The Link between Environment and Organizational Architecture for Decision-Making in Educational Institutions: A Systemic Approach

Abstract

Numerous organizations employ decision-making processes to support operational activities; however, decisions and mistakes can significantly impact Market Performance (MP) due to the oversight of organizational architecture and the environment. This becomes particularly critical in the realm of strategic management, where improper practices and a lack of management understanding can lead to substantial losses. Hence, a systemic investigation was undertaken to explore the repercussions of not adopting such an approach concerning organizational architecture and the environment. Employing a quantitative analysis via hierarchical regression involving Confirmatory Factor Analysis and Ordinary Least Squares, using data gathered from a survey encompassing 134 collaborators from Brazilian Federal Universities. The findings show that the organizational environment positively impacts decision-making, leading to better MP. Additionally, organizational architecture partially mediates the link between the organizational environment and decision-making. Remarkably, national literature lacked research combining Student Assistance Program (PNAES) actions with MP improvement to assess Brazilian Federal Universities’ effectiveness.

1. Introduction

The systemic method approaches a phenomenon by integrating aspects such as the worldview of those involved to trace improvement actions considering the relationship with the environment [1]. The systemic approach meets the need of organizations to develop models that enable greater speed in decision-making and integration of organizational areas [2].

It is impossible to think about the organization without considering the decision-making process, which occurs constantly [3]. The current context of uncertainty and competitiveness has highlighted decision-making processes as a systemic approach [4].

Organizational performance is a central construct in strategic management research, and while previous studies have established the multidimensionality of organizational performance, the relationships between performance and its dimensions remain unexplored [5]. Understanding the factors that impact organizational performance is one of the research challenges that still motivates many researchers and organizational managers [6]. The objective of organizations extends beyond mere profit maximization, aiming instead for sustainability achieved through ongoing growth and sustained organizational performance.

Measuring organizational performance can yield improved asset management, heightened organizational knowledge, and enhanced customer value. Moreover, the actions taken by an organization to enhance performance can significantly impact its reputation. The data generated by organizations are now systematically gathered and stored within databases, contributing to informed organizational decision-making [7]. Organizations are inserted in environments in constant mutation, for this reason, the adoption of strategies in order to review their processes continuously is highly recommended, aiming to maintain the efficient use of their resources [8].

This study aims to investigate the effects of integrating the principles of organizational environment and organizational architecture with decision-making within a systemic approach framework to enhance the market performance of Brazilian federal universities. The research employed a quantitative analysis using Confirmatory Factor Analysis (CFA) and Ordinary Least Squares (OLS) hierarchical regression based on survey responses from 134 employees, directors, and planning technicians.

To achieve the objective, the following steps were followed:

(i)

Literature review: An extensive literature review was conducted to understand the relationship between organizational environment, organizational architecture, decision-making, and market performance, as well as the benefits of a systematic approach.

(ii)

Development of the research instrument: Based on the literature review, a questionnaire was developed to measure the variables of interest, ensuring their validity and reliability.

(iii)

Data collection: The questionnaire was emailed to those responsible for planning at Brazilian federal universities. Responses were obtained from 134 participants, covering 31 federal universities from the five regions of Brazil.

(iv)

Data analysis: The data were analyzed using confirmatory factor analysis (CFA) and hierarchical ordinary least squares (OLS) regression, allowing the evaluation of relationships between variables and testing of study hypotheses.

(v)

Interpretation of results: The results showed that the integration of organizational environment, organizational architecture, and decision-making positively contributes to the market performance of Brazilian federal universities.

The results were promising regarding the systemic approach to integrating Organizational Environment, Decision Making, and Organizational Architecture, highlighting their contributions to the Market Performance of Brazilian Federal Universities.

2. Current Literature on Organizational Environment, Organizational Architecture, and Organizational Decision Making

In this section, we provide a literature overview of the intersection between Organizational Environment (OE), Organizational Architecture (OA), and Organizational Decision Making (OD) based on a comprehensive literature review spanning from 2012 to 2023.

For this literature research, we conducted searches in the Scopus and Web of Science databases. The search queries used were as follows: “Organizational Environment” AND “Organizational Architecture” AND “Organizational Decision Making” (resulting in 0 articles in both databases); “Organizational Environment” AND “Organizational Decision Making” (resulting in 6 articles in Scopus and 0 in Web of Science); “Organizational Architecture” AND “Organizational Decision Making” (resulting in 0 articles in both databases). Out of the 6 articles, only 3 were scientific articles. Hence, the total sample resulted in 3 articles. We thoroughly reviewed these 3 articles to assess their relevance to the research theme. The inclusion criterion focused on whether the research aimed to study and/or establish correlations between Organizational Environment and/or Organizational Architecture with Organizational Decision Making to contribute to improving Market Performance. Articles that solely addressed individual topics or did not intend to enhance Organizational Decision-Making through the integration of Organizational Environment and/or Organizational Architecture were excluded. Consequently,

Table 1. Current literature on OE and OA in OD. Source: Authors.

