Causal Modeling of Academic Activity and Study Process Management

Abstract

This article presents a causal modeling approach for analyzing the processes of an academic institution. Academic processes consist of activities that are considered self-managed systems and are defined as management transactions (MTs). The purpose of this article is to present a method of causal modeling of organizational processes, which helps to determine the internal model of the current process under consideration, its activities, and the processes’ causal dependencies in the management hierarchy of the institution, as well as horizontal and vertical coordination interactions and their content. Internal models of the identified activities were created, corresponding to the MT framework. In the second step, based on the causal model, a taxonomy of characteristics is presented, which helps to systematize the process quality assessment and ensures the completeness of the characteristics and indicators. Predefined structures of characteristic types are the basis of activity content description templates. Based on the proposed method, two causal models are created: the “to-be” causal model of the target study process (based on expert knowledge) and the “as-is” documented (existing) model of the study process used to evaluate the study process’s quality. The principles and examples of comparing the created “to-be” causal model with the existing study process monitoring method are presented, enabling the detection of the shortcomings in the existing method for assessing academic performance. Causal modeling allows for the rethinking of existing interactions and the identification of necessary interactions to improve the quality of studies. The comparison based on causal modeling allows for a systematic analysis of regulations and the consistent identification of new characteristics (indicators) that evaluate relevant aspects of academic processes and activities.

1. Introduction

The activities of academic institutions (universities and colleges) include teaching and research; a hierarchy of academic activities is formed. Individual activities need to be coordinated by identifying the necessary interaction content to ensure the efficiency and quality of results. The analysis of the academic process aims to improve the quality of the study content, the conditions created for students and their achievements, and graduation rates, as well as increase teachers and students’ satisfaction. The aim of academic process assessment should be “to educate and improve student performance, not merely to audit it” [1].

This article examines only a part of academic activities; we concentrate on study process management. “The information gained should be put to good use by informing decisions about curriculum and instruction at the all management levels of academic institution and ultimately improving student learning” [2]. The result of the analysis of the academic process is summarized as a set of characteristics (indicators) describing the efficiency and quality of the educational institution’s activities, infrastructure (working conditions for teachers and study resources for students), and study content (study programs). The indicators should include the activities and organizational units of all levels of the institution’s management hierarchy, starting with students and teachers (professors) and ending with the activities of the rectorate. The conclusions of the analysis of the academic process should be supported by useful information obtained from the data on the evaluation process, including indicators of school performance [3].

Typical key indicators of school performance, such as students achievements, discipline rules, attendance rates, graduation rates, and teacher satisfaction [4], capture only a fraction of higher education institution (HEI) activities or a few aspects of an academic institution. A broader approach to HEI activities is described as the Key Performance Indicators Framework in [5], which defines five perspectives (with the associated specific indicators): academic, financial, research, facilities (infrastructure), and sustainability.

Another framework for KPIs in higher education institutions is the 2023–2024 Baldrige Excellence Framework (Education), which includes six major categories (perspectives): strategy; leadership; operations (students learning and process results); workforce (workforce results); customers (customer results); and measurement, analysis, and knowledge management (financial, marketplace, and strategy results) [6]. It should be noted that in the proposed framework, the criteria are quite general and require more detail.

An analysis of the existing assessment methodology of the academic process shows that the efficiency and quality of the educational institution’s activities are determined through the set of characteristics (indicators) described in the regulations and their generalization [7].

Academic performance characteristics should also assess the interactions between academic institutions and the marketplace, i.e., the employment rate of graduates and the prediction of needs for new specialties, knowledge, and skills. Anticipating changes in the market demand for new knowledge and technologies is an important aspect of academic performance analysis, as it enables the adaptation of study content to future market needs.

This article aims to present a method of causal modeling of organizations, which helps to determine the internal model of the current process under consideration, its activities, their causal dependencies in the management hierarchy of the institution, horizontal and vertical coordination interactions, and their content. Internal models of the identified activities were created, corresponding to the MT framework. This method provides a conceptual basis not only for the monitoring of academic processes but also for the formation of a system of performance indicators based on a causal interaction model.

The causal modeling approach considers a HEI to be a self-managed system, where each academic activity is a self-managed component defined as a management transaction (MT) [8,9]. The enterprise management causal model was developed in [9] and has already been used to study the causality of academic activities (processes) and the content of informational interactions [10,11]. Causality is considered a chain of cause-and-effect relationships related to a specific domain, as discussed in more detail in [12]. The peculiarity of the method is the use of the concept of circular causality, which is a complex construct formalized as a management transaction (MT).

A comparison of the developed causal academic process model with the methodology applied in practice is presented. Such a technique allows for the detection of gaps in the existing methodologies of academic performance evaluation [7,13]. Such a comparison with the causal model allows for the systematic and consistent identification of new indicators for evaluating relevant aspects of the activity.

The methodology developed based on causal modeling can be applied not only to academic but also to other types of organizations to analyze and evaluate their process management and activities.

The rest of this paper is organized as follows: Section 2 discusses the academic process analysis, modeling, and approaches used in the study process analysis and evaluation. In Section 3, causal modeling principles and frameworks are described. The basic steps of a causal modeling approach for modeling and evaluating processes in academic institutions are presented. Section 4 describes the causal hierarchy model of study process management in detail, with examples. The taxonomy of academic activity characteristics is presented. In Section 5, using the causal modeling method, the existing study process evaluation methodology is clarified. A comparison of “as-is” and “to-be” academic process evaluation models is discussed, and the experimental results are precisely analyzed. Finally, Section 6 presents the main features and generalizations of the method and discusses future work.

2. Related Works

2.1. Academic Process Analysis Issues

Academic activities include research and teaching processes, the study process, study process management, and the organizational management of activities. Academic activities are carried out by organizational units (university, college, school, etc.), aimed at ensuring the quality of the study process for students and support for teachers and research. Departments (units) at different levels of the management hierarchy carry out the internal activities of academic institutions; they communicate with each other during the management control of the institution. The coordination of interactions between these units and the content of transmitted information is the main factor for ensuring the quality of study. Of course, the internal infrastructure of these units (the quality of teaching equipment) and staff qualifications are also important.

All countries or universities have their own experience-based approach to the analysis and modeling of academic processes [14,15,16,17,18,19,20,21,22]. The content of its internal processes and the features of its interaction with the external environment (market and employers) characterize the quality of an academic institution. Usually, the set of performance characteristics is defined by self-analysis documents regulated by an external institution (higher management-level authority) and internal documents of the institution itself (for example, senate resolutions, rector’s orders, or dean’s orders).

The focus of this article is on the study process (our experience is associated with computing degree programs) and study process management, analysis, and modeling. A method based on a causal process model is presented using the concept of a management transaction. It aims to identify the necessary activities and their necessary causal interactions, create internal models of these activities, and specify the content of coordination interactions.

The causal model is a conceptual model representing the inherent causal dependence of activities within a specific real-world domain. Causality and enterprise causal modeling constructs are consistently explored in our work [12]. Here, we briefly present the essential descriptions that explain the causal modeling method. Causality is considered a chain of cause-and-effect relationships associated with regularity in a specific area of the real world. A separate connection between cause and effect is considered as only a fragment of regularity, an element of causal knowledge. The peculiarity of the method is the use of the concept of circular causality, which is formalized in control theory as a control system with a feedback control loop.

Circular causality is a key feature of a well-known business management model, Deming’s PDCA cycle (Plan, Do, Check, and Act) [23]. Circular causality is a common feature of the management and control processes in complex cyber systems such as CPS (cyber–physical systems), CES (cyber–enterprise systems), and CSS (cyber–social systems). One more example is the autonomous computing approach of IBM based on circular causality [24].

Circular causality is abstracted in the concept of a “transaction” in systems engineering and software engineering [25,26]. The topology of the generalized transaction is a wheel graph [27]. A transaction is a key concept for discovering the subject domain’s deep properties (causality). In enterprise modeling, a management transaction is a logical unit comprising management and a control cycle of activity interactions [8,9,25]. The presented method uses the concept of a management transaction to determine the goal of the process and relevant activities and their necessary causal interactions, create internal models of these activities, and specify the content of coordination interactions.

2.2. Study Process Analysis and Evaluation

Many computing degree programs teach different computing disciplines. Computing disciplines must attract quality students from a broad cross-section of the population [13,14,15,16,17,18,19,20,21,22].

The IS2010 Curriculum Guidelines for Undergraduate Degree Programs in Information Systems model has been widely used for nearly a decade [17]. The Association for Computing Machinery (ACM), Association for Information Systems (AIS), formerly DPMA (AITP), and International Federation for Information Processing (IFIP) have contributed greatly to the development of quality training programs [18]. However, their value may be decreasing as new approaches to model curricula have been introduced in IS2020, “Competency-Based Information Systems Curriculum Guidelines” [21].

