Journal of Occupational

Assessment of Noise Effect on Employee Comfort in an Open-Plan Office: Validation of an Assessment Questionnaire

Mohammad Javad Sheikhmozafari¹, Parsa Mohammad Alizadeh¹, Omran Ahmadi^2*, Behnaz Mazloomi³

1. MSc in Occupational Health and Safety Engineering, Faculty of Medical Sciences, Tarbiat Modares University, Tehran, Iran.
2. Assistant Prof., Dept. of Occupational Health and Safety Engineering, Faculty of Medical Sciences, Tarbiat Modares University, Tehran, Iran.
3. BSc in Occupational Health and Safety Engineering, School of Public Health, Bam University of Medical Sciences, Bam, Iran.

* Corresponding author: Omran Ahmadi; E-mail: o.ahmad@modares.ac.ir   

Abstract
Background: Today, open-plan offices are among the most common work environments. Although the noise in these environments is usually below the standard level, it is one of the critical annoyance factors due to the nature of the mental work. Accordingly, this study aimed to assess noise effects on employee comfort and validate the Persian version of the assessment of noise effects on employee comfort in the open-plan office questionnaire in Iran.
Materials & Methods: The present descriptive study was conducted in an open-plan office in Shiraz, in 2021. We distributed questionnaires among 66 employees. Besides, we translated the questionnaire using the backward-forward technique, with the alpha coefficient and the intra-class correlation used to measure reliability. Finally, we used the Kappa, Spearman, and Mann-Whitney tests in SPSS V22.0 for data analysis.
Results: The validity of the questionnaire items was evaluated as acceptable using Kappa and Spearman's coefficients. Besides, the reliability of the questionnaire, using the ICC and the alpha coefficient, was 0.876 and 0.930, respectively. In this study, the employees were sensitive to noise, and their overall satisfaction with their physical conditions was moderate. In total, 47% of the employees considered the noise level of their working environment high and very high, and 35% considered it annoying or very annoying.
Conclusions: The findings of the present study showed that the questionnaire was an effective and reliable tool for noise annoyance assessments in office environments. Besides, we can use this tool to determine improvements required in open-plan offices and to evaluate the efficiency of these improvements.

Keywords: Hearing, Noise, Questionnaire, Validation.
 

