ArticlePDF Available

Trajectories of subject-interests development and influence factors in higher education

Authors:

Abstract and Figures

It is a well-studied phenomenon, that throughout the course of studying at university, the motivation for the study program decreases. Correlation between motivation and learners’ behaviour, for example the learning process, achievement or, in the worst case, dropout exist. So there is a need for understanding the development of motivation in detail, like that of subject-interests, and for identifying influence factors, especially for higher education. This panel study examined the development of 4,345 students in higher education. Growth mixture models for subject-interests identify two classes of trajectories: “descending interest” and “continuously high interest”. In a next step, the analysis shows that gender, university entrance score, academic field and occupational aspiration influence membership of the classes. The results are discussed with respect to their consequences for education programs, but also with respect to possible new research questions.
Content may be subject to copyright.
Vol.:(0123456789)
1 3
Current Psychology
https://doi.org/10.1007/s12144-021-02691-7
Trajectories ofsubject‑interests development andinfluence factors
inhigher education
SteenWild1
Accepted: 30 December 2021
© The Author(s) 2022
Abstract
It is a well-studied phenomenon, that throughout the course of studying at university, the motivation for the study program
decreases. Correlation between motivation and learners’ behaviour, for example the learning process, achievement or, in the
worst case, dropout exist. So there is a need for understanding the development of motivation in detail, like that of subject-
interests, and for identifying influence factors, especially for higher education. This panel study examined the development
of 4,345 students in higher education. Growth mixture models for subject-interests identify two classes of trajectories:
“descending interest” and “continuously high interest”. In a next step, the analysis shows that gender, university entrance
score, academic field and occupational aspiration influence membership of the classes. The results are discussed with respect
to their consequences for education programs, but also with respect to possible new research questions.
Highlights
• Two trajectory classes of subject-interest development in higher education are identified.
• Gender and academic major influence membership in the two different classes.
• The cognitive factors university entrance score and GPA affect the development of subject-interests.
• Occupational aspiration is an important factor for class affiliation.
Keywords Interest development· Higher education· Trajectories· Subject-interests
Introduction
A timeless challenge for education research is how to
improve the academic performance of individuals (Hidi &
Harackiewicz, 2000). Motivation is a key to understanding
(Richardson etal., 2012) as it often decreases over time
in education programs (Gaspard etal., 2020) and that this
decrease in the worst case can lead to student drop out (Sch-
nettler etal., 2020). To look at it in more detail, interest
is seen as a crucial dimension within motivation theories
that influences learning. Scientists have shown its impact
on attention, goals and levels of learning (Hidi & Ren-
ninger, 2006; Renninger & Hidi, 2019). Further research
results show that content-specific interests can be seen as an
important factor in college students’ academic choices and
performance (Harackiewicz etal., 2002).
The importance of interest is supported by other academic
disciplines. For example, neuroscientists have detected interest
as a motivator that influences learning and achievement and
thus suggest that educators should focus on how they can best
support their students’ interest development (Hidi, 2006). A
reason is, that well developed individual interests can help indi-
viduals overcome a lack of ability and/or perceptual disabilities
in math or reading (Renninger etal., 2002). Furthermore, teach-
ers who recognize the potential benefits of increased academi-
cally relevant interests may be best positioned to enhance their
students’ learning. Research data from educational psychology
further supports this claim (Hidi, 2006).
Research on interest still results predominantly from cross-
sectional studies (Dotterer etal., 2009). Yet, for an ontoge-
netic analysis of the development of interests it is important to
consider intraindividual developmental processes (cf. Krapp,
2002). Occasionally, this is possible for researchers analysing
* Steffen Wild
steffen.wild@tu-dortmund.de
1 Chair ofPsychological Diagnostics, Faculty 13
Rehabilitation Sciences, TU Dortmund University,
Emil-Figge-Str. 50, 44227Dortmund, Germany
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
longitudinal data. Another problem is that most studies on
interest are situated in the field of primary and secondary
education research. The preponderance of researchers reverts
to the approaches and results of these studies. Longitudinal
studies on higher education are the exception and can only be
found very sporadically in this research field (Harackiewicz
etal., 2002; Liebendörfer & Schukajlow, 2017).
This study extends previous research by examining the devel-
opment of interest over time in a panel design. The major goal of
this study is to analyze data based on the findings in primary and
secondary school research and examine whether these effects
also exist in higher education. Furthermore, I integrate new
aspects in this study such as occupational aspiration, to provide
a broader picture of the role of developing interest. Last but not
least I try to find out whether there are not only two groups of
either interested and uninterested people but also other interme-
diate forms with unspecific developmental trajectories.
Interest
Theoretical Assumptions andDelimitation
The importance of interest for education has been recog-
nized since the late nineteenth century (Hidi, 2006). Today,
one way to depict interest describes it as a special interac-
tion with the environment, either a PersonObjectInterac-
tion (leading to the development of “individual interest”)
or a PersonStimulusInteraction (leading to “situational
interest”) (Krapp, 2007). This approach is called Person
ObjectTheory (POI) and is visualized in Fig.1. Its assump-
tions are on the one hand that individual interest is a per-
son’s characteristic and conceptualized as a stable personal
disposition, and on the other hand that situational interest
is based on interesting stimuli described as a momentary
specific motivational/psychological state or object within a
person. Both types of interest can be seen as a development
and influence from each other (Pany etal., 2019). A deeper
understanding of POI following Krapp (1992) and Schiefer
etal. (2018) suggest that interest (e.g. subject-interest) has
three components: First, there is the object of interest, which
defines the concrete content of the interest (e.g. the content
of the subject Economy). Second, there are actions of inter-
est that are carried out to engage with the object of interest
(e.g. reading a book or writing a text). And finally, there
are concrete objects that are used to deal with the object of
interest (e.g. a video or a poem).
According to Krapp (2002), this concept of interest also
contains a combination of emotional and value-oriented
components: the person assigns a personal value to the
object of interest and feels positive emotions triggered by
the sum of the object-related actions when dealing with the
object. Especially with regard to the value component, it is
also assumed that people with a stably developed interest
identify with the object of interest and that it becomes part
of their self-definition. While situational interest is only pre-
sent in a specifically interesting situation, individual interest
is anchored in the person's interest self-system.
To understand interest more precisely, researchers need
to look at other motivational variables. The development
of new interest starts with the triggering of attention to a
specific content. An external impulse can help to raise the
chance that an individual continues to deal with this con-
tent (Renninger & Hidi, 2019). Curiosity can also trigger a
person’s attention, but while feeling curious, a person might
encounter, and recognize, a knowledge gap (Loewenstein,
1994). Loewenstein’s theoretical framework of information
gap holds that curiosity functions like other drive states, like
hunger, which motivates eating. Building on this assump-
tion, Loewenstein suggests that a small amount of infor-
mation serves as a priming dose that increases curiosity.
Consumption of information is rewarding, but, eventually,
Fig. 1 Main constituent parts of
Person-Object-Theory of Inter-
est (POI)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
when enough information is getting, satiation occurs, and
information serves to reduce more curiosity. There is an
ongoing debate on the exact relation of interest and curi-
osity. According to Grossnickle (2016) a difference is that
interest develops over time compared to curiosity. Adding to
that, Renninger and Hidi (2016) work out that information-
seeking processes as well as the basis and outcomes of the
search differ. Curiosity is seen as a desire to seek and learn
new information by exploring novel and uncertain environ-
ments. Individual interest focuses the motivation to seek and
learn new information because it is linked to some form of
existing knowledge, which then continues to develop. Fur-
ther research shows a reciprocal relation between interest
and goal setting (Harackiewicz etal., 2008), self-efficacy
(Hidi & Ainley, 2008), and self-regulation (Sansone & Tho-
man, 2005). Renninger and Hidi (2019) conclude that these
variables are distinct, and that in earlier phases of interest
they may appear to be unrelated while in later phases of
interest they are coordinated and mutually supportive.
This study conceptualises subject-interest as individual
interest. According to Hoffmann (2002), subject-interest can
be conceptualized in two different ways: on the one hand
as interest in the topics of the respective subject and on
the other hand as interest in the entire teaching of the sub-
ject—how it is taught and what is learned. Subject-interest
in this research is conceptualized in the first manner. I define
subject interest in this study as the match between personal
interests and learning opportunities in one's own field of
study (cf. Fellenberg & Hannover, 2006).
Theoretical Framework fortheDevelopment
ofInterest
Findings in research about education programs often report
the decrease in interest. For example, pupils enter elementary
and secondary school with a high level of interest in indi-
vidual school subjects. This interest declines throughout the
course of their schooling (Dotterer etal., 2009; Frenzel etal.,
2010, 2012). A variety of different assumptions are made to
explain this phenomenon and react.
One possible explanation for these findings is that extra-
curricular areas of interest increase and compete with school
interests (e.g. Hartinger & Fölling-Albers, 2002). Another
suggestion, the stage-environment-fit approach (Eccles etal.,
1993), emphasizes a mismatch between the needs and inter-
ests of young people and the supply structures of the school
context. Finally, Daniels (2008) proposes that pupils undergo
a process of differentiation and hence solely focus on a few
areas of interest in their schooling time so that only selected
subject areas of interest stay high.
Interest theories use different approaches to explain these
findings. The framework of person object theory of inter-
est distinguishes two interrelated subsystems: 1. emotional
experiences and 2. conscious-cognitive factors (Krapp,
2005). These systems explain the start of an activity in a
certain domain triggered by situational interest, and the
continuous engagement in a specific object area because of
stable individual interest. Emotional experiences based on
a biological component. Here, emotions give a feedback on
the organism's state of functioning in a situation. Conscious-
cognitive factors refer to the process of rational–analytic
intention. This is important when a person’s own controlled
actions are responsible, in a consciously effortful way, in
overcoming obstacles during a goal-oriented activity or an
uninteresting, but significant task. When both subsystems
are positive, interest develop.
According to the “Fourphase model of interest develop-
ment” (Hidi & Renninger, 2006; Renninger & Hidi, 2019),
individual interest develops in four stages (Fig.2). First,
there is a triggered or “catch”component of situational
Fig. 2 Levels of Interests in the
FourPhase Model of Interest
Development
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
interest: a person comes into contact with a stimulus that
raises their attention. The second phase consists of main-
taining situational interest or a “hold”component of situ-
ational interest: out of attention an experience develops that
combines a growing sense of value with an epistemic ori-
entation toward the content—the person is willing to know
more about the object of interest. The third phase is seen
as emerging individual interest. Here, the person develops
positive feelings, has stored knowledge, and attaches a per-
sonal value to the interesting object. In addition to this, the
person generates their own “curiosity” questions concerning
the content of an emerging individual interest. The fourth
and final phase is characterized as welldeveloped individual
interest. Here, a person engages with the object of interest
against the background of their own set of values, increases
the stored knowledge and starts to search and create answers
to their own curiosity questions.
