**4. Results**

#### *4.1. Survey Method*

In 2021, The BATM assessment was applied to an engineering company engaged in general contracting. Data were acquired using the questionnaire survey method, and 50 participants were selected for the survey from project staff who had worked in the company for more than 5 years. About half of the participants had used BIM software or participated in BIM training, and the other half were project managers and other management personnel who knew something about BIM but had no experience of operating BIM software. Their information is listed in Table 4.

**Table 4.** Participant information table.


After understanding the purpose of the survey, the participants carefully determined the PBM and PBA maturities of 37 desired outcomes according to the definition of maturity levels and their understanding of the actual PBM and PBA maturities in the company. Ultimately, a total of 49 valid questionnaires were acquired (in an unqualified questionnaire, some options were not answered).

#### *4.2. Statistical Analysis*

Content validity analysis: Because the BATM questionnaire was developed based on the PMBOK 7th edition, which is an acceptable standard, the researchers organized a pilot study to evaluate the internal validity. In this pilot study, researchers conducted comprehensive interviews with seven project personnel who had participated in at least two projects adopting BIM. Based on the positive assessment of these project managers, the conclusion obtained from them was that the content of the questionnaire was closely related to the PBM and PBA maturities, its structure was simple and clear, and its operability was appropriate.

Reliability analysis: The reliability analysis was conducted using Cronbach's alpha coefficient. According to Kim and Feldt [56], when the internal consistency coefficient of the data reaches 0.70 or higher, the data can be considered to have sufficient reliability. In the study, the item scale was internally consistent because all of Cronbach's alpha coefficients exceeded the threshold value (0.70) (Table 5).

**Table 5.** Descriptive statistics, Cronbach's alpha, and correlation coefficients.


Data analysis: To understand the data more effectively, we conducted a descriptive data analysis and a correlation analysis, and the results are listed in Table 5. The mean values reflect the maturity of each performance domain, in which the PBM maturities change from 2.15 to 2.71 and PBA maturities change from 1.29 to 1.79. Correlation analyses were conducted to verify the correlations between the PBM and PBA maturities. As can be seen from Table 5, the correlation coefficients in all performance domains are between 0.51 and 0.84; the *t*-test statistics are between 4.14 and 10.9, which are greater than ta/2 (α = 0.05; ta/2 = 2.008) and indicate clear correlations.

Table 6 shows the significant difference among different performance domain data and among different participant data, in which the results were obtained by the analysis of variance with two factors. The fact that all of the F-values were greater than the values of "F-crit," and all of the *p*-values were less than the significance level of 0.05, implies that statistically significant differences existed among the maturities of the eight performance domains and among the feedback of the 49 participants.

**Table 6.** Analysis of variance with two factors.