Introduction
Noise  is an unwanted sound affecting people's physical and mental health [1]. Today, noise pollution is one of the major problems in societies, which affects the quality of all people's life [2]. The effects of noise on human health and the environment are increasing so dramatically that if left uncontrolled, it can lead to hazardous conditions [3]. To enumerate some noise effects on human health, one can refer to hearing loss, social irritability, speech disorders, sleep disorders, cardiovascular disorders, mental disorders, metabolic disorders, cancer exacerbation, oxidative stress, concentration reduction at work, and the like [4-6]. Noise has always been the main detrimental factor in work environments. Besides, in work environments where speech communication is one of the main ways to communicate, noise can be very annoying [2]. Nowadays, to make better use of available space and to consider economic aspects, open-plan offices are among the most common working environments [7]. Since the formation of open-plan offices, their acoustic problems have been among the major causes of job dissatisfaction. Although office employees experience noise below the hazardous level, their productivity can decrease by 66% when they are exposed to conversations of their colleagues [8]. The noise level in open-plan offices usually varies between 46 and 58 dB. In general, if the background noise level exceeds 45 dB, people's acoustic satisfaction will decrease [9]. Other problems of open-plan offices, except acoustic problems, include their effects on people's well-being, job satisfaction, and job performance [10]. A study reported that noise has a negative effect on office employees' productivity [11]. Haapakangas stated that noise had a negative effect on employees' ability to focus on the manner of performing their tasks [12]. In addition, Seddigh reported that office noise increased frustration and stress among employees [13]. A study reported noise as the sixth most important factor and the second annoying physical factor in the workplace in terms of comprehensibility, with 31% of employees having been dissatisfied with their workplace noise [9]. Oseland stated that 92% of employees reported that noise made their job performance less productive [14]. Other studies show that the conversation between individuals and their colleagues is the primary source of noise annoyance for other employees [10, 15, 16]. Iannace examined the effects of noise, heat, and ventilation systems in open-plan offices on employee comfort, According to him, when ventilation systems are active and inactive, the noise level reaches about 49 and 39 dB, respectively, which shows a significant difference [8]. Haapakanga found that the noise produced by conversations among employees plays a significant role in other employees' overall perception of noise [10]. Nezami found that the primary source of noise annoyance was the noise of conversations among people with each other, which led to a decrease in people's concentration and perception [7]. ČEKAN evaluated noise effects on employee comfort in an open-plan office [2]. Blasio reported that unnecessary and irrelevant conversations among people led to a rise in acoustical annoyance and the prevalence of symptoms related to mental health, which decreased job performance and wellbeing among employees in open-plan offices [15]. Similarly, Pierrette reported that background noise levels were high in the workplace, with unnecessary and irrelevant conversations among colleagues being the primary source of noise annoyance in open-plan offices [17]. In the same vein, many other studies have been conducted on the effects of noise on office employees [18-22].
ISO 3382-3 2012 (En) fully describes the method of measuring acoustic parameters in open-plan offices [23]. This standard considers the factors affecting the acoustic performance of open-plan offices, including the layout, acoustic absorption coefficient, and background noise. However, the way people in that environment perceive the noise of their work environment has not been taken into account. It has been stated in some studies that the perceived noise level by employees contributes to their overall perceived noise annoyance only by 20 to 25%, so other important factors must be considered as well. Thus, we should consider the factors, such as the source (s) of noise, the type of tasks performed by employees, noise sensitivity among individuals, and the work environment in studying and evaluating the noise annoyance occurring in open-plan offices [17]. All of the abovementioned factors emphasize the importance of the subjective assessment of noise by individuals and the need for a reliable tool for collecting required data. Since objective noise assessments using measurement devices do not consider the noise annoyance caused by people, mental evaluation must be carried out. Besides, the use of such tools is easier, faster, and cheaper, being possibly doable by people working in that work environment [24]. Questionnaires are among the most reliable tools for mental and qualitative noise assessments in the workplace. Pierrette developed a questionnaire and proved its validity and reliability. Many studies have used this type of questionnaire [10, 12, 15, 18, 24-28]. Using this questionnaire, Braat found that 38% of people suffered from noise annoyance in their work environment. Besides, he showed a significant relationship between acoustic characteristics of the environment and noise annoyance [29]. N. Perrin et al, using this questionnaire, found that 56% of employees considered their office noise high, and 58% considered it annoying [30]. Since there is no reliable tool to be used for physical evaluation in open-plan offices in Iran, we decided to localize and validate this questionnaire in our country. To use and localize a questionnaire from the original language into Persian, we should first examine its validity and reliability. Against this backdrop, the present study aims to assess noise effects on employee comfort; to this end, we translated and validated the Persian version of the questionnaire on the assessment of noise effects on employee comfort in open-plan offices in Shiraz, Iran.

