Differences in teachers’ satisfaction with indoor environmental quality and their well-being in new,renovated and non-renovated schools

Abstract

Most studies on buildings’ renovations in relation to indoor environmental quality (IEQ) and occupants’ well-being have focused on offices, while those investigating schools focused on students rather than teachers. Most of these studies limited their assessment of well-being to occupants’ satisfaction with IEQ factors, overlooking essential aspects related to psychological, social and physical well-being. This article presents results of a research conducted in 32 schools in Manitoba, Canada, to assess teachers’ IEQ satisfaction and well-being in new, renovated and non-renovated schools. The research involved adapting and refining an IEQ satisfaction survey and developing and refining three new surveys to assess teachers’ psychological, social and physical well-being. The results of the refined surveys showed statistically significant differences in teachers’ satisfaction with IEQ factors between the new and renovated schools on one hand and the non-renovated ones on the other. However, no statistically significant differences were found in teachers’ psychological, social and physical well-being perceptions between all pairs of the three school categories analysed. Association analyses suggested a potential indirect impact of schools’ renovations on teachers’ well-being via their satisfaction with IEQ. The results of this study should aid school managers in making strategic decisions about the maintenance of their existing schools.

Keywords

Indoor environmental quality maintenance renovations schools teachers well-being

Introduction

The working condition of building elements such as walls, windows and floors declines during the operation phase of buildings due to known and unknown factors including visible defects and exposure to adverse environmental conditions.¹ This will affect buildings’ performance with respect to energy consumption and primarily, the provision of safe and comfortable indoor environments to occupants. Hence, the maintenance (i.e. renovation, repair, refurbishment and retrofit) of buildings during their service life is fundamental to accomplishing their functional requirements.² Educational buildings constitute a significant proportion of public infrastructure. In Canada, they represented half of public infrastructure as of 2008.³ From 2012 to 2014, Canada spent an average of CAD$ 1.69 billion⁴ on the repair of educational infrastructure.

Several studies^5–7 assessed the relationship between buildings’ maintenance and their energy performance given the potential economic benefits of reduced energy consumption. Others^8–10 investigated occupants’ satisfaction with indoor environmental quality (IEQ) in relation to buildings’ maintenance. While some studies^11–13 focused on investigating the impact of energy retrofitting on IEQ, others^10,14,15 focused on IEQ-related maintenance in general. Although results generally suggested that improving IEQ improved occupants’ satisfaction, these studies usually focused on office buildings, with limited empirical evidence on how the maintenance of school buildings influenced occupants’ IEQ satisfaction and well-being.

The substantial expenditure on school buildings’ maintenance calls for empirical research focused on end user benefits to justify the outlay and improve strategic maintenance decision-making. IEQ research in schools has historically focused on students^16–18 despite a study¹⁹ in the United States showing that improvements in school facilities could lead to teacher retention more than salary increases. Another study²⁰ reported a positive association between teacher effectiveness and student test scores, thus, reinforcing the need to improve the working condition of school buildings to enhance teachers’ performance.

This research focused on investigating the impact of IEQ-related renovations on teachers’ IEQ satisfaction and well-being, by assessing the differences in teachers’ IEQ satisfaction and well-being in new, renovated and non-renovated schools in Manitoba, Canada. IEQ-related renovations in this research refer to renovations implemented with the aim of enhancing buildings’ IEQ performance and ultimately their occupants’ IEQ satisfaction. Examples include improving the thermal properties of building envelopes with insulation to minimize indoor temperature fluctuations and thus improve occupants’ thermal comfort. They also include using mechanical ventilation to decrease the concentration of indoor air pollutants, increase indoor ventilation rates and thus improve indoor air quality satisfaction.

This research is part of a large study aiming to analyse the physical and IEQ conditions of schools in relation to teachers’ IEQ satisfaction and well-being. As a part of the large study, an existing IEQ satisfaction survey was adapted and refined to assess school teachers’ IEQ satisfaction. Additionally, three new well-being surveys were developed and validated to assess teachers’ psychological, social and physical well-being. This research was based on responses from these surveys. Hence, this research should be of interest to school divisions as well as design and operation and maintenance professionals looking to renovate existing schools, or design and build new ones. It should also be of interest to IEQ researchers looking to study the link between the physical condition of existing buildings and their occupants’ IEQ satisfaction and well-being.

Background information

Buildings are renovated for several reasons such as to enhance their aesthetic appearance, to extend their service life and to increase their rental value.²¹ Another driving force of building renovation is the potential cost savings associated with reducing buildings’ energy consumption.^9,12 Zinzi et al.²² noted 80% energy savings in a three-storey school building after deep energy retrofitting. Energy retrofits generally involve enhancing the thermal performance of building envelopes, installing energy efficient heating, ventilation, air-conditioning (HVAC) systems and lighting upgrades.^8,11 Evidence in the literature^11,14,23 points to the growing research interest in building renovations aimed at improving IEQ.

The articles and studies reviewed in this section were sourced from the electronic databases: Engineering Village, Science Direct and Scopus using the ‘AND’ Boolean logic and a combination of keywords including ‘air quality’, ‘energy retrofit’, ‘indoor environment quality’, ‘indoor air quality’, ‘occupants’ satisfaction’, ‘renovation’, ‘refurbishment’, ‘retrofit’, ‘schools’, ‘school building’ and ‘thermal comfort’. Studies that did not include an assessment of occupants’ IEQ satisfaction in relation to IEQ renovations were excluded. This resulted in a total of 10 peer-reviewed journal papers or studies, three of which focused on school buildings. These 10 studies were categorized into one of three types: (a) renovated versus non-renovated, (b) pre-renovation versus post-renovation and (c) post-renovation. Renovation was used in this context to represent building maintenance interventions including retrofits, refurbishments and repairs.

Renovated versus non-renovated

Studies under this category investigated occupants’ IEQ satisfaction in renovated versus non-renovated buildings of the same occupancy type, in the same geographic area, and with the same climate. Table 1 summarises the studies under this category. These studies support the thesis that IEQ-related renovations could improve occupants’ IEQ satisfaction, though, none of them reported the statistical significance of their observed differences. Moreover, these studies involved surveying occupants’ in the renovated buildings post-renovation only. Pre-renovation differences were not accounted for.

