Science Education and Beyond: Citizen Science in Primary School Potentially Affects Conceptual Learning and Socio-Emotional Development

Abstract

Citizen science (CS) is considered a powerful supplement for teaching natural sciences (and beyond) at school. Even though involving children of primary school age in scientific activities is still uncommon, previous studies confirmed that they can contribute meaningful data as citizen scientists. Yet, the administrative efforts of organising the pupils‘ participation in research activities are high for both, schools and scientists. Typically, some children benefit enormously from participating in a CS project; however, others don’t. To enable decisions for school representatives and funding agencies, empirical tests of the learning benefits of involving CS in routine teaching are needed. This chapter focuses on CS in the education context and wraps up the results of critical tests of (i) factual learning during a project on the social behaviour of a free-living bird species, that is, Greylag geese (Anser anser), (ii) conceptual learning, that is, the transfer of knowledge to new contexts and the children’s concepts of ‘friendship’ and (iii) impulsive behaviour control in primary school children involved in a project as citizen scientists.

1. Introduction

Citizen science (CS) is increasingly popular as a supplement for teaching natural sciences (and beyond) at school [1, 2]. However, involving children in primary school age is rather rare yet [3]. Teachers and scientists still tend to doubt the young children’s capacity of contributing to CS data, although Frigerio et al. [4] emphasized that even 6- to 10-years old children may contribute serious data as citizen scientists if they were well introduced and guided through the process. Thus, CS may benefit from involving children at school as citizen scientists and meanwhile, CS involving schools has emerged as a promising field [5]. The actors involved, that is, scientists, teachers and parents tend to believe that the children will benefit in terms of learning from CS activities. There is empirical evidence that the collaboration between education and research increases the students’ motivation for out-of-school learning [6, 7]. However, often it is assumed that the children will gain knowledge from participating in CS and that the benefits may go beyond learning specific contents by the potential transfer of facts to new insights into the children’s personal environment and by inducing personal interest [8, 9]. These beliefs rarely have been challenged, which inspired us to conduct the studies reported in this chapter. We attempted to ask, whether CS indeed is more effective than the routine of science education in the classroom. Do extracurricular contents and the experience with CS propose extra value in (science) education at school?

2. The educational perspective

Typically, some children benefit enormously from participation in a CS project [2, 10, 11], while others don’t. This may be viewed as the given individual variance in the data observed from a sample and considered acceptable when the results are within satisfying effect size range. However, from an educational perspective, every child in the class must be included in the teaching activities [3] and thus, educational research has a particular focus on those individuals, who were lacking effects of CS participation [12]. The identification of factors that contribute to the efficacy of CS projects in the school context in terms of successful learning is indeed challenging [11]. Even though this was not the focus of the studies presented here, a better understanding of such factors is desirable, and unraveling individual learning outcomes on top of science education [10] might add pieces to the puzzle. To enable decisions for school representatives and funding agencies, empirical studies are needed to test the assumed (learning) benefits of these efforts, as well as the potential developmental benefits of involving CS in routine teaching.

This chapter wraps up the results of critical tests involving primary school children in a CS project on the behaviour of free-roaming greylag geese (Anser anser) at the Konrad Lorenz research center (KLF) in Austria. The focus is on pedagogical aspects of CS in terms of learning specific facts about greylag geese (study 1), conceptual transfer based on the children’s knowledge of goose behaviour (study 2), and experimental tests of impulsive behaviour control in children involved in the CS project with greylag geese and in a control group (study 3). We aimed at testing different hypotheses concerned with the learning outcomes of primary school-aged children participating in a CS project on the behaviour of greylag geese and the relationships between individual birds in the studied goose flock (Box 1). Study 1 represents tests of the basic learning assumptions for children as citizen scientists. Study 2 goes beyond that by testing one aspect of meaningful learning, which is relevant for science education, and study 3 focuses on developmental and behavioural aspects, which are relevant for successful learning.

Box 1.

Overview hypotheses.

Study 1—Learning outcomes: Participation in the CS project has different effects on learning specific facts about greylag geese than being instructed based on pictures and books. Learning effects are independent of age, and similar effects are observed in children of primary levels I and II (level I: class 1 and 2 and level II: class 3 and 4). The gained factual knowledge is sustainable and applicable to new contexts.

