Critical indoor environmental factors affecting productivity: perspectives from university staff and postgraduate students

ABSTRACT Indoor environmental quality (IEQ) of buildings plays an important role in affecting building occupants’ comfort, health and well-being. However, there is limited understanding of the critical IEQ factors that affect occupants’ work productivity in university office buildings. By surveying 204 people (72 university staff members and 132 postgraduate students) in a tertiary education institution in New Zealand, this research aimed to identify and rank the critical IEQ factors concerning their perceived productivity and how the perceived importance of these factors differed between staff and student cohorts. Statistical analysis identified a total of 16 IEQ factors as important in affecting work productivity, and 15 of them, except cleanliness of the office, could be grouped into four categories: (1) thermal comfort and lighting; (2) acoustics and privacy comfort; (3) spatial comfort and (4) aesthetics and views. Based on the identified critical IEQ factors, a framework was developed and can be used by building facility managers and designers in the tertiary education sector to optimize design solutions to improve building performance that is more conducive to their occupants. The findings from this research can inform the inclusion of design features that will enable staff and students using these buildings to achieve better productivity.


Introduction
Workplace productivity has been a growing concern over the past decade following the recognition of the potential business benefits from a healthy and happy workforce (UK Green Building Council, 2017;World Green Building Council, 2014). According to a report from the Centre for Mental Health in 2011, reduced employee productivity due to mental health problems led to a loss of £15 billion (approximately USD$23 billion) to the UK economy per year (Centre for Mental Health, 2011). In addition, there is a wealth of evidence linking the office environment with workers' occupational health, safety and well-being (Franke & Nadler, 2021;Sadick et al., 2020;Shan et al., 2018;Wu et al., 2020). Evidence also shows that a poor office environment is likely to cause bad moods, lack of concentration, and decreased work motivation and performance among office workers (Al Horr et al., 2016;Clements-Croome et al., 2019;Roskams & Haynes, 2021).
In 2016, the World Green Building Council called for action to create better working environments that support better health and productivity outcomes (World Green Building Council, 2016). This vision, however, has not been eventuated in many countries. For example, in New Zealand, a report investigating the development of office buildings released by the Building Research Association of New Zealand in 2016 (Berg et al., 2016), showed that there had been insufficient consideration of employee productivity in the design of new buildings. This issue was brought forth by recent legislative and regulatory reforms in New Zealand to improve legislation by incorporating an aspect of well-being into building performance criteria (Nuth, 2020). Similarly, in the UK, the effects of building environments on workers' health and well-being have only been recognized in recent years (UK Green Building Council, 2017) and thus warrant a need to incorporate productivity into the design and operation of office buildings.
Among the commonly cited factors affecting workplace productivity, there are indoor air quality and thermal comfort (Kaushik et al., 2020;Soto Muñoz et al., 2022), lighting (Boubekri et al., 2020), noise and acoustics (Lou & Ou, 2020), and interior and active design (Kang et al., 2017). Studies on how an office environment affects workplace productivity can be segmented in terms of workplace context, including building type (Elnaklah et al., 2020;Mallawaarachchi et al., 2017), office layout (Kang et al., 2017) and occupation of office workers (Sadick et al., 2020). For instance, occupants in various office types may have different requirements for the working environment based on the nature of their work (Lou & Ou, 2019). It is also interesting that the effects of indoor factors on workers' productivity may vary according to demographic characteristics, such as age, gender and personality type (Andargie & Azar, 2019;Chen et al., 2020;Kallio et al., 2020;Khoshbakht et al., 2021).
There is a substantial and growing body of research investigating the influences of the office environment on workers' productivity, yet little is known about how it affects occupants engaged in academic research, specifically within a tertiary education building setting (Kang et al., 2017;Sadick et al., 2020;Shan et al., 2018). Academic research work, however, is expected to require higher levels of concentration, creativity, inspiration and consistency than many other types of work (Kang et al., 2017). Consequently, postgraduate students and academics engaged in research work could likely have different preferences concerning the nature of indoor working environments than people engaged in other types of work (Lou & Ou, 2019). To improve such an understanding, this research aims to investigate the critical IEQ factors affecting the productivity of both university staff members and postgraduate students by addressing the following two questions: (1) What critical IEQ factors affect the productivity of staff members and postgraduate students in tertiary education buildings? (2) To what extent do university staff members and postgraduate students perceive the effects of indoor environmental quality (IEQ) on their productivity differently?
By gaining the perspectives of staff members and postgraduate students in a tertiary education institution in New Zealand, this research develops a framework of the IEQ factors affecting the productivity of university staff members and postgraduate students. Building facility managers in the tertiary education sector can use such a framework to optimize their solutions to make buildings more conducive to their users in respect of productivity. Office users at tertiary educational buildings can also use the framework to identify what solutions they can put in place to improve their health and well-being and, ultimately, their work performance. The findings are expected to inform the importance of incorporating university office users' perspectives and voices into the design consideration for developing tertiary educational buildings.