Creating a high-quality study program even with these guidelines is the first serious task, the success of which depends on the qualifications of the staff of the academic institution and the efficiency of the management of activities. The second part is the arrangement of the program content for the students, the development of competences, and the control of knowledge. All these activities must form an integrated study system that is effectively managed. Constraints on time, resources, prioritization, mentoring, and capacity are the challenges that teachers and learners face. When academic institutions are improving their management, it is useful to understand the evaluations of their performance, which are unbiased and objective, as they encourage improvement.

Five evaluation categories (logic model, theory of change, project plan, metrics calculator, and A+ research model) are synthesized in the program evaluation framework [19]. Different types of evidence are presented to clarify the need for better instructional support related to teacher performance data. They describe how the online curriculum development and implementation activities meet needs, whether the activities are being implemented as planned, and whether the intended curriculum outcomes are being achieved. Using the methods described in this paper, a plan was developed to ensure that the curriculum was evaluated comprehensively. The recommendations could be adapted to support the planning of program evaluations for various degree programs [19].

It is common practice to conduct evaluations of training programs as a means of evaluating their effectiveness. There is a tendency to focus on the importance of the intervention and the amount of learning the individual achieves. One review [20] examined existing evaluation frameworks that have been used to evaluate educational interventions and, in particular, to assess the outcomes of such activities.

The external evaluation and accreditation of studies in accordance with European Higher Education Area Quality Assurance Standards and Guidelines (ESG) in Lithuania is performed by the SKVC [Study Quality Assessment Center] in the Republic of Lithuania [7]. The Study Evaluation Commission evaluates study fields in definite universities at least once every 7 years. Short, second-cycle, complex, and professional studies are evaluated and accredited separately. Points (in a 5-point system) according to seven areas of evaluation assess the quality of the study field: study objectives, outcomes, and content; links between science (art) and study activities; students admission and support; studying, academic achievement, and graduate employment; teaching staff; study facilities and learning resources (material resources); and study quality management and publicity. The decision on the evaluation and accreditation of the study field and cycle is made considering the evaluation report and the proposal of the Study Evaluation Commission. The SKVC is obliged to conduct an annual analysis of the monitoring indicators of study areas. The annual study field monitoring report contains data obtained from the Education Management Information System [22]. The annual monitoring indicators report of study areas is prepared and includes the following indicators of the study process:

• The number of students admitted and studying for majors and degrees in the last 3 years;
• The percentage of students who dropped out during their first year of study in the last 3 years (undergraduate and graduate studies);
• The proportion of students who completed their major and degree studies on time;
• The percentage of students who left for part-time studies (minimum of 15 study credits) of all students in the field of study during the last 3 years;
• The share of teachers in the field of study who left to teach and conduct scientific research in foreign scientific and study institutions from the total number of teachers included in the field of study (in the last 3 years); note: this indicator is not yet included in the annual study field monitoring reports, as there are no data in the registers to assign teachers to the field of study.

We also determined the number of graduates in the field of study [by groups: professional bachelors, bachelors; masters, graduates; short-degree graduates] employed within 12 months after graduation, as well as the share (in percent) of self-employed persons of all graduates and those who did not continue their studies.

2.3. Refinement of the Content of Any Study Program

The content of any study program depends on the definite knowledge about particular features of the problem domain (i.e., organization, enterprise, and physical system). The development of an attractive and competitive study program could first be considered as the acquisition and modeling of some definite knowledge from a particular problem domain.

Knowledge acquisition and management methods for refining the content of any study program are hot topics nowadays. When a specific study program is created or reorganized, many aspects are considered: the legal basis of higher education, market needs, graduate employment level and career opportunities, advanced knowledge, mandatory and optional elements, introductory and basic subjects, preparation opportunities, and credits. The development of study programs is a complex interdisciplinary problem, and an effective solution requires significant human, information, and other resources.

It is necessary to mention the MOCURIS (Modeling Curriculum and Study Process) project that developed a model for managing curriculum- and master’s-level study processes within a university. MOCURIS uses business rule-based analysis and modeling of academic processes at the master’s level to enhance educational quality and students outcomes [13].

In a rapidly changing environment, the timely modernization of study programs, quality assurance, and updating of knowledge content are particularly relevant. This requires the cooperation of all interested parties and effective feedback from the business area.

Another method of determining study content, described in [10,11], is based on discovering causal interactions in the real-world area (subject area) related to the study program. Domain causality is discovered and modeled using the concept of management transactions (MTs) [8]. The internal dependencies of the domain elements are specified by the knowledge model, which is used to shape the content of the studies and then constantly adjust it.

3. Research Method Description

3.1. Principles of Enterprise Causal Modeling

The conceptual model of academic institutions was developed based on the enterprise causal modeling approach [9]. This causal modeling method is intended for enterprise management modeling, where knowledge about enterprise internal activities and their causal dependencies is discovered, including the content of management information [8,9]. The causal modeling approach considers an organizational system (any type of enterprise) as a system of self-managed activities, defined here as management transactions (Figure 1) [8,9].

The enterprise causal modeling method uses two levels of detail for the conceptual modeling of activities [8,9]:

1.

A management transaction (MT) is the main modeling level (level 1) of a self-managed activity and includes closed-loop interaction between two internal parts: the management function (F) and physical process (P) of the management activity;

2.

The elementary management cycle (EMC) is a detailed model of a self-managed activity that reveals the internal steps (inside the closed loop) and information transmitted between steps of the management activity (level 2).

3.1.1. Definition of a Management Transaction

The key concept in the causal modeling approach is a management transaction (MT), depicted in Figure 2 [8,9]. Any activity (in the case of the university, an academic activity) is defined as MT = (F, P, A, V), where F is the management function (domain activity that performs data processing and decisions); P is the process domain activity controlled by F; A = (a(1), a(2), …, a(n)) is an information flow of process P state attributes; and V = (v(1), v(2), …, v(m)) is an information flow of controls (impacts) to process P. This level of detail includes an explicit specification of the goal related to the definite activity. It is important that any activity in the problem area be considered a self-managed system with a defined goal G. Therefore, an MT is associated with a definite goal G relevant to the enterprise’s (e.g., academic institution) activity.

The process P = P(I, O) is a real-world domain activity (e.g., an academic activity) that involves the transformation of physical objects or material flows (real-world domain objects), where I is the input flow (energy and materials), and O is the output flow (products and services). Process P is monitored and controlled by the management function F due to the closed-loop interaction created by information flows A State attributes and V Controls. The management function F is a data-processing and decision-making activity that transforms data flow A = (a(1), a(2), …, a(n)) (the state attributes of P at a particular point in time) and makes decisions related to goal G and, next, transfers decisions (a flow of controls V = (v(1), v(2), …, v(m)) to determine the target state of the process P at a given instant. When describing the features of the MT framework, it is important to distinguish between two types of flows: (a) internal information flows A and V, which form a feedback loop between the management function F, and (b) the managed process P and external (material and physical) flows I (input) and O (output) of process P, which connect process P with the environment (other processes and MTs). Various quantitative estimates—performance or quality indicators—can be defined and used to describe all these MT flows (internal and external). Choosing the right modeling paradigm is important for further analysis of the academic performance. If real-world activities are modeled using an external modeling paradigm (i.e., a black-box approach), the modeler cannot observe and evaluate the internal elements of the activity because only inputs (I-flow) and outputs (O-flow) are observed. Only by applying the internal modeling paradigm (i.e., causal modeling approach) can the modeler observe and evaluate internal elements (function F, flows A and V, and process P) and external flows (I and O).

3.1.2. Definition of the Elementary Management Cycle

A management transaction (see Figure 2) is decomposed and defined in detail as the elementary management cycle (EMC) [8,9], depicted in Figure 3. The elementary management cycle (EMC) framework reveals the internal steps of the enterprise management function Fj (G), internal information flows, and their interactions with enterprise process Pi (G). All elements and interactions depend on enterprise management goals G. Four types of information transformation steps are identified in the EMC framework by decomposing the internal structure of management function Fj: IN—interpretation of state data (data acquisition), DP—data processing, DM—decision making, and RE—realization of decisions (controls). The EMC framework includes an enterprise management goal (G) and goal-driven management information flows (A, B, C, D, and V):

EMC(Fj, Pi) = (Pi(A, G), IN(A, B, G), DP(B, C, G), DM(C, D, G), RE(D, V, G), Pi(V, G))

(1)

where A is the flow of state attributes of process Pi, i.e., a raw data set, which is required in terms of goal G; B is the flow of systematized (interpreted) raw data required in terms of goal G; C is the flow of processed data, the output of the data-processing (DP) step, which corresponds to goal G; D is the flow of management solutions, the output of the decision-making procedure (DM), which corresponds to goal G; and V is the flow of controls for process Pj, the output of the realization step (RE), which corresponds to enterprise goal G.