References	OE	OA	OD	Contribution
[9]	X		X	This article sought to analyze perceived organizational politics and abandonment plans through an examination of the buffering roles of relational and organizational resources.
[10]	X		X	This research sought to demonstrate how the different features of the garbage can model manifest themselves within organizations managing numbers.
[11]	X		X	This article identifies key factors driving the organizational adoption of social recruiting technologies, such as LinkedIn, Facebook, and Twitter.

provides a comprehensive overview of this literature research, highlighting the covered fields and contributions of the 3 selected articles.

Prior studies have explored the OE and OD relationship. Three Figure 1 articles link OE–OD integration to market performance but lack a systemic approach. No articles focus on OA aiding decision-making. Ref. [9] examines the link between employees’ beliefs, guided decision-making, and turnover. Ref. [10] analyzes garbage can metrics and organizational uncertainty. Ref. [11] assesses factors driving social media tech adoption in recruitment decisions. Studies highlight the need for theoretical frameworks linking OE–OA for OD support in Brazilian federal universities aiming at market performance. The current literature lacks a systemic OE–OA–OD approach, justifying this research.

3. Hypotheses Development

To bridge this gap, we have taken a systemic approach by integrating Organizational Environment (OE), Organizational Architecture (OA), and Organizational Decision Making (OD) with the aim of enhancing Market Performance (MP) in Brazilian federal universities.

In order to guide our study, we crafted a conceptual framework (see Figure 1) as a basis for forming hypotheses. Rooted in systems theory, this framework draws from multidisciplinary perspectives to examine how subsystems and agents mutually influence one another within a comprehensive macro-process [12]. Through this systemic lens, we aim to comprehensively understand the intricate dynamics and interdependence among practices, OE, OA, OD, and MP, contributing to existing knowledge.

Definition of Constructs

In this section, we define the key constructs used in our conceptual framework (see Figure 1) and explain their relationships.

• Organizational Environment (OE): The OE refers to the external factors and conditions that affect an organization’s operations and performance. It includes economic, demographic, cultural, ecological, and social aspects that influence how an organization functions within its environment.
• Organizational Architecture (OA): OA encompasses the structure, processes, and strategies that define how an organization operates internally. It includes aspects related to knowledge management, skill development, and the design of organizational processes and systems.
• Organizational Decision Making (OD): OD involves the process of making choices or selecting a course of action from among alternatives. It includes planning, information gathering, and the evaluation of alternatives to make informed decisions that align with organizational goals.
• Market Performance (MP): MP refers to the effectiveness and efficiency of an organization in achieving its objectives within its market or industry. It includes indicators such as student graduation rates, academic performance, student retention, and success rates, which reflect the overall performance of the university in the market.

In Figure 1, these constructs are depicted as interconnected, highlighting the interdependence and influence they have on each other. The arrows indicate the direction of influence, showing how changes in one construct can impact others.

Amidst fierce global competition, rising costs, and rapid technological advances, organizations adopt innovative business environments to thrive [13]. These environments foster continuous innovation as a survival necessity [14,15]. Prior studies highlight the potential of OE, OA, and Decision Making in relation to MP [16,17,18,19,20]. Within this context, we posit Hypothesis H1: Organizational Environment has a positive association with Decision-Making, leading educational organizations to improve their Market Performance.

The systemic approach optimizes processes and enhances competitiveness [21]. During Decision-Making phases, integrating external with internal knowledge significantly impacts MP and promotes organizational synergy [22]. SA principles bolster superior decision-making performance [23,24,25,26,27,28]. Thus, we investigate the positive association between Organizational Architecture and Decision-Making, impacting Market Performance. Hypothesis H2: Organizational Architecture has a positive association with Decision-Making, and leading organizations would obtain Market Performance improvement.

Next, we explore Decision Making’s role as a mediator between OE and Market Performance, characterized by continuous improvement and stakeholder engagement [25]. OE plays a vital role, in enhancing efficiency in marketing and process development, ultimately boosting Market Performance. We posit Hypothesis H3: Decision Making mediates the relationship between Organizational Environment and Market Performance leading to an improvement in MP.

Our subsequent hypothesis delves into the potential of OE and associated technologies to enhance Market Performance [29,30]. Simultaneously, OA technologies improve productivity, resource efficiency, and decision-making [31,32]. Hypothesis H4: Decision-Making mediates the relationship between Organizational Architecture and Market Performance leading to an improvement in MP.

4. Research Method

4.1. Sampling

For this study, a survey was carried out with those responsible for managing the planning of the Federal Universities of Brazil in Brazil (IFES). The research instrument, developed on the Google Forms^® platform, was sent by email to a total sample of 672 respondents in the 69 existing IFES. The result presented a sample of 134 respondents covering 31 IFES from the five regions of Brazil, from September to October 2022, resulting in a response rate of 19.9%. We acknowledge the limitations of convenience sampling, as it may introduce bias; however, to mitigate this risk, we employed randomization techniques during data analysis, as proposed by [33]. These techniques helped ensure the statistical robustness of our findings and reduce the likelihood of response bias.

Regarding sample size, we performed a power analysis to assess the statistical robustness of our model, employing the approach suggested by [34], focusing on the largest number of independent predictors. Setting the parameters to reflect an average effect size (f² = 0.15) and considering the total sample size of 134, we obtained a test power of 1 − β ≈ 0.80, in line with the recommendation of [34]. Thus, we can validate the feasibility of using an OLS approach with the proposed sample size.

