Triggers of Students’ Efficacious Interaction in Collaborative Learning Situations

Abstract

This article presents a case study that investigated primary school students’ social interaction while working in small groups in science. The aim was to identify what characterizes and triggers students’ efficacious interaction in collaborative learning situations. This was done by exploring and analyzing students’ and groups’ task involvement and the quality of their activity. The micro-level analysis proceeded first by identifying episodes of group task involvement and evaluating the quality of these interaction episodes. Next, the transitions between episodes were explored by identifying what triggered the shifts between task involvement levels. Ten hours of video observation data captured fourth-grade students, aged 9 to 10, in two groups of 3 students studying “the vital conditions of life” for 5 weeks. The main findings indicated that efficacious interaction demanded collaboration between group members and required active participation and productive on-task working. Three types of triggers were found to cause changes in students’ activity, of which the group progress trigger was the most powerful for increasing, and the contextual trigger the most powerful for decreasing the activity. These findings shed light on certain aspects of efficacious interaction that will help us to identify details of efficacious interaction in future studies.

Keywords

collaborative learning efficacious interaction triggers of interaction interaction analysis video analysis

Introduction

Previous small group interaction research has extensively described practices, processes, and conditions that lead to effective interaction (Cohen, 1994; Greeno, 1997; Soller, 2001; Tuyay, Jennings, & Dixon, 1995). The focus of these interaction studies has been on investigating the form and the content of individuals’ behavior, specifically social interaction. Thus the main interest has been in investigating the features, forms, and nature of group interactions (e.g., Mercer, 2000; Webb, Troper, & Fall, 1995) and identifying different patterns and episodes that are conducive to learning in small groups (e.g., Fisher, 1993; Hogan, Nastasi, & Pressley, 2000; Mercer, 1996). According to Johnson and Johnson (1996), effective small group interaction can be described as group work in which all group members contribute to solving a problem and spend minimal time doing off-task activities. Soller (2001) extends this definition by suggesting factors indicating effective interaction in collaborative learning: namely, participation, social grounding, active learning conversation skills, performance analysis, and group processing. Interaction is effective when it fulfills the characteristics of these factors. In other words, to be able to measure effectiveness, these objectives need to be achieved.

Effective small group interaction can also be discussed from the point of view of efficacious interaction. The concept of efficacious interaction originates from sociocognitive theory, which emphasizes that self-efficacy beliefs have a central role in human performance (Bandura, 1986). According to this approach, self-efficacy influences several aspects of agentic behavior that are important for learning (Bandura, 1986, 1997; Zimmerman, 2000). In this study, these beliefs are seen as creating a mental capacity for students to participate and engage in social interaction and in productive on-task working. We argue that efficacious interaction depends on engagement in collaboration in which group processes, such as working methods, tools, shared goals, and roles, are socially negotiated within the group’s learning situation. Therefore, we believe that efficacious interaction can be reached in authentic learning situations by exploring and analyzing students’ and groups’ task involvement and the quality of their activity.

Self-Regulated Learning and Efficacious Interaction

The field of educational research is actively seeking solutions to promote each learner’s active engagement in learning––in other words, each one’s efficacious learning. Self-regulated learning (SRL) research has provided valuable information regarding how efficacious learners develop learning skills and use them effectively (Boekaerts, Pitrich, & Zeidner, 2000). By integrating cognition and motivation, engaged learners are empowered to take a purposeful role in their own learning (Schunk, 2001; Zimmerman, 2000). Cognitive studies have emphasized investigating individuals’ knowledge structures and cognitive strategies. These studies have indicated, for example, a number of cognitive strategies that enhance learning (Paris, Byrnes, & Paris, 2001; Weinstein, Husman, & Dierking, 2000) and have shown how these skills can be learned and developed (Winne, 2001). From a motivational point of view, research has provided important information concerning how motivational constructs such as self-efficacy, attributions, intrinsic motivation, and goals trigger a learner’s cognitive skills and further influence achievement and learning (Eccles, Wigfield, & Schiefelt, 1998; Graham & Weiner, 1996; Pintrich & Schunk, 2002). In all, learning is a proactive process in which self-motivated efficacious learners act as agents of their own behavior and, by using effective learning strategies, achieve desired results (Zimmerman, 2001).

Efficacious Interaction in Collaborative Learning

Previous studies of SRL have typically acknowledged the social and contextual nature of learning. Recent research has emphasized practices, processes, and conditions that lead to effective social interaction and the construction of knowledge in different learning situations (Hmelo-Silver, 2003; Hmelo-Silver & Barrows, 2008; Soller, 2001). Collaborative learning is generally seen as mutual engagement of participants in a joint effort to solve the problem together (Roschelle & Teasley, 1995), where sharing authority among group members is very important (Dillenbourg, 1999; Roschelle & Teasley, 1995). In learning situations where the aim is to work in groups, collaborate, and create something together, individual motivation and learning skills are further challenged by the learner’s social skills. A key to successful collaboration in these learning situations is social interaction (Kreijns, Kirschner, & Jochems, 2003). For example, studies of collaborative learning and peer interaction in different settings and contexts have identified specific interactions that seem to promote learning (see Barron, 2000; Dillenbourg, 1999; Roschelle & Teasley, 1995; Stahl, 2006). Mercer (2000) described how learners participated in educational activities by characterizing the features, forms, and nature of group interactions. Hiltz’s (1994) findings underlined the importance of social interaction stating that the process of developing shared understanding through interaction is the natural way for people to learn. These studies have shown that when students are interacting, they elaborate on their understandings and become aware of their thinking processes. Learning is seen not only as taking place in the learner’s mind but also as a process of meaning-making within social practices (Kumpulainen & Wray, 2002). The practices of sharing and constructing perspectives in collaborative interactions seem to also promote reflection, planning, and metacognition (Arvaja, Häkkinen, Eteläpelto, & Rausku-Puttonen, 2000).