This method of causal modeling considers an enterprise as a system of self-managed activities described in our previous publications [8,9]. It can be used for various purposes, e.g., for the analysis of enterprise management (the reorganization of business processes), the identification of the content of management information transferred between organizational units, and requirement specification for application software development. This method of causal modeling was also applied to academic institution management analysis and modeling the causality of internal activities and their interactions from the point of view of information flow and transformation [10,11].

3.2. Causal Modeling of Study Content

Causal modeling is applied to the acquisition of knowledge in areas related to the study program and the adaptation of the content of learning objects (LOs) [11]. Two components of the causal model were obtained. First, top-level domain knowledge is discovered and defined using a management operations (MT) framework. Second, the deep structure of the management transaction is revealed through a comprehensive framework called the elementary management cycle (EMC). This approach and causal modeling frameworks are suitable for discovering knowledge related to the content of the study program.

A causality-driven approach is applied here for LO content analysis and renewal (adaptation). The prerequisite for analyzing knowledge content is normalizing knowledge structures using MT and EMC frameworks. Algorithms of the LO content adaptation are presented in two phases. Traceability matrices specify the mismatch of the LO content (old knowledge) and the actual captured domain knowledge. The classification of the content discrepancies and an example of the study program content analysis are presented. The main result of the causal modeling approach to content analysis is the effectiveness of discovering deep knowledge, which is determined by the causal dependencies of the subject area. In addition, a method of updating causal knowledge required in the field of study programs (educational domain) is proposed based on the MT modeling framework. A two-level causal model is created to describe the business area related to the study program content analysis [11]. This method has been verified in practice by creating a knowledge database of the CASE tool, which is designed based on the causal model of the educational domain. The semi-formal procedure for preparing study programs is designed to solve one of the most pressing problems of computer science studies in Lithuania.

3.3. Causal Modeling of the Academic Institution Processes

3.3.1. Study Process Management Steps and Activities

The type of enterprise in this study is an academic institution (university or college). In an academic institution, several main processes (study process, financial management process, research, etc.) are performed; these processes consist of academic activities. To analyze the activities of the academic institution from the point of view of causality, we use the concept of MT, the conceptual model described above (Figure 2). The main steps of the causal modeling approach, which is applied here to model and evaluate the processes of academic institutions, are as follows:

1.

Determining the target model of the current process under consideration (expertly):

1.1.

Identifying the main activities that need to be performed to make the current process efficient (expertly), creating a target process model as a matrix of interaction of activities (“to-be”);

1.2.

Defining the content of the coordination flows of activities in the activity interaction matrix, considering the levels of the management hierarchy where the activities are carried out;

1.3.

Ranking the importance of causal interactions (coordination flows) between activities by analyzing the matrix;

1.4.

Based on the causal model of the study process and the definition of the management transaction, a taxonomy of performance characteristics is proposed, on the basis of which it is possible to systematically and consistently define the quantitative indicators of specific characteristics.

2.

Creating the existing (documented by regulations) process model (as it is) used to assess the quality of the process:

2.1.

Naming the main activities of the process defined in the regulations;

2.2.

Compiling the causal interaction matrix of the activities (“as-is”), recording the content of the coordination flows disclosed in the regulations;

2.3.

Comparing the created “to-be” causal model with the existing (in practice) process monitoring methodology; this makes it possible to detect the shortcomings of the existing methodology for evaluating the academic performances of universities.

When describing this method, we chose the management of the study process and its internal activities as an example. In the first step, the expert group creates a target model of the study process, the main activities of which are presented in

Table 1. Study process management activities.

Types of Study Process Activities (and the Corresponding Organizational Units of Management)	ID
Administration of the study process (university level)	MT1
Students in the learning process (students level)	MT2
Market needs analysis (faculty/department level)	MT3
Study program design (faculty/department level and study program committee)	MT4
Prediction of study program content (faculty/department level and study program committee)	MT5
Study program teaching (staffing level and professors)	MT6
Microcredit design (faculty/department level and study program committee)	MT7
External study evaluation (ministry level and external institutions)	MT8

.

Each activity in Table 1 is understood as a management transaction MT = (F, P, A, V), of which the specific content of elements depends on its purpose. An example of the management transaction MT4, study program designing (faculty/department level and study program committee), is provided in Figure 4. When creating or improving a specific study program, it is necessary to consider different and conflicting aspects (the existing legal framework of higher education, global and local market needs, available knowledge, etc.) depicted in the study process management concept map (Figure 5).

The concept map presented in Figure 5 shows the essential aspects and provides links to other necessary sources of study process management.

3.3.2. An Example of a Causal Model of Academic Activity

We explain the internal causal interactions in the model of activity MT4, study program designing, shown in Figure 4. MT4 comprises a closed-loop interaction between the management function F4, management of SP development and process P4, development of SP content, which is associated with goal G4, compilation of SP qualification requirements that meet the needs of the market. The management control of this activity, study program designing, is provided by the feedback between F4 and P4, which is created by flow A4, data of SP development (state of P4), and flow V4, qualification requirements (controls). Making full use of the causal modeling constructs described in [8,9], the conceptual model of the academic activity, study program designing, can be decomposed, revealing more detailed granularity (as shown in Figure 6) using the concept of “elementary management cycle” (EMC) [8,9].

Such a decomposition of an MT is theoretically justified because the management function (F) is complex and consists of four internal steps: the interpretation (IN) of the process P state attributes, data processing (DP), decision making (DM), and decision realization (RE) (implementation of the impact to process P) [8,9]. The internal elements of the management function (F4) are identified and depicted in Figure 4. This results in the EMC4 “Study program designing” model, with the following elements: IN1—interpretation step, DP1—data-processing step, DM1—decision-making step, and RE1—decision-implementation (realization) step. Furthermore, the goal (G4) (the same as in MT4) and its interactions with the internal elements of F4 are clearly specified. Such a detailed representation of self-managed activities is called the elementary management cycle (EMC) and is described in detail in [8,9,12].

Thus, the causal model of the activity “study program designing” using the EMC framework (Figure 6) is being developed with the overall goal G4, “compilation of SP qualification requirements that meet the needs of the market”. The contents of the EMC4 elements are listed in

Table 2. The detailed specification of the activity, “designing the study program” in the EMC.

Elements of EMC	ID	Content of Steps (Business Rules BR)	Information Flows (Attributes)
Goal G1	G4	Goal setting: analysis of study program purpose, demand, and defining goals
Management function F1 = “management of the study program development”	IN1	Elicitation of SP requirements: IN1.1 {BR11} = the identification and systematization of study program requirement sources IN1.2 {BR2} = the study program requirements’ specification and description	A1 = the data on SP state over time B1 = the primary study program requirements with a description
	DP1	Analysis of primary SP requirements (B1): DP1.1 {BR3} = the classification of requirements defining primary types of requirements DP1.2 {BR4} = the formulation of specifications for the study program and its different components	C1 = the study program content requirements
	DM1	SP content definition and approval: DM1.1 {BR5} = the drafting of the study program content project DM1.2 {BR6} = the description of requirement types and document types DM1.3 {BR7} = the creation of requirement documents and the formation of requirements	D1 = the approved e-document for SP content requirements
	RE1	Implementation of solutions: RE1.1 {BR8} = the definition of study program qualification requirements (features) RE1.2 {BR9} = the traceability of groups of qualification requirements	V4 = the qualification requirements (SP quality controls)
Process P1 = development of SP content	P4	The development of study program content by staff (real academic activity) based on real-world needs	A4 = the data of the SP development state over time V4 = qualification requirements (SP quality controls)

, namely, the internal steps IN1, DP1, DM1, and RE1 (information transformation rules) and internal flows A4, B1, D1, and V4 (attributes and data) of the feedback loop.

The content of EMC4 is aligned with the specification of the same activity as MT4 due to the same identifiers F4, A4, and V4 (Figure 5).

3.4. Coordination of Academic Institution Activities

The coordination of organizational activities is defined as communication between two different processes (run by units) that are independent (i.e., self-managed) and that seek to reconcile their actions because they pursue a common enterprise goal (sub-goal). Coordination is based on the transfer of definite information that ensures that different organizational units (departments and groups) work toward a common goal; these academic activities are aligned with the institution’s goals. In causal modeling, any academic activity (i) is considered a management transaction (MTi). The coordination is considered a mutual exchange of information, where two MTs transmit and receive information, which affects the behavior of each MT (the MT elements’ content) (Figure 7).

