A longitudinal study investigating the role of basic psychological needs and autonomous motivation in explaining students' achievement and dropout from teacher education

Abstract

This pre-registered longitudinal study investigated the relations between basic psychological need satisfaction and motivational quality and objective measures of achievement and dropout, to extend principles of self-determination theory in the classroom to modelling of psychological need satisfaction alongside long-term objective academic outcomes. Participants were first-year and fourth-year student teachers in Norway, a demographic known for having high attrition rates. Unexpectedly, we found that autonomous motivation and amotivation were negatively related with achievement, whereas gender (males) and previous grades were positively related with it. Controlled motivation and gender (males) was, conversely, positively related with remaining on the study program. As expected, amotivation was related with dropout. Finally, the effect of autonomous motivation on remaining in education was mediated by basic psychological needs, suggesting that autonomous motivation indirectly reduces dropout through the satisfaction of the basic needs. We discuss the limitations of our study and implications for future research.

Full text

MOTIVATION, ACHIEVEMENT AND DROPOUT
A longitudinal study investigating the role of basic psychological needs and autonomous motivation in
explaining students’ achievement and dropout from teacher education
Lucas M. Jeno1, Chantal Levesque-Bristol2, Jorun Nylehn3, Zeljana Pavlovic1, Dag Roness1,
& Netta Weinstein4
1Department of Education, University of Bergen
2Center for Instructional Excellence, Purdue University
3Department of Biological Sciences, University of Bergen
4School of Psychology and Clinical Language Sciences, University of Reading
Authors footnotes
Correspondence regarding this article should be addressed to Lucas M. Jeno, Department of
Education, University of Bergen, Norway. Email: lucas.jeno@uib.no
Acknowledgment
This study was funded by UiB’s Humaniorastrategy 2018-2022 for the project “The role of
motivation in explaining teacher students’ study quality and dropout” awarded to the first
author.
Draft version June 19th, 2024. This manuscript has not been peer reviewed.

MOTIVATION, ACHIEVEMENT AND DROPOUT
Abstract
This pre-registered longitudinal study investigated the relations between basic psychological
need satisfaction and motivational quality and objective measures of achievement and
dropout, to extend principles of self-determination theory in the classroom to modelling of
psychological need satisfaction alongside long-term objective academic outcomes.
Participants were first-year and fourth-year student teachers in Norway, a demographic
known for having high attrition rates. Unexpectedly, we found that autonomous motivation
and amotivation were negatively related with achievement, whereas gender (males) and
previous grades were positively related with it. Controlled motivation and gender (males)
was, conversely, positively related with remaining on the study program. As expected,
amotivation was related with dropout. Finally, the effect of autonomous motivation on
remaining in education was mediated by basic psychological needs, suggesting that
autonomous motivation indirectly reduces dropout through the satisfaction of the basic needs.
We discuss the limitations of our study and implications for future research.

MOTIVATION, ACHIEVEMENT AND DROPOUT
A longitudinal study investigating the role of basic psychological needs and autonomous
motivation in explaining students’ achievement and dropout from teacher education
Growing concern for educational quality in schools worldwide cites insufficient graduation
rates and subsequent recruitment of qualified teachers (Alatalo et al., 2024; UNESCO, 2013).
Traditionally, teacher education programs at universities have admitted students with
relatively good grades from upper secondary schools (NOKUT, 2022; Skagen & Elstad,
2023). However, for decades there has been an alarming trend of declining admission to
university colleges and of less qualified students being admitted (NOKUT, 2022), a trend that
has increased further after the COVID-19 pandemic.
Adding to this, there is a higher proportion of students leaving their teacher education and
fewer students applying for admission to teacher education than in previous years (Bakken,
2022; Statistics Norway, 2024). In Norway, approximately 24% of student teachers drop out
from their study program, and only 34% of the student teachers graduate within the stipulated
time (Bakken, 2022). The majority of dropouts occur at the start of the program – that is
between the first and second year of study (55% of the dropouts) – with men, older students,
and low-performing students from upper secondary education having a greater likelihood of
dropping out. A large proportion of this dropout rate can be explained by the level of
educational and high school achievement levels from high school (Andresen & Lervåg, 2022).
The economic situation for students has also worsened over recent decades in Norway in
particular, and students are forced to spend a lot of time working alongside their studies
(Wikan & Bugge, 2014). Simultaneously, study and contact time have decreased, which may
be further contributing to dropout and poorer performance (Wikan & Bugge, 2014). Recently,
Böhn and Deutscher (2022) conducted a meta-analysis and systematic review of dropout
within initial vocational education and training (VET) and found that professional factors that
contributed to dropout included changes in interest, wrong career choice, different
expectations, and perceived support or guidance from the educational institution. This
suggests that, although there are some fixed factors that explain dropout (i.e., gender, age,
prior academic achievement), there are some motivational factors (choice, interest,
expectancies, support) that might also be predictors of dropout beyond the effect of fixed
factors.
While factors such as gender and age are more stable, motivational factors are more
malleable (Robinson, 2023). Specifically, when it comes to teacher education, motivation has
been shown to be highly important for choosing the study program, studying, and for
continuation on the program (Elstad et al., 2023; Roness, 2011). Given the importance of

