A Comparison of Actual and Preferred Classroom Environments as Perceived by Middle School Students
ABSTRACT
BACKGROUND
A good classroom environment can promote students' learning motivation and affect their academic efficacy and adaptation. This study compares the perceptions of Taiwanese middle school students regarding actual and preferred classroom environments and explores the association with sex and grade level.
METHODS
Data were collected using cross-sectional research design from a national sample of 1932 middle school students. Data of 1897 valid questionnaires from the Chinese Elementary and Middle School Inventory of Classroom Environment were analyzed.
RESULTS
The actual and preferred classroom environments perceived by students differed significantly (p < .001). The overall mean score of actual classroom environment was lower than that of preferred classroom environment. Differences between the actual and preferred environments were greater for girls than for boys (p < .001). Moreover, girls had higher scores than those of boys for both actual (p = .036) and preferred (p < .001) environments. After adjusted for region, seventh- and eighth-grade students had a higher score than that of ninth-grade students on preferred classroom environment (p = .002).
CONCLUSIONS
This study identified a gap between the ideal and actual classroom environments. We suggest that the government, schools, and health education teachers improve classroom environments during school health programs to satisfy students' expectations and thus increase their learning efficacy and overall well-being.
Schools are among the most important learning and living environments for students. Factors such as the school environment, teacher-student relationship, schoolwork, and peer interaction can have a significant influence on young people's mental health; therefore, these factors should not be underestimated.1, 2 Studies have shown that classroom environments are closely related to a number of aspects of student learning, such as student achievements, motivation, and goal achievement.3-9 Good classroom environments can promote students' learning motivation and behavior and can affect students' academic efficacy and adaptation.10-12 A positive classroom environment can not only boost students' self-esteem,3 but also can also facilitate the learning of social skills and stabilize social groups.13 A healthy classroom environments contributes to a relaxed-learning environment, helping students develop self-esteem, confidence, and mutual respect, increasing their participation and sense of belonging.14-16
Students' learning outcomes can be affected by their perception of classroom environments. When teachers and students have an amicable relationship, where students feel comfortable and well-treated, students' social and academic performance can be improved, leading to a more healthy and vibrant class.17, 18 Positive learning environments are also crucial for effective instruction in all content areas including health education. Moreover, it is suggested that health education teachers should provide classroom environments in which students feel comfortable, safe, and supported in order to implement the National Health Education Standard.19
The earliest study of classroom environments investigated the formation of groups of children and social interaction.20 With changing research directions, studies have shifted their attention to teacher leadership and the classroom environment, and the relationship between teachers' verbal behavior and interaction.21 Since the 1970s, studies have focused on students' perception of learning settings; the development of related measurements; the factors that influence classroom environments; and the relationship between the classroom environment, learning achievements, and attitudes. Among these, issues pertinent to “the development of classroom environment instruments” and “the relationship between classroom environments and students' learning achievements” have received more attention.4, 5, 22
To explore the status of classroom environments, previous studies examined personal observations, teachers' perceptions, and students' perceptions.23-27 Because the classroom is an environment shared by teachers and students, student perceptions of classrooms can provide abundant information.28, 29 Consequently, studies using this approach to understand learning environments have increased.30-32
The perception of classroom environments of students by sex and grade level has also drawn significant attention. Most studies have indicated that girls' perceptions of classroom environments are more positive than boys' perceptions.27, 33, 34 However, a number of studies have found no differences between sex.7, 35 For grade level differences, the lower the student's grade level, the more positive perception of classroom environments.7, 35
In the last few decades, studies have focused on students' preference regarding classroom environments, in addition to their perception of actual classroom environments. A number of studies have indicated that students' classroom preferences are positively associated with their learning orientation.36 Studies also have examined the gap between actual and preferred classroom environments, because an increasing gap suggests a higher expectation of better classroom environments.37, 38 Furthermore, Dorman26 claimed that the more positive students' perception of the actual classroom environment was, the smaller the gap between their perception of actual and preferred classroom environments. Previous studies indicated that students have better learning outcomes and more positive learning attitudes when the gap between actual and preferred classroom environments is reduced.39-43 In summary, the greater the congruence between the actual and preferred classroom environments, the better student learning outcomes. Therefore, decreasing the discrepancy between actual and preferred classroom environments, or creating a classroom environment consistent with students' preferences is important for promoting their learning outcomes and achieving educational goals.