Definition: In the general case, the coordination C(MTi, MTj) of two different management transactions, MTi and MTj, includes the two-directional transfer of a certain set of attributes: flow C(i, j) from MTi(Fi, Pi, Ai, Vi) to MTj(Fj, Pj, Aj, Vj) and feedback flow C(j, i) from MTj(Fj, Pj, Aj, Vj) to Mi(Fi, Pi, Ai, Vi).

In this way, from a systemic point of view, a reciprocal interaction between two different MTs is formed. Mutual information transfer is especially important if departments at different levels of the hierarchy perform these coordinated activities. For example, in the study process model, the activities of MT6 and MT2 are at different levels of the hierarchy, where MT2 is subordinate to MT6 (Figure 8). We will define the principle of how the received coordination flow can affect (correct) the internal structure of the coordinated activity (the internal MT structure).

Assumption: The received coordination flow C(.,.) is analyzed using MT(F, P, A, V) management function F, makes a decision, and changes the content of its elements: the structure of flows A and/or V and the structure of function F (i.e., the behavior of the entire MT is adjusted accordingly).

The general case of coordination interaction between activity MT and activity MT’ is depicted in Figure 7. Coordination flow C(i, j) from the coordinating activity MTi(Fi, Pi, Ai, Vi) affects the behavior of the coordinated activity MTj(Fj, Pj, Aj, Vj). In turn, the coordinated activity MTj(Fj, Pj, Aj, Vj) also provides feedback C(j, i) (the corresponding response) to the coordinating activity MTi(Fi, Pi, Ai, Vi):

C(MTi, MTj) = {C(i, j); C(j, i)}

(2)

Therefore, the effect of relativity (uncertainty) is created because coordinating activity MTi only “roughly knows” the state of MTj, since it can estimate only a part of MTj attribute values obtained by the feedback flow C(j, i) but seeks to influence (control) MTj behavior or evaluate the status of MTj (Figure 7). “Roughly” means that in the general case, the coordination flow C(i, j) is associated only with a part of MTi internal elements (a subset of F, P, A, and V) and a part of MTj internal elements (Fj, Pj, Aj, and Vj) and their characteristics. In general, the coordination flow can include not only information but also (if necessary) the transfer of materials (support and equipment) or other resources.

4. Causal Model of the Study Process

4.1. Academic Process as a System of Management Transactions

The assumptions on which the conceptual model of academic processes was developed are as follows:

1.

The academic process consists of activity types (see Table 1), which are considered MTs with a predefined internal structure MT = (F, P, A, V).

2.

Academic activities communicate by transferring information to each other, as they aim to coordinate their activities (decisions) in order to achieve the common institution goal (e.g., ensuring the quality of studies).

3.

Academic activities form a hierarchical structure (aggregation hierarchy), which ensures the management of the academic institution and the achievement of the institution’s goal.

4.

The status of academic activity at each level of the management hierarchy is evaluated using indicators (KPIs) specific to that level.

4.2. The Coordination Interactions of Study Process Activities

The main study process activities are presented in Table 1 and illustrated in Figure 8. The recommended coordination interactions of study process activities (state “to-be”) determined based on causal modeling and expert knowledge are shown in Table 3. Below is a brief explanation of this recommended study process management coordination model:

• C(i, j) in the cells of Table 3 represents a causal link between two activities: the content of coordination information from an academic activity defined as a management transaction (MTi) (in column i) to an academic activity defined as a management transaction (MTj) (in row j).
• Information that flows from MTs at a higher level of the management hierarchy to MTs at a lower level includes requirements, restrictions, or controls, called “impact”.
• Information that flows from the lower-level MTs of the management hierarchy to MTs at a higher level is called “state information” (information about the current situation).
• The regulations define the academic quality conditions of the study process and the necessary information (data and characteristics) related to the resources and activities of the study process, which are used to calculate the relevant study performance and quality indicators (KPIs).

Table 3. The coordination of study process activities (state “to-be”).

Academic Process Activities (MTs)	From/To	MT1	MT2	MT3	MT4	MT5	MT6	MT7	MT8
University administration	MT1		C(2, 1)	C(3, 1)	C(4, 1)	C(5, 1)	C(6, 1)	N	C(8, 1)
Student in the learning process	MT2	C(1, 2)		C(3, 2)	N	N	C(6, 2)	N	N
Analysis of market needs (department level)	MT3	C(1, 3)	C(2, 3)		C(4, 3)	C(5, 3)	C(6, 3)	C(7, 3)	N
Study program design (department level)	MT4	C(1, 4)	C(2, 4)	C(3, 4)		C(5, 4)	C(6, 4)	C(7, 4)	N
Prediction of study content (department level)	MT5	C(1, 5)	C(2, 5)	C(3, 5)	C(4, 5)		C(6, 5)	C(7, 5)	N
Study program teaching (professors)	MT6	C(1, 6)	C(2, 6)	C(3, 6)	C(4, 6)	C(5, 6)		C(7, 6)	C(8, 6)
Microcredit development (department level)	MT7	N	C(2, 7)	C(3, 7)	C(4, 7)	C(5, 7)	C(6, 7)		N
External study evaluation [ministry level]	MT8	C(1, 8)	C(2, 8)	C(3, 8)	N	N	N	N

Table 3. The coordination of study process activities (state “to-be”).

Academic Process Activities (MTs)	From/To	MT1	MT2	MT3	MT4	MT5	MT6	MT7	MT8
University administration	MT1		C(2, 1)	C(3, 1)	C(4, 1)	C(5, 1)	C(6, 1)	N	C(8, 1)
Student in the learning process	MT2	C(1, 2)		C(3, 2)	N	N	C(6, 2)	N	N
Analysis of market needs (department level)	MT3	C(1, 3)	C(2, 3)		C(4, 3)	C(5, 3)	C(6, 3)	C(7, 3)	N
Study program design (department level)	MT4	C(1, 4)	C(2, 4)	C(3, 4)		C(5, 4)	C(6, 4)	C(7, 4)	N
Prediction of study content (department level)	MT5	C(1, 5)	C(2, 5)	C(3, 5)	C(4, 5)		C(6, 5)	C(7, 5)	N
Study program teaching (professors)	MT6	C(1, 6)	C(2, 6)	C(3, 6)	C(4, 6)	C(5, 6)		C(7, 6)	C(8, 6)
Microcredit development (department level)	MT7	N	C(2, 7)	C(3, 7)	C(4, 7)	C(5, 7)	C(6, 7)		N
External study evaluation [ministry level]	MT8	C(1, 8)	C(2, 8)	C(3, 8)	N	N	N	N

The performed expert assessment showed the significance (importance) of the coordination flows C(i, j) presented in the conceptual model (Table 3). The degree of need for coordination information is measured on a scale as follows: N—unnecessary, 1—useless, 2—low impact, 3—medium impact, 4—high impact, and 5—decisive (determining).

After surveying the students, the following estimates of the importance of coordination interactions were obtained (the calculated average):

• The students’ need to receive information from university departments was measured using the importance ratings of coordination flows in the MT2 row: C(1, 2) = 2.3; C(3, 2) = 3.6; C(6, 2) = 4.1;
• After surveying the students, the following estimates were also obtained about the importance of students’ opinions on study activities (study content): C(2, 1) = 2.9; C(2, 2) = 3.6; C(2, 3) = 2.8; C(2, 4) = 3.0; C(2, 5) = 3.4; C(2, 6) = 3.0; C(2, 7) = 2.4; C(2, 8) = 3.1. The students are thus greatly influenced by peers’ opinions, C(2, 2) = 3.6.

The survey of the university’s academic staff based on this model showed estimates of the importance of the coordination interactions obtained (the calculated average). Some of the more important activities are as follows:

• The need to receive information from other activities for “study program designing” (MT4) is assessed as follows: C(1, 4) = 4.3, C(2, 4) = 2.7, C(3, 4) = 4.3, C(5, 4) = 4, C(6, 4) = 3.7, C(7, 4) = 3; C(8, 4) = 1.3;
• The need to receive information for study program teaching (MT6) from other activities is as follows: C(1, 6) = 4.7, C(2, 6) = 3.7, C(3, 6) = 3, C(4, 6) = 2, C(5, 6) = 2, C(7, 6) = 3, C(8, 6) = 4.7;
• The need to receive information for microcredit development (MT7) (the department level) from other activities is as follows: C(1, 7) = 1, C(2, 7) = 2.7, C(3, 7) = 4.7, C(4, 7) = 4.3, C(5, 7) = 3.7, C(6, 7) = 1.7, C(7, 8) = 1.

