Effectiveness of Blended Learning in Nursing Education
Effectiveness of Blended Learning in Nursing Education
net/publication/339651952
CITATIONS READS
10 235
3 authors, including:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Maria consuelo Saiz-Manzanares on 10 March 2020.
Article
Effectiveness of Blended Learning in Nursing
Education
María Consuelo Sáiz-Manzanares 1, * , María-Camino Escolar-Llamazares 1 and
Álvar Arnaiz González 2
1 Departamento de Ciencias de la Salud, Facultad de Ciencias de la Salud, Universidad de Burgos, C/
Comendadores s/n, 09001 Burgos, Spain; [email protected]
2 Departamento de Ingeniería Informática, Escuela Politécnica Superior, Universidad de Burgos, Avda.
Cantabria s/n, 09006 Burgos, Spain; [email protected]
* Correspondence: [email protected]; Tel.: +34-673192734
Received: 15 January 2020; Accepted: 26 February 2020; Published: 1 March 2020
Abstract: Currently, teaching in higher education is being heavily developed by learning management
systems that record the learning behaviour of both students and teachers. The use of learning
management systems that include project-based learning and hypermedia resources increases safer
learning, and it is proven to be effective in degrees such as nursing. In this study, we worked with
120 students in the third year of nursing degree. Two types of blended learning were applied (more
interaction in learning management systems with hypermedia resources vs. none). Supervised
learning techniques were applied: linear regression and k-means clustering. The results indicated
that the type of blended learning in use predicted 40.4% of student learning outcomes. It also
predicted 71.9% of the effective learning behaviors of students in learning management systems.
It therefore appears that blended learning applied in Learning Management System (LMS) with
hypermedia resources favors greater achievement of effective learning. Likewise, with this type
of Blended Learning (BL) a larger number of students were found to belong to the intermediate
cluster, suggesting that this environment strengthens better results in a larger number of students.
BL with hypermedia resources and project-based learning increase students’ learning outcomes and
interaction in learning management systems. Future research will be aimed at verifying these results
in other nursing degree courses.
1. Introduction
In approximately the last decade there has been a marked interest in investigating ways of
teaching other than traditional face-to-face. The incorporation of technological resources such as virtual
platforms and hypermedia resources, combined with other innovative, methodological techniques
such as project-based or problem-based learning, have revolutionized the teaching–learning process.
The aim is to teach in the most efficient way possible and to make the most of resources while ensuring
sustainability. These technological and methodological resources have been applied to different
disciplines, especially in the field of health sciences (medicine, pharmacy, psychology, veterinary,
etc.). However, in recent years these resources have been incorporated into nursing studies. Next, an
approach will be made for the most relevant concepts of teaching in a virtual platform, which has been
called blended teaching, as well as the implementation of methodological resources for project-based
learning. Likewise, special importance will be given to its application for the formation of future
nursing programs by analyzing the pros and cons of this form of teaching and learning in the society
of the 21st century. For this reason, the most relevant concepts of these new forms of teaching and their
Int. J. Environ. Res. Public Health 2020, 17, 1589; doi:10.3390/ijerph17051589 www.mdpi.com/journal/ijerph
Int. J. Environ. Res. Public Health 2020, 17, 1589 2 of 15
specific application to the nursing degree will be dealt with below. The final objective of this work is to
study the effectiveness of different blended learning environments in the teaching of future nurses.
Twenty-first century society requires students and graduates to develop a series of skills related
to two important leitmotifs: collaborative work and operation of information and communications
technology (ICT). It is increasingly necessary to possess effective and rapid problem-solving skills
and to develop digital competences [1]. The use of learning management systems (LMS) is, therefore,
a reference in instructional practice, especially in higher education, as is the implementation of
collaborative work in these methodological settings for the resolution of tasks and problems. A good
example might be the use of project-based learning (PBL) methodology [2]. Recent investigations have
confirmed that if such a methodology is accompanied by the use of hypermedia resources (e.g., flipped
learning experiences, quizzes, use of wikis, on-line glossaries, etc.), then acquisition of deep learning is
strengthened in students [3]. Deep learning is a concept developed in the framework of the taxonomy
of Bloom [4]. It corresponds to the highest level of learning competences (comprehending, analyzing,
summarizing, and evaluating their own learning). One of the currents of thought in LMS learning
environments suggests that learning in these environments implies deeper learning from the point of
view of cognitive and metacognitive complexity, as these facilitate self-regulated learning (SRL) and
meaningful learning [5].
Likewise, LMS permit a more precise analysis of interactions which are logged in records (or logs).
The logs represent units of information and registration that stores precise data on the frequency of
user interactions and their duration [6]. LMS also facilitates the inclusion of hypermedia resources [7].
The use of these resources is especially relevant in health science degrees (nursing, medicine, pharmacy,
etc.) since it implements practical assumptions in the work, which has been proven to reduce errors in
the workplace [8].
There are various stages in this instruction process that will facilitate or inhibit the efficiency and
depth of the learning process. One of them is the design of learning tasks in LMS [9,10]. Another
essential element is that the teacher plans for process-oriented feedback [11].
