Sage Journals: Discover world-class research

Abstract

The tradition of developing typologies has been prominent in research, particularly within the fields of psychology and sociology, for decades. A typology is formed by grouping cases or participants into types on the basis of their common features. Despite the prominence of typologies in research, methodological guidance on the process of developing a typology, particularly as a qualitative method for analysing data, is scarce. Ideal-type analysis is a relatively new addition to the family of qualitative research methods, which offers a systematic, rigorous method for constructing typologies from qualitative data. In our approach to ideal-type analysis, the methodology consists of seven steps: becoming familiarised with the dataset; writing the case reconstructions; constructing the ideal types; identifying the optimal cases; forming the ideal-type descriptions; checking credibility; and making comparisons. This article is a summary of our approach to conducting ideal-type analysis. We hope that this article will help researchers to consider whether using ideal-type analysis may be a suitable approach for their own studies.

Keywords

mixed methods typologies qualitative methods ideal-type analysis critical realism

Introduction

In selecting an appropriate methodology for a piece of qualitative analysis, decisions may be based on the researcher’s own training history (or that of a supervisor), the degree to which a methodology is recognised and accepted by certain institutions, or whether there is clear guidance on how a particular research method can be used. Whilst all of these factors are relevant, it has been argued that qualitative researchers would do well to take a pragmatic approach to selecting a research method, which “looks at each method in terms of what questions each can help the researcher answer—a ‘whatever works’ position” (Pistrang & Barker, 2012, p. 7). Thus, ensuring that there is a good fit between methodology and research question should be a key aspect of the decision-making process.

For many qualitative research projects, a methodology that is broadly speaking ‘thematic’ may seem appropriate, as it allows for in-depth exploration of participants’ diverging and shared experiences or perspectives. Examples of thematic approaches include thematic analysis (e.g., Braun & Clarke, 2019), grounded theory (e.g., Charmaz & Henwood, 2017), or interpretative phenomenological analysis (IPA; e.g., Eatough & Smith, 2017). Despite some important differences between these data analysis techniques, they all share a fundamental approach to analysing data across cases or participants to identify patterns in their experiences and perspectives (Elliott & Timulak, 2005). However, in taking a thematic approach, what can be lost is the specific meaning of experiences for each individual, in the context of their own life (or narrative). This is why some qualitative researchers prefer to use case study methods, for example, as this avoids the risk of de-contextualising participants’ experiences and allows for in-depth exploration of meaning within-case or in context. Yet, this approach, in turn, also has its limitations – including the fact that it reduces the possibilities for cross-case analysis and understanding.

One way of bridging the gap between within-case and cross-case approaches is through the use of typologies. A typology is “a hierarchical system of categories used to organize objects according to their similarities and dissimilarities” (Mandara, 2003, p. 132). A typology is formed by grouping cases or participants into different types on the basis of their common features, with consideration of how each unique individual represents a particular pattern of features. The tradition of developing typologies has been prominent in research, particularly within the fields of psychology and sociology, for decades. Examples of typologies in psychology include parenting styles (Baumrind, 1991), personality types (McCrae & Costa, 1987), and attachment (Ainsworth & Bell, 1970), and in sociology, economic systems (Kang, 2005) and religious organisations (Pew Research Center, 2018). In the context of psychology, where our work has been situated, typologies seek to understand human behaviour by paying close attention to individual cases (e.g., what are the unique features of this person that represent a particular personality type?), combined with a cross-case approach to explore similarities and differences between cases (e.g., how does this person relate to others in the dataset?). Yet, despite the prominence of typologies in existing research, minimal methodological guidance has been developed to explain how to construct a typology, particularly as a qualitative data analysis technique (Kluge, 2000).

In this article, we present an approach to constructing typologies: ideal-type analysis. This approach is not prominent within existing qualitative research, yet we propose that it is an important addition to the range of qualitative research methods available, as it is a systematic, rigorous approach for developing typologies from qualitative data (Gerhardt, 1994). This article is a summary of our approach to conducting ideal-type analysis. We have published a book providing a detailed guide to conducting ideal-type analysis (Stapley et al., 2021). Here, we present a brief overview of the technique, explain how this approach compares to other qualitative data analysis techniques, and present two example studies to illustrate the diverse ways in which it may be usefully applied in research.

What is Ideal-Type Analysis?

Max Weber (1904), one of the early pioneers in the field of sociology, introduced the ideal type approach. Weber regarded the ideal type - a description derived from observations of an empirical reality or a social phenomenon - as the initial step in one’s analysis of a new or under-studied topic (Psathas, 2005; Swedberg, 2018). The ideal type was in essence a hypothesis about a particular phenomenon, which different instances or examples of that phenomenon could be compared, tested, or measured against to facilitate interpretation (Gerhardt, 1994; Stuhr & Wachholz, 2001; Wachholz & Stuhr, 1999; Werbart et al., 2016). In this way, the ideal type could be construed as a ‘methodological tool’ or ‘yardstick’ used to facilitate comparisons between instances of phenomena unfolding in different time periods and places (Kvist, 2007; Psathas, 2005; Werbart et al., 2016).