Evaluations regarding the importance of the content of one activity (department and unit) to other activities were also obtained. Some of the more important activities are as follows:

• The importance of the content of study program designing (MT4) to other activities is as follows: C(4, 1) = 3.7, C(4, 2) = 1, C(4, 3) = 2.7, C(4, 5) = 2.7, C(4, 6) = 2, C(4, 7) = 4.3, C(4, 8) = 1;
• The importance of the result (content) of microcredit development (MT7) to other activities is as follows: C(7, 1) = 1, C(7, 2) = 1, C(7, 3) = 3.5, C(7, 4) = 3, C(7, 5) = 1, C(7, 6) = 3, C(7, 8) = 1.

The meaning (content) of information C(i, j) transmitted in the coordination interaction depends on the type (purpose) of specific activities (i, j); for example, the content of C(7, 2) can be “individual study plans with microcredits”, C(2, 6)—feedback information from students to professors.

By following such a methodology, highlighting strategically important causal interactions, it is possible to compile (based on expert experience) the institution’s target model (state of the art) with activity priorities (importance indicators).

Following the causal model of the study process, after a detailed analysis of the existing regulations, the coordination interactions defined there (the as-is state) were identified and are presented in

Table 4. Template of the MT4 “study program designing” internal model.

Internal Elements of MT4	Content of MT4 Elements	Description, Comments
Management of SP development (F4—function)	<SP data processing and decision making>	F4: <…meaningful description…>
Development of SP content (P4—process)	Development of SP description	P4: <…meaningful description…>
Input I: information corresponding to the causal model in Table 2	I1 = C(2, 4); I2 = C(3, 4); I3 = C(5, 4); I4 = C(6, 4); I5 = C(7, 4)	I1: <…>; I2: <…>; I3: <…>; I4: <…>, I5: <…>
Output O: results corresponding to the causal model in Table 2	O1 = C(4, 3); O2 = C(4, 6)	O1: <…>; O2: <…>
A4—state data of SP development	A1 = C(4, 1); A2 = C(4, 5); A3 = C(4, 7)	A1: <…>; A2: <…>; A3: <…>
V4—controls: qualification requirements	V4 = C(1, 4)	V4: <…>

. In table C(i, j), the contents of cell C(i, j) are indicated only if the relevant university regulations (university administration, faculty council, department, etc.) are approved. An empty cell (i, j) means that the interaction C(i, j) is not defined in the documents.

The conceptual model in Figure 8 is a visualization of the coordination interactions from Table 3, created by analyzing the university’s study process (management) activities with the causal modeling method.

It is necessary to note that the study process management model in Figure 8 is shown from an external modeling perspective, where study process activities (MT1–MT8) are considered black boxes, so the internal structure is not shown. Therefore, only external (coordination) relationships between different MTs and the input flow (I) and output flow (O) of each MT can be observed in this model, measured, and evaluated (see Figure 1). Note: the input (I) and output (O) of the MT are not shown in the diagram (Figure 8) for simplicity.

Considering the hierarchy of the management units in Figure 8, the meanings of coordination flows differ:

(a)

The meaning of flow C(i, j), if sent from a lower-level unit (activity MTi) to a higher-level unit (activity MTj), is “state parameters”;

(b)

The meaning of flow C(i, j), if sent from a higher-level unit (activity MTi) to a lower-level unit (activity MTj), is “instructions or requirements” affecting the activity of MTj.

We now discuss the content of the causal model presented in Table 3, which shows the recommended coordination interactions C(i, j) between study process management activities (MTs). The content of the university administration’s activity (defined as MT1) communication with institution units could be as follows:

○

Coordination flow C(1, 2) is the impact of university administration (MT1) on the students activity in the learning process (MT2) and indicates the academic conditions that the student has the right to use. To ensure the quality of the study process, coordination flow C(2, 1) is mandatory because it includes feedback about the student’s academic performance and his/her opinion (needs) about real academic conditions. A causal interaction loop between MT1 and MT2 is created.

○

Coordination flow C(1, 3) uses the influence of university administration (MT1) to analyze market needs at the department level (MT3): rights and resources are granted for market analysis and support for the department’s activities. The coordination flow C(3, 1) is the necessary feedback information about department activities (MT3) of market needs analysis required by the university administration. A causal loop of interaction between MT1 and MT3 is created.

○

Coordination flow C(1, 4) influences the university administration (MT1) in SP design at the department level: conditions and resources are granted, and restrictions are specified. The coordination flow C(4, 1) includes mandatory feedback about the state of the study program design (MT4) for the university administration. A causal loop of interaction between MT1 and MT4 is created.

○

Coordination flow C(1, 5) in study content forecasting is mandatory, but feedback information C(5, 1) from the department level to the university administration (MT1) separately from SP is not required.

○

Coordination flow C(1, 6) impacts university administration (MT1) to study program teaching at the department level: conditions and resources are granted, and restrictions are specified. The coordination flow C(6, 1) provides mandatory feedback information about study program teaching (MT4) for the university administration.

○

Coordination flow C(8, 1) provides mandatory information for external study evaluation (the ministry level). The coordination flow C(1, 8) provides feedback about the results of the study process evaluation. A causal interaction between MT1 and MT8 is created.

The currently valid study quality indicators (in the Republic of Lithuania) are defined by the “Methodology of external evaluation of study areas” [7]. Based on this document, universities organize internal accounting and evaluation activities of the study process. This section includes some examples of different types of indicators. An example of the contents of C(1, 8) is given in Appendix A. Several indicators are presented: 1. Objectives, results, and contents of studies includes 13 indicators (MT01–MT13). 2. Interfaces of science and study activities includes six indicators (MT14–MT19). 3. The admission and support of students includes 10 indicators (MT20–MT29). 4. Studies, study achievements, and graduates’ employment includes 11 indicators (MT30–MT40). 5. Teachers includes 10 indicators (MT41–MT50); 6. The material resources of studies includes eight indicators (MT51–MT58). 7. Study quality management and publication includes 13 indicators (MT59–MT71).

The student is considered a focus of studies process management, and his/her activity is conceptualized as a management transaction (MT2): the student in the learning process (see Table 1). The recommended interactions of the student’s learning process (MT2) with the rest of the study activities are as follows:

• Coordination flow C(2, 1) includes mandatory information feedback to university administration activity (MT1) about the students academic performance and his/her opinion (experience and needs) about real academic conditions.
• Coordination flow C(2, 3) includes the students opinion (experience and needs) about real market needs (feedback), which is potentially valuable for the analysis of market needs (MT3) at the department level.
• Coordination flows C(2, 4) and C(2, 5) are not necessary (from the point of the causal model); however, they may be considered on a case-by-case basis.
• Coordination flow C(2, 6) includes valuable feedback on study program teaching activities (MT6).
• Coordination flow C(2, 7) includes the students opinion (experience and needs) about the development of microcredits, which is potentially valuable for the development of microcredits at the department level (MT7).
• Coordination flow C(2, 8) includes the students opinion (experience and needs) about study process features, which is mandatory in the evaluation process (MT8) of the university’s activities.

The recommended causal interactions between the analysis of market needs, MT3, and other study activities are as follows:

• Coordination flow C(3, 1) includes valuable feedback information for university administration activity (MT1) about market needs analysis at the department level (MT3).
• Coordination flow C(3, 2) from the market needs analysis of the department (MT3) could be valuable information to motivate students (MT2).
• Coordination flows C(3, 4), C(3, 5), C(3, 6), and C(3, 7) contain the results of market needs analysis (MT3); they are valuable feedback information for study program design (MT4), the prediction of study content (MT5), and microcredit development (MT7) at the department level.

The interactions of study program designing (MT4) with other study activities include the following:

• Information flow C(4, 1) from study program design (MT4) at the department level to university administration activities (MT1) is valuable feedback.
• Coordination flows C(4, 6) and C(4, 7) from study program design (MT4) to study program teaching (MT6) and microcredit development (MT7) are not necessary but are useful at the department level to improve the quality of studies.
• Interactions of the management transaction MT4 with MT3 and MT5 are not defined; however, they may be considered on a case-by-case basis.

Coordination flows from the prediction of study content (MT5) include the following:

• Information flow C(5, 1) to university administration activity (MT1) about the revealed trend of the study content changes.
• Coordination information C(5, 3) for the analysis of market needs (MT3) and C(5, 4) study program design (MT4) (SP preparation and update) is needed to validate the conclusions obtained about market needs and trends to gain knowledge about new specialties (at the department level).
• Coordination flows C(5, 6) to study program teaching (MT6) and C(5, 7) microcredits development (MT7) (department level) are needed to base conclusions on the received trends of market needs and new, specialized knowledge.