Table 1. Preliminary elements to take into account for the design of learning activities.
1.2. Application of Artificial Intelligence Techniques to Analyze the Teaching and Learning Process
Development of the internet and information and communications technology (ICT) has expanded
learner access to information, and they have changed the way that information is taught and the way
it is learned [14]. A learning management system (LMS) is an interactive learning environment that
facilitates both teaching and learning. In addition, these software environments record all the actions
performed by the teacher and by the students, under individual and group headings. However, those
logs store a lot of data and learning analytics have to be used in order to study them in a flexible and
accurate manner. These techniques can be simple, such as the ones usually found in LMS (descriptive
statistics). However, more complex analytical techniques can be used, such as machine learning
techniques (a subset of artificial intelligence). The latter are analogous to the computational thought of
the human brain and operate with what is known as artificial intelligence. Machine learning techniques
of classification and clustering [15] are among the most widely applied techniques for data analysis in
educational environments. The use of these techniques for the analysis of both student and teacher
behaviors will provide the teacher and those responsible for educational institutions with ideas to
introduce improvements into the learning environment [11].
In brief, machine learning techniques are used, as these techniques are currently considered
to provide the researcher with more data in the field of cognitive psychology and learning than
traditional statistical techniques [16,17]. In particular, machine learning techniques permit personalized
learning and provide individualized information on the development of student learning. Prediction
techniques facilitate early detection of at-risk students and, therefore, personalized [18] help from the
teacher [19,20]. Machine learning techniques also provide information on the effects of predicting the
independent variable over each of the dependent variables in percent effects [21].
Int. J. Environ. Res. Public Health 2020, 17, 1589 4 of 15
1.4. Extraction and Analysis of Information on the Teaching–Learning Process Recorded in LMS
As we have mentioned earlier, development of the teaching through LMS will facilitate the student
in learning recording and follow-up behaviours [35]. Many of these learning managers use supervised
machine learning techniques, techniques such as multiple regression analysis (MRA), neural network,
and SVM. Those techniques help with the detection and subsequent prediction of successful and risk
behaviours. Behaviours of the students in LMS that have been related to successful learning are, among
others [23]:
Therefore, the frequency and systematicness of student interactions and their interactions with
LMS are directly related to effective learning [36]. Along these lines, recent investigations [23] have
revealed differences in predicting learning results in relation to the variable “teaching methodology”
(understood in terms of the pedagogic structure of the teaching, the evaluation procedures, and
feedback). The type of activities and the evaluation tests (quizzes, tests, projects, presentations . . . ) are
understood to determine the effectiveness of behavioural learning logged on the LMS.
As previously mentioned, application of machine learning techniques to study the logs will
allow the teachers to analyze the behavioural learning of their students and to detect at-risk students.
In these cases, early intervention will presumably improve student learning responses. Recent studies
have confirmed [17,35] that following up with student behavioural learning in the LMS facilitates the
identification of at-risk students with an explained variance of 67.2%.
In summary, the use of machine learning techniques will permit the study of behavioral learning
of both students and teachers on the platform, which will facilitate the application of prediction
techniques to the learning results [37]. Reviewing the investigations presented earlier, we consider
it important to study the behavior of PBL in the LMS. As has been indicated, there are few studies
in that field, and more information is needed that will help to improve teaching practices in these
environments [38]. Project work and personalization of learning in LMS have been proven to have
significant effects on the quality of learning. Particular relevance has been in Health Science degrees,
such as nursing or medicine, etc., since it facilitates work on clinical cases in a collaborative way and
optimizes the results applied to real learning contexts [39].
This research study was performed to analyze data of students’ online and face-to-face (F2F)
activity in a blended nursing learning course. We applied two types of blended learning: Blended
Learning type 1 [in which the interaction between the teacher and students is 80% in the LMS and 20%
Face to Face (F2F)] and Blended Learning type 2 [in which the interaction between the teacher and
students is 20% in the LMS and 80% Face to Face (F2F)].
In light of the above, the hypotheses in this study were the following:
H 1: The types of blended learning (Blended Learning type 1 vs. Blended Learning type 2) used
will predict student learning outcomes;
H 2: The types of blended learning (Blended Learning type 1 vs. Blended Learning type 2) used
will predict the learning behaviours logged on the LMS; and
H 3: The type of clusters will be different for each type of blended learning used (Blended Learning
type 1 vs. Blended Learning type 2).
2.1. Design
A quasi-experimental post-treatment design with an equal control group (in terms of metacognitive
skill) was used. Likewise, learning outcomes (learning outcomes in the development of project-based
learning; learning outcomes in exhibition of project-based learning; learning outcomes in the test; and
learning outcomes total) and behavioral learning in the LMS were the dependent variables (access to
complementary information; Access to guidance to prepare PBL; Access to theoretical information;
Access to teacher feedback; and mean visits per day).