In this study, efficacious interaction is investigated within collaborative learning situations, where one of the key features is collaborative problem solving. In collaborative problem solving, students co-construct knowledge through productive interaction (Hmelo-Silver, 2004). When students work in small collaborative groups, they learn what they need to know to solve a problem. There are a variety of techniques that might help establish productive collaboration and effective interaction. For example, scripted cooperation and the use of student roles have all been used to support effective collaborative learning and interaction in small groups (King, 2007; Weinberger, Ertl, Fischer, & Mandl, 2005). These roles help ensure that all group members are cognitively engaged in joint problem solving (Hmelo-Silver, 2003). Collaborative learning situations create opportunities for efficacious interaction by engaging students to productively work on-task and by encouraging them to ask questions of each other, explain and justify their opinions, articulate their reasoning, and elaborate and reflect on their knowledge, and thereby motivating and improving learning (Dillenbourg, 1999).

The Aim of This Study

Efficacious interaction has not been comprehensively studied in earlier studies. Despite extensive studies of effective small group interaction, there seems to be little, if any, evidence on the elements of students’ efficacious interactions in small-group learning contexts. The earlier research on efficacy has been dominated by the individual psychometric tradition where the measurements of efficacy have been targeted to the individual interpretations of students’ own efficacy, not as an interactive construct observed in student interactions. Therefore, the aims of the present study are to identify (a) what characterizes and (b) what triggers efficacious interaction in collaborative learning situations. The specific research questions are

Research Question 1a: How do off-task and on-task involvement occur in collaborative learning situations?

Research Question 1b: What is the quality of on-task involvement?

Research Question 1c: Is task involvement related to the quality of interaction?

Research Question 2a: What are the different kinds of factors that trigger changes in activity?

Research Question 2b: What triggers students to engage in on-task activity?

Method

Participants

This case study focuses in detail on two 3-student (N = 6) small group interactions while working on a science task. The participants were fourth-grade primary school students aged 9 to 10 who participated in a science project for 5 weeks. Students were familiar with each other and were used to working together during various school lessons. Teachers assessed members’ learning success, so that both groups contained high and low achievers, and determined how the groups would best collaborate. According to the teacher’s evaluations and ratings, all group members possessed the same level of prior knowledge and had equal learning skills. Group A consisted of two girls, Hanna and Anna, and one boy, Jake. Group B was composed of two boys, Simo and Aaron, and one girl, Sara.

Context

The students studied a science topic from the learning kit called vital conditions of life, which was divided into seven different subtopics: water, air, nutrition, heat, light, family, and human rights (Malmberg, Järvenoja, & Järvelä, 2010). The task was to construct shared understanding of the vital conditions of life. The pedagogical task structure provided guidelines for the groups regarding how to get started. This was done by asking the groups to assign roles to each group member. Three roles were provided: leader, recorder, and information seeker. The leader was responsible for keeping the group focused on the assigned task and for making sure that all members of the group were participating and contributing to the group’s assignment. Jake and Aaron were chosen for these roles. The recorder was responsible for writing up notes of the material that was contributed by each group member and for preparing materials for the presentation, which took place at end of the project. Anna and Sara were chosen for these roles. The information seeker was responsible for preparing and organizing the information search, gathering materials or resources for the group, and monitoring the use of materials. Hanna and Simo were chosen for these roles.

The working process during the collaborative group work included four phases. First, students internalized the requirements of the task by discussing and planning their goals and strategies for achieving them. This phase also included negotiating each other’s roles. Next, they researched the topic. In the third phase, students brought together their knowledge of the topic and decided on which subtopic(s) to concentrate. In the last phase, students prepared a presentation of chosen subtopic(s) for the whole class. In all these phases, the students were simultaneously involved in the group’s task execution and shared tools and activities.

Data Collection

This case study consisted of video observation data of the groups’ interactions in these collaborative learning situations. The groups were videotaped with three cameras. Two cameras were focused on the group’s primary working stations while the third camera captured students working elsewhere in the classroom. Both groups’ working processes were videotaped during 20 lessons, resulting in above 10 hr of video observation data.

Data Analysis

Qualitative Analysis

This case study was based on the assumption that efficacious interaction is reached when students are actively and collaboratively involved in social interaction. The micro-level analysis proceeded first by sequencing the whole set of transcribed video observation data into episodes of groups’ task involvement (Järvelä, Veermans, & Leinonen, 2008). Functions identified in the students’ actions and utterances were coded into four task involvement categories: no one is working (OFF), one student is working (ON1), two students are working (ON2), and the whole group is working (ON3).

The second phase of the analysis concentrated on analyzing the quality of the groups’ on-task episodes. This analysis was grounded in the theoretical perspectives of social learning (Levine, Resnick, & Higgins, 1993) and on recent developments of collaborative learning (Dillenbourg, 1999). Through the theory-driven content analysis (Chi, 1997), collaborative on-task categories (ON2, ON3) were divided into five categories of collaborative and cooperative interaction. Collaborative interaction refers to activity where group members worked together as a group. This includes three categories: (a) equal participation and productive on-task work, which appeared typically when a group’s goals and roles were clear for each member and when group members were equally engaged and collaborated to work with their assignment, (b) student helping and assisting another student, and (c) silent and intensive work, which occurred only if preceding episodes contained discussions or negotiations regarding the goals or roles of each group member and if students were aware of how their individual work would contribute to the success of the group’s assignment. Cooperative interaction refers to activity in which group members worked separately. It includes two categories: (a) unequal participation and unproductive on-task work, which occurred when students disagreed with either the goals or roles of their working and (b) working on-task individually, which typically occurred when students did not want to use the same tools or a shared working space or when, for some reason, they were not interested in collaborating with one another.