Study 2—Concepts of friendship: Participation in CS enables 8- to 10-years old children to transfer the biological concepts of social alliances (as observed in the model bird species) to their own experiences with friendship.

Study 3—Executive functions: Participation in the CS project with free-living greylag geese promotes the children’s capacity to control impulsive behaviour, that is, the project participation has an effect on the socio-emotional development of 6- to 7-year old children.

2.1 Monitoring learning and beyond

Pupils from four primary schools were invited to participate in the CS project on the behaviour of free-roaming greylag geese. The flock of 140 individually marked greylag geese can be found along the river and on the meadows around the Konrad Lorenz research center in Grünau (Austria). The semi-tame and free-living geese are employed as a model species for animal social systems. The social relationships among these birds have been studied for decades; however hitherto, the focus was primarily on pair partners and hierarchical relationships [13]. In the presented CS project, the focus was on social allies in the goose flock (i. e. non-sexual partners), which were discussed with the children in the context of ‘friendships’. The CS project (2017–2020) aimed at gathering data on the local abundance of individual geese in the valley and the modulation of activity patterns by the social context (Box 2).

Box 2.

Additional information about the study site, the focal species and the main CS project.

Study site: The study area is located at 550 m above sea level in the valley of the river Alm at the northern edge of the Austrian Limestone Alps (47°48΄E and 13°56΄N). Some of the data for the main CS project were collected at the Cumberland Wildpark (approx. 60 ha area), a game park neighboring the Konrad Lorenz research center (KLF) and used by the model avian species during the breeding season.

Study species: The greylag goose is one of the model species of the KLF, where scientists conduct basic research on the complexity of avian social systems [13]. The flock of greylag geese was introduced in the valley by late Konrad Lorenz and co-workers in 1973 [14]. The birds are unrestrained and experience natural predation mostly from red foxes (Vulpes vulpes) with losses of up to 10% of the flock and 90% of the goslings per year [15]. The birds are individually marked with a unique combination of coloured leg bands. No two individuals have the same combination of coloured leg bands. The birds are habituated to the close presence of humans, that is, they can generally be approached at less than 5 m. Individual life-history data have been monitored since 1973 and provide reliable information about the age of the breeding birds and social relationships among individuals within the flock (i. e. paired or not, parental and sibling relationships), as well as information on reproductive performance (i. e. breeding attempts, clutch sizes and number of fledged offspring) [15, 16]. The flock does not migrate. Rather, it is regularly fed twice a day year-round and included approximately 130 individuals at the time of data collection.

CS project: The main research objective of the project (2017–2020) was the investigation of the social behaviour of a highly social and long-lived vertebrate, the greylag goose (Anser anser), which can be considered a model for studying animal social systems. In fact, in group living vertebrates social context is among the major stress factors affecting physiology, behaviour, fertility and the immune system of the single individuals. Several different school classes of regional primary schools actively participated in this project. The participants contributed to surveying the temporal patterns of the local abundance of the greylag geese during the breeding season. Previous results have shown that the data collected by citizen scientists are reliable sources of information [17]. The results of the CS project that was studied here, indicated seasonal preferences of the greylag geese for certain locations. In addition, the results have shown that breeders and non-breeders often visit the same places. This aspect was particularly important for population management and public administration bodies (e. g. hunting laws and regulations). The manifold of actors and tasks involved in this project demonstrates the long-term cooperation of the KLF with several regional educational institutions and provides the basis for further research projects.

The children from six classes from three different schools (ABC in Table 1) and additional two classes from a fourth school D (see Section 2.1.3) repeatedly visited the research center in Grünau (Austria) and conducted field observations of the behaviour and the relationships between individuals in the free-roaming flock of greylag geese. For assessing the basic scientific learning (study 1), a total of seven classes from three different schools were subject to the presented evaluation of learning outcomes (ABC in Table 1). In study 1, three classes were aged between 6 and 7.9 years, that is ‘primary level I’ according to the Austrian education system, and three classes were between 8 and 10 years, that is ‘primary level II’. For the assessments of dynamic conceptual understandings (study 2), a subsample of older children (i. e. primary level II) was chosen because this part involved open questions, and children were required to have already firm writing skills. In the subsample of 8- to 10-year old children (primary level II in school C) study 1 and study 2 were conducted in cooperation by collecting answers for both research questions during the same exams (questionnaires, which were not part of formal assessments). Study 3 was independent of the assessments in studies 1 and 2 and tested aspects of executive functions in another cohort of 6- to 7-year old children (primary level I) from school D (Table 1).

