirical material. (B) Dissertations where concepts had a contextualising or comparing function, related to different potential ways of understanding the empirical material. (C) Dissertations with a theoretical focus that permeated the texts, from aim to analysis and implication, with an elaborative function that could result in the launch of new concepts or new understandings of the phenomenon in focus. Three categories thus emerged, and in formulating them, the first author interpreted them as similar to how Wolcott (1994) has described different ways of understanding and analysing ethnographic material: Description, Analysis, and Interpretation.

As a third stage of the analysis, the first author also conducted an in-depth reading of the dissertations that had previously been analysed as not explicitly mentioning any theory. The analysis took inspiration from Bowen (2006:15), aiming “to discover, understand, and interpret what is happening in the research context”, specifically highlighting the interdisciplinary context of the subject area. Inspiration was further taken from Strauss and Corbin (1990), pinpointing the value of not only analysing what is repeatedly being present, but also what is being significantly absent. This analysis was, like the second stage, inspired by Charmaz (2012), looking for processes, actions, and meanings, and focus was specifically directed at how the aim, analysis, discussions and conclusions/implications were expressed and motivated in the texts, when not explicitly using or referencing any theoretical concepts.

In coding, one pattern emerged. This differed from (A), (B) and (C) described above, and involved a focus on empirical variables and the elaborating on hypothesised relationships between different variables, directing attention to future outcomes. The analysis thus evoked not only the question of how theories can be used, but of how theory can be understood, in all the dissertations analysed. The analytical process therefore took a step back, to challenge and refine all of our interpretations, questioning and developing the analysis in an interplay between the empirical material and theory, about theory and theorising. As a result of this third stage of the analysis, the two groups previously having been categorised related to whether they entailed any explicitly elaborating on or referencing of theory or theoretical concept - were now categorised as: (1) Understanding theory as the use of variables to predict outcomes, and (2) Understanding theory as understanding society, social interactions and social change.

Although the first author had the main responsibility for the second and third stages of the analysis, all interpretations were extensively discussed among all authors, in several meetings, until consensus was reached. Through this analytical process, the analysis matured – from being tentative and inductive, to becoming deductively analytical, as the categories became theoretically saturated (Charmaz 2017).

As a result of the three analytical stages described, the texts were divided into two overarching groups:

(1) Understanding theory as the use of variables to predict outcomes.

(2) Understanding theory as understanding society, social interactions and social change.

The use of theory within the second group was, as described earlier, analysed as divided in three categories:

(A) Using theory as Description

(B) Using theory as Analysis

(C) Using theory as Interpretation

As the categories: Description, Analysis and Interpretation, are based on the work of Wolcott (1994), we briefly describe his concepts below.

Description, Analysis, and Interpretation

Wolcott (1994) outlines three approaches to ethnographic empirical material in research: Description, Analysis, and Interpretation. These approaches are understood as asking the material different questions and positioning yourself as a researcher in different ways in relation to the material.

The first approach: Description, is described as focused on asking the question: What is going on here? In staying quite close to the empirical material, the researcher uses the theoretical concepts to “flag items with possible importance’s” (Wolcott, 1994:14).

In the second approach: Analysis, the researcher rather asks the question of how things work and/or why? with a “mind set” (Wolcott, 1994:29) where the focus becomes increasingly selective, as the findings are highlighted in relation to the analytical framework. In other words, in an Analysis, the concepts play a role in selecting material for analysis: for what becomes highlighted, as well as for how. The concepts can also be used to contextualise the material, by comparing it to other cases or concepts, or become directed at the analytical process itself.

In the third approach: Interpretation, the question becomes more focused on what is to be made of it? (Wolcott 1994). Wolcott (1994:36) describes this as a leap, or as a ”threshold in thinking and writing at which the researcher transcends factual data and cautious analyses and begins to probe into what is to be made of them”. The analysis thus becomes extended, by an explicit turn to theory, where the interpretative processes are analysed, or the analytical process is interpreted.

Results and analysis

In our analysis of the 52 dissertations, we found two overarching ways of understanding theory. The first category: Understanding theory as the use of variables to predict outcomes, is analysed as being influenced by the health sciences and is presented first.