In the third phase, the qualitative content analysis (Chi, 1997) focused on identifying the triggers that affected student interaction during collaborative group work. In this study, triggers were factors or events regarding a group’s interaction that influenced its members’ task involvement and the quality of their interaction. Triggers were investigated in the transitions between episodes (Järvelä et al., 2008; Järvelä, Salonen, & Lepola, 2001) by exploring the reasons why students’ task involvement levels changed—that is, why students decided to start or quit working on a task. These reasons were coded into three trigger categories made up of factors that either increased or reduced a student’s efficacious activity. (a) An individual progress trigger refers to an individual’s positive or negative comments or actions that affected one’s own activity. Examples of the factors included in this category include making comments in regard to knowing what to do, actions such as changing one’s working method or an individual’s physiological state. (b) Group progress trigger refers to positive or negative evaluations, comments, or actions made by others; these are implemented, for example, through social reinforcement. (c) Contextual trigger includes factors within the pedagogical structure, such as the presence of the teacher, help from people other than group members, cameras and microphones, or filling out a motivation scaffolding sheet.

Reliability of Qualitative Proceedings

The analysis of reliability covered all three phases of analysis. The primary coder instructed a second coder about the coding principles by providing coding schemes that included categories, definitions, and a list of what to identify. These principles were negotiated before the second coder started to classify the reliability of the coding. However, the intercoding process was blind.

The reliability analysis was applied to refine the coding scheme and improve the results of the analysis. In this study, the reliability of the coding scheme was analyzed by measuring internal consistency using Cronbach’s alpha. The average correlation of each coding scheme item with other coding scheme items was determined; 0.7 is acceptable. The reliability of data analysis was analyzed using Cohen’s Kappa, which is commonly used to quantify intercoder reliability; 0.5 is considered to be acceptable. In the first phase, the reliability coding aimed to analyze task involvement episodes. Internal consistency of the coding scheme was 0.986; intercoder reliability was 0.922. The second phase of reliability analysis focused on the quality of interaction. Internal consistency of the coding scheme was 0.846; intercoder reliability was 0.742. In the third phase, the reliability analysis focused on triggers that initiated transitions between episodes. Internal consistency of the coding scheme was 0.846; intercoder reliability was 0.509.

Statistical Measures

The qualitative analysis was complemented by statistical measures. Cross-tabulation was conducted to measure the relations between both task involvement and quality of interaction and triggers and changes in activity. One episode represented one unit of analysis. One unit of analysis included coding for task involvement and quality of interaction as well as coding for triggers and changes in activity. The effect sizes for χ² were calculated using Cramer’s V.

Results

Groups’ Task Involvement

In the task involvement analysis, the groups’ interactions were coded as being in either off-task (OFF) or on-task (ON1, ON2, ON3) categories, resulting altogether in 241 task involvement episodes. Figure 1 illustrates one example of task involvement episodes. The example is from a video that lasted 20 min and contained eight episodes. Off-task activity referred to a situation in which students’ activities did not focus on the task—this included playing around, discussing free-time activities, or doing absentminded activities. Functions that were related to on-task activity referred to instances in which students worked on the task by planning and discussing the task or the topic and when they searched for information, worked with notes, made a mind map, or practiced their presentation.

Figure 1.

Example of how task involvement episodes appeared in video observation data

On the ON1 level, one group member worked on a task while others were on the OFF level. On the ON2 level, two group members worked on a task––either individually with notes or together sharing the responsibility of the task. One group member was on the OFF level. On the ON3 level, the whole group focused on doing the task either individually or collaboratively.

The groups’ task involvement is described in more detail in Table 1. For both groups, the most typical task involvement level was ON1, describing 31% of the groups’ use of their time. Group A had a total of 134 episodes, during which this group was active in some of the on-task levels 76% of the time. The most frequent task involvement level was ON2, which described the group’s activity 29% of the time. Group B had 107 episodes, of which the group worked on task 77% of the time. Within group B, the predominant level of task involvement was ON1, covering 28% of all episodes.

Table 1.

Groups’ Task Involvement

	Group A				Group B
	Hanna Information seeker/recorder	Jake Leader	Anna Recorder	All	Simo Information seeker/recorder	Sara Recorder	Aaron Leader	All
	f	f	f	f	f	f	f	f
OFF	40	64	81	32	44	57	59	24
ON1	21	5	3	27	16	8	7	31
ON2	37	29	14	39	21	16	15	26
ON3	36	36	36	36	26	26	26	26
Total	134	134	134	134	107	107	107	107
On task	69.6%	51.9%	36.3%	76%	58.5%	46.2%	44.3%	77%
Off task	30.4%	48.1%	63.7%	24%	41.5%	53.8%	55.7%	23%

The task involvement analysis showed that group members internalized the task’s requirements by discussing, planning, and negotiating their goals, strategies, and roles for working. It is notable that Hanna and Simo were the most active members in their groups and worked on the ON1 level more often than others. Although both of them were designated as information seekers, they also adopted other roles to ensure that the group work would be productive.

The Quality of On-Task Involvement

The quality of the groups’ social interactions was analyzed by looking at the 127 ON2 and ON3 task involvement episodes. The quality of the groups’ task involvement is summarized in Table 2.

Table 2.

Quality of Interaction

	Collaborative				Cooperative
	A	B	Both Groups		A	B	Both Groups
Equal participation and productive on-task work	30 40%	21 41%	41%	Unequal participation and unproductive on-task work	9 12%	9 18%	14%
Student(s) helping and assisting another student	9 12%	4 8%	10%	Working task individually	27 36%	14 27%	33%
Intensive work, silent episodes	—	3 6%	2%
Total	39/75 52%	28/52 55%	67/127 53%		36/75 48%	23/52 45%	59/127 47%

Data analysis shows that most of the time groups’ social interactions were collaborative in nature, the category of equal participation and productive on-task working being the most typical (41%). In group B, the collaborative interaction could appear as silent episodes when the group members worked intensively for 5 to 24 min without talking. Although the working in these episodes was done silently, during preceding episodes group members discussed and negotiated goals for each individual’s work and were aware of how their individual work would contribute to the group’s assignment. For example, the comic-drawing episodes were characterized as such. In cooperative interaction, the predominant category was working on a task individually (33%).

The following examples illustrate the characteristics of collaborative interaction. In Example 1, the group had difficulties beginning their work, and one group member sought help from the teacher. The teacher encouraged group members to start planning the comics together, thereby enabling the students to change the direction of their activity so that it became more productive. Example 2 illustrates how one student helped and assisted another student in the gStudy environment and enabled her to continue with the task.