	CS groups	N	Control groups	N
Primary level I	School A: 1a	12	School B: 1b	18
	School A: 1b	17	School A: 2b	19
	School B: 2b	17	School B: 2a	19
Primary level II	School B: 3a	19
	School B: 3b	23	School C: 4a*	21*
	School C: 3a*	21*	School C: 4b*	21*
Total N (study 1)		109		98
Total N (study 2)*		21*		42*
Primary level I	School D: 1a	23	School D: 1b	24
Total N (study 3)		23		24

Table 1.

Sample sizes for CS and control classes by primary level I and II, which were tested in the three studies.

*The subsample of children, who were also tested in study 2 (school C only).

Study 3 was conducted with two classes from a different school than studies 1 and 2 (school D). Control groups were from the same schools, age-matched and received direct indoors instructions based on images rather than participating in the CS activities.

Initially, the scientists’ team visited the school classes and provided indoor introduction to explain the biology of this bird species and its behaviour (February 2018). Then, the children started outdoor activities and visited the KLF three times, that is, during the mating season in March, the breeding season between April and early May, and the parenting season in late May/June 2018 (Table 2). During the visits to the KLF, the children learned to identify individual birds in the flock of 140 individually marked greylag geese and to conduct behavioural observations (Figure 1).

Timeline	Activity	Indoors (I); Outdoors (O)	Study
Jan	Questionnaires (t1) experimental tests for EF	I	1, 2, 3
Feb	Introduction to Greylag geese (biology, behaviour)	I	1, 2, 3
Mar	First visit KLF (mating season, observing behaviour)	O	1, 2, 3
April/May	Second visit KLF (breeding season, observing behaviour)	O	1, 2, 3
June	Third visit KLF (goose families, observing behaviour)	O	1, 2
	Questionnaires (t2) experimental tests for EF	I	1, 2, 3
July/August	National summer holidays		1, 2, 3
Sept	Third visit KLF for EF cohort	O	3
	Questionnaires (t3) experimental tests for EF	I	1, 3

Table 2.

Overview of CS project activities with school classes.

EF: executive functions, KLF: Konrad Lorenz research center, t1, t2, t3: sampling of the presented data.

Figure 1.

Children recording the behaviour of two greylag geese at a meadow around the Konrad Lorenz research center (photo: Archiv KLF).

2.1.1 Study 1: learning outcomes

In study 1, we tested the specific knowledge of 109 pupils (aged 6 to 10 years; Table 1) about greylag geese and their behaviour before the project input had started (January 2018) and after six months of CS project participation (June 2018). Additionally, for assessing the long-term effects of CS participation on learning, we repeated the tests after nine weeks of summer holidays (September 2018). Children in control groups did not participate in the CS project (N = 98, 5 parallel school classes, same school, age-matched; Table 1), that is, they were not visiting the flock of greylag geese and were taught greylag goose behaviour by teacher students providing direct instructions and using images. The children from control groups were tested with the same questions and at the same times as the children who participated in the CS project (Table 2). The children’s knowledge of greylag geese was assessed in form of written questionnaires with ten multiple choice questions. The questionnaires were not part of the formal assessments by the teachers. With the school beginners (level I, 6- to 7 years old), the teacher assisted the children with reading the questions and answers to choose (multiple choice) in the exam. In school C, the total number of questions was reduced to three questions to allow combined assessment with study 2 within a reasonable time. Therefore, the results from school C were not entirely comparable with the plots of schools A and B and thus, were not included in Figure 2.

Figure 2.