The research instrument was divided into four groups of questions being subdivided into: 30 questions for Organizational Environment, 15 questions for Organizational Architecture, 07 questions for Decision Making, and 06 questions for MP. The questions utilized a 1 to 5 scale for assessment: 1—Minimal development, 2—Low development, 3—Moderate development, 4—Significant development, and 5—Consistent development. The evaluation scale ranged from 1—Very low, 2—Low, 3—Average, 4—High, to 5—Very high.

The control variables refer to the average number of students benefiting from National Student Assistance Program (PNAES) resources in the last 3 years in binary evaluations (0, 1): low investment and high investment. PNAES stands for the National Student Assistance Program of the Brazilian federal government, which aims to promote the inclusion and retention of low-income students in public higher education institutions. It offers various resources and benefits to help students overcome the financial and socioeconomic barriers that can hinder the completion of their courses [35].

When addressing missing and discrepant data, such as responses consisting of only a single scale, we meticulously examined all responses and weeded out those that did not comprehensively or adequately address all questions, as suggested by [7].

4.2. Survey Instrument

The questionnaire was developed based on constructs and reference articles on subjects consolidated in the literature. The constructs were: Organizational Environment [18,19,23,36,37], Organizational Architecture [38,39], Decision Making [40,41,42,43,44,45,46] and MP [47,48,49,50,51,52,53,54,55].

The Organizational Environment construct was subdivided into economic, demographic, cultural, ecological, and social practices. The organizational architecture construct was subdivided into knowledge and skill. The decision-making construct was not subdivided but contained seven questions. The MP construct was not subdivided, but contained six questions, and was also used as a dependent variable.

Table 2. Measurement Validation. Source: Authors.

Code	Questions	Factor Loading
01—Organizational Environment/Economic	Does the MHDI (Human Development Index of the Municipality) interfere with the socioeconomic vulnerability conditions of students?	0.65
	Is the university able to develop a region socio-economically?	0.43
	Does the Economic Policy of the Federal Government affect the performance of university teaching, research and extension?	0.47
	Does monetary stability influence the development of the university’s actions?	0.56
02—Organizational Environment/demographic	Does the income distribution of the population interfere with the socioeconomic vulnerability of students?	0.70
	Does the geographical distribution of the population, where the university is located, interfere with the profile of the students?	0.67
	The lower the income distribution of the population, where the university is inserted, the greater the need for inclusion actions and the permanence of students?	0.71
	The greater the social diversity of the population, where the university is inserted, the greater the need for policies of inclusion and permanence of students?	0.58
03—Organizational Environment/Cultural	Does the university allow staff to propose solutions?	0.71
	Does the university value teamwork?	0.86
	Does the university encourage the professional growth of staff?	0.67
04—Organizational Environment/Ecological	Does the university develop the rational use of natural resources?	0.77
	Does the university develop environmental management with cost savings?	0.82
	Does the university develop the 5R’s policy (Reuse, Recycle, Reduce, Reclaim, Rethink)?	0.82
	Does the university develop environmental management to improve the quality of life of staff and students?	0.84
	Does the university develop awareness among managers about socio-environmental responsibility?	0.83
05—Organizational Environment/Social	Does the university develop practices for the social inclusion of students with socioeconomic vulnerability?	0.46
	Does the university develop leadership practices with the training of qualified professionals?	0.61
	Does the university develop an environment of institutional trust with knowledge sharing among the university community?	0.79
	Does the university interconnect teaching, research and extension practices, favoring citizen education?	0.70
	Is the university an institution embedded in social practice, ensuring its inclusion at local, regional, national and international levels?	0.70
06—Organizational Architecture/Knowledge	Is the university an institution that serves society by providing knowledge?	0.53
	Does the university enable knowledge to be disseminated among staff?	0.39
	Does knowledge management enhance institutional results and improve the quality of processes?	0.63
07—Organizational Architecture/Skill	Does the social process at university demand the building of interpersonal relationships that meet people’s needs?	0.60
	Are skills in the academic context as a basic condition for generating learning and solving social demands?	0.65
	What is the degree of influence of the following skills practices on the organizational architecture of the university?	0.59
	Does interpersonal relationship represent the human behavior that generates trust and participation of people?	0.70
08—Decision-Making/Planning and information	At the university, does the decision-maker make an effort to collect complete information?	0.76
	Are the criteria for evaluating alternatives for decisions known and defined in advance at the university?	0.83
	Does the university adopt rules based on experiences with previously planned decisions?	0.83
	Are decisions made at the university structured and routine?	0.64
	Does the university manager have little information for decision making?	0.72
	Are the answers for decision making at the university not standardized?	0.51
	At university information is inaccurate, not expressed	0.92
09—Market Performance	The number of students graduating from undergraduate courses in the last 3 years?	0.83
	Academic performance in undergraduate courses in the last 3 years?	0.80
	The total number of students entering undergraduate courses in the last 3 years?	0.73
	The retention rate in undergraduate courses in the last 3 years?	0.52
	The success rate in undergraduate courses in the last 3 years?	0.76

presents the questions and factor loadings. Factor loadings, or loading factors, in Confirmatory Factor Analysis (CFA) represent the relationships between observed variables and latent factors. They indicate how much each observed variable contributes to explaining the respective latent factor. Factor loadings are calculated as the covariance between the observed variable and the latent factor divided by the variance of the latent factor. In simple terms, factor loadings represent the weight or importance of each observed variable in defining the latent factor. Values close to 1 indicate that the observed variable is strongly related to the latent factor, while values close to 0 indicate a weak relationship [55].