In summary, the development of interest starts with the
triggering of attention and is followed by searching for infor-
mation. In earlier phases, interest development can be facili-
tated by the structure of the environment or by interacting
with other people. In contrast, in later phases, interest is built
on content knowledge. However, in any phase of interest
development, the level of interest can be expected to pla-
teau or fall. This happens if individuals have no opportunity
to engage with content that allows them to further develop
their understanding. Moreover, problems for the develop-
ment of interests arise when individuals maintain compet-
ing interest or if they do not receive the support they need
to make meaningful connections to content (Renninger &
Hidi, 2019).
Current State ofResearch onInterest inEducation
Research on interest exists almost solely for primary and
secondary education (Renninger & Hidi, 2019; Schiefele,
2009). For higher education, on the other hand, publications
are rare. Fortunately, longitudinal studies have been used
more often in the scientific discourse as numerous panel
studies have been initiated in recent years (Blossfeld etal.,
2019). The issue of diminishing interest over time, which has
already been discussed in detail in the previous sub-chapter,
is seen as important in science. Research on higher educa-
tion confirm this (Xu etal., 2021). Here, other topics, such as
performance or academic choices, are also often considered
when dealing with interest (Schiefele, 2009).
Krapp (2002) and Schiefele (2009) report that subject-
interest development is affected by subject areas, context
conditions, school type and gender. This research is very
focused and hard to generalize. Todt etal. (1974) show a
decreasing interest for girls throughout secondary school for
zoological and botanical topics in biology, but an increased
interest in topics related to human beings and ecology
biology. Other research in subject areas shows a low inter-
est in physics teaching within a scientific context (with an
emphasis on the validity of general physical laws) and a
strong interest when the teacher is able to relate physical
principles and facts to practical problems and the students’
everyday life (Hoffmann & Lehrke, 1986; Hoffmann etal.,
1985). According to Høgheim and Reber (2019) girls report
lower levels of individual interest in mathematics than boys
do. A slight trend reported by Krapp (2002) implies that
girls’ interests decrease faster than boys.
The situation is different for the link between interest and
performance. Schiefele etal. (1993) report a correlation of
r = 0.30 in a meta-analysis. However, simple correlations
are used for this analysis. Schiefele (2009) argue that in the
current empirical state of research on interest the causal
direction between interest and achievement remains an unre-
solved issue. Longitudinal research by Maurice etal. (2014)
shows effects of achievement on interest for the 3rd Grade.
Reciprocal instead of unidirectional effects between inter-
est and achievement are reported by Scherrer etal. (2020).
Schiefele (2009) concludes for lower secondary school level
that interest is either a nonsignificant or weak antecedent
of achievement, while for higher education, Rotgans and
Schmidt (2011) show effects of interest on achievement.
Academic choices are also related to interest. Whereas in
lower secondary schools, students’ motivation is mostly reg-
ulated by extrinsic incentives and values, in upper secondary
school, interest gains more influence on the regulation of
learning activities and highly interested students more often
choose an advanced course. These findings are in line with
research on academic choices by Bong (2001), Durik etal.
(2006) and Eccles (1983), who presented evidence that the
effects of motivational characteristics on academic choices
are more substantial than those of achievement or learn-
ing. Moreover, the theoretical framework of the Wisconsin
model (Hauser etal., 1983; Sewell etal., 1969) from social
psychology assumes the importance of social origin, occupa-
tional aspirations and academic performance for educational
attainment and outcome. Furthermore, empirical research
underlines the existence of such effects (Kim etal., 2019;
Sabates etal., 2011; Yates etal., 2011).
Researchers in higher education have often called for ana-
lyzing students over an extended period of time (Xu etal.,
2021). But research in this field has mostly focused on evalu-
ating learning gains and much less frequently on estimating
growth rates of individual constructs (Coertjens etal., 2017;
Kyndt etal., 2015) and developmental relationships between
constructs (Kyndt etal., 2019). This situation highlights a
significant challenge for research on students’ motivational
development, especially for the construct interest, in higher
education. In Europe exist no consistent findings according
to this specific research field. Liebendörfer and Schukajlow
(2017) published results, with a small sample size (N = 92)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
from lower secondary school teachers, showing that stu-
dents’ interest in mathematics remained stable during the
first academic year in Germany. Xu etal. (2021) show for
students in Educational Sciences, Speech Pathology and
Audiology in Belgium a decreasing of statistic interest over
time by using a latent growth curve analysis for analysing
data over two years. Furthermore, this study reports, that the
rate of decrease in interest was positively associated with
rates of growth in cognitive competence and utility value.
However, the different development of subgroups within
the sample was not analyzed. Further longitudinal stud-
ies on interest in the higher education sector indicate that
mastery goals, e.g. the desire to develop new skills (Ames
& Archer, 1988), are unrelated to academic performance
but that they predict interest in the course. Students who
adopted work avoidance goals report less subject-interests
in psychology (Harackiewicz etal., 2002). Analyses for dif-
ferent developing groups of trajectories in subject-interests
are not consistent. Research based on higher education is
not well develop and as a consequence I use studies focused
on elementary and secondary school. Research of Musu-
Gillette etal. (2015) analyzes trajectories of interest in math
from 4th grade to the 2nd year of college. Results show three
latent classes with a curvilinear trajectory. The first class is
called “High Interest Trajectory” (40 percent of the sample)
starts with the highest reported levels of interest in math.
That interest declines over time, especially between 6th
grade and 10th grade, but shows an increase from late high
school on into college. The trajectory could be interpreted as
a u-shaped change. The second latent class is called “Slow
Decline Interest Trajectory” (22 percent of the sample) and
students in this class begin with moderate levels of interest
in math in elementary school. Persons in this class show a
decline in their reported interest over time with a curvilinear
trend and stagnation in high school and into college. For
the third latent class, “High Self-Concept Trajectory” (38
percent of the sample), initially high levels of interest in
math, which decline rather steeply over time, are reported.
In contrast, Schiefer etal. (2018) identify five different
latent classes for subject-interests in German (native lan-
guage), mathematics and English at grade 4 to grade 11.
Taken together, this underlines the importance of studying
students’ interest development.
The Present Study
Drawing on the Wisconsin model, the POI and a fourphase
model of interest development, this study examines the tra-
jectories in subject-interests in higher education. It aims at
establishing different latent classes and linking these classes
of subject interests to academic performance, demographic
conditions, learning conditions and career choices. This
study uses the well-known growth mixture modeling (GMM)
to investigate qualitatively distinct trajectory classes (Guo
etal., 2018; Musu-Gillette etal., 2015; Wang etal., 2017).
GMM analyzes not only an average trend in the data, but
also interindividual differences between student groups as
well as intraindividual differences. Therefore, GMM is con-
sidered an appropriate approach for this study, as it allows to
account for potential heterogeneity in developmental trajec-
tories between groups of students as well as intraindividual
differentiation across domains.
Two research questions are investigated. The first ques-
tion is: “Can qualitatively different latent trajectory classes
of subject-interest be identified?”. Because of the scarcity
of previous research on this topic no specific predictions
about the number and shape of such trajectories are made.
However, I expect to find classes in which subject-interest
remains high or decreases by controlling for the cognitive
factors (Protsch & Solga, 2015), measured by Grade Point
Average, and the context factor Quality of instruction (cf.
Gaspard etal., 2020). I expect different trajectories against
the background of the fourphase model and the POI,
because in earlier phases of a study program interest devel-
ops through the formal structure of the environment and
through interacting with other people. In the last years of a
study program interest is built on relatively open structures,
opportunities and support, which students use in different
ways.
The second question is: “Are individual (gender, occu-
pational aspiration), cognitive (university entrance score),
family background (social origin) or academic field factors
related to affiliating with a specific latent class?” Based on
previous research on links between these variables and the
development of students’ subject-interest, I expect that high
university entrance score and occupational aspiration predict
higher levels of subject-interest as predicted in the Wiscon-
sin model. For academic field, social origin and gender, no
directional hypothesis is assumed, since, so far, no consistent
research results on their links to subject-interest develop-
ment have been attained.
This study closes a research gap by analyzing the devel-
opment of subject interest in the whole bachelor program of
students in higher education. Similar research that focuses
this topic over such long periods of time with a similar num-
ber of measurement points as well as a similarly large sam-
ple size and for these academic majors is not known.
Methods
Participants andProcedure
The present research uses data from the panel study “Study
Process – Crossroads, Determinants of Success and Barriers
during a Study at the DHBW” from 2016 to 2019 (Deuer
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
etal., 2020). The study is specifically designed to explore
determinants of academic success in cooperative educa-
tion at Baden-Wuerttemberg Cooperative State University
(DHBW). These bachelor’s degree programs are made up of
210 ECTS (European Credit Transfer System) credits in six
semesters (three years) and every three months, a coopera-
tive student rotates between academic training at the univer-
sity and workplace training at the company (Wild & Alvarez,
2020). Participation have been voluntary and privacy policy
is protected. Every fiftieth student who answered more than
one question received a 10 € coupon as an incentive for
participation.
The numbers of students have been continuously rising
since 2007 and as a consequence this study is done during
the years 2016 and 2019 (Destatis, 2020, p. 31), including
for cooperative education (AusbildungPlus, 2020, p. 11),
and student dropout has become increasingly relevant as a
topic in higher education research and policy (Neugebauer
etal., 2019;Wild & Schulze Heuling, 2020). Four cohorts
are included in this four-year study, since there is only a
chance to get enrolled one time per year. Data collection is
conducted once a year for economic reasons, resulting in
four waves of data collection. Differences in the data sets
are analyzed by measurement invariance and the results are
presented in the following chapter “Measures”.
For the present analyses, data from all students who
reported their subject-interest more than one time in the
four waves is used. Every year, all 34,000 enrolled students
at DHBW are invited to participate in the survey by two
emails, separated by a two-week interval, which included a
link to a questionnaire. Panel wave 1 is conducted in summer
2016 (response rate 19.7 percent), panel wave 2 in spring
2017 (response rate 18 percent), panel wave 3 in spring 2018
(response rate 24.3 percent) and panel wave 4 in spring 2019
(response rate 22 percent). A total of 4,345 students (58 per-
cent female) from four cohorts (n = 565 for Cohort starting
the study program in October 2014; n = 1,432 for Cohort
starting the study program in October 2015; n = 1,526 for
Cohort starting the study program in October 2016; n = 822
for Cohort starting the study program in October 2017)
participated and contribute to the estimation of the GMM.
Table1 gives an overview of the collected data for each
wave and cohort. In cohort 2015 and cohort 2016, more
than 1,000 participants per wave take part in the study. In
cohort 2014 only 565 students and in cohort 2017 822 par-
ticipated. Range of the means in age is in cohort 2016 with
Mwave2 = 21.71 (wave 2) and Mwave4 = 23.94 widest. Standard
deviation ranges from SD = 2.71 (cohort 2015 in wave 2) to
SD = 3.18 (cohort 2016 in wave 4). Attrition is a huge prob-
lem. Calculations show that only 512 students participated
in all waves from wave 1 to wave 3 during their regular
enrolment time of three years.
