Business & Management Studies: An International Journal
Yazarlar: Zeynep AYTAÇ, Hasan Şakir BİLGE
Konular:-
Anahtar Kelimeler:Big Data Big Data and Education Büyük Veri,Büyük Veri ve Eğitim
Özet: LITERATURE Big data is defined as “Datasets whose size is beyond the ability of typical database software tools to capture, store, manage and analyze (Manyika et al., 2011). Big data’s 5V characteristics are volume, velocity, variety, veracity, and value (Gahi, Guennoun & Mouftah, 2016). RESEARCH SUBJECT In 2015, there were more than 2 million 350 thousand job postings for data science and analytics jobs in the USA. It is estimated that there will be close to 2 million 720 thousand job postings by 2020 (Miller & Hughes, 2017). Nearly 60 universities have data science programs at the undergraduate level in the USA (DataScience, 2020). In Turkey’s Strategy and Budget Presidency’s Strategic Plan (2019-2023), “Increasing the capacity of expert/ data scientists experienced in big data analysis and strengthening the infrastructure” is defined as a requirement (Strateji Ve Bütçe Başkanlığı, 2018). Accordingly, in this study, the expectations of public and private institutions from universities and students regarding the big data field were evaluated. RESEARCH PURPOSE AND IMPORTANCE This study aims to examine the use of big data analytics in Turkey's public and private institutions and determine the institutions' expectations. Within the scope, the most preferred big data tools and the expectations the field of big data have towards students and universities in Turkey were determined. This study is essential in terms of being a resource for determining the expectations of institutions from universities and students in big data. CONTRIBUTION of the ARTICLE to the LITERATURE The evaluation of public and private institutions’ expectations from universities and students in Turkey is expected to contribute to immense data analytics literature. DESIGN AND METHOD RESEARCH TYPE This study is an applied research and was carried out through exploratory research. RESEARCH PROBLEMS In this study, the following research questions are defined. What are the big data tools and technologies preferred by public and private institutions that offer big data solutions? What are the expectations of public and private institutions that provide big data solutions from students studying in computer science and related departments and universities in the field of big data? DATA COLLECTION METHOD The Snowball sampling method was chosen in the interviews. The criterion in sample selection is that the institution offers big data solutions. Meetings were held with the general managers or prominent data project managers of ten institutions. Considering the adoption of big data by sectors, institutions in different sectors have preferred the research. In this study, interviews were conducted with the prominent data department managers of two public and eight private institutions that offer big data analytics solutions in Ankara. QUANTITATIVE / QUALITATIVE ANALYSIS Qualitative research methods, namely the grounded theory approach, was used to analyze the data. The transcription of the voice recordings was collected to analyze the interview data. ter the interview transcripts were examined, they were subjected to the coding process using NVivo 12 software. RESEARCH MODEL A qualitative research method was used in this study. The data obtained from the interviews were generalized inductively step by step. In this study, data collection and analysis were carried out together. After the data was collected, the concepts and processes that emerged with the analysis were included in the next data collection stage. FINDINGS AND DISCUSSION FINDINGS as a RESULT of ANALYSIS The findings obtained with the data collected from the institutions were stated under three main parts: the current situation of the big data analytics of the institutions, their expectations from the universities and students, and their suggestions. The current status of the big data analytics category of the organizations covered the positions preferred by the big data team, the number of employees in these teams, the adequacy of this number, and the technologies they use. The category of expectation included the undergraduate programs preferred in big data and the abilities and skills expected from the graduates. The suggestion section included suggestions for students considering working in this field and for universities. It was defined that a wide variety of position names were used, and it was revealed that the most preferred position name was a software engineer. It has been observed that most of the institutions use Hadoop Ecosystem. Spark, Cassandra, Hive, and Kafka are among the most preferred tools. The most preferred departments by institutions for recruitment are Statistics, Mathematics, and Computer Science. Among the skills expected from graduates, awareness and analytical thinking were the most expressed as the interviewed institutions' technical skills expectations are the database skills (SQL) and necessary programming skills. Among the suggestions of the interviewees to the students are to have awareness, to make an extra effort other than what is taught at school, to work part-time or internship, to have continuity of learning and to be eager to learning, to participate in online courses, to participate in seminars, courses, and events. The interviewees' recommendations for universities include computer science, statistics, mathematics, and management information systems to renew the curriculum, long-term internships, and open more graduate programs. DISCUSSING the FINDINGS with the LITERATURE The U.S. Bureau of Labor Statistics states that positions covering data science and analytics do not have a clear definition. It is stated that positions such as data scientists and data engineering in data science and analytics can not be followed (Miller & Hughes, 2017). It was revealed that the institutions interviewed can not make a clear distinction between positions such as data scientists, data engineers, and software engineers. It was concluded that the tools used by the institutions and the tools used around the world (HDFS, Apache Hadoop, HBase, Cassandra, Hive, Oracle Big Data ve R) (Bhadani ve Jothimani, 2016) are similar. The institutions' recommendations are to be aware of the development of big data and analytics and follow innovations. The recommendations of the institutions for universities are to renew the curriculum and to open more graduate programs. Lawler and Molluzzo (2015) stated that with a good curriculum design, students studying in computer science and information system departments could contribute to big data analytics. CONCLUSION, RECOMMENDATION, AND LIMITATIONS RESULTS of the ARTICLE It has been concluded that data scientists, data engineers, and data analysts could not be made clearly by institutions, yet when determining employees' positions in big data projects. It has been concluded that they generally prefer the position as a software engineer. Apache Hadoop, Spark, Hive, Kafka, and Cassandra were the most used tools by the interviewed institutions. It has been revealed that the institutions prefer the personnel who work in this field mostly from computer science, then mathematics, statistics, and management information systems. It is concluded that the institutions have suggested having awareness about the big data field for students and curriculum renewal for universities. SUGGESTIONS BASED on RESULTS It is recommended to open more graduate programs and undergraduate programs in interdisciplinary fields such as data science and redesign the relevant departments' curriculum to include big data analytics. The renewal of the curriculum will help institutions make more precise decisions in determining positions and recruitment. Another suggestion is to have more course alternatives in the field of data science in related departments. It is also recommended for universities to add long-term internships to relevant departments. LIMITATIONS of the ARTICLE In this research, interviews were done with the prominent data project team leaders of public and private institutions that offer big data solutions in Ankara. One of the limitations of this study is the limited number of institutions that produce solutions in big data. Another limitation is to find participants in the criteria determined due to the workload of the institutions.