Understanding theory as the use of variables to predict outcomes

In this category, we find dissertations where no explicitly articulated theoretical concepts were identified. However, the authors often use words that can be used as theoretical concepts, but that are also used in an everyday and perceived self-explanatory manner; most recurring are words like crisis, grief, gender, and social network, but also interprofessional collaboration, teamwork, emotions, conflict, and roles. In the dissertations analysed within this category, definitions of these words are not discussed or further elaborated on, e.g., via references to theoreticians or other scholars. This is interpreted as the words being used as variables, alongside other variables, e.g., age. In these dissertations, the aim is generally expressed in line with an endeavour to understand different variables, such as:

With a growing number of women surviving breast cancer each year in combination with limited healthcare resources, clinics are increasingly pushed to optimize psychosocial support as life after a breast cancer diagnosis not only is a question of survival, but also a question of how well one survives. In light of this pressure, it becomes important to gain an understanding of women's own psychosocial support needs. Today, clinics remain unsure how to optimize psychosocial interventions. This may in part be caused by a knowledge gap of women's own preferences. (Stinesen-Kollberg, 2014:iv)

The example quoted above is analysed as linking previous research and clinical experiences to the phenomenon “psychosocial support”. This could, like grief or crisis, be understood as a theoretical concept, but in the dissertation quoted above, it is not being defined or referenced but rather linked to “how to optimize psychosocial interventions” by gaining knowledge on “women’s own preferences”. It is further stated that the dissertation assumes hypotheses, as expressed in one of the articles included:

… we hypothesized the following: (1) there were personal and healthcare related factors associated with low psychological well-being, and (2) the mothers with the youngest children more often experienced a need for psychosocial intervention compared with those mothers who had older children and children not living at home. (Stinesen-Kollberg et al., 2013:2117)

This directing of attention to “factors associated with” and “more often experienced a need for” is analysed as an understanding of theory being associated with empirical variables, and a hypothesised relationship between these variables (cf. Portney and Watkins, 2015), rather than theory as a way of understanding society or social relations (cf. Abend, 2008).

Another example of a variable used is gender. Gender often functions as a theoretical concept in the dissertations analysed (e.g. Ekman, 2018; Sjödin et al., 2017), but within this category of dissertations, gender is sometimes rather used as a variable. One example can be found in Krekula et al. (2009:349), where the authors in the Discussion state that: “women are more inclined to talk about emotional issues”, followed by referring to previous research on gender differences in family communication about organ donation. Gender is in this case thus analysed as being used as a variable, becoming a synonym to how one could use “sex”.

Another example is found in Stenström (2008:86), where the word ethnicity appears alongside age and gender as subheadings under the main heading “Background factors regarding heavy drug users in the big cities” (2008:81). Ethnicity is not elaborated on as a theoretical concept, but the subheading is followed by the description “In Sweden it is estimated that nearly 12% of the population are born abroad” (2008:86). The category ‘country of birth’ is re-worded here as ethnicity and treated as a variable, rather than as a theoretically defined concept.

In these dissertations, the analysis of the empirical material is generally directed at previous research, which is presented as background as well as used as a reference, primarily to compare the results of the study: confirming or contradicting them. Previous research, as well as the empirical material, are thus used as analytical points of reference and also as basis for implications; as in the example mentioned earlier in the endeavour to explore “how to optimize psychosocial interventions” (Stinesen-Kollberg, 2014:iv). This analytical focus, on gaining new and useful knowledge by summarising different variables and drawing conclusions, is analysed as an understanding of theory focused on the use of variables to predict, and affect, future outcomes, such as in another previously mentioned example, by Krekula et al. (2009:350):

… future campaigns should be designed to reach different subgroups, focusing on motivating people to inform about their decision on organ donation – to finally go from words to action.

This way of understanding theory, as the use of variables to predict outcomes, is analysed as predominantly influenced by the health sciences rather than the social sciences.

To sum up: Understanding theory as the use of variables to predict outcomes is characterised by an understanding of theory that is similar to the understanding of a hypothesis, based on previous research and clinical experience, where the aim of the knowledge produced is to better predict and impact future outcomes. This can be understood as based on an ideal of analysis to build knowledge from testing and comparing pre-set variables thereby gaining better support for empirically based generalisations. Theory can thus be used synonymous with hypothesis, and theorising can be analysed as the enabling of predicting outcomes, by testing your hypothesis and building empirical generalisations. The potential knowledge gained from Understanding theory as the use of variables to predict outcomes can thus be understood as: to know more, is to be able to do better.