Literature review
The literature search was conducted in February 2021 using the search terms '(productivity OR performance) AND (office OR workplace)' which were the most relevant and widely used in previous literature (Al Horr et al., 2016;Colenberg et al., 2021;Sui et al., 2019). Web of Science was chosen as the main search engine for its comprehensive coverage of high-quality, interdisciplinary literature (Falagas et al., 2008). The initial search was conducted in the field of 'paper title, abstract, and keywords' and resulted in 40,336 publications published in English after 2007.
The article title and abstract were screened to check the relevance of selected articles. Only articles focusing on IEQ factors and their effects on office work performance were included to establish a firm base for the research findings. Articles investigating productivity gains or losses from workplace behavioural or social environments, building energy performance, and nonoffice work performance were excluded. The initial screening returned 229 articles which were exported to EndNote reference management software. Furthermore, only journal articles published in first and second quartile ranked journals (according to SCImago Journal Rank 2020) were considered to ensure publications were high quality, as was applied in a previous systematic review (Chen et al., 2022;Royle et al., 2013).
A full-text assessment was conducted to further exclude irrelevant articles. Firstly, this review primarily reviewed publications related to tertiary educational office buildings, since this study focused on university staff members and postgraduate students. Furthermore, publications using surveys as a data collection method were included. According to the author's knowledge, the current efforts investigating the impacts of IEQ factors on productivity commonly use occupant surveys or a combination of surveys and lab experiments with sensors. Compared to a lab experiment, an occupant survey is more powerful in investigating the effects of intangible IEQ factors, such as office furnishings, visual privacy and outside views from windows (Franke & Nadler, 2021). Finally, a snowballing approach was utilized in Google Scholar to identify any relevant articles missing from the previous search process by looking at the reference lists in the selected articles. In total, 15 publications were included and classified according to the type of buildings, office layout and focus groups considering that the effects of IEQ on productivity might differ across the classifications mentioned above (see Table 1).
Across different types of office layouts and occupants, the widely studied IEQ aspects among tertiary educational buildings included: (1) air quality; (2) thermal comfort (temperature, relative humidity); (3) acoustics and privacy (speech noise); (4) lighting (natural and electric lighting) and visual comfort; and (5) spatial comfort (number of occupants, comfort of the furnishings, collaboration, privacy, friendship and cleanliness and maintenance). Detailed review is presented in reference (Liu et al., 2023). These IEQ aspects can collectively support office users' work by providing good quality of indoor environments, facilitating privacy and collaboration, and promoting happiness and well-being. Inadequate office environments have an adverse effect on office users' work performance by slowing their information processing speed (Michael et al., 2005), introducing physical discomfort (Thach et al., 2020), and distracting office users from concentration (Blasio et al., 2019;Jahncke et al., 2011;Kaarlela-Tuomaala et al., 2009), while supportive and favourable workplace environments will contribute to office users' well-being and motivation ultimately inducing higher performance (Gu et al., 2022;Harris, 2019).
Common IEQ aspects across the buildings (1) Air quality (2) Thermal comfort (temperature, relative humidity) (3) Acoustics and privacy (speech noise) (4) Lighting (natural and electric lighting) and visual comfort (5) Spatial comfort (number of occupants, comfort of the furnishings, collaboration, privacy, friendship, cleanliness and maintenance).
Although it has been identified that the effects of the IEQ factors on workplace productivity vary across office tasks, only two articles investigated the differences between office users engaged in academic work and administrative work concerning the effects of IEQ on work productivity (Lou & Ou, 2019;Sadick et al., 2020). Nevertheless, the research findings from these two articles reaffirmed that office users who engage in academic work generally had higher requirements for IEQ factors. Therefore, there is still a need for studies investigating the effects of IEQ factors on workplace productivity among university staff members and postgraduate students (Kang et al., 2017;Lou & Ou, 2019).