In Taiwan, middle school is the final 3 years of compulsory education. The current average class size is 35 students. A homeroom teacher, who is responsible for managing the class and instructing a specific professional subject, is assigned to each class by the school. To date, only a few studies have investigated the classroom environment at middle schools in Taiwan, and most of these studies primarily focused on students' perception of the actual classroom environment for a single subject, such as the science classroom environment.44, 45 A few studies investigating both actual and preferred classroom environments selected university and high school students as participants and primarily focused on the subject of Earth science.40, 46 Under these circumstances, the issues regarding middle school students' perceptions of actual and preferred classroom environments, and the disparity between these 2 perceptions, remains unclear. Moreover, whether students' demographic characteristics (ie, sex and grade level) have an impact on their perceptions of actual and preferred classroom environments is worth investigating. Therefore, this study investigated and compared Taiwanese middle school students' perceptions of actual and preferred classroom environments, and further explored whether students of different sex and in different grade levels varied in these 2 perceptions.
METHODS
Research Design and Participants
Data were collected using a cross-sectional survey research design. Participants were drawn from a national population of 950,855 public middle school seventh- to ninth-grade students in Taiwan. After reviewing relevant research,27, 44 we selected approximately 2000 students as the research sample. The samples were recruited from schools in 6 regions; northern, central, southern, and eastern Taiwan, as well as Taipei City and Kaohsiung City, because the proportion of students differs significantly in these regions (5-30%), and classroom cultures and resources also differ. Multistage sampling was adopted. In the first step, stratified proportional sampling was used with regions as the strata. The number of samples required in the 6 regions was determined based on their student proportions. According to these values, we determined that 60 classes and 20 schools were required. Subsequently, class served as sampling unit, and 1 class was selected randomly from each grade of every school, resulting in national samples of 1932 students. Among the students selected, 14 students refused to participate and 21 did not complete the questionnaire. Therefore, the effective sample was 1897, yielding an overall response rate of 98.2%.
Instruments
Data were collected using the self-report version of the Chinese Elementary and Middle School Inventory of Classroom Environment (C-ICE) questionnaire. Lai et al47 developed the C-ICE through a forward/backward translation, and its use was authorized by Sinclair and Fraser.27 Then, seventh- to ninth-grade students were recruited to complete the C-ICE, and a psychometric evaluation was conducted. The C-ICE includes 2 sections, actual classroom environment and preferred classroom environment, and each section comprises 19 parallel items. The item is scored from 1 (seldom) to 3 (frequently) or 3 (seldom) to 1 (frequently) depending on its positive and negative statements. The higher a participant's score, the more positive the perception.
The psychometric evaluation of the C-ICE included a structure and criterion-related validity, and internal and test-retest reliability. On the basis of factor analysis, each section of the C-ICE was divided into 4 subscales: Cooperation (3 items), Teacher Empathy (4 items), Involvement (5 items), and Task Orientation (7 items). The definition and content of each subscale is provided in Table 1. The internal consistency coefficient, Cronbach's α, for the actual and preferred classroom environment scales was .86, and .92, respectively. The correlation coefficients of the test-retest reliability, with an interval of 2 to 3 weeks, were .68 and .72 for the actual and preferred environment scales, respectively.
| Example | Cronbach's α | ||||
|---|---|---|---|---|---|
| Scale | Description | Actual | Preferred | Actual | Preferred |
| Cooperation | The extent to which students get along with each other during class and activities | I make friends with students in class | I want to make friends with students in class | .66 | .72 |
| Task orientation | The extent to which students coordinate courses or class activities | I understand how to complete my work in class | I want to understand how to complete my work in class | .69 | .78 |
| Involvement | The extent to which students participate in class activities and discussions | I discuss ideas in class | I want to discuss ideas in class | .73 | .81 |
| Teacher empathy | The extent to which the teacher helps, encourages, and supports students | My homeroom teacher likes me | I want my homeroom teacher to like me | .87 | .89 |
| C-ICE | .86 | .92 | |||
- C-ICE, Chinese Elementary and Middle School Inventory of Classroom Environment.