Uta Gerhardt (1994) later developed a qualitative research method within sociology, drawing on Weber’s concept of the ideal type: ideal-type analysis. Though ideal-type analysis has its roots in sociology, and has often been used by psychotherapy researchers within the field of psychology (McLeod, 2011), ideal-type analysis is a flexible method that can be used with a range of different qualitative data sources and research fields. In a nutshell, it involves the systematic comparison of cases or participants within a qualitative dataset to form ‘ideal types’, or groupings of similar cases. Together, the ideal types form a typology. Through constructing ideal types, we are able to illustrate the different patterns of behaviours, thinking, and feelings that participants exhibit or describe, and compare between participants.

Ideal-Type Analysis And Epistemology

Ideal-type analysis is not underpinned by one particular epistemological perspective, but as with all research, it is important for researchers to try to be explicit and reflexive about the assumptions that they make at the outset of their study. In our own work, we have taken a critical realist perspective (Maxwell, 2010). In other words, we recognise that there is a pre-social world that exists, but knowledge of it can only be understood through our own experiences of it and so it is always subjective. From a critical realist perspective, the researcher would develop ideal types as an attempt to make sense of or explain some aspect of reality, without considering these types to be an ‘objective’ version of the world. Indeed, the word ‘ideal’ in an ideal-type analysis context does not mean ‘best’, but rather it refers, philosophically, to an ‘idea’ (McLeod, 2011; Werbart et al., 2016). In this way, we can regard ideal types as generalisations or mental representations of a social phenomenon that will never be identical with reality, but which will help to make that reality understandable (Stapley et al., 2021). Thus, in an ideal-type analysis, ‘the ideal’ is seen as something that is constructed, through a process of empirical investigation of things in the world, and it is understood as a social construct that may help our understanding of the phenomenon under review.

If we take a critical realist perspective, we accept that in constructing ideal types, the researcher inevitably brings their own perspective and view of the world to their interpretation of the data. Decisions about the validity of that interpretation would partly depend on how fully their interpretation is rooted in the data, but would also depend on how well the types constructed contribute to our understanding of reality, or help to make predictions about the world. For example, the ‘parenting styles’ identified by Baumrind (1991) are without doubt shaped by certain cultural assumptions and may not be objectively ‘true’ (Sorkhabi & Mandara, 2013), but they can nevertheless help us to make sense of certain key differences in styles of parenting, and allow us to make predictions. For example, authoritative and authoritarian parenting styles have been shown to have predictive value in terms of child academic achievement (Pinquart, 2016).

When taking a critical realist perspective, we therefore recognise that different researchers may legitimately construct different ideal types regarding the same phenomena, depending on their context and view of the world (Frommer et al., 2017; Psathas, 2005). Consequently, researcher reflexivity and transparency about the context within which ideal types are constructed is important. Moreover, when conducting an ideal-type analysis, researchers should be transparent about how they view the data that they are analysing, and remain reflective throughout the analysis process on their own role in shaping the analysis and interpretation of the data.

Comparisons to Existing Qualitative Data Analysis Techniques

Table 1 illustrates how ideal-type analysis compares to other widely used qualitative data analysis techniques. For example, ideal-type analysis differs from thematic approaches to qualitative data analysis because it does not seek to identify a set of themes across the dataset, rather it seeks to identify groupings of participants (who share similar experiences) within the dataset. Thematic approaches look for commonalities (and points of divergence) across participants’ experiences, or search for patterns of shared meaning across a dataset (Braun & Clarke, 2006). By contrast, ideal-type analysis looks for commonalities (and points of divergence) between participants’ entire accounts of their experiences. In this way, ideal-type analysis shares some common ground with narrative analysis approaches, which seek to analyse data in the context of participants’ stories or narratives (Sharp et al., 2018). However, in a narrative analysis, findings can be presented thematically and/or in terms of individual stories, whereas in an ideal-type analysis, findings are always presented in terms of participants’ individual accounts of their experiences, with a focus on how these relate to the experiences of other participants in the dataset. Furthermore, unlike other qualitative data analysis techniques, the starting point for an ideal-type analysis requires a relatively large sample (typically 30+ cases), which is required to facilitate the development of meaningfully distinct groupings of participants within the dataset.

Table 1.

Similarities and differences between ideal-type analysis and other qualitative data analysis techniques.