Patterns of learning specific biological facts about greylag geese did not differ between CS and control groups. Plots show the mean proportions of correct answers per class (± SEM) and are presented as age matched units of analyses, upper panel 6- to 7-years, mid panel 8 years, bottom panel 9- to 10-years old children. Different letters at the bottom of each plot indicate significant changes within groups (Repeated measures ANOVA primary level I, first classes, upper panel, within groups: X² = 11.8, df = 2, P = 0.003; between groups: X² = 2.8, df = 2, P = 0.247; primary level I, second classes, mid panel, within groups: X² = 14.4, df = 2, P < 0.001; between groups: X² = 0.8, df = 2, P = 0.667; primary level II, bottom panel, within groups: X² = 11.7, df = 2, P = 0.003; between groups: X² = 13.5, df = 1, P < 0.001).

The assessments consisted of nine questions addressing factual knowledge on greylag geese and one question to test transfer knowledge. This one question included in the questionnaires dealt with identifying the images of the feet of a greylag goose among images of the feet of other bird species. The morphology of the feet of the geese was not directly addressed during the CS activities or subject of instructions in the control classes. Thus, this question is aimed at the children’s ability to identify the feet of a greylag goose (i. e. of waterbirds) in a comparative context. The issue was not discussed in the classroom/CS activities intentionally, to be able to ask one question, that demanded a transfer of factual knowledge to a meaningful novel context. This question was also included in the questionnaires of school C.

2.1.2 Study 2: concepts of friendship

Study 2 assessed in a subset of 8- to 10-years old children from school C the concepts of friendship with regard to both, relationships between individual greylag geese and transferred to the children’s own social experiences and their subjective views of friendship (N = 63; Table 1). We chose an age-based subsample of children at primary level II (Table 1) for this part, as it was crucial for us that the children were sufficiently able to read and write down answers to the open questions. The children were asked to explain how to identify a social alliance (a ‘friendship’) between individual geese (open question format), whether they thought that ‘friends’ were important for a greylag goose (multiple choice), and whether friendship was important in their own lives (multiple choice). In the subsample, the answers to three questions from study 1 were assessed in combined questionnaires. To assess the effects of the CS activities on the children’s concepts of friendship, the questionnaires were employed twice, before and after the children’s CS activities, that is, in January and June 2018. The children’s answers to the open format question (‘How to identify friends in a goose flock?’) were categorized in a qualitative approach by assessing phrases in the texts, which were indicative of the children’s individual conceptions of friendship. This approach allowed for applying two categories of individual concepts related to social affiliation between individual geese (which were also applicable to concepts of friendship in humans): relationship concepts based on social support [18] or relationship concepts based on spatial proximity between the allies [19]. A third category was employed when children indicated that friendship was ‘not detectable in geese’.

2.1.3 Study 3: executive functions

The focus of study 3 was on the children’s executive functions [20, 21] and thus, the effects of the CS activities on their socio-emotional development. Executive functions (EF) are predestined during early childhood and are needed to regulate a number of socio-emotional competencies and behaviours that are needed for successful learning [20]. Study 3 employed experimental tests of executive functions with 6–7 years old children from the fourth school before and after participation in the CS project (N = 23) and in an age-matched control group (N = 24) in the same school.

This research question emerged during the planning phase for participating in the CS activities. The teacher of this class recognized already at this point that the focal cohort of school beginners might have problems with impulsive behaviours [22] when approaching the free-living greylag geese. To be able to observe the free-roaming birds’ behaviour, it is necessary to quietly approach the birds and behave appropriately. If children start running towards or approaching a goose in a boisterous and thoughtless way, the entire group of birds would probably fly off and stay away from the location until the next day. Thus, to be able to observe the birds’ behaviour and participate in the CS project, the children were demanded to approach the birds slowly and behave appropriately. The children were very interested in participating in the CS project, and the teacher thought that interacting with the free-living birds would be a great motivation for the children to potentially improve their ability to control their impulsive behaviours (for example, see [23]). To test this assumption, a battery of individual experimental tests was conducted (N = 47) three times in total. As in study 1, these tests were run before the CS project had started (January 2018), after CS activities (June 2018, i. e. after three visits to the KLF) and after the summer holidays in September 2018.