Example 1

Aaron:

How should we start planning this [drawing comics]?

Teacher:

First, you should think about what kind of things you want to present in your comic. What does nutrition mean to you?

Aaron:

But how should we start doing this?

Teacher:

You could start by listing what nutrition means.

Simo:

Fruits, bad nutrition, meat, fish, healthy nutrition . . .

Teacher:

What is the title of your comic?

Sara:

What happens in first panel? Who is in it? What kinds of characters?

Simo:

Basic stick figures.

Sara:

What are they saying? How many stick figures do we have?

Simo:

It seems to me that we have eleven subtopics. Eleven. How are we gonna fit them into six panels?

Sara:

We don’t have to do them all.

Simo:

Okay, should we take two subtopics? How about healthy and bad nutrition?

Aaron:

Okay, that’s a good idea!

Example 2

Anna:

If I want to add water to this, how do I get it there? Hanna, how do I get water there? I don’t understand how to do this.

Hanna:

To [the] first row, you need to type the name of the group you’re in. All the basic information has to be completed. But “my observation” is not even the correct basis for your note. You should use the basis of “my findings.”

Anna:

Well, what is a finding anyway?

Hanna:

It’s your own explanation of the topic.

Anna:

Okay. Now I get it!

The following examples present descriptions of the episodes that characterized cooperative interaction. In Example 3, the leader abused his role and caused unequal participation and unproductive on-task working between group members.

Example 3

The group has chosen nutrition as their subtopic of “Vital conditions of life.” Aaron makes sure that Sara and Simo know what they are expected to do. Instead of participating in the work, he just sits back and watches the others do their share of the work. Sara is working intensively with nutrition notes in gStudy while Simo is looking for information in textbooks. However, Simo has lost his focus on the group’s task and is concentrating on information about water, which is not the group’s subtopic.

Example 4 presents an episode where students worked on-task individually in the group but are not aware of each other’s activities.

Example 4

Group members are working individually. Aaron and Simo are looking for nutrition information. Aaron is reading a textbook while Simo is looking on the Internet. He is also working with the group’s notes in gStudy. Sara is writing the plot of the group’s comic in the group’s notebook. The group members are not communicating with each other about the information they have found or about the plot of the comic.

It can be seen that in these episodes the students did not communicate at all. Instead of engaging in collaborative learning and productive on-task working, they tended to concentrate on completing the task. In these episodes, the individual working processes seemed to be more important than working on a task collaboratively.

The Relationship Between Task Involvement and the Quality of Interaction

The results indicate a statistically significant relation between task involvement and the quality of interaction. Table 3 shows that working on the ON2 level were most typically cooperative in nature. When students were working on the ON3 level, the quality of their interaction was typically collaborative in nature. In other words, when the working level was ON2, the group members worked on the assignment individually, and when working on the ON3 level, they implemented equal collaboration between group members. The findings were similar in both groups.

Table 3.

Relation Between Task Involvement and the Quality of Interaction

	Collaborative				Cooperative
	A^a		B^b		A^a		B^b
	f	%	f	%	f	%	f	%
ON2	16	41	10	40	23	59	15	60
ON3	23	64	18	69	13	36	8	31
Total	39	52	28	55	36	48	23	45

χ²= 3.920, df = 1, p = 0.048, Cramer’s V = 0.048.

χ²= 4.398, df = 1, p = 0.036, Cramer’s V = 0.036.

Triggers Affecting Students’ Activity

Triggers were analyzed by locating the transitions between the episodes and by exploring the reasons for the changes from one task involvement episode to another. Altogether, 235 transitions were analyzed and coded into trigger categories, which are summarized and described in Table 4. The results of the trigger analysis showed that triggers either increased (f = 164) or decreased (f = 71) students’ social interaction.

Table 4.

Triggers Causing Changes in Activity

Trigger/definition	Increased activity	Decreased activity
Individual progress	Student has a clear vision of the task and knows what to do and how to do it.	Student has difficulty concentrating on task or is no longer interested.
Consists of an individual’s positive or negative comments or actions that affect one’s own efficacious activity.	Hanna: “I’m gonna start doing this and you can choose whether you’ll do the same or continue doing nothing.”	Aaron: “I’m bored. Are we still doing the same thing? I don’t want to.”
Group progress	Group members’ joint discussion spurs student(s) to continue working. Student receives help or support and is able to continue on task.	Student needs to wait for other group members in order to continue on task or feels she or he has already completed the task.
Consists of positive or negative evaluations, comments, and actions from others. Social reinforcement can take the form of verbal praise or criticism or non-verbal communication such as smiles, frowns, and gestures.	Hanna: “I brought papers for all of us. Now, everyone, start to write information about something. You’ll write about water.”	Sara: “I just finished my first comic. Can I start drawing the next one?”
	Jake: “We have only one note about water but two about nutrition.”	Aaron: “No, you need to wait until we’re ready.”
	Hanna: “Yeah, that’s why me and Anna will write about nutrition. Anna will write about the information we have in gStudy and I’ll write about the information I found from this book. Is everything clear? Remember: good handwriting!”	Sara: “How long is it gonna take?”
		Aaron: “As long as it takes!”
Consists of contextual factors that affect students’ efficacious activity either positively or negatively.	The presence of or help from an adult encourages student to start working on task.	The presence of the cameras, an adult or a student from another group, a lack of time, or filling in the motivational scaffolding sheet interrupts students’ work.
	Teacher: “How are you doing?”	Teacher: “Time is up. Please fill in the motivation scaffolding sheet before going on break!”
	Hanna: “Just fine. Now our task is to decide the questions that we want to find answers for.”
	Jake: “And why do we need to ask those questions?”
	Hanna: “Because they help us to find the right kind of information.”
	Teacher: “Brilliant, Hanna!”

