Decision Support Systems: A Content Analysis of Graduate Theses in Turkey Türkiye’deki Lisansüstü Tezlerin Betimsel İçerik Analizi

This study aims to conduct a descriptive analysis and evaluation of graduate theses regarding decision support systems carried out between 1989 and 2020 in Turkey. The qualitative research methodology was applied, the theses were analyzed through the descriptive content analysis technique. Forty-eight graduate studies accessible from the national thesis center database of YÖK were included in the analysis. The theses were coded according to the date of publication, university, institute, department, degree level, the academic title of the thesis supervisor, thesis language, research methodology, research sub-areas. Graduate studies on DSS have increased in the last 15 years, and 42 studies have been conducted in the last 15 years. Selçuk University ranks first with the most studies on DSS. Half of the studies were carried out in the Institute of Science, and most of the studies that were produced in the universities were master's theses. Most supervisors were "Prof. Dr." titled faculty members. Most of the theses were written in Turkish, and primarily experimental studies were conducted. The business administration department produced most of the theses on DSS. Business and environment studies were the primary disciplines that produced theses. These were carried out in 7 institutes and 26 different departments. The findings of this study will guide other researchers who are willing to work in the decision support systems field.


INTRODUCTION
Information systems have become more significant to support managerial decisions in recent years. Companies' decisions involve the management and use of information systems and data interpretation from the business and its environment. The essential characteristic of a manager is her/his ability to make fast and correct decisions. Among the managers' duties, the most crucial task is to make the best decision in the shortest time possible. A good decisionmaking process can be achieved by producing accurate, UpToDate, and timely information.
Today, managers have to comprehend their organisation's capabilities and uses of information significantly to comprehend budgeting financial resources (Pearlson, Saunders and Galletta, 2019).
Decisions are classified as unstructured, semi-structured, and structured. Unstructured decisions "are those in which the decision-maker must provide judgment, evaluation, and insight to solve the problem" (Laudon & Laudon, 2020). Each of these decisions is novel, essential, and non-routine, and there is no well-understood or agreed-on procedure for making them. On the other hand, structured decisions "are repetitive and routine, and they comprise a definite procedure for handling them so that they do not have to be treated each time as if they were new" (Laudon & Laudon, 2020). Many decisions have elements of both types of decisions and are classified as semi-structured decisions. In semi-structured decisions, only a part of the problem has a definite answer provided by an accepted procedure (Laudon & Laudon, 2020). A decision support system (DSS) is "an information system at the organisation's management level that combines data and sophisticated analytical models or data analysis tools to support semi-structured and unstructured decision making" (Laudon & Laudon, 2018). Gorry and Scott Morton, in 1971 first mentioned DSS, and it has been widely used in many applications (Gorry and Morton, 1971). DSS is intended to support decision-makers to assist and improve their decisions regarding the process and the outcome of their business activities (Laudon & Laudon, 2020, Biswas, 2020. DSSs serve the middle and higher management, operations, and planning levels of an organisation. DSS do not replace the decision-maker or the manager; they are the systems that only assist and support the decisionmaker in their decision (Rainer, Prince, and Watson, 2017).
Information as a vital source of the development of competitive advantage is crucial for businesses to be managed. Thus, businesses have to accept this change and continuously renew themselves to compete with their rivals in rapidly developing competitive conditions. Otherwise, they are doomed to have difficulty competing with their rivals and disappear (Rainer, Prince and Watson, 2017). Increasing the opportunities of any business to gain the aforesaid competitive advantage is possible using DSS in critical decision-making conditions. agricultures (Hafezalkotob, Hami-Dindar & Rabie, 2018), fisheries and marine affairs (Hozairi & Krisnafi,2017). This tendency suggests that multipurpose DSS applications support the decision-making process (Teniwut & Hasyim, 2020).
Decision Support Systems as a fundamental topic needs to be handled with various branches of science. This study aimed to investigate academic work at the graduate level that has been done on the decision support system over the last 31 years in Turkey. A total of 48 Master's and Doctoral theses were analysed, and the findings are resented in this paper.
This study is significant as it examines the postgraduate studies conducted in the field of DSSs in Turkey. It provides descriptive information about the nature of the studies and offers recommendations for future studies on decision support systems.

Research Questions and the Design of the Study
This study addresses the following research question and its sub-questions. This study employed descriptive content analysis of the qualitative research approach.
Content analysis is a process of summarising and reporting written data. The researcher reads, organises, and digitises the data per the codes, categories, and themes previously created (Dawson, 2019). Thus, researchers gained the opportunity to interpret data from the existing sources using the keywords in the searched studies. They were able to show the frequency of the importance of the topic under investigation.

Population
Purposive sampling was used as a sampling strategy. The population consists of theses and dissertations accessible from the YÖK's "National Thesis Centre." The first dissertation on decision support systems was published in 1989 in the National Thesis Centre. Therefore, the study's time frame was determined as from 1989 to 2020 to include all published theses accessible from the centre on DSS. Theses within the scope of the research were downloaded between 01.12.2020 and 30.12.2020 by the researchers. 48 Theses, 32 of which were masters' theses, 15 were doctoral dissertations, and only one was a PhD thesis on medical speciality.