MOTIVATION, ACHIEVEMENT AND DROPOUT
basic psychological need satisfaction for motivation (Bureau et al., 2021; Vansteenkiste et al.,
2023), and the strong effect of motivation on academic adjustment (Howard et al., 2021), this
study examines whether basic psychological need satisfaction and motivation act as predictive
factors for achievement and have a shielding effect against dropping out of a program.
To our knowledge, no previous studies have investigated the longitudinal effects of basic
psychological needs and motivation on actual achievement and dropout among student
teachers in Norway. Most international longitudinal studies investigate the role of motivation
on either achievement (e.g., Alamer & Alrabai, 2022; Corpus et al., 2022) or the intention to
dropout and achievement (e.g., Corpus et al., 2020; Messerer et al., 2023), without
considering the role of basic psychological need satisfaction. Thus, we extend the literature in
this field by including the role of basic psychological needs as a potential predictor of
achievement and dropout and as a mediator between motivation and achievement and
dropout. As such, the primary aim of this study is to longitudinally investigate student
teachers’ motivation during study periods which are known to have high attrition rates and
investigate the impact of motivation on achievement and dropout (e.g., Tinto, 1993).
Analyzing the motivational quality that explains achievement and dropout in a sample with
high attrition rates may be especially important for targeting intervention strategies that can be
employed to support motivation over time.
The correlates of basic psychological needs and motivation in higher education
Our theoretical starting point is informed by Self-Determination Theory (SDT; Ryan &
Deci, 2017). SDT is a broad theory of human motivation that emphasizes the role of the
impact of social environments on the individual’s basic psychological need satisfaction and
motivation, and the differential impact of motivation on academic outcomes.
SDT states that satisfaction of basic psychological needs for autonomy (i.e., to feel a sense
of agency), competence (i.e., to feel effective), and relatedness (i.e., to feel cared for) are
important conditions for students to thrive psychologically and academically (Ryan & Deci,
2020). Studies within SDT have shown that satisfaction of these basic psychological needs are
related to well-being (Chen et al., 2015; León & Núñez, 2013; Yu et al., 2017), prosocial
behavior (Jang et al., 2019) and autonomous motivation (Slemp et al., 2020). A further
assumption within SDT is that the satisfaction of these basic psychological needs mediates the
effect of the social context on outcomes (Ryan & Deci, 2017) – in other words, individuals’
perception of the social context and the effect of outcomes is dependent upon the relative
satisfaction or frustration of these basic psychological needs. Studies within higher education
have consistently shown that basic psychological need satisfaction is beneficial for study

MOTIVATION, ACHIEVEMENT AND DROPOUT
intentions (Young-Jones et al., 2019), intrinsic motivation (Hernández et al., 2022), and
grades (Filak & Sheldon, 2008; Guiffrida et al., 2013), whereas basic psychological need
frustration reduce students’ well-being (Longo et al., 2016) and study engagement (Hughes et
al., 2023).
In contrast with most other motivational theories (e.g., Cook & Artino Jr, 2016), SDT
differentiates between three qualities of motivation, depending on the relative degree of
internalization – that is, the extent to which they are self-driven – and autonomy (i.e., self-
determination). These three different qualities of motivation have different motivational
forces, emotional experiences, involvement levels, and correlates associated with them
(Koestner & Losier, 2002).
The least internalized (i.e., self-driven) quality of motivation is controlled motivation.
Controlled motivation is regulated by either external control such as compliance, rewards, or
avoiding punishment, or internal control such as self-control, ego-involvement, or attaining
self-esteem (Ryan & Deci, 2020). Previous studies show that controlled motivation is
negatively correlated with optimal learning indicators such as time management,
concentration, and attitude (Vansteenkiste et al., 2005), and positively correlated with dropout
intentions (Jeno et al., 2018), surface approaches to learning (Kusurkar et al., 2013), and test-
anxiety (Iraola-Real et al., 2022). Conversely, while some studies have linked it to negative
learning outcomes, others have suggested that controlled motivation may have some
energizing effects, which in turn predicts greater achievement (e.g., Botnaru et al., 2021; Jeno
et al., 2021). However, researchers also suggest that this form of motivation is fragile because
the associated emotional experience is negative and the underlying motivational force is
driven by compliance and compulsion (Koestner & Losier, 2002).
In contrast to controlled motivation, autonomous motivation reflects more internalized
motivation and is characterized by a high level of personal volition, is endorsed and driven by
the self, and feels choiceful (Ryan & Deci, 2017). Autonomous motivation has been largely
correlated with positive outcomes within academic literature. When autonomously motivated,
individuals are regulated by personal importance and value, congruence in behaviors and
actions, and interest and enjoyment (Ryan & Deci, 2017). Indeed, studies have shown that
autonomous motivation is related to academic satisfaction (Breva & Galindo, 2020),
perceived knowledge transferability (Levesque-Bristol et al., 2020), adaptive cognitive
strategies (Manganelli et al., 2019), vitality (Jeno et al., 2021; Johansen et al., 2023), and
higher grades (Wang et al., 2022). Furthermore, autonomous motivation has been negatively