Understanding theory as understanding society, social interactions and social change

A clear majority (90.4%) of the dissertations included in this study have been analysed in the second group: Understanding theory as understanding society, social interactions, and social change, characterised by the explicit use of theoretical concepts. This understanding of theory is analysed as primarily influenced by the social sciences rather than the healthcare sciences. The category is in turn divided into three different ways of theorising, presented below: Using theory as Description, Analysis, and Interpretation.

Using theory as Description

Description is understood as theoretical concepts filling the function of presenting the empirical material in a certain way. The approach, which in the example below is stated as “to characterize”, is sometimes explicit in the study´s aim:

The aim of this study was to characterize young dating violent offenders (DVO), and to compare them to the general population and to young offenders with violent crimes directed against other victims. (Sjödin et al., 2017: 83)

In these dissertations, the aim is generally not presented theoretically, even if theories are mentioned, as in the dissertation quoted above, e.g., social learning theory. When presenting the concepts used, they are not discussed or compared with other concepts but rather used to sort and describe the empirical material. The concepts are often explained in a straightforward manner:

Antonovsky’s salutogenic model is a global orientation reflecting the extent to which an individual is able to use his or her general resources in handling stressful situations. … Sense of coherence is a personal resource (Anmyr, 2014:13)

In the example above, “Sense of coherence” is presented and used as a theoretical concept; it is referenced to a theoretician, Antonovsky, and placed within a broader theoretical framework of salutogenesis. This choice of concept is further clearly stated, in the same straightforward manner, as having been “seen and chosen”:

When planning this thesis, the children’s SOC (CSOC) was seen and chosen as an important individual resource (Anmyr, 2014:13).

In using theory as Description, the researchers’ analysis further remains close to the empirical material and often functions as a way to describe an understanding of the area in focus. The concepts are thus used to “flag items with possible importances´” (Wolcott, 1994:14), for instance regarding how to understand expressions of grief as strategies of coping:

The process of managing living with cancer in the family can change over time in accordance with changing circumstances. The time when the threat appears can affect their psychological well-being. Cognitive theory focuses on what the person is thinking, doing or feeling in a specific situation [81]. It is possible to distinguish between problem-focused and emotion-focused coping. Problem-focused coping implies grappling with the problem by acting or seeking information. Emotion-focused coping implies regulating one’s distress by avoiding feelings or situations reminding one of the source of this distress, and it can also be a question of seeking support [81-83]. (Sjölander 2012:19-20)

The theoretical concept “coping” is here analysed as being used to show or suggest to the reader a certain understanding of the world, in this case of how to understand the “process of managing living with cancer”. The two different kinds of coping used as theoretical concepts are described and references are included, in a way that makes the use of the concepts similar to “facts” found in the empirical material: the concepts fill the explicit function to describe.

A Description is sometimes followed by articulated implications for clinical practice, or conclusions directed at applicability: they might, for example, state that professionals ought to be aware of the described problem and display an understanding of the reactions that they might face from patients and/or loved ones. Or, as below, highlight the need to elaborate working methods in relation to specific problems:

Brief alcohol interventions may result in a reduction of AUDIT score to a small extent in psychiatric patients with hazardous or harmful alcohol use. Results suggest that BI may be of some value in the psychiatric outpatient setting. Still, more profound forms of alcohol interventions with risky-drinking psychiatric patients need elaboration. (Nehlin et al., 2013:1)

In this example, focus is directed at a brief intervention program (BI), which is described as having its theoretical base in Motivational Interviewing (Nehlin et al., 2013:1). However, this theoretical base is described as a background, and it is not further elaborated on, either in general or in relation to what impact the theory might have on the results or implications presented.

To sum up: in the dissertations using theory as Description, the aim is generally not theoretically defined or outlined. Instead, the aim tends to be to gain knowledge in a general sense. The phenomenon in focus is framed by a theoretical concept that is defined, but not further elaborated on. When clinical implications are presented, concepts are understood as fulfilling an informing function: telling professionals new “facts” regarding the field in focus. The potential knowledge gained from using theoretical concepts as Description can thus be understood as: from facts informing, to act.