2.3.Data Collection Procedure and Analysis
During the data collection procedure, the keywords for the DSS concept were entered both in Turkish and English in the search engine as "karar destek sistemleri, decision support systems". 72 Theses were located in the National Thesis Centre, 48 of which were open to access (Appendix 1). All 48 theses with access permission within the field of decision support systems registered in the database of YÖK were included in the study. Out of 48 theses, 32

Decision Support Systems: A Content Analysis of Graduate Theses in Turkey
Hüseyin GÖKAL, Volkan CANTEMIR, Ahmet ADALIER 16 were masters' theses, 15 were doctoral dissertations, and only one was a PhD thesis on medical speciality. 24 theses out of the initially available 72 were eliminated from the study because their full-texts were not accessible due to the access limits placed by the authors. The figure 1 shows the data collection procedure visually.
The theses were uploaded to the computer in PDF document format. The unit of analysis was identified as descriptive information regarding the nature of the graduate theses on DSS. Each thesis was assigned with numbers starting from 1 to 48. The abstracts and full texts of 48 theses were read and analysed thoroughly. Nine codes for the analysis were identified. The codes are as follows: the year, the university, the institute, the department, the degree level, supervisors' academic title, language, the research methodology, and the research sub-area. Then, the frequency of each code was counted. The information on the frequency of the codes enabled the researcher to compare them. The comparison showed the focus of academic work regarding DSS over 31 years.

FINDINGS
In this section, the descriptive findings of the graduate theses on DSS written between 1989 and 2020 are presented. The graphs presented below involve descriptive information about the year, the university, the institute, the department, the degree level, supervisors' academic title, the theses' language, the research method, and the research sub-areas.    As shown in Figure 6, regarding the degree levels of theses produced in universities on DSS, 32 (66.67%) were the master's theses, 15 of them were doctoral dissertations (31.25%), and only a single (2.08%) PhD dissertation in medical specialty was carried out according to YÖK's "National Thesis Centre".

Figure 7: Graduate Theses According to Advisors' Academic Titles
As shown in figure 7, 25 (52.08%) of the theses were supervised by "Prof. Dr." titled faculty members. In other words, more than half of these thesis studies are supervised by "Prof.
Dr." titled faculty members. Faculty members with the titles of "Assoc. Prof. Dr." supervised 9 (18.75%) theses. The rest of the 8 (16.67%) theses were supervised by "Assist. Prof. Dr.", and 6 (12.50%) were supervised by "Dr." titled faculty members.  As shown in Figure 9, experimental studies constitute the maximum with 20 (41.67%) studies, 15 (31.25%) studies were carried out quantitatively, and 9 (18.75%) were conducted qualitatively. Only 4 (8.33%) of the theses followed a case study design. As shown in Figure 10, most studies on DSS were related to business. 9 Studies (18.75%) were conducted in business. 8 (16.67%) studies were carried out in environmental studies. A total of 31 studies were conducted in the sub-areas of banking, health, military, logistics, information technology, maintenance, disaster management, warehouse management, map space, model and method, organization, marketing, planning, problem-solving, restoration, student affairs, production, data mining, web base, artificial intelligence, and management.

DISCUSSION AND CONCLUSION
According Dr." titled faculty members. Most of the theses were written in Turkish, and most experimental studies were conducted. The business administration department produced most of the theses on DSS. Business and environment studies were the primary disciplines that produced theses.
Initially, more studies were expected to be found in other areas such as marketing and student affairs in higher education. In such areas, DSS could ensure a competitive advantage.
Inevitably DSS is a tool for businesses to improve as management gets the opportunity to develop their organizations by making appropriate decisions with the support of the system. Discovering that there were only nine studies over 31 years in the business field was surprising.
According to the results of this study, the number of master's or doctorate thesis are very few in healthcare/medicine subjects in Turkey. In the last few years, substantial progress has been made in artificial intelligence field and the machine learning context. Features provided in practical applications range from supporting the user while making decisions, e.g., in a recommender system, to making decisions fully autonomously. Applying machine language algorithms to new large datasets can expose novel tendencies and relationships that may have practical effects for clinical practice in medicine/healthcare decision support systems (Brusko, Kolcun, & Wang, 2018). Researchers have studied the application of machine language methods in healthcare decision support systems and have demonstrated the significant impact of machine language in making enhancements to healthcare safety, quality, and DSSs (Buchlak, Esmaili, Leveque, Farrokhi, Bennett, Piccardi, & Sethi, 2019;Miotto, Wang, Jiang, & Dudley, 2018;Liang, Zhang, Huang, & Hu, 2014). Therefore, the universities should focus on doing more research in healthcare subjects, either master or doctorate level.

23
Arnott & Pervan (2016) indicated in their research that; using information technologybased systems to support managers and senior personnel's decision-making activities has been a significant feature of information system (IS) research and practice since the IS discipline appeared in the 1960s and 1970s. In the future, this trend will continue as especially business intelligence is presently rated as the most vital information technology subject for CIOs worldwide, and DSS research is currently over 10% of the IS discipline (Arnott & Pervan, 2016). The decision support system is an important research topic for many various disciplines and departments. Many departments in various disciplines worked on this subject. They revealed the importance of decision-making and decision support systems, regardless of discipline, program, and faculty. Remarkably, most of the studies conducted were experimental.
Today, the increase in complexity of machine learning techniques based on deep learning provides enlightenments for scientists to comprehend the consequences of these methods is appealing essential in decision support systems.

Recommendations
Future research could concentrate on the studies published in internationally indexed journals. Comparisons concerning the methodology, field of studies, and studies' contributions might be analysed. Postgraduate students, namely masters and doctorate, could be encouraged to review DSS studies in different countries and write comparative studies.
Such studies could contribute to both practitioners and decision-makers. Moreover, the research scope can be expanded by examining the articles on the subject at the national and international levels. More studies can be written jointly by various disciplines. As mentioned earlier, the new developments in the world lead all companies and countries to concentrate more on the decision making as technology is the main driving force of the developments and competitive advantage. Therefore, studying DSS in the fields such as transportation, higher education, and service industries is necessary.

Limitations
This study is limited to postgraduate studies, which could be accessed from the Thesis