The survey data was analyzed in three separated phases, which linked together with the growing knowledge and understanding about the issue of children’s participation in early childhood education setting.
In the first phase I analyzed the two first sets of quantitative variables for gaining knowledge about the frequency of children’s participatory experiences and educators’ pedagogical practices. Because of the similar Likert scales, these sets of data seemed best to analyze together. I used exploratory factor analysis with SPSS-software to build a model of children’s participation levels. The results of this analysis have been published in my master thesis study and are therefore not explained here throughout (Leinonen, 2010). To be mentioned, important finding was that the descriptive data collected from Finnish early childhood
29
education settings was theoretically valid with international participation research. The results could be explained based on Shier’s (2001) model of Pathways to participation and with that results it seems that Shiers model could be used also in early childhood education.
On the second phase of analysis the researchers focused in the open-ended variables of the whole data set and with abductive content analysis formed frameworks of children’s participation key-elements in ECE together with models for participatory pedagogy. In Finnish educational research tradition Tuomi & Sarajärvi (2009) state that the content analysis is suitable method for summarizing the contest from the data with systematical and objective viewpoint. Content analysis gives also room for human viewpoints and meaning making process i.e. focusing in to essential core of the data (Hsieh & Shannon, 2005). Inside of the content analysis tradition can be found different approaches of using the methodology.
For example Tuomi and Sarajärvi (2009) have found the quantifying approaches, where the themes and items are counted for summarizing or comparison. This approach is called summative approach by Hsieh & Shannon (2005) and it involves counting and comparisons through keywords, followed by the interpretation of the summative underlying context. The context analysis approach, the data is described with qualified expression rather than quantified counting (Tuomi & Sarajärvi, 2009). According to Hsieh & Shannon (2005) this context approach can be divided into two sub-approaches: The directed approach starts the analysis with a theory or relevant research findings as guidance for initial codes, while the conventional content analysis approach coding categories are derived directly from the text data. In this research the data was first concerned through summative approach (quantifying approach) because the amount of data was huge. At the end of the analysis was implemented hand-by-hand with both approaches, where the quantification was used to support the context analysis to open and explain the results. The analysis can be better referred through the abductive content analysis, which is a mixture of inductive and deductive analysis approaches (Kovách & Spence, 2005; Fann, 1970). It is rarely recognized in Finnish educational research tradition. For example Tuomi & Sarajärvi (2009) state that in abductive content analysis the common tread is hardly findable. In my research I have, however, leaned on international methodology, which recognized the abductive analysis. In an inductive approach, data about the phenomenon under investigation guide the analysis process, and in a deductive approach, the analysis is theoretically derived and it is a theory testing process (Hyde, 2000; Mayring, 2000), whereas in abductive analysis, the key element is the interaction between the theory and the data to acquire new knowledge about the
30
phenomenon (Fann, 1970). The researcher creates a framework of different theories together with the data in a creative process of “theory matching” (Kovách & Spens, 2005), Figure 2.
Figure 2 Three dimensions of research process (Spens & Kovách, 2005)
In this process, the prior theoretical knowledge was a guideline for the phenomenon studied through the real-life observation, this case the reported practices of participants. It is important to notice, that both the theoretical knowledge and the depth of understanding the survey data were both developing simultaneously, through cycles of analysis (see Fann, 1970). Kóvach & Spens (2005) states that the process of abduction starts when theoretical knowledge and the real-life observations does not match. In this process the interaction between theories and observation data are forming ongoing cycles of analysis, where intuition and the researcher’s cultural understanding have essential role (see Fann, 1970;
Kóvach & Spens, 2005). Finally, the findings in each specific research items are results of individual processes of going through the data and the theoretical background, but also connected to each other through evolving knowledge of enhancing children’s participation in early childhood education contexts.
31
On the final phase of the analysis focus of research was transferred back to the quantitative variables sets and those were analyzed together with qualitative analysis results to create a mixed methods approach. This kind of method mixing is described by Johnson and Christensen (2004) to include both quantitative and qualitative analysis phases systematically in the process of gaining knowledge about the research items. In this phase the interpretations of the data analysis was done with dialogic way with both findings from quantitative variables sets and qualitative open-ended questions together with theoretical background to create new knowledge of the process of enhancing children’s participation from educators perspectives. This kind of approach is described recently by Tolan and Deuch (2015) who state that mixed methods approach seems advantageous when exploring and understanding phenomena is developmental sciences. They mention that with this approach it is possible to seek knowledge combining both the universal level of the phenomenon (through quantitative analysis) and uniqueness of voice and conceptions of each participant (through qualitative analysis) (Tolan and Deuch, 2015). Through mixed methods approach it seemed possible to reach the qualitative experienced phenomenon together with the incidence and frequencies of statistically measured variables about enhancing children’s participation. Finally through the abductive approach both data sets were used and content analysis of the qualitative data with the different statistical methods, like factor analysis, comparing of means and basic descriptive statistics from quantitative data were founded useful to build a big picture about the phenomenon of participatory pedagogy.
The results of this mixed-model analysis phase will be published in five research papers.
First one focuses to compare the Curriculum of Pre-school education in Finland and the participants’ practices in pre-primary school groups. The second one is focusing on children’s chances to design their learning together with teachers, i.e. plan, implement and evaluate the pedagogical activities in participatory practices. In third paper the focus is on everyday constraints that prevent children’s participation and teachers’ opportunities to influence on those. Fourth paper that is so far not yet accepted for publishing aims to find connections between children developing skills in self-regulation and teachers’ support for that development with participatory pedagogy. Final one aims to explain a tentative framework of participatory pedagogy and teachers duties and responsibilities in it.
32