2.2. Participants
A sample of 120 university students was assembled following the third year of their nursery
degree in Spain (the degree has four years) during one semester (9 weeks): 63 followed the Blended
Learning type 1 methodology and 57 followed the Blended Learning type 2 methodology (see Table 2).
Int. J. Environ. Res. Public Health 2020, 17, 1589 6 of 15
The students were assigned to each blended learning group (Blended Learning type 1 vs. Blended
Learning type 2) by means of convenience sampling. The work was developed in the subject of
“Quality management methodology of nursing services.”
Table 2. Group assignment and descriptive statistics for age, a n = 60. b n = 62.
Men Women
Groups
n Mage SDage n Mage SDage
Experimental Group, Blended Learning type 1 (a n) 7 23.29 2.56 56 22.30 2.13
Control Group, Blended Learning type 2 (b n) 9 24.67 4.12 48 23.83 5.13
Note. Mage = Mean Age; SDage = Standard Deviation.
2.3. Instruments
a. LMS UBUVirtual version 3.1. A Moodle-based learning management system (LMS) was used
that began with a constructivist approach and was developed through a modular system. It is a
personalized Moodle-based LMS. An LMS is a modular learning environment that permits interaction
and feedback between teacher and students, in many cases in real time, and in addition it facilitates
the process of automated feedback.
b. The (ACRAr) Scales of Learning Strategies by Román & Poggioli [40]. This widely tested
instrument identifies 32 strategies at different points in the information processing cycle. The reliability
indicators on the scale were between α = 0.75 to α = 0.90 and the indicators of content validity were
between r = 0.85 and r = 0.88. The subscale of metacognitive skills was applied in this study; this
scale incorporated 17 strategies about the use of metacognitive skills into the problem solving tasks.
A reliability index of α = 0.80 was obtained in this study; the reliability indicator on this subscale was
α = 0.90 and the indicator of validity was r = 0.88.
c. Student learning results: the results were recorded in the different evaluation procedures. (1)
Multiple-choice tests on the theoretical contents of the subject (test) were assigned a weight of 30%
of the final grade. The test had 10 multiple-choice questions (four possible answers) with only one
correct response. As well, five questionnaire-type quizzes were administered, one for each thematic
unit. Cronbach’s Alpha reliability of the test was α = 0.81. (2) Development of PBL, with a weight of
25%, was measured with a rubric, which can be seen in Supplemental Material Table S1. (3) Likewise,
the exhibition of the PBL, with a weight of 20%, was also measured with a rubric and can be seen in
Supplemental Material Table S2. In the final mark, Cronbach’s Alpha reliability of PBL was α = 0.62.
This result is lower because there was less dispersion among the scores in this type of evaluation test.
Since the performance of the groups was quite uniform, this aspect can be checked in the results section
and it is in accordance with the philosophy of PBL. Finally, the learning outcomes total covered the
weighted scores of all the results (over 10 points). 4) The students solved five practices, and this part
was 25% of the final grade. However, in this part all students had the highest qualification since the
teacher reviewed the practices continuously, and if they were not correct the teacher ordered them to
be repeated. Therefore, because it is not discriminate it has not been included in the analysis. Examples
of the PBLs developed can be found at this link https://riubu.ubu.es/handle/10259/3753/discover.
2.4. Procedure
Convenience sampling was followed for the choice of the sample. This was due to the possibility
of working with this methodology by a specialist teacher who attended to both groups, and in this
way the “type of teacher” effect was avoided. Before the instructional intervention, the two groups
(Blended Learning type 1 vs. Blended Learning type 2) were scored on the metacognitive skills
Scale of ACRAr [40], with the aim of establishing the similarities between both groups in terms of
metacognitive skills.
As stated in the introduction, Blended Learning type 1 was applied to the experimental group,
a learning environment in which the interactions between teacher and student were 20% F2F and 80%
Int. J. Environ. Res. Public Health 2020, 17, 1589 7 of 15
LMS. Likewise, Blended Learning type 2 was applied to the control group, a learning environment in
which the interactions between teacher and students were 20% LMS and 80% F2F. In the experimental
Group, hypermedia resources were used such as videos, and feedback was through the LMS. In contrast,
classroom interactions between teacher and students and feedback in the control group were all F2F. In
both groups, PBL methodology was followed. The difference, as has been pointed out, consisted of
the type of blended learning in use (Blended Learning type 1 vs. Blended Learning type 2). Project
development was done in both (the control and the experimental) groups in a collaborative way.
The project work was completed in small groups of students of between 2 and 5 members.
3. Results
Table 3. Indicators of asymmetry and kurtosis in the experimental group and control group.
Blended Learning Type 1 (Experimental Group) Blended Learning Type 2 (Control Group)
M SD A ASE K SEK M SD A AES K SEK
80 23.85 −0.464 0.441 −0.973 0.858 75 28.75 −0.333 0.306 −0.957 0.604
Note. M = Mean Age; SD = Standard Deviation; A = Asymmetry; K = Kurtosis; ASE = Asymmetry Standard Error;
SEK = Kurtosis Standard Error.
