DATA ANALYSIS

This chapter provides an overview of the data analysis methods in the four sub-studies. For the purposes of the single-country study, a narrative synthesis was conducted. In sub-study 2, to map the field of cross-national comparative research, a narrative synthesis was conducted, and a coding process carried out. In sub-study 3, to compare the response of the German and Finnish school social workers to an exemplary case of child maltreatment, computer-aided content-structuring content analysis and computer-computer-aided analysis with MAXQDA were conducted. In sub-study 4, a coding process was carried out.

Narrative synthesis

In sub-studies 1 and 2, a narrative synthesis was conducted; this method was seen as appropriate as these two sub-studies did not aim at summarising the effectiveness of interventions but to provide an overview of terms and concepts in German SSW, as well as relevant aspects of its CYW and education system (sub-study 1) and SSW practice themes that transcend national boundaries (sub-study 2). Moreover, the included publications and studies

were highly heterogeneous, which is why a method was needed that allowed combining them “in the way that seems most feasible or appropriate”; thus, it was decided to conduct a narrative synthesis and, hence, to use a textual approach instead of a statistical summary, although a narrative synthesis can include statistical data as well; the term “narrative” is understood as denoting the opposite of a “statistical” summary, although there is, to this day, “no consensus on the constituent elements […] and the conditions for establishing a systematic and transparent approach to the synthesis process”

(Rodgers & Sowden et al., 2009, p. 50). Thus, this method can be applied to investigate several issues, “not only those relating to the effectiveness of a particular intervention” (Popay et al., 2006, p. 5).

In sub-study 1 (Beck, 2017), the findings were, first of all, preliminarily synthesised by generating tables. The publications were analysed regarding the terms used in publication headings that describe the specific form of cooperation between CYW and school other than SSW [Schulsozialarbeit]. In addition, different scholars, legal bases, concepts, and target groups were organised in a table. Second, relationships between the included publications were examined (Rodgers & Sowden et al., 2009, p. 58) and interpretations carried out regarding the nature of SSW in Germany, based on the preliminary synthesis. This method conforms to the systems theoretical perspectives applied; thus, it supports “thinking in contexts” in order to view SSW from a

“networked and holistic perspective” (Barth, 2007, p. 235, this researcher’s translation).

In sub-study 2 (Beck & Hämäläinen, 2020), a preliminary synthesis was conducted by developing tables as well as organising specific aspects in groupings; a table was generated that includes the year and type of publication, knowledge source and number of countries under comparison. Another table provides insight into the criteria upon which a comparison took place, the comparative countries and the findings. In order to show the differences of

“countries and continents under comparison between all publications with and without the surveys conducted by the INSSW, a Two-Cases Model was generated with the software MAXQDA”; to present key areas of previous publications, a classification system (Meeuwisse & Swärd, 2007) was applied that distinguishes “comparisons based on social policy, profession, and

practice”; finally, to visualise these areas, “vote counting, in the form of ticks”

was done (Beck & Hämäläinen, 2020, p. 7). Second, after the findings were identified, listed, tabulated and counted, the relations between and within the publications were examined (Rodgers & Sowden et al., 2009, p. 58) with a specific focus on SSW practice themes (see the section about the coding process below).

Coding process

In the sub-studies 2 and 4, the data were analysed by using a coding process, starting with selecting relevant text as well as identifying repeating ideas and themes which comprise groups of repeating ideas (Auerbach & Silverstein, 2003). This was deemed an adequate method to investigate SSW practice themes that transcend national boundaries to meet the need to tackle the diversity of different ideas concerning the main foci of SSW practice (sub-study 2). A narrative synthesis was conducted before the relationships between and within the publications were explored (Rodgers & Sowden et al., 2009, p. 58).

Also, it was deemed an adequate method to investigate the work-related stressors that accompany school social workers while assessing children’s well-being (sub-study 4) as the semi-structured interviews were already transcribed and several statements therein were repeatedly expressed by the study participants.

In sub-study 2, at first, the included literature was investigated and repeating ideas identified (SSW practice themes) before they were assigned to 16 superordinate themes: behavioural problems and delinquency; physical and mental health issues; social-emotional challenges; meeting basic needs;

divorce and single parenting; domestic violence and child maltreatment;

parental physical and mental health issues; poverty; attendance, drop out and motivation; enrolment and completion; quality of education; violence in schools; crime, violence and war; decaying neighbourhoods; a lack of community services; and racism. Finally, based upon these themes, four categories were built, “namely, child-, family-, school-, and community-related issues” (Beck & Hämäläinen, 2020, p. 7). Table 5 provides an example of the data analysis in sub-study 2 (see Table 5).