- Actual, actual form of C-ICE; Preferred, preferred form of C-ICE.
Procedure
The questionnaire was anonymous and designed to be completed by the entire class during a group administration. Two days before the survey, parental consent forms were brought home by students. During the survey, the investigators first introduced themselves to the class before explaining the research purpose and how to complete the questionnaire. Then, the questionnaire was distributed to the students with signed parental consent. If a participant had any questions when answering the questionnaire, the investigator was available to help. After students had completed and returned the questionnaire, the investigator checked to ensure that all the items had been answered.
Data Analysis
Statistical analysis was conducted using SPSS software (version 17.0; SPSS Inc., Chicago, IL) for Windows. A goodness-of-fit test was conducted to examine the sample and population distribution regarding region, sex, and grade. To understand variable distribution, the percentage, the mean, and the standard deviation were calculated. Paired t-test was conducted to examine the mean differences in actual and preferred classroom environments. Student's t-test was performed to explore sex difference of perceptions of the classroom environments. An analysis of covariance (ANCOVA) was performed to assess grade-level differences in perceptions of the classroom environment adjusted for region. If the significance level was reached for ANCOVA, we further conducted post hoc tests for pair-wise comparison. All tests use α = .05 as the level of statistical significance.48
RESULTS
Participant Characteristics
The distribution of participants' sex (χ2 = 1.71, p = .191) and region (χ2 = 5.30, p = .381) did not significantly differ from the population. However, the distribution of participants' grade levels was significantly different from that of the population (χ2 = 16.18, p < .001). Furthermore, the percentage of boys and girls who participated in this study was similar (49.5% and 50.5%, respectively). Eighth-grade students accounted for the largest number of participants (36.4%), and ninth-grade students represented the lowest percentage (29.3%). The proportions of participants from the northern (29.4%), southern (26.4%), and central (26.7%) regions were comparatively higher, and the proportions of students from the other three regions were below 10%.
Actual and Preferred Classroom Environments and the Differences Between Them
Table 2 shows that the overall mean score for the actual classroom environment was 2.27 ± 0.34 (maximum = 3), in which the mean scores of the 4 subscales all exceeded the median value of 2, with the highest mean score for the Cooperation subscale, and the Involvement subscale scored the lowest. The overall mean score for the preferred classroom environment was 2.39 ± 0.44, in which the mean scores of the 4 subscales were between 2.12 and 2.61, with the highest mean score for the Cooperation subscale, and the Involvement subscale scored the lowest.
| Variable | Mean | Standard Deviation | t | p -Value |
|---|---|---|---|---|
| Overall scale | −14.10 | <.001*** | ||
| Actual | 2.27 | 0.34 | ||
| Preferred | 2.39 | 0.44 | ||
| Cooperation | −14.04 | <.001*** | ||
| Actual | 2.44 | 0.47 | ||
| Preferred | 2.61 | 0.47 | ||
| Task orientation | −8.21 | <.001*** | ||
| Actual | 2.41 | 0.42 | ||
| Preferred | 2.50 | 0.52 | ||
| Involvement | −10.91 | <.001*** | ||
| Actual | 2.00 | 0.49 | ||
| Preferred | 2.12 | 0.55 | ||
| Teacher empathy | −9.71 | <.001*** | ||
| Actual | 2.30 | 0.51 | ||
| Preferred | 2.42 | 0.53 |
- *** p < .001.
The differences between actual and preferred classroom environments are displayed in Table 2. The results show that the overall mean scores of actual and preferred classroom environments differed significantly (p < .001). The overall score for the preferred environment was higher than that of the actual environment. In addition, significant differences were observed between the corresponding subscales in the actual and the preferred classroom environments, including for Cooperation, Task Orientation, Involvement, and Teacher Empathy (p < .001).
