Predictive Analytics and Decision-Making in Higher Education Report

This report delves into the realm of predictive analytics and inferential statistics, specifically focusing on their applications within higher education. It begins by defining big data and its characteristics, highlighting the challenges conventional database techniques face in handling such large datasets. The report then explores the four categories of big data analytics: descriptive, diagnostic, predictive, and prescriptive, emphasizing the significance of predictive analytics in forecasting future trends and informing decision-making processes. The paper examines the use of predictive analytics in healthcare and higher education. It references various systems and tools, such as Hortonworks, Apache Hadoop, and Tableau, and stresses the importance of predictive modeling combined with statistical methods in decision-making. The report also addresses the ethical considerations surrounding the use of big data, including privacy and security concerns, and concludes by highlighting the need for further research in evolutionary methods and predictions from diverse data sources.

Running head: PREDICTIVE STATISTICS 1
Predictive Analytics and Inferential Statistics
Student’s Name
Course
Institution
Date of Submission

PREDICTIVE STATISTICS 2
Introduction
At present, data refers to excessively fast and big. Conventional database techniques,
therefore, cannot process such data efficiently. Thus techniques to manage, distribute, store,
capture, and analyze diverse data is known as big data. The key features of big data involve
volatility, validity, variety, veracity, velocity, and volume.
Big data process refers to the process of organizing, gathering, and analyzing huge sets of
data known as the big data to discover useful information and patterns, and Big Data is important
since it can assist the company in comprehending the information contained in data. Big data
analytics is important in identifying the data that is most crucial to the business as well as future
decisions of the business (Chiang, 2012). Analytics who work with big data normally want the
information or knowledge coming from data analysis. Big data analytics is divided into four
categories: These are Descriptive, Diagnostic, Predictive, and Prescriptive (Chiang, 2012).
The Prescriptive Analytics
Predictive analytics make use variables that can be analyzed and measured to predict the
behavior of machinery, individuals, or any other entities. For instance, an insurance company is
capable of taking into consideration the safety variables such as driving record, type of vehicle,
location, gender, and age, when issuing and pricing auto insurance policies. Many variables are
incorporated into a predictive model to assess future probabilities with a certain level of

PREDICTIVE STATISTICS 3
reliability. Predictive analytics relies on methodologies, algorithms, such as decision trees, time
series analysis, and regression models.
Prescriptive analytics shows the actions that ought to be undertaken. Predictive
analytics refers to the analysis of situations or scenarios of what is likely to happen (Chiang,
2012). It is also the use of algorithms, data, and machine learning methods to identify the
chances of results of future based on past data (Chiang, 2012). It uses known results to create a
model that can be utilized to forecast the values of new or data. It involves modeling results in
predictions that denotes the probability values for new or different data. The predictive analytics
is used to predict the behavior, drive decision making, improve performance, and predict trends.
Prediction in health care
YiChuanWang concluded that big data come as a result of controlling big volume of data.
Data from different sources are structured and arranged in the first layer and are obtained and
converted by use of transformation engines and kept. The data is mapped in the analytics and
processed. Cases from a different area under different parameters are gathered, which are then
analyzed (Chiang, 2012). The analysis is done according to a particular criteria map decreased
algorithms by use of Apache Hadoop.
Today, Higher Education is operating in a competitive and complex environment, and
thus, has to face its problems. Different stakeholders in higher education have created a way

PREDICTIVE STATISTICS 4
for big data to play an important role. The huge amount of data that keeps coming in every day
can be used only via big data (Baillie, 2015). Available information on social media and
generated content on social media can be utilized to solve everyday life problems. Using Big
Analytics will reframe our comprehension of the Higher education field. This paper attempts to
discuss the application of big data in decision making, particularly in higher education.
Systems such as Horton works, Apache Hadoop, tableau, and map-reduce are important
and powerful programs that can be utilized without prior technical skills or knowledge. In higher
education, big data can have impending impact on theory of decision making; huge available
operational and administrative data can be assessed and processed to forecast future performance
as well as locate areas in learning, research, and academic programming (DeTure, 2013). ).
Predicting modeling combined with a huge scale, statistical methods is important in decision
making. Big data plays a vital role in Predictive model
The predictive analysis attempts to figure out and unravel the possibilities. Based on
predictive analysis, processed data is given to stakeholders and students. The importance of data
will be based on the creation of leading structures as well as the creation of better and
progressive policies (DeTure, 2013). Privacy and security are some of the challenges that come
with predicting big data. Buxton and Greenberg emphasized that there is need for higher

PREDICTIVE STATISTICS 5
education to advance its way of doing things. Information technology has to utilize to apply
advanced techniques to analytics in backing up management and making the decision.
The research community, in a similar context, has to bring transparency to efficient work
of analytics to prevent bad uses of available data in online courses. According to Kellerman,
showed that accreditors are trying to set up pace with new rules of federal to give oversight on
online platforms that need colleges to verify that learners learn as much in face to face courses
and distance courses (DeTure, 2013). Such requirements put pressure on institution to oblige
new rules and to give appropriate evaluations of quality education. Furthermore, the trainer
should be known what is going on in the course; the utilization of predictive analytics will
produce information concerning the progress of the student as well as the instructional process.
Siemens stressed that the online community needs guiding the way as to how analytics
are utilized in evaluating and defining data in courses. This issue involves the need for data
defining, new analytics tools and methodologies, sharing and visualizing the nature of analytics
output, and setting up efficient practice that leads to making decisions concerning the
performance of the learner (Sinn and Scamp, 2019).
Conclusion
Complicated tasks have become common for various users. This paper discussed key
applications that depend on the solutions of predictive analytics. Nevertheless, instead of looking

PREDICTIVE STATISTICS 6
the application cases as well as the computing of the user, available frameworks are made mainly
for working with big datasets in different applications. Thus, future research will focus on
distinct evolutionary methods and predictions from different data.

PREDICTIVE STATISTICS 7
References
Baillie, A. J. (2015). Predictive gender and education bias in Kessler's psychological distress
scale (K10). Social Psychiatry and Psychiatric Epidemiology, 40(9), 743-748.
Chiang, H. M. (2012). Predictive factors of participation in postsecondary education for high
school leavers with autism. Journal of autism and developmental disorders, 42(5), 685-
696.
DeTure, M. L. (2013). Investigating the predictive value of cognitive style and online technology
self-efficacy in predicting student success in online distance education courses (pp. 1-66).
The University of Florida.
Suinn, R. M., & Oskamp, S. (2019). The Predictive Validity of Projective Measures.