Table 5 Example of the data analysis for the category school-related issues in sub-study 2.

Level Content

Category 3 School-related issues Theme 1 Violence in school Repeating idea 1 Bullying

Ideas Playground bullying (Huxtable & Blyth, 2002); bullying (Jarolmen, 2014; Villarreal-Sosa et al., 2017); bullying through internet and other technologies (Jarolmen, 2014) Repeating idea 2 Abuse by staff

Ideas Corporal punishment (Huxtable & Blyth, 2002; Jarolmen, 2014)

Accordingly, single ideas such as playground bullying (Huxtable & Blyth, 2002), bullying (Jarolmen, 2014; Villarreal-Sosa et al., 2017) and bullying through internet and other technologies (Jarolmen, 2014) were grouped under the repeated idea category 1 bullying. Also, single ideas such as corporal punishment (Huxtable & Blyth, 2002; Jarolmen, 2014) were grouped under the repeated idea category 2 abuse by staff at school. Then, the repeated idea categories 1 and 2 were grouped under the theme 1 violence in schools, which was, in turn, grouped under the category 3 school-related issues. By using this analysis method, it was possible to focus on the primary addressees of SSW, namely, children.

In sub-study 4, the interview transcripts were analysed and eight repeating ideas (work-related stressors that were repeatedly named by participants) identified. These were emotional distress, great responsibility, great workload, insufficient time and personnel resources, lack of accessibility of youth welfare departments, lack of timely reaction by youth welfare departments, no decision-making power and a too far-reaching scope of the parental right. In a second step, three superordinate themes (work-related stressors that were repeatedly named by the participants in connection with each other) resulted in the visual tool Code Map (CM) (MAXQDA; see section

about computer-aided analysis for a detailed description). The findings are to be seen in close relation with the investigated roles of German and Finnish school social workers against the background of welfare regimes and child welfare systems (study 3). Table 6 shows the category system of sub-study 4 (see Table 6).

Table 6 Category system of sub-study 4.

Work-related stressors Connected stressor 1 Emotional distress Great responsibility No decision-making power Connected stressor 2 Lack of accessibility to youth

welfare department Lack of timely reaction by youth welfare department Too far-reaching scope of parental right

Connected stressor 3 Great workload Insufficient time and personnel resources

Hence, three superordinate themes appeared: connected stressor 1 comprises a perceived great responsibility associated with emotional distress but no decision-making power; connected stressor 2 includes the lack of accessibility to, and timely reaction by youth welfare departments and a too far-reaching scope of the parental right; connected stressor 3 comprises great workload connected with insufficient time and personnel resources. Table 7 provides an example of a category (see Table 7).

Table 7 An exemplary category in sub-study 4.

Category Examples in the transcripts

Great responsibliity • “I see that I have a great responsibility here, especially in these situations, because I am very often the first contact or, in this case, the second contact after the classmate. Nevertheless, I am the person who is most likely to hear about it first. And I think that I have the responsibility to work very sensitively thereon. I am also liable if I do not direct the information further [to the youth welfare department] or if I do not care about [him]” (G1).

• “There is a very high responsibility, especially in a case like this involving pornography. Because such cases aren’t solved overnight, so at first you really feel the burden until you get to the point where you know: Ok, such and such has been done and a good solution for everyone has been found. But until it is found, you are often left thinking: How does he feel right now, or is there any sign of his even worse endangerment? Then, there’s always the thought: What if nothing gets better (comment: for him)? Or what else can I do? Did I over-look something?” (G3).

Accordingly, the category great responsibility included all answers that referred to the great responsibility perceived by the German professionals.

Different colours were assigned to the categories for reasons of clarity.

Table 8 illustrates the colour scheme used in MAXQDA for the purpose of sub-study 4 and provides information about the percentages of all tiles in each document (interview) in the respective colours. The grey-coloured cells show the topics addressed most by the interviewees (see Table 8).

Table 8 The applied colour scheme and percentages of all tiles in the doc-ument in the respective colours in sub-study 4.

Category Color G1 G2 G3 G4 F1/F2

Emotional distress 5,25 4,92 6,58

Great responsibility 26,33 16,67 11,67 25,83

No decision-making power

6,75

Great workload 26,42

Insufficient time and

personnel resources 6,58

Lack of accessibili-ty of youth welfare department

9,58

Lack of timely reac-tion by youth welfare department

28,67 7,33 11,58 6,33

Too far-reaching scope

of the parental right 27,33 2,08 6,00 0,92

Thus, participants G2, G3 and G4 talked mainly about their great responsibility while assessing children’s well-being. G1 talked primarily about the lack of timely reaction by the youth welfare department. In contrast, F1 and F2 talked mainly about their great workload. This topic was followed by the topic of insufficient time and personnel resources.