Measures
Subject‑Interest
The subject-interest is measured with a modified instru-
ment by Fellenberg and Hannover (2006) in every wave.
Reliability on items of a 5-point Likert scale with values
ranging from 1 (strongly disagree) to 5 (strongly agree) is
excellent for all four waves (ω = 0.89–0.91). The modifi-
cation of the instrument of originally 13 items is neces-
sary for reasons of, because of test efficiency and research
Table 1 Sample description by
cohort, and age at each panel
wave
The average age was computed as of January 1 of each year shown (e.g., January 1, 1988)
Time and Variable Cohort 2014 Cohort 2015 Cohort 2016 Cohort 2017
Panel Wave 1 (July 2016)
n565 1,049
M23.12 22.04
SD 3.26 3.02
Panel Wave 2 (March 2017)
n565 1,090 1,074
M23.79 22.67 21.71
SD 3.26 2.71 2.93
Panel Wave 3 (March 2018)
n1,144 1,312 822
M23.71 22.78 21.78
SD 2.99 2.91 3.16
Panel Wave 4 (March 2019)
n1,138 822
M23.94 22.78
SD 3.18 3.16
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
group discussion of better face validity. Table2 shows the
original instrument and the items used in this study. As a
next step, I investigate the cut-off criteria by Chen (2007)
by χ2-Test and Δ CFI 0.005 in combination with Δ
RMSEA 0.010 for the level of measurement invari-
ance (Table3). Using χ2-Test is problematic, because χ2
increases in power to reject the null hypothesis as the sam-
ple size increases. Having a larger total sample, this may
lead to over-rejection of measurement invariance tests if
the change in χ2 is the only criterion used to evaluate fit
(Putnick & Bornstein, 2016). The χ2-Test in every model
comparison is significant by p < 0.001. According to the
benchmarks of Δ CFI ≤ − 0.005 and Δ RMSEA ≥ 0.010
for the level of measurement invariance, scalar invariance
for the factors academic year and panel wave is indicated.
Time
The information regarding the time of measurement is
obtained from the survey software. To be able to model the
time period in the study program to the exact day in the
statistical model, the survey date was compared with the
participants’ date of starting the study program.
Occupational Aspiration
To measure occupational aspiration of the students, this
study employs a proxy variable. The item text is “The sub-
ject I am studying has been my "desired subject" from the
very beginning.”. A 5-point Likert scale with values ranging
from 1 (strongly disagree) to 5 (strongly agree) is used and
the data was collected one time for each cohort.
Gender
The university’s administration provides the data for gen-
der with male and female. I do not receive any data for
gender diverse persons. The study matches this data to the
collected data from the survey.
Table 2 Formulations of the
Items in original Instrument
(Fellenberg & Hannover, 2006)
and used in this research for the
scale subject interest
Presented are translations of the original German items that are not yet validated in the English language;
* = used in this research; (-) = inverse item
Item formulation
My field of study matches with my interests. *
I cannot imagine a more interesting subject than my field of study. *
My subject is exactly the right one for me. *
For me, dealing with the content of my subject is more of a frustration than a pleasure. *(-)
I enjoy dealing with topics in my subject
I often think about certain topics in my field of study
I enjoy exchanging ideas with others about topics in my subject. *
I have doubts about whether my subject really matches my interests. *(-)
I enjoy dealing with certain questions and problems in my field of study.*
The subject I study does not necessarily reflect my main interests. (-)
My subject of study is also my hobby. *
The interest in my subject of study is not excessively strong in me. *(-)
Actually, I am more interested in other subject contents than in those of my subject. (-)
Table 3 Measurement invariance for the scale subject interest on four panel wave and academic year (n = 8,547)
Satorra-Bentler-scaled χ2-difference test. Academic year: one (n = 3,408), two (n = 2,797), and three (n = 2,342). Panel wave: one (n = 916), two
(n = 1,293), three (n = 3,715), and four (n = 2,623)
χ2df χ2/df Δ χ2Δ df p CFI RMSEA Δ CFI Δ RMSEA
Academic year
Configural Invariance 1102.3 69 15.98 < .001 .969 .078
Metric invariance 1154.1 85 13.58 51.817 16 < .001 .968 .072 -.001 .007
Scalar invariance 1274.2 101 12.62 120.124 16 < .001 .965 .069 -.003 .003
Panel wave
Configural Invariance 1100.1 92 11.95 .970 .077
Metric invariance 1152.1 116 9.93 51.949 24 < .001 .970 .070 -.001 .008
Scalar invariance 1308.0 140 9.34 155.905 24 < .001 .966 .068 -.004 .002
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
Academic Major
The test persons are enrolled in the three academic majors
of Economy, Engineering and Social Work. These data are
obtained from the university administration and matched
to the survey data. Economy and Engineering is chosen,
because those are the academic majors in Germany with
the highest number of students enrolled in 2020 (Destatis,
2020, p. 31). Social Work is integrated in this research,
because a shortage of skilled workers is expected in this
field (Vogler-Ludwig etal., 2016). These academic majors
vary in their didactics and teaching–learning methodolo-
gies. For example, Economy uses management simula-
tions. Engineering uses technical laboratories, for example
in the context of materials science. Social Work strongly
reflects its works in the context of case management. Dif-
ferences in the academic majors, like Economy and Engi-
neering, exists for example in the basic needs from Self-
Determination Theory (Wild & Neef, 2019).
Social Origin
Social origin is measured via parental education. This
study distinguishes three origin groups: “low” if mother
and father complete a lower or no school leaving certifi-
cate, “medium” if at least one parent gained a higher edu-
cation entrance qualification and “high” if at least one par-
ent has a degree in higher education. This data is collected
one time during the panel survey.
University Entrance Score andGPA
German university entrance scores in the survey vary
between 1 (equivalent to A in Great Britain and United
States of America) and 4 (equivalent to E (GB) or D (US))
and GPA varies between 1 (equivalent to A in Great Brit-
ain and United States of America) and 5 (equivalent to E
(GB) or D (US)). The date is recoded for better interpreta-
tion so that 5 is the best score and 2 (university entrance
scores) or 1 (GPA) are the lowest score. Data for the GPA
and the university entrance scores is provided by the uni-
versity administration for GPA.
Quality ofInstruction
An adjusted scale by Thiel etal. (2008) measures Quality
of instruction with eight items that vary between 1 (strongly
disagree) and 5 (strongly agree). The reliability in all four
waves shows good values (ω = 0.81–0.82; item example: In
general, the courses are well structured.). The data is col-
lected in every wave.
Table 4 Descriptive statistics and correlations (r) among metric key variables between academic years (n = 1,450 4345)
GPA Grade Point Average; †p < .10; *p < .05; **p < .01; ***p < .001
1234567891011
1. Subject-interests (Academic year 1)
2. Subject-interests (Academic year 2) .72***
3. Subject-interests (Academic year 3) .65*** .76***
4. Quality of instruction (Academic year 1) .34*** .26*** .23***
5. Quality of instruction (Academic year 2) .21*** .31*** .24*** .56***
6. Quality of instruction (Academic year 3) .17*** .24*** .27*** .48*** .64***
7. GPA (Academic year 1) .10*** .12*** .13*** .06** .05** .02
8. GPA (Academic year 2) .12*** .15*** .16*** .05* .08*** .05* .86***
9. GPA (Academic year 3) .09** .16*** .17*** .04 .07** .03 .82*** .94***
10. Occupational aspiration .42*** .38*** .31*** .11*** .08*** .08*** .00 .02 .05**
11. University entrance score .04* .05 .02 .06** .05** .08*** .33 .39*** .42*** .01
M3.70 3.59 3.56 3.70 3.50 3.42 3.85 3.86 3.92 3.90 3.84
SD 3.66 3.74 3.77 3.53 3.58 3.60 3.60 3.52 3.45 1.16 3.59
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
Data Analysis Strategy
To achieve the main objective of the present study, I use
GMM for analysing research question 1 (Fitzmaurice etal.,
2009). A hierarchical logistic regression is estimated to
test the assumptions of research question 2 (Tabachnick &
Fidell, 2013). Each step is described in more detail in the
following section. According to Richard etal., (2003; p. 339)
I interpret the effect size of r = 0.10 – 0.19 as small, r = 0.20
– 0.29 as medium and r ≥ 0.30 as large in the first paragraph
of the chapter results.
All analyses are conducted with R version 3.6.2. The
reliability analysis of ω (McDonald, 1999) is done with the
package “MBESS”. The GMM is analyzed with the package
“lcmm” and logistic regression analyses are conducted with
the package “margins”.
To address the first research question, this study follows
the proposed approach of Ram and Grimm (2009) for the
analysis of the GMM. Thus, this study examines the shape of
growth over time using growth curve analyses for one single
group. Both linear and quadratic growth models are tested.
As a next step, a model specification based on previous
models is conducted to identify unobserved subgroups of
individual trajectories. The analysis compares models with
increasing numbers of classes. Comparisons across models
are conducted with the Akaike information criterion (AIC),
the Bayesian information criterion (BIC) fit statistics, and
the sample-adjusted BIC (SABIC), with smaller values indi-
cating superior fit to the data. The entropy value is measured
(ranging from 0 to 1) as an indicator of classification accu-
racy. Values > 0.70 indicate a good classification accuracy
(Reinecke, 2006) and the diagonal of the average latent class
probabilities for most likely class membership near 1 (Jung
& Wickrama, 2008). Finally, plots of the group trajectories
are inspected for the plausibility of the results.
Subsequently, this study investigates differences in the
sets of student characteristics and outcomes across the
previously identified classes (Research Questions 2). For
this, hierarchical logistic regression and Likelihood Ratio
Tests are used (Glover & Dixon, 2004). Average Marginal
Effects (AME) are estimated, because this procedure leads
to satisfactory results in many different scenarios (Best &
Wolf, 2012; Mood, 2010).
The results for estimated fixed effects model are reported
by four decimal places (ten thousandths). The reason for
this is, that the effect is small yet still positive or negative
and does not actually include zero. Only two decimal places
with zero and statistically significant results would confuse
the reporting. Especially in the case large sample size, where
almost everything will be significant.
The percentage of missing values of each variable in the
dataset of the GMM is below 0.1 percent. A Missing Values
Analysis indicates that Little’s (1988) test of Missing Com-
pletely at Random (MCAR) is not significant (χ2 = 2.071,
df = 2, p = 0.36). For the logistic regression dataset, the per-
centage of missing values on each variable vary between 0
and 2.9 percent. Little’s (1988) test is not significant here,
either (χ2 = 16.826, df = 25, p = 0.88). Thus, there is no
evidence that the data was not MCAR. The missing data
is estimated using the R Package “Amelia” and the EMB
algorithm method, which combines the classic Expectation
Maximization (EM) algorithm with a bootstrap approach
(Honaker etal., 2011), using m = 5 imputed datasets. GMM
models are estimated based on z-scores.