MOTIVATION, ACHIEVEMENT AND DROPOUT
correlated with dropout intentions (Jeno et al., 2018) and performance anxiety (Vansteenkiste
et al., 2005).
Alongside these two forms of motivation is amotivation, which is characterized by a lack
of any type of motivation. Students that are amotivated act with no intention, value, or
control, and feel incompetent (Ryan & Deci, 2017). Research has shown that amotivation is
negatively related to student’s grades and knowledge transfer (Wang et al., 2022), well-being
(Breva & Galindo, 2020), and behavioral intentionality (Guay et al., 2000). In a prospective
study among Canadian college students, Vallerand and Bissonnette (1992) found that students
who had dropped out had significantly higher amotivation scores compared to persistent
students.
Longitudinal effects of basic psychological needs and motivation
Motivation is sensitive to the learning context (Janke et al., 2022; Roeser et al., 2009) and
can fluctuate within a semester, as well as between semesters (e.g., Busse & Walter, 2013;
Larose et al., 2006; Robinson et al., 2019). These fluctuations may impact learning and
dropout because these changes in motivation impact the energy for action, the value and
quality of the behavior, and the experiential quality of the actions (Brown & Ryan, 2007).
Several studies informed by SDT have investigated the longitudinal effects of motivation
in higher education. Some studies have looked at the longitudinal role of autonomous
motivation on achievement. For instance, Alamer and Alrabai (2022) investigated how
autonomous motivation for learning English as a foreign language predicted achievement
across three time points. Using cross-lagged panel modeling, they found that, at the within-
level, directional effects at Time 1 and Time 3, autonomous motivation predicted achievement
at Time 1, but not the other way around, and achievement at Time 2 predicted autonomous
motivation at Time 3, but not the other way around. Comparing two identical cohorts across
four time points, Corpus et al. (2022) found that autonomous regulation remained relatively
high across the measurement times, whereas controlled regulation and amotivation remained
relatively moderate to low. However, the regulations did not affect students’ grades. Garn and
Morin (2021) measured kinesiology students’ motivational regulation every two weeks for
seven weeks and found a quadratic model exhibiting high autonomous motivation in the
beginning, followed by a decline, before increasing again. Students with higher grade point
averages (GPAs) reported higher levels of motivation regulation at the start of the semester.
Other studies have investigated the trajectories of motivational regulation and its effect on
achievement and dropout. Specifically, Messerer et al. (2023) found that, among German
freshman students, intrinsic motivation at Time 1, predicted intrinsic motivation for learning