Associations Among Sex, Grade Level and Perceptions of Actual and Preferred Classroom Environments
For the actual classroom environment, significant sex difference were found for the overall scale (p = .036) (Table 3), indicating that girls had a higher mean score than males. Furthermore, sex differences were found in the subscales where girls' mean score was higher than that of boys for the Task Orientation (p < .001) and Teacher Empathy (p = .007) subscales. After adjusted for region, no significant grade-level differences were observed for the overall actual classroom environment scale, but a significant difference was observed for the Cooperation subscale (p = .045). The post hoc results indicated that the seventh- and eighth-grade students had a higher mean score than that of the ninth-grade students.
| Sex | Grade Level | ||||
|---|---|---|---|---|---|
| Boys (N = 937) | Girls (N = 956) | Seventh (N = 651) | Eighth (N = 691) | Ninth (N = 555) | |
| Overall scale | |||||
| Mean ± SD | 2.25 ± 0.36 | 2.28 ± 0.33 | 2.28 ± 0.36 | 2.28 ± 0.33 | 2.24 ± 0.34 |
| t/F | −2.09 | 2.07† | |||
| p -value | .036* | .127 | |||
| Cooperation | |||||
| Mean ± SD | 2.44 ± 0.46 | 2.44 ± 0.48 | 2.46 ± 0.47 | 2.46 ± 0.47 | 2.40 ± 0.46 |
| t/ F | −0.15 | 3.11† | |||
| p -value | .884 | .045* | |||
| Post hoc | Seventh, eighth > ninth | ||||
| Task orientation | |||||
| Mean ± SD | 2.36 ± 0.44 | 2.46 ± 0.40 | 2.43 ± 0.42 | 2.41 ± 0.42 | 2.40 ± 0.43 |
| t/F | −5.19 | 1.17† | |||
| p -value | <.001*** | .312 | |||
| Involvement | |||||
| Mean ± SD | 2.03 ± 0.50 | 1.98 ± 0.47 | 2.02 ± 0.50 | 2.01 ± 0.49 | 1.99 ± 0.47 |
| t/F | 1.98 | 0.40† | |||
| p -value | .053 | .670 | |||
| Teacher empathy | |||||
| Mean ± SD | 2.27 ± 0.52 | 2.33 ± 0.50 | 2.31 ± 0.51 | 2.32 ± 0.48 | 2.27 ± 0.53 |
| t/F | −2.68 | 1.59† | |||
| p -value | .007** | .205 | |||
- * p < .05;
- ** p < .01;
- *** p < .001.
- † Adjusted for region.
Table 4 shows the results of the preferred classroom environment. Girls had higher mean scores than boys in the preferred classroom environment and its 4 subscales (p < .05). Similarly, the overall scores for the preferred classroom environment scale differed significantly by grade level (p = .002) after adjusted for region. The post hoc results indicated that the seventh- and eighth-grade students had a higher mean score than that of the ninth-grade students. Additionally, significant grade-level differences were observed for the 4 subscales (p < .05). Among these, the seventh- and eighth-grade students scored higher than the ninth-grade students did for all 4 subscales; the seventh-grade students scored higher than the eighth-grade students did for the Task Orientation and Teacher Empathy subscales.
| Sex | Grade Level | ||||
|---|---|---|---|---|---|
| Boys (N = 937) | Girls (N = 956) | Seventh (N = 651) | Eighth (N = 691) | Ninth (N = 555) | |
| Overall scale | |||||
| Mean ± SD | 2.33 ± 0.46 | 2.44 ± 0.41 | 2.42 ± 0.44 | 2.39 ± 0.44 | 2.33 ± 0.43 |
| t/F | −5.46 | 6.08† | |||
| p -value | <.001*** | .002** | |||
| Post hoc | Seventh, eighth > ninth | ||||
| Cooperation | |||||
| Mean ± SD | 2.56 ± 0.50 | 2.65 ± 0.43 | 2.62 ± 0.49 | 2.64 ± 0.44 | 2.55 ± 0.49 |
| t/ F | −4.11 | 4.65† | |||
| p -value | <.001*** | .010* | |||
| Post hoc | Seventh, Eighth > ninth | ||||
| Task orientation | |||||
| Mean ± SD | 2.43 ± 0.55 | 2.55 ± 0.48 | 2.55 ± 0.50 | 2.49 ± 0.52 | 2.44 ± 0.52 |
| t/ F | −5.03 | 5.85† | |||
| p -value | <.001*** | .003** | |||
| Post hoc |
Seventh > eigh Seventh, eigh > ninth |
||||
| Involvement | |||||
| Mean ± SD | 2.08 ± 0.57 | 2.14 ± 0.53 | 2.14 ± 0.56 | 2.14 ± 0.56 | 2.06 ± 0.52 |
| t/ F | −2.45 | 3.06† | |||
| p -value | .015* | .047* | |||
| Post hoc | Seventh, eighth > ninth | ||||
| Teacher empathy | |||||
| Mean ± SD | 2.34 ± 0.55 | 2.49 ± 0.50 | 2.47 ± 0.52 | 2.41 ± 0.53 | 2.36 ± 0.54 |
| t/ F | −6.15 | 5.89† | |||
| p -value | <.001*** | .003** | |||
| Post hoc |
Seventh > eighth Seventh, eighth > ninth |
||||
- * p < .05;
- ** p < .01;
- *** p < .001.