Table 5 Fixed-effects models for changes in subject-interests in study
program (full sample)
Standard errors are shown in parentheses; β = standardized beta coef-
ficients; GPA Grade Point Average; †p < .10; *p < .05; **p < .01;
***p < .001
Model 1 Model 2
β β
Time .0218 (.0049)*** .0095 (.0046)*
Time (quadratic) .0004 (.0001)*** .0002 (.0001)†
GPA .1672 (.0134)***
Quality of instruction .3739 (.0123)***
Intercept 3.8477 (.0319)*** 1.7456 (.0747)**
Number of Persons 4,345 4,345
Number of Observations 9,581 9,581
Table 6 Fit indices from estimated growth mixture models
AIC Akaike information criterion, BIC Bayesian information criterion (BIC), SABIC Sample-adjusted Bayesian information criterion
Model AIC BIC SABIC Entropy Minimum size
of class in
percent
maximum
size of class in
percent
minimum average latent
class probabilities for
most likely latent class
membership
maximum average latent
class probabilities for most
likely latent class member-
ship
1-Class 26043.95 26082.21 26063.14 - - - - -
2-Class 23612.01 23688.53 23650.40 .74 27 73 89 94
3-Class 22550.95 22665.73 22608.54 .72 12 53 86 90
4-Class 22113.53 22266.57 22190.31 .73 5 52 83 88
5-Class 21998.20 22189.50 22094.17 .69 2 43 75 87
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
Results
Table4 provides descriptive statistics and correlations (r)
for all variables across academic years. The means on the
5-point Likert scale (1 = “strongly disagree” to 5 “strongly
agree”) range between M = 3.42 for quality of instruction
and M = 3.92 for GPA – each in the third academic year. As
a next step, I inspect the correlations. According to Rich-
ard etal. (2003), large correlation effect sizes exist between
occupational aspiration and subject-interest (r = 0.31 to
r = 0.42). In this study, the measurement for the third aca-
demic year also shows medium and large effect sizes of the
correlations for subject-interest and quality of instruction
between r = 0.27 and r = 0.34. The two performance meas-
urements of university entrance score and GPA correlates
with r = 0.33 and r = 0.42.
Step 1 of the analysis involves exploring the functional
form of the growth curve in students’ subject-interests
across the full sample. Table5 shows the results of the two
estimated fixed-effects models without (Model 1) and with
control variables (Model 2). I see that the linear slope factor
for time is negative for Model 1 (β = 0.0218) and Model
2 (β = 0.0095). Both models are significant p < 0.05. The
quadratic slope factor for time is seen as very small. In
Model 1 it is β = 0.0004 and significant (p < 0.001). In Model
2 the effect is β = 0.0002 and marginally significant (p < 0.1).
The effects for the control variables GPA (β = 0.1672;
p < 0.001) and quality of instruction (β = 0.3739; p < 0.001)
are larger, than the time effect. Because of these results a
quadratic growth factor for subject-interest is assumed and
the two control variables in the model are kept for further
analysis.
Step 2 of the analyses addresses research question 1 con-
cerning how many classes of students’ trajectories can be
identified. The study estimates growth mixture models vary-
ing from a one-class solution to a five-class solution. Table6
summarizes the results. AIC, BIC and saBIC decrease in all
estimated models, which indicates a solution with only a few
classes. Entropy varies between two classes and four classes
in a range from 0.72 to 0.74. The researcher finally chose a
two class solution, because its entropy of 0.74 is the highest
of all class solutions, an appropriately large class size rang-
ing from 24 to 73 percent—the smallest class sizes in the
other estimated models could be seen as marginal groups or
outliers—and highest large average latent class probabilities
for the most likely latent class membership between 89 and
94 percent.
Fig. 3 Estimated average tra-
jectories of the growth mixture
modeling (2-Class solution).
Notes: Sample size of class 1
“descending interest” (below
trajectory) is 27 percent; Sam-
ple size of class 2 “continuously
high interest” (above trajectory)
is 73 percent
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
Figure3 plots the result for the two class solution. I use
z-scores in our plot. The trajectories for class 1 (27 percent
of the sample) starts at nearly z = 0.50 and decreases in the
following months. In month 15, the score is under z = 1.
Although the score rises again from around the 20th month, it
never rises above z = 1 in the subsequent period. It is specu-
lated that subject-interests will increase, because of the thesis
here. The trajectory for class 2 (73 percent of the sample size)
starts at about z = 0.50 with less changes. This trajectory is
always above the mean with a z-score > 0. As a consequence
of these results, I name class 1 “descending interest” and
class 2 “continuously high interest” for further analyses.
Table7 shows the estimated development trajectories
of the two classes solution as fixed-effects models. The
intercept for the class “descending interest” is β0 = 0.5451
and for the class “continuously high interest” β0 = 0.4401.
The development of the class “descending interest” shows
a decreasing slope (β = 0.0427; p < 0.001) with a quad-
ratic trend (β = 0.0008; p < 0.001). In contrast, the class
“continuously high interest” shows a smaller negative
effect (β = 0.0055; p > 0.10). GPA (Class “descending
interest”: β = 0.1239; p < 0.001; Class “continuously high
interest”: β = 0.0867; p < 0.001) with larger effect against
time and quality of instruction (Class “descending inter-
est”: β = 0.3489; p < 0.001; Class “continuously high inter-
est”: β = 0.2158; p < 0.001) with largest effect in the whole
model are significantly positive for these two variables and
both classes. This result underlines the importance of these
variables from a theoretical point of view, for their practical
implications and finally for the estimated model.
To work on research question 2 and to identify individual,
cognitive and background factors for membership in differ-
ent classes, this study uses logistic regression analysis, which
is depicted in Table8. To check the robustness of the results,
I systematically extend the estimated models by adding
variables. Results of Model 1 (McFadden's adj. R2 = 0.13;
Nagelkerke R2 = 0.19; Cox & Snell R2 = 0.12) show t hat
occupational aspiration (AME = 0.10; p < 0.001) and the
academic majors of social work (AME = 0.16; p < 0.001)
and engineering (AME = 0.08; p < 0.001) influence belong-
ing to the class “continuously high interest”. However, there
is a significant effect for female against male participants
being in the class “descending interest” (AME = 0.09;
p < 0.001). These effects remain almost unchanged in the
other estimated models. Cognitive factors depict by the uni-
versity entrance score are integrated in Model 2. However,
this effect is negative (AME = 0.05; p < 0.001) and students
with better scores belong to the class “descending interest”.
A likelihood ratio test in between Model 1 and Model 2
shows a modest improvement in model fit (McFadden's adj.
R2 = 0.12; Nagelkerke R2 = 0.20; Cox & Snell R2 = 0.13; χ2
(1) = 15.99, p < 0.001). Model 3 includes family background
factors. The effect of social origin “middle” (AME = 0.05;
p < 0.10) and social origin “high” (AME = 0.04; p < 0.10)
is not as high as the other variables in the models and only
marginally significant. A likelihood ratio test between Model
2 and Model 3 shows no improvement in model fit (McFad-
den's adj. R2= 0.12; Nagelkerke R2= 0.20; Cox & Snell
R2 = 0.14; χ2 (2) = 3.47, p > 0.10). Figure4 plots the results
of Model 3 in Table7 using a coefficient plot.
Discussion
This study is one of the first studies to examine develop-
mental processes of subject-interest in higher education in a
longitudinal perspective analyzing inter- and intraindividual
Table 7 Fixed-effects models for changes in subject-interests in study
program (2-Class solution)
Standard errors are shown in parentheses; table shows standard-
ized beta coefficients; GPA Grade Point Average; †p < .10; *p < .05;
**p < .01; ***p < .001
descending interest continuously high
interest
β β
Time .0427 (.0079)*** .0055 (.0047)
Time (quadratic) .0008 (.0002)*** .0002 (.0001)
GPA .1239 (.0174)*** .0867 (.0099)***
Quality of instruction .3489 (.0177)*** .2158 (.0096)***
Intercept .5451 (.0643)*** .4401 (.0379)***
Number of Persons 4,345
Number of Observations 9,581
Table 8 Logistic regression for prediction on membership on class
“continuously high interest” (n = 4,345)
Standard errors are shown in parentheses; table shows AME Average
Marginal Effect. †p < .10; *p < .05; **p < .01; ***p < .001
Model 1 Model 2 Model 3
AME AME AME
Occupational aspiration .10 (.01)*** .10 (.01)*** .10 (.01)***
Gender: female (ref.
male) .09(.01)*** .09(.01)*** .09(.01)***
Academic major (ref. economy)
social work .16(.02)*** .14(.02)*** .14(.02)***
engineering .08(.01)*** .09(.01)*** .09(.01)***
University entrance
score .05(.01)*** .05(.01)***
Social origin (ref. low)
middle .05 (.03)†
high .04(.02)†
Cox-Snell R2.13 .13 .14
Nagelkerke R2.19 .20 .20
McFadden's adj. R2.12 .12 .12
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
differences. The first major aim of this study is to identify
different types of trajectories. In a subsequent step, deter-
minants are tested that influence the membership in these
trajectory types as second major aim.
This research identifies a decreasing subject-interest over
all observations. GMM models are estimated for analys-
ing research question 1 detect two classes of trajectories.
Class 1, classify as “descending interest”, shows a decreas-
ing curve from the beginning of university on, and a slow
rise from the 20th month onward. This curve is in line with
previous findings from motivation research (Dotterer etal.,
2009; Frenzel etal., 2010). The second class, characterize as
“continuously high interest”, shows an almost parallel curve
above the overall mean. The interest in this class stays on a
continuously high level. Gaspard etal. (2020) publish simi-
lar results for elementary and secondary school, with two
trajectory lines for motivational variables. Consequently, the
results replicate the state of current research. Influence fac-
tors for the membership of different classes are individual
(gender, occupational aspiration), environmental (academic
field) and cognitive (university entrance score) variables.
In contrast, family background is not found to be a strong
influencing factor. While the positive effect of the cognitive
factor as well as the occupational aspiration can be expected
from former research, the effects of gender and academic
field can be seen as additions of the current state of research
that provide hypotheses for future research.
The two trajectories must be seen against the theoretical
background of the “Fourphase model of interest develop-
ment” (Renninger & Hidi, 2019). The class "continuously
high interest" seems to be able to maintain its continuously
high interest. This group is permanently in the highest
phases of the “Fourphase model of interest development”.
In contrast, the "descending interest" class already seems
to be in a lower phase at the beginning of the study. In the
further course of studies, a descent into a lower phase of the
“Fourphase model of interest development” seems to be
characteristic for this class. It may only be possible to catch
this group over a period of time with situational interest.
Towards the end of their studies, when they are working
towards their final thesis, many people move up again to a
higher level.