Research design
The mixed qualitative and quantitative methods employed for this study included: (1) a literature review; (2) a pilot study; (3) a questionnaire survey; and (4) qualitative interviews (see Figure 1), which considers the multidisciplinary nature of the research questions and has benefits over a single method (Mallawaarachchi et al., 2016). The literature review aimed to capture a holistic picture of the IEQ factors influencing work productivity for university office occupants as well as inform a survey instrument, which was the main method for data collection, followed by qualitative interviews as a supplementary data collection technique to validate and enhance the interpretation of the survey results. Previous studies have affirmed that the combination of structured questionnaires and semi-structured interviews performs well in generating confirmatory results (Harris & Brown, 2010;Mallawaarachchi et al., 2016). Survey data were analysed using one-sample ttest, Mann-Whitney U test and exploratory factor analysis (EFA) to identify and rank IEQ factors. A framework of IEQ factors affecting the productivity of university office users was proposed based on the survey analysis results, and thematic content analysis and Kendall's concordance analysis were used for interview data to validate the proposed framework.

Data collection
Eligible participants for this study were university staff members and postgraduate students who worked in one of four university office buildings (Buildings 401,405,901,906) at the University of Auckland, New Zealand. Detailed information on the selection of studied areas can be found in Liu et al. (2023). In prior of the formal survey distribution, a pilot study was undertaken to validate and refine the factors derived from the literature review and improve the user-friendliness of the questionnaire survey. A random sampling approach was taken for the pilot study among university staff members and postgraduate students working in the previously mentioned buildings. Recruitment for pilot study participants was advertised across the selected buildings, with the participant criteria clearly described. A total of 15 people, including three academic staff members and 12 postgraduate students, expressed their willingness to participate in the pilot study and participated. Some questions were modified in response to feedback from the pilot study participants, and additional attention was given to the question sequence, formulation, wording and layout.
The formal survey comprised five sections and survey questions regarding participant demographics and perceptions of the importance of indoor factors on productivity were included in the analysis for this research (see Table 2). The finalized IEQ factors used in the questionnaire were listed accordingly. It should be noted that this survey used a perception approach to measure the critical IEQ factors concerning work productivity. Respondents were asked to estimate the importance of each indoor element (IEQ factor) regarding their ability to concentrate and be productive using a seven-point Likert scale (from 1 = completely unimportant to 7 = extremely important).
The survey was administered for two months, between 15 July and 15 September 2021. Participants were invited using an email distribution list, following approval by faculty leadership. The survey was distributed to a sample of university staff members and postgraduate students in which pilot study participants were included. Since the pilot study participants were invited to participate in the pilot study to only validate the survey and improve its relevance and user-friendliness, it is not a repeated measurement if some of the pilot study participants have also participated in the survey. After the survey was open for completion for two weeks, door knocking was undertaken to increase participation. Only one response was permitted per participant, either via electronic format or hard copy (collected by the author within three days of distribution). The collection of hard copies was affected by a COVID-19 lockdown in Auckland. The study was approved by the University of Auckland Human Participation Ethics Committee (Reference Number UAHPEC22572).
After the survey results were analysed, respondents who had indicated their willingness to participate in an interview were contacted by email for one-to-one interviews via Zoom. Interviewees were selected based on their roles at the university and located office buildings to enhance the representation of the sample (see Table 3). In total, 42 out of 91 participants were invited, and 14 people, including six staff members and eight postgraduate students, finally participated in interviews.  of the desk or chairs, functionality of stationary equipment) E14-The size of your working space E15-Storage space E16-Cleanliness of your office Note: In this survey, productivity refers to how well someone performs their expected work tasks in a timely manner.
The semi-structured interview is expected to last for 15 min. Firstly, the interviewer will briefly explain the framework developed from the survey results, then several interview questions were asked to confirm the reliability of the critical indoor elements and its corresponding categorization, including: (1) How do you think of the framework developed?
(2) How will you re-rank the IEQ categories? Why? (3) How will you re-rank the subfactors under each IEQ category? Why? (4) Any others?
Data cleaning Survey data were analysed using IBM SPSS Statistics 27 for Windows. Data cleaning was conducted before analysis to increase the reported results' external validity (Rousseeuw & Hubert, 2011). A total of 246 responses remained after excluding participants who completed less than 90% of the survey questions, representing a response rate of approximately 47.86% (246/514) of the targeted population. Records with missing values concerning questions presented in Table 2 were further excluded. Additionally, univariate outliers and multivariate outliers were excluded using standardized scores (|z-scores| > 3.29) and Mahalanobis distance statistics (p-value < .001) (Tabachnick & Fidell, 2019), respectively. Finally, 204 responses were analysed after excluding cases with missing values (8), univariate outliers (31) and multivariate outliers (3). Of those returning questionnaires, 91 (44.61%) indicated their willingness to participate in a follow-up interview.