Materials and Methods
The present descriptive-analytical study was conducted in 2021. Out of the total 79 employees working in the open-plan office in Shiraz, we selected 66 employees using an online sample size calculator for assessment. To use a questionnaire in the original language, two steps must be taken. Firstly, its linguistic validity must be determined, which was done by translating the questionnaire from the original language into Persian. Secondly, the psychometric aspects of the questionnaire must be examined. Linguistic validation and psychometric validation include the translation and cross-cultural adaptation processes of a tool into the target language so that the tool remains conceptually and semantically equivalent to the source version. Both aforementioned steps are complementary and taken to ensure the equivalents of the translated and original versions. In the first step, we are supposed to translate the original questionnaire to determine its linguistic validity. There are two methods to translate each questionnaire from the original language into another language, which include forward-backward (FB) and dual-panel (DP) methods. Accordingly, we used the backward-forward method to prepare and translate the questionnaire from the original language into Persian  [31]. In this technique, two separate experts translated the items in this questionnaire into Persian. In the next step, two Persian language experts fluent in the English language translated the Persian questionnaires into the original language, i.e. English, without seeing the first original version. Next, after checking the linguistic and content validity of the second questionnaire translated into English, we merged the translations with the original version of the questionnaire and produced the final version of
the questionnaire.
In addition, we used SPSS V22.0 to analyze the data. To evaluate the reliability of the questionnaire, we used the intra-class correlation coefficient (ICC) test and the alpha coefficient. Besides, to use the ICC test, a single group answered the questionnaire items twice under the same conditions. In addition, using the ICC test, the scores obtained were compared and used as a reliability coefficient. We also used the Kappa and Spearman's correlation coefficients to evaluate the validity of the questionnaire items. Other statistical analyses were done using Mann-Whitney and Spearman's tests.
In this study, we used the random sampling method. To determine the sample size, Cochran's formula was used. We distributed the final questionnaires among 66 employees of an open-plan office in Shiraz. At first, we briefed the occupational health officer on all conditions and objectives of the study in that office. Next, after talking to the staff and explaining the objectives of the study, the individuals participated in the study voluntarily. The inclusion criteria were having at least one year of work experience, having no mental illness, having normal hearing, and not being pregnant. On the other side, the exclusion criteria included taking certain medical drugs, suffering from congenital hearing impairment, and completing the questionnaire improperly.
The questionnaire on the assessment of noise effects on employee comfort in open-plan offices: The questionnaire generally consisted of 4 sections and 67 questions. The 5 sections of the questionnaire included studying the demographic information of the people and their workplace, assessing physical conditions of the work environment, assessing the ambient noise in the workplace, examining the relationship between employees and noise (at home, at work, and when sleep), and examining the health effects of noise on employees.

Fig. 1.  A view of the studied open-plan office and its employees

In the first part of the questionnaire, general information about people and their workstation, such as gender, age, work experience in the office, work experience in the current position, number of people working in the workstation (room), and job duties were considered. In the first part of the questionnaire, people's job satisfaction with physical conditions of their work environment was evaluated as well.
There were a total of 14 items for measuring job satisfaction on a 5-point scale (from 1 = not satisfactory at all to 5 = quite satisfactory). These 14 items measured satisfaction based on the two dimensions of control/privacy (7 items) as well as comfort/functionality (7 items). For the dimension of control/privacy, "possibilities available for noise control" were considered, and for the comfort/functionality dimension, "equipment available in the workspace" was mentioned. After studying the answers to job satisfaction questions among the employees in terms of physical conditions of the work environment on a 5-point scale, three responses were obtained, being "average global satisfaction", "average control/privacy satisfaction", and "average comfort/functionality satisfaction".
The second part of the questionnaire evaluated environmental noise made by the employees in the workplace. In the first step, the overall perceived noise level by the individuals and then the overall noise annoyance level were assessed. In this part of the questionnaire, environmental noise was broken down into 5 noise sources that included operating machines (ventilation, computers, printers, etc.), ringing telephones, intelligible conversations, unintelligible conversations, and walking people. This classification was based on the frequency of noise perception, the level of annoyance generated by each single noise, effects of these noises on the work, and if these noises were more annoying for some tasks than others. Several studies have mentioned these five noise sources as the primary sources of noise annoyance in open-plan offices [17, 32]. The perception frequency was assessed on a 5-point scale (1 = never to 5 = permanently). Besides, annoyance was assessed on a 5-point scale (1 = not annoying at all to 5 = quite annoying). Next, the employees were asked to give "yes" or "no" answers to the question if the given noise was more annoying for some work activities, and if yes, the employees were supposed to mention the activities.
In the next step, the employees were asked to sort the noise sources from the most annoying to the least annoying ones. In addition to the five noise sources mentioned above, two other sources were added, including the noise produced by people (keyboards a well as opening and closing drawers) and the noise related to a specific person.
In the third part of the questionnaire, the participants' reactions to the noise or their sensitivity to the noise were assessed. In this section, four separate questions were asked about each of the three subjects, i.e. sleep, personal habits, and work, which totaled 12 questions. These 12 questions examined the individuals' agreement with the question on a 4-point scale (1 = strongly disagree to 4 = strongly agree). The score for each question ranged from 0 to 3, which was considered as the noise sensitivity score. A score less than 1.1 indicated that the person was not sensitive to noise, and a score more than 1.63 indicated that the person was sensitive to noise.
The last (fourth) part of the questionnaire assessed the way the employees perceived their own health. This section, which included 15 questions, allowed the employees to assess their physical and mental health. These questions were used to calculate the general perceived health score of the employees, with the respondents having been supposed to express their agreement with the questions on a 5-point scale. This section calculated 4 specific scores for factors, including perceived physical health, perceived psychological health, perceived symptoms, and perceived stress. The scores in this section ranged from 1 to 5, with a score closer to 1 meaning very poor health, and a score closer to 5 indicating very satisfactory conditions. In other words, scores greater than or equal to 3.5 indicated good health, scores greater than or equal to 2.5 and less than 3.5 indicated moderate health, and scores less than 2.5 indicated poor health. The scores were reversed for the last 10 questions.
The present article is the result of a research project approved by the Ethics Committee of Tarbiat Modares University, Tehran, Iran (IR.MODARES.REC.1400.195).