Table 1.

Summary of renovated versus non-renovated buildings research studies.

Study	Building/renovation details	Results
Liu et al.⁸	Buildings: multi-family buildings; one energy retrofitted and five non-energy retrofitted. Retrofit details: improved thermal properties of building envelope and installed energy efficient HVAC.	Occupants of energy retrofitted buildings were the most satisfied with thermal comfort, air quality and acoustics; thermal comfort satisfaction difference was markedly high.
Földváry et al.¹³	Buildings: naturally ventilated homes; six energy retrofitted and six non-energy retrofitted. Retrofit details: thermally insulated facades, energy efficient windows and localised radiator control.	Occupants of energy retrofitted homes were the most satisfied with thermal comfort; however, they were the most dissatisfied with air quality due to improved air tightness of envelopes and low ventilation rate.
Ghita and Catalina^11 a	Buildings: naturally ventilated schools; one new, one renovated and one old school. Renovation details: improved thermal properties of building envelope.	Students in new and renovated schools were the most satisfied with thermal comfort.

This study was included in this category as new buildings represent a special case of non-renovated buildings.

Pre-renovation versus post-renovation

Pre-renovation versus post-renovation refers to studies that surveyed the IEQ satisfaction of the same group of occupants before and after a renovation intervention. The advantage of these studies is their ability to determine the specific effect of IEQ-related renovations due to their use of the same group of respondents in their pre-post research design. Their findings, summarised in Table 2, confirmed findings of the renovated versus non-renovated studies and the thesis that IEQ-related renovations could improve occupants’ IEQ satisfaction.

Table 2.

Summary of pre-renovation versus post-renovation research studies.

Study	Building/renovation details	Results
Hongisto et al.¹⁴	Buildings: open-plan office; one refurbished office building. Refurbishment details: improved air quality, thermal comfort, acoustics and occupant density.	Statistically significant improvement in occupants’ IEQ satisfaction and job satisfaction.
Noris et al.¹⁰	Buildings: apartments; 16 energy retrofitted apartments. Retrofit details: retrofitted to improve energy performance and indoor air quality.	Majority of respondents reported improved indoor air quality.
Schreuder et al.²⁴	Buildings: office buildings; moved from 11 old to three new office buildings. New building details: environmentally friendly with superior IEQ.	Statistically significant improvement in IEQ satisfaction and well-being.
Zinzi et al.²²	Buildings: school building; three storeys high renovated. Renovation details: insulated building envelope, installed mechanical ventilation, energy efficient windows and new sun shades.	Proportion of students satisfied with thermal comfort, air quality, daylighting and acoustics before the renovations increased substantially after the renovations.

Post-renovation

Post-renovation studies summarized in Table 3 investigated occupants’ IEQ satisfaction in renovated buildings only without comparing them to any other group of occupants or to the same group prior to the renovation, thereby limiting the implication of their results. Their results did not consistently confirm the thesis that IEQ-related renovations could improve occupants’ IEQ satisfaction.

Table 3.

Summary of post-renovation research studies.

Study	Building/renovation details	Results
Singh et al.²⁵	Buildings: homes; 80 renovated homes. Refurbishment details: double glazed windows and insulated roofing.	Majority of occupants were dissatisfied with thermal comfort.
Salleh et al.²⁶	Buildings: school buildings; 240 renovated buildings. Retrofit details: single unit homes, single and double-storey terrace homes and office buildings converted to kindergarten.	Teachers and other staff were most satisfied with daylight but were dissatisfied with air quality and acoustics due to proximity of schools to busy roads and violations of minimum classroom space requirements for children in Malaysia.
Wells et al.⁹	Buildings: homes; 12 low-income new homes. New building details: six deep energy (DER) retrofitted and six energy star (ES) retrofitted.	Occupants of ES retrofitted homes were most satisfied with thermal comfort relative to those of DER due to latter’s unfamiliarity with the newly installed heating system.

Indoor environmental quality satisfaction and well-being

The term occupant well-being is not explicitly defined in the IEQ literature. The term tends to be used interchangeably with the terms occupant comfort or satisfaction to refer to occupants’ satisfaction with the four main IEQ factors namely thermal comfort, air quality, lighting and acoustics.^27–29 It is generally assessed using questionnaire surveys.

Although the term well-being is defined in the social sciences literature, there is no universally accepted definition for it. Dodge et al.³⁰ defined individual well-being as ‘the balance point between an individual’s resource pool (i.e. psychological, social and physical) and the challenges faced (i.e. psychological, social and physical)’. Hence, for building occupants, such as school teachers, that balance can be offset by adverse indoor conditions such as high temperature, and require a resource such as operable window to restore it. A review of the instruments used to assess well-being in the social sciences literature suggests that assessment of occupant well-being should not be limited to IEQ satisfaction factors. For instance, Ryff’s Scale of Psychological Well-Being includes the dimensions ‘positive relations with others’ and ‘environmental mastery’,³¹ while the RAND-36 health survey’s physical well-being components includes the dimensions ‘physical functioning’ and ‘energy/fatigue’.³² These dimensions are not captured in IEQ satisfaction surveys, thus, calling into question the comprehensiveness of these surveys for assessing occupants’ well-being. Nevertheless, one limitation of the well-being instruments used in the social sciences is that they are mostly context-free. Therefore, they are likely to exclude aspects of personal well-being that are unique to teachers in school environments, thus the need to develop new tools to account for these aspects.

Methods

This two-stage, mixed-method research was conducted in collaboration with the Government of Manitoba Public School Finance Board and two Manitoba school divisions. Stage 1 involved developing new surveys to assess teacher’s psychological well-being (TPWB), teacher’s social well-being (TSWB) and teacher’s physical well-being (TPhWB) in schools using an inductive approach. These well-being surveys excluded questions on teachers’ personal characteristics such as current health status due to restrictions by the University of Manitoba Research Ethics Board on collecting such data. Additionally, an existing IEQ satisfaction survey was adapted (from the National Research Council Canada (NRC)³³) to assess teachers’ IEQ satisfaction in schools. The second stage involved administering an online survey composed of the adapted IEQ satisfaction survey and the new teacher well-being surveys developed in stage one. Additionally, the administered surveys were refined through factor analysis.³⁴ This research presents results from the refined surveys.