MOTIVATION, ACHIEVEMENT AND DROPOUT
at Time 2, and in turn dropout intention at Time 3, but not actual dropout. Corpus et al. (2020)
assessed students’ motivational trajectories on achievement and retention over four time
periods within the first year of college and found that increases in amotivation predicted lower
grades and a lower likelihood of remaining enrolled, whereas identified regulation and
intrinsic motivation was associated with higher grades and the likelihood of remaining
enrolled.
We also found several studies that investigated basic need satisfaction using a longitudinal
design. Sheldon and Krieger (2007) investigated the effect of different law schools on law
students’ need-satisfaction for autonomous motivation, self-reported achievement and well-
being. Results showed that supporting basic psychological needs positively predicted need
satisfaction. Changes in autonomous motivation was explained by changes in autonomy,
whereas achievement was uniquely explained by changes in competence. Pan and Gauvain
(2012) collected data on Chinese freshman students’ autonomous motivation, parental
autonomy support, and peer relatedness in a three-wave longitudinal study. Results showed
that although autonomous motivation declined, both parental autonomy support and peer
relatedness positively predicted autonomous motivation. In a recent study across the first four
semesters for German undergraduate students, Janke (2022) found that increases in needs
frustration were associated with decreases in students’ learning goal orientation. Furthermore,
basic need satisfaction in the initial semester predicted both an increase in learning goal
orientation over time (direct effect) and amplified the adaptability of this goal orientation
(moderation effect). Finally, using latent transition analyses, Gillet et al. (2020) found support
for five stable need-satisfaction profiles. Results showed that a profile characterized by high
relatedness and high global basic need satisfaction, and average competence displayed the
highest levels of students’ interest in their studies, educational satisfaction, highest class
attendance, and lowest dropout intentions.
In summary, there have been a number of longitudinal studies assessing the general SDT
model on the effects of higher education students’ achievement and/or dropout. However, in
evaluating past research, we identified several research gaps. First, few studies have used
objective registry data (data collected at the time of admission or during/after the semester
from the registrar’s office) on both achievement and dropout. This may be a cause for concern
given that registry data is less vulnerable to subjective evaluation than self-reported data
(Angel & Gronfein, 1988). Second, in teacher education, few studies have investigated the
longitudinal effects of basic psychological needs and motivational qualities on grades and
dropout. We sought to fill these gaps in a Norwegian educational context. The Norwegian

MOTIVATION, ACHIEVEMENT AND DROPOUT
educational system has undergone several reforms and changes in recent decades but still
faces challenges in recruitment and retainment of educators (Caspersen & Smeby, 2023),
making it a useful context for testing.
The present study
In response to the lack of longitudinal studies on student teachers in the Norwegian higher
education system and the use of registry data, we conducted a pre-registered longitudinal
study investigating the impact of students’ basic psychological needs and motivation qualities
on registry data for dropout and achievement. The benefit of pre-registration is that it
increases credibility and reproducibility (Nosek et al., 2019) by allowing us to specify our
hypotheses and methodological and analytical plan before data collection. We hypothesized
that autonomous motivation will be a predictor of higher grades and lower dropout (H1a),
whereas controlled motivation and amotivation will be a predictor of lower grades and higher
dropout (H1b). Given that some studies suggest that gender predicts dropout (e.g., Andresen
& Lervåg, 2022), we hypothesized that males will have higher dropout rates than females,
whereas females will have higher grades (H1c). We investigated the effect of the year of
study on students’ motivation. Given that dropout is greater in some years compared to others
(e.g., Bakken, 2022), it is important to investigate whether different motivational qualities can
explain this difference. Thus, we hypothesized, from a theoretical point of view, that students
in their fourth year will have higher autonomous motivation and lower controlled motivation
compared to first-year students (H2) given their greater degree of maturity and self-
understanding (Vansteenkiste et al., 2018). Furthermore, we also investigated the individual
variability in the effect of motivation on dropout and achievement. Specifically, we
hypothesized that the effect of motivational qualities on grades and dropout will vary across
individuals (H3). Individuals have different starting points for their motivation, and the
trajectory may also vary over time. Accounting for this within-person variation is important
for understanding differences in grades and dropout. Finally, we investigate the structural
relationship between our motivational variables (i.e., basic psychological needs and
motivational qualities) on achievement and dropout. We hypothesized that basic need
satisfaction will positively correlate with autonomous motivation, and negatively with
controlled motivation and amotivation (H4), and that basic need satisfaction will mediate the
effects of motivation on grades and dropout across time (H5).

MOTIVATION, ACHIEVEMENT AND DROPOUT
Methods
Participants and procedure
Student teachers on a five-year master program at a large university in Norway participated
in the present study. We collected data from first-year, second-year, and fourth-year students,
where the dropout rates are at the highest levels. The total sample size comprised 272
students. The baseline mean age was 21.01 (min=19, max=39). In our study sample, 166
(61.02%) were females, whereas 97 (35.6%) were males. Our sample consisted of students
from first year (n=110), second year (n=65), and fourth year (n=75). Included in our sample
are also students (n=13) that are from other study years, either third or fifth year. These were
coded as N/As for the study year factor given that our primary interest was between first-year,
second-year, and fourth-year students.
Data collection was performed at three time points. First, in the middle of the spring
semester of 2023 (T1), then at the end of the spring semester of 2023 (T2), and finally at the
beginning of the autumn semester of 2023 (T3). At these three time points, the same questions
were asked concerning the students’ basic psychological needs and motivation. At the end of
the data collection period, we collected registry data from all students who had consented to
participate in the study. Registry data was collected through the university and included
grades and study points accumulation from the spring semester of 2023, and semester
registration for autumn 2023.
Aims, sampling plans, hypotheses, and analytical plans are provided in our pre-
registration, freely available at the Open Science Framework (OSF)
(https://doi.org/10.17605/OSF.IO/3AK94). Data were collected through SurveyXact
(Ramboll, 2024). One deviation from our pre-registration is that, as opposed to the pre-
registration where we stated that students not consenting to participate at Time 1 would be
removed, we asked students that did not respond at Time 1 to participate at either Time 2 or 3.
We did this to increase the total sample size and increase the measurement power at each time
point.
Ethical concerns were addressed by registering our project with the institutional ethical
review board (System for Risk and Compliance), in line with local guidelines. We collected
registry data only from students who consented to give us access to collect registry data. At
each time point, three reminders were sent out. Students who responded at two or more time
points were given a gift card (150 NOK= ~14 USD) for their participation.