Features	Ideal-type Analysis	Thematic Approaches	Case Study Approaches	Narrative Approaches
Aims	To identify groupings (‘types’) of participants within the dataset to organise and understand people according to their within-group similarities and between-group differences	To identify thematic patterns across the dataset and use these themes to examine people’s shared or divergent views or experiences	To conduct an in-depth investigation of a single person, group, event, or community	To analyse data as stories to understand how people experience or make sense of life events or their everyday lives
Types of research questions	Exploratory – with the aim of developing a typology, e.g	Exploratory – with the aim of deriving themes, e.g	Exploratory – with the aim of presenting a case in-depth, e.g	Exploratory – with the aim of presenting findings from participants’ stories, e.g
	• In what ways does the emotional wellbeing of adolescents in receipt of a resilience-building intervention change over time?	• What are the lived experiences of young adults who are diagnosed with bipolar disorder in early adulthood?	• What does an individual young person’s journey through school-based mental health support look like?	• What are the stories of parents during lockdown in the Covid-19 pandemic?
	• What are the different types of reasons for psychotherapy dropout?	• What do people experience as helpful and unhelpful about a weight management programme?	• How does a public health programme designed to improve take-up of vaccinations operate?	• What are the day-to-day experiences of people living with chronic fatigue syndrome?
Types of data	Interviews/focus groups, observations, field notes/case notes, naturally occurring data	Most frequently used with interviews/focus groups	Interviews/focus groups, observations, field notes/case notes, naturally occurring data	Most frequently used with interviews
Sample size	Medium to large (e.g., 30+ cases)	Small to large (e.g., 3+ cases)	Single case	Small to large (e.g., 3+ cases)
Presentation of results	Overview of types, including illustrative cases	Overview of themes and subthemes, including illustrative quotes	In-depth understanding of a single case	Individual participant stories and/or overview of cross-story themes and subthemes

What Are The Steps To Carrying Out An Ideal-Type Analysis?

In our approach, there are seven steps to conducting an ideal-type analysis, which are outlined below. These steps are described in more detail in Stapley et al.’s (2021) book. We present an outline of each of the seven steps here. The steps are presented linearly, but in reality, the researcher may revisit steps multiple times over the course of their analysis to make refinements as necessary. Our approach draws on and builds on the methodology for ideal-type analysis outlined in previous studies conducted by Gerhardt (1994), Lindner and colleagues (Lindner, 2006; Lindner & Briggs, 2010; Lindner et al., 2006), Stuhr and Wachholz (1999, 2001), and Werbart and colleagues (Philips, Werbart, et al., 2007; Werbart et al., 2011, 2016), but also differs from these previous studies in several ways (Stapley et al., 2021). First, our approach adds a familiarisation with the dataset step at the outset of the analysis. This aligns with the first step of a thematic analysis, as described by Braun and Clarke (2006), which advocates that the time spent by the researcher in becoming familiar with the extent and content of the data collected, including recording initial reflections and ideas, is “the bedrock for the rest of the analysis” (p. 87). Our approach also specifies the development of the ideal-type descriptions as a distinct step in the analysis process, as this helps the researcher to clarify the essential features of each type. The process of comparing cases within and between the ideal types is often conducted in previous studies (e.g., Stuhr & Wachholz, 2001; Wachholz & Stuhr, 1999; Werbart et al., 2011, 2016); however, our approach formalises this as the final step in the analysis and a core part of the methodology.

Becoming Familiarised With The Dataset

The researcher begins by familiarising themselves with the dataset. Familiarisation can be achieved, for example, by conducting interviews, listening to interview audio files, and transcribing interviews.

Writing The Case Reconstructions

The researcher reconstructs or summarises each participant’s narrative (e.g., their interview transcript) into a case reconstruction. Thus, a case reconstruction is essentially a written summary or description of the data available for each participant. A case reconstruction could summarise an entire interview transcript or it could focus solely on the sections of an interview transcript that are relevant to the study aim(s) or research question(s). It could also include more than one interview transcript for each participant, such as in the context of a longitudinal study whereby interviews are conducted at multiple timepoints.

Constructing The Ideal Types

The case reconstructions are used to form the ideal types, through systematically comparing and contrasting each case reconstruction with each other. The researcher’s intention is to explore the similarities and differences between cases (participants) to identify patterns across the dataset in terms of participants’ whole accounts of their experiences or perspectives. This process leads to the formation of groups (or ideal types) of similar cases - participants with similar experiences or perspectives. There should be clear distinctions, differences, and heterogeneity between groups, with homogeneity within groups (Stuhr & Wachholz, 2001; Wachholz & Stuhr, 1999). Not all cases within each group will have had the exact same experience nor will share the exact same perspective, however there must be something fundamentally similar about the cases within each group that links them together and apart from the other groups of cases (Stapley et al., 2021). Each participant should only belong to one ideal type.