The reasons for the downward trend of subject-interest
is complex, but potential reasons for explanations can be
Fig. 4 Coefficient plot for
prediction of membership in
class “continuously high inter-
est” based on logistic regression
(n = 4,345). Notes: table shows
AME Average Marginal Effect,
OA Occupational aspiration,
UEC University entrance score,
Eng engineering (ref. economy),
SW social work (ref. economy),
SOM Social origin middle (ref.
Social origin low), SOH Social
origin high (ref. Social origin
low)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
offered. It is possible that the subject-interests shift to a prac-
tical application of the subject content, so that the reduced
shift in interest results from this. Another explanation for
this development of the class “descending interest” can be
seen in the POI. The start of a study program is linked to a
variety of changes for freshmen’s social and achievement
situation. For example, requirements arise in regulating
one’s learning process and achievement motivation as less
individual support by universities is a characteristic of the
new learning environment (Pillay & Ngcobo, 2010). Con-
sequently, students have to cope with other tasks, which are
often put before subject interests. The upward-trend before
finishing the study program is possibly a result of acquiring
grit.
Analyses of determinants for the membership in classes to
answer research question 2 are in line with the current state
of research. The strong effect of occupational aspiration to
the class membership underline links between interest and
academic choices. This replicates findings by Bargel etal.
(1989) and Schiefele (2009). As already mentioned in the
previous section, this analysis detects an effect of cognitive
factors on interest. Results show that university entrance
score is an important factor influencing class membership.
However, people with a higher university entrance score are
more likely to belong to the class “descending interest”. It is
difficult to explain this result, but maybe there is a mismatch
with the choice of the academic major or potentially too
low demands in the study program. Against the background
of the discussions in educational policy on interest in the
STEM field and gender effects, the present results are likely
to attract particular attention, especially in Germany (Got-
tfried etal., 2001; Krapp, 2018; Su etal., 2009). Findings in
this study show that men as well as students from the aca-
demic fields engineering and social work more often belong
to the class “continuously high interest” than women and
students from other academic fields do.
Against the backdrop of the developmental trajectories
of subject-interest and the influencing factors that affect
it, new research questions occur that need to be addressed
in the future. The question of the correlation between the
development of subject-interests and research-based learn-
ing arises (Wessels etal., 2020). Furthermore, the influence
the COVID-19 pandemic has on this research field as well
as on the entire teaching at universities should be investi-
gated (Ortiz, 2020). It remains open how far the research
results can be generalised. Can the results, for example, also
be transferred to traditional students or regions? Replica-
tion studies need to be initiated here. One possible answer
could be that the result of this study can be generalized to
other geographies with the following characteristics. Rein-
hard etal. (2016) emphasize a generalization of such results
from cooperative education program especially to countries
having a rich history in cooperative education programs,
like South Africa or Namibia. Graf etal. (2014) propose
a transfer of such results from cooperative education pro-
grams to other countries and education systems depending
on the general economic conditions in a target country and
a general interest in the expansion of the tertiary education
sector, for instance through (education) policy reforms and
initiatives in the target country. Against the background of
research in learning and instruction, it would also be inter-
esting to explore the correlation between learning difficulties
and interest.
The present study has several limitations. Firstly, only
students from a single university with twelve locations in
one federal state of Germany were interviewed. Further-
more, this study is only able to use three academic majors
in the datasets. The data is based on cooperative students
that are recruited by the partner companies of the universi-
ties (Kupfer, 2013). Therefore, a generalization of the results
is difficult so that it has to be replicated, for example with
traditional students. This suggestion is already addressed in
the last paragraph. Furthermore, the imprecise measurement
of social origin could contribute to the fact that no signifi-
cant effect was found here. Also, an influence of the different
teaching methods in the three academic majors on interest
cannot be ruled out and should be more investigated in fur-
ther studies. Moreover, a large sample size is used here. In
such situation it must be consider, that in large enough sam-
ple sizes even tiny and practically irrelevant effects become
statistically significant.
However, the study also possesses several strengths. I am
able to use panel data analyzing intraindividual changes of
participants. Furthermore, I do research in higher educa-
tion, which is very rare in the research field of interest. I am
able to access extensive data from the university adminis-
tration and to integrate it into analyses, which increases the
quality of this study. The large sample size of 4,345 partici-
pants as well as the reliable measurements in the field study
≥ 0.81) should also be positively emphasized. The use of
innovative and complex analytical methods further under-
lines the importance of this study.
At this point, I would like to present practical implica-
tions that enable students to increase interest in their field of
study. Even before they start their studies, students have the
opportunity to match their interests with the subject they are
studying. Online self-assessments are a successful method
for this (Ćukušić etal., 2014). A further strengthening of
subject-specific interests can be attained by focusing on the
didactic offers of university lecturers. Based on targeted
offers, the teaching quality can be increased, which can in
turn push situational as well as individual interests (Biggs
& Tang, 2011). A further practical implication to increase
student’s interest and motivation or prevent burnout could
be accompanying coaching programs at the beginning of
their student program at universities. First approaches to
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
implement this are available, but a more precise adaptation
to the topic as well as to the target group is necessary (Unter-
brink etal., 2012).
The results of this study build in important ways on the
extant literature on the development of students’ subject-
interest across the higher education years and offer important
new findings. The present study can be used as a starting
point for further research. I hope that the findings have pro-
duced the first important results for addressing the raised
research question.
Acknowledgements I would also like to thank Joanna Bedersdorfer
(University Heidelberg) for proofreading and for her many helpful sug-
gestions for improving the manuscript.
Funding Open Access funding enabled and organized by Projekt
DEAL.
Data Availability Data is available. Please contact the corresponding
author.
Code Availability Syntax is available. Please contact the correspond-
ing author.
Declarations
Ethical Statement The study was conducted in accordance with the
Declaration of Helsinki. It was approved by Baden-Wuerttemberg
Cooperative State University (8th July 2015) and local heads of the
research groups for ethical standards. All the subjects gave their digital
informed consent.
Consent to Participate Before the participants responded, informed
consent was obtained and the anonymity of responses ensured.
Conflict of Interests Author of this study declare that he has no bio-
medical or financial conflicts of interest to disclose.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article's Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
References
Ames, C., & Archer, J. (1988). Achievement goals in the classroom:
Students’ learning strategies and motivation processes. Journal
of Educational Psychology, 80(3), 260–267. https:// doi. org/ 10.
1037/ 0022- 0663. 80.3. 260
AusbildungPlus. (2020). Duales Studium in Zahlen 2019. Trends und
Analysen [Cooperative Education in Figures2019. Trends and
Analysis]. Federal Institute for Vocational Education and Training.
Bargel, T., Framheim-Peisert, G., & Sandberger, J. U. (1989). Stu-
dienerfahrungen und studentische Orientierungen in den 80er
Jahren. Trends und Stabilitäten [Study experiences and student
orientations in the 1980s. Trends and Stabilities]. Bock.
Best, H., & Wolf, C. (2012). Comparing nested models and interpreting
results from logit and probit regression. KZfSS Kölner Zeitschrift
Für Soziologie Und Sozialpsychologie, 64(2), 377–395. https://
doi. org/ 10. 1007/ s11577- 012- 0167- 4
Biggs, J., & Tang, C. (2011). Teaching for quality learning at Univer-
sity (4th ed.). Open University Press.
Blossfeld, H. .─P., van Maurice, J., & Schneider, T. (2019). The
national educational panel study: need, main features, and research
potential. In H. .─P. Blossfeld & H. .─G. Roßbach (Eds.), Educa-
tion as a lifelong process (2nd ed., pp. 1–16). Springer.
Bong, M. (2001). Role of self-efficacy and task-value in predicting
college students’ course performance and future enrolment inten-
tions. Contemporary Educational Psychology, 26(4), 553–570.
https:// doi. org/ 10. 1006/ ceps. 2000. 1048
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of
measurement invariance. Structural Equation Modeling, 14, 464–
504. https:// doi. org/ 10. 1080/ 10705 51070 13018 34
Coertjens, L., Donche, V., De Maeyer, S., van Daal, T., & Van Pet-
gen, P. (2017). The growth trend in learning strategies during
the transition from secondary to higher education in Flanders.
Higher Educucation, 73, 499–518. https:// doi. org/ 10. 1007/
s10734- 016- 0093- x
Ćukušić, M., Garača, Z., & Jadrić, M. (2014). Online self-assessment
and students’ success in higher education institutions. Comput-
ers & Education, 72, 100–109. https:// doi. org/ 10. 1016/j. compe
du. 2013. 10. 018
Daniels, Z. (2008). Entwicklung schulischer Interessen im Jugendalter
[Development of school interests in adolescence]. Waxmann.
Destatis (= Federal Statistical Office). (2020). Bildung und Kultur.
Studierende an Hochschulen [Education and Culture. Students
at universities]. Federal Statistical Office.
Deuer, E., Meyer, T., Walkmann, R., & Rahn, S. (2020). Das Studi-
enverlaufspanel an der DHBW (2015–2019) [The study panel at
the DHBW (2015–2019)]. In E. Deuer & T. Meyer (Eds.), Stu-
dienverlauf und Studienerfolg im Kontext des dualen Studiums
[Study process and study success in cooperative education] (pp.
5–20). WBV.
Dotterer, A. M., McHale, S. M., & Crouter, A. C. (2009). The develop-
ment and correlates of academic interests from childhood through
adolescence. Journal of Educational Psychology, 101(2), 509–
519. https:// psycn et. apa. org/ doi/ 10. 1037/ a0013 987 .
Durik, A. M., Vida, M., & Eccles, J. S. (2006). Task values and abil-
ity beliefs as predictors of high school literacy choices: A devel-
opmental analysis. Journal of Educational Psychology, 98(2),
382–393. https:// doi. apa. org/ doi/ 10. 1037/ 0022- 0663. 98.2. 382 .
Eccles, J. S. (1983). Expectancies, values, and academic behaviors.
In J. T. Spence (Ed.), Achievement and achievement motives (pp.
75–146). Freeman.
Eccles, J. S., Midgley, C., Wigfield, A., Buchanan, C. M., Reuman, D.,
Flanagan, C., & Mac Iver, D. (1993). Development during adoles-
cence: The impact of stage-environment fit on young adolescents
experiences in schools and in families. American Psychologist,
48(2), 90–101. https:// doi. org/ 10. 1037/ 0003- 066X. 48.2. 90
Fellenberg, F., & Hannover, B. (2006). Kaum begonnen, schon zer-
ronnen? Psychologische Ursachenfaktoren für die Neigung von
Studienanfängern, das Studium abzubrechen oder das Fach zu
wechseln [Easy come, easy go? Psychological causes of students´
drop out of university or changing the subject at the beginning of
their study]. Empirische Pädagogik, 20(4), 381–399.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
Fitzmaurice, G., Davidian, M., Verbeke, G., & Molenberghs, G.