3.2. Previous Statistical Analysis of Homogeneity between the Groups before the Intervention
Significant differences between both groups (experimental and control) in their use of metacognitive
strategies were anlayzed before application of the different types of blended learning (Type 1 vs.
Type 2). To do so, a single-factor ANOVA with fixed-effects was performed (blended learning type) on
the results. No significant differences were found between both, so they can be considered similar
groups (F1 , 119 = 0.276; p = 0.601; η2 = 0.002) in the ACRAr subscale of metacognitive skills.
Int. J. Environ. Res. Public Health 2020, 17, 1589 8 of 15
Similarly, in order to study which type of supervised learning technique would be the most
appropriate, the Goodness-of-fit indices were measured in the structural equation modeling (SEM)
that was used to study the settings of the machine learning technique to predict the learning results.
The calculations were performed with the Statistical Package for the Social Sciences (SPSS) AMOS
v.24, as may be seen in Table 4, and no dependent relations between the observed values and the
different prediction methods (LR, DT, RBFN, and kNN) were found for any of the four prediction
models. Among these possibilities, the following were applied in the MRA.
3.3. Hypothesis 1.
MRA was performed to study the predictive value of the variable blended learning type applied to
the student learning outcomes. An R2 = 0.404 was found, which indicates that this variable explained
40.04% of the variance in the learning results. The Tolerance (T) values were within an interval of 0.106
and 0.336 and the Variance Inflation Factor (VIF) between 3.491 and 9.45, so none of the variables had
to be removed. Likewise, the highest partial correlation was found in the Learning Outcomes Total
(r = 0.586; p = 0.000), see Table A3.
3.4. Hypothesis 2.
MRA yielded a figure of R2 = 0.719 in the study of the predictive value of blended learning applied
to student behaviors on the platform. This figure indicated that the blended learning type in use
explained 71.19% of the variance in the learning behaviors of students on the platform. The Tolerance
(T) values were situated within an interval between 0.136 and 0.539 and the Variance Inflation Factor
(VIF) between 1.472 and 7.346, so that no variable had to be removed. The highest partial correlation
was found in Access to Teacher Feedback (r = 0.448), see Table A4.
3.5. Hypothesis 3.
A k-means clustering technique was applied in each type of blended learning in use (Blended
Learning type 1 vs. Blended Learning type 2), as seen in Table 5. Three clusters are shown in Table 5
that were found in the two types of blended learning (Cluster 1, Sufficient; Cluster 2, Intermediary;
and Cluster 3, Excellent. The classification of Cluster type was according to the maximum possible
value in each learning outcome and number of accesses obtained). Higher values for performance
were found in the Blended Learning type 1 rather than the Blended Learning type 2 in all three clusters,
specifically in Learning Outcomes Total. Likewise, with regard to the learning behaviors developed
by students in the type of blended learning in use (Blended Learning type 1 vs. Blended Learning
type 2), as may be seen in Table 5, a higher number of log-ons to the platform in the Blended Learning
Int. J. Environ. Res. Public Health 2020, 17, 1589 9 of 15
type 1 rather than the Blended Learning type 2 environment were found, except for student queries on
theoretical information provided by the teacher (see Table 6).
Table 5. Centers of final clusters for the learning results variable in Blended Learning type 1 and type 2,
Blended Learning type 1: a n = 1; b n = 45; c n = 17; Blended Learning type 2: A lost value is observed. a n
= 9; b n = 30; c n = 18
Table 6. Centers of final clusters and the variable behavioral learning on the LMS in Blended Learning
type 1 and type 2. Blended Learning type 1: a lost value is observed. a n = 31; b n = 27; c n = 5; Blended
Learning type 2: Two lost values were observed. a n = 36; b n = 16; c n = 6.
Figure 1 shows the scores in the two groups: experimental group (red color) and control group
(blue color). As can be seen, there was a greater homogeneity of higher scores in the experimental
group for different types of Learning outcomes. Similarly, Figure 2 points to the distributions of LMS
behavioral learning scores in different resources.
Note; PBL = Project‐Based Learning Development.
Figure 1 shows the scores in the two groups: experimental group (red color) and control group
(blue color). As can be seen, there was a greater homogeneity of higher scores in the experimental
group for different types of Learning outcomes. Similarly, Figure 2 points to the distributions of LMS
Int. J. Environ. Res. Public Health 2020, 17, 1589 10 of 15
behavioral learning scores in different resources.
Figure 1. Distribution of scores in the different types of Learning outcomes. Note. Development PBL =
Figure 1. Distribution of scores in the different types of Learning outcomes. Note. Development PBL
Development Project-Based Learning outcomes; Exhibition PBL = Exhibition Project-Based Learning
= Development Project‐Based Learning outcomes; Exhibition PBL = Exhibition Project‐Based Learning
Int. J. Environ. Res. Public Health 2020, 17, x FOR PEER REVIEW 10 of 16
outcomes; Test = Test Learning outcomes; Total LO = Learning outcomes Total.
outcomes; Test = Test Learning outcomes; Total LO = Learning outcomes Total.
