Children`s activities in Norwegian kindergartens. Part 2: Focus on variations

Introduction

The Norwegian early year's educational field is known at one with high quality which can be associated with equal opportunities for every child. In the Norwegian Kindergarten Act §1 and in Framework Plan for Kindergarten in Norway (7), it is stated that: Kindergarten will promote democracy and equality and combat all forms of discrimination. In these key articulations, formal political norms are understood as rules of behaviour and expectations that are devised politically. The motivation for studying what is going on in the Norwegian kindergartens comes from a curiosity about state of affair, to get closer to the empirical facts of variations between vulnerable groups of children. How are everyday life activities for girls and boys, for children with special needs and for children with origins outside Europe? According to the Nordic Database such information and knowledge are lacking. Based on an overall picture of children's activities in kindergarten (se part 1 in this special issue), this study further explored whether particular groups of children were involved in different types of activities. The investigated group variables were age, gender, ethnicity and physical as well as psychological disabilities. In other words, the reported analyses show whether younger children were involved in different types of activities compared with older children, and whether the reported activities were different for boys, girls or in gender-mixed groups. We explored possible differences in activities for children with a deviant ethnicity or some type of disability. Possible consequences of such variations in activities should be critically discussed. Theoretically, we raise the question whether the reported patterns of activities reflect different processes of cultural formation for these groups of children.

Theoretical framework

Socio-cultural perspectives are seen as central for framing theorizing and thinking, and for developing didactics for current kindergarten practice (3, 5 and 8). In addition to being focused on the child, a socio-cultural approach also encompasses social and relational factors, artifacts such as toys and materials, institutional factors such as traditions, certain ways of doing things, and patterns of culture. One of the central concepts of Bourdieu's cultural sociology is "habitus", a set of dispositions that generate culturally-based practice (1and 2). Early childhood education practice can, in such a cultural perspective, be understood as a social practice formed by actions, organizational structures, mediation of artifacts and embodiment. Personal and local practice will therefore be socio-cultural and habitual practices. Through participating in early year's institutions, children's cultural formation, learning and development takes place. Thus such practices can be studied through observing everyday activities (8 and 5).

The term "activity" is crucial in the design of the study; both in the instructions to the observers as well as in the analysis. Socio-cultural theoretical conceptualization is well suited when researching children's and teachers' activities in an institutional setting (4, 8 and 9). According to Barbara Rogoff development can be seen as transformation of participation in cultural activities (8). A consequence of seeing the term activity in a sociocultu­ral frame is that even what seems to be individual form of activity it is considered social because the institutionalized setting conditions the way a child can be related to time, space and human relations. Individual activities are not separate from the world and it is not only a question of being influenced. "Activity" in this study equals "cultural activity" and embraces spontaneous activities in small groups and alone, as well as planned teacher activities with in socio-cultural framework.

The analysis does however not follow any specific kind of activity system analysis, but are rather considered close to the empirical base of this study. Rogoff refers to "the processes and systems of involvement between people as they communicate and coordinate efforts while participating in culturally valued activity" and names this kind of process guided participation (10). The "guidance" referred to in guided participation involves the direction offered by cultural and social values, as well as social partners. Following Rogoffs suggestion of interpersonal focus of analysis, we are interested in teachers and children's everyday activities; being together in kindergarten. The analytic lens is first chosen by the teachers. They selected dominant and less dominant activities. The second lens is chosen by the researcher by focusing on variance. A key to Rogoff's analytic approach is also the emphasis on process in human activity. Following her analytic approach in the design, kindergarten is seen as a dynamic cultural community. Teachers and children are participating in planned and spontaneous activities during a day and by studying observations with the lens of variance we are able to get more information about conditions for different groups of children.

Data and participants

Many kindergartens in Norway have a formal teacher training collaboration with their local teacher training university. All local kindergartens were invited to participate in our project. We required that the observers had to be educated as kindergarten teachers, and they were asked to report observations in their group of children (the organizational unit to which they belonged) 3—4 times a day, every day of one particular week (or the following week). We specified the ideal number of observers dependent on the total size of the kindergarten (roughly 1 observer for 20 children), but we accepted participation also with a smaller number of observers. Thirty-one observers at 18 kindergartens participated in our study. It is clear that only a small percentage of the invited kindergartens participated (about 25 percent) and some of them with a smaller number of observers than we asked for. We will argue, however, that the present study should not be regarded as a traditional survey. It should be noted that the participants were not asked to simply fill in a questionnaire, they were asked to contribute observational reports requiring hours of work during a whole week. We will therefore argue that it is not reasonable to compare the percentage participation in our study directly with traditional survey studies. On the other hand, it is a critical issue, whether the sampled observations show a representative picture of the content in Norwegian kindergartens, and this issue is further commented on in our discussion part below.