Identifying The Optimal Cases

Once the ideal types are formed, the researcher then identifies a single case reconstruction as the ‘optimal case’ to represent each ideal type. Optimal cases are those which most closely illustrate the pattern of similar cases that each group represents. The optimal case for each ideal type is the orientation point to which the researcher compares all of the other cases within that type to explore how and to what degree they resemble the optimal case (Philips, Werbart, et al., 2007; Stuhr & Wachholz, 2001; Wachholz & Stuhr, 1999; Werbart et al., 2011, 2016). This process facilitates the researcher in conducting a thorough exploration of the characteristics, perspectives, and experiences of all participants who represent each ideal type.

Forming The Ideal-Type Descriptions

The researcher then names and develops a detailed description for each ideal type. The description closely represents the optimal case and in this way illustrates the group of similar cases to which it belongs (Stuhr & Wachholz, 2001; Wachholz & Stuhr, 1999). The description can also include elements of other cases within the group as necessary to comprehensively describe the characteristics of that ideal type. Each case within an ideal type will represent or reflect the ideal-type description to a greater or lesser degree (Kühnlein, 1999).

Checking Credibility

Using the ideal-type descriptions, an independent researcher - who has not been involved in the analysis so far - then attempts to regroup the cases into the ideal types (Philips, Werbart, et al., 2007; Werbart et al., 2011, 2016). The purpose of this is not to check inter-rater reliability, but to ensure that the descriptions of the ideal types are appropriately grounded in the data, and that the descriptions are clear and differentiated enough to allow someone else to use them to be able to group the existing cases into the proposed ideal types. Where there is disagreement, this should be discussed until consensus can be reached (Philips, Werbart, et al., 2007; Werbart et al., 2011, 2016). This may lead to refining of the definitions of the ideal types, regrouping cases, rewording ideal-type descriptions, or identifying additional types. The aim of this step is to assess the clarity of the ideal types, rather than to ascertain the ‘correctness’ of the typology, including checking that interpretations adequately reflect the data.

Making comparisons

The write-up of the study should include a summary of the similarities (and differences) between the cases within each type, compared with each other and with the optimal case. It should also include a summary of the differences (and similarities) between the ideal types themselves. To facilitate comparisons of the cases within and between the ideal types, the researcher could also draw on additional data collected about participants, such as demographic or clinical outcomes data, where appropriate.

Examples Of Studies Using Ideal-Type Analysis

To illustrate how ideal-type analysis can be used by researchers, we present two examples of studies using this method. Both studies were underpinned by a critical realist perspective, taking the view that participants’ interviews provided a subjective window into their experiences and perspectives, with details selected by participants in the moment of their interview, depending on what they remembered or were willing or able to share. From this perspective, it was also understood that the researchers’ analysis of participants’ accounts would inevitably be shaped by their own prior experiences, knowledge, and research interests. For example, the research teams in both studies were all educated within the field of psychology in the UK, which may bring with it certain assumptions about the psychological world.

In the first study, the intention was to examine the subjective experiences of parents over the course of their teenage child’s journey through depression and their therapy at child and adolescent mental health services (CAMHS), as part of a wider PhD study interested in the perspective of parents of adolescents diagnosed with depression. In the second study, the intention was to explore the reasons that adolescents gave for why they dropped out of their therapy for depression at CAMHS, in the context of a wider PhD study about dropout from adolescent psychotherapy. In this study, reflexive discussion within the research team led to an awareness that the term ‘dropout’ already frames young people’s behaviour in particular ways, and this awareness helped the research team to become more open to the idea that ‘dropping out’ could have alternative, including potentially more positive, meanings.

Example 1

A longitudinal study to explore the experiences of parents whose adolescent children had received psychological therapy for depression

Our first example is taken from work undertaken by the first author (ES) as part of her PhD. This study has been published (Stapley et al., 2017), and the methodology has been described in detail elsewhere (Stapley et al., 2021). Here we provide an overview of the study and its findings, using this example to demonstrate ideal-type analysis as a qualitative method.

This study aimed to examine patterns in the lived experiences of 28 parents of adolescents who had been referred for psychological therapy for depression at CAMHS in England, in the context of a randomised controlled trial (RCT; Goodyer et al., 2017). The dataset for this study consisted of three interviews with the adolescent’s mother or father, or both parents, over the 3-year period of the study (one interview per year). The first interview (Time 1) took place around the point of the adolescent’s referral to CAMHS, the second interview (Time 2) took place approximately 36 weeks after Time 1, and the third interview (Time 3) took place approximately 1 year later.