MOTIVATION, ACHIEVEMENT AND DROPOUT
Measures
Basic psychological needs
To measure students’ basic psychological needs, we employed an adapted three-item
(Martela & Ryan, 2023) measure of autonomy (“I am able to do the things that I really want
and value in teacher education”), competence (“I can do things well and achieve my goals in
teacher education”), and relatedness (“I feel close and connected with other people who are
important to me in teacher education”). The students responded on a seven-point Likert scale
ranging from 1 (completely disagree) through 4 (neither / nor) to 7 (completely agree). We
created a composite variable of these three items to form a variable of psychological needs.
Omega reliability was satisfactory for each time point (Table 2). All the items are available at
the OSF link.
Motivation
We used the Self-Determination Index (SDI) (Levesque-Bristol, 2021) to measure three
qualities of motivation. The SDI consists of 18 items measuring the different regulations
conceptualized within SDT, which were outlined above. The scale can also be used to create
classes of motivation depending on their level of self-determination, amotivation, controlled
motivation, and autonomous motivation, respectively. In the present study, the students were
asked at each time point why they were studying to become a teacher. An example of an item
for amotivation is “I don’t know. I wonder if I should continue”. An example of an item for
controlled motivation (external regulation) is “because I feel I have to”. An example of an
item for autonomous motivation (identified regulation) is “because it allows me to develop
skills that are important to me”. We created composite variables of amotivation, controlled
motivation (external regulation and introjected regulation), and autonomous motivation
(identified regulation, integrated regulation, and intrinsic motivation). The students responded
on a seven-point Likert scale ranging from 1 (completely disagree) through 4 (neither / nor) to
7 (completely agree). Omega reliability was satisfactory for each time point for each
motivational class (Table 2). All the items are available through the OSF link.
Achievement
The registry data of student achievement collected was in the form of grades achieved
during the spring semester. If the students had more than one exam, an average was created.
In Norway, grades range from A (the highest) to F (fail). In this study, only passing grades
were collected from the registry data, meaning grades A to E. The grades were transformed
into numerical values ranging from 5 (the highest) to 1 (the lowest).

MOTIVATION, ACHIEVEMENT AND DROPOUT
Dropout
Two registry data measurements were collected as proxies for dropout. First, we collected
how many study credits (ECTS) the students completed during the spring semester. The
normal amount for progression is 30 ECTS, and anything less is considered a failure to
complete that semester. We dichotomized this measure with £ 29 being considered a failure,
and ³ 30 being considered a pass. Finally, we collected a dichotomous measure of student
registration in the following semester. If the students were not registered for the autumn
semester, this was an indication that the students had dropped out of the current (autumn)
semester.
Control variables
The following control variables were collected from the registry office, which were
recorded when the students were accepted on to the program: gender, age, and high-school
GPA. Students reported their gender and age at the start of the teacher education program.
Their high-school GPA was registered when the students applied for the university. The
scores range from 1 (lowest grade) to 6 (highest grade).
Analytical strategy
All preliminary and primary analyses were performed using the open and freely available
statistical software R, version 4.3.2 (R Core Team, 2018). All R-scripts reported in our study
are openly available on GitHub (link). To analyze our primary hypotheses, we employed
linear mixed-effect modeling using the “nlme” package (Pinheiro et al., 2023), generalized
linear models (R Core Team, 2018) and the “lavaan” package (Rosseel, 2012) for path
analysis. To evaluate global fit measures for the path-model, we used conventional cut-off
criteria (Hu & Bentler, 1999), with satisfactory values above .90 for comparative fit index
(CFI) and Tucker-Lewis index (TLI), and values below .08 for root mean square error of
approximation (RMSEA) and standardized mean square residual (SRMR). Furthermore, a
non-significant chi-square (c2) is recommended (Schermelleh-Engel et al., 2003), however,
chi-square is sensitive to samples and sample size (Kline, 2011), and as such caution is
advised for these goodness-of-fit indices.
Results
Missing data
Of the total 427 students, 272 students (63.7%) responded at any of the three time points.
Of the total sample size of 272 respondents, the number of complete cases across all three
time points is 83, whereas 188 cases are incomplete. We present the overview of the missing

