Page 204 - Contributed Paper Session (CPS) - Volume 6

CPS1883 Christina A. et al.
interchange of the activities taking place in different locations. We can express
the flipped classroom as an approach, where initial out-of-class activities
include individual, often computer-based preparatory activities, followed by
in-class group-centered interactive actions (Bishop and Verleger, 2013). This
means that a kind of more passive knowledge acquisition takes place outside
the classroom and hands-on related activities are dominating inside the
classroom (Lockwood and Esselstein, 2013). The flipped classroom strategy
has been described and evaluated, mostly with a positive result, in many
reports (Strayer, 2012; Bergman and Sams, 2012; Roehl, 2013), including the
application to undergraduate statistics courses (Wilson, 2013; Touchton, 2015;
Cilli-Turner, 2015; Winquist and Carlson, 2014; Chen et al., 2015, Phillips and
Phillips, 2016; Vidic and Clark, 2016).
Usually, the flipped statistics classroom has been applied in such a context
that the out-of-class activities take place without a physical classroom,
whereas the in-class activities take place in a real, physical classroom, where
the students and lecturers meet face-to-face (Wilson, 2013; Touchton , 2015;
Cilli-Turner, 2015; Winquist and Carlson, 2014; Chen et al., 2015, Phillips and
Phillips, 2016; Vidic and Clark, 2016). In our case, we show how the flipped
classroom can be applied to a statistics course, which is taught completely
online. This means that we still flip the classroom concerning the nature of the
learning process, i.e. the activities traditionally taught during the class take
place out of class and vice versa. But, the whole teaching framework of the
course is completely web-based, without face-to-face phases, in order to
ensure the students better possibilities for time and location independent
studies. In this paper, we present how the flipped classroom approach was
applied to an advanced course in multivariate data analysis, in order to reduce
the problems occurring in statistics courses by introducing active learning
components in the course. We describe the settings of the flipped online
course as well as lessons learned in the first in-class usage of the approach.

2. The Online Course Multivariate Data Analysis
In the era of data science, the statistics courses play an essential role in the
computer science curricula at both B.Sc. and M.Sc. level at Frankfurt University
of Applied Sciences (FRA-UAS). This is e.g. the case in the curriculum of the
program High Integrity Systems (HIS), which is a two-year international M.Sc.
program in computer science with an all-English curriculum (High Integrity
Systems, 2018). This degree gives the students many challenging professional
opportunities in different areas of software application. The concept high
integrity systems refers to complex systems, controlled by software, having a
large impact on humans and society in case of a failure, such as safety critical
systems, which are directly influencing the life of humans, as e.g. aerospace,
automotive, railway and marine systems or medical technology. We also talk

193 | I S I W S C 2 0 1 9