Figure 2. Distribution of scores in the different types of behavioral learning in the LMS. Note. CI =
Figure 2. Distribution of scores in the different types of behavioral learning in the LMS. Note. CI =
Access to Complementary Information scores; CPBL = Access to guidance to prepare PBL scores; TI =
Access to Complementary Information scores; CPBL = Access to guidance to prepare PBL scores; TI =
Access to Theoretical Information scores; F = Access to Teacher Feedback; MVD = Mean Visits per day.
Access to Theoretical Information scores; F = Access to Teacher Feedback; MVD = Mean Visits per
day.
4. Discussion
4. Discussion
In the blended learning environments, the type of teaching design appears to be a predictive
factor in both the learning results and the learning behaviors that the students develop in the LMS.
In the blended learning environments, the type of teaching design appears to be a predictive
blended learning with 80% of interactions in LMS appeared to be more effective, both with respect to
factor in both the learning results and the learning behaviors that the students develop in the LMS.
the learning results of the students and the effectiveness of the learning behaviors that they develop.
blended learning with 80% of interactions in LMS appeared to be more effective, both with respect to
This type of pedagogic design includes the use of hypermedia resources that strengthen teacher
the learning results of the students and the effectiveness of the learning behaviors that they develop.
feedback in real time, which furthers the development of SRL strategies [24,35]. This aspect is of special
This type of pedagogic design includes the use of hypermedia resources that strengthen teacher
relevance for teachers in nursing higher education, and the implicit message is that they would be
feedback in real time, which furthers the development of SRL strategies [24,35]. This aspect is of
well advised to design their materials for use in a blended learning environment [27–30], as those
special relevance for teachers in nursing higher education, and the implicit message is that they
environments appear to have increased the effectiveness of active methodologies, especially PBL with
would be well advised to design their materials for use in a blended learning environment [27–30],
hypermedia resources in LMS [38]. In addition, Blended Learning type 1 (80% the interaction in the
as those environments appear to have increased the effectiveness of active methodologies, especially
LMS) strengthens students’ use of learning-based projects that have been considered more effective in
PBL with hypermedia resources in LMS [38]. In addition, Blended Learning type 1 (80% the
the LMS [2,23]. These behaviors range from access to feedback given by the teacher to tasks carried
interaction in the LMS) strengthens students’ use of learning‐based projects that have been
out by the student or the collaborative groups and the average number of visits per day [23]. All of
considered more effective in the LMS [2,23]. These behaviors range from access to feedback given by
this indicates that the Blended Learning type 1 design increases the interaction of the student in the
the teacher to tasks carried out by the student or the collaborative groups and the average number of
LMS and that interaction also facilitates student access to feedback from the teacher, as the LMS
visits per day [23]. All of this indicates that the Blended Learning type 1 design increases the
can be consulted as many times as necessary when learning, an aspect that is less feasible with F2F
interaction of the student in the LMS and that interaction also facilitates student access to feedback
instruction [9,10,18]. In this way, the teachers can structure their help and prepare specific materials
from the teacher, as the LMS can be consulted as many times as necessary when learning, an aspect
for each group.
that is less feasible with F2F instruction [9,10,18]. In this way, the teachers can structure their help
In addition, machine learning techniques have been used in this study, in view of their effective
and prepare specific materials for each group.
use with what is known as data mining [13,18,36]. In particular, supervised and unsupervised machine
In addition, machine learning techniques have been used in this study, in view of their effective
learning techniques have been used (linear regression and clustering k-means methods, respectively).
use with what is known as data mining [13,18,36]. In particular, supervised and unsupervised
Prediction and clustering studies, among others, can be conducted with these techniques, which
machine learning techniques have been used (linear regression and clustering K‐means methods,
help the teacher to gain knowledge of the learning characteristics of students and to predict at-risk
respectively). Prediction and clustering studies, among others, can be conducted with these
students [23,34,38]. Even so, it is true that these techniques should be used throughout the whole
techniques, which help the teacher to gain knowledge of the learning characteristics of students and
to predict at‐risk students [23,34,38]. Even so, it is true that these techniques should be used
throughout the whole teaching process to be able to develop personalized actions for student learning
[35,36]. In subsequent studies, therefore, development of the learning process among students at the
start, in the middle, and at the end of the study module will be analyzed with machine learning
techniques [12,13].
Int. J. Environ. Res. Public Health 2020, 17, 1589 11 of 15
teaching process to be able to develop personalized actions for student learning [35,36]. In subsequent
studies, therefore, development of the learning process among students at the start, in the middle,
and at the end of the study module will be analyzed with machine learning techniques [12,13].
5. Limitations
This study has limitations, but the results of this study should, nevertheless, be given prudent
consideration. Limitations include the following: methodological intervention was in one university,
the students were from a specific country, convenience sampling was applied, the knowledge area of
the students was specific, and the type of design (quasi-experimental) was also specific. Although,
it must be taken into consideration that there are few specific studies to test the effectiveness of this type
of methodology in nursing students. Studies that have been carried out have similar characteristics
that are justified from the specificity of this research [28–30].