Study location and buildings

The two school divisions that participated in this research included a total of 32 elementary to secondary schools, mainly in rural Manitoba, southeast of Winnipeg. A two-stage classification strategy was employed to stratify these 32 schools for analysis purposes. The first stage involved classifying them based on their start date of operation, which ranged from 1950 to 2012, representing a period of 62 years. This period was divided into three approximate 20-year periods, resulting in three school strata as follows: new schools from 1992 to 2012 (7 in number), middle age schools from 1971 to 1991 (6 in number) and old schools from 1950 to 1970 (19 in number). The 20-year period seemed logical given that the National Building Code of Canada (NCB) had historically been revised approximately every 5 to 10 years from 1941 to 2015. Given that the 32 schools were constructed to different versions of the NCB, and that successive versions of the NBC are expected to improve the performance of buildings, it is logical to expect a substantial performance gap between these three school strata. The second stage of the classification was based on the schools’ renovation records supplied by the two school divisions. From 2007 to 2014, 12 old schools and 5 middle age ones underwent substantial IEQ-related renovations to improve their IEQ performance. Table 5 summarizes the nature of these renovations and their potential impacts. Therefore, the schools were reclassified by merging the 17 old and middle age schools that were renovated into one stratum (i.e. renovated) and the remaining 8 old and middle age schools that were not renovated (i.e. not-renovated) into another stratum. The new school stratum from the first stage remained unchanged. Table 4 summarizes these three school strata.

Table 4.

Strata of schools.

Name of stratum	Abbreviated stratum name	Start date of operation	Number of schools	Number of students	Number of teachers
1. New schools	New	1992–2012	7	2520	172
2. Old/middle age recently renovated	Renovated	1971–1950	17	7892	557
3. Old/middle age not recently renovated	Non-renovated	1971–1950	8	3322	123

Table 5.

IEQ-related renovations executed in stratum 2 schools and their expected impact.

Renovations	Expected impact on IEQ	Number of schools
Building envelope		17
• Partially replacing existing walls • Partially or completely reinsulating building envelopes • Completely replacing existing glazed windows with new low U-value and high visible light transmittance glazed windows	• Minimize thermal loss of building envelope to avoid rapid fluctuations in indoor air temperature levels • Increase the amount of daylight indoors • New walls would likely increase attenuation of sound from adjoining classrooms and outdoors • Tightening of building envelope will likely decrease exfiltration of indoor air pollutants	17
Finishes		13
• Using low volatile organic compound (VOC) materials for painting, flooring and ceiling • Replacing existing sound absorbing finishes and fixtures (e.g. acoustics ceiling)	• Decrease the concentration of VOCs in classrooms • Enhance the decay of sound indoors to minimize reverberation	13
HVAC		4
Refurbishing or replacing existing HVAC systems	• Enhance ventilation and decrease concentration of indoor air contaminants • Provide the required heating and cooling indoors	4
Lighting		13
Replacing lighting fixtures	• Enhance the quality of vision indoors	13

Schools in stratum 1 (i.e. new) were constructed in accordance with the NBC versions 1985 to 2010. Three of the oldest new schools were substantially upgraded within the last 10 years to enhance their IEQ while the rest only experienced routine maintenance such as replacing broken items (e.g. damaged window glazing). Schools in stratum 2 (i.e. renovated) were a mix of old and middle age schools constructed to NBC versions 1950 to 1980. All 17 schools in this stratum experienced substantial IEQ-related renovations as shown in Table 5. Stratum 3 schools (i.e. non-renovated) were also a mix of old and middle age schools constructed to NBC versions 1950 to 1980. However, they did not undergo the substantial IEQ-related renovations experienced by stratum 2 schools from year 2007 to 2014.

The new schools’ teachers were expected to be the most satisfied with IEQ. Additionally, the IEQ satisfaction level of teachers in the renovated schools were expected to be similar or second to teachers in the new schools given the substantial IEQ-related renovations implemented in them, while teachers in the non-renovated schools were expected to be the least satisfied. Consequently, this research aimed to test the hypothesis that teachers’ IEQ satisfaction and their well-being were statistically significantly higher in the new and renovated schools than in the non-renovated ones. It also aimed to test the hypothesis that teachers’ IEQ satisfaction and well-being in the new and renovated schools were not statistically significantly different.

Study sample and online survey

The 852 teachers in all the 32 schools were invited to participate in this study. The online survey was divided into four modules. Module 1 included the adapted IEQ satisfaction survey and demographic information. The core module of the original NRC survey³³ included 18 items for assessing ventilation and thermal comfort, lighting and acoustics and privacy. In adapting this survey, the wording of items was slightly modified to suit school teachers and school environments, with an additional item included for daylight thus increasing the number of items from 18 to 19. Each of the 19 items was assessed using a seven-point scale from ‘Very Unsatisfactory’ (−3) to ‘Very Satisfactory’ (+3), like the NRC survey. Modules 2, 3 and 4 consisted of the newly developed TPWB, TSWB and TPhWB surveys, respectively, for assessing those three subdivisions of teachers’ personal well-being principally in school environments. Items of well-being surveys were assessed using a six-point scale from ‘Strongly Disagree’ (1) to ‘Strongly Agree’ (6). Most respondents completed the survey within 15 to 30 min. A total of 234 responses was recorded, resulting in a response rate of 27.5%. The survey was administered online using Qualtrics during the winter of 2016. Each respondent was assigned two modules: module 1 and an additional module assigned randomly.

Refining of occupants’ satisfaction and well-being surveys

The adapted IEQ satisfaction survey and three newly developed well-being surveys were refined through factor analysis to extract the significant factors that defined their structure. Additionally, these three well-being surveys were validated by correlating them to their corresponding existing context-free validated well-being surveys. Details of factors, final surveys and their validation are given in Sadick and Issa.³⁴ The final IEQ satisfaction survey consisted of 12 items and three factors namely ‘Lighting’ (L_sat), ‘Ventilation and thermal comfort’ (VTC_sat) and ‘Acoustics and privacy’ (AP_sat). The TPWB survey consisted of 12 items and two factors namely ‘Job satisfaction and professionalism’ and ‘Feeling of accomplishment and recognition’. The TSWB survey consisted of 12 items and three factors namely ‘Social Functioning in School’, ‘Social Connectedness in School’ and ‘Classroom Socialization’. The TPhWB survey also included 11 items and two factors namely ‘Role and Work Environment Outcomes’ and ‘Physical Functioning in Class’.