- † Adjusted for region.
Effects of Sex and Grade-Level on the Gap Between Actual and Preferred Classroom Environments
The sex and grade-level differences across scores for the actual and preferred classroom environments are shown in Table 5. The overall gap between actual and preferred classroom environments differed significantly by sex (p < .001). Also, the gap between the actual and preferred environments was larger for girls than for boys in the Cooperation, Involvement, and Teacher Empathy subscales (p < .05). Similarly, the overall gap between the actual and the preferred classroom environments differed significantly by grade level (p = .043) after adjusted for region. The post hoc results indicated that the gap between the seventh-grade students' perception of actual and preferred classroom environments was greater than that of ninth-grade students. Furthermore, significant differences were found by grade level for the Task Orientation and Teacher Empathy subscales (p < .05, respectively). According to the post hoc results, the gap for seventh- and eighth-grade students in the Task Orientation subscale was greater than that for ninth-grade students; the gap for seventh-grade students in the Teacher Empathy subscale was greater than that for eighth-grade students.
| Sex | Grade Level | ||||
|---|---|---|---|---|---|
| Boys (N = 937) | Girls (N = 956) | Seventh (N = 651) | Eighth (N = 691) | Ninth (N = 555) | |
| Overall scale | |||||
| Mean ± SD | 0.08 ± 0.38 | 0.15 ± 0.34 | 0.14 ± 0.37 | 0.12 ± 0.34 | 0.09 ± 0.38 |
| t/F | −4.64 | 3.15† | |||
| p -value | <.001*** | .043* | |||
| Post hoc | Seventh > ninth | ||||
| Cooperation | |||||
| Mean ± SD | 0.13 ± 0.51 | 0.21 ± 0.51 | 0.16 ± 0.53 | 0.18 ± 0.50 | 0.16 ± 0.51 |
| t/F | −3.63 | 0.25† | |||
| p -value | <.001*** | .778 | |||
| Task orientation | |||||
| Mean ± SD | 0.07 ± 0.45 | 0.09 ± 0.43 | 0.11 ± 0.44 | 0.08 ± 0.44 | 0.05 ± 0.44 |
| t/F | −0.94 | 3.47† | |||
| p -value | .348 | .031* | |||
| Post hoc | Seventh, eighth > ninth | ||||
| Involvement | |||||
| Mean ± SD | 0.06 ± 0.44 | 0.16 ± 0.43 | 0.12 ± 0.45 | 0.13 ± 0.44 | 0.07 ± 0.43 |
| t/F | −5.25 | 2.44† | |||
| p -value | <.001*** | .087 | |||
| Teacher empathy | |||||
| Mean ± SD | 0.08 ± 0.57 | 0.16 ± 0.52 | 0.17 ± 0.55 | 0.10 ± 0.50 | 0.10 ± 0.59 |
| t/F | −3.47 | 3.21† | |||
| p-value | .001** | .041* | |||
| Post hoc | Seventh > eighth | ||||
- * p < .05;
- ** p < .01;
- *** p < .001.
- † Adjusted for region.