Using theory as Analysis

An Analysis is, compared to a Description, understood as characterised by theoretical concepts being used in a systematic manner throughout the texts. This is often explicit in the aim, e.g., to “describe and analyse” the phenomenon in focus, with stated references to theoretical, and often also methodological, concepts:

The aim of the thesis is to describe and analyse how a group of women experience that their every-day lives are affected during and after primary breast cancer treatment. … The analyses of the interviews were inspired by grounded theory and narrative analysis. (Lilliehorn, 2013:v)

Theoretical concepts are, like above, often mentioned already in the aims and research questions, or when presenting the dissertation and its endeavours. The concepts can also be presented as something the researcher makes use of conceptually:

In this thesis, I have been inspired by and have used several theoretical frameworks, to “look at” and broaden the conceptual understanding when discussing the study results. (Åsander, 2010:12)

The concepts are defined, referenced to a theoretician, and used in an analytical manner that goes beyond merely sorting the material. Their functions are rather to compare or contextualise different understandings of the phenomenon in focus. The concepts are also often clearly stated as an inspiration:

I have primarily been inspired by Aspers (2007), who suggests that a qualitative empirical material can be understood from a range of theoretical concepts, which in themselves highlight different aspects of the research question in focus, thereby enabling the empirical material to be seen in a new context. (Ritenius Manjer, 2017:48-49)

The analysis can still be close to the empirical material, but the theoretical concepts make the approach to the material more selective in analysing what, related to the methods used, is understood as e.g., manifest, latent, or absent in the material. The concepts play an articulated role in how the material is presented, regardless of whether the methodology is deductive, inductive, or abductive. The analytical processes are thus described and elaborated, with the same analytical gaze, or in the words of Wolcott (1994:29), a “mind set”, as the analysis itself.

Compared to a Description, where reactions to difficult situations can be described as coping-strategies, an Analysis involves further elaborating on different possible theoretical understandings of the phenomenon, such as when the authors quoted below discuss potential understandings of “hope” and “denial”:

The definition of hope can vary but is an idea of a positive orientation toward the future. [7,8] Hope is different from denial. Denial is a psychological defense mechanism in which the person does not take in the existing reality and suppresses it. [9] (Benkel et al., 2010:1120)

As will be shown below, an Analysis is further often followed by clinical implications, following or as a part of an analysis, that are theoretically rather than empirically motivated. The implications outline different potential understandings of the phenomenon that has been discussed and based on the theoretical analysis of the empirical material, the authors advocate for the benefits of one of them. Such as when the authors above continue writing:

The study shows that the manner in which the coping strategies are used is individual and also depends on how loved ones can cope with the concept of a dying person with whom they are very close. When loved ones have a need for support outside their personal network, it is important to understand that this need is directly related to coping strategies and that it is not a result of denial. (Benkel et al., 2010:1119)

When using theory this way, the concepts become a part of the analysis. They are elaborated on in relation to the empirical material, and vice versa, implying and sometimes clearly expressing an underlying theoretically anchored assumption of different potential ways of understanding and making sense of the world and of the material. This is analysed as using theory as a way of understanding, in this case, social relations.

To sum up: The analytical gaze in an Analysis is understood as not aiming to lay out the facts, as in a Description, but as an interplay between different potential theoretical understandings, both ontologically and epistemologically – we can understand the world in different ways, as well as understand the analytical work in different ways. Theories are used to help understand how, often expressed in a research aim to describe and analyse the phenomenon in question. The potential knowledge gained from using theoretical concepts as Analysis can thus be understood as: from enhanced understanding, to theoretically informed decision.

Using theory as Interpretation

As outlined above, an Analysis is understood as a more thorough use of theoretical concepts, compared to a Description; a “mindset” (Wolcott 1994:29) where the empirical material is presented more selectively, compared and/or contextualised. In an Interpretation, the analytical focus takes yet another step, and the analysis becomes extended. Generally, the theoretical concepts, such as the below mentioned “power” and “how clients are constructed”, are expressed in the overall aim:

The general aim has been to contribute to an understanding of the everyday practice of maintenance treatment, how power is exercised and how clients are constructed in a local, outpatient treatment context. (Petersson, 2013:21)

In this example, the theoretical concepts are explicitly described as being used to “contribute to an understanding of the everyday practice”. In the dissertation, the concepts are thoroughly elaborated on and function as the common thread throughout the text. The use of the concept of power, a word that can also be used in an everyday and atheoretical fashion, is for example described theoretically already in the abstract:

Foucault’s works on power and the distinction between discipline and border control occupy a central place in the analysis throughout the dissertation (Petersson, 2013).