All observers received information about the project and detailed instructions for reporting observations. The day was divided into four parts and we asked the observers to report 1—4 observations for each of the four time periods. Each activity was reported on a separate standardized form with both fixed alternatives and open space for their own descriptions. It was emphasized that we were interested in their qualitative description of the reported activity including characteristics of participants, what was said and done, responses, use of equipment, physical location and so on.

A dominant trait by Norwegian kindergartens is that a lot of different activities take place at the same time. In the project-planning it was therefore a major challenge to elaborate guidelines for selecting a specific activity. In the final guidelines it was stated that an activity could involve one, several or many children. The activity could be planned, spontaneous or just daily routines (such as eating and cleaning up, etc.). The observers were instructed to first select the most obvious and distinct activity at a certain time point (reported as First Activity in Table 1). An obvious and distinct activity was defined as an activity easy to recognize due, for example, to the number of children involved, type of equipment, loud voices etc. It should be noted that an obvious and distinct activity does not necessarily mean participation of many children, but it is clear to the observer "what's going on". Such activities are referred to as clear in the following, and examples of clear activities could be children painting pictures and a child "reading" a book. A simple definition of unclear activities is everything that is not clear. Many activities in kindergarten change rapidly and may also be a playful mixture of different activities. In other words, it may be difficult for the observer to report "what's actually going on". In our guidelines we explicitly asked the observers to look for such activities and report them as Second Activity in a particular time period (Table 1). If the observers didn't recognize any clear/unclear activity at a certain time point, they were asked to report the type of activities that were actually taking place and mark the observation as clear or unclear. If the observers reported more than two activities during a spell of work, we didn't ask specifically for clear or unclear activities. The guidelines for selecting an activity were tested and discussed with observers in three kindergartens before data collection (pilot), and the observers had an open support service (phone number) for some hours every day of the data collection week. This service was only used twice, which indicates that the information given was clear. The 31 observers contributed 798 observations in our study.

The overall picture of reported activities is presented in in Part 1 of our study (11), but a brief description of results is available in Table 1 below. The analyses in the present report are based on data where the observers reported standardized information about the children involved in an activity. We asked about age, gender (for groups of six children or more this information was given as "most girls/boys or mixed" and age from/to or on average). We asked whether some of the children had inadequate language for their age or physical or psychological disabilities. It was also reported whether some of the children had a cultural belonging that could to some extent be regarded as deviant from Western European culture. We applied the formulation "to some extent" to underline that all degrees of cultural affiliation outside Norway are relevant and should be marked on the observational form. It should therefore be noted that the children in this category include both children with a completely different culture and children with only weak ties to another country outside Western Europe.

The observers also filled in a questionnaire the last day of their reporting job, which is referred to as Dataset 2 in the present report. Based on these data, we know their attitude to the performed observational work, their experience with selecting activities and reporting them in our standardized form. We also asked them in general about the particular week (that was reported), the characteristics of the total group of children (where only some of them were included in the reported activities) and finally we asked the observers to assess the pedagogical quality of the kindergarten and their own satisfaction with their workplace. All this information is useful for placing the reported activities in a broader context described by the observers.

Method and analysis

Overview of observational reports

All observations and questionnaires were returned to the project by post in an anonymous envelope. The data analyses were performed with the statistical program SPSS. Each observation was coded as a case (labeled with a random number for the actual kindergarten and observer).

In the result section we first present simple descriptive statistics to provide an overview of the observational reports. These results show how the kindergarten teachers fulfilled their observational task (how many reports on average at each time point, Table 1), and furthermore we highlight some characteristics in the overall picture of reported activities.