(2009). Longitudinal data analysis. Handbooks of modern sta-
tistical methods. Taylor & Francis Group. https:// doi. org/ 10.
1201/ 97814 20011 579
Frenzel, A. C., Goetz, T., Pekrun, R., & Watt, H. M. G. (2010).
Development of mathematics interest in adolescence: Influences
of gender, family, and school context. Journal of Research on
Adolescence, 20, 507–537. https:// doi. org/ 10. 1111/j. 1532- 7795.
2010. 00645. x
Frenzel, A. C., Pekrun, R., Dicke, A. L., & Goetz, T. (2012). Beyond
quantitative decline: Conceptual shifts in adolescents’ develop-
ment of interest in mathematics. Developmental Psychology,
48(4), 1069–1082. https:// doi. org/ 10. 1037/ a0026 895
Gaspard, H., Lauermann, F., Rose, N., Wigfield, A., & Eccles, J.
S. (2020). Cross-domain trajectories of students’ ability self-
concepts and intrinsic values in math and language arts. Child
Development, 91(5), 1800–1818. https:// doi. org/ 10. 1111/ cdev.
13343
Glover, S., & Dixon, P. (2004). Likelihood ratios: A simple and flex-
ible statistic for empirical psychologists. Psychonomic Bulletin
& Review, 11(5), 791–806. https:// doi. org/ 10. 3758/ BF031 96706
Gottfried, A. E., Fleming, J. S., & Gottfried, A. W. (2001). Continu-
ity of academic intrinsic motivation from childhood through late
adolescence: A longitudinal study. Journal of Educational Psy-
chology, 93(1), 3–13. https:// doi. org/ 10. 1037/ 0022- 0663. 93.1. 3
Graf, L., Powell, J. J. W., Fortwengel, J., & Bernhard, N. (2014). Dual
study programmes in global context: Internationalisation in Ger-
many and transfer to Brazil, France, Qatar, Mexico and the US.
DAAD.
Grossnickle, E. M. (2016). Disentangling curiosity: Dimensionality,
definitions, and distinctions from interest in educational contexts.
Educational Psychology Review, 28(1), 23–60. https:// doi. org/ 10.
1007/ s10648- 014- 9294- y
Guo, J., Wang, M.-T., Ketonen, E. E., Eccles, J. S., & Salmela-Aro,
K. (2018). Joint trajectories of task value in multiple subject
domains: From both variable- and pattern-centered perspectives.
Contemporary Educational Psychology, 55, 139–154. https:// doi.
org/ 10. 1016/j. cedps ych. 2018. 10. 004
Harackiewicz, J. M., Barron, K. E., Tauer, J. M., & Elliot, A. J. (2002).
Predicting success in college: A longitudinal study of achieve-
ment goals and ability measures as predictors of interest and
performance from freshman year through graduation. Journal of
Educational Psychology, 94(3), 562–575. https:// doi. org/ 10. 1037/
0022- 0663. 94.3. 562
Harackiewicz, J. M., Durik, A. M., Barron, K. E., Linnenbrink-Garcia,
L., & Tauer, J. M. (2008). The role of achievement goals in the
development of interest: Reciprocal relations between achieve-
ment goals, interest, and performance. Journal of Educational
Psychology, 100(1), 105–122. https:// doi. org/ 10. 1037/ 0022- 0663.
100.1. 105
Hartinger, A., & Fölling-Albers, M. (2002). Schüler motivieren und
interessieren – Ergebnisse aus der Forschung, Anregungen für
die Praxis [Making pupils motivated and interested - Results from
Research, Suggestions for Practice]. Julius Klinkhardt.
Hauser, R. M., Tsai, S.-L., & Sewell, W.-H. (1983). A model of strati-
fication with response error in social and psychological variables.
Sociology of Education, 56(1), 20–46. https:// doi. org/ 10. 2307/
21123 01
Hidi, S. (2006). Interest: A unique motivational variable. Educational
Research Review, 1(2), 69–82. https:// doi. org/ 10. 1016/j. edurev.
2006. 09. 001
Hidi, S., & Ainley, M. (2008). Interest and self-regulation: Relation-
ships between two variables that influence learning. In D. H.
Schunk & B. J. Zimmerman (Eds.), Motivation and self-regu-
lated learning: Theory, research, and application (pp. 77–109).
Erlbaum.
Hidi, S., & Harackiewicz, J. M. (2000). Motivating the academically
unmotivated: A critical issue for the 21st century. Review of Edu-
cational Research, 70(2), 151–179. https:// doi. org/ 10. 3102/ 00346
54307 00021 51
Hidi, S., & Renninger, K. A. (2006). The four-phase model of interest
development. Educational Psychologist, 41(2), 111–127. https://
doi. org/ 10. 1207/ s1532 6985e p4102_ 4
Hoffmann, L. (2002). Promoting girls’ interest and achievement in
physics classes for beginners. Learning and Instruction, 12(4),
447–465. https:// doi. org/ 10. 1016/ S0959- 4752(01) 00010- X
Hoffmann, L., & Lehrke, M. (1986). Eine Untersuchung über Schüler-
interessen an Physik und Technik [An investigation of students‘
interest in physics and technics]. Zeitschrift Für Pädagogik, 32,
189–204.
Hoffmann, L., Lehrke, M., & Todt, E. (1985). Development and change
of pupils’ interest in physics: Design of longitudinal study (grade
5–10). In M. Lehrke, L. Hoffmann, & P. L. Gardner (Eds.), Inter-
ests in science and technology (pp. 71–80). Institute for Science
Education.
Høgheim, S., & Reber, R. (2019). Interesting, but less interested: Gen-
der differences and similarities in mathematics interest. Scandi-
navian Journal of Educational Research, 63(2), 285–299. https://
doi. org/ 10. 1080/ 00313 831. 2017. 13364 82
Honaker, J., King, G., & Blackwell, M. (2011). Amelia II: A program
for missing data. Journal of Statistical Software, 45(7), 1–47.
https:// doi. org/ 10. 18637/ jss. v045. i07
Jung, T., & Wickrama, K. A. S. (2008). An introduction to latent class
growth analysis and growth mixture modeling. Social and Per-
sonality Psychology Compass, 2(1), 302–317. https:// doi. org/ 10.
1111/j. 1751- 9004. 2007. 00054. x
Kim, S., Klager, C., & Schneider, B. (2019). The effects of alignment
of educational expectations and occupational aspirations on labor
market outcomes: Evidence from NLSY79. The Journal of Higher
Education, 90(6), 992–1015. https:// doi. org/ 10. 1080/ 00221 546.
2019. 16153 33
Krapp, A. (1992). Das Interessenkonstrukt: Bestimmungsmerkmale
der Interessenhandlung und des individuellen Interesses aus der
Sicht einer Person-Gegenstands-Konzeption [The construct of
interest: determinants of interest action and individual interest
from the perspective of a person-object conception]. In A. Krapp
& M. Prenzel (Eds.), Interesse, Lernen, Leistung: Neuere Ansätze
der pädagogisch-psychologischen Interessenforschung [Interest,
Learning, Performance: Recent Approaches of Interest Research
in Educational Psychological] (pp. 297–329). Aschendorff.
Krapp, A. (2002). Structural and dynamic aspects of interest develop-
ment: Theoretical considerations from an ontogenetic perspec-
tive. Learning and Instruction, 12(4), 383–409. https:// doi. org/
10. 1016/ S0959- 4752(01) 00011- 1
Krapp, A. (2005). Basic needs and the development of interest and
intrinsic motivational orientations. Learning and Instruction,
15(5), 381–395. https:// doi. org/ 10. 1016/j. learn instr uc. 2005. 07.
007
Krapp, A. (2007). An educational–psychological conceptualisation of
interest. International Journal for Educational and Vocational
Guidance, 7(1), 5–21. https:// doi. org/ 10. 1007/ s10775- 007- 9113- 9
Krapp, A. (2018). Interesse [Interest]. In D. H. Rost, J. R. Sparfeldt, &
S. R. Buch (Eds.), Handwörterbuch Pädagogische Psychologie
[Handbook of Educational Psychology] (5th ed., pp. 286–297).
Beltz.
Kupfer, F. (2013). Duale Studiengänge aus Sicht der Betriebe: Praxis-
nahes Erfolgsmodell durch Bestenauslese [Cooperative education
programmes from the point of view of the companies: Practical
success model through selection of the best]. Berufsbildung in
Wissenschaft und Praxis – BWP, 47(4), 25–29.
Kyndt, E., Coertjens, L., van Daal, T., Donche, V., Gijbels, D., & Van
Petegem, P. (2015). The development of students’ motivation in
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
the transition from secondary to higher education: A longitudinal
study. Learning and Individual Differences, 39, 114–123. https://
doi. org/ 10. 1016/j. lindif. 2015. 03. 001
Kyndt, E., Donche, V., Coertjens, L., van Daal, T., Gijbels, D., & Van
Petegem, P. (2019). Does self-efficacy contribute to the develop-
ment of students’ motivation across the transition from secondary
to higher education? European Journal of Psychology of Educa-
tion, 34, 457–478. https:// doi. org/ 10. 1007/ s10212- 018- 0389- 6
Liebendörfer, M., & Schukajlow, S. (2017). Interest development dur-
ing the first year at university: Do mathematical beliefs predict
interest in mathematics? ZDM, 49, 355–366. https:// doi. org/ 10.
1007/ s11858- 016- 0827- 3
Little, R. J. (1988). A test of missing completely at random for multi-
variate data with missing values. Journal of the American Statis-
tical Association, 83, 1198–1202. https:// doi. org/ 10. 1080/ 01621
459. 1988. 10478 722
Loewenstein, G. (1994). The psychology of curiosity: A review and
reinterpretation. Psychological Bulletin, 116(1), 75–98. https://
doi. org/ 10. 1037/ 0033- 2909. 116.1. 75
Maurice, J., Dörfler, T., & Artelt, C. (2014). The relation between
interests and grades: Path analyses in primary school age. Inter-
national Journal of Educational Research, 64, 1–11. https:// doi.
org/ 10. 1016/j. ijer. 2013. 09. 011
McDonald, R. P. (1999). Test theory: A unified treatment. Lawrence
Erlbaum.
Mood, C. (2010). Logistic regression: Why we cannot do what we think
we can do, and what we can do about it. European Sociological
Review, 26(1), 67–82. https:// doi. org/ 10. 1093/ esr/ jcp00 6
Musu-Gillette, L. E., Wigfield, A., Harring, J. R., & Eccles, J. S.
(2015). Trajectories of change in students’ self-concepts of abil-
ity and values in math and college major choice. Educational
Research and Evaluation, 21(4), 343–370. https:// doi. org/ 10.
1080/ 13803 611. 2015. 10571 61
Neugebauer, M., Heublein, U., & Annabel, D. (2019). Higher education
dropout in Germany: Extent, causes, consequences, prevention.