DISCUSSION
Results of this study indicate that participants' score for the actual classroom environment exceed the median value of 2, showing a positive inclination. This finding corresponds to those of previous studies on middle school students in Taiwan, Australia, and the United States.27, 44 Moreover, we found that participants scored the highest for the Cooperation subscale and the lowest for the Involvement subscale. This finding is consistent with that of Sinclair and Fraser.27 They also found that students scored higher for the Cooperation, Task Orientation, and Teacher Empathy subscales, and scored lower for the Involvement subscale. Additionally, Aldridge and Fraser44 recruited Taiwan and Western Australia students to respond to the “What Is Happening in This Classroom (WIHIC)” scale. Their results showed that students in both countries scored similarly for the Student Cohesiveness subscale. However, Australian students scored higher than Taiwanese students for the Involvement, Task-Orientation, Cooperation, and Equity subscales. Generally, studies have shown that Taiwanese middle school students have lower involvement, task-orientation, and cooperation in the science learning environment.44
The culture and education systems in Taiwan differ from those in other countries, such as the United States and Australia. Thus, as expected, students' perceptions of their classroom environments differ from that of students from other countries. Additionally, some distinctive features of education in Taiwan can lead to lower student involvement, such as a large number of middle students per class, low teacher-student ratio, high competitiveness among students, and the focus on student academic performance.49 These features differ significantly from features that encourage student involvement in a classroom environment, such as a small number of students in a class, welcoming classroom environment, student-centered classroom, and nonthreatening assessments.50, 51 Furthermore, teacher-centered classrooms where students are passive learners and teachers instill knowledge in students and dominate their learning are commonly seen in Asia.52, 53 However, recent studies have shown that Taiwanese students do not prefer a student-centered teaching approach.46, 54 These features may have induced the lower involvement from Taiwanese students compared with that of other students from other country.
Moreover, results of this study show that the mean score of the preferred classroom environment exceeded the median, which suggests that participants prefer positive learning environments. This finding supports previous studies using different scales for classroom environments, confirming that students expect a better classroom environment.39 Also, our results indicate that students' scores for all 4 subscales exceeded the median, which suggests that students prefer a classroom environment with better interaction between students, more schoolwork, more involvement and discussions in class, and more teacher support. Previous researchers have highlighted that changing classroom environments, according to student preferences, can improve learning outcomes.41 Similarly, Chan et al39 also claimed that students feel safe, welcome, and develop a sense of belonging in their preferred classroom environment where teachers and students can work cooperatively and induce a diverse range of interests rather than competitive activities. This can contribute to more successful learning for students. Therefore, schools and teachers should consider how to satisfy students' expectations and create supportive classroom environments that facilitate student cooperation and participation.
This study also found that different sex significantly differed in their perception of actual classroom environments, with girls scoring higher than boys did. This finding is consistent with numerous studies suggesting that girls' perceived classroom environment is better than that of boys, particularly for the Task Orientation and Teacher Empathy subscales.27, 33, 34 However, no sex difference was found in the Cooperation subscale in this study. This contradicts the findings of Lee et al7 who concluded that girls perceived more cooperation than boys did. Additionally, the results showed that girls scored higher than boys did for the overall scale and the 4 subscales in the preferred classroom environment. This suggests that girls may have a stronger desire than boys do to have a preferred classroom environment. Presumably, this could be because of the impact of traditional sex socialization where girls value social identity and value the importance of having good interaction with people around them is important.55-57 This leads girls to be more dedicated to maintaining and establishing good relationships with others.
Contrary to some investigations, this study did not observe differences across different grade levels for the actual classroom environment.7, 23 A possible explanation for this result is the lower number of ninth-grade participants. However, this study found that students in different grade levels scored differently for the Cooperation subscale of the actual classroom environment. This might be because students in different grade levels experience academic competition differently. Therefore, their perspective on the importance of cooperation in learning activities varies. The trade-off between students' academic competition and cooperation is worth further investigation. However, this study found differences in different grade levels for the overall preferred classroom environment. The results were similar to that of previous studies.58 Moreover, the results show that the seventh- and eighth-grade students scored higher than the ninth-grade students for the preferred classroom environment scale and its 4 subscales. This indicates that the seventh- and eighth-grade students may have higher expectations for ideal classroom environment, particularly in their relationships with classmates, coordination and involvement with curricular activities, and teacher care. One explanation is that the ninth-grade students who prepare for high school entrance examinations appeared to have less expectations of an ideal learning environment. Additionally, apathy problem and nonchalant attitude of secondary schools and students might be another explanation for less expectations of an ideal learning environment among the ninth-grade students.59, 60 One longitudinal study pointed out that middle school students feel bored with schoolwork.61 Studies also showed that disengagement and avoidance related to schoolwork varied and increased with grade levels.59, 62 The other explanation is that the seventh-grade students might be craving for more teacher supportive learning environment which is similar to their elementary schools.63 This finding implies that teachers should establish a positive preferred classroom environment as soon as students transit to middle schools.