Data analysis

The scores for each of the three factors of IEQ were calculated by averaging ratings of items making up each factor. The factor scores were subsequently averaged to reach the overall IEQ satisfaction scores. The scores for TPWB, TSWB and TPhWB were each calculated by averaging ratings of multiple items making up each survey. The IEQ satisfaction ratings were recoded from their original scale of ‘Very Unsatisfactory’ (−3) to ‘Very Satisfactory’ (+3) to ‘Very Unsatisfactory’ (1) to ‘Very Satisfactory’ (7) for consistency and ease of comparison with the well-being ratings. Furthermore, the IEQ satisfaction ratings were categorized into three zones: the ‘dissatisfaction zone’ (1.00–3.49), the ‘neutral zone’ (3.50–4.49) and the ‘satisfaction zone’ (4.50–7.00). Similarly, the well-being perception ratings were categorized into three perception levels: a ‘low level’ (1.00–2.49), a ‘medium level’ (2.50–4.49) and a ‘high level’ (4.50–6.00). These categorisations did not involve recoding the IEQ satisfaction and well-being ratings. Outliers were determined using standardized ratings with an absolute value of 3.29 as cut-off.³⁵

The differences in the mean ranking of teachers’ IEQ satisfaction and well-being levels between the three strata of schools were assessed using the Kruskal–Wallis H test because these ratings were not normally distributed based on the Shapiro-Wilk test. Statistically significant differences were further investigated using pairwise comparisons with Bonferroni correction for multiple comparisons to determine the pairs of school strata that were statistically significantly different. The tests for the well-being levels were all statistically insignificant hence further analyses of differences were conducted based on the gender and age of respondents using either the Kruskal–Wallis H test or the Mann–Whitney U test. The five age groups shown in Table 6 were divided in two (i.e. 18–39 and 40–60+) for this analysis. Effect sizes were calculated to determine the magnitude of these differences in IEQ satisfaction and well-being for all statistically significant pairwise comparisons.³⁶ The Spearman rho correlation coefficients (ρ) were calculated to determine the level of association between the three IEQ satisfaction factors and three subdivisions of well-being.

Table 6.

Demographic information.

	School strata
	Stratum 1 (new)	Stratum 2 (renovated)	Stratum 3 (non- renovated)
School division
Division 1	24	59	14
Division 2	28	15	37
Gender
Female	38	57	41
Male	14	17	10
Age
18–29	6	10	8
30–39	16	31	23
40–49	12	20	9
50–59	16	13	10
60 or over	2	0	0
Number of years in current school
Less than a year	5	12	7
1–2 years	5	9	3
3–5 years	23	15	17
More than 5 years	19	38	24
Class time per week
Less than 10 h	7	9	1
10–20 h	3	6	4
11–30 h	7	7	7
More than 30 h	33	51	39

Results

The section starts with the demographic information related to the survey, then proceeds to describe the results on the differences in IEQ satisfaction and well-being between the three strata of schools.

Demographic information

Table 6 presents the demographic information of respondents included in the analysis.

Table 6 shows that the number of responses from the renovated schools exceeded those in the new and non-renovated schools. This was expected given that 17 of the 32 schools were renovated schools. The age of respondents was fairly distributed from 18 to 59 years old, with most of them having worked in their current school for 3 or more years, and spent on average more than 30 h per week in class. The number of female respondents was more than twice the number of males but fits the general observations made during site visits to 15 of the 32 schools.

Stratum level satisfaction with overall IEQ and its factors

Figure 1 depicts the distribution of overall IEQ satisfaction ratings for teachers in the three strata of schools. Teachers in the new and renovated schools were the most satisfied with overall IEQ, with about 78.85% and 71.62% of respondents respectively in the satisfaction zone. For the non-renovated schools, only 47.06% of respondents were satisfied. The most notable observation in Figure 1 is the relatively low median and mean level of overall IEQ satisfaction for teachers in the non-renovated schools.

Figure 1.

Stratum level distribution of overall IEQ satisfaction ratings.

The histogram in Figure 2 depicts respondents’ satisfaction with the three factors of IEQ. The figure shows that the satisfaction levels with the three IEQ factors of teachers in the new and renovated schools were all above 60%, while those of teachers in the non-renovated schools were all below 60%. The proportion of teachers satisfied with the L_sat factor was substantially higher in both the new and renovated schools than in the non-renovated schools. The items assessed as part of the L_sat factor included the amount of daylight and access to a view of outside. The respective mean rating of these two items were 5.34, 5.19 for new schools; 4.78, 5.05 for renovated schools; and 3.90, 3.76 for non-renovated schools. These mean ratings show that most teachers in the non-renovated schools were dissatisfied with these two items. This is not surprising given that most of these schools did not undergo major renovations within the last 10 years and thus probably have glazed windows with low light transmittance.

Figure 2.

Stratum level teachers’ satisfaction with IEQ factors.

As shown in Figure 2, teachers in the non-renovated schools were the least satisfied with AP_sat; however, the proportion of teachers dissatisfied with AP_sat (i.e. 19.61%) was lower than those dissatisfied with L_sat (i.e. 39.22%). The main source of dissatisfaction with AP_sat in the non-renovated schools was background noise with a mean rating of 4.19 and was probably caused by the noise emanating from the ageing HVAC systems in these schools. Figure 2 also shows that VTC_sat was the factor with the lowest ratings in all three strata of schools, and with the most telling difference in satisfaction levels. Although teachers in the new and renovated schools were exceedingly more satisfied with VTC_sat than those in the non-renovated, none of the strata achieved the American Society of Heating, Refrigeration and Air-conditioning recommendation of 80% satisfaction level.³⁷ As shown in Figure 2, the percentage of teachers dissatisfied with VTC_sat was the highest of the three IEQ satisfaction factors in each school strata, with the non-renovated schools recording the highest dissatisfaction level. Figure 3 depicts the mean rating for the four items making up the VTC_sat factor.