Prevalence of different types of activities and the number of children/adults involved An important aim of our study is to present an overall picture of the kind of activities that were reported often or seldom. The observers were asked to both label the activity and describe the content of the activity, and, based on this information, we developed and adapted categories of activities in a continuous "dialogue" with data. This procedure required of course the researchers' subjective assessment (first author). Regarded as traditional quantitative research, this procedure may be seen as critical, but regarded as qualitative research this interpretation of data is the normal research activity. Main categories and sub-categories were constructed. For example, "Painting" was set up as a sub-category under "Art performance" etc. This result is reported Table 2, which shows main categories of activities and percent of the total observations distributed across different activities. In further analyses of activities we apply only the main categories of activities presented in Table 2. Each main activity was recoded as to dummy variable (occur or not) across all cases (observations).

Analyses of variance between kindergartens and observers

Based on an overall picture of children's activities in kindergarten, it is important to investigate whether there are significant differences between kindergartens with regard to the analyzed activities. Is there, for example, any evidence that children's participation in reading activities is dependent on the kindergarten they belong to? Reports on reading activities may also vary significantly between different organizational units within each kindergarten, a finding that would be indicated as a significant variance between observers when examining their reports on, for example, reading activities.

Estimating intraclass correlations is a simple approach for investigating variance between groups, in our case between kindergartens and between observers (reporting from different organizational units within kindergartens). We applied an ANOVA (analysis of variance) to provide the statistical measures needed for calculating intraclass correlations (rho) with the following formula: rho = т / (о2 + т), where т and о2 are the between-school and within-school variances, respectively (for details about procedure see 7).

It should be underlined that significantly high intra­class correlations would require estimates of multilevel models applying statistical multilevel programs (as mlwin, or hlmwin). Such analyses could, for example, reveal whether differences between groups (as kindergartens) were a result of systematic differences in, for example, children's ages. As an example, such a finding would show that differences between kindergartens (with regard to a particular activity) are a result of observational reports of children of different ages rather than a organizational characteristic where some kindergartens have high or low prevalence of a particular activity.

Multilevel analyses are in fact the correct procedure when data has a hierarchical structure as in our study (each kindergarten had one or several observers who in turn reported many observations). However, estimating such models presupposes reliable variance between groups (at different levels). It can therefore be argued that calculating intraclass correlations is an adequate procedure for examining whether the data should be further explored in multilevel analyses. An intraclass correlation was calculated for each main activity, and this procedure was performed at both levels (kindergarten and observers).

Analyses of variance between different groups of children

In the present study we have focused on certain individual characteristics for the children involved in the reported activities. These characteristics were gender, age, ethnicity, and physical and psychological disabilities. We performed an independent t-test to examine whether participation in activities was related to the group variables mentioned above. For some of the variables we also performed a correlation analysis (Person corr), and for gender we present a table that shows the percent of observations that involve both girls and boys (Table 3). A main issue in our study is to explore possible differences in cultural formation for different groups of children. It should be noted, however, that the teachers were not asked to focus on particular groups of children in their observational work. It is therefore relevant to ask whether these children's activities were reported to an extent that corresponds roughly to their proportion of the total group of children observed. To highlight this issue, we apply information about the whole group of children based on Dataset 2, where the teachers reported the number of children with cultural ties outside Western Europe, inadequate language, physical or psychological disability. Based on this information we make a very rough estimate of the total size of the analyzed groups.

Results

Overview of observational reports

The kindergarten teachers were asked to report 1 — 4 observations four times a day for one week. The number of reports each day was relatively equal, with the highest number on Tuesday day (174 observations) and lowest on Friday (141 observations). The day was divided into four parts or spells of work. The teachers report some more observations in the middle of the day, which is natural because not all of them are at work when the kindergarten opens in the morning and closes in the afternoon. Most observations were reported in the second spell of work (302) and lowest at the end of the day (123 observations). The observers were asked to report 1—4 observations for each spell of work (as mentioned above). Normally they reported 1—2 activities as shown below (also reported tin Part 1).

Table 1

Number of observations reported as first, second, third or fourth activity at a particular time point

The table above does not tell us how many of the observers who reported four activities within a spell of work (all 31 observers could have done it once or just a few of them several times). Closer inspection of data shows that 11 teachers (of 31) reported four activities one or several times, and 19 teachers reported three activities one or several times. This information will be taken into account when we describe the overall picture of children's activities.