The battery of tests for individual scores in the context of executive functions included three elements: (i) a classical test for impulsive control, the ‘silly sound stroop test’ using a flipbook with 14 images of dogs and cats [24]. To assess working memory function, we chose (ii) a test for age adequate (first graders) mental arithmetics with a predetermined time limit (up to 45 calculations in three minutes) and (iii) a visual working memory task with a time limit (6 images within 45 seconds; Figure 3). The latter tests are similar to the mosaique tests, which are elements of the WISC-V (Wechsler Intelligence Scale for Children, former HAWIK; [25]). The WISC-V intelligence test assesses different cognitive functions that are also relevant for neuropsychological diagnostics [26].

Figure 3.

Examples of images employed for the visual working memory task in study 3 (mosaique tests).

2.2 Effects of CS on learning and beyond

2.2.1 Study 1: learning outcomes

In sum, we observed that teaching facts about the greylag geese and their behaviour was equally effective whether using conservative classroom instructions with materials, such as pictures and books or when children participated in outdoor experiences as citizen scientists (Figure 2). Figure 4 demonstrates with one exemplary question (e. g. young greylag geese are called goslings) that children at primary level II from control groups may reach equal proportions of correct answers as the children from CS classes. The learning progress lasted until September in both CS and control groups, who were retested after the nine weeks of Austrian national summer holidays (Figures 2 and 4).

Figure 4.

One selected question to demonstrate that indoor classroom instruction in control groups resulted in similar proportions of correct answers (e. g. young greylag geese are called goslings) in children from control groups as in CS-groups (primary level II).

Another selected question is worth being discussed, as higher proportions of children who participated in the CS project were able to distinguish the feet of a goose from other (non-waterbird) species feet than in the control groups (identify the feet of a goose based on multiple choice of images of the feet of a crow, a goose and a chicken, Figure 5). Both groups were not specifically instructed about the feet of a goose and the question was regarded as a quest for applying factual knowledge in a new context. In this case, the children from CS classes scored higher for the transfer of learnt content to a new context than children in control groups.

Figure 5.

Higher proportions of children at primary level II who participated in the CS project displayed correct answers than in the control groups (The task was to identify the feet of a goose based on multiple choice of images of the feet of a crow, a goose, and a chicken).

Thus, at the level of reproducing facts classroom instructions based on pictures and books were well effective for learning. These patterns were similar in all age groups. Yet, when the aim is to apply factual knowledge to new contexts, that is, to reach insights and competencies rather than reproducing mere facts, the participation in CS allowed more effective learning than the conservative instructions. This dimension of learning is considered connectable (e.g., see [27]) and useful for further science education (i. e. in secondary school), and it corresponds to the demands of the Austrian curricula [28] for the learning outcomes of science education at the primary level I and II. In a previous study, we observed an additional effect of CS activities on the children’s knowledge of bird diversity [11], which indicated that personal interest in the subject was triggered by the CS experiences. The results of the current study add that, potentially, CS activities stimulated the children’s skills for applying factual knowledge in a new context and beyond the model bird species.

2.2.2 Study 2: concepts of friendship

The children’s concepts of friendship between individual animals changed in a different way in the CS class than in the control group (Figure 6). The experiences of detecting social allies (i. e. friends) within a group of geese by observing the birds’ social behaviour led CS children to describe the nature of 'friendships’ with concepts of social support and spatial proximity (Figure 6a), to rate higher scores for the benefits of having an ally in geese, as well as the value of having a friend in their own lives (Figure 6b). This result was clearly different in one of the control groups; however, the patterns of the two control groups were inconsistent. In contrast to control groups, the pattern of the children with CS experiences confirmed the hypothesis that biological concepts of social alliances may be transferred to the children’s own experiences with friendship (Box 1).

Figure 6.

Children who had participated in the CS activities (black bars) had clear concepts of how to observe friends in the flock of greylag geese (upper panel, categories of the children’s statements answering an open format question, Figure 6a) and their understanding of the importance of social allies for the birds (bottom panel, children indicating high relevance of social allies for greylag geese and of friendship for own experiences, Figure 6b). Bars show percent change from t1 (in January) to t2 (in July). Numbers underneath bars are proportions of answers at t1.