Data analysis and validation
Mean values of the 16 IEQ factors were calculated using the seven-point Likert scale and one-sample t-test was conducted to identify the critical IEQ factors. A Mann-Whitney U test was undertaken to explore whether the perceived importance of each indoor environment factor significantly differ based on roles at the university. A Mann-Whitney U test is a commonly used non-parametric test to compare two independent groups (McCrum-Gardner, 2008), with a p-value less than .05 indicating significant differences (Kang et al., 2018;Pereira & Leslie, 2010). As a supplement, Cliff's delta (d) was calculated as the effect size measurement to further illustrate the magnitude of differences due to its high robustness and power (Chi et al., 2020;Cliff, 1993). Additionally, the nature of the effect size will aid future studies that seek to compare and consulate results of related studies.
Exploratory factor analysis (EFA), a preferred method for identifying latent factor structures (Fabrigar & Wegener, 2011), was employed to extract the latent factors from the 16 identified IEQ factors adopted in previous literature. Principal axis factoring was chosen for factor extraction, given its relative tolerance with respect to the assumption of normality (Costello & Osborne, 2005;Watkins, 2018). Given that the IEQ factors were expected to be correlated, promax rotation was chosen, as it can offer a more realistic representation of the data (Fabrigar & Wegener, 2011;Loehlin & Beaujean, 2017). Data from 15 of the indoor elements (E1-E15) were appropriate for factor analysis. The 'cleanliness of the office' variable was removed as the factor loading for this variable failed to meet the 0.4 threshold to ensure that the factors were practically and statistically significant (Bandalos & Gerstner, 2016;Norman & Streiner, 2014). The results of Bartlett's test of sphericity suggested that the correlation matrix was not random (χ 2 (105) = 1261, p < .001), and the Kaiser-Meyer-Olkin (KMO) measure indicated the sampling accuracy was also good (KMO = 0.809 > 0.7) (Bartlett, 1954;Hoelzle & Meyer, 2012;Kaiser, 1974;Susana, 2017). Therefore, the data were suitable for factor analysis.
The appropriate number of latent factors (components) to retain was guided by MAP (minimum average partials), parallel analysis, eigenvalues greater than one, and a scree plot, as suggested in previous research (Fabrigar & Wegener, 2011;Sadick et al., 2020;Watkins, 2018). The original 1976 MAP test suggested that three components should be retained. However, the revised 2000 MAP test suggested that two components should be extracted. Parallel analysis, eigenvalue and scree plots indicated that the number of components to be retained should be three, four and six, respectively. Finally, a four-factor solution was accepted as an appropriate structural representation given the conceptional  5  20  2  1  28  Invited via email  3  11  1  1  16  Interviewed  1  3  1  1  6  Postgraduate students  Willing to participate  1  40  5  17  63  Invited via email  1  17  3  5  26  Interviewed  1  5  1  1  8 similarities of the indoor elements within each component. Interview notes were analysed using thematic content analysis (Gone, 2009;Malik et al., 2019). Analysis was conducted manually, considering the limited data volume. Interview transcripts were reviewed several times to pinpoint, examine and record patterns to avoid possibly missing patterns. Identified patterns were extracted and classified. Finally, Kendall's concordance analysis was used to measure the agreement concerning the level of