4.3. Variable Operationalisation, Reliability, and Validity of Measures

To analyze unidimensionality, a Confirmatory Factor Analysis (CFA) was utilized. Our model showed the goodness of fit as the reference values for comparative fit index (CFI), Root mean squared error of approximation (RMSEA), average variance extracted (AVE), composite reliability (CR), and Cronbach’s alpha fell in the acceptable values [7], as shown in

Table 3. CFA Metrics. Source: Authors.

Construct	AVE	CR	Alpha	RMSEA	CFI	TLI
Organizational Environment/Economic	0.29	0.60	0.59	0.063	0.912	0.896
Organizational Environment/Demographics	0.45	0.76	0.76
Organizational Environment/culture’s	0.56	0.79	0.78
Organizational Environment/Ecological	0.67	0.91	0.91
Organizational Environment/Socials	0.44	0.79	0.78
Organizational Architecture/Knowledge	0.27	0.52	0.56	0.079	0.952	0.909
Organizational Architecture/Skill	0.40	0.72	0.72
Decision-Making/Planning	0.59	0.85	0.85	0.083	0.972	0.952
Decision-Making/information	0.54	0.77	0.74
Market Performance	0.54	0.85	0.84	0.080	0.988	0.969

. To have consistency in the CFA, it is necessary to pay attention to the metrics: Root Mean Square Error of approximation (RMSEA), Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Average Variance Extracted (AVE), Cronbach’s Alpha, Composite Reliability (CR), and Factor Loading.

The mean square error of approximation is a parsimony-adjusted index, where values closer to zero represent good fits. The RMSEA value, normally used, varies between 0.03 and 0.08; thus, it is noted that the construct on Decision-Making/Planning and information is within the range, but the other constructs obtained values close to the range defined in the literature. The comparative fit index (CFI) compares the fit of a target model with the fit of an independent or null model. Values greater than 0.90 are recommended; thus, it was observed that the Organizational Environment and Organizational Architecture constructs are within the range, but the Organizational Environment and Organizational Architecture constructs present values close to the range indicated in the literature [56].

The unnormed fit index (TLI) is preferable for small samples and has reference values above 0.90 [57] or greater than 0.95 [58]. Therefore, all constructs showed adherence. Extracted mean variance (AVE) is a measure of the amount of variance that is captured by a construct relative to the amount of variance due to measurement error. The literature recommends an AVE of up to 0.5. However, when a multiconstruct measure is performed, the individual AVEs\thresholds may not present values above 0.5 [7].

Please note that all constructions are adherent. Cronbach’s Alpha is a measure of reliability that varies from 0 to 1, with values of 0.60 and 0.70 considered good levels of acceptance. Overall, the constructs showed values close to or above 0.60. It measures the correlation between the responses of a questionnaire by analyzing the answers given by the respondents, presenting an average correlation between the questions [7]. Therefore, it is noted that all construct items are above the lower limit. The Composite Reliability is a measure of the internal consistency of the scale items, as well as Cronbach’s Alpha [59]. Note that the CR value was appropriate for presenting values above 0.70.

Table 4. Bivariate Correlation Matrix. Source: Authors.

n°	Variables	Mean	SD	Skewness	Kurtosis	1	2	3	4	5	6	7	8	9	10	11
1	Students_Benefit_PNAES	0.366	0.483	0.564	−1.707	-
2	ECONOMIC	4.636	0.429	−2.288	6.939	0.012	-
3	DEMOGRAPHIC	4.485	0.618	−1.556	2.331	−0.101	0.474 **	-
4	CULTURAL	4.264	0.614	−1.346	3.106	0.138	0.219 *	0.060	-
5	ECOLOGY	2.825	0.672	−0.617	0.308	0.078	0.202 *	0.109	0.464 **	-
6	SOCIAL	3.393	0.469	−0.801	−0.081	0.098	0.218 *	0.133	0.350 **	0.536 **	-
7	KNOWLEDGE	4.249	0.515	−0.584	0.533	0.156	0.345 **	0.212 *	0.449 **	0.426 **	0.432 **	-
8	SKILL	4.026	0.543	−0.814	2.921	0.078	0.186 *	0.210 *	0.316 **	0.277 **	0.260 **	0.394 **	-
9	PLANNING	3.614	0.767	−0.452	0.273	0.120	0.084	0.019	0.510 **	0.529 **	0.405 **	0.552 **	0.283 **	-
10	INFORMATION	2.575	0.854	0.082	−0.618	−0.160	−0.091	0.115	−0.405 **	−0.392 **	−0.263 **	−0.442 **	−0.235 **	−0.539 **	-
11	PERFORMANCE	3.436	0.673	0.123	−0.11	0.241 **	0.048	−0.102	0.321 **	0.213 *	0.249 **	0.341 **	−0.007	0.314 **	−0.242 **	-

* indicates significance at p < 0.05. ** indicates significance at p < 0.01.

presents the correlation matrix between the independent variables of the model: Organizational Environment (2, 3, 4, 5, 6), Organizational Architecture (7, 8), Decision Making (9, 10), and Market Performance-MP (11) as a construct. MP is the response variable, it was not reported in the matrix. The control variable (Students_Benefit_PNAES) was included in the analysis to verify the existence of a relationship with the variables.

