**3. Method**

For the purpose of selecting study buildings for the research, a detailed evaluation of four recently constructed Green-Star rated academic buildings in Australia was carried out to observe their performance and management structure in reality. Four tertiary academic institutions were selected in Melbourne, Australia as case study buildings, incorporating post-occupancy evaluation techniques and appropriate project management strategies. The buildings were completed within the 2010–2014 period. The key features (Building Type, Academic Faculty, Green Star Certification Score, Volume, Area, and number of Floors) of the study buildings are described in **Table 1**. The buildings are denoted as Buildings A, B, C, and D. The results provide a model framework for design stakeholders to achieve desired building performance targets for green buildings, as they are designed to perform as per specific standards but can be applicable to conventional or non-green buildings as well, drawing upon lessons learned from this study.

The study has adopted a mixed-methods approach utilizing both quantitative and qualitative techniques. For quantitative data, survey research has been deployed. Surveys have been used to evaluate occupant satisfaction levels in the case study buildings.

Currently, Building User Survey (BUS) and Building Occupants Survey System Australia (BOSSA) are the only two officially accredited post-occupancy evaluation (POE) instruments within the National Australian Built Environment Rating System (NABERS) for commercial buildings used in Australia [34]. These surveys are robust and accessible Australian alternatives to other surveys currently in use by NABERS and


#### **Table 1.**

*Description of case study buildings.*

Green Star Performance rating tools [35]. However, BOSSA has wider applications, BUS surveys were used in this study due to the narrower scope, more-wider presence globally and acceptance of the respective survey, and the license rights available with or for the case study buildings [28, 34]. Apart from the BUS itself, an online survey deployed through Survey Monkey (cloud-based interface) was used by one of the case study buildings. Although the survey used was different for one building, their project managers allowed the researcher to add or modify survey questions as required before they were finally deployed. Hence, to ensure comparative study across all the buildings, their questionnaire was modified, and questions were added or removed to match the themes of the BUS format. So, four buildings used BUS and one building used an additional online survey as well as the standard BUS. BUS is available as hard copy survey and online; the buildings used the online survey rather than hard copy.

The surveys were used to evaluate the feedback of case study buildings, focusing on staff rather than staff and students. This was because staff spend more hours in a building either for teaching, research, or undertaking other professional activities. The BUS surveys used a 5-point Likert scale for measurement with 1 rating the highest dissatisfaction to 5 being the highest level of satisfaction. Statistical analysis of the occupant satisfaction survey data was carried out using SPSS Statistics (version 26) analysing via one-way ANOVA. Many studies [36–38] have proved that the Likert scale data can be analysed using parametric tests, as the analysis can vary based on how the responses are formulated. For example, in this research, the distance between each item category is considered "equivalent," and when a Likert scale is assumed to present symmetry of variables and is equidistant, it behaves more like an intervallevel measurement in practice, being suitable for parametric tests [37, 38]. In this study, the distance between each item category is "equivalent" and the Likert scale was used to run parametric tests on the data collected. The other key primary criteria were to assess if the data were normally distributed within all groups. Hence, the Descriptive and Levine's Test for Homogeneity of Variances was carried out for all variables [36, 37]. The output displayed normal distribution using the mean values as the measure of central tendency, that is, the probability distribution. Correspondingly, as the data represented a Gaussian population (normally distributed), one-way ANOVA was used to determine the statistical significance of data [36, 37].

In addition to the surveys, qualitative tools in the form of walk-in discussions, stakeholder interviews with the building facilities department and design stakeholders, and focus group discussions were undertaken. The walk-in discussions assessed occupant's (including building's staff and students) responses in a more informal

**Figure 1.** *Data collection approach.*

## *Highlighting the Design and Performance Gaps: Case Studies of University Buildings DOI: http://dx.doi.org/10.5772/intechopen.102779*

manner, followed by the outcomes from the stakeholder interviews, assessing the role and responses of management and finally the focus group results. The focus group discussions were conducted to share the preliminary results and discuss preliminary results of the actual versus expected outcomes with key management stakeholders. These qualitative results assist in understanding and validating the occupant survey findings and facility management perspectives concerning sustainable and green building concepts. The study hence adopted an "Action Research" framework for data analysis. The steps involved in the data collection are summarized in **Figure 1**.