To conclude, this analysis method enabled the juxtaposition of the findings, which would allow a follow-up comparison later and was, therefore, in line with the overall methodological considerations regarding cross-national

comparative research in social work (Friesenhahn & Kniephoff-Knebel, 2011).

Furthermore, it aligns with systems theoretical perspectives that require

“thinking in context” (Barth, 2007, p. 235; this researcher’s translation) as well as the ecological systems theoretical perspectives which focus on the relationship between a person and their environment (Bronfenbrenner, 1995, p. 623) and on the necessity to focus on how settings are subjectively perceived (Bronfenbrenner, 1979, p. 22); for example, the German school social workers perceived the parental right to be too far-reaching. Thus, it was possible to concentrate the focus on the work-related stressors of the participants within the context of the German and Finnish welfare regimes and child welfare systems and on how the assigned roles are subjectively perceived. It was also possible to obtain insights into the conflicts between these assigned roles and the subjective perceptions of these by the school social workers. Thus, this method supported answering the main research question, namely, to investigate and compare the nature and role of SSW in the context of welfare regimes and child welfare systems.

Computer-aided content-structuring content analysis

In sub-study 3, the semi-structured interviews were investigated using the seven-step content-structuring content analysis developed by Kuckartz (2016) and the software MAXQDA. This decision was made due to the fact that MAXQDA was developed by Kuckartz; therefore, the greatest possible compatibility between the analysis method and software could be assumed.

First, interviews were transcribed, text passages that appeared particularly important for the studies marked and memos were written. Second, categories were built up to generate a content-related structuring of the transcribed data (Kuckartz, 2016, p. 101). The categories were developed independently from the material but were based on an already existing systematisation, namely, an interview guide, which is a common procedure in a theory-oriented (deductive) building-up of categories, as opposed to an empirically oriented (inductive) one (Flick, 2016; Kuckartz, 2016, p. 64).

Accordingly, the initial main categories derived from the research questions provided direction for the data collection (Kuckartz, 2016, p. 101).

For example, the participants were asked to name the main areas that they would focus on in their work, which is why the category “most important focus areas” was logically one of the primary topics of data analysis. In addition, to answer the main research question about the nature and role of SSW within the context of the German and Finnish welfare regimes and child welfare systems, specific categories were seen as necessary for inclusion, which is why a theory-oriented building-up of initial categories was carried out.

Third, a first coding process took place in which all text passages were added to their respective categories. Fourth, all text passages that were added to one specific category were compiled (Kuckartz, 2016, pp. 102–

106). Fifth, subcategories, derived directly from the material, were built for each category. Sixth, a second coding process was carried out that took the differentiated categories into account (Kuckartz, 2016, pp. 106–111). Seventh, a category-based analysis was conducted, which yielded findings with regard to the main categories and clarified what was said on certain topics and which topics were only rarely mentioned; in addition, relationships between different subcategories were analysed and visualisations with MAXQDA created (Kuckartz, 2016, pp. 118–119). Table 9 shows the category system of sub-study 3 (see Table 9).

Table 9 Category system of sub-study 3.

Category Subcategories Sub-subcategories

First reactions Sexual abuse Daniel

Focus areas Most important focus

areas Sexual abuse

Nutrition All focus areas Sexual abuse

Nutrition Hygiene

Violent environments Daniel

Other needs

Hence, the category first reactions comprises the two subcategories sexual abuse and Daniel; the category focus areas comprises the subcategory most important focus areas, which, in turn, includes the sub-subcategories sexual abuse and nutrition; the subcategory all focus areas comprises the sub-subcategories of sexual abuse, nutrition, hygiene, violent environment, Daniel and other needs. The findings of this study support the understanding of the findings of sub-study 4, which focuses on work-related stressors that accompany school social workers while assessing children’s well-being.

Accordingly, work-related stressors are closely related to the role that the German and Finnish school social workers must fulfil within the context of welfare regimes and child welfare systems. Table 10 provides an example of a deductive and an inductive category (see Table 10).

Table 10 An exemplary deductive and inductive category in sub-study 3.

Category Examples in the transcripts First reactions

(deductive) • “Well, first of all, I would try to ask Daniel what he thinks about it. For example, if he would come, together with Florian, to me. Or if I should ask Florian directly, which I would see as a second step. So, I would really try to let it be done through Daniel, because I noticed that he is a person of trust for Florian. And if Daniel says: Well, look, you can really have good conversations with her [the school social worker].

