**1. Introduction**

Since its emergence, the new coronavirus has had various adverse effects on the mental health of the general public. In the first year of COVID-19, the prevalence of depression and anxiety disorders is estimated to have increased by 25% [1,2]. Studies [3] show that people have had varied degrees of psychological problems throughout the outbreak. Stress, anxiety, and depression values remained high after two weeks and did not decrease over time. This was exacerbated by the isolation and confinement caused by the epidemic, with symptoms such as mood disorders, depression, stress, poor mood, irritability, insomnia, and post-traumatic stress disorder [4]. Of all populations, students are among the most prone to have severe psychological problems [5]. A comparative study found that children who experienced isolation had post-traumatic stress values four times higher than those who did not [6]. In an Italian study [7], late bedtimes and late wakeups were particularly common among the student population, and sleep quality was also affected during the period of isolation, with 27.8% reporting depression symptoms and 34.3% showing anxiety symptoms. In a pre-and post-closure survey of Chinese university students, the mean PANAS-NA (negative affect) scale score fell from 2.38 (0.79) to 2.24 (0.80), and the mean anxiety-depression score on the PHQ-4 scale changed from 0.95 (0.65) to 0.76 (0.61), with significant reductions in both values. This indicates that there is a significant increase in anxiety and depression symptoms among students, and the negative effects of school

**Citation:** Tao, W.; Wu, Y.; Li, W.; Liu, F. Influence of Classroom Colour Environment on College Students' Emotions during Campus Lockdown in the COVID-19 Post-Pandemic Era—A Case Study in Harbin, China. *Buildings* **2022**, *12*, 1873. https:// doi.org/10.3390/buildings12111873

Academic Editor: Diego Pablo Ruiz Padillo

Received: 29 August 2022 Accepted: 26 October 2022 Published: 3 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

closure are becoming more prominent [8]. The epidemic school closure greatly harms the mental health of college students and requires urgent interventions.

China has been implementing the "Dynamic Zero-COVID" approach since the pandemic outbreak in 2020. In the context of that, universities in China have also been undergoing a dynamic lockdown for nearly three years. Because of numerous elements like Chinese universities offering practically every student housing and the Chinese population figure being relatively significant, Chinese universities also have a higher volume of students living on campus than foreign universities. Therefore, if no protection measure is conducted, the virus is more likely to transmit on the Chinese campus and cause a disastrous outcome. Based on all the facts and according to policies in districts, the university operates in closure, resulting in students being forced to stay in campus walls. During the closure, students reside in dormitories and can wander around inside the university, conducting ordinary tasks including studying indoors and playing sports outside, etc. If needed, medical resources are supplied. This action is temporary, and the campus will be reopened once the external epidemic is under control. However, we can observe complaints from students on social media like Weibo, writing about anxiety from long-time closure. Some mention that their psychological condition gets worse because of the feeling of being restricted on campus. College students' mental health is affected by campus lockdown. However, there is a lack of post-pandemic closure studies with Chinese characteristics. The current studies globally on epidemic closure mostly focus on two contexts—home isolation and confinement [4]—whereas there is a lack of discussion on campus lockdown when students are required to stay in university. In addition, most of the studies focus on the early stage of the new corona outbreak [3,5], and there is a lack of studies on the normalization of closure in the post-epidemic period. Furthermore, studies focusing on college students' mental health during post-pandemic college lockdown are also relatively few; the existing literature mostly analyses the changes in the psychological state of adolescents [9,10] during school closure or isolation. Some studies have focused on the psychological changes of college students [8], but also failed to suggest creative strategies of mitigation.

Existing literature indicates that colour can relieve pressure to some extent. Colour psychology is a branch of psychological science that believes that colour has various psychological and behavioural effects [11]. Some psychologists [12] have shown that 83% of the information humans obtain is from visual sources, and that colour predominates in this visual information. Extensive research discusses the psychological, cognitive, physiological, and behavioural effects of colour [13,14]. Some studies indicate the psychological impact of colour in terms of dynamism, size and quantity, and warmth and coolness [15,16]. Since each colour has its own wavelength and frequency, when the body absorbs its specific energy, it might change the original energy in the body. Therefore, colour can be used for the treatment of physical illnesses and psychological problems [17], which has led to the concept of colour healing. Colour therapy, derived from colour psychology, is a technique of psychotherapy that promotes recovery by allowing the patient to see and feel a colourful environment that causes stimulation of the brain and emotions [18]. Colour has been found to have features that affect the patients' physiological activities, emotions in daily life, cognitive processing, and other changes in mental activity [19]. In addition to the use of colour in the medical environment, the use of colour as an adjuvant to therapy to relieve the psychological distress of patients has been increasingly recognized and applied in existing research [20]. It has been shown [17] that colour therapy can have the same or a similar effect on any group of people, regardless of the cultural background. It is known that colour preference could be influenced by differences in age, sex, and geographical region. Additionally, factor analysis and cluster analysis indicated some relation between colour preference and the subjects' lifestyles [21]. For instance, Great Britain has a strong preference for G categories and a warm-greyish colour image is preferred. Italy has a preference for R and Y categories and a warm-clear image is preferred [22]. However, Chinese people have specific colour preferences. For example, black on red signifies happiness to Chinese people, and therefore the colour combination is commonly used for