In the context of this study, the independent variables refer to the theoretical constructs that represent different aspects of the organization under investigation: Organizational Environment, Organizational Architecture, and Organizational Decision Making. These constructs were measured using specific questions in a questionnaire developed based on existing literature on the subject. For example, the Organizational Environment was assessed through questions related to the organization’s economic, demographic, cultural, ecological, and social context. Organizational Architecture was evaluated based on questions about organizational knowledge and skills, while Organizational Decision Making was measured through questions about the decision-making process within the organization.

Table 4 shows the correlation coefficients for the different variables. Significant relationships, that is, had p-values less than 0.05 or less than 0.01, were reported in Table 4.

Furthermore, in Table 4, descriptive statistics of the model were reported, using the mean and deviation techniques standard. Input values were unstandardized variable values. Data normality was tested by kurtosis and asymmetry techniques, also reported at the end of Table 4. The results suggest that our independent variables are normally distributed, as all values are between [−2.58, +2.58], which represents 0.01 significance [7], except for control variables 2 and 11; however, this cannot be a problem as they are control variables and not main variables of the model.

4.4. Response Bias

To analyze the consistency of the model, the Harman test was applied [60]. A post hoc Harman factor analysis is commonly used to see whether the variation in the data can be largely attributed to a single factor. Harman’s test is used to collect data for dependent and independent variables. An analysis is performed using EFA on all construct items to check for total variance. If the total variance extracted by a factor exceeds 50%, there is a common method bias in the study [7,33]. As a result, it was possible to notice that the extracted variance was not greater than 50% (42%), thus indicating that there is no inconsistency in the construct.

We also employed the marker variable technique to assess Common Method Variance (CMV). This technique involves including a variable in the survey that is theoretically unrelated to the substantive variables being measured in the model [61]. In this sense, we introduced the variable “occupational satisfaction” as a marker, using it as a proxy to assess respondents’ feelings about their occupation in the decision-making department. This marker variable was included in all models, and the results were compared to those obtained without the marker. Notably, the inclusion of the marker variable did not produce significant changes in the models, indicating that response bias is unlikely to be a concern in this dataset. Thus, we conclude that response bias is not a substantial problem in our findings.

4.5. Endogeneity and Robustness Checks

Addressing endogeneity in this context is crucial, as the presence of endogeneity would require the use of alternative methods such as two-stage least squares regression (2SLS). We identified potentially endogenous constructs, namely Organizational Environment and Organizational Architecture resource practices, as they can be influenced by Market Performance activities carried out by organizations. To address this concern, we extract potential instruments from the Brazilian Innovation Survey [62] that are associated with activities known to influence Market Performance resources and practices.

We had chosen “product and/or processes” and “improvement in product quality”. These variables were chosen based on the logic that companies that frequently engage in product and process innovation and prioritize product quality are more likely to develop Organizational Environment and Organizational Architecture features and implementation practices. In addition, they are not directly related to planning and information from decision-making. In our initial tests, the instrumental variables exhibited strong explanatory power, with p-values less than 0.01 and a minimum F-value of 22.23 (p = 0.000). Subsequently, we evaluated whether the explanatory variables should be treated as endogenous and instrumented, following the approach proposed by the 2SLS regression model. For that, we performed the Durbin test and the Wu-Hausman test, which showed values of 2.39 (p = 0.30) and 1.02 (p = 0.36), respectively.

These results indicate that our potential endogenous constructs (Organizational Environment and Organizational Architecture) are, in fact, exogenous, as all p-values are greater than 0.05. Therefore, based on the comprehensive results of our endogeneity tests, we conclude that the OLS procedure is appropriate for testing the hypothetical relationships in our study. Also, to ensure model consistency, we perform a robustness check on the model. We checked whether the results of our regression analysis could vary by (i) removing control variables and (ii) analyzing individual predictors. In the first approach, we remove control variables to see if our predictors are influenced by them.

We found stable results because they did not show significant changes in the coefficients of our model; moreover, all significance relationships remained the same without the presence of control variables. The second approach was contemplated by the individual analysis of the relationship of effects between each construct; in Table 4, we find consistency with our main results. Control variables did not show significance, overall, when compared to predictor variables. When comparing the predictor variables, we found a strong relationship of significance, in general, between them.

4.6. Data Analysis

We conducted a hierarchical OLS analysis to examine our hypotheses. To ensure consistency, we normalized the independent variables using a mean deviation Z score, assessing all relationships (see

Table 5. Results of Regression Analysis. Source: Authors.