Figure 3.

Stratum level mean teacher ratings of VTC_sat items.

Figure 3 shows that teachers’ ratings of these four items in the non-renovated schools were generally low, with the mean rating of three out of the four items falling in the lower region of neutral. In contrast, the mean rating of three of the four items for teachers in the new and renovated schools fell in the lower region of the satisfaction zone, thus explaining the general low levels of satisfaction with VTC_sat in comparison to the other IEQ satisfaction factors. In general, respondents’ limited ability to control VTC_sat parameters was the root cause of their dissatisfaction. The low ratings for occupants’ control of their thermal environment across the three strata of schools was not surprising given that air temperature in almost all the schools was controlled remotely or onsite by the school divisions rather than by teachers themselves.

Stratum level differences between IEQ satisfaction factors

The Kruskal–Wallis H test showed statistically significant differences in the mean ranks of IEQ satisfaction factors across the three strata of schools. First, the mean ranks for L_sat were 102.45 (n = 52), 95.01 (n = 74) and 66.53 (n = 51) for the new, renovated and non-renovated schools, respectively, and were statistically significantly different, H(2) = 14.43, p = 0.001. Second, the mean ranks for VTC_sat were 97.85 (n = 52), 100.33 (n = 74) and 63.54 (n = 51) for the new, renovated and non-renovated schools, respectively, and were statistically significantly different, H(2) = 17.82, p = 0.0005. Lastly, the mean ranks for AP_sat were 97.88 (n = 52), 94.73 (n = 74) and 71.63 (n = 51) for the new, renovated and non-renovated schools, respectively, and were statistically significantly different, H(2) = 8.40, p = 0.015.

The post hoc pairwise comparisons with Bonferroni correction for multiple comparisons are presented in Table 7, along with their effect sizes. The table shows statistically significant differences between the new and non-renovated schools and between the renovated and non-renovated schools for the three factors of IEQ satisfaction. None of these differences between the new and renovated schools were statistically significant.

Table 7.

Post hoc pairwise comparison of IEQ satisfaction variables.

Pairwise comparisons	Test statistics	Adjusted significance	Effect size
L_sat
New-non-renovated	35.893	0.001*	0.344
Renovated-non-renovated	28.455	0.007*	0.277
New-renovated	7.438	1.000	0.076
VTC_sat
New-non-renovated	34.307	0.002*	0.327
Renovated-non-renovated	36.792	0.0005*	0.359
New-renovated	−2.485	1.000	0.019
AP_sat
New-non-renovated	26.257	0.027*	0.255
Renovated-non-renovated	23.102	0.039*	0.223
New-renovated	3.155	1.000	0.031

*Denotes significant correlation at level 0.05.

The effect sizes for all pairwise comparisons showed that the largest differences in teachers’ IEQ satisfaction were in VTC_sat, followed by L_sat and finally AP_sat. The effect sizes of VTC_sat between both the new and non-renovated schools and the renovated and non-renovated schools suggest medium to high differences in satisfaction levels. Furthermore, the effect size of L_sat suggests a medium to high difference between the new and non-renovated schools and a low to medium difference between the renovated and non-renovated schools. Lastly, the effect sizes of the two statistically significant differences in AP_sat between the new and non-renovated schools and between the renovated and non-renovated schools suggest medium differences in satisfaction levels.

Stratum level perception of the subdivisions of personal well-being

Figure 4 presents the breakdown of teachers’ perception levels for the three subdivisions of well-being. Except for TPhWB in the new schools, teachers in all three strata of schools rated their well-being as high or medium. Figure 4 shows that the highest rating in each school stratum was for TSWB with a mean rating of 5.22, 5.09 and 4.96 in the renovated, non-renovated and new schools, respectively. This contradicted the consistently high IEQ rating in the new and renovated schools relative to the non-renovated. The TPWB was the second highest rated subdivision of well-being in each stratum with a mean rating 4.79, 4.74 and 4.67 in the renovated, new and non-renovated schools, respectively. TPhWB was the lowest rated subdivision in each stratum with a mean rating of 4.53, 4.07 and 4.00 in the renovated, new and non-renovated schools, respectively.

Figure 4.

Stratum level teachers’ perception of the three subdivisions of personal well-being.

Figure 5 shows the mean rating of factors of TPWB, TSWB and TPhWB for the three school strata. As shown, the mean rating of the three factors of TSWB for each school stratum were in the high level zone, thus explaining the general high mean rating for TSWB. The ‘Role and work environment outcomes’ factor of the TPhWB was the least rated of all the well-being factors and explains why TPhWB was the lowest rated subdivision of well-being. The items of the factor ‘Role and work environment outcomes’ described mainly the negative consequences of teachers working in a school’s indoor environment including ‘I frequently get running nose, dry throat and lips, or watery eyes in school’. The low ratings for this TPhWB factor across all school strata corresponded with the low rating of the VTC_sat factor in this study: a result that is in line with those of previous IEQ studies.^38,39 Examination of both Figures 4 and 5 shows marginal differences in well-being perception levels between the three strata of schools.

Figure 5.

Stratum level mean rating of factors of personal well-being subdivisions.

Stratum level differences between subdivisions of individual well-being

The Kruskal–Wallis H test showed no statistically significant differences between TPWB, TSWB and TPhWB across the three school strata. Therefore, between strata and within strata analyses were conducted to determine if the two individual characteristics of teachers’ gender and age influenced their well-being ratings. The Kruskal–Wallis H test between the three strata for TPWB and TSWB showed no statistically significant difference for gender and the two age groups. For TPhWB, the test showed a statistically significant difference between females and the 40–60+ age groups. First, the mean rank for females were 18.55 (n = 8), 28.62 (n = 17) and 18.72 (n = 16) for the new, renovated and non-renovated schools, respectively. Lastly, the mean rank for the 40–60+ age groups were 10.42 (n = 6), 17.71 (n = 12) and 9.50 (n = 8), respectively, for the new, renovated and non-renovated schools, respectively. Pairwise comparison with Bonferroni correction for multiple comparisons showed that none of the pairs of strata were statistically significantly different for both females and the two age groups. The Mann–Whitney U test within each stratum for TSWB and TPhWB showed no statistically significant difference for gender and the two age groups. Additionally, the test showed no statistically significant difference in TPWB within each stratum for gender and the two age groups except in the non-renovated schools, where TPWB mean ranks were significantly different between females (mean = 4.86) and males (mean = 4.13), U = 6.00, z = − 2.10, p = 0.04.