importance of indoor elements between the survey data and interview data. Table 4 presents the demographic information of the study participants. A significant number of participants were in the age ranges of 20-29 years (77, 37.75%) and 30-39 years (75, 36.76%), with the majority of participants (105, 51.47%) having resided in New Zealand for the last 1-5 years. This is understandable, considering that most of the participants who completed the survey were postgraduate students (132, 64.71%), many of whom were international. In addition, there were more males (145, 71.08%) than females (59, 28.92%) among respondents, with Asians (114, 55.88%) making up the highest proportion of the participants, followed by New Zealand Europeans (36, 17.65%).

Descriptive analysis
A one-sample t-test indicated that all IEQ factors were considered important to work productivity, since all mean values were higher than the neutral value of 4.0 with a level of significance less than .05. The 16 IEQ factors were ranked according to the perceived importance concerning work productivity against the mean values (see Table 5). In addition, 10 IEQ factors (lighting (6.08), level of noise from others (6.01), freshness of air (5.94), temperature (5.83), the set-up of workstations (5.81), cleanliness of office (5.78), air flows (5.69), soundproofing of the office (5.65), exposure to sunlight (5.63), visual privacy (5.58)) were ranked as top critical IEQ factors according to the criticality cut off point of 5.56, which was the average mean value of all IEQ factors. This approach has been adopted in previous studies when selecting critical factors (Ayodele et al., 2022).
The results of Mann-Whitney U test and effect size (d) are presented in Table 6. Significant differences were found in the perceived importance of four IEQ factors, including temperature in the office (U = 3588.5, p = .002, d = −0.245), air flows (U = 3957.5, d = −0.167, p = .037), exposure to sunlight (U = 3908.5, d = 0.178, p = .030) and the size of the working space (U = 3739.0, d = −0.213, p = .009) between university staff members and postgraduate students. Postgraduate tudents rated the importance of temperature in the office, air flows and the size of working space higher in relation to their work productivity, while staff  members scored higher in terms of the perceived importance of exposure to sunlight. According to the thresholds provided by Romano (Chi et al., 2020;Romano et al., 2006), the magnitude of differences is relatively small (|d| < 0.33) for these four IEQ factors, and the top two largest differences were found in temperature in the office and the size of the working space by comparing the effect sizes of all IEQ factors. No statistically significant differences were found between the two cohorts for the rest of the IEQ factors. Table 7 presents details of the 15-item four-factor solution. The remaining 15 IEQ factors were found to be adequate with respect to their usefulness in interpreting their respective components, given that all factor loadings were greater than 0.4. The internal reliability of each of the four components was also adequate, given the Cronbach's alpha was greater than 0.70 (Hair et al., 2005). This four-factor solution explained 65.37% of total variance, which indicated a strong sense of predication of these IEQ factors building on the factor analysis (Rosenthal, 2011). The names of these four latent factors were proposed based on previous literature (Al Horr et al., 2016;Lou & Ou, 2019;Sadick et al., 2020) considering the conceptionally similar indoor element compositions.