The teachers marked each observation as a clear/unclear activity in line with the definition outlined in our introduction. They were also asked to report first a clear activity and secondly a less clear activity if both types of activities were taking place during a particular spell of work. Our results show that 73 percent of the observations were marked as a clear activity. The activities were normally regarded as very common or quite common by the teachers (over 80 percent of the observations). About two-thirds of the activities took place indoor (68 percent) and only 7 percent took place outside the kindergarten (trips and excursions).

Prevalence of different types of activities.

Table 2      below (also reported in Part 1) shows the percent of observations distributed across the main categories of activities. Indoor and outdoor activities are separated in the table, but it should be noted that role­playing is included in both arenas. We suggest that the measures in Table 2 can give an overall picture of what characterize children's activities in kindergarten. From a methodological viewpoint it may be argued, however, that some of the teachers influence this overall picture more than others because they contributed a higher number of observational reports. We therefore decided to also examine the frequencies when restricting the number of observations to a maximum of two within each spell of work. These results are listed under the heading "selected observations" in the table below. Based on the results we conclude that the most eager observers do not influence the total data material and we therefore include all observations in further analyses.

Table 2

Percent of observations distributed across activities

The results in Table 2 are commented in more details in Part 1 of our study (11).

Analyses of variance between kindergartens and observers

Intraclass correlations for kindergartens and observers (as group variables) were examined and the results showed only very low intraclass correlations (the highest of .05). There is considerable agreement in the literature that attention to group differences should be based on intraclass correlations of at least .10 (see e.g. 6), so we conclude that in the present study there is no empirical support for further multilevel analyses. Our findings suggest that children\s activities in kindergarten were not significantly different across kindergartens and units within them.

Analyses of variance between different groups

of children

Age

We examined possible age differences in children's activities as reported in the present study. All main activities (recoded to dummy variables) were correlated with age and, based on the result, a few variables were selected for an independent t-test. The analyzed groups were children under three years of age compared with children older than three. We found that the average level of reading activities and music activities were significantly lower for the oldest children. (T-value for reading activities was 2.32, and p = .020; for music activities t-value was 3.38 and p = .001). The results also show that the youngest children were significantly less involved in activities in the sandpit (t-value -2.59 and p = .010).

Gender

Gender differences in the reported activities were explored. In general, we found that both girls and boys participated in almost half of the observations (45 percent) and for the rest of the observations girls (or mostly girls) were involved in 24 percent of the activities and boys (mostly boys) in 31 percent. Table 5 below shows the percent of observations including both girls and boys within each main activity.

Table 3     

Percent of observations including both girls
and boys within each activity

Mixed gender activities are in particular music activities, trips/excursions, art performance and reading activities. It may be noted that of these activities, art performance and reading have the highest prevalence. We also want to point out that construction activities were reported relatively frequently but boys and girls cooperate only in 28 percent of the observations. Closer inspection shows that the remaining observations involved mainly boys.

Focusing on indoor activities t-tests showed a significant difference (p = .000) between girls and boys with regard to art performance (30 observations of girls and 12 of boys). It may be repeated, however, that the most common was mixed gender within art performance (61 observations). Turning to outdoor activities, we see that boys tend to play more with balls (t-test, p = .019), but it should be noted that in this category we have only a few cases (only 11).

Ethnicity and inadequate language

In our results, 15.4 percent of the observations include children with a varying degree of cultural belonging (ties) to a culture outside Western Europe. In these observations we normally find one to two children with a deviant cultural background (68 and 27 percent). These children typically have some ties to Asia (68 percent), Latin or South America (11 percent). These children were normally not adopted (only in 10 percent of the cases). Children with cultural ties outside Western Europe seem to be very clearly represented in the observations based on our estimate of their proportion of the total group of children (35 children in the total group of children). It may be added, however, that this group of children were not represented in 11 of the 31 groups included in our study. This finding shows that many children may have very little kindergarten experience with children representing cultures outside Western Europe.

Our main finding is that the activities for this group of children (cultural ties outside Western Europe) were not significantly different from other children. T-tests (independent on dichotomized activity variables) showed only two exceptions from this picture and in both cases these children were more involved than the average child. Children in the studied group participated significantly more in music activities than children in general (t-value -2.27 and p = .023). Music activities were reported in 9 percent of the cases when we select observations where minority-cultural children participated compared to 4 percent for children in general.