The coordination interactions of study program teaching (MT6) with other activities (see Table 1) are as follows:

• Coordination flow C(6, 1) from the department to university administration activities (MT1) about the state of the teaching results of study program courses, requirements for students knowledge, and calculations of specific indicators (defined in the university regulations).
• Coordination flow C(6, 2) informs students (MT2) about updated content regarding teaching study program courses and requirements for students knowledge and task results.
• Information flows C(6, 4), C(6, 5), and C(6, 7) regarding the state of teaching results for study program courses are valuable to study program design (MT4), the prediction of study content (MT5), and microcredit development (MT7).

Coordination interactions of activity microcredit development (MT7) with the activities MT3, MT4, MT6, and MT8 (see Table 1) are as follows:

• Coordination flows C(7, 3), C(7, 4), and C(7, 6) provide information about microcredit development (MT3) results at the department level. They are valuable feedback for the MT3, MT4, and MT6 activities.
• There is also valuable feedback information to MT7 in coordination flow C(3, 7) about market needs from MT3, in coordination flow C(4, 7) about study program design results from MT4, in coordination flow C(5, 7) about the prediction of study content from MT5, and coordination flow C(6, 7) of the teaching results from MT6.

Coordination information C(8, 1) and C(8, 6) from external study evaluation creation (MT8) are mandatory requirements for university administration activities (MT1) and study program teaching (MT6) (department level and study program committee).

4.3. An Example of Study Activity Content Specification

The study process management model in Figure 8 shows a hierarchy of study process activities defined as management transactions (MTs) in Table 1. The scheme contains five management hierarchy levels (aggregation levels i1, i2, i3, i4, and i5) and their purpose names. When refining our study process causal model (Figure 8), all activities identified in this model can be decomposed, and their internal models can be created. The conceptual model of management transactions is presented in Figure 2. The content of the elements of management transaction depends on the goal of the simulated activity (defined as the internal goal of MT) and on their coordination interactions C(i, j), which are defined in the study process model (Table 1 and Figure 8).

The internal model of any MT is created according to the same logic, which is based on the definition of the MT’s internal structure in Figure 2. Below are two examples. The main steps for defining the content of MT6, study program teaching, and MT2, a student in the learning process, are presented in the context of the enterprise analysis and modeling (EAM) course. MT2 is at level i = 1 (students knowledge control), and MT6 is at level i = 2 (study program teaching control/staffing). The study program teaching activity has an internal structure, defined as MT6 = (F6, P6, A6, V6) (teaching activity as a self-managed system) and is associated with goal G6. The students in the learning process have an internal structure defined as MT2 = (F2, P2, A2, V2) (students’ activity as a self-managed system), which is associated with goal G2.

The internal model of the MT6 study program teaching in the case of the course “enterprise analysis and modeling (EAM)” is created according to the following logic:

• The goal of activity MT6 is defined in the syllabus; in the case of the EAM course, goal G6 means “to master the methods and tools of the enterprise systemic analysis and enterprise modeling”.
• The content of management function F6 comprises the compilation and preparation of study subject content for teaching based on SP qualification requirements.
• In the case of EAM teaching, the semantics of F6 is “Controlling the delivery of EAM content, collecting and analysis of students knowledge level indicators, and practical ability indicators”.
• P6 is a physical teaching process that presents the content of a certain SP course and tests students knowledge, the accumulation of knowledge level indicators, and practical ability indicators. Process P6 input flow I6 includes a few components: 1. SP qualification requirements. 2. Training content: up-to-date methods and standards, case studies and tools, and practical tasks. 3. Tests for students knowledge control. Output flow O6 includes the following components: 1. students reports on practical tasks; 2. students testing results; 3. students feedback on SP-related needs.
• In the case of EAM, input flow I6.2, training content, includes the basics of business process modeling, DFD, BPMN, DMN, UML, SySML specifications, enterprise architecture frameworks MODAF, UAF, and practical tasks using modeling tools (Lucidchart, Camunda, MagicDraw, etc.).
• Information flow A6, state attributes, includes a set of indicators that reflect the quality of course delivery, including the characteristics of students acquisition of knowledge of the subject of study and perceived professional skills (status indicators): A6 = (a(61), a(62), a(23), …, a(6n)).
• In the case of EAM course teaching, examples of state attributes, A6, include the following: a(61)—knowledge of enterprise modeling notation, a(62)—enterprise domain modeling skills, and a(63)—domain modeling errors.
• Control flow V6 content includes recommendations for the methods used in the teaching process, requirements and limitations for the course, and tools for teaching (controls of the teaching process, P6): V6 = (v(61), v(62), v(63), …, v(6m)).
• In the case of EAM course instruction, examples of V6 include the following: v(61)—business enterprise analysis methods, v(62)—user requirement specification templates, and v(63)—system requirement specification in UML.

The internal model of MT2, students in the learning process (MT2 is at level i = 1, students knowledge control), in the case of the course “Enterprise analysis and modeling” (EAM), is created according to the following logic:

–

The MT2 activity objective is determined by the student him-/herself; in the case of the EAM course, goal G2 is defined as “To fully master the theory and practical tools of EAM”.

–

The content of management function F2 includes the assimilation of knowledge of the subject of study, self-monitoring of the understanding of the subject knowledge, and the analysis of the state of one’s knowledge.

–

P2 covers student activities of studying enterprise modeling methods and tools, solving and describing practical tasks. Process P2 input flow I2 consists of the following: 1. educational material, 2. subject knowledge requirements, and 3. the schedule of practical tasks. Output flow O2 consists of the following: 1. student reports on practical tasks, 2. student testing results, 3. student self-testing results, and 4. student feedback on EAM course-related needs.

–

The content of flow A2 is the student’s own understanding of the content of the taught subject: a self-assessment of subject knowledge and practical abilities of the subject of study: A2 = (a(21), a(22), a(23),…, a2n)).

–

In the case of the EAM course, the content of stream A2 is as follows: a(21)—assimilated knowledge of DFD, DSD, and FH notations; a(22)—assimilated knowledge of BPMN and DMN specifications; and a(23)—skills in using business domain modeling tools, etc.

–

The content of stream V2 includes students’ conclusions regarding their understanding of the content of the subject, weak areas in their knowledge, and additional steps in studies.

–

In the case of the EAM course, the content of stream V2 is as follows: v(21)—examine more closely the characteristics of the modeling system of the highest DFD level, and v(22)—examine the requirements of the models of the first DFD level.

–

The content of flow V2 includes feedback to students about the results of their knowledge control, a discussion of the results, and recommendations: V2 = (v(21), v(22), v(23),…, v(2m)).

In the causal model of the study process, there is a coordination interaction between MT6 and MT2. The general case of coordination is shown in Figure 7. An example of coordination interactions between MT2 and MT6 when teaching the course “Enterprise analysis and modeling” is shown in Figure 9.

Since MT6 is a study process activity of a higher organizational level (aggregation level i = 2) than MT2 (aggregation level i = 1; see Figure 8), the coordination flow C(6, 2) from MT6 to MT2 includes (a) requirements for students knowledge and skills and (b) post-test notes on students knowledge gaps. The feedback flow C(2, 6) includes the results of the assessment of students knowledge as well messages from students to teachers regarding the taught subject.

The content of C(6, 2) affects the students behavior in study activities, as this is how specific knowledge gaps are indicated. The students feedback is important for the C(2, 6) teacher, as it allows for an objective evaluation of the progress of the subject studies and connects the students level of understanding with specific parts of the taught subject.

4.4. An Example of Linking a Study Activity to the Causal Model

Compiling the internal models of study activities according to the internal structure of MT(F, P, A, V) is important for the university. However, this extensive expert work is beyond the scope of this article. However, the created conceptual study process (management) model (Table 3) provides a basis for a systematic and consistent methodology. An example of the template of the internal model of the activity “study program designing” (MT4 in Figure 4) is presented in Table 4. MT4 is a part of the coordination model of the entire study process (Figure 8). First, the internal purpose of activity MT4 is defined; for example, G(4) = “SP evaluating market tendency and technology development prediction”, which gives meaning to other activity elements. In the second step, all interactions of MT4 with other activities (according to Table 3) are selected from the created causal model of the entire study process and classified, respectively, as internal elements of MT4 (F4, P4, A4, and V4), as shown in Table 4.

This example shows a systematic way of identifying the specific content of all MT4 elements when the content is obtained from the causal domain model compiled earlier. Now, the SP design (MT4) model, which is in Figure 4, acquires specific content; as each element of MT4 is specified in more detail, MT4 becomes a specification of requirements (template). Based on this methodology, the specifications (templates) of the management activities of all study processes can be drawn; this systematizes the design and implementation of the internal interactions of transactions MT1–MT8, as well as the compatibility of the content of all activities (and data exchange) according to the causal model.