Finally, results of this study show that participants' score for the preferred classroom environment was higher than that for the actual classroom environment. This finding is consistent with findings from numerous studies conducted in many countries.26, 39, 64-66 Furthermore, the scores for the 4 subscales in the preferred classroom environment were all greater than the corresponding subscale scores in the actual classroom environment. This finding suggests that Taiwanese middle school students' perception of the actual classroom environment is below their expectations. Previous studies have shown that teachers could improve learning environment successfully by using data of the students' perception of the actual and preferred classroom environment.67 Therefore, how to establish a classroom environment that caters to student preferences can be an important future objective for Taiwanese middle school teachers and education system. The national education policy should focus on classroom environments and teachers' abilities. Educational authorities should organize practical workshops to enhance teachers' related abilities, and encourage teachers to devote attention to students' expectations of classroom environments and to design and promote management plans.
Limitations
All dependent variables were measured using Likert-type scales, and they were vague quantifiers. The difference in vague quantifiers is not identical to variance in real life.68 In addition, the significant differences in sex and grade level were found with relatively small values. Therefore, one should be cautious when applying the findings to practice.
There are other limitations of this. Except for grade level, the distribution of the participants regarding region and sex was similar to that of the population. The ninth-grade students had a relatively lower response rate. This difference may have resulted from the lower rate of parental consent for the ninth-grade students because many were preparing for senior high school entrance examinations. Therefore, any assumptions regarding grade-level differences in the preferred and actual classroom environments should be conducted with caution.
In addition, most of the correlations between grade level and the 3 dependent variables were significant. However, differences in actual classroom environments and several subscales were not significant; this may have been caused by the limited sample size.
Moreover, this study recruited only students. Thus, only the students' perceptions of classroom environments were explored whereas the teachers' opinions of the classroom environment and their willingness to improve were not examined. Future studies should include both teachers and students and investigate the similarities and differences between their needs for learning environments. By finding a balance between students and teachers, learning environments can be improved.
Conclusion
This study found that the gap between actual and preferred classroom environments was significant, and the actual and preferred classroom environments were influenced by sex and grade level. Given these findings, developing and implementing classroom environment projects tailored to students' expectations, reduce the gap, and enhance students' academic performance as students start their middle school life is one key task for middle schools and teachers.
IMPLICATIONS FOR SCHOOL HEALTH
Classroom environments are closely associated with students' school life adaptation and learning outcomes. When the gap between the actual and preferred classroom environments is reduced, students can be immersed in a friendly, cooperative, participating, and caring learning environment, which leads to better physical and mental development, and better learning outcomes.26 Positive learning environments are also critical for the implementation of effective health education instruction and the National Health Education Standard. Results of this study revealed differences between the actual and preferred classrooms. This suggests that students generally expect a better learning environment. Moreover, this study shows that the seventh-grade students expect to have a more ideal learning environment. It implies that it is a vital period for building positive learning environments and providing support to fulfill students' psychological needs when students enroll in the middle school. School administrators should put a learning environment improving plan into practice in the new semester and encourage all teachers to get involved. Homeroom teachers should hold class meetings and discuss with students about the vision of ideal classroom environments. Additionally, homeroom teachers can create a classroom code of conduct, teach positive actions, and reinforce respectful behaviors. To enhance teachers' empathy and positive teacher-student relationships, teachers should express enthusiastic, warm, caring, fair, and positive attitudes toward each student, and arrange personal interviews to completely understand students' learning problems and concerns. Health education teachers can create a safe and supporting classroom environment by developing and practicing learning activities which facilitate positive peer interaction and reciprocal relationship building. For example, incorporating cooperative learning into a specific health issue can reinforce collaboration among group members. Finally, school health professionals can play an important role in developing and implementing in-service training programs to promote teachers' classroom management abilities.
Human Subjects Approval Statement
This study was reviewed by the Institute Review Board of Taipei Medical University. Students were only included in the study after we received the consents of parents and students.