Associations between well-being and IEQ satisfaction factors

Table 8 shows that VTC_sat was not statistically significantly associated with TPWB; however, it had the highest statistically significant moderate positive associations with TSWB and TPhWB. AP_sat had the second highest statistically significant moderate associations with both TSWB and TPhWB, whereas L_sat was moderately associated with TPhWB only.

Table 8.

Correlations between IEQ variables and well-being measures.

New teachers’ well-being measures	Correlation coefficient (ρ) and significance level (p)
New teachers’ well-being measures	L_sat	VTC_sat	AP_sat
Teachers’ psychological well-being (TPWB)	0.199 (0.125)	0.230 (0.075)	0.239 (0.064)
Teachers’ social well-being (TSWB)	0.207 (0.106)	0.485** (0.000)	0.351** (0.005)
Teachers’ physical well-being (TPhWB)	0.348** (0.010)	0.582** (0.000)	0.383** (0.004)

The values in parenthesis denote level of significance for correlation coefficients.

Denotes significant correlation at level 0.01.

Discussion

The results of this research proved the hypothesis that teachers’ satisfaction with IEQ was statistically significantly higher in the new and renovated schools than in the non-renovated schools. It also proved the hypothesis that the difference in teachers’ satisfaction with IEQ in the new versus the renovated schools was statistically insignificant. As building codes improve over the years, newer versions are generally expected to improve the overall performance of buildings, which should translate to more comfortable, safer and healthier indoor environments for occupants. Hence, the new schools were expected to have the best indoor environmental conditions and thus the most satisfied teachers given that they followed recent versions of the NBC relative to the renovated and non-renovated schools. However, the insignificant difference between teachers’ IEQ satisfaction in the new and renovated schools suggests that the IEQ-related renovations of the latter probably improved their IEQ performance to the level of the IEQ performance of new schools. This was evident in the pattern of IEQ satisfaction in both strata of schools, where for example both groups of teachers were concurrently satisfied and dissatisfied with items of VTC_sat.

Given that the renovated and non-renovated schools were similar (i.e. mix of old and middle age schools), the statistically significant difference between their teachers’ IEQ satisfaction can be attributed to the IEQ-related renovations of the renovated schools. These renovations involved tightening building envelopes with insulation to minimize thermal loses and enhance thermal comfort. While this strategy would have reduced both the exfiltration of indoor air contaminants and infiltration of outdoor air, the refurbishing or replacing of existing HVAC systems in the renovated schools would have increased the supply of outdoor air for ventilation, reducing the concentration of indoor air contaminants. Hence, the synergy between both strategies probably led to significantly higher VTC_sat for teachers in the renovated schools than the non-renovated schools. Similar propositions can also be used to explain the differences in both L_sat and AP_sat between teachers in renovated and non-renovated schools. These findings are consistent with the ones in the literature^11,12,24 regarding the likely impact of IEQ-related renovations on occupants IEQ satisfaction. This study did not include objective parameters of IEQ such as air temperature, reverberation time and carbon dioxide concentration. Hence, future studies should include these objective parameters to determine the potential direct impact of the IEQ-related renovations on indoor environments of the renovated schools.

The research also proved the hypothesis that the difference in teachers’ well-being in the new versus the renovated school was statistically insignificant but disproved the hypothesis that teachers’ well-being was statistically significantly higher in the renovated schools than in the non-renovated. This suggested that IEQ-related renovations did not lead to noteworthy improvements in the well-being perception of teachers in the renovated schools like they did for their IEQ satisfaction. It may also be that the renovated schools experienced substantial improvement in well-being without that improvement leading to statistically significant differences because the physical and IEQ conditions of the renovated schools, and their teachers’ well-being were worse prior to the renovations than the non-renovated ones.

Further analyses suggested that gender and age of teachers influenced their well-being perception. TPhWB scores for both females and teachers within the 40–60+ age bracket was statistically significantly different across the three strata of schools. Although the post hoc analysis showed no significant differences between pairs of school strata, this suggests that the impact of IEQ-related renovations on teachers’ well-being is probably highly dependent on personal characteristics such as gender, age and other characteristics not included in this study. The results also showed that female teachers’ perception of TPWB was statistically significantly higher than male teachers in the non-renovated schools, but not in the new and renovated schools.

The study results also showed statistically significant weak to medium associations of TSWB and TPhWB with the three IEQ variables. Although the significant correlations do not imply causation, one potential explanation is that the IEQ-related renovations may have enhanced IEQ satisfaction and subsequently enhanced teachers’ TPhWB and TSWB perceptions. The insignificant associations between TPWB and IEQ variables do not imply a lack of relationship as these may become statistically significant with a larger sample size. They may also be due to TPWB being the least physically observable of the three well-being subdivisions.

With respect to IEQ satisfaction, the effect sizes in this study showed that the magnitude of the observed difference in VTC_sat in renovated versus non-renovated schools was the largest, followed by L_sat and AP_sat. Hence, based on results of this research, IEQ-related renovations are likely to have the most positive impact on VTC_sat followed by L_sat and AP_sat. Although AP_sat significantly correlated with TSWB and TPhWB while L_sat significantly correlated with TPhWB only, L_sat was ranked second to VTC_sat not based on its effect size but because of the reported biological benefits of light generally, and particularly its impact on learning in schools. Bright light, both natural and artificial, synchronizes the internal body clock with the daily day and night routine so that people are awake during the day and sleep at night.⁴⁰ Hence, bright light, especially daylight exposure during the day contributes to alertness during daytime and quality sleep at night.⁴⁰ For schools, in particular, Tanner⁴¹ found a significant positive relationship between daylighting and students’ academic achievement.