Individual progress trigger refers to an individual’s positive or negative comments or actions that affected one’s activity by either increasing or decreasing it. For example, activity usually increased when a student had a clear vision of the task and knew how to proceed with it. On the other hand, when a student had difficulty concentrating, or when the task was no longer interesting, his or her activity was decreased. Group progress trigger refers to positive or negative evaluations, comments, and actions from others that influenced some of the group members’ task involvement. Typically, this included different forms of social reinforcement, such as verbal praise, or nonverbal communication, such as smiles, frowns, and gestures, that increased students’ activity. Contextual trigger refers to factors derived from the pedagogical structure, such as the presence of the teacher or help from people other than group members that increased the student’s activity. Factors from the surrounding context, such as cameras and microphones, or factors from the pedagogical task structure, such as filling in a motivation scaffolding sheet, were most often the cause of a decrease in student activity. Figure 2 illustrates how triggers occurred in video observation data. The example presents the same video that was introduced in Figure 1 but goes on to display triggers and the change in the direction of activity (increase/decrease).

Figure 2.

Transitions and triggers between task involvement episodes

Triggers of On-Task Activity

Interaction was increased in 164 transitions. Group progress trigger most frequently influenced the groups’ on-task activity (54%). Generally, at times when group progress triggers increased the group’s activity, the triggers were found to contain elements such as encouragement, getting help or support, and positive impact on modeling others. At such times, students discussed the task or the topic, or worked with notes or their presentation. In Example 5, the group progress trigger is illustrated. The group members are discussing the task requirements and planning the working processes. Individual progress trigger influenced the groups’ on-task activity 40% of the time. The students progressed when they had a clear vision of the task and when they self-regulated their course of action. Usually, individual progress triggers prompted students to work individually with notes or search information. In Example 6 below, Hanna tried to start working and asked for help from peers and adults several times. For some reason, she was not able to start working and became bored. She decided to start writing notes from the textbook into the group’s notebook. Contextual triggers increased the groups’ activity 5% of the time. This is illustrated in Example 7, in which the teacher’s help allowed the group to overcome the setback of not knowing how to proceed.

Example 5

Hanna:

Okay, now we sit down and then we start planning this on the basis of this paper.

Do group members understand the task? [reads directly from the instructions]

Jake:

How should we start doing this?

Hanna:

We need to start thinking of— The task is to produce a common understanding of vital conditions of life. Anna, do you understand this?

Anna:

Well . . . .

Hanna:

It’s not specific enough.

Jake:

Well, what then?

Hanna:

It’s just not enough. We must think about this question together. We need to understand the vital conditions of life.

Example 6

Hanna:

I’m gonna start doing this, and you can choose whether you’ll do the same or continue doing nothing.

Example 7

Teacher:

How are you doing? What have you accomplished today?

Simo:

Found out that meat, berries, salad, and fruits are healthy.

Teacher:

What else can you do besides listing all the foods? Have you read the information other group members have found about nutrition?

Aaron:

Yes.

Teacher:

Maybe you should think about the most important nutrition-related information and plan who will research each one.

Aaron:

Actually, we have done that.

Teacher:

Okay, what is your part?

Aaron:

I need to find out— Simo, what was I supposed to do?

Teacher:

Precisely! Could it be something you boys start to do next?

Simo:

Yes.

Findings from statistical analysis complemented the results found in the qualitative data analysis (see Table 5). Results indicated that there was a statistically significant relation between triggers and increased activity.

Table 5.

Relation Between Triggers and Activity, Results of χ² Tests, and Effect Size

	Increased activity				Decreased activity
	A^a		B^b		A^a		B^b
	f	%	f	%	f	%	f	%
Individual progress trigger	45	51.7	21	51.2	42	48.3	20	48.8
Group progress trigger	50	96.2	39	88.6	2	3.8	5	11.4
Contextual trigger	5	12.2	4	13.3	36	87.8	26	86.7
Total	100	55.6	64	55.7	80	44.4	51	44.3

χ² = 66.448, df = 2, p = .000, Cramer’s V = 0.608.

χ² = 41.491, df = 2, p = .000, Cramer’s V = 0.601.

The trigger that most often increased activity was the group progress trigger, which only rarely decreased activity. The contextual trigger had the reverse affect: it usually triggered a decrease and only a few times increased activity. The effect of the individual progress trigger equally increased and decreased activity. Findings were similar in both groups.

Discussion

This study focused on investigating small group interaction by identifying the characteristics and triggers of efficacious interaction in collaborative learning situations. The main findings propose that efficacious interaction is defined by active participation and productive on-task working and is enhanced and maintained through collaboration between group members. Three types of triggers were found affecting change in students’ activity, and the group progress trigger was related to an increase in students’ collaborative interaction during 54% of the increased-activity transitions. These findings are new and complement the traditional research on classroom engagement.

Previous SRL and social interaction research has indicated that the relationship between learning and interaction is extremely complex. These processes are constructed by the participants and shaped by their intentions and interpretations (Kumpulainen & Wray, 2002). Recent research has shown how collaborative learning gives students more opportunities to participate in, observe, reflect on, and practice socially shared ways of knowing and thinking (e.g., Dillenbourg, 1999; Hmelo-Silver & Barrows, 2008). As a result, the role of practices that support collaborative learning and social interaction have increased in today’s classrooms. It is important to better understand the processes and conditions that take place during interactions between students in the context of small group learning (Greeno, 1997; Tuyay et al., 1995). Yet it seems that merely providing opportunities to participate in social interaction does not necessarily lead to meaningful and effective learning experiences (Kumpulainen & Wray, 2002). We argue that when studying students’ social interactions in collaborative learning situations, we need to consider new concepts and new ways to analyze it to measure effectiveness. In this study, active participation and productive on-task working was used to describe efficacious interaction. It was analyzed through task involvement and quality of interaction and by identifying triggers causing the changes in students’ activity. We believe that by doing this we will be able to see more in social interaction and in overall learning process.

