The impact of flipped classroom based on collective problem solving promotion on students’ learning performance and interaction patterns

The impact of flipped classroom based on collective problem solving promotion on students’ learning performance and interaction patterns

（Effects of a collective problem-solving promotion based flipped classroom on students’ learning performances and interactive patterns）

2019

1. Concept analysis

1. Collaborate to solve problems

Collaborative problem solving is a learning method in which students form groups and complete learning tasks through discussion and collaboration between groups.

2. Mechanism to promote collective problem solving

This study proposes a collective problem-solving promotion mechanism that uses (interactive response system) IRS for collaborative problem-solving and inter-group competition, and applies it to flipped classroom learning activities.

2. Research content

1. Research background:

Flipped classroom teaching is one of the teaching methods that has attracted much attention in recent years. However, when students engage in classroom learning, they may not perform as well as expected without the appropriate educational tools to support them.

2. Research purpose:

This study proposes a collective problem-solving promotion mechanism that utilizesInteractive Response System (IRS) to improveStudents’ interaction and learning engagement in flipped classroom.

3.ResearchMethod:

Adoptquasi-experimental design, with college students as the research subjects. Students in the experimental group adopted this method, while students in the control group adopted the traditional flipped learning method.

This study adoptsa combination of qualitative and quantitative research methods; quantitative content analysis (QCA) and Conduct behavioral analysis of students' collaborative discussions. Lag - sequence analysis (LSA)

4. Research questions:

(1) Can the flipped learning method that uses a collective problem-solving promotion mechanism improve students’ academic performance?

(2) Can the flipped learning method that uses a collective problem-solving promotion mechanismenhance students’ self-regulation ability?

(3) Can the flipped learning method that uses a collective problem-solving promotion mechanismimprove students’ sense of collective efficacy?

(4) Can the flipped learning method that uses a collective problem-solving promotion mechanismimprove students’ interactivity?

(5) Can the flipped learning method that uses a collective problem-solving promotion mechanismpromote students’ higher-order learning behaviors of discussion and interaction?< /span>

5. Research results:

The results show that in the flipped classroom based on the promotion of collective problem solving, students not only have betterlearning performance and collective efficacy, but also There is a higher level of knowledge construction and deeper interactions. The research results can provide reference for future teaching and research.

3. Experimental design

1.Experimental subject:

This study takestwo classes of students from a university in northern Taiwan as the research subjects; one class is the experimental group and the other class is the control group. Each class consists of 36 students.

2.Experimental method:

Research Design This study adoptedquasi-experimental design, including an experimental group and a control group.

The experimental group adopted the collective problem-solving promotion mechanism in the flipped learning method, and the control group adopted the traditional Flipped learning method.

In order to avoid the weaknesses of using only one research method, weadopted both quantitative and qualitative research methods to strengthen the research design and ensure better research results.

Quantitative research methodsUsing questionnaires to collect statistical data; in addition, analysis of variance (ANCOVA) is used to eliminate the impact of learners’ prior knowledge or cognitive differences on learning before the experiment impact on results.

Qualitative research methodsadoptedcontent analysis and lagged sequence analysis to understand students' discussion behavior patterns.

3.Questionnaire design:

Tools for this study includeSelf-Regulation Questionnaire, Collective Efficacy Questionnaire and Interaction QuestionnaireQuestionnaire. This study also used an adaptive behavioral coding scheme to analyze students’ interactive learning discussion behaviors. pre-test and post-test of learning performance, and

Self-regulated learning questionnaireThe proposed online self-regulated learning questionnaire was used. There are 24 test items in total. Using7-point Likert scale;

Collective efficacy questionnaireThe collective efficacy questionnaire proposed by Wang and Hwang in 2012 was used. Using a 7-point Likert scale; Interactive questionnaire, with a total of 18 questionnaires item, it uses a 7-point Likert scale;

Test aspect of learning performance, this study used pre-test and post-test of learning performance . The Kuder-Richardson formula 20 (KR-20) in the pretest was 0.83 and the posttest was 0.81, which has good internal consistency .

4.Experimental concept:

This study proposes a collective problem-solving promotion mechanism that uses IRS for collaborative problem-solving and inter-group competition, and applies it to flipped classroom learning activities. As shown in Figure 1, the mechanism includes the following stages.

Collaboration (primary and secondary); cooperation (no obvious primary and secondary)

5. PracticeProcess：

Figure 2. Experimental process

4. Results

(1) Learning performance

This study uses analysis of covariance (ANCOVA) for data analysis to explore learning performance. The homogeneity test of the regression coefficients within the group verified the hypothesis (F = 1.07, p = 0.304 >.05), and further ANCOVA statistical tests can be performed.