Therefore, future studies will be directed at increasing the size of the sample and the diversity
of the nursing degree course level. Therefore, this profession is subject to continuous theoretical
and technological advances that require systematic research on how to teach better in order to learn
more effectively.
6. Conclusions
This research study has identified the characteristics to design an effective LMS in the nursing
degree. The use of prediction and clustering techniques is very important to facilitate personalized
learning and to analyze how resources are better utilized in the blended learning space. This type of
analysis can be automatically generated in LMS environments, such as Moodle, and could be integrated
in modules and plugins. These tools would facilitate rapid and straightforward generation of those
analyses, which would be of great utility for the teacher and would assist with the early detection of
at-risk students, as well as behavioral analyses of both the individual student and the collaborative
groups of students, which would foreseeably increase the teaching quality and learning outcomes.
This need has been underlined in such studies as those by Peña-Ayala [12] and Romero et al. [17], and
they have to be approved by university management, but they will virtually be a necessity in 21st
century teaching as we move closer to personalized on-line teaching, both in F2F teaching and in virtual
learning environments. In summary, this teaching design is especially significant in the nursing degree
since project work is a practice that has proved very effective in the training of future professionals.
Good results have been obtained in all assessment tests in the two types of blended learning.
However, the type of blended learning that applied automated feedback and hypermedia resources
obtained even better results (more percentage of work in the LMS) [28–30]. One explanation may
be that the student can access information in the LMS at any time, which is not possible for F2F
interaction, and this facilitates personalization of learning and motivates the student [7,8,39]. Therefore,
incorporating these forms of work in teaching in the field of health is a very effective option.
The results obtained are in line with those found in the research of Oh & Lee [28]. The use of PBL
methodology in blended learning environments empowers nursing students to acquire practical skills
that are of great help for nursing work in real intervention environments [28]. This form of teaching is
flexible [30] because it facilitates development and tests hypotheses in the resolution of tasks similar
to those they will encounter in a working environment, and in addition, group work facilitates the
acquisition of collaborative work skills, which they will also encounter in such working environments.
All this increases the self-efficacy and critical thinking skills of these professionals. Recent studies
recommend the application of this methodology within the nursing degree curricula [27].
In sum, it can be concluded that this way of teaching seems to be effective for nursing students.
Although, more studies are needed in this field aimed at studying the effectiveness of blended learning
in teaching in the nursing degree.
Int. J. Environ. Res. Public Health 2020, 17, 1589 12 of 15
Appendix A
Blended Learning Type 1 (Experimental Group) Blended Learning Type 2 (Control Group)
Learning Outcomes
Minimum Maximum M SD α Minimum Maximum M SD α
Learning outcomes in the
1.75 2.45 2.23 0.21 0.79 1.88 2.38 2.25 0.17 0.81
preparation of PBL
Learning outcomes in
1.00 1.95 1.74 0.17 0.76 0 1.90 1.75 0.19 0.75
presentation of PBL
Learning outcomes in test 1.78 3.00 2.43 0.35 0.80 1.23 2.80 2.17 0.42 0.78
Learning Outcomes Total 7.00 0.72 9.03 0.44 0.73 6.08 9.57 8.64 0.66 0.70
Note: M = Mean; SD= Standard Deviation; PBL = Project-Based Learning; α = Reliability index of Cronbach’s Alpha.
Blended Learning Type 1 (Experimental Group) Blended Learning Type 2 (Control Group)
Behavioral Learning in the LMS
Minimum Maximum M SD Minimum Maximum M SD
Access to Complementary
0 28 5.07 11.60 0 22 5.07 4.60
Information
Access to Guidance for the
0 32 9.46 7.59 0 15 3.60 3.01
Preparation of PBL
Access to Theoretical Information 1 70 14.84 10.23 0 37 13.81 7.49
Access to Teacher Feedback 0 194 90.90 29.21 0 82 18.43 21.14
Mean Visits per day 0.41 6.15 3.04 1.02 0.06 2.81 1.08 0.58
Note: M = Mean; SD = Standard Deviation
Int. J. Environ. Res. Public Health 2020, 17, 1589 13 of 15
Table A3. Coefficients in the prediction of learning outcomes with variable types of Blended Learning.
Table A4. Coefficients in the prediction of Behavioral learning with variable types of Blended Learning.
References
1. Siddiq, F.; Scherer, R. Revealing the processes of students’ interaction with a novel collaborative problem
solving task: An in-depth analysis of think-aloud protocols. Comput. Human Behav. 2017, 76, 509–525.
[CrossRef]
2. Sáiz, M.C.; Escolar, M.C.; Marticorena, R.; García-Osorio, C.I.; Queiruga, M.A. Aprendizaje Basado en
Proyectos utilizando LMS: una experiencia en Ciencias de la Salud [Project Based Learning using LMS: an
experience in Health Sciences]. In Temas actuales de investigación en áreas de la Salud y de la Educación [Current
research topics in the areas of Health and Education]; SCINFOPER: Almería, Spain, 2017; pp. 739–746.