Predictors	Organizational Architecture (AO)				Decision Making Organizational (OD)							Market Performance
	Knowledge		Skills		Planning			Information
	M1	M2	M1	M2	M1	M2	M3	M1	M2	M3	M4	M1	M2	M3	M4
BEN_PNAES	0.166 (p = 0.072)	0.102	0.088	0.054	0.191	0.058	0.003	−0.283 (p = 0.065)	−0.144	−0.088	0.277 (p = 0.005)	0.336 (p = 0.005)	0.249 (p = 0.031)	0.224 (p = 0.046)	0.224 (p = 0.048)
Economic		0.091 (p = 0.036)		0.009		−0.058	−0.107 (p = 0.074)		−0.051	−0.004			0.012	−0.020	−0.012
Demographic		0.041		0.092 (p = 0.070)		−0.01	−0.033		0.156 (p = 0.040)	0.184 (p = 0.013)			−0.087	−0.080	−0.078
Cultural		0.130 (p = 0.003)		0.117 (p = 0.024)		0.251 (p = 0.000)	0.179 (p = 0.004)		−0.224 (p = 0.004)	−0.148 (p = 0.055)			0.160 (p = 0.012)	0.139 (p = 0.030)	0.126 (p = 0.060)
Ecological		0.075		0.053		0.245 (p = 0.000)	0.203 (p = 0.002)		−0.213 (p = 0.011)	−0.171 (p = 0.036)				0.012	−0.004
Social		0.107 (p = 0.017)		0.056		0.103	0.044		−0.035	0.024			0.102	0.076	0.072
Knowledge							0.277 (p = 0.000)			−0.263 (p = 0.001)				0.195 (p = 0.004)	0.174 (p = 0.019)
Skills							0.010			−0.044				−0.132 (p = 0.027)	−0.133 (p = 0.027)
Planning											0.167 (p = 0.011)				0.061
Information											−0.051				0.003
F-Value	3.290 (p = 0.072)	11.787 (p = 0.000)	0.805	4.146 (p = 0.000)	1.933	13.417 (p = 0.000)	14.037 (p = 0.000)	3.472 (p = 0.065)	7.215 (p = 0.000)	7.437 (p = 0.000)	7.324 (p = 0.000)	8.157 (p = 0.005)	4.495 (p = 0.000)	5.008 (p = 0.000)	4.032 (p = 0.000)
R2	0.024	0.357	0.006	−0.001	0.014	0.388	0.473	0.025	0.254	0.322	0.145	0.058	0.175	0.243	0.247
R2 adjusted	0.016	0.327	0.164	0.124	0.007	0.359	0.440	0.018	0.219	0.279	0.125	0.051	0.136	0.194	0.186
Change in R2	0.024 (p = 0.072)	0.333 (p = 0.000)	0.006	0.158 (p = 0.000)	0.014	0.374 (p = 0.000)	0.085 (p = 0.000)	0.025 (p = 0.065)	0.228 (p = 0.000)	0.068 (p = 0.002)	0.058 (p = 0.002)	0.058 (p = 0.005)	0.117 (p = 0.004)	0.068 (p = 0.005)	0.004

). The hierarchical regression was carried out in five main stages. In Model 1, only control variables were included. Model 2 introduced the Organizational Environment. In Model 3, Organizational Architecture was added. Model 4 encompassed Organizational Decision Making. Therefore, our model comprises one control variable, nine independent variables, and one dependent variable (MP). We checked the assumptions of normality, linearity, and homoscedasticity in our regression analysis. We analyzed normality using Kurtosis and Skewness values.

Linearity was investigated by plotting partial regression for the independent variables, while homoscedasticity was visualized by examining plots of standardized residuals against predicted values. Finally, multicollinearity was assessed for our independent variables [7]. For the purpose of mediation, we examined the Organizational Decision-Making construct as a mediator between the Organizational Environment and Market Performance constructs. Subsequently, we also assessed the role of the Organizational Decision-Making construct as a mediator between the Organizational Architecture and Market Performance constructs.

To present the mediation of effects, we used the Process macroby [63]. To assess the mediation of effects, we calculated the indirect effects of relationships as suggested by [64]. Process analysis allows us to bootstrap to examine the condition of indirect effects. Bootstrapping is a resampling method used to approximate the normal distribution in the sample for a statistical survey. With this, it allows the calculation of population means from a sample redistribution (Central Limit Theorem). We defined 5000 bootstrap samples in the sample, as suggested by [63].

5. Results

From Table 5, we can see the results of our regression analysis. Unstandardized coefficients are reported because all scale values have been standardized with Z-scores that represent standardized effects.

The F test is a statistical test used to check whether the variances of two populations or two samples are equal or not, indicating whether the regression model provides a better fit than a model that does not contain independent variables. Analyzing the F-value of all models shows significance in most of them. R square is a statistical measure that represents the proportion of the variance of a dependent variable that is explained by an independent variable. The model with the highest R squared in the sample was in the third stage in the variable “Planning” (0.473), followed by the second stage in the variable “knowledge” (0.357). Fitted R square is a corrected measure of the goodness of fit for linear models. As for the ranking of the explanation proportion of the independent variables, it was the same presented for the R squared. In general, no major differences were found between R-squared and adjusted R-squared in the models.

Finally, the last metric was the R-changed. It represents how much the model improved with the addition of more predictor (independent) variables in the hierarchical regression. Since all models in all stages were already significant (considering the last stage), it was not possible to illustrate a situation in which the model changes from non-significant to significant with the introduction of predictor variables. With this, all analysis requirements were checked in the database to perform the regression analysis. To assess the mediation of effects, we calculated the indirect effects of the relationships as suggested by [63].