Zeitschrift Für Erziehungswissenschaft, 22, 1025–1046. https://
doi. org/ 10. 1007/ s11618- 019- 00904- 1
Ortiz, P. A. (2020). Teaching in the time of COVID-19. Biochemistry
and Molecular Biology Education: A Bimonthly Publication of
the International Union of Biochemistry and Molecular Biology,
48(3), 201. https:// doi. org/ 10. 1002/ bmb. 21348
Pany, P., Lörnitzo, A., Auleitner, L., Heidinger, C., Lampert, P., &
Kiehn, M. (2019). Using students’ interest in useful plants to
encourage plant vision in the classroom. Plants, People, Planet,
Special Issue: Standing in the Shadows of Plants: New Perspec-
tives on Plant Blindness, 1(3), 261–270. https:// doi. org/ 10. 1002/
ppp3. 43
Pillay, A. L., & Ngcobo, H. S. B. (2010). Sources of stress and support
among rural-based first-year university students: An exploratory
study. South African Journal of Psychology, 40, 234–240. https://
doi. org/ 10. 1177/ 00812 46310 04000 302
Protsch, P., & Solga, H. (2015). How employers use signals of cogni-
tive and noncognitive skills at labour market entry: Insights from
field experiments. European Sociological Review, 31(5), 521–532.
Putnick, D. L., & Bornstein, M. H. (2016). Measurement Invariance
conventions and reporting: The state of the art and future direc-
tions for psychological research. Developmental Review: DR, 41,
71–90. https:// doi. org/ 10. 1016/j. dr. 2016. 06. 004
Ram, N., & Grimm, K. J. (2009). Methods and measures: Growth
mixture modeling: A method for identifying differences in longi-
tudinal change among unobserved groups. International Journal
of Behavioral Development, 33(6), 565–576. https:// doi. org/ 10.
1177/ 01650 25409 34376 5
Reinecke, J. (2006). Longitudinal analysis of adolescents’ deviant and
delinquent behavior. Methodology, 2, 100–112. https:// doi. org/ 10.
1027/ 1614- 2241.2. 3. 100
Reinhard, K., Pogrzeba, A., Townsend, R., & Pop, A. P. (2016). A
comparative study of cooperative education and work integrated
learning in Germany, South Africa, and Namibia. Asia-Pacific
Journal of Cooperative Education, 17(3), 249–263.
Renninger, K. A., & Hidi, S. (2016). The power of interest for motiva-
tion and engagement. Routledge.
Renninger, K., & Hidi, S. (2019). Interest development and learning.
In K. Renninger & S. Hidi (Eds.), The Cambridge handbook of
motivation and learning (pp. 265–290). Cambridge University
Press. https:// doi. org/ 10. 1017/ 97813 16823 279. 013
Renninger, K. A., Ewen, L., & Lasher, A. K. (2002). Individual interest
as context in expository text and mathematical word problems.
Learning and Instruction, 12(4), 467–490. https:// doi. org/ 10.
1016/ S0959- 4752(01) 00012- 3
Richard, F. D., Bond, C. F., & Stokes-Zoota, J. J. (2003). One hun-
dred years of social psychology quantitatively described. Review
of General Psychology, 7(4), 331–363. https:// doi. org/ 10. 1037/
1089- 2680.7. 4. 331
Richardson, M., Abraham, C., & Bond, R. (2012). Psychological cor-
relates of university students’ academic performance: A system-
atic review and meta-analysis. Psychological Bulletin, 138(2),
353–387. https:// doi. org/ 10. 1037/ a0026 838
Rotgans, J. I., & Schmidt, H. G. (2011). Situational interest and aca-
demic achievement in the active-learning classroom. Learning
and Instruction, 21(1), 58–67. https:// doi. org/ 10. 1016/j. learn instr
uc. 2009. 11. 001
Sabates, R., Harris, A. L., & Staff, J. (2011). Ambition gone awry:
The long-term socioeconomic consequences of misaligned and
uncertain ambitions in adolescence. Social Science Quarterly,
92(4), 959–977. Retrieved June 18, 2021, from http:// www. jstor.
org/ stable/ 42956 558
Sansone, C., & Thoman, D. B. (2005). Interest as the missing motivator
in self-regulation. European Psychologist, 10(3), 175–186. https://
doi. org/ 10. 1027/ 1016- 9040. 10.3. 175
Scherrer, V., Preckel, F., Schmidt, I., & Elliot, A. J. (2020). Develop-
ment of achievement goals and their relation to academic inter-
est and achievement in adolescence: A review of the literature
and two longitudinal studies. Developmental Psychology, 56(4),
795–814. https:// doi. org/ 10. 1037/ dev00 00898
Schiefele, U. (2009). Situational and individual interest. In K. R. Wen-
tzel & A. Wigfield (Eds.), Handbook of motivation at school (pp.
197–222). Routledge.
Schiefele, U., Krapp, A., & Schreyer, I. (1993). Metaanalyse des
Zusammenhangs von Interesse und schulischer Leistung [Meta-
analysis of interest and academic achievement]. Zeitschrift Für
Entwicklungspsychologie Und Pädagogische Psychologie, 25(2),
120–148.
Schiefer, I. M., Becker, S., & Artelt, C. (2018). A person centered
approach for analyzing the development of students’ subject-inter-
ests in language arts, Mathematics, and English as a foreign lan-
guage from Grade 4 to 11. Psychologie in Erziehung Und Unter-
richt, 65(2), 141–157. https:// doi. org/ 10. 2378/ PEU20 17. art24 d
Schnettler, T., Bobe, J., Scheunemann, A., Fries, S., & Grunschel, C.
(2020). Is it still worth it? Applying expectancy-value theory to
investigate the intraindividual motivational process of forming
intentions to drop out from university. Motivation and Emotion,
44, 491–507. https:// doi. org/ 10. 1007/ s11031- 020- 09822- w
Sewell, W.-H., Haller, A. O., & Portes, A. (1969). The educational
and early occupational attainment process. American Sociological
Review, 34(1), 82–92.
Su, R., Rounds, J., & Armstrong, P. I. (2009). Men and things, women
and people: A meta-analysis of sex differences in interests. Psy-
chological Bulletin, 135(6), 859–884. https:// doi. org/ 10. 1037/
a0017 364
Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics
(6th ed.). Pearson.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Current Psychology
1 3
Thiel, F., Veit, S., Blüthmann, I., Lepa, S. & Ficzko, M. (2008). Ergeb-
nisse der Befragung der Studierenden in den Bachelorstudiengän-
gen an der Freien Universität Berlin - Sommersemester 2008
[Results of students in Bachelor's degree programmes survey at
Freie Universität Berlin - Summer term 2008]. Retrieved February
27, 2021, from https:// www. geo. fu- berlin. de/ studi um/ Quali taets
siche rung/ Resso urcen/ FU_ bache lorbe fragu ng_ 2008. pdf
Todt, E., Arbinger, R., Seitz, H., & Wildgrube, W. (1974). Untersu-
chungen über die Motivation zur Beschäftigung mit naturwissen-
schaftlichen Problemen (Sekundarstufe I: Klassen 5–9) [Studies
on the motivation to work with science problems (secondary level:
Grades 5–9)]. Universität, Psychologisches Institut.
Unterbrink, T., Pfeifer, R., Krippeit, L., Zimmermann, L., Rose, U.,
Joos, A., Hartmann, A., Wirsching, M., & Bauer, J. (2012). Burn-
out and effort-reward imbalance improvement for teachers by a
manual-based group program. International Archives of Occupa-
tional and Environmental Health, 85(6), 667–674. https:// doi. org/
10. 1007/ s00420- 011- 0712- x
Vogler-Ludwig, K., Düll, N., & Kriechel, B. (2016). Arbeitsmarkt
2030. Wirtschaft und Arbeitsmarkt im digitalen Zeitalter. Prog-
nose 2016. Kurzfassung. [Labour market 2030. Economy and
labor market in the digital age. Prognosis 2016. Summary].
Economix.
Wang, M. T., Chow, A., Degol, J. L., & Eccles, J. S. (2017). Does
everyone’s motivational beliefs about physical science decline
in secondary school?: Heterogeneity of adolescents’ achieve-
ment motivation trajectories in physics and chemistry. Journal
of Youth and Adolescence, 46(8), 1821–1838. https:// doi. org/ 10.
1007/ s10964- 016- 0620- 1
Wessels, I., Rueß, J., Gess, C., Deicke, W., & Ziegler, M. (2020). Is
research-based learning effective? Evidence from a pre–post anal-
ysis in the social sciences. Studies in Higher Education. https://
doi. org/ 10. 1080/ 03075 079. 2020. 17390 14
Wild, S., & Alvarez, S. (2020). Cooperative education in the higher
education system and Big Five personality traits in Germany.
International Journal of Work-Integrated Learning, 21(1), 37–49.
Wild, S., & Neef, C. (2019). The role of academic major and academic
year for self-determined motivation in cooperative education.
Industry and Higher Education, 33(5), 327–339. https:// doi. org/
10. 1177/ 09504 22219 843261.
Wild, S., & Schulze Heuling, L. (2020). Student dropout and retention:
an event history analysis among students in cooperative higher
education. International Journal of Educational Research, 104.
https:// doi. org/ 10. 1016/j. ijer. 2020. 101687
Xu, C., Lern, S., & Onghena, P. (2021). Examining developmental rela-
tionships between utility value, interest, and cognitive competence
for college statistics students with differential self-perceived math-
ematics ability. Learning and Individual Differences, 86, 101980.
https:// doi. org/ 10. 1016/j. lindif. 2021. 10198 0
Yates, S., Harris, A., Sabates, R., & Staff, J. (2011). Early Occupa-
tional Aspirations andFractured Transitions: A Study of Entry
into‘NEET’ Status in the UK. Journal of Social Policy, 40(3),
513–534. https:// doi. org/ 10. 1017/ S0047 27941 00006 56
Publisher's note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
... Third, in terms of student motivation, it has been shown that there is a clear relationship between this variable and university dropout cross-culturally, obtaining the same clear positive tie between high motivation and a lower probability of university dropout in different countries and diverse cultural settings [65][66][67][68][69][70][71][72][73]. In fact, a correlation has been found between motivation and positive behavior of students, as well as favorable involvement in the learning process and academic achievement [7,46,[74][75][76][77][78][79][80][81], in such a way that the most motivated students show less disruptive and/or challenging behavior, greater commitment to the learning process, and higher probabilities of achieving academic achievement. In the same way, recent studies have shown that the predictive value of motivation with respect to the probability of dropping out is clear, revealing that students with greater motivation are more resistant to the problem of dropping out, showing greater probabilities of completing the academic year [7,82,83]. ...
... In fact, a correlation has been found between motivation and positive behavior of students, as well as favorable involvement in the learning process and academic achievement [7,46,[74][75][76][77][78][79][80][81], in such a way that the most motivated students show less disruptive and/or challenging behavior, greater commitment to the learning process, and higher probabilities of achieving academic achievement. In the same way, recent studies have shown that the predictive value of motivation with respect to the probability of dropping out is clear, revealing that students with greater motivation are more resistant to the problem of dropping out, showing greater probabilities of completing the academic year [7,82,83]. ...