Results
In this study, 66 employees (44 men and 22 women) working in an open-plan office in Shiraz were surveyed. The participants' average age was 33 years (± 6.75), which indicated that the study population was young. The average work experience of the participants was about 6 years. A total of 52% of the people confirmed they had a fixed workstation. Besides, 35%, 32%, and 11% of the employees confirmed that 7 to 15, 3 to 6, and 16 to 49 people in their workstations worked together, respectively. The reliability of the questionnaire was determined using the ICC and alpha coefficients of 0.876 and 0.930. The closer the ICC was to 1, the greater the reliability would be. Table 1 lists the reliability values of all questions. Table 2 shows validation results for the answers given by the employees for the first and second series of questions using the Kappa and Spearman's tests. It should be noted that we used the Kappa and Spearman's tests to examine the agreement and the correlation between employees' responses, respectively.

The average score obtained from the noise sensitivity assessment section was 1.80 (± 0.4). Besides, the reliability values of this part of the questionnaire using the ICC and the alpha coefficient were 0.87 and 0.93, respectively.
In addition, the overall score obtained from the perceived health assessment section was 4.18 (± 0.45). The reliability values of this part of the questionnaire using the ICC and the alpha coefficient were 0.89 and 0.94, respectively.
The score of the overall satisfaction with the workplace was 2.895 (± 0.76). Besides, we examined the scores of this section according to the two dimensions of control/privacy and comfort/functionality separately, which were 2.787 (± 0.87) and 3 (± 0.78), respectively. Reliability values of the overall satisfaction assessment with the workplace using the ICC and the alpha coefficient were 0.87 and 0.93, respectively. Besides, reliability values for the control/privacy and comfort/functionality dimensions were 0.84
and 0.91 as well as 0.90 and 0.95, respectively.
Concerning the employees' perception of workplace noise, 47% (almost half of them) considered the noise level of their work environment high (27%) or very high (20%), and 35% of the people considered that annoying (23%) or very annoying (12%) (Figs. 2 and 3).

Fig. 2. Employees' perception of the noise level in the workplace (N = 66)



Fig. 3. Employees' perception of noise annoyance in the workplace (N = 66)

The results of the Spearman's statistical test indicated that the noise level perceived and the annoyance felt were positively and strongly related ( , ).
We requested the employees to assess the frequency of various noise sources present in their workplace. Accordingly, it seemed that all five noise sources in the workplace more or less existed in the environment and had a high significance (Table 3). Besides, the noise source heard most frequently was the telephone ringing sound with 32%, which was constantly heard by the employees.