4.5. Taxonomy of Academic Activity Characteristics

Based on the presented study process model, the characteristics of academic activities can be classified and used to assess the performance and quality of the activity characteristics that can be observed, measured, analyzed, and evaluated. The taxonomy of the characteristics is presented while considering the selected paradigm of modeling (analysis) of processes: the external view (black box) or internal view (white box). An example of the hierarchy of study process activities (state “to-be”) is shown in Figure 8 from an external modeling perspective, where each academic activity is a black box because its internal structure is not represented. Therefore, only external relationships (coordination interactions) between different MTs and the input flow (I) and output flow (O) of each MT can be observed, measured, and evaluated (see Figure 1). Note: the input (I) and output (O) of the MT are not shown in the diagram (Figure 8) for simplicity. When applying internal modeling, academic activity is a self-managed system defined as an MT and has the internal structure MT(F, P, A, and V) (see Figure 2). Therefore, from the internal modeling perspective, the MT is a white box, the characteristics of the internal elements (F, P, A, and V) of the MT can be analyzed, and then their evaluation can be carried out. It is also necessary to assess the hierarchy of the institution’s management units when classifying because they are performers of specific activities, and there are vertical subordinate relationships between them.

The discussed modeling assumptions allow for defining the following types of activity characteristics (indices):

• The external characteristics of activity MTj:

The external characteristics of activity MTj include the following:

(a)

The management hierarchy level index (h) for activity MTj(h); value (h) can be the hierarchy level name, e.g., h = “rectorate level” or “faculty level”, etc.;

(b)

The input content MTj(I) of activity MTj, where I = {i(1), i(2),…, i(n)};

(c)

The output content MTj(O) of activity MTj, where O = {o(1), o(2),…, o(m)};

(d)

The description MTj(D) of the activity MTj content, where D = (<description of MTj>).

• The internal characteristics of activity MTj:

The characteristics of the internal elements of activity MTj(Fj, Pj, Aj, Vj) include the following:

(a)

Specification Fj(S) of management function Fj content, where Fj(S) = <content of Fj>;

(b)

Specification Pj(S) of process Pj content, where Pj(S) = <content of Pj>);

(c)

Specification Aj(S) of the state attributes Aj(S), where Aj(S) = {a(1), a(2),… a(m)};

(d)

Specification Vj(S) of the controls Vj(S), where Vj(S) = {v(1), v(2),… v(n)}.

Note: each activity MTj is at a certain management level and has an index h: MTj(h).

• The characteristics of subject area process coordination (a process involving several activities).

The types of coordination relationships between different MTs are described for the general case, taking the study process management hierarchy as an example (Figure 8). It is appropriate to divide coordination interactions C(i, j) into three types:

(a)

The type of coordination interaction “impact“: the flow C(i, j) sent by the higher-level MTi to the lower-level MTj is the information that creates an effect (an impact on the lower-level activity);

(b)

The type of coordination interaction “feedback”: the flow C(j, i) sent by the lower-level MTj to the higher-level MTi has state parameters of MTj (feedback and status information);

(c)

The type of coordination interaction “conversation”: flows C(j, i) and C(i, j), sent between two activities, MTi and MTj, at the same management hierarchy level, include communication and the exchange of information about the status parameters of each MT (status information).

The characteristics of the coordination relationships of activity MTj with activity MTi include the following:

(a)

Status data of activity MTj required by the higher-level activity MTi: C(j, i) = {<set of MTj state attributes>};

(b)

Impacts of the higher-level activity MTi on lower-level activity MTj: C(i, j) = {<set of MTj controls>};

(c)

Conversation content C(i, x) between activity MTi and other activities at the same hierarchy level (h): C(i, x) = {<set of MTi attributes>; <set of MTx attributes>}.

Note: each activity MTj is at a certain management level, which is indicated by the index h.

Using the defined types of characteristics, it is possible to systematically and consistently observe, measure, and analyze the relevant academic processes and the interactions of the activities that make them up, as well as conduct their evaluation. Various metrics can be used; activity performance indicators can be calculated and assessed according to the selected scale of values. Templates of characteristics defined in

Table 5. The taxonomy of activity characteristics.

Templates of Characteristic Types
External Characteristics of Activity MT	Internal Characteristics of Activity MT(F, P, A, V)	Coordination Flows Related to MT Activity (See Table 6)
Management level: h	Management level: h	Management level: h
Activity ID: MTj(h), D = (<description of MTj>; Input I = {i(1), i(2), …, i(n)}; Output O = {o(1), o(2), …, o(m)}.	Activity ID: MTj(h). Fj(S) = <specification of Fj>; Pj(S) = <specification of Pj>; Aj(S) = {a(1), a(2), …, a(m)}; Vj(S) = {v(1), v(2), …, v(n)}.	Activity ID: MTj(h). Coordination flows: Management level h* (h* ≠ h); Activity ID: MTi; Status data C(j, i) = {<set of MTj state attributes>}; Impacts C(i, j) = {<set of MTj controls >}; Conversation content C(j, i) = {<set of MTj attributes>; <set of MTi attributes>}.

can be linked to a specific evaluation methodology (measurement scale), and quantitative estimates (KPIs—key performance indicators) can be calculated. Specific sets of KPIs are determined by the institution and the external evaluation body.

Based on the organizational hierarchy of the study process in some universities and the analysis of the existing evaluation regulations, definite characteristics were revealed (see

Table 6. The study process characteristics refined through the analysis of regulations.

External Characteristics of Activity MT	Internal Characteristics of MT(F, P, A, V)	Coordination Flows Related to MT
Students knowledge control level h = 1, MT2, students in the learning process
MT2(1): D = <description of MT2>. I: characteristics of study conditions (resources) O: subject study grades, grade point average, and compliance of final theses	MT2(1): F: motivation and expectations P: study habits A: skills, dutifulness, and attentiveness. V: intellectual level	MT2(1) coordination flows with the following: Study program teaching MT6(2): state data: C(2, 6); impacts: C(1, 2), C(6, 2)
Department/staffing level h = 2, MT6, study program teaching
MT6(2): D = <description of MT6> I: characteristics of teaching conditions (resources) O: knowledge delivered; knowledge acquired by students	MT6(2): F: indicators of compliance of final theses P: teaching process characteristics A: state of teaching process V: requirements for the students knowledge	MT6(2) coordination flows with the following: Students in the learning process MT2(1): state data: C(6, 2) = {MT8-10, MT8-19,…, MT8-33}; controls (impacts): C(2, 6) = {MT8-34, MT8-35, MT8-36. MT8-37} Study program design MT4(3): C(6,4) = {MT8-01, MT8-08}. Controls (impacts): C(1, 6) = {MT8-01, MT8-04,…, MT8-70}, C(8, 6) = {MT8-67, MT8-68,…, MT-71} Conversations: <not defined>
Department/SP committee level: h = 3, MT4, study program design
MT4(3): D = <description of MT4> I: characteristics of SP needs and qualification requirements O: content of SP	MT4(3): F: study program content characteristics P: study program development state A: assessment of SP completeness V: SP content changes	MT4(3) coordination flows with the following: University administration MT1(4): controls (impacts): C(1, 4) = {MT8-41, MT8-51,…, MT8-62} Study program teaching MT6(2): state data: C(6, 4) = {MT8-01, MT8-08} Department/SP committee level h = 3: conversations: <not defined>
Institution management level: h = 4, MT1, university administration
MT1(4): D = <description of MT1> I: characteristics of entries O: characteristics of graduates	MT(1): F: study management rules and criteria; P: study process characteristics; A: characteristics of studies status; V: impacts to study process	MT1(4) coordination flows with the following: External study evaluation MT8(5): state data: C(1, 8) = {MT8-01, … MT8-71}; controls/impacts: C(8, 1) = {MT8-01, …, MT8-71}

), and examples of the required coordination interactions of study process activities are presented in

Table 7. The required coordination interactions of study process activities (status of “as-is”).