In this study, efficacious interaction was supported by reciprocal attempts to create joint meaning while working collaboratively on task. This is in accordance with many studies that have stressed the reciprocal understanding in collaborative interaction. For example, Häkkinen and Järvelä (2006) found that reciprocal understanding and sharing was critical for higher level collaborative web-based discussions in a higher education context. Furthermore, Byman, Järvelä, and Häkkinen (2005) showed that reciprocity was an essential component not only of social interaction but also of successful virtual interaction. In individualistic interaction, instead of engaging in joint reasoning and meaning-making, the students tended to concentrate on completing the task. In these episodes, an individual’s working processes seemed to be more important than working on a task collaboratively. This reflects the need for individual, self-regulated learning skills as well as for shared, group-level regulation (Hadwin, Järvelä, & Miller, 2011). The results support the idea that the social context created by a collaborative task structure engages students in a social interaction. However, the dynamics of small group interaction seem to pose both opportunities and challenges for students engaged in productive and effective learning. These findings indicate the importance of students’ social and cognitive skills in addition to their level of motivation in maintaining productive and task-focused interaction.

The earlier research on efficacy has been dominated by the individual psychometric tradition where the main method used has been self-report questionnaires (e.g., Bandura, 1997, 2006). Thus the measurements of efficacy have been targeted to the individual interpretations of students’ own efficacy (Bandura, 1997), not as an interactive construct observed in student interactions. We have not found other empirical studies using video data analysis on efficacious interaction. The methodological challenge was whether the episodes of efficacious interaction could be identified from video observation data. Therefore, the micro-level analysis concentrated first on the elements of group interactions in terms of task involvement. The analysis of task involvement was carried out by identifying the episodes of task involvement (Järvelä et al., 2008) and exploring the quality of group interactions (Dillenbourg, 1999; Levine et al., 1993)—in other words, considering to what degree the work was shared during these episodes. The main findings of this case study indicate that simply working on the task did not guarantee that the group’s work was collaborative in nature. To reach the level of efficacious interaction, students needed to be actively involved in the task, and the quality of their work needed to be collaborative in nature. Collaborative interaction contains many of the same elements that have been found to be effective in learning, such as shared ways of knowing and thinking as well as participation in joint meaning-making (Fisher, 1993; Greeno, 1997).

Analyzing the total amount of time spent on task involvement levels did not reveal enough about the nature of social interactions. To gain a deeper understanding of the connection between task involvement and efficacious interaction, the trigger analysis concentrated on investigating the transitions between task involvement episodes (Järvelä et al., 2008) by exploring the reasons why students stopped working or decided to start working again. Three trigger categories were found. All three types of triggers were found to increase the groups’ efficacious interaction, although the role of the contextual trigger was minimal. The appendix summarizes the manner in which the qualitative analysis was carried out by providing a coherent picture of how the video observation data was analyzed. The intention was to show how the triggers functioned in the interaction and how the quality of the interaction was analyzed.

Reliability and validity issues of all kinds apply to video-based data just as they apply for any other kind of quantitative or qualitative data analysis (Derry et al., 2010). The reliability of qualitative analysis was performed to all three phases of analysis and to whole sets of video data. An independent coder coded 100% of the videos. Cronbach’s alpha and Cohen’s kappa were used to establish internal consistency and intercoder reliability for a second coder. The qualitative analysis was complemented by statistical measures. Crosstabulation was conducted to measure the relations between, on one hand, task involvement and quality of interaction and, on the other, triggers and changes in activity. Also, the effect sizes for χ² were calculated using Cramer’s V. Our study contributes to the existing body of research concerning how to use both theory-driven content analysis and pure content analysis to analyze social interactions recorded through video observations.

There are some limitations in this case study that are derived mainly from the research design. Because of the small sample size, the results of this study cannot be generalized. Therefore, this case study has focused on describing the elements of efficacious interaction through micro-level analysis. Because of the limited data regarding students’ experiences of social interaction, it was not possible to analyze students’ prior relationships. After all, given children of this age group, the group members’ prior relationships could be one possible component influencing the nature of the group interactions.

In the future, the aim is to identify the characteristics of efficacious students in different kinds of collaborative learning situations and to observe students in an authentic classroom environment. The objective for future research is to shed light on different factors that influence and are related to the overall progress of efficacious interaction in collaborative learning. Examining both efficacious interaction and efficacious students requires a research approach that combines situated measures of how individuals experience learning activities in collaborative learning situations with conventional measures of individuals’ cognitive skills and affective beliefs. Although the video observation method will have a strong role in future data collections, other data sources, such as self-reports or stimulated recall data, need to be used to gain further insight into the processes of efficacious interaction in collaborative learning situations.