MOTIVATION, ACHIEVEMENT AND DROPOUT
data for each motivational variable at each time point in Table 1. Given that our rate of
missing data is greater than 5%, we did not proceed to impute missing values. All missing
data were handled through full information maximum likelihood (FIML).
Table 1. Overview of missing data
Variable
Time 1
Time 2
Time 3
Compl.
Incompl.
Compl.
Incompl.
Compl.
Incompl.
Basic need satisfaction
142 (52.2)
130 (47.8)
153 (56.2)
119 (43.8)
209 (76.8)
63 (23.2)
Amotivation
138 (50.7)
134 (49.3)
150 (55.1)
122 (44.9)
208 (76.5)
64 (23.5)
Controlled motivation
138 (50.7)
134 (49.3)
150 (55.1)
122 (44.9)
208 (76.5)
64 (23.5)
Autonomous motivation
138 (50.7)
134 (49.3)
150 (55.1)
122 (44.9)
208 (76.5)
64 (23.5)
Note: Compl=Complete, Incompl=Incomplete. Percentages are shown in parenthesis.
Preliminary analysis
We provide basic descriptive statistics for our main motivational variables for each time
point in Table 2. All variables show signs of normality with acceptable values of skewness (<
3) and kurtosis (< 3). Furthermore, all variables have satisfactory reliability measures
(Omega) with values > .70. Finally, the mean scores suggest that basic need satisfaction and
autonomous motivation decrease slightly at the end of the spring semester (between Time 1
and Time 2), and then increase again at the start of the autumn semester (Time 3). However,
for amotivation and controlled motivation, we see the opposite with mean scores increasing at
the end of the spring semester and then decreasing at the start of the autumn semester.
Descriptive statistics for both dropout indicators showed that 14 participants (5.3%) in our
sample were not registered, whereas 35 participants (14.1%) had less than 30 ECTS from the
spring semester. Across study years, the distribution was as follows for non-registered
students: in first year n=8, second year n=4, and fourth year n=0. Two (n=2) non-registered
students came from an unspecified study year. For study points, the distribution was as
follows for those with less than 30 ECTS: in first year n=19, second year n=12, and fourth
year n=4.
Table 2. Descriptive statistics of the main motivational variables
n
M
SD
Min
Max
Range
Skew.
Kurt.
se
w
1. Basic needs_T1
142
4.95
1.07
2.00
7.00
5.00
-.39
-.24
.09
.73

MOTIVATION, ACHIEVEMENT AND DROPOUT
2. Amotivation_T1
138
2.26
1.36
1.00
6.00
5.00
.96
-.20
.12
.83
3. Controlled motivation_T1
138
2.40
1.12
1.00
5.67
4.67
.83
-.07
.09
.89
4. Autonomous motivation_T1
138
4.97
1.07
1.78
7.00
5.22
-.57
.15
.09
.92
5. Basic needs_T2
153
4.75
1.18
1.00
7.00
6.00
-.75
.33
.10
.77
6. Amotivation_T2
150
2.49
1.53
1.00
6.00
5.00
.84
-.48
.12
.84
7. Controlled motivation_T2
150
2.71
1.19
1.00
6.50
5.50
.60
-.45
.10
.91
8. Autonomous motivation_T2
150
4.82
1.08
1.67
7.00
5.33
-.42
-.18
.09
.92
9. Basic needs_T3
209
4.91
1.12
1.00
7.00
6.00
-.86
1.19
.08
.77
10. Amotivation_T3
208
2.36
1.51
1.00
7.00
6.00
1.03
.04
.10
.88
11. Controlled motivation_T3
208
2.57
1.26
1.00
5.67
4.67
.60
-.68
.09
.92
12. Autonomous motivation_T3
208
4.86
1.08
1.67
7.00
5.33
-.58
-.02
.07
.93
Primary analysis
Grades
To test Hypotheses 1a, 1b,1c, and 3, where we investigate the role of autonomous
motivation (H1a), controlled motivation (H1b), gender differences (H1c), and individual
variability (H3) on grades, we conducted a range of mixed-effects model building from a
simple model to a more complex model (Ryoo, 2011). However, after testing a null model
(intercept only), we found that the intraclass correlation (ICC) was 0.97, indicating that 97%
of the variance in grades could be attributed to a student (Hausknecht et al., 2008). This
suggests that only 3% of the variance is left unexplained after taking the between-level
variables into account. We thus conducted a multiple regression analysis with cluster robust
standard errors to account for some of the within-level effects (using students as a grouping
factor). We dummy-coded gender, with males as the reference variable (female=0, males=1),
and mean-centered amotivation, controlled motivation, and autonomous motivation for ease
of interpretation (Raudenbush & Bryk, 2002). We also controlled for mean-centered students’
high-school grades. Model comparison between the traditional and cluster robust standard
error showed almost identical results, and thus we retained the multiple regression analysis
without cluster robust standard errors so that we could obtain R2 values. Results showed that
the model as a whole explained 21% of the variance in grades, R2=0.21, F(5, 452)= 23.41, p<
.001, adj R2= 0.20. The model’s intercept when amotivation, controlled motivation,
autonomous motivation, and high school grades are at mean, and for females, is 3.42 (95% CI
[3.32, 3.51], t(452)= 71.7, p< .001. The results within this model showed that both