Academic Process Activities (MTs)	From/To	MT1	MT2	MT3	MT4	MT5	MT6	MT7	MT8
University administration	MT1								C(8, 1) = {MT8-01: MT8-71}
Student in the learning process	MT2	C(1, 2) * = List-C(1, 2)					C(6, 2) * = List-C(6, 2)
Analysis of market needs (department, SP committee)	MT3	C(1, 3) = {MT8-62}			C(4, 3) = {MT8-18}	C(5, 3) = {MT8-18}
Study program design (department, SP committee)	MT4	C(1, 4) * = List-C(1, 4)					C(6, 4) = {MT8-01, MT8-08}
Prediction of study content (department level)	MT5	C(1, 5) = {MT8-14, MT8-18}			C(4, 5) = {MT8-14, MT8-18}
Study program teaching (department, professors)	MT6	C(1, 6) * = List-C(1, 6)	C(2, 6) = List-C(2, 6)						C(8, 6) * = List-C(8, 6)
Microcredit development (department, SP committee)	MT7	C(1, 7) = {MT8-62}
External study evaluation (ministry level)	MT8	C(1, 8) = {MT8-01 − MT8-71}

* List-C(1, 2) = {MT8-19, MT8-40, MT8-64, MT8-61}; List-C(1, 4) = {MT8-41, MT8-51, MT8-57, MT8-59, MT8-62}; List-C(1, 6) = {MT8-01, MT8-04, MT8-05, MT8-08, MT8-10, MT8-14, MT8-33, MT8-35, MT8-47, MT8-50, MT8-51, MT8-57, MT8-59, MT8-65, MT8-67, MT8-68, MT8-69, MT8-70}; List-C(2, 6) = {MT8-34, MT8-35, MT8-36. MT8-37}; List-C(6, 2) = {MT8-10, MT8-19, MT8-30, MT8-34, MT8-33}; List-C(8, 6) = {MT8-67, MT8-68, MT8-69, MT8-70}.

(according to management levels indicated in Figure 8).

5. Analysis of the Existing Study Process Assessment Methodology

5.1. Study Process Activities in the Existing Regulations

The coordination interactions between the different academic activities established in the academic regulations (at the time of our study in one of the universities) are summarized in Table 7. The coordination interaction C(i, j) in cell (i, j) provides the identifiers of the required information (characteristics) that academic activity MTi (column name i) must transmit to academic activity MTj (row name j), as defined in some type of academic regulation (a document approved by a higher authority). For example, academic activity MT1 “University administration” sends coordination information C(1, 2) = (MT8-19, MT8-40, MT8-61, MT8-64) to academic activity MT2, “Student in the learning process”. These are the mandatory requirements in regulations MT8-19, MT8-40, MT8-61, and MT8-64 (coordination impact). Academic activity MT1 transmits the characteristics and indicators specified in the external evaluation methodology of study areas to all other university departments (for activities MT2–MT7) via C(1, 3)–C(1, 7). However, the meaning of coordination flow C(1, 8) is different; it is a report (feedback) from an academic institution (identified as MT1) to a higher organization (MT8) identified as MT1) and includes data for mandatory indicators that were specified through coordination flow C(8, 1).

Examples of mandatory characteristics of academic activity, which are defined in the current regulations [7], are presented in Appendix A (part of the list). Table 7 presents the analysis of the study process based on the causal model, which shows that activities MT1–MT8 and the content of their interaction determined in the academic regulation of mandatory information (characteristics) are transmitted. The analysis of the “as-is” model reveals several important features:

-

The regulation includes well-defined requirements (controls) for institution C(8, 1), C(8, 6) sent by the external study evaluation unit and controls sent to units by the university administration: C(1, 2), C(1, 3), C(1, 4), C(1, 5), C(1, 6), and C(1, 7);

-

The regulation includes well-defined feedback (status data) from the institution to the external study evaluation unit: C(1, 8);

-

The regulation includes only a few instructions to provide data on the coordination of internal activities: C(4, 3), C(4, 5), C(5, 3), C(6, 2), and C(6, 4).

Comparing this model of the existing system (Table 6 and Table 7) with the proposed causal model (Table 3) and the taxonomy of activity characteristics in Table 5, the existing evaluation system has the following characteristics:

-

It includes process analysis and the external characteristics of the activities involved in the process but does not include an analysis of the activity structure (internal elements), and characteristics of activity elements are not included in the analysis and evaluation.

-

It does not include the analysis of the coordination of the activities that make up the process and the characteristics of mutual interactions that affect the quality. Empty columns MT3 and MT7, as well as sparsely filled columns MT2, MT4, and MT5, show this.

The presented example of study process analysis based on the causal model reveals the possibilities of this method to design systematically ordered processes, specify activities using predefined structures (templates), and create consistent evaluation methods of the entire process and its components (activities).

5.2. Comparison of the Academic Process Evaluation Models

The study process model created using the causal modeling method allows for systematizing the internal processes of the institution and their interactions. By comparing our proposed (through expert evaluation) causal model of study process management (“to-be” model in Table 3 and Figure 8) with study process activities defined in the existing regulations (“as-is” model in Table 7), systematic differences are identified. They show the differences between the implementation and monitoring of the existing study process from the desired (target) study process management (monitoring and evaluation) model. Of course, each institution can create an original model for monitoring and evaluating the study process using the causal modeling framework and conceptual framework we offer. After systematizing the interaction of mandatory “as-is” activities of study process management (Table 7) and comparing it with our constructed causal model (Table 3), it becomes clear that only a small part of the internal interactions is regulated in the documents of the academic institution. The main observations are as follows:

-

Column MT2, a student in the learning process, contains only one interaction C(2, 6), which shows that the student’s opinion is polled only according to the external study evaluation instructions; the institution itself does not define the rules and format of students feedback. This is left to the department’s competence (or teachers).

-

Columns MT3 and MT7 are empty. This shows that the activity (MT3) analysis of market needs and the (MT7) development of microcredits are not systematically integrated, since the institution itself does not define the rules and format.

-

Study quality improvement activities such as (MT4), study program design (department level, SP committee), and (MT5), the prediction of study content (department level), are weakly integrated, indicating unregulated cooperation with other activities.

Above is a small example of the comparison of the created causal model with the methodology applied in practice, which allows for the detection of the gaps in the methodology used for evaluating universities’ academic performances.

6. Conclusions

Academic high schools’ (universities and colleges) teaching and research activities must be coordinated by determining the necessary interactions to ensure the efficiency and quality of results. A causal modeling approach was developed for enterprise management modeling and application software development [8,9]. In this study, causal modeling was applied to analyze the academic institution’s performance. The focus is on creating a study process management and control model based on the discovered causal dependencies between academic activities. Academic activities are considered self-managed processes and are defined as management transactions with a predefined internal structure. Any academic activity is defined as a management transaction MT = (F, P, A, V), where F is the management function (an information-transformation and decision-making step), and P is a physical process that converts inputs (resources, energy, materials, and students) into outputs (products, services, and graduates). The management function F monitors and controls process P via a feedback loop involving the state attributes A of process P and the controls (effects) V in process P. Each activity defined as an MT is also associated with a goal description that is not strictly specified at this level of model detail. In this way, the internal structures of all academic activities and internal information interactions between parts of their structure are unified (normalized). This makes it possible to create templates for specifying the internal elements of all academic activity types. In this way, the information (data, knowledge, and goals) about each necessary academic activity is systematized. The main principles of constructing a causal model of academic activities by studying the dependencies of these activities are presented and illustrated herein. The study management process of one university was studied in detail, and the most important types of activities, MT1–MT8, were identified. A causal model of the hierarchy of academic activities enables the coordination of the study process and the identification of the interactions necessary to ensure the efficiency of studies and the quality of results. The importance of the informational (coordinating) mutual interactions of the activities of the recommended model was assessed using a survey of experts (staff members) and students (separately) on a scale from 0 (unnecessary) to 5 (determining). The model of the study process created using this method is an example of a causal system that allows for rethinking the coordination of the study process according to the goals set by the institution. The causal model determines the content of interactions between processes and activities and considers the levels of the institution management hierarchy (Figure 8). The content (and data exchange) compatibility of all activities can be systematized according to the causal model. The principles of creating and using the causal model allow for analyzing the existing study process evaluation methodologies, objectively evaluating the importance of the indicators defined in these methodologies, and detecting specific evaluation methodology and communication gaps in the institution’s internal processes. The causal model of existing coordination interactions was revealed (Table 7) after the existing regulations were analyzed. The comparison of the created causal model with the methodology used in practice allows for the detection of the shortcomings of the existing methodology used for evaluating the academic performances of universities. The comparison allows for a systematic analysis of regulations and the consistent identification of the relevant aspects of academic activity and new indicators.

The presented method of causal modeling reveals the possibilities of design processes organized according to causality, specifying the content of activities using predetermined structures (templates) to create consistent evaluation methods of the entire process and its components (activities). This method provides a conceptual basis for the monitoring of academic processes and the formation of a system of performance indicators based on a causal interaction model.

The method itself enables the discovery of organizational process management information (data and knowledge) and can be applied to modeling various organizational systems (enterprise modeling domain), as it is not only for academic institutions. The method and developed constructs (MT and EMC frameworks) can be applied to other types of organizations (production, governmental, non-governmental, business organizations, etc.) by analyzing and evaluating their process management. However, at the same time, the development of such analytical abilities may also limit the application of the proposed method, as the destination organization must reach a certain level of maturity in the field of process management and thoroughly master the MT and EMC concepts presented here.

Activity-specification templates based on a causal model are essential in building process modeling software with AI features to help analyze and evaluate business domain activities.