This study had several elements that contributed to the decision to use ideal-type analysis to analyse the data. First, the aim of this study was to look at patterns across parents’ whole accounts of their experiences over the course of their teenage child’s depression and therapy at CAMHS. In line with this, ideal-type analysis is used to create heterogenous groupings of cases or participants with similar experiences – each group or ideal type represents a different pattern of experiences within a dataset. Second, the dataset for this study was relatively large: 84 interviews in total across 28 families over the three years of the study. Sample sizes in studies that use ideal-type analysis are typically large and heterogeneous to provide sufficient variation and breadth of cases to enable the researcher to group cases into distinct types (Stapley et al., 2021). Third, the dataset in this study was longitudinal, which can present a dilemma for qualitative researchers in terms of whether to use a trajectory or a repeated cross-sectional approach to analyse the data (Grossoehme & Lipstein, 2016). Ideal-type analysis has been used in previous studies to make sense of longitudinal qualitative data and to explore change over time in participants’ experiences and perspectives (e.g., Werbart et al., 2011, 2016).

The seven steps outlined in the previous section were followed in this study to conduct the ideal-type analysis. The steps and findings of this study are also described elsewhere (Stapley et al., 2017, 2021), but are summarised here. ES familiarised herself with the data through conducting the interviews and quality checking interview transcripts. Case reconstructions (approximately one page in length each) were then written by ES to describe and summarise each parent’s experiences as discussed during their interviews at Times 1, 2, and 3. The case reconstructions were written after all of the data had been collected, with each participant’s interviews summarised chronologically in their case reconstruction. This meant that each of the 28 case reconstructions (one per family) consisted of three subsections, each representing one timepoint. This process involved constant comparison between the original interview transcripts and the summary document (the case reconstruction) to ensure that the essential features of each participant’s experience at each timepoint were captured.

Then, through comparing and contrasting the case reconstructions with each other, the research team developed the ideal types. The research team took a reflexive and consensus-building approach throughout their analysis. This involved keeping an audit trail of the analysis process and having discussions with each other about the assumptions that each researcher was making or bringing to the dataset, and the possible bearing of these on the analysis, to ensure that the types remained grounded within or representative of the data. The research team also explored together the rationale behind different possible ways of grouping participants to determine the best fit for the dataset. This ultimately led to the development of three ideal types, which captured three distinct patterns in parents’ experiences over the 2-year period of the study. An optimal case for each of the three ideal types developed throughout this process was selected, which most clearly illustrated each of the three patterns. Names and descriptions for each of the three ideal types were then developed, with the optimal case for each in mind.

An independent researcher then attempted to regroup all of the case reconstructions into the ideal types, using the ideal-type descriptions. The percentage agreement between the original researchers and the independent researcher on which type each case belonged to was 75%. The descriptions of the ideal types were refined accordingly (for instance, more detail was added for clarification) following discussions with the independent researcher, and agreement was then reached on the seven cases that the original researchers and the independent researcher had assigned to different ideal types. Thus, the calculation of percentage agreement in this study was used as a tool to facilitate discussion, reflection, and refinement where necessary of the typology, rather than as an indication of its ‘correctness'.

The first type was the ‘learning curve’ parents (N = 12), who had found the support that they and their child had received from CAMHS to be life-changingly helpful, as they developed new perspectives over time and adapted accordingly. The second type was the ‘finding my own solutions’ parents (N = 6), who tended to find CAMHS disappointing and ultimately ended up solving their child’s difficulties themselves. The third type was the ‘stuck’ parents (N = 10), who did not find CAMHS particularly helpful, but who also did not know where to turn next for further help for them and their child.

Finally, similarities and differences were explored between the three ideal types in terms of participant demographic data. This comparison process showed that there were demographic differences between the parents representing each type. For example, “17% of the learning curve parents had experienced mental health issues themselves or described their child’s other parent as having experienced mental health issues, compared to 33% for the finding my own solutions parents, and 70% for the stuck parents” (Stapley et al., 2017, p. 1436).

Example 2

A mixed methods study to explore types of treatment dropout in adolescents receiving psychological therapy for depression

Our second example is taken from work undertaken by the second author (SOK) as part of her PhD. This study has been published (O'Keeffe et al., 2019), and the methodology has been described in detail elsewhere (Stapley et al., 2021). Here we provide an overview of the study and its findings, using this example to demonstrate how ideal-type analysis is compatible with mixed methods studies.

This study aimed to investigate the concept of treatment dropout, to understand the reasons why young people drop out of psychological therapies. Treatment dropout is defined as the premature ending of therapy when the client decides to end treatment without the agreement of their therapist (Warnick et al., 2012). However, the concept of dropout is poorly defined in the literature, and little is known about the meaning behind adolescents’ decisions to stop treatment. This study sought to investigate whether there are more meaningful categories of dropout that could be derived from qualitative data.