The results of this study can be used by school divisions as part of their maintenance decision-making process. Nielsen et al.⁴² proposed goal setting as the first stage of developing a sustainable renovation decision-making tool. This involved defining clients’ sustainability objectives, criteria and appropriate weightings. School divisions looking to renovate their schools can define objectives of their renovation interventions in terms of specific improvements they would like to see in teachers’ IEQ satisfaction and well-being. They can use the study’s IEQ and well-being surveys to establish suitable benchmarks for IEQ satisfaction and well-being. They can also use effect sizes for the differences in each IEQ factor between the renovated and non-renovated schools as the respective weighting for prioritization. These effect sizes may be reassessed for prioritization purposes as indoor conditions improve. Nielsen et al.⁴² also proposed the registration and ranking of buildings as part of the sustainable renovation decision-making tool. Registration involves collecting information on the renovation needs of buildings. With this information, school divisions can categorize each defect in each school building into three based on the IEQ factor that is expected to improve when a defect is repaired. This should allow them to prioritize the maintenance needs of each school, and rank each school based on those needs as part of their maintenance decision-making process.

Conclusions and recommendations

This research investigated teachers’ IEQ satisfaction and well-being in new, renovated and non-renovated schools. The results suggested that IEQ-related renovations in existing schools can be an effective strategy for improving teachers’ IEQ satisfaction and probably their well-being. Considering that the renovated and non-renovated school strata were each composed of both old and middle age schools, the statistically significant differences in teachers’ IEQ satisfaction between the two strata of schools can be attributed to the IEQ-related renovations in the renovated schools. This assertion is further supported by the statistically insignificant difference in teachers’ IEQ satisfaction between the renovated and new schools probably due to the improved indoor conditions in the former because of the IEQ-related renovations they underwent. However, this study suggested that IEQ-related renovations would require planned integration of different strategies to yield the desired IEQ satisfaction outcome. For example, the likely synergy between reinsulating building envelopes and refurbishing or replacing existing HVAC systems in the renovated schools seem to account for the statistically significantly higher VTC_sat for teachers in the renovated schools in comparison with the non-renovated schools. Although this study excluded IEQ field measurements of these different schools, the statistically significant IEQ satisfaction differences between the renovated and non-renovated schools, as well as the statistically insignificant differences between the renovated and new schools seems to affirm the effectiveness of the IEQ-related renovations.

Teachers well-being perception ratings were not statistically significantly different across the three strata of schools. The statistically insignificant difference in well-being perception and the statistically significant difference in IEQ satisfaction between teachers’ in the renovated and non-renovated schools suggested that well-being perception and IEQ satisfaction were not similarly affected by the IEQ-related renovations. However, our results suggested that personal characteristics such as gender and age may be essential in determining the impact of IEQ-related renovations on occupants’ well-being. The test of associations also suggested a plausible indirect impact of school’s IEQ-related renovations on teachers’ well-being via improved IEQ satisfaction. The most promising indirect positive impact is likely that of IEQ-centred renovations on TPhWB via improvement of their VTC_sat. These results confirm that occupants’ IEQ satisfaction and their well-being are probably not the same and should therefore be assessed separately.

This study used post-renovation data to investigate the impact of IEQ-related renovations on the IEQ satisfaction and well-being of teachers in renovated schools relative to teachers in new and non-renovated schools. The lack of pre-renovation data implies that the actual improvement in IEQ satisfaction and well-being of teachers in the renovated schools were not accounted for in this study. Hence, future studies should adopt a pre-post research design that would account more accurately for actual improvements in IEQ satisfaction and well-being due to specific renovation interventions. Not only was the population for this study limited, but each respondent was assigned the IEQ satisfaction survey in addition to only one of the three well-being surveys deployed as part of this study. This deployment strategy shortened the completion time of the online survey; however, it resulted in fewer respondents completing each of the three well-being surveys, which may explain most of the statistically insignificant differences in well-being perception between the three strata of schools. Given that the research used an adapted and refined teachers’ IEQ satisfaction survey, and newly developed and refined well-being surveys, future research should deploy these refined surveys to a larger population of teachers to confirm their structural validity and enable the generalisation of their results.

Footnotes

Authors’ contributions

All authors contributed equally in the preparation of this manuscript.

Acknowledgement

This research would not have been possible without the support of the Government of Manitoba Public School Finance Board, the two School Divisions, and the 32 schools. The authors are particularly grateful to all teachers who participated in the online survey.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada.

References

Langevine R, Allouche M and AbouRizk S. Decision support tool for the maintenance management of buildings. In: Joint international conference on computing and decision making in civil and building engineering, Montréal, Canada, 14–16 June 2006, ASCE, ICCCBE, DMUCE, CIB-W78 and CIB-W102, pp.2292–3301, 2006.

Mendell

Heath

. Do indoor pollutants and thermal conditions in schools influence student performance? A critical review of the literature. Indoor Air 2005; 15: 27–52.

Gaudreault

Overton

Trstenjak

. Age of education infrastructure: recent trends. Investment and Capital Stock Division. Report no. 11-621-M, no. 81, Ottawa: Statistics Canada, 2009.

Statistics Canada. Capital and repair expenditures by North American Industry Classification System (NAICS), Canada, http://www5.statcan.gc.ca/cansim/a26 (2016, accessed 24 October 2016).

Exner D, D’Alonzo V, Paoletti G, Pascual R and Pernetti R. Building-Stock Analysis for the Definition of an Energy Renovation Scenario on the Urban Scale. In: SSPCR 2015, the 1st International Conference on Smart and Sustainable Planning for Cities and Regions, Bolzano, Italy, November 19–20, 2015, Springer, pp. 33–54, 2017.

Ott

Bolliger

. Pitfalls in the Economic and Ecological Evaluation of Energy Related Building Renovation Strategies and Measures. Energy Procedia 2015; 78: 2340–2345.

Angela

Abbas

David

. Linking energy and maintenance management for sustainability through three American case studies. Facilities 2011; 29: 243–254.

Liu

Rohdin

Moshfegh

. Evaluating indoor environment of a retrofitted multi-family building with improved energy performance in Sweden. Energy and Buildings 2015; 102: 32–44.

Wells

Berges

Metcalf

Kinsella

Foreman

Dearborn

Greenberg

. Indoor air quality and occupant comfort in homes with deep versus conventional energy efficiency renovations. Building and Environment 2015; 93, Part 2: 331–338.