MOTIVATION, ACHIEVEMENT AND DROPOUT
amotivation (b= -0.17, 95% [-0.27, -0.07], t(452)= -3.30, p< .001) and autonomous
motivation (b= -0.18, 95% [-0.27, -0.08], t(452)= -3.58, p< .001) were negatively related to
students’ grades. The effect of controlled motivation was not significant (b= 0.02, 95% [-0.46,
0.08], t(452)= 0.65, p= .513). The effect of both gender (males) (b= 0.23, 95% [0.05, 0.41],
t(452)= 2.54, p= .012) and high school grades (b= 0.43, 95% [0.34, 0.51], t(452)= 9.87, p<
.001) were positive predictors of students’ grades.
Dropout
To test Hypotheses 1a, 1b, 1c, and 3, where we investigate the role of autonomous
motivation (H1a), controlled motivation (H1b), gender differences (H1c), and individual
variability (H3) on dropout, we conducted logistic regression analyses. We tested two
identical models (i.e., amotivation, controlled motivation, autonomous motivation, and
gender) in predicting students’ study point accumulation (³ 30 ECTS which indicates passing
that semester, and £ 29 ECTS which indicates failure) and registration (either registered, or
not registered). In both models, our null model (intercept only model) suggested high ICC
(>95%). Thus, we continued with standard logistic regression analysis without considering
within-person variation. See Table 4 for results of the logistic regressions.
To test if our model predicted study point accumulation, we fitted logistic regression using
maximum likelihood. The explanatory power (pseudo R-square) of the model was low,
McFadden= .01, Cox & Snell= .01, Nagelkerke= .02. None of our predictors had significant
coefficients.
To test if our model predicted semester registration, we fitted logistic regression using
maximum likelihood. The explanatory power (pseudo R-squared) of the model was
substantial, McFadden= .25, Cox & Snell= .10, Nagelkerke= .29. Within this model, the
results showed that amotivation was a significant and negative predictor of semester
registration, whereas controlled motivation and gender (males) were positive and significant
predictors. Autonomous motivation was a non-significant predictor. Furthermore, the results
show that if the students’ amotivation increased by one unit, the odds ratio for being
registered decreased by .41. In contrast, if the students’ controlled motivation increased by
one unit, the odds ratio for being registered increased by 1.51. Finally, the odds ratio for being
registered if you were a male compared to a female was 4.36 times higher.