wedding invitations [23]. Red is not only consistently associated with "active," "hot", and "vibrant", but it also conveys additional meaning ("pleasant") in China [23]. There are also some scholars who point out that the "red preference" phenomenon is observed in Chinese adults. Light colours are preferred the most in terms of chroma-lightness level [24]. Based on all the facts, colour therapy has great potential in reality. However, the current application of colour healing in China is limited. Although there are now discussions on the application of colour psychology for campus space design [25], they are failing to incorporate the current state of college student's mental health and failing to apply various theories of colour healing. Therefore, we must consider the possibility of applying colour therapy in the teaching space to alleviate the anxiety and depression of college students. If the environmental colours of interior teaching spaces can be used to reduce anxiety and depression values, it will considerably enhance the mental health of college students.

To assess students' psychological condition, the HAD scale is used as the measurement tool. The Hospital Anxiety and Depression Scale (HAD) was created by Zigmond and Snaith in 1983 to screen for anxiety and depression in general hospital patients. The scale consists of two separate scales. One is the Hospital Anxiety Scale (HADA) and the other is the Hospital Depression Scale (HADD). It has been translated into several national versions and is widely used in medical assessment. There are studies of national versions of the scale, such as in Spanish [26], Chinese [27], Norwegian [28], and Arabic [29], as well as studies of applicable populations, such as patients with fibromyalgia syndrome [30], tinnitus [31], office workers [32], the elderly [27], and patients with oral burning syndrome [33]. In a Spanish study [30], the subscales "anxiety" and "depression" were evaluated separately, and both scales were found to be highly reliable and accurate (HADA = 0.80, HADD = 0.85). Some studies have found that the HAD scale has 80% sensitivity and 90% specificity, considering it a good screening tool for anxiety and depression in older adults in Cantonesespeaking areas [27]. It has also been suggested that the HAD scale is more useful in the assessment of depression [31]. A Norwegian study [28] revealed the high internal consistency of the scale with a substantial sample size (65,648 participants). In summary, most of the studies corroborate the scientific validity, reliability, and validity of the HAD scale and therefore support the application of the scale in the assessment of anxiety and depression in various domains. Yet there are few cases of applying the HAD scale to research in China, more focus being on particular patients [34,35] and application in the field of education being neglected. There are no examples of assessing students' mental health. Therefore, it is practical and feasible in this paper to apply the HAD scale to assess anxiety and depression among college students.

The research specifically focuses on college students who must stay inside campus because of the pandemic prevention policy. The group's features are quite different from those who can only stay in the dormitory or those home commuting subjects. Therefore, this study has its speciality in geography, timing, groups, and so on. This study will supplement the gap of existing research. Based on the mental health problems of college students in the post-epidemic school closure normalization, using the HAD scale to assess the relevant indicators, we lead to conclusions of colour healing to provide a reference for subsequent space design and psychotherapy. In conclusion, this study aims to provide references and suggestions for the development of campus teaching space environment design in the post-epidemic era, and it is also an innovative attempt to intervene in the mental health of college students from the perspective of colour healing. The following hypotheses are proposed and tested:

**H1.** *During the closure of colleges and universities due to the epidemic, there is a high prevalence of anxiety and depression in the college student population.*

**H2.** *Neutral, warm, and cold teaching spaces and teaching spaces with adjusted lightness shifts have different effects on college students' emotions.*

**H3.** *There is a significant difference between different anxious and depressed groups in judging the effect of teaching space on mood.*

**H4.** *There are significant differences between demographic characteristics in judging the effect of instructional space on mood.*

#### **2. Methods**

In this study, a questionnaire was distributed and filled out through the "Questionnaire Star" platform to collect the subjects' emotional evaluation of the teaching space with different colour characteristics. The research idea is shown in Figure 1.

#### **Figure 1.** Research scheme.

#### *2.1. Questionnaire Setting*

The questionnaire (Appendix A) for this investigation consisted of three parts. Part I: The subjects were asked about their demographic information, including gender, education, and major. Part II: The subjects were tested on the Hospital Anxiety and Depression Scale (HAD scale). The test contains 14 questions (Table 1), and subjects make choices based on their past week. From the outcomes we obtained the subject's level of anxiety and depression. The HAD scale consists of two subscales, anxiety and depression, for anxiety (A) and depression (D), each with 7 questions. Each item is assessed on a 4-point scale, with single-sign ratings summing to anxiety ratings and double-sign ratings summing to depression ratings. A single scale score of 0–7 indicates no depression or anxiety, a total score of 8–10 indicates possible or "borderline" anxiety and depression, and a total score of 11–20 indicates possible significant anxiety or depression.