10.

Noris

Adamkiewicz

Delp

Hotchi

Russell

Singer

Spears

Vermeer

Fisk

. Indoor environmental quality benefits of apartment energy retrofits. Building and Environment 2013; 68: 170–178.

11.

Ghita

Catalina

. Energy efficiency versus indoor environmental quality in different Romanian countryside schools. Energy and Buildings 2015; 92: 140–154.

12.

Thomsen

Rose

Mørck

Jensen

SØ

Østergaard

Knudsen

Bergsøe

. Energy consumption and indoor climate in a residential building before and after comprehensive energy retrofitting. Energy and Buildings 2016; 123: 8–16.

13.

Földváry

Bukovianska

Petráš

. Analysis of energy performance and indoor climate conditions of the Slovak Housing Stock before and after its renovation. Energy Procedia 2015; 78: 2184–2189.

14.

Hongisto

Haapakangas

Varjo

Helenius

Koskela

. Refurbishment of an open-plan office - Environmental and job satisfaction. Journal of Environmental Psychology 2016; 45: 176–191.

15.

Singh

Attia

Mahapatra

Teller

. Assessment of thermal comfort in existing pre-1945 residential building stock. Energy 2016; 98: 122–134.

16.

Allen

Marples

. Notched-noise and quiet detection thresholds as early indicators of auditory system damage in school-aged children. Canadian Acoustics - Acoustique Canadienne 1998; 26: 62–63.

17.

Bradley

Sato

. The intelligibility of speech in elementary school classrooms. The Journal of the Acoustical Society of America 2008; 123: 2078–2086.

18.

Toyinbo

Shaughnessy

Turunen

Putus

Metsämuuronen

Kurnitski

Haverinen-Shaughnessy

. Building characteristics, indoor environmental quality, and mathematics achievement in Finnish elementary schools. Building and Environment 2016; 104: 114–121.

19.

Buckley

Schneider

Shang

. Fix it and they might stay: School facility quality and teacher retention in Washington, DC. The Teachers College Record 2005; 107: 1107–1123.

20.

Harris

Sass

. Skills, productivity and the evaluation of teacher performance. Economics of Education Review 2014; 40: 183–204.

21.

Eweda

Zayed

Alkass

. Space-based condition assessment model for buildings: Case study of educational buildings. Journal of Performance of Constructed Facilities 2015; 29: 04014032–04014032.

22.

Zinzi

Agnoli

Battistini

Bernabini

. Deep energy retrofit of the TM Plauto School in Italy—A five years experience. Energy and Buildings 2016; 126: 239–251.

23.

Almeida

RMSF

de Freitas

. Indoor environmental quality of classrooms in Southern European climate. Energy and Buildings 2014; 81: 127–140.

24.

Schreuder

van Heel

Goedhart

Dusseldorp

Schraagen

Burdorf

. Effects of newly designed hospital buildings on staff perceptions: A pre-post study to validate design decisions. Health Environments Research and Design Journal 2015; 8: 77–97.

25.

Singh

Mahapatra

Teller

. Relation between indoor thermal environment and renovation in liege residential buildings. Thermal Science 2014; 18: 889–902.

26.

Salleh

Kamaruzzaman

Riley

Ahmad Zawawi

Sulaiman

. A quantitative evaluation of indoor environmental quality in refurbished kindergarten buildings: A Malaysian case study. Building and Environment 2015; 94: 723–733.

27.

Bluyssen

Aries

van Dommelen

. Comfort of workers in office buildings: The European HOPE project. Building and Environment 2011; 46: 280–288.

28.

Turunen

Toyinbo

Putus

Nevalainen

Shaughnessy

Haverinen-Shaughnessy

. Indoor environmental quality in school buildings, and the health and wellbeing of students. International Journal of Hygiene and Environmental Health 2014; 217: 733–739.

29.

Vieira

EMdA

Silva

LBd

Souza

ELd

. The influence of the workplace indoor environmental quality on the incidence of psychological and physical symptoms in intensive care units. Building and Environment 2016; 109: 12–24.

30.

Dodge

Daly

Huyton

Sanders

. The challenge of defining wellbeing. International Journal of Wellbeing 2012; 2: 3–3.

31.

Ryff

. Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology 1989; 57: 1069–1069.

32.

Hays

Morales

. The RAND-36 measure of health-related quality of life. Annals of Medicine 2001; 33: 350–357.

33.

Veitch

Charles

Farley

KMJ

Newsham

. A model of satisfaction with open-plan office conditions: COPE field findings. Journal of Environmental Psychology 2007; 27: 177–189.

34.

Sadick

A-M

Issa

. Occupants’ indoor environmental quality satisfaction factors as measures of school teachers’ well-being. Building and Environment 2017; 119: 99–109.

35.

Tabachnick

Fidell

. Using multivariate statistics, 4th ed. Boston: Allyn & Bacon, 2001.

36.

Field

. Discovering statistics using SPSS statistics, 2nd ed. London: Sage Publication, 2005.

37.

ASHRAE 55:2010. Thermal environmental conditons for human occupancy, Atlanta: the American Society of Heating, Refrigerating and Air-Conditioning Engineers, 2010.

38.

Witterseh

Wyon

Clausen

. The effects of moderate heat stress and open-plan office noise distraction on SBS symptoms and on the performance of office work. Indoor Air 2004; 14: 30–40.

39.

Madureira

Alvim-Ferraz

MCM

Rodrigues

Gonçalves

Azevedo

Pinto

Mayan

. Indoor air quality in schools and health symptoms among Portuguese teachers. Human and Ecological Risk Assessment: An International Journal 2009; 15: 159–169.

40.

van Bommel

WJM

. Non-visual biological effect of lighting and the practical meaning for lighting for work. Applied Ergonomics 2006; 37: 461–466.

41.

Tanner

. Explaining relationships among student outcomes and the school’s physical environment. Journal of Advanced Academics 2008; 19: 444–471.

42.

Nielsen

Jensen

Larsen

Nissen

. Early stage decision support for sustainable building renovation–A review. Building and Environment 2016; 103: 165–181.