Footnotes

Appendix

Example of the Analyzed Data

Time	Participation	Transcribed interaction	Context	Task involvement	Quality of interaction	Trigger
3:15	Simo	“I’m gonna look at our notes. Notes, notes. We’ll take that and that one. I’ll check that each has . . .”	Simo is working with the computer group’s notes in the learning kit and checking that each note has all the information needed.	ON1 Simo	COOPERATIVE Simo is working individually on the task	ON1 to ON2 Individual.breakAaron knows what to do and likes the work.
3:52	Aaron	“I need to check how many panels we’ll need for our comics.”	Aaron comes over and wants to start planning the group’s comics. Simo would like to participate, but Aaron ignores Simo’s attempts. Simo continues working with the group’s notes and Aaron starts to draw panels for the comics.	ON2	COOPERATIVE	ON2 to ON1
	Simo	“Six.”		Simo, Aaron	Simo and Aaron lack a shared understanding of how to proceed and are working on the task individually	Group.
	Aaron	“Yeah, but I’ll check that myself. Let’s see…” [Starts drawing the panels].		ON1		Simo finishes the notes and would like to draw the comics. Since the panels are not ready, he needs to wait.
				Aaron
						ON1 to ON3
						Contextual
						Group is getting help from the teacher.
6:57	Teacher	“Hey, you’ll need to start thinking about how to draw together. Aaron can’t be the only one who is doing the drawing.”	Teacher comes to see how the group is proceeding. He is not pleased that Aaron is the only one who is working and that others need to wait. The group is planning how to draw the comics together and the teacher realizes that they can do the panel with the help of a computer to speed up the work.	ON3	COOPERATIVE	ON3 to OFF
	Aaron	“But I want to do that. I’ll be fast.”		Simo, Aaron, Sara	Students have problems regarding how to proceed; the participation is unequal.	Contextual.
	Teacher	“How could you do that together? How many panels do you have in your comics?”		OFF		Teacher leaves to go to the printer.
	Simo	“Six.”				OFF to ON3
	Teacher	“Well then, maybe Simo could draw two, Sara two, and Aaron the last two panels. What do you think?”				Contextual.
						Teacher comes back and helps group to start working.
						Group.
						Group members’ joint discussion and planning spurs students to start working.
	Sara	“Yes.”
	Teacher	“Or we could do the panels with computer and print them out. You could start drawing the comics then.”
	Simo	“Okay!”
12:25	Teacher	“Is the plot of your comic ready?”	Teacher comes back with the printed panels. Simo and Aaron are planning the comic figures and Sara is writing down the plot.	ON3	COLLABORATIVE
	Aaron	“Yes. Hey look, everyone. The skinny one looks like this and the fat one like that.”		Simo, Aaron, Sara	Students are equally participating in drawing.
	Teacher	“Good, you have agreed how the figures look alike. That way we can recognize them. Really smart!”
	Aaron	“Now we need to decide which panels each one of us will draw. Let’s draw the kind of fingers that Simo did. Those are great extra!”

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study has been funded by Academy of Finland research project no. 135100.

Bios

Elina Määttä, MEd, is a researcher at the Learning and Educational Technology Research Unit (LET), University of Oulu, Finland. Her research interest is in self-regulated learning, especially in efficacious interaction for self-regulated learning among young children.

Hanna Järvenoja, PhD, is a postdoctoral researcher at the Learning and Educational Technology Research Unit (LET), University of Oulu, Finland. Her research interest is in self-regulated learning, especially in emotion and motivation regulation in individual and socially shared learning situations.

Sanna Järvelä, PhD, is a professor and head of the Learning and Educational Technology Research Unit in the Department of Educational Sciences, University of Oulu, Finland. Her research interest deals with learning processes in technology-enhanced learning social and motivational processes in learning, self-regulated and computer-supported collaborative learning.

References

Arvaja

Häkkinen

Eteläpelto

Rausku-Puttonen

(2000). Collaborative processes during report writing of a science learning project: The nature of discourse as a function of task requirements. European Journal of Psychology of Education, 15, 455-466. doi:10.1007/BF0317298710.1007/BF03172987

Bandura

(1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: W. H. Freeman.

Bandura

(2006). Guide for creating self-efficacy scales. In Pajares

Urdan

(Eds.) Self-efficacy beliefs of adolescent (pp. 307-338). Greenwich, CT: IAP-Information Age.

Barron

(2000). Achieving coordination in collaborative problem-solving groups. Journal of the Learning Sciences, 9, 403-436. doi:10.1207/S15327809JLS0904_210.1207/S15327809JLS0904_2

Boekaerts

Pintrich

P. R.

Zeidner

(Eds.). (2000). Handbook of self-regulation. San Diego, CA: Academic Press.

Byman

Järvelä

Häkkinen

(2005). What is reciprocal understanding in virtual interaction? Instructional Science, 33(2), 121-136. doi:10.1007/s11251-004-7689-810.1007/s11251-004-7689-8

Chi

(1997). Quantifying qualitative analyses of verbal data: A practical guide. Journal of the Learning Sciences, 6, 271-315. doi:10.1207/s15327809jls0603_110.1207/s15327809jls0603_1

Cohen

(1994). Restructuring the classroom: Conditions for productive small groups. Review of Educational Research, 64(1), 1-35. doi:10.2307/117074410.2307/1170744

10.

Derry

S. J.

Pea

R. D.

Barron

Engle

R. A.

Erickson

Goldman

. . . Sherin

B. L.

(2010). Conducting video research in the learning sciences: Guidance on selection, analysis, technology, and ethics. Journal of the Learning Sciences, 19, 3-53. doi:10.1080/1050840090345288410.1080/10508400903452884

11.

Dillenbourg

(Ed.). (1999). Collaborative learning: Cognitive and computational approaches. Amsterdam, Netherlands: Pergamon.

12.

Eccles

J. S.

Wigfield

Schiefele

(1998). Motivation to succeed. In Eisenberg

(Ed.), Handbook of child psychology: Volume 3—Social, emotional, and personality development (5th ed., pp. 1017-1095). Hoboken, NJ: John Wiley.

13.

Fisher

(1993). Distinctive features of pupil-pupil talk and their relationship to learning: How discursive exploration might be encouraged. Language and Education, 7, 239-257. doi:10.1080/0950078930954136310.1080/09500789309541363

14.

Graham

Weiner

(1996). Theories and principles of motivation. In Berliner

Calfee

(Eds.), Handbook of Educational Psychology (pp. 63-84). New York, NY: Simon & Schuster Macmillan.

15.

Greeno

J. G.

(1997). On claims that answer the wrong question. Educational Researcher, 26, 5-17. doi:10.3102/0013189X02600100510.3102/0013189X026001005

16.

Hadwin

A. F.

Järvelä

Miller

(2011). Self-regulated, co-regulated, and socially shared regulation of learning. In Zimmerman

Schunk

(Eds.), Handbook of Self-Regulation of Learning and Performance (pp. 65-84). New York, NY: Routledge.

17.

Häkkinen

Järvelä

(2006). Sharing and constructing perspectives in web-based conferencing. Computers and Education, 47, 433-447. doi:10.1016/j.compedu.2004.10.01510.1016/j.compedu.2004.10.015

18.

Hiltz

S. R.

(1994). The virtual classroom: Learning without limits via computer networks. Human-Computer Interaction Series. Norwood, NJ: Ablex.

19.