... This finding is consistent with that obtained by other recent studies since it has been shown that the persistence and university dropout of university students depends on a combination of individual, institutional, and economic factors, whose effects on the decision to drop out are mediated by the student's ability to successfully integrate into the academic system [1]. In addition, beyond the situation of vulnerability or social exclusion present in university students, other variables have been detected, such as motivation, which plays a key predictive role in the academic achievement of students [7,68,70,73,143], thus preventing dropout or failure and empowering the students to overcome difficulties, even when starting from disadvantaged social situations. ...
Article
Full-text available
The failure and dropout of university studies are issues that worry all nations due to the personal, social, and economic costs that this they entail. Because the dropout phenomenon is complex and involves numerous factors, to reverse it would involve a comprehensive approach through interventions aimed at the factors identified as key in the decision to drop out. Therefore, the main objective of this work is to determine the profile of students who enter the EPN (STEM higher-education institution) to analyze the characteristics that differentiate students who drop out early in their career and those who stay in school. A sample of 624 students who accessed the EPN leveling course (a compulsory course at the beginning of their studies) participated in the study. A total of 26.6% of the participants were women. A total of 50.7% of the participants passed the course. Data referring to social, economic, and academic variables were analyzed. Comparison techniques, as well as artificial neural networks, were used to compare characteristic profiles of students who passed the leveling course and those who dropped out. The results showed significant differences between the profiles of the students who passed and those who dropped out with regard to the variables related to previous academic performance and motivational and attributional aspects. The artificial neural networks corroborated the importance of these variables in predicting dropout. In this research, the key variables predicting whether a student continues or leaves higher education are revealed, allowing the identification of students at possible risk of dropping out and thus promoting initiatives to provide adequate academic support and improve student retention.
Chapter
The most common cause of death among patients with cardiovascular diseases is myocardial infarction (MI). Identifying predictors for in-hospital mortality is an essential step toward MI prevention and consequent reduction in mortality. We aimed to develop machine learning (ML) methods for predicting in-hospital mortality in MI patients and apply novel techniques for models’ interpretability to detect the predictive importance of the variables. Random forest (RF) and extreme gradient boosting (XGB) are applied to a dataset of 2035 MI patients who underwent percutaneous coronary intervention. When comparing the models’ AUC (RF—0.9712 vs. XGB—0.9666) and accuracy (RF—97% versus XGB—98%), both techniques achieved similar performance. However, the RF model obtained a higher sensitivity (86%) than the XGBoost classifier (80%). Hypertension, cardiogenic shock, ejection fraction were identified as some of the main contributors to the outcome. Our paper contributes to the global effort of reducing mortality in patients with myocardial infarction by proposing two interpretable ML models that accurately predict in-hospital mortality in MI patients. These results are essential steps in improving current preventive strategies. However, future studies on larger datasets enriched with both categorical and continuous variables, and models validated on external data from other centers are needed to accurately assume generalizability in clinical practice.
Article
Full-text available
Research-based learning (RBL) is regarded as a panacea when it comes to effective instructional formats in higher education settings. It is said to improve a wide set of research-related skills and is a recommended learning experience for students. However, whether RBL in the social sciences is indeed as effective as has been postulated for other disciplines has not yet been systematically examined. We thus administered a pre–post-test study to N = 952 students enrolled in 70 RBL courses at 10 German universities and examined potential changes in cognitive and affective-motivational research dispositions. Latent change score modelling indicated that students increased their cognitive research dispositions, whereas most affective-motivational research dispositions decreased. The instructors’ interest in the students’ work served as a significant predictor of changes in research interest and joy. Practical implications for designing RBL environments can be inferred from the results.
Article
Full-text available
The intraindividual process of study dropout, from forming dropout intention to deregistration, is of motivational nature. Yet typical studies investigate interindividual differences, which do not inform about intraindividual processes. Our study focused on the intraindividual process of forming dropout intention, and applied expectancy-value theory to analyze its motivational underpinnings. To expand research, we considered associations of intraindividual deviations in expectancy, intrinsic value, attainment value, utility value, and cost to intraindividual deviations in dropout intention. A total of 326 undergraduate students of law and mathematics rated motivational variables and dropout intention three times from semester start to the final exam period. Multilevel regression analyses revealed that intraindividual changes in intrinsic value, attainment, and cost, but not in expectancy and utility, related to intraindividual changes in dropout intention. Further, we considered students’ demographics as moderators. Only age moderated the association between intrinsic value and dropout intention. Our results stress the crucial role of certain value components, including cost, for emerging dropout intention.
Article
Full-text available
Adolescence is important for the development of achievement motivation, including achievement goal pursuit. Longitudinal research is scarce on adolescents' goal development and its implications for academic outcomes. In our research, we first present a systematic review of findings on achievement goals in adolescence. Then we report 2 longitudinal studies with German adolescents in which we investigated the separate as well as joint development of achievement goals, interest, and achievement in the domain of mathematics. Study 1 comprised 745 students assessed in 4 waves in grades 5-7 (43% female; age MT1 = 10.66). Study 2 comprised 1,420 students assessed in 4 waves in grades 5-8 (47% female; age MT1 = 10.58). Students reported their mastery, performance-approach, and performance-avoidance goals and their interest in mathematics. Mathematics achievement was assessed by school grades (Studies 1 and 2) and standardized test scores (Study 1). Data were analyzed using latent growth curve, multiple process, and cross-lagged models. Findings for the 2 studies evidenced a substantial degree of consistency. All goals decreased and the decrease became smaller over time in most instances. Controlling for nonfocal goals and demographic variables (socioeconomic status, gender), multiple process models revealed that change in mastery and performance-approach goals was positively related or unrelated to change in interest and achievement, whereas change in performance-avoidance goals was negatively related or unrelated to change in interest and achievement. Cross-lagged models revealed that relations between the achievement goals on the one hand and interest and achievement on the other hand were reciprocal rather than unidirectional. (PsycINFO Database Record (c) 2020 APA, all rights reserved).
Article
Full-text available
Different cross‐domain trajectories in the development of students’ ability self‐concepts (ASCs) and their intrinsic valuing of math and language arts were examined in a cross‐sequential study spanning Grades 1 through 12 (n = 1,069). Growth mixture modeling analyses identified a Moderate Math Decline/Stable High Language Arts class and a Moderate Math Decline/Strong Language Arts Decline class for students’ ASC trajectories. Students’ intrinsic value trajectories included a Strong Math Decline/Language Arts Decline Leveling Off, a Moderate Math Decline/Strong Language Arts Decline, and a Stable Math and Language Arts Trajectories class. These classes differed with regard to student characteristics, including gender, family background, and math and reading aptitudes. They also resulted in different high school math course enrollments, career aspirations, and adult careers.
Article
Full-text available
The phenomenon known as “plant blindness” leads to a restricted view on nature, which may in turn result in people and society having less understanding of important environmental and conservation issues. Biology educators must tackle this problem directly. By exploring students’ interests in useful plants (spice plants and stimulant herbal drugs), we present a route by which educators could counteract plant blindness by taking students’ plant‐related interests into account when preparing a teaching unit. We reveal that educators must carefully select the plant materials used as examples in the classroom to effectively encourage plant vision in students. Summary The term “plant blindness” describes the phenomenon that many people tend to overlook plants in their everyday life. Therefore, they often do not understand key biological concepts to their full extent and may develop a more restricted view of nature. Here, we provide a possible method of counteracting plant blindness in schools by capitalizing on students’ interest in useful plants to impart botanical content in biology lessons. We investigated 172 students aged 10–18 years old in order to explore the links between students’ individual interests in spice plants and stimulant herbal drugs and their situational interest during an inquiry‐based botany teaching unit by using a questionnaire and additional semi‐structured interviews. Data analysis shows that, for students with low and medium individual interest, stimulant herbal drugs and spice plants can be used to generate higher situational interest. Furthermore, students with initially high individual interest should get the opportunity to follow their own questions concerning a plant group before working on a structured learning task. Using study objects from different subgroups of useful plants for introducing botanical content in biology lessons, and giving highly interested students time and space to follow their own interests, seems to be a promising way to counteract plant blindness and, thus, to help enable students to develop a more comprehensive view of nature. The phenomenon known as “plant blindness” leads to a restricted view on nature, which may in turn result in people and society having less understanding of important environmental and conservation issues. Biology educators must tackle this problem directly. By exploring students’ interests in useful plants (spice plants and stimulant herbal drugs), we present a route by which educators could counteract plant blindness by taking students’ plant‐related interests into account when preparing a teaching unit. We reveal that educators must carefully select the plant materials used as examples in the classroom to effectively encourage plant vision in students.
Article
Although there is a growth of interest in student motivation in higher education, not enough is known about the developmental relationships between motivational components among college students. In this study, a sample of 503 students in a five-year combined Bachelor's/Master's degree program at a Belgian university reported their statistics cognitive competence, utility value, and interest at the beginning of or prior to each of three progressively more advanced statistics courses. Results from latent growth curve analysis (LGCA) revealed that students' self-perceived mathematics ability at the beginning of the first course was positively associated with initial levels of statistics interest, utility value, and cognitive competence but not with their developmental trends. Moreover, results from multigroup LGCA suggested that students who reported lower mathematics ability began the first statistics courses with lower initial levels of interest, utility value and cognitive competence, but experienced development in these constructs at similar rates as their peers reporting higher mathematics ability. We also found that rates of growth in cognitive competence or value were positively associated with rates of decrease in interest. Implications of these findings for instructional practice are discussed.
Article
Although there are many reasons why students dropout of higher education, those reasons may be unique for students who are enrolled in a cooperative study program. The issue of a second educational player, which is the workplace training side, and the above-average study load may be factors influencing the student’s decision to leave the program. Looking at four waves of panel data collected between 2016 and 2019, we analyse a sample of 19,769 cooperative students in Germany, including 852 dropouts. A Cox regression analysis was applied to identify indicators for dropout risk. The results show that personality and commitment to one’s company matter alongside the known determinants of cognitive skills. Recommendations for agents (student advisory services, companies, etc.) that wish to counteract dropout and motivate students’ workplace commitment are incorporated in the conclusions.
Article
Students in psychology are often surveyed for personality. Empirical results show relevance of this topic, because personality influences academic success. In contrast, we know much less about the personality of students of cooperative education. So, we collected data from 5,863 students at Baden-Wuerttemberg Cooperative State University in August 2016. Multivariate analysis of covariance (MANCOVA) indicated that gender (partial η 2 = .06), academic field (partial η 2 = .01) and covariate university entrance scores (partial η 2 = .05) have an effect on Big Five personality traits controlling age and social class. The results can be used for selecting students in cooperative education for academic fields and compare them with other types of study.