In this study, the sample consisted of 32 adolescents who had been offered psychological treatment in the context of the same RCT referenced in the previous example, and who had been classified as having dropped out of treatment by their therapists. The adolescents (and their therapists) had been interviewed about their experiences of the therapy, including how the therapy had ended. Data were analysed following the seven steps to conducting an ideal-type analysis outlined earlier in this article. This led to the construction of three dropout types: ‘dissatisfied’, ‘got what they needed’, and ‘troubled’. ‘Dissatisfied’ dropouts were those adolescents who reported stopping therapy because it failed to meet their needs (N = 18); ‘got what they needed’ dropouts were those who reported stopping therapy because they felt better (N = 10); and ‘troubled’ dropouts were those who reported stopping therapy because they felt that it was not the right time for them to engage in treatment due to having too much else going on in their lives at that time (N = 4).

Having followed the steps described in this article to construct this typology, the research team began to make hypotheses about the dropout types during the final step of the analysis (the ‘making comparisons’ step). They hypothesised that ‘got what they needed’ dropouts would have better clinical outcomes compared with the ‘dissatisfied’ dropouts. Clinical outcomes were measured in the RCT using the Mood and Feelings Questionnaire (MFQ), a measure of self-reported depression (Angold et al., 1987). The research team found that ‘got what they needed’ dropouts had better clinical outcomes compared with ‘dissatisfied’ dropouts at 36-weeks post-therapy, providing tentative support for this hypothesis (O'Keeffe et al., 2019). These findings must be viewed with caution due to the small sample size, yet they provide an illustration of how quantitative data may be integrated into the analysis at this stage to compare cases across ideal types.

In a subsequent study, SOK conducted further exploration of the dropout types, this time to retrospectively explore the interactions between ‘dissatisfied’ dropouts and ‘got what they needed’ dropouts with their therapists prior to the therapy ending (O'Keeffe et al., 2020). Specifically, the research team were interested in whether there were markers of ruptures in the therapeutic alliance that were indicative of different types of treatment endings. Ruptures are defined as deterioration in the alliance between a client and therapist (Safran & Muran, 1996). Audio recordings of therapy sessions for those same cases were rated using observational measures of the therapeutic alliance and instances of rupture-repair in the therapeutic alliance. In integrating the ideal types with this observational data, the research team indeed found greater unresolved ruptures in the sessions prior to ‘dissatisfied’ dropout, compared with ‘got what they needed’ dropout. This finding provided further evidence that the ‘dissatisfied’ and ‘got what they needed’ dropout types, constructed using ideal-type analysis of semi-structured interviews, should be regarded as distinct phenomena – given that different interactional patterns were observed in their therapy sessions prior to them stopping treatment (O'Keeffe et al., 2020).

Together, these studies show how various forms of data, for instance demographic information, outcome data, and observational data, can be integrated into an ideal-type analysis to provide a rich understanding of a complex psychological phenomenon.

Conclusions

We argue that ideal-type analysis, while a relatively unknown and underused method within qualitative research, offers an important contribution to the range of qualitative research methods available. The construction of typologies is not new in research, yet there has been a lack of guidance in the literature on how researchers might go about constructing a typology. We propose that ideal-type analysis provides a clear, rigorous, and systematic approach to constructing typologies using qualitative data. The use of case reconstructions allows researchers to work with a relatively condensed form of qualitative data, which facilitates the management and analysis of large qualitative datasets. The emphasis on checking credibility ensures that interpretations are grounded within the data. The identification of optimal cases helps to illustrate the core features of the ideal types. These elements can make it a particularly useful approach, including for those less experienced in qualitative methods – as each step can be easily recorded, audited, and overseen by other team members and supervisors.

As demonstrated in our case study examples given here, ideal-type analysis is a flexible approach that can be adapted to fit with the aims and data of a given study. Often researchers will need to decide between carrying out a cross-case or within-case approach to analysing qualitative data; with the former risking losing the voice of the individual, while the latter risks overstating the importance of the experience of one individual. Aligning with Gerhardt (1994), we argue that ideal-type analysis bridges the gap between cross-case and within-case approaches to data analysis, as its focus is on exploring similarities and differences in participants’ whole accounts of their experiences. We hope that this article will help researchers to consider whether using ideal-type analysis may be a suitable approach for their own studies.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Emily Stapley

References

Ainsworth

M. D.

Bell

S. M

(1970). “Attachment, Exploration, And Separation: Illustrated By The Behavior Of One-Year-Olds In A Strange Situation”. Child Development, 41(1), 49–67. https://doi.org/10.2307/1127388.

Angold

Costello

E. J

Pickles

Winder

(1987). The Development Of A Questionnaire For Use In Epidemiological Studies In Children And Adolescents. MRC Child Psychiatry Research Unit.

Baumrind

(1991). “The Influence of Parenting Style on Adolescent Competence and Substance Use”. Journal of Early Adolescence, 11(1), 56–95. https://doi.org/10.1177/0272431691111004.

Braun

Clarke

(2006). “Using Thematic Analysis in Psychology”. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa.

Braun

Clarke

(2019). “Reflecting On Reflexive Thematic Analysis”. Qualitative Research in Sport, Exercise and Health, 11(4), 589–597. https://doi.org/10.1080/2159676X.2019.1628806.