So that he [Florian] then comes along with him [Daniel]” (G2).

• “Well, I would listen to exactly what happened. I would ask after the point in time and the previous weekend. Yeah.

And– Yes. If he [Daniel] could also motivate him [Florian] to come to see me” (G1).

Category Examples in the transcripts Daniel

(inductive) • “In addition, I think that one must make contact with Daniel as well. Daniel might be not able to cope with the situation, might not know what will happen now, and, and, and– feels co-responsible maybe” (G1).

• “Mhm– first of all, I can imagine that he [Daniel] is highly concerned. Also, a bit upset, which is why I would try to calm him down. To see, that he feels reasonably good, that he can trust me […] Also, I would try, if I also notice that he seems affected, also to involve him. So, I would say that he can feel free to talk about his feelings. So that it is not just about Florian but also, of course, about him as a person entrusted with a secret” (G2).

Accordingly, the category first reactions was seen as an important topic for answering the main research question and included all answers that referred to the first reactions provided by the German and Finnish professionals. In contrast, there were further categories built up inductively like the category Daniel, which comprised all answers that referred to him as a person of concern.

Different colours were assigned to the categories for reasons of clarity.

Table 11 illustrates the colour scheme used in MAXQDA for the purpose of sub-study 3 and provides information about the percentages of all tiles in each document (interview) in the respective colours. The grey-coloured cells show the topics addressed most by the interviewees; the framed cells show the topics that were highlighted as the most important topics to work on in the exemplary child maltreatment case (see Table 11).

Table 11 The applied colour scheme and percentages of all tiles in the document in the respective colours in sub-study 3.

Category Color G1 G2 G3 G4 F1/F2

Sexual abuse 29,33 12,58 16,58 11,25 37,75

Failure to provide

adequate nutrition 31,92 19,43 27,33 31,75 18,42

Failure to provide

adequate hygiene 17,17 8,08 5,92 3,33 9,33

Exposure to violent

environments 8,00 2,17 1,50

Statements concerning

Daniel 12,42 33,83 2,83 7,25

Other topics suggested

as relevant 4,00 11,00

Thus, participants G1, G3 and G4 talked mainly about the failure of Florian’s parents to provide him with adequate nutrition, despite the fact that all the German participants highlighted sexual abuse as the most important topic, except for one school social worker who highlighted sexual abuse and nutrition as equally important (G4). G2 had a different approach and focused predominately on Daniel, followed by the topic of the failure to provide a child with adequate nutrition. In contrast, participants F1 and F2 talked primarily about the topic of sexual abuse, which they highlighted as the most important one. This topic was followed by the topic of the failure to provide adequate nutrition.

To conclude, this analysis method enabled the juxtaposition of the findings, which would allow a follow-up comparison later and was, therefore, in line with the overall methodological considerations regarding cross-national comparative research in social work (Friesenhahn & Kniephoff-Knebel, 2011). It further aligns with the systems theoretical perspectives that require “thinking in context” (Barth, 2007, p. 235; this researcher’s translation) and ecological systems theoretical perspectives which focus on the relationship between a person and their environment (Bronfenbrenner, 1995, p. 623). It was possible

to concentrate the focus on SSW’s role in cases of child maltreatment, hence, this method supported answering the main research question, namely, to investigate and compare the nature and role of SSW within the context of welfare regimes and child welfare systems.

Computer-aided analysis with MAXQDA

In sub-study 3, computer-aided qualitative data analysis software (CAQDAS) was used predominately for two reasons. First, it was assumed that it would allow data analysis to be carried out in a clearer and more systematic manner; the possibility of building up a category system and including memos concerning ideas on specific text passages was deemed an important feature to aid reflection upon the data obtained. Second, from the beginning, it was intended to visualise specific findings to facilitate a comparison as well as to enhance the transparency and credibility of the qualitative research.

Interestingly, CAQDAS software has raised the interest of new groups of people in qualitative research: for example, CAQDAS software is often used by female junior researchers for qualification works, such as doctoral dissertations; many of these researchers have little or no experience with traditional, non-electronic methods and, for the most part, are not limited to one single analytical style or scientific discipline (Kuckartz, 2007, p. 27).

The computer-aided analysis of qualitative data has a long history (Flick,

The computer-aided analysis of qualitative data has a long history (Flick,

In document School social work in the context of the German and Finnish welfare regimes and child welfare systems (sivua 118-145)