Concepts of friendship are not easily studied in 8- to 10-years old children, and some variance was probably added by the individually different verbal abilities in children between eight and ten years. It is also possible that the children in one control group (school C 4a in Figure 6) were providing random answers to the questions dealing with friendship. Alternatively, the answers of the children in this control group were already different from the other two groups at t1 (in January; Figure 6). Thus, we suggest interpreting these results with caution and future tests of this hypothesis will be needed. However, if the comparison of the conceptual transfer in the CS class and the remaining control group (school C 4b in Figure 6) is at focus, the results demonstrate clear effects of CS activities at the mechanistic level (how to detect social allies in geese), however, no effects at the functional level, that is, importance of social allies.

The potential benefits of extracurricular contents for science education have been shown before with regard to factual learning and personal interest [11]. The results presented here suggest that learning at the level of conceptual insights is potentially possible, and future focus should be on further understanding the additional factors needed to reinforce the efficacy of CS for conceptual skills in science education [27]. One promising factor is to inform and involve the teachers who plan to participate in a CS project during early project phases [29]. A close cooperation with the teachers involved is desirable, yet teachers perceive cooperation with scientists as an additional task and tend to refrain from additional tasks as time is constrained. However, the time spent with CS activities may be regarded worthy, if teachers were encouraged to connect the learning outcomes of CS activities with curricular demands [30]. Scientists who manage CS for schools need to develop approaches to catch the personal interest of the teachers involved.

2.2.3 Study 3: executive functions

The experimental tests of selected aspects of executive functions, such as impulsivity in ‘silly sounds stroop tests’, and working memory in mental arithmetics and visual mosaique tests revealed significantly better scores in children of the CS class than in the control group. These differences between groups were particularly evident after the third visit to the research station in September (t3 in Figure 7). The capacity to control impulsive behaviour was essential to involve the children in the project with the free-living greylag geese.

Figure 7.

Six- to 7-years old children with CS experience (t3) score higher in experimental tests of three aspects of executive functions than the children in the control group. (a) control of impulsive behaviour in silly sound stroop tests [24]; (b) working memory in mental arithmetics tests; (c) working memory in visual working memory tests (mosaique test as in [25]). (One-way ANOVA between groups df = 1: (a) t1: X² = 1.2, P = 0.279; t2: X² = 6.3, P = 0.012; t3: X² = 5.4, P = 0.020; (b) t1: X² = 0.1, P = 0.776; t2: X² = 3.4, P = 0.066; t3: X² = 12.9, P < 0.001; (c) t1: X² = 14.3, P < 0.001; t2: X² = 8.6, P = 0.327; t3: X² = 11.9, P < 0.001).

The observed effects of the CS participation on the children’s behaviour are clearly connected to the project experiences rather than due to maturation because they differ from the control group. The curious decrease of successful hits in the mental arithmetics task (Figure 7a) is conceivable, as the teacher gradually increased the difficulty of the task to complement the ongoing developments of the pupils during the school year. Even if this difficulty resulted in falling patterns, the CS group was clearly ahead (Figure 7b).

The results confirm the hypothesis that participation in the CS project with free-living greylag geese promotes the children’s capacity to control impulsive behaviour (Box 1). The CS experiences were an effective chance for training impulsivity and the individual development of the ability for socially adjustable behaviour [25]. We assume that this effect was strongly mediated by the animal-assisted approach of the CS activity (e.g., see [23]). Most children love to interact with living animals, a phenomenon that may be understood in the context of biophilia [31, 32]. Kelemen-Finan et al. [10] have shown that CS activities of primary school-aged children affected their motivation and attitudes towards wild animals, such as wild bees. The results of the study presented here demonstrate that involving children in CS activities may be beneficial at various levels, by adding the level of socio-emotional development and hence, there are indeed potential effects beyond the factual learning level of science education.