Further, in dissertations using theoretical concepts as Interpretation, a general leap is taken in the analysis, from the empirical material to theoretically motivated questions of what is to be made of the knowledge produced – a “threshold in thinking” (Wolcott, 1994:36) that is generally clearly defined and described by the authors as an integral part of the research process, rather than as isolated implications:

My main hope is for the knowledge produced in this thesis to be useful to professionals who meet such young mourners of suicide in their practice, and perhaps more importantly that this knowledge can contribute constructively to parentally suicide-bereaved youths’ own management of grief, as they become aware of how the telling of their experiences has the potential to construct alternative meanings and identities from experience – and to resist the prevailing ‘suicide-stigma’ in society. (Silvén Hagström, 2016:xx)

Our analysis of the text quoted above, and our understanding of it using theory as Interpretation, builds on the text focusing on how “the telling of … experiences” can help “construct alternative meanings and identities”. This is in turn related to the theoretical concepts of “grief” as well as of “suicide-stigma” – concepts that are themselves analysed, rather than presented, in the dissertation.

Doing an analysis as Interpretation can also result in developing new concepts, such as when Järkestig Berggren (2010), inspired by a named theoretician, offers a new concept to understand the professionalism of a specific professional group:

The PO [personal ombudsman] services brought about a new idea of professionalism, suggested in conclusion, which could be called ”user-mandated-professionalism” in an analogy with the logic of organizational professionalism and occupational professionalism as described by Julia Evetts (2006). User-mandated-professionalism is then denoted first and foremost by the lack of an academic knowledge base, but on the contrary has a knowledge base built together by the professional PO and the service user. (Järkestig Berggren, 2010:69)

Drawing from the empirical material and theoretical concepts, the new concept of ”user-mandated-professionalism” is here launched as a result of the theoretical analysis – offering a new concept to understand the specific professional group, but also describing a potentially more general – and theoretically based – “new idea of professionalism”.

The interpretative gaze can also be directed at the analytical process itself, particularly when the dissertation has a strong focus on methodology or has an articulated theoretically explorative aim. However, an Interpretation should not be understood as distanced from practice. An analysis as Interpretation can result in presenting clinical implications derived from a generation of theory, and as a result of the interpretative analysis. These implications can be directed at professionals, but they can also be directed at the scientific community:

Regardless of whether the actions are understood as a public health problem, an individual problem, or a social problem, my conclusion is that it is not the actions in themselves that are the primary problem. The youth’s stories indicate that what is often primarily needed, is a change of the problematic social situation that led to the self-harming acts. Social work as an academic discipline, focusing on peoples’ stories, social context and interaction, has the potential to develop the research field of self-harm, to broaden the perspectives and create an understanding that goes beyond the established medicalization of this social phenomenon. (Ekman, 2018:80)

In the example quoted above, different potential understandings of the phenomenon in focus are presented: “a public health problem”, “an individual problem”, or “a social problem”. The problem is then however also framed within an overarching conceptualising of social sciences and social work as exploring “peoples’ stories, social context and interaction”, that is linked to the theoretical concept of “medicalization”, which is further elaborated in the dissertation, resulting in a theoretically motivated implication.

To build knowledge in this case thus goes beyond the testing and comparing of pre-set variables to gain better support for empirically based generalisations, such as in Understanding theory as the use of variables to predict outcomes. When doing an Interpretation, the scope of the analysis becomes wider, by articulating a theoretically motivated question of “what is to be made of” (Wolcott 1994:36) the knowledge produced. The focus lies on theoretically motivated transferability, rather than generalisations in a more objectivist sense. Both ways of understanding the possibility to build on knowledge with knowledge might mean for theorising to “go somewhere” (Krause, 2016:28), but perhaps to different places.

To sum up: using theory as Interpretation can be described as the text being immersed in theory and theoretical concepts. Theoretical concepts do not make up a part of these dissertations; instead, they permeate the texts and are generally mentioned already in the aim. The concepts are not used as tools for laying out the facts, as in Description, or for comparing different perspectives, as in Analysis, but as a way to “transcend factual data” (Wolcott 1994:36), by thoroughly analysing the interpretative processes, and often also interpreting the analytical processes. The potential knowledge gained from using theoretical concepts as Interpretation can thus be understood as: from interpreting, to changing the understanding of a phenomenon.