Learning performance raw data:

Learning Performance Covariance Analysis

Yishu

Table 2.Covariance analysis of learning performance

variable	Group	N	Mean	SD	After adjustment M	standard error	F	h2
posttest	test group	36	80.78	3.27	80.78	0.545	116.325	0.624
	control group	36	72.47	3.27	72.47	0.545

***p<.001, **p<.01, *p <.05.

Results: Table 2 shows the results of the variance analysis of learning performance. The result shows that F = 116.33, p = 0.00. The average academic performance of the experimental group (M = 80.78, SD = 3.27) was significantly higher than that of the control group (M = 72.47, SD = 3.27). < 0.05, indicating that there is a significant difference in the academic performance of the two groups of students

(2) Self-regulation

This study used analysis of covariance (ANCOVA) to analyze the self-regulation of the two groups of students. The homogeneity test of the regression coefficients within the group verified the hypothesis of self-regulation (F = 2.79, p = 0.10 >0.05), and further analysis of covariance can be performed. Table 3 shows the results of the self-regulation questionnaire (F = 8.895, p = 0.04 < 0.05).

self-regulating covariance

Table 3. Analysis of covariance for self-regulation

variable	Group	N	Mean	SD	After adjustment M	standard error	F	h2
self-regulation	test group	36	5.17	0.94	5.17	0.171	8.895	0.113
	control group	36	4.44	1.11	4.44	0.171

Since the self-regulation questionnaire consists of six aspects, this study further examined whether these six aspects have independent results, so a covariance analysis was performed on these six factors. Initially, the results tested the homogeneity of the regression coefficients of these six factors; goal setting (F = 1.14, p = 29 > 0.05), environment (F = 2.112, p = 0.29 > 0.05), task strategy (F = 0.68, p = 0.41> 0.05), time management (F = 1.32, p =0.26 > 0.05), help seeking (F = 1.14, p = 0.29 > 0.05), and self-evaluation (F = 3.82 , p = 0.06 > 0.05), all corresponding hypotheses. Analysis of covariance can be carried out further. Table 4 presents the analysis results of the six self-regulation aspects.

Raw data on the six elements of self-regulation:

The average of the six elements of self-regulation:

Analysis of covariance of environment:

Looking for help with analysis of covariance:

Table4.Covariance analysis of six aspects of self-regulation

variable	Group	N	Mean	SD	adjusted mean	Error	F	h2
Target setting	test group	36	4.75	1.08	4.75	0.269	0.266	0.004
	control group	36	4.61	1.20	4.61	0.269
environment	test group	36	5.39	1.08	5.39	0.228	32.886	0.320
	control group	36	4.08	0.84	4.08	0.228
mission strategy	test group	36	5.22	0.96	5.22	0.212	0.616	0.009
	control group	36	5.39	0.84	5.39	0.212
time management	test group	36	4.69	1.49	4.69	0.283	0.473	0.007
	control group	36	4.50	0.81	4.50	0.283
Ask for help	test group	36	5.14	1.09	5.14	0.326	9.953	0.124
	control group	36	4.11	1.62	4.11	0.326
Self-evaluation	test group	36	4.39	1.48	4.39	0.292	0.227	0.003
	control group	36	4.25	0.94	4.25	0.292

***p < .001, **p < .01, *p < .05.

The results show thatthere is a significant difference between environmental factors and help-seeking factors; the average environmental score of the experimental group (M = 5.39, SD = 1.07) was significantly higher than the control group (M = 4.08, SD = 0.84). In terms of seeking helpthe experimental group was better than the control group (M = 5.14, SD = 1.10) (M = 4.11, SD = 1.62).

(3) Collective effectiveness

Raw data for collective effectiveness:

Collective efficacy covariance analysis:

Table 5.Covariance analysis of collective efficacy

variable	Group	N	Mean	SD	After adjustment M	Error	F	h2
collective efficacy	test group	36	5.84	1.11	5.84	0.186	14.243	0.169
	control group	36	4.90	1.01	4.90	0.167

Results: The collective efficacy hypothesis was verified (F = 2.69, p = 0.11 <0.05), followed by analysis of covariance. Table 5 presents the results of the covariance analysis of collective efficacy. The result shows: F = 14.243, p = 0.0001, SD = 1.04.90) was significantly higher than the control group (M = 1.11, SD = 84 < 0.01; There is a significant difference in the collective efficacy of the two groups of patients. The average collective efficacy of the experimental group (M = 5.