3. Conclusions

The results of the three empirical studies emphasize that participation in CS projects may affect the learning outcomes at different levels in primary school-aged children. In sum, primary school-aged children may benefit enormously from being involved in research activities with scientists, particularly when animals are involved [4, 33]. The presented studies investigating the benefits of CS participation from an educational perspective demonstrated that these benefits are not exclusively at the level of factual learning. Regarding factual learning, the children in control groups (who were taught the same contents with pictures and books) reached similar scores as the CS groups. However, at a conceptual transfer level (i. e. transfer to new context and the children’s concepts of friendship) and with regard to socio-emotional development (i. e. impulsive behaviour control) and working memory the children in CS groups clearly scored better than the age-matched control groups. In addition, previous results showed from the scientific perspective that primary school-aged children produced high-quality data within a CS project, hence the children as citizen scientists were satisfying scientific standards [4]. The results presented in this chapter complement this observation from the educational perspective by showing that also the primary school-aged children may benefit from CS activities at various levels of learning—hence the CS project was satisfying educational standards, as well—even beyond curricular demands.

It may be argued that the efforts of organizing the participation of schools in CS research activities are high for both, teachers and scientists. Even if the costs including time resources may seem high at first, CS is a very strong supplement for conceptual teaching at school and potentially promotes competencies and advanced personal skills (even) in primary school children. Nevertheless, there was still relatively large variance in the presented data and education should keep a focus on those children who scored below the average effect. We noted some factors that contributed to the efficacy of CS projects in the school context from discussions with colleagues, teachers and children who took part in the CS project. For example, it is recommended to involve the teachers in (i) elaborated introduction to and discussions about the subject, which should include (ii) agreements on potential learning outcomes and (iii) enable the teachers to practical and theoretical guidance during CS activities with formal training before the CS activities start. Another recommendable feature of the presented CS project was (iv) the long-term and repeated nature of CS activities, which probably supported the long-term effects, that is, the children’s knowledge lasted until after nine weeks of break during the summer holidays. It also seems helpful if the involved teachers develop a genuine interest in the topic and activities in order to identify connections to the curriculum, as well as in their own teaching. More understanding of the factors reinforcing these potential effects is yet needed if the aim is to teach skills and concepts, which may be transferred to new contexts in the children’s personal environment.

In sum, the results of the presented studies suggest that the participation of children in CS as part of their science education at school bears great potential to advance the learning outcomes from the level of factual knowledge to contextual learning and hence, to promote the development of scientific and personal skills. The concept of contextual teaching and conceptual learning has recently been discussed in the literature on science education with a focus on advancing science education in the field of biology. The participative and motivating nature of pupils acting as citizen scientists in the field of biology satisfies the contextual teaching strategies of ‘relating, experiencing, applying, cooperating and using knowledge in a new context’ [34], which were originally proposed for teaching mathematics; however, they also match the educational requirements for primary science education very well [35]. The contexts experienced in the presented CS project may be related to several biological core concepts as the analyses of the observed goose behaviours may relate to past and future knowledge in the contexts of behaviour with structure and function, information flow, systems and evolution [27, 36]. These discipline-specific core concepts are not yet at the focus of citizen scientists; however, it is a great chance to advance an efficient collaboration between education and science and to encourage teachers to participate in CS activities. Finally, we want to highlight to all actors involved in CS (may it be in a school context or with adult citizen scientists) to consider potential additional learning outcomes about and beyond scientific facts for citizen scientists, for example, through the transfer of learned facts to new contexts within the citizens’ personal experiences that offer meaningful insights about their environment.

Acknowledgments

We gratefully acknowledge M. Bauer, S. Kaiblinger, E. Kornfeind and S. Kögler for assistance with designing the questionnaires to assess the effects of CS activities and data collection. G. Gegendorfer and J. Rittenschober for assistance with the birds in the field and A. Szabo, K. Oberroithner and S. Amering for conducting the indoors instructions in the control classes. We also thank B. Lankmaier and the staff of the Cumberland Wildpark Grünau im Almtal for the willingness and engagement to build up a long-term cooperation with the KLF and their CS activities. Funding for the CS project was provided by the program Sparkling Science, Project SPA-06/155 to DF. Permanent support came from the ‘Verein der Förderer der Konrad Lorenz Forschungsstelle’ and the ‘Herzog von Cumberland Stiftung’. We greatly appreciated the interest in the studies by the schools involved in the CS activities and data collection and the support by J. Leeb and K. Soukup-Altrichter. I. Fertschai added valuable comments on an earlier draft of the manuscript.

Conflict of interest

The authors declare no conflict of interest.