Table 6. Indirect effects (bootstrapping outcome). Source: Authors.

Interactions	Bootstrap Outcome			95% Confidence Interval		Total and Direct Effects	Sig.	Conclusion
	Mean	SD	Sig.	LLCI	ULCI
ECONOMIC → PLANNING → MARKET PERFORMANCE	0.0146	0.0188	0.0100	−0.0111	0.0620	TOTAL EFFECT	0.5812	NO MEDIATION
ECONOMIC → INFORMATION → MARKET PERFORMANCE	0.0062	0.0166	0.3043	−0.0087	0.0370	DIRECT EFFECT	0.8366	NO MEDIATION
DEMOGRAPHIC → PLANNING → MARKET PERFORMANCE	0.0035	0.0203	0.0070	−0.0321	0.0508	TOTAL EFFECT	0.2420	NO MEDIATION
DEMOGRAPHIC → INFORMATION → MARKET PERFORMANCE	−0.0066	0.0114	0.3888	−0.0329	0.0136	DIRECT EFFECT	0.2461	NO MEDIATION
CULTURAL → PLANNING → MARKET PERFORMANCE	0.0598	0.0399	0.984	−0.0066	0.1498	TOTAL EFFECT	0.0002	NO MEDIATION
CULTURAL → INFORMATION → MARKET PERFORMANCE	0.0177	0.0299	0.5099	−0.0402	0.0794	DIRECT EFFECT	0.0349	NO MEDIATION
ECOLOGY → PLANNING → MARKET PERFORMANCE	0.0839	0.0454	0.0308	0.0095	0.1849	TOTAL EFFECT	0.0134	COMPLETE
ECOLOGY → INFORMATION → MARKET PERFORMANCE	0.0251	0.0302	0.3412	−0.0266	0.0908	DIRECT EFFECT	0.6030	NO MEDIATION
SOCIAL → PLANNING → MARKET PERFORMANCE	0.0562	0.0333	0.0476	0.0012	0.1307	TOTAL EFFECT	0.0036	COMPLETE
SOCIAL → INFORMATION → MARKET PERFORMANCE	0.0166	0.0202	0.3390	−0.0184	0.0625	DIRECT EFFECT	0.1188	NO MEDIATION
KNOWLEDGE → PLANNING → MARKET PERFORMANCE	0.0583	0.0671	0.1412	−0.0141	0.1402	TOTAL EFFECT	0.0001	NO MEDIATION
KNOWLEDGE → INFORMATION → MARKET PERFORMANCE	0.0166	0.0330	0.5736	−0.0445	0.0858	DIRECT EFFECT	0.0224	NO MEDIATION
SKILL → PLANNING → MARKET PERFORMANCE	0.0540	0.0277	0.0051	0.0086	0.1150	TOTAL EFFECT	0.9389	NO MEDIATION
SKILL → INFORMATION → MARKET PERFORMANCE	0.0184	0.0175	0.2389	−0.0130	0.0570	DIRECT EFFECT	0.1885	NO MEDIATION

presents the standardized error estimates, significance level and their corresponding lower (LLCI) and upper (ULCI) confidence intervals for the mediation effect.

Analyzing the results in Table 6, it was observed that only two interactions demonstrated complete mediation. Both belong to the Organizational Environment construct.

Table 7. Hypotheses Evaluation. Source: Authors.

Hypotheses	Outcome	Supported Relationship
H1: Organizational Environment has a positive association with Decision Making, leading educational organizations to improve in PM.	Partially Supported	Demographic → Information Cultural → Planning Ecology → Planning
H2: Organizational Architecture has a positive association with Decision-Making, and leading organizations would obtain PM improvement.	Partially Supported	Knowledge → Planning
H3: Decision Making mediates the relationship between Organizational Environment and Market Performance	Partially Supported	Ecology → Planning → MP Social → Planning → MP
H4: Decision Making mediates the relationship between Organizational Architecture and Market Performance	Not Supported	No relationship

presents the results of our hypotheses.

The outcomes of this study garnered through regression analysis and hypothesis testing, unveil certain affirmative associations between Organizational Environment and Organizational Architecture practices and the various phases of Decision Making. This partially corroborates hypotheses H1, H2, and H3. These findings carry significant implications for sector managers, signaling that the adoption of Organizational Environment and Organizational Architecture practices can enhance Market Performance within the context of Decision-Making.

Significantly, the absence of empirical evidence simultaneously linking these three domains (Organizational Environment, Decision Making, Organizational Architecture), as indicated in Figure 1, underscores the novelty and significance of this research. This study bridges a research gap, furnishing empirical substantiation for the interplay among Organizational Environment, Decision Making, and Organizational Architecture, elucidating the synergistic effects of their integration and their practical ramifications. In this vein, our study underscores the existence of relationships and impacts among the addressed topics.

6. Discussion

Our findings reveal promising outcomes concerning the systemic approach of integrating Organizational Environment, Decision Making, and Organizational Architecture, highlighting their contributions to organizations’ Market Performance. Firstly, we observed that organizations focused on acquiring Organizational Environment capabilities, such as advanced technologies and applications, are more adept at navigating all stages of Decision-Making. This aligns with prior studies by [40,65], showcasing how the integration of Organizational Environment and Decision Making can enhance potential contributions to Market Performance. Thus, our results partially support Hypotheses H1, H2, and H3, indicating that Organizational Environment resources significantly influence all stages of Decision-Making, ultimately impacting Market Performance.