(4) Interaction

Interaction raw data

interaction covariance analysis

Table 6. Interaction covariance analysis

variable	Group	N	Mean	SD	After adjustment M	Error	F	h2
interaction	test group	36	5.34	1.64	5.34	0.240	3.408	0.046
	control group	36	4.75	1.08	4.75	0.214

**p < .01.

Results: For the analysis of the interaction questionnaire, this study also used covariance analysis to understand the differences in interaction between the experimental group and the control group. First, the homogeneity test of the regression coefficients within the group shows that F = 2.46, p = 0.12 >0 .05; the results verify the hypothesis and indicate that analysis of covariance can be performed. The average interaction effect of the experimental group (M = 5.34, SD = 1.64) was significantly higher than the control group (M = 4.75, SD = < a i=8>1.08). This also shows that the collective problem-solving promotion mechanism used in this study can indeed have an impact on students' interactions.

Interactive three-factor raw data

Three-factor analysis of covariance

The interaction questionnaire includes three aspects, includinglearner-learner interaction, learner-teacher interaction and learner-content interaction. Therefore, this study also further explored whether these three aspects have independent results and conducted a covariance analysis on these three aspects. The homogeneity results of the regression coefficients in these three aspects show that learner-learner interaction (F = 3.4, p =0.069 >0.05), learner-teacher interaction (F = 0.09, p = 0.76 >0.05) , learner-content interaction (F = 1.89, p = 0.17 > 0.05) all meet the assumption of homogeneity of regression coefficients; therefore, further covariance analysis can be conducted. Table 7 illustrates the results of the interaction of these three aspects.

5. Conclusion

This study proposes a flipped learning method based on a collective problem-solving promotion mechanism. Judging from the experimental results, the students' learning effects are consistent with what was proposed.

In terms of self-regulation, students who adopted the flipped learning method scored significantly higher than students who used the traditional flipped learning method in the two aspects of "environment" and "seeking help." The possible reason is that the competition mechanism encourages students to pay more attention to their academic performance; therefore, they have higher standards for the learning environment so that they can devote themselves to the learning environment. This is consistent with a previous study that showed that the right environment can help learners focus more, thereby improving their learning outcomes. In addition, in order to complete learning tasks, students who use this method are more willing to seek help when encountering learning problems than students who use traditional flipped learning methods, including discussing with peers online, discussing with teachers via email, and sharing at any time own problems and solutions. Peer interactions, including help-seeking behaviors, can be promoted during collaborative learning activities in flipped classrooms. There were no significant differences between the two groups on other aspects of self-regulation, namely intrinsic characteristics. Therefore, we can infer that students' environment and help-seeking perceptions are more susceptible to the collective problem-solving promotion mechanism in the flipped classroom because they are external factors.

In terms of collective efficacy, the collective efficacy of students using the flipped learning method was significantly higher than that of students using the traditional flipped learning method. That is to say, due to the promotion mechanism of collective problem solving, students more actively use group cooperation time to complete learning tasks. They reasoned that collective efficacy could promote students' group commitment to completing learning tasks. As a result, students believed that by working in groups, they were more capable of completing learning tasks and could obtain higher scores; this was the same as previous research results.

In terms of interaction patterns, we found that students using flipped learning methods had different interaction patterns than students using traditional flipped learning methods. Because the facilitation mechanism of group problem solving was combined with IRS, which allowed students to set time limits and competition rules, students showed more interaction and participation in the learning situation. This is consistent with previous research showing that the more engaged students are in class, the better they perform academically.

By further investigating students' discussion content and behavior, we found that the discussion behavior of the two groups of students in class was different. Students who adopted this approach showed more behavioral patterns related to higher cognitive levels and deeper knowledge building than the other group of students. That is, in group learning, there are more higher-level interactive behaviors such as negotiation and comparison.

In contrast, students using traditional flipped learning methods seemed to have more discussion content at lower knowledge building levels, especially content unrelated to the topic. It can be inferred that due to the inter-group competition in the classroom, students are more involved and interactive in group collaboration, and the discussion content is deeper and higher-level. This finding is consistent with the results of previous researchers, who noted that incorporating effective learning strategies into flipped classrooms has the potential to promote higher-order thinking in students.

In summary, this study proposes a collective problem-solving promotion mechanism for flipped learning to promote students' learning performance, collective efficacy, and interaction. It provides a good reference for teachers and researchers to implement effective flipped classrooms, and also points out a new direction for flipped learning research on effective strategies. It can also provide reference for higher education institutions or policy makers in planning and designing courses.

Original article link:

tandfonline.com/doi/abs/10.1080/10494820.2019.1568263https://www.tandfonline.com/doi/abs/10.1080/10494820.2019.1568263