Discussion

The framework presented in this article can serve as a tangible, yet creative, conceptualisation of different ways of using theory, which hopefully can contribute for theorising to “go somewhere” (Krause, 2016:28). The framework may be especially useful for research supervisors and students as a focal point, both when elaborating on emerging research and when refining almost completed work. It is particularly relevant to the articulation and adjustment of research aims and questions, as well as to linking these to the analysis and conclusions. For theorising to be possible, Abbott (2004:3) has emphasised the need for a “conversation between rigour and imagination,” which calls for creativity and the development of skills or craftsmanship. However, academia can often seem both mysterious and rule-governed, and the kind of academic self-confidence that theorising requires may not be easily gained.

In line with Järkestig Berggren et al. (2021), our intention with the framework presented is not to take a stand for any general ideal for every researcher, in every field, for every research question. However, we hope that the framework contributes to a more in-depth discussion about theorising that can be applicable in many fields, maybe particularly in research areas that, like health social work, are characterised by ideals of applicability and theory-practice transference (cf. Alley et al., 2010; Zambo, 2014).

Different approaches to theory and theorising can be relevant, depending on what knowledge we wish to gain. In health social work, a need for knowledge about which and how different variables intersect concerning social inequalities and health can involve research understanding theory as the use of variables to predict, and thereby be able to change future outcomes. However, researchers must be conscious of the risks of over-simplifications when using variables. Not doing so could result in both naïve empiricism (Carleheden, 2016) and, worse, the impeding rather than the promoting of social change and development, social cohesion, and the empowerment and liberation of people.

Understanding theory as understanding society, social interactions and social change, and using theory as Description, Analysis and Interpretation, needs to be adapted with the same thorough awareness. Using theory as Description can be relevant, such as when knowledge needs to be gained about how people experience a health social work intervention, linking theory to practice. However, using theory as Description should not be chosen because it seems like an easier way to go than an Analysis or an Interpretation. Describing the world and someone’s experiences of it, such as when aiming to understand the effects of a social work intervention, is an important and difficult endeavour that calls for discussions on e.g., the theoretical basis of the intervention, to avoid research becoming over-simplifying, over-generalised, or speculative (cf. Hicks, 2016). One example could be analysing crisis interventions that do not encompass social or economic factors and fail to address the potential effects of this in the analysis; taking the theoretical basis of the intervention as a given.

When using theory as Analysis, the same caution must be used. An Analysis can be fruitful to enhance our understanding of a phenomenon, and in health social work this could entail contributing to the multidisciplinary context that might be characterised by a focus on the physiology of the individual without its social embedding. Adding the social dimensions of health can render knowledge improving the quality of care. However, the aims expressed should be in line with the potential knowledge gains. When different theoretical concepts are used, such as to analyse how the world and our analytical work can be understood in different ways, researchers need to be conscious of the choices made, and how the different theoretical concepts might be in conflict related to their ontological points of departure.

Using theory as Interpretation can be fruitful, perhaps particularly to change the understanding of a phenomenon. In the material analysed here, this way of theorising is predominantly used when criticising or challenging pathologisation, medicalisation or psychologisation of social problems. However, regardless of which theories are used, when using theory as Interpretation, the researcher needs to be cautious in avoiding the research becoming characterised by theoretical apriorism and over-generalisations (cf. Carleheden, 2016). Another challenge with this endeavour, that might affect health social work and other fields characterised by applicability and theory-practice transference, is if research does not consider practitioners’ working conditions, or if practitioners are not given the opportunity to read and critically discuss the research and how to make use of it. The knowledge gained might still be of importance but not become used to the benefit of social work clients.

Considering that social work is both a science and a practice-based profession, efforts to achieve theory-practice transference are required from all actors involved, and regardless of which theory is used, or how, it is essential to foster the development of knowledge through the interplay between research and practice. Inspiration can be drawn from James Secord (2004), who criticizes a tendency to focus too much on the emergence of new knowledge, rather than on how knowledge moves, develops, and acquires new meanings. Knowledge is often viewed as fixed and static, produced and conveyed by some, and received by others. Secord (2004) instead emphasizes how knowledge circulates, and how communication is integral to the creation of knowledge. Future research could therefore focus not only on the use of theories, but also on the movement and transformation of knowledge.