We also found that children in this group were more included in trips/excursions than the average child (t- value -2.65 and p = .008). Inspection of data shows that for the studied group of children, 10 percent of the activities were trips/excursions compared to 4 percent for the children in general. It may be added that older children participate more in this activity than younger children, and the average age in the studied group of children tend to be older compared to all children, but probably this fact explains the finding only partly (age influence music- activities the opposite direction).

Children with inadequate language were reported in 14 percent of the observations, and in these cases normally one child. We expect this group of children to have a high overlap with the ethnicity group above, and this assumption is confirmed in results showing that these children participate relatively more in music activities and trips/excursions. In addition we found that this group of children was more involved in reading activities (t-value -2.87 and p = .004) than children in general. Of the selected activities reported for children with inadequate language (for their age), 18 percent of the observations were reading activities compared to 9 percent of the observations for other children.

Physical and psychological disabilities

Children with physical disability were included in about 3 percent of the observations (21 observations) and in these cases one child had this trait. We know from Dataset 2 that 10 children had physical disabilities in the participating kindergartens (in 7 of 31 groups). Our assessment is therefore that this group of children has been clearly represented in our observations.

The main finding is that the activities for this group of children were not significantly different from activities in general, with one exception. Children with a physical disability were significantly more represented in music activities (independent t-test, t-value -6.59 and p = .000). Inspection of data shows that music activities were reported in 33 percent of the observations compared to 4 percent for children in general.

Children with psychological disabilities were included in only 2 percent of the observations and in these cases one child (14 observations, in 15 cases the teachers were not sure but these cases were not included). The teachers reported in Dataset 2 that only seven children had psychological disabilities, which means that this group also seems to be clearly represented in our observations.

Again we found that the activities for this group of children were not significantly different from children in general, with one exception. Children with psychological disabilities were more represented in reading activities than children in general (independent t-test, t-value=-2.40 and p = .017). The results show that for the studied group of children, reading activities were reported in 29 percent of the observations whereas this was the case for 9 percent for children in general.

Discussion

Methodological considerations

The present study claims to highlight children's activities in Norwegian kindergartens, which is of course a very important knowledge base for all professional discussions about quality and policy within this area. Our contribution to this field should be assessed in light of all methodological challenges briefly outlined in five main points below (a-e). First (a) we have a selection of kindergartens invited to the study and finally included in our sample. We invited kindergartens with a formal kindergarten teacher training partnership with Bergen University College, and it may be suggested that these kindergartens do not represent the average Norwegian kindergarten. It may be suggested that these kindergartens are more oriented to change and more updated on challenges and strategies discussed within early childhood education. Kindergartens in Bergen may also be deviant from kindergartens in more rural areas both with regard to employees and places to visit in their near surroundings. We agree that our sample may have this bias, but on the other hand, our results show kindergartens that we would call "quite traditional" with regard to children's activities. It may also be added that the 18 kindergartens represent different ownerships and a wide range of sizes and organizational structures. We hope that future contributions from studies based on other samples will supplement our study, but meanwhile we cannot see any obvious reason to take sample precautions when discussing our main results.

Secondly (b), in the present design kindergarten teachers select observations to be reported. In many cases they could probably report some other activities instead of the selected ones. It may be added that the observers had the opportunity to report several activities (up to four) during each spell of work, but at any rate they make selections. It should be noted that this methodological problem will be present in all studies of field observations. To handle this problem we elaborated an instruction to guide the observers as outlined in our introduction. We emphasized the importance of also including in reporting "the daily life" in kindergartens activities that were unclear to the observer. In our view, the teachers have reported activities in line with our instructions, and Dataset 2 shows that the observers themselves did not experience the selection of activity as a problematic issue. Only two teachers expressed that it was rather difficult to select observations to report and only one teacher experienced some difficulties reporting the observations in our standardized form. We conclude that selection of observations is necessarily a methodological issue in this kind of research, but the reported observations seem to be clearly in line with the aim of the study.

Furthermore (c), we may discuss how the researcher (first author) influences the data by defining and categorizing the activities. The observer gave a qualitative description of the activity along with a suggestion of labeling. Based on this information the researcher categorized the activities, and normally it was not difficult to sort out the reported activities. We are not claiming, however, that the activities were mainly one dimensional. In many cases the described activities were multidimensional, which means that several activities took place simultaneously (as also mentioned in the introduction), but in most observations it was rather evident what was the main activity. This assessment is also supported by the teachers who regarded 73 percent of the observations as "clear".