Part III: Conducting the observation of virtual teaching spaces with different colour differences was carried out. In this experiment, two different teaching spaces were used as prototypes. The initial model was built with Revit 2021 then rendered and post-adjusted with Lumion 11. According to the variation of hue and lightness, 14 different virtual spaces are constructed (Figure 2). Immediately after the observation of each set of virtual teaching spaces, the subjects filled out a questionnaire on the level of pleasure, relaxation, and mental attention for the scene to obtain their subjective feelings about the pictures. The chromaticity analysis was conducted to explore the effect of neutral, warm, and cold classrooms on human emotions. The brightness analysis was conducted to investigate the effect of warm and cold classrooms on human emotions at both high and low brightness levels. The reference data is the mean and standard deviation of the questionnaire scores. In the colour and brightness selection section, three pairs of two-level adjectives, "pleasant/unpleasant", "relaxed/unrelaxed" and "focused/unfocused", were used to evaluate different colour

teaching spaces. The standard is a 5-point semantic difference, using a scoring system from 1 to 5. The lower the score, the more negative is the emotion.

**Table 1.** HAD scale.


#### *2.2. Participants*

One hundred and ten participants were recruited to fill out the questionnaire through the Questionnaire Star platform in April–May 2022. During the period from 11 April to 15 May, most of the subjects were quarantined on campus due to the epidemic closure and were unable to enter or leave the campus freely.

A total of 110 valid questionnaires were returned in this study, with an effective rate of 100%. The numerical characteristics of the demographic variables can be seen according to the analysis results in Table 2, which reflect the distribution of the respondents in this survey and where the mean value represents the trend among them and the standard deviation represents the fluctuation. According to the results of the frequency analysis of each variable, it can be seen that the distribution meets the requirements of the sample survey. For example, among the gender survey results, the proportion of males is 65.5%, and the proportion of females is 34.5%. This shows that the results of this survey focus on male colour preference. In terms of academic distribution, the largest category is undergraduates, including the highest proportion of junior students. In terms of professional distribution, the highest proportion is engineering students, up to 73.6%, indicating that the subjects are mainly science and technology students.

**Figure 2.** Virtual environment modelling of teaching spaces. Note: This computer model is built based on actual classrooms in a university in Harbin, China. The size, furniture, and material are in accord with reality. There are two types of classrooms. One is small and could contain no more than 20 students. The other is large and could hold no more than 200 students. The pictures above are taken from two angles which is of both eyes' perspective, aiming to offer subjects a more immersive experience. Since most Chinese classrooms are decorated with coating materials/paint (Figure 3), we built models without considering the materials' influence. Because wall colour takes up the most percentage of classroom colour, this time we only picked wall colour as the variation. The wall is changed from neutral colour to warm colour and cold colour. Then the lightness of different colours is altered.

**Figure 3.** Teaching spaces in China (online) [36–42].



### *2.3. Data Analysis*

The data analysis software used for the study was SPSS 27. The reliability validity of the dependent variables was first examined, and correlation tests were used to assess whether there was a relationship between the dependent variables, after which the mean and standard deviation of the data were calculated. A repeated measures ANOVA was used to assess the effect of differences in the colour of the teaching space environment on participants' emotions in that context. A one-way ANOVA was used to test whether there were significant differences in the emotional perceptions of the teaching environment space between different anxious and depressed groups. One-way ANOVA was performed afterwards to test whether there were significant differences in the effects of education and major on the participants' emotions; independent samples *t*-test was used to test whether there were significant differences in the effects of gender on the participants' emotions in the difference of colour in the teaching space. The data obtained were presented in graphical or tabular form.

#### *2.4. Reliability Validity Test*

SPSS 27 was used to implement the process of reliability and validity analysis. First, we conducted reliability statistics on 14 HAD scale items and 21 questionnaire items

respectively. It was found that their standard reliability coefficients were 0.907 and 0.925, which were very close to 1, meaning that the reliability was very high. Then, we conducted reliability statistics on all the questions in the questionnaire. According to the results of the reliability analysis of the overall scale, the Cronbach α coefficient based on standardized items was 0.837. It shows that the analysis results have high reliability. The validity analysis of the questionnaire was carried out by the test process through the approach of exploratory factor analysis in SPSS 27. According to the results of the exploratory factor analysis, the coefficient of the KMO test was 0.823, and the range of the coefficient of the KMO test was between 0 and 1. The closer to 1, the better is the validity of the questionnaire. According to the significance of the sphericity test, it can also be concluded that the significance of this test is infinitely close to 0. The significance is significantly less than 0.005, and the original hypothesis is rejected, indicating the questionnaire has good validity.