Factor analysis of critical IEQ factors
The 15 IEQ factors within each category were reranked according to the calculated mean values of the perceived importance of each IEQ factor (see Appendix  Table A1). Furthermore, a framework of IEQ factors affecting the productivity of university staff members and postgraduate students was proposed building on the category and ranking of each IEQ factor (see Figure 2). The category of thermal comfort and lighting (5.754) was considered as the most important category concerning work productivity, followed by acoustics and privacy comfort (5.747), spatial comfort (5.397), and aesthetics and views (5.243). It can be seen that 'thermal comfort and lighting' and 'acoustics and privacy comfort' comprised most of the top critical IEQ factors. It should be noted that although the IEQ factor 'cleanliness of the office' was not grouped into a denotes the number of times.
(3) CI: Confidence interval. category, it was still considered in the framework and listed separately due to its high perceived importance concerning productivity amongst university office users.

Validation results
The qualitative findings from the follow-up interviews were summarized in Appendix Table A2, supported with some key extracts from the interviews. In general, all 14 interviewees opined that this framework was clear and reasonable regarding the coverage of IEQ factors that are important to work productivity. There were two participants suggested additional factors that could be included in the framework. One participant suggested to add 'noise of evacuations' or 'construction noise' in the category acoustics and privacy comfort since 'there was undergoing construction work outside the office building'. The finalized framework, however, does not include this sub-factor for the following reasons. One is that this framework is designed for indoor environment factors while construction noise is a factor outside the office. In addition, the influences of construction noise to some extents are related to the soundproofing of the office, which is one indoor factor that the framework has already included. Another participant suggested including the comfort of chair and a suitable height of the desk in the category of spatial comfort. This has already been considered in 'the set-up of workstation' which incorporates the conditions of furnishings. Further adjustments of the ranking and category of IEQ factors were promoted, including promoting higher importance on the category of aesthetics and views and moving exposure to sunlight to the category of thermal comfort and lighting. Additional attention should be paid to visual privacy, as interviewees differ a lot in its importance to work productivity. Some people indicate that they attach higher importance or even top on visual privacy within the acoustic and privacy category, while the others argue that visual privacy is not an issue for them. Generally, interviewees are likely to adjust the importance of certain IEQ factors concerning their productivity based on their personal preferences, the ability to control the indoor environment to their own comfort level, and their satisfaction with the current office environment. (2) PG signals Postgraduate students attach significantly higher importance than staff members, S signals staff members attach significantly higher importance than postgraduate students.
Despite the various adjustments that interviewees would prefer to make, the overall importance ranking of the IEQ factors from the two groups of interviewees showed positive and strong agreement with the original ranking in the framework. As shown in Appendix Table  A3, Kendall's coefficients were greater than 0.9 (w was greater than 0 and close to 1 (Adafin et al., 2016)) for the four latent categories, as well as the IEQ factors within the categories of thermal comfort and lighting, spatial comfort, and aesthetics and views. Slightly different from the survey results (w = 0.778), visual privacy acquired higher importance from staff members and postgraduate students. The interviewees further explained that they prioritized visual privacy in the context of the confidentiality requirement of some of their work, and not from the perspective of productivity. In addition, visual privacy was more difficult to deal with than soundproofing issues where they could wear headphones. Therefore, the proposed framework is generally considerable from the perspective of productivityoriented.