At a more detailed level, it may be added that role play was in some cases difficult to categorize because this activity is so embedded in children's play. We often found an aspect of role-play when children were playing in the sandpit, when they were climbing on frames and playing with cars etc. It can therefore be discussed whether the prevalence of role-play actually is underestimated in our study. Another viewpoint is that role­play could be regarded not as a category but an aspect that more or less is integrated in other activities. We have applied role-play as a category because the distribution of roles and the play in itself is very explicit in many of the observations.

Another methodological issue (d) is that children's daily life in kindergarten is described as prevalence of activities in the present study. It may be argued that the link between prevalence estimates and cultural formation is not simple and straightforward as also mentioned above (introduction). Activities that go on for a longer time and/or involve many children may have an impact on children that is underestimated in the present study. From this viewpoint, some of the planned activities perhaps get too little attention in our study. On the other hand, data shows clearly that most activities were performed within small groups of children and it is our general impression that activities in kindergartens normally last for a relatively short period of time. We therefore regard the prevalence estimates as meaningful measures in exploring children's daily life in kindergarten.

Finally (e) our study was performed one particular week in October 2008. Is it reasonable to apply these data to present general descriptions of children's activities in kindergartens? A majority of the teachers reported (Dataset 2) that the actual week had been "a normal" week for the year, but we also asked them whether children's activities vary significantly across different times of the year. For indoor activities about 30 percent of them gave a positive answer (meaning yes, to a high or relatively high degree), whereas almost 60 percent gave a positive answer with regard to outdoor activities. On this background we suggest that the time of the year should be taken into account when interpreting the results. It should be noted, however, that most of the activities reported in our data were indoor activities which normally do not vary by time of the year.

Our conclusion at this point is that the present study can highlight children's activities in kindergarten and point out some important issues for debate, in spite of methodological restrictions in our research design. In the following we will discuss some main findings.

Focus on variations

Kindergartens in Norway are allowed to have a particular subject profile (as for example "outdoor kindergarten") and they vary in size, structure and ownership. Such variations are also represented in our sample. It is therefore interesting to note that our study does not find systematic differences between kindergartens with regard to the type of activities reported. This finding suggests that children in Norwegian kindergartens experience to a high degree the same patterns of activities

A major issue in the present study was exploration of possible differences between groups of children. For Norwegian kindergartens it is an explicit policy to counteract traditional differences across gender. From this perspective it is a positive result that almost half of the activities included both girls and boys. On the other hand, we also found more girls-only groups participating in art performance and more boys-only groups participating in construction activities and playing ball outside.

With regard to age, two findings should be noted in our view. First, that a relatively low level of music activities was observed for the oldest children. These activities were often planned and typically initiated by an adult. We therefore suggest that adults working with older children critically consider more use of music activities. Secondly, we are worried about the result showing a negative relationship between age and reading activities. In our view, reading activities should not be restricted to the youngest children.

Finally, the present study pays attention to the minority groups: children with a different ethnicity, inadequate language, physical or psychological disabilities. In general these groups do not have an activity pattern different from other children, which is a positive result in our view. When we focus on the few exceptions from this main picture, it should be noted that these groups were never underrepresented in any of the activities. They were instead overrepresented in some activities, in particular music and reading activities. For children with inadequate language (for their age), reading activities can be seen as a very relevant activity and were probably often initiated by adults. Reading activities may also be a positive stimulation for children with psychological disabilities.

Conclusion

Our study suggests that the Norwegian kindergartens seem to be rather homogenous which may be interpreted as rather egalitarian conditions for cultural formation. On the other hand this finding may suggest a strong national kindergarten-tradition defining what is regarded as natural kindergarten activities for children. From this perspective, it can be a challenge to question children's activity pattern Our results also show a very positive integration of the minority groups we have focused on (even though the analyses were based on relative few observations). We found that boys and girls cooperate in almost half of the reported activities, a result that invites us to a discussion of what is actually the final aim from a gender equality perspective? With regard to age, we have above encouraged kindergarten teachers to reflect on older children's declining participation in reading and music activities. The present study provides, however, only an overview of some aspects of variation in children's activities in kindergarten, which means that further explorative research is needed to better understand conditions for cultural formation.