3. Lau, C.; Sinclair, J.; Taub, M.; Azevedo, R.; Jang, E.E. Transitioning Self-regulated Learning Profiles in
Hypermedia-learning Environments. In Proceedings of the Seventh International Learning Analytics &
Knowledge Conference, 17 March 2017, Vancouver, BC, Canada; Association for Computing Machinery:
New York, NY, USA, 2017; pp. 198–202. [CrossRef]
4. Krathwohl, D.R. A Revision of Bloom’s Taxonomy: An Overview. Theory Pract. 2002, 41, 212–218. [CrossRef]
5. le Roux, I.; Nagel, L. Seeking the best blend for deep learning in a flipped classroom—viewing student
perceptions through the Community of Inquiry lens. Int. J. Educ. Technol. High 2018, 15, 1–28. [CrossRef]
6. Scoular, C.; Care, E.; Hesse, F.W. Designs for operationalizing collaborative problem solving for automated
assessment. J. Educ. Meas. 2017, 54, 12–35. [CrossRef]
7. Sung, T.-W.S.; Wu, T.-T. Learning With E-books and Project-based Strategy in a Community Health Nursing
Course. Comput. Inform. Nurs. 2018, 36, 140–146. [CrossRef]
8. Feather, R.; Carr, D.; Reising, D.; Garletts, D. Team-Based Learning for Nursing and Medical Students: Focus
Group Results From an Interprofessional Education Project. Nurse Educ. 2016, 41, E1–E5. [CrossRef]
9. Sáiz, M.C.; Montero, E. Metodologías activas en docencia universitaria: Diseño de una asignatura de Ciencias de la
Salud en la plataforma virtual; [Actives Methodologies at the university: Design of a subject of Health Sciences
in the virtual platform]; Servicio de Publicaciones de la Universidad de Burgos: Burgos, Spain, 2016.
10. Von Davier, A.A. Computational psychometrics in support of collaborative educational assessments.
J. Educ. Meas. 2017, 54, 3–11. [CrossRef]
11. Oh, Y.; Oh, Y.K. A computational model of design critiquing. Artif. Intell. Rev. 2017, 48, 529–555. [CrossRef]
12. Peña-Ayala, A. Educational data mining: A survey and a data mining-based analysis of recent works. Expert.
Syst. Appl. 2014, 41, 1432–1462. [CrossRef]
13. Romero, C.; Ventura, S. Educational data mining: A survey from 1995 to 2005. Expert. Syst. Appl. 2007, 33,
135–146. [CrossRef]
14. Bernard, P.; Broś, P.; Migdał-Mikuli, A. Influence of blended learning on outcomes of students attending a
general chemistry course: summary of a five-year-long study. Chem. Educ. Res. Pract. 2017, 18, 682–690.
[CrossRef]
15. Asif, R.; Merceron, A.; Ali, S.B.; Haider, N.G. Analyzing undergraduate students’ performance using
educational data mining. Comput. Educ. 2017, 113, 177–194. [CrossRef]
16. Condell, J.; Wade, J.; Galway, L.; McBride, M.; Gormley, P.; Brennan, J. Problem solving techniques in
cognitive science. Artif. Intell. Rev. 2010, 34, 221–234. [CrossRef]
17. Romero, C.; Espejo, P.G.; Zafra, A.; Romero, J.R.; Ventura, S. Web Usage Mining for Predicting Final Marks of
Students That Use Moodle Courses. Comput. Appl. Eng. Educ. 2013, 21, 135–146. [CrossRef]
18. Sáiz, M.C.; Marticorena, R.; García-Osorio, C.I.; Díez-Pastor, J.F. How Do B-Learning and Learning Patterns
Influence Learning Outcomes? Front. Psychol. 2017, 8, 1–13.
19. Mozer, M.C.; Lindsey, R.V. Predicting and improving memory retention: Psychological theory matters in
the big data era. In Big Data in Cognitive Science; Jones, M., Ed.; Oxford University Press: Oxford, UK, 2017;
pp. 34–64.
20. Dramiński, M. ADX Algorithm for Supervised Classification. In Challenges in Computational Statistics and
Data Mining; Matwin, S., Mielniczuk, J., Eds.; Springer: Basel, Switzerland, 2016; pp. 39–52.
21. Hu, H.; Wen, Y.; Chua, T.S.; Li, X. Toward scalable systems for big data analysis: A technology tutorial.
IEEE Access 2014, 2, 652–687.