Charmaz

Kathy.

Henwood

Karen

(2017). The SAGE Handbook of Qualitative Research in Psychology Grounded Theory Methods for Qualitative Psychology (pp. 238–256). https://dx.doi.org/10.4135/9781526405555.

Eatough

Smith

(2017). Interpretative Phenomenological analysis. The SAGE Handbook of Qualitative Research in Psychology, 2nd Edition, 193–209. https://doi.org/10.4135/9781526405555.n12.

Elliott

Timulak

(2005). “Descriptive And Interpretive Approaches To Qualitative Research”. In Miles

Gilbert

(Eds), A handbook of research methods for clinical and health psychology (pp. 147–159). Oxford University Press.

Frommer

Joerg.

Langenbach

Michael.

Streeck

(2017). “Qualitative Psychotherapy Research in German-Speaking Countries”. Psychotherapy Research, 11(2), 65–68.

10.

Gerhardt

Uta.

(1994). “The Use of Weberian Ideal-Type Methodology in Qualitative Data Interpretation: An Outline for Ideal-Type Analysis”. Bulletin de Méthodologie Sociologique, 45(1), 74–126. https://doi.org/10.1177/075910639404500105.

11.

Grossoehme

Lipstein

(2016). “Analyzing Longitudinal Qualitative Data: The Application Of Trajectory And Recurrent Cross-Sectional Approaches”. BMC Res Notes, 9(1), 136. https://doi.org/10.1186/s13104-016-1954-1.

12.

Kang

B. W.

(2005). “An Examination Of An Intermediate Sociopolitical Evolutionary Type Between Chiefdom And State”. Arctic Anthropology, 42(2), 22–35. http://www.jstor.org/stable/40316644.

13.

Kluge

Susann.

(2000). Empirically Grounded Construction Of Types And Typologies In Qualitative Social Research. http://www.qualitative-research.net/fqs/.

14.

Kühnlein

Irene.

(1999). “Psychotherapy As A Process Of Transformation: Analysis Of Posttherapeutic Autobiographic Narrations”. Psychotherapy Research, 9(3), 274–287. https://doi.org/10.1093/ptr/9.3.274.

15.

Kvist

Jon.

(2007). “Fuzzy Set Ideal Type Analysis”. Journal of Business Research, 60(5), 474–481. https://doi.org/10.1016/j.jbusres.2007.01.005.

16.

Lindner

(2006). “Suicidality In Men In Psychodynamic Psychotherapy”. Psychoanalytic Psychotherapy, 20(3), 197–217. https://doi.org/10.1080/02668730600868948.

17.

Lindner

Briggs

(2010). Forming Ideal Types By Understanding: The Psychoanalytic Treatment Of Suicidal Men (11, p. 12). https://www.qualitative-research.net/index.php/fqs/article/view/1278/2977.Forum: Qualitative Social Research 2

18.

Lindner

Fiedler

Altenhofer

Gotze

Happach

(2006). “Psychodynamic Ideal Types Of Elderly Suicidal Persons Based On Counter Transference”. Journal of Social Work Practice, 20, 347–365. https://doi.org/10.1080/02650530600932011.

19.

Mandara

Jelani.

(2003). The Typological Approach In Child And Family Psychology: A Review Of Theory, Methods, And Researchclinical Child And Family Psychology Review.

20.

Maxwell

J. A.

(2010). A Realist Approach For Qualitative Research. SAGE.

21.

McCrae

R. R.

Costa

P. T

(1987). “Validation Of The Five-Factor Model Of Personality Across Instruments And Observers”. Journal of Personality and Social Psychology, 52(1), 81–90. https://doi.org/10.1037/0022-3514.52.1.81.

22.

McLeod

(2011). “Variants of Grounded Theory” Qualitative Research in Counselling and Psychotherapy (pp. 161–166): Sage Publications.

23.

O’Keeffe

Martin

Midgley

(2020). Rupture-Repair Processes in the Therapeutic Alliance Prior to Dropout in Adolescents Receiving Therapy for Depression. Psychotherapy, 57, 471–490.

24.

O'Keeffe

Martin

Target

Midgley

(2019). “‘I Just Stopped Going’: A Mixed Methods Investigation Into Types of Therapy Dropout in Adolescents With Depression”. Frontiers in Psychology, 10(FEB), 1–14.

25.

Pew Research Center (2018). The Religious Typology. Retrieved from https://www.pewforum.org/2018/08/29/the-religious-typology/.

26.

Philips

Werbart

Wennberg

Schubert

(2007). “Young Adults’ Ideas Of Cure Prior To Psychoanalytic Psychotherapy”. Journal of Clinical Psychology, 63(3), 213–232. https://doi.org/10.1002/jclp.20342.

27.