Discussion
The findings of this research confirmed the significance of the 16 IEQ factors for workplace productivity for university staff members and postgraduate students. The university office users perceived lighting, level of noise from others, freshness of air, temperature and the setup of workstations as the top five critical IEQ factors for their productivity, which generally satisfied the fundamental requirements of an office to support work activities (Clements-Croome et al., 2019). In particular, university staff members attach a higher level of importance to exposure to sunlight than postgraduate students, this possibly because most staff members are allocated offices without external-facing windows. Additional attention should be given to this explanation, as it is based on how staff members and postgraduate students perceive the importance of IEQ factors to their productivity may be related to their satisfaction level with IEQ aspects in the office. As suggested by Chen et al. (2020) based on data from university offices across six countries, IEQ satisfaction has a strong positive influence on occupants' perceptions of the link between the five IEQ aspects (indoor temperature, air, natural and electric lighting and acoustics) and productivity.
The proposed framework of 16 IEQ factors in respect of their importance to work productivity was considered reasonable and comprehensive by the interview results. However, this ranking differs from Kang et al. (2017), who indicated that the quality of the acoustic environment has the largest effect on postgraduate students in open-plan research offices, and Rasheed et al. (2019) who suggested that the perception of the noise level was the key factor affecting productivity for occupants in all office types (solo, duo, 2-4, 5-8 and 8+ offices). One possible explanation is that most of the university staff members and students surveyed were satisfied with their acoustic environment, and it is easier for them to take personal measures to prevent the effects of noise in comparison with the effects of thermal comfort and lighting. This again indicates the possibility of how university office users perceive the importance of IEQ factors to their productivity may be related to their satisfaction with the current IEQ conditions.
The identified top critical IEQ factors in the framework were across four different IEQ categories and most were within top two categories of 'thermal comfort and lighting' and 'acoustics and privacy comfort'. This finding may imply that a productivity-oriented workplace can only be achieved by adhering to all four IEQ categories, and that thermal comfort, lighting and acoustic aspects in the office space are more important. This is in line with empirical findings from Roskams and Haynes (2021), who indicated that the employees rated their highest level of productivity when they were satisfied with temperature, lighting, air quality and noise levels within the office. In addition, IEQ categories with more tangible, sensor-measurable factors had higher importance among university office users, which is understandable since these tangible IEQ aspects are mostly the fundamental requirements to undertake office activities. However, this finding differs from Franke and Nadler (2021), who suggested that workplace productivity is more strongly affected by intangible factors (privacy and ergonomics) than by tangible factors (temperature and indoor air quality) based on a laboratory-controlled test among university students. Possible reason could be due to the disparities in the control of the IEQ aspects among participants in laboratory control experiments and a real office room.
This article captured a group of building occupants (university staff members and postgraduate students) that have received very limited research attention regarding the perceived effects of IEQ on their productivity. This article has the potential to add to our understanding and could make a significant contribution to the knowledge of the effects of building IEQ on workforce productivity from tertiary education building users' perspectives. The developed framework can be used as a checklist for tertiary education building facility managers and designers to consider design solutions that can improve building performance in respect of enhanced productivity for users. For example, thermal comfort and lighting were given the highest importance in relation to work productivity. This could indicate a need for better and more user-friendly thermal control and lighting systems for tertiary education buildings. Although aesthetics and views are also important to work productivity, they acquired the least importance than other IEQ aspects, which may indicate the possibility of limiting investment in aesthetic and view design when this aspect had already met the basic requirements in building standards.
The framework could also be used by university staff members and postgraduate students at tertiary educational buildings to identify solutions for improving their health and well-being and, ultimately, their work performance. For instance, improving the acoustics and privacy of their office environment in order to be able to focus and be more productive could involve establishing office etiquette that considers noise reduction. However, the Mann-Whitney U test revealed that staff members and postgraduate students perceived the importance of several IEQ factors differently. It is therefore important to take their respective needs into design considerations when selecting orientation and ventilation systems for the building to be 'fit for purpose'. It is also hoped that the findings from this research will raise the awareness of tertiary building users, especially postgraduate students, about the effects of IEQ on their productivity which is a critical contributor to the overall research reputation of tertiary education institutions. Postgraduate students should be engaged and consulted at the initial design phase of a building, and their views on how IEQ can affect their well-being and productivity should be considered.
There are several limitations. First, this study only focused on IEQ factors. Other social factors, such as connection between co-workers and support from the organizations and outdoor environmental factors, were not included in the analysis, which may also affect workplace productivity based on existing literature. Another concern is that the findings are based on subjective measures. The findings of this study should be viewed as presenting the views of those who work in university office buildings. As this article is part of the work of ongoing PhD research, an investigation of environmental data measured by sensors will be conducted in future work to enrich the empirical findings and gain more precise assessments.