22. Margulieux, L.E.; McCracken, W.M.; Catrambone, R. A taxonomy to define courses that mix face-to-face and
online learning. Educ. Res. Rev. 2016, 19, 104–118. [CrossRef]
Int. J. Environ. Res. Public Health 2020, 17, 1589 15 of 15
23. Cerezo, R.; Sánchez-Santillan, M.; Paule-Ruiz, M.P.; Núñez, J.C. Students´ LMS interaction patterns and their
relationship with achievement: A case study in higher education. Comput. Educ. 2016, 96, 42–54. [CrossRef]
24. Moos, D.C.; Bonde, C. Flipping the Classroom: Embedding Self-Regulated Learning Prompts in Videos.
Technol. Knowl. Learn. 2016, 21, 225–242. [CrossRef]
25. Järvelä, S.; Malmberg, J.; Koivuniemi, M. Recognizing socially shared regulation by using the temporal
sequences of online chat and logs in CSCL. Learn. Instr. 2016, 42, 1–11. [CrossRef]
26. Sáiz, M.C.; Marticorena, R. Metacognition, self-regulation and feedback for Object-Oriented Programming
problem-solving. In Metacognition: Theory Performance and Current Research; Benson, J., Ed.; Nova Science
Publishers: New York, NY, USA, 2016; pp. 43–94.
27. Álvarez-García, C.; Álvarez-Nieto, C.; Kelsey, J.; Carter, R.; Sanz-Martos, S.; López-Medina, I.M. Effectiveness
of the e-NurSus Children Intervention in the Training of Nursing Students. Int. J. Environ. Res. Public Health
2019, 16, 4288. [CrossRef] [PubMed]
28. Oh, E.G.; Yang, Y.L. Evidence-based nursing education for undergraduate students: A preliminary
experimental study. Nurse Educ. Pract. 2019, 38, 45–51. [CrossRef] [PubMed]
29. Häggman-Laitila, A.; Mattila, L.R.; Melender, H.-L. Educational interventions on evidence-based nursing in
clinical practice: A systematic review with qualitative analysis. Nurse Educ. Today 2016, 43, 50–59. [CrossRef]
[PubMed]
30. Leidl, D.M.; Ritchie, L.; Moslemi, N. Blended learning in undergraduate nursing education – A scoping
review. Nurse Educ. Today 2020, 86, 1–9. [CrossRef]
31. Pitigala-liyanage, M.P.; Lasith-Gunawardena, K.S.; Hirakawa, M. Detecting Learning Styles in Learning
Management Systems Using Data Mining. J. Inf. Process. 2016, 24, 740–749.
32. Sáiz, M.C.; Montero, E.; Bol, A.; Carbonero, M.Á. An analysis of Learning to Learning competencies at the
University. Electron. J. Res. Educ. Psychol. 2012, 10, 253–270.
33. Harrati, N.; Bouchrika, I.; Tari, A.; Ladjailia, A. Exploring user satisfaction for e-learning systems via
usage-based metrics and system usability scale analysis. Comput. Human Behav. 2016, 61, 463–471. [CrossRef]
34. Sáiz, M.C.; Marticorena, R.; García-Osorio, C.I.; Díez-Pastor, J.F. Does the Use of Learning Management
Systems With Hypermedia Mean Improved Student Learning Outcomes? Front. Psychol. 2019, 10, 1–14.
35. Strang, K.D. Beyond engagement analytics: which online mixed-data factors predict student learning
outcomes? Educ. Inf. Technol. 2016, 22, 917–937. [CrossRef]
36. Yücel, Ü.A.; Usluel, Y.K. Knowledge building and the quantity, content and quality of the interaction and
participation of students in an online collaborative learning environment. Comput. Educ. 2016, 97, 31–48.
[CrossRef]
37. Jones, M.N. Big Data in Cognitive Science; Routledge: New York, NY, USA, 2017. [CrossRef]
38. Saqr, M.; Fors, U.; Tedre, M. How learning analytics can early predict under-achieving students in a blended
medical education course. Med. Teach. 2017, 39, 757–767. [CrossRef] [PubMed]
39. Hodges, H.F.; Massey, A.T. Interprofessional Problem-Based Learning Project Outcomes Between Prelicensure
Baccalaureate of Science in Nursing and Doctor of Pharmacy Programs. J. Nurs. Educ. 2015, 54, 201–206.
[CrossRef] [PubMed]
40. Román, J.M.; Poggioli, L. ACRA (r): Escalas de Estrategias de Aprendizaje; [Learning Strategies Scales];
Publicaciones UCAB (Postgraduate Doctorate in Education): Caracas, Venezuela, 2013.
41. IBM Corp. SPSS Statistical Package for the Social Sciences (SPSS), Version 24; IBM: Madrid, Spain, 2016.
42. IBM Corp. AMOS Statistical Package for the Structural Equation Modeling (AMOS), Version 24; IBM: Madrid,
Spain, 2016.
43. RapidMiner Studio. Available online: https://rapidminer.com/why-rapidminer/ (accessed on 6 September
2019).
44. Bandalos, D.L.; Finney, S.J. Item parceling issues in structural equation modeling. In New Development and
Techniques in Structural Equation Modeling; Marcoulides, G.A., Schumacker, R.E., Eds.; Lawrence Erlbaum
Associates: Mahwah, NJ, USA, 2001; pp. 269–296.
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).