Pinquart

(2016). Associations of Parenting Styles and Dimensions with Academic Achievement in Children and Adolescents: A Meta-analysis. Educ Psychol Rev, 28, 475-493. https://doi.org/10.1007/s10648-015-9338-y.

28.

Pistrang

Barker

(2012). “Varieties Of Qualitative Research: A Pragmatic Approach To Selecting Methods”. In Cooper

Camic

P. M.

Long

D. L.

Panter

A. T.

Rindskopf

Sher

K. J.

(Eds), Research designs: Quantitative, qualitative, neuropsychological, and biological (Vol. 2, pp. 5–18). American Psychological Association. https://doi.org/10.1037/13620-001.

29.

Psathas

George

(2005). The ideal type in weber and schutz.” Explorations of the life-world: 144–69. http://link.springer.com/chapter/10.1007/1-4020-3220-X_7.

30.

Safran

J.D.

Muran

J.C

(1996). “The Resolution of Ruptures in the Therapeutic Alliance”. Journal of Consulting and Clinical Psychology, 64(3), 447–458. http://ezproxy.usherbrooke.ca/login?url=https://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=1996-05271-003&site=ehost-live.

31.

Sharp

N. L.

Bye

R. A.

Cusick

(2018). Narrative analysis. In Liamputtong

(Ed.), Handbook of research methods in health social sciences (pp. 1-21). Springer. https://doi.org/10.1007/978-981-10-2779-6_106-1.

32.

Sorkhabi

Mandara

(2013). “Are The Effects Of Baumrind's Parenting Styles Culturally Specific Or Culturally Equivalent?” In Larzelere

R. E.

Morris

A. S.

Harrist

A. W.

(Eds), Authoritative Parenting: Synthesizing Nurturance and Discipline for Optimal Child Development (pp. 113–135). American Psychological Association. https://doi.org/10.1037/13948-006.

33.

Stapley

O’Keeffe

Midgley

(2021). Essentials of ideal-type Analysis: A qualitative approach to constructing typologies. American Psychological Association.

34.

Stapley

Target

Midgley

(2017). “The Journey Through and Beyond Mental Health Services in the United Kingdom: A Typology of Parents’ Ways of Managing the Crisis of Their Teenage Child’s Depression”. Journal of Clinical Psychology, 73(10), 1429–1441. https://doi.org/10.1002/jclp.22446.

35.

Stuhr

Ulrich.

Wachholz

Sylvia

(2001). “In Search for a Psychoanalytic Research Strategy: The Concept of Ideal Types”In Qualitative Psychotherapy Research - Methods and Methodology.

36.

Swedberg

Richard.

(2018). “How to Use Max Weber’s Ideal Type in Sociological Analysis”. Journal of Classical Sociology, 18(3), 181–196. https://doi.org/10.1177/1468795x17743643.

37.

Wachholz

Stuhr

(1999). “The Concept Of Ideal Types In Psychoanalytic Follow-Up Research”. Psychotherapy Research, 9(3), 327–341.

38.

Warnick

E.M.

Gonzalez

Robin Weersing

Scahill

Woolston

(2012). “Defining Dropout from Youth Psychotherapy: How Definitions Shape the Prevalence and Predictors of Attrition”. Child and Adolescent Mental Health, 17(2), 76–85. https://doi.org/10.1111/j.1475-3588.2011.00606.x.

39.

Weber

Max.

(1904). On Methodology Of Social Sciences. Routledge.

40.

Werbart

Andrzej.

Brusell

Iggedal

Lavfors

Widholm

(2016). “Changes in Self-Representations Following Psychoanalytic Psychotherapy for Young Adults: A Comparative Typology”. Journal of the American Psychoanalytic Association, 64(5), 917–958. https://doi.org/10.1177/0003065116676765.

41.

Werbart

Andrzej.

Grunbaum

Jonasson

Kempe

Kusz

Linde

O’Nils

K. L.

Sjovall

Svenson

Theve

Ulin

Ohlin

(2011). “Changes In The Representations Of Mother And Father Among Young Adults In Psychoanalytic Psychotherapy”. Psychoanalytic Psychology, 28(1), 95–116. https://doi.org/10.1037/a0022344.

Developing Typologies in Qualitative Research: The Use of Ideal-type Analysis

Abstract

Keywords

Introduction

What is Ideal-Type Analysis?

Ideal-Type Analysis And Epistemology

Comparisons to Existing Qualitative Data Analysis Techniques

What Are The Steps To Carrying Out An Ideal-Type Analysis?

Becoming Familiarised With The Dataset

Writing The Case Reconstructions

Constructing The Ideal Types

Identifying The Optimal Cases

Forming The Ideal-Type Descriptions

Checking Credibility

Making comparisons

Examples Of Studies Using Ideal-Type Analysis

Conclusions

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iD

References