Conclusion
Overall, lighting, level of noise from others, air freshness, temperature and the set-up of workstations were rated as the top five critical IEQ factors in relation to productivity. University staff tended to indicate higher importance for exposure to sunlight in relation to work productivity, while postgraduate students valued temperature in the office, air flows and the size of the working space more. A framework was further developed, comprising 16 critical IEQ factors that were grouped into four ranked categories as: (1) thermal comfort and lighting; (2) acoustics and privacy comfort; (3) spatial comfort; and (4) aesthetics and views. This framework was validated as being comprehensive and clear regarding IEQ factors affecting work productivity by the interviewees.
Findings from this research enhance the body of knowledge on IEQ literature regarding the effects of IEQ factors on workplace productivity from the perspectives of university staff and postgraduate students and provides insights related to the design of university buildings, particularly in New Zealand. The developed framework can be used as a checklist for tertiary education building facility managers and designers to consider design solutions that can improve users' productivity. University staff members and postgraduate students are expected to use the framework to inform their behaviours by identifying solutions they can implement to improve their productivity.

Disclosure statement
No potential conflict of interest was reported by the author(s). . 'In general, the framework is clear, reasonable, and easy to understand.' (B9604DT0112, B4511CX0107, B4511KN0108)

Funding Appendices
. 'I think this framework is reasonable. It looks like a normative result.' (B4100CM0930) . 'Definitely I agree with the top category, thermal comfort and lighting and then acoustics and privacy comfort.' (B4509PA1003) . 'In general, thermal comfort, lighting, and acoustic are essential factors, which can directly affect one's motivation on working. Aesthetics and views and spatial comfort are bonus factors which can only have an impact on one's productivity when we have worked for a certain time.' (B9105JZ0112) . 'With respect to 'thermal comfort and lighting', I generally think that the sub-factors within this category are reasonable and important.' (B4511ZS0108, B9105JZ0112)

Supplemental subfactors
Additional sub-factors including 'construction noise', 'the comfort of a chair' and 'a suitable height of the desk' were suggested to be added in the framework.
. 'There was undergoing construction work outside the office building during the survey period, how about add noise of evacuations or construction noise in the category acoustics and privacy comfort.' (B9604DT0112) . 'Regarding category 'spatial comfort', it might be better to include furniture and equipment, like 'the comfort of chair' and 'a suitable height of the desk'.' (B9105JZ0112)

Adjustment of IEQ factors
Adjusting the category of 'exposure to sunlight' 'Exposure to sunlight' is proposed to be in category 'thermal comfort and lighting'.
. ''Exposure to sunlight' may be classified into category 1 'thermal comfort and lighting', as this might refer to the sunlight in a room which affects the lighting and temperature in a room.' Higher importance is given to 'aesthetics and views' from the concern of occupants' preferences on natural elements.
. 'I think aesthetics and views would be my number two category. My reasoning is for sunlight, the sunlight is the number one in this category.' (B4100CM0930) . 'I would rank aesthetics and views more important above spatial comfort, because exposure to sunlight is really big, and having a nice view and greenness.' (B4509PA1003)