On the other hand, concerning Organizational Architecture principles, our results suggest that educational organizations require a balanced approach, incorporating lightweight practices, to navigate effectively to Decision Making. This finding resonates with the work of [66], demonstrating that synergy between information and planning practices leads to improved operational performance. This is reasonable, given that these phases demand robust process monitoring and control tools to ensure Decision Making quality, with top management support and employee commitment being crucial in later Decision Making stages. Understandably, final Decision Making stages require effective knowledge management, especially in market monitoring. These findings align with decision-making scholars’ perspectives.

Interestingly, our results do not support Decision-Making development when organizations concentrate on implementing Organizational Architecture and Organizational Environment practices (i.e., Model 3). A plausible explanation is the challenge of integrating a planning approach that combines Organizational Architecture and Organizational Environment into Decision-Making, which often operates according to market demands. This limited perspective among managers hampers their ability to visualize how a coordinated planning approach of Organizational Architecture and Organizational Environment could effectively drive Decision Making development. Managers often prioritize the immediate outcomes of adopting practices and technologies, overshadowing potential synergistic benefits.

However, examining Hypothesis 3 reveals intriguing findings, suggesting that mediation plays a significant role in the relationship between Decision Making and Organizational Architecture practices, particularly in the Decision-Making context. Hypothesis 4’s result definitively confirms that Decision-Making does not mediate the relationship between Organizational Architecture and Market Performance, leading educational organizations to achieve Market Performance.

Our results indicate that both the technological aspects of the Organizational Environment (i.e., technologies) and the operational aspects of Organizational Architecture can coexist but do not mutually enhance each other in Decision-Making development. This underscores the interdependence of resources in driving successful Decision Making.

Remarkably, in contrast to numerous studies on Organizational Environment, Decision Making, and Organizational Architecture, our findings suggest that, in the given context, these factors are interdependent and essential for achieving superior market performance. Additionally, we note that the Organizational Environment dimension (Model 2) also partially acts as a mediator, facilitating integration. This observation aligns with the logical progression of Decision-Making and literature, where customer data and feedback play a crucial role in optimizing market processes and relationships [15,67]. This is consistent with the research conducted by [68], who explored the mediating effect between Organizational Environment and strategic practices for sustainable organizational performance. Moreover, we find that the implementation of the Organizational Environment (Model 3) can also act as a mechanism to improve Organizational Architecture (Model 2) during the process. This is a significant observation, indicating that pre-planning and decision-making information can enhance effectiveness and support. This underscores the importance of a strategy to enhance tools facilitating successful decision-making. Although complete support for Hypothesis H3 was not obtained, it is worth noting that managers can leverage interdependent practices in different ways to support various phases of Decision Making.

In terms of practical value, our results suggest that organizations, particularly educational institutions, can benefit from a systematic integration of their Organizational Environment, Organizational Architecture, and Decision-Making processes. By focusing on acquiring advanced technologies and applications within their Organizational Environment, organizations can improve their Decision-Making processes, ultimately leading to enhanced Market Performance. This approach can help in better management of budgetary resources, reducing waste, and improving operational efficiency, which are crucial aspects for educational organizations, especially in the context of federal universities.

7. Conclusions

Our study significantly contributes to the field by consolidating a theoretical model and empirically validating the interdependence of Organizational Architecture and Organizational Environment resources in supporting the development of Decision-Making in educational organizations. Unlike previous studies (e.g., [69,70,71,72]) that focus primarily on the moderating effect of well-established operations, our empirical findings demonstrate the importance of incorporating Organizational Architecture, and Organizational Environment features [73,74].

We also provide information on how specific Organizational Environment and Organizational architecture practices can be integrated into Decision-Making, enabling managers to make informed decisions based on their strategic objectives [75,76]. This synergy between Organizational Environment and Organizational architecture practices offers valuable insights for process managers who seek to obtain better results in Decision Making [77,78].

Adopting a systemic view, our study allows managers and professionals to understand the subsystems of and how they can contribute to superior market performance. For example, managers can use our findings to determine whether to prioritize operational aspects, strategic aspects, or both, we demonstrate the relevance of the operational side in supporting interdependence with other aspects [79,80].

Furthermore, our study suggests that the operations and technology literature can benefit from our findings by examining other specific aspects of organizations’ operations, as we provide evidence of the interdependence between contexts, challenging the prevailing notion of a reinforcement effect only. It is important to recognize some limitations of our study. First, our research focused only on educational managers in Brazil, which may limit the generalizability of the results. Future studies should consider larger samples and investigate the introduction of new constructs to explore additional hypotheses and relationships that may influence the dependent variable of the MP [81].

Furthermore, the incorporation of use cases in future research has the potential to generate more insightful findings after our empirical analysis, further enriching the understanding of the interaction between the themes studied. Finally, given that research is relatively peripheral in the literature, we encourage further empirical studies to place Decision-Making/Planning and information as a central object of investigation.