Hmelo-Silver

C. E.

(2003). Analyzing collaborative knowledge construction: Multiple methods for integrated understanding. Computers & Education, 41, 397-420. doi:10.1016/j.compedu.2003.07.00110.1016/j.compedu.2003.07.001

20.

Hmelo-Silver

C. E.

(2004). Problem-based learning: What and how do students learn? Educational Psychology Review, 16, 235-266. doi:10.1023/B:EDPR.0000034022.16470.f310.1023/B:EDPR.0000034022.16470.f3

21.

Hmelo-Silver

C. E.

Barrows

H. S.

(2008). Facilitating collaborative knowledge building. Cognition and Instruction, 26, 48-94. doi:10.1080/0737000070179849510.1080/07370000701798495

22.

Hogan

Nastasi

B. K.

Pressley

(2000). Discourse patterns and collaborative scientific reasoning in peer and teacher-guided discussions. Learning and Instruction, 6, 345-358. doi:10.1207/S1532690XCI1704_210.1207/S1532690XCI1704_2

23.

Johnson

D. W.

Johnson

R. T.

(1996). Cooperation and the use of technology. In Jonassen

D. H.

(Ed.), Handbook of research for educational communications and technology (pp. 170-198). New York, NY: Simon & Schuster Macmillan.

24.

King

(2007). Scripting collaborative learning processes: A cognitive perspective. In Fischer

Kollar

Mandl

Haake

J. M.

(Eds.), Scripting computer-supported collaborative learning: Cognitive, computational, and educational perspectives (pp. 13-37). New York, NY: Springer.

25.

Kreijns

Kirschner

P. A.

Jochems

(2003). Identifying the pitfalls for social interaction in computer-supported collaborative learning environments: A review of the research. Computers in Human Behavior, 19, 335-353. doi:10.1016/S0747-5632(02)00057-210.1016/S0747-5632(02)00057-2

26.

Kumpulainen

Wray

(Eds.). (2002). Classroom interaction and social learning: From theory to practice. London, UK: Routledge & Kegan Paul.

27.

Levine

J. M.

Resnick

L. B.

Higgins

E. T.

(1993). Social foundations of cognition. Annual Review of Psychology, 44, 585-612. doi:10.1146/annurev.ps.44.020193.00310110.1146/annurev.ps.44.020193.003101

28.

Malmberg

Jarvenoja

Jarvela

(2010). Tracing elementary school students’ study tactic use in gStudy by examining a strategic and self-regulated learning. Computers in Human Behavior, 26, 1034-1042. doi:10.1016/j.chb.2010.03.00410.1016/j.chb.2010.03.004

29.

Mercer

(1996). The quality of talk in children’s collaborative activity in the classroom. Learning and Instruction, 6, 359-377. doi:10.1016/S0959-4752(96)00021710.1016/S0959-4752(96)000217

30.

Mercer

(Ed.). (2000). Words and minds. London, UK: Routledge.

31.

Paris

S. G.

Byrnes

J. P.

Paris

A. H.

(2001). Constructing theories, identities, and actions of self-regulated learners. In Zimmerman

Schunk

(Eds.), Self-regulated learning and academic achievement (pp. 253-287). New York, NY: Springer-Verlag.

32.

Pintrich

P. R.

Schunk

D. H.

(2002). Motivation in education: Theory, research, and applications. Columbus, OH: Merrill-Prentice Hall.

33.

Roschelle

Teasley

(1995). The construction of shared knowledge in collaborative problem solving. In O’Malley

(Ed.), Computer supported collaborative learning (pp. 69-97). NATO ASI Series F: Computer and system sciences (Vol. 128). Berlin, Germany: Springer-Verlag.

34.

Schunk

D. H.

(2001). Social cognitive theory and self-regulated learning. In Zimmerman

B. J.

Schunk

D. H.

(Eds.), Self-regulated learning and academic achievement: Theoretical perspectives (pp. 125-152). Mahwah, NJ: Lawrence Erlbaum.

35.

Soller

(2001). Supporting social interaction in an intelligent collaborative learning system. International Journal of Artificial Intelligence in Education, 12, 40-62.

36.

Stahl

(2006). Group cognition: Computer support for building collaborative knowledge. Cambridge, MA: MIT Press.

37.

Tuyay

Jennings

Dixon

(1995). Classroom discourse and opportunities to learn: An ethnographic study of knowledge construction in a bilingual third grade classroom. Discourse Processes, 19, 75-110.

38.

Webb

N. M.

Troper

Fall

J. R.

(1995). Constructive activity and learning in collaborative small groups. Journal of Educational Psychology, 87, 406-423. doi:10.1037/0022-0663.87.3.40610.1037/0022-0663.87.3.406

39.

Weinberger

Ertl

Fischer

Mandl

(2005). Epistemic and social scripts in computer-supported collaborative learning. Instructional Science, 33, 1-30. doi:10.1007/s11251-004-2322-410.1007/s11251-004-2322-4

40.

Weinstein

C. E.

Husman

Dierking

D. R.

(2000). Interventions with a focus on learning strategies. In Boekaerts

Pintrich

P. R.

Zeidner

(Eds.), Handbook of self-regulation (pp. 727-747). San Diego, CA: Academic Press.

41.

Winne

P. H.

(2001). Self-regulated learning viewed from models of information processing. In Zimmerman

B. J.

Schunk

D. H.

(Eds.), Self-regulated learning and academic achievement: Theoretical perspectives (pp. 153-189). Mahwah, NJ: Lawrence Erlbaum.

42.

Zimmerman

B. J.

(2000). Attaining self-regulation: A social cognitive perspective. In Boekaerts

Pintrich

Zeidner

(Eds.), Handbook of self-regulation (pp. 13-39). New York, NY: Academic Press.

43.

Zimmerman

B. J.

(2001). Theories of self-regulated learning and academic achievement: An overview and analysis. In Zimmerman

B. J.

Schunk

D. H.

(Eds.), Self-regulated learning and academic achievement: Theoretical perspectives (pp. 1-38). Mahwah, NJ: Lawrence Erlbaum.