MOTIVATION, ACHIEVEMENT AND DROPOUT
Table 4. Logistic regression models of dropout
B
SE
Wald’s
Z
P
Deviance
(DF)
Odds
ratio
CI for odds
ratio
Study points
329.29
(462)
Amotivation
-0.14
0.11
-1.25
.20
0.86
0.69, 1.08
Controlled
motivation
-0.07
0.12
-0.62
.53
0.92
0.73, 1.18
Autonomous
motivation
-0.17
0.15
-1.09
.27
0.83
0.60, 1.14
Gender (males)
-0.46
0.29
-1.56
.11
0.62
0.35, 1.13
Semester
registration
150.76
(480)
Amotivation
-0.88
0.16
-5.50
.000***
0.41
0.29, 0.55
Controlled
motivation
0.41
0.20
2.06
.03*
1.51
1.03, 2.29
Autonomous
motivation
0.09
0.22
0.41
.68
1.09
0.70, 1.69
Gender (males)
1.47
0.59
2.48
.01*
4.36
1.50, 16.20
Autonomous motivation
To test Hypothesis 2 (whether students in their fourth year will have higher autonomous
motivation and lower controlled motivation compared to first-year students), we fitted a
mixed-effects model using autonomous motivation as a dependent variable and study year and
time points as predictors (fixed effects). The results are presented in Table 5. We started with
the simplest model (null model) and then increased the complexity. However, the best-fitting
model was the random intercept model in which we included students as random effects. We
added an interaction term between study year and time points, however, this model was not
significantly better than the model without interaction effect, p= .17. We also tested if we
could add study year as a random intercept, however, this model was not significantly better
than the previous model, p= .12. We thus retained our random intercept model.
The results showed that the intercept for the random intercept model is 5.227 across
students at study year (first year) and Time-point (Time 1). The fixed effect of Time-point
(both Time 2 and Time 3) was non-significant, p> .05. The fixed effect of study year (second
year) is non-significant, whereas the effect of the fourth year (study year) is significant (
b
= -
.64, 95% CI [-0.90, -0.38], t= -4.78(239), p < .001. The variance around students’ intercept is
0.84, whereas the variance around students’ slope is 0.29. The model’s explanatory power is
substantial with a conditional R2 of 73%. See Figure 1 for a visualization of the effect of study
year on students’ autonomous motivation.

MOTIVATION, ACHIEVEMENT AND DROPOUT
Table 5. Mixed effects model of students’ autonomous motivation
Null model
Random
intercept model
Random
intercept/interaction
model
Random intercept
and random slope
model
(Intercept)
4.902***
(0.061)
5.227***
(0.106)
5.247***
(0.128)
5.226***
(0.101)
Study year (second
year)
-0.076
(0.153)
-0.028
(0.195)
-0.073
(0.141)
Study year (fourth
year)
0.683***
(0.143)
0.754***
(0.181)
0.680***
(0.151)
Time-point (Time 2)
-0.093
(0.073)
-0.075
(0.127)
-0.089
(0.073)
Time-point (Time 3)
-0.113
(0.071)
-0.155
(0.119)
-0.112
(0.071)
Study year (second
year) x Time-point
(Time 2)
0.032
(0.189)
Study year (fourth
year) x Time-point
(Time 2)
-0.045
(0.168)
Study year (second
year) x Time-point
(Time 3)
-0.167
(0.184)
Study year (fourth
year) x Time-point
(Time 3)
0.202
(0.163)
SD (Intercept
Subject)
0.896
0.836
0.840
0.767
SD (Study year:
second year Subject)
0.996
SD (Study year:
fourth year Subject)
1.089
Cor (intercept~
Study year: second
year Subject)
-0.674
Cor (intercept~
Study year: fourth
year Subject)
-0.493
-0.276
SD (observations)
0.546
0.547
0.538
0.549
R2 marginal
0.000
0.092
0.099
0.090
R2 conditional
0.729
0.727
0.738
0.730
AIC
1271.7
1182.8
1190.9
1188.3
ICC
0.7
0.7
0.7
0.7
Note: * p <0.05, ** p < 0.01, *** p < 0.001. Standard errors within parentheses.
Highlighted in bold is the final model we retained. Subject is each individual student.

MOTIVATION, ACHIEVEMENT AND DROPOUT
Figure 1. Autonomous motivation across semesters as a function of study year
Note: NAs = 13 participants who were not sampled from either first, second, or fourth year and were
thus removed.
To test Hypothesis 4, regarding whether psychological need satisfaction correlates
positively with autonomous motivation, and negatively with controlled motivation and
amotivation, we conducted a correlational analysis using individual students as a grouping
factor. Results are presented in Table 6. In line with our predictions, we found that both
between-person and within-person correlations are in the expected directions. That is, basic
need satisfaction correlated negatively with amotivation, and positively with autonomous
motivation. Additionally, to investigate whether these results were consistent within time
points, we tested time points as a grouping factor. The results of the correlational direction
were in line with our hypothesis. Controlled motivation was unrelated to basic need
satisfaction at the between-person level but negatively related at the within-person and within-
time level.
Table 6. Correlational matrix of the main motivational variables
Variable
Mean (SD)
ICC
1
2
3
4
1. Basic need
satisfaction
4.87 (1.13)
.58
-
-.19***
(-.40***)
-.13**
(-.15**)
.22***
(.59***)

MOTIVATION, ACHIEVEMENT AND DROPOUT
2. Amotivation
2.37 (1.48)
.74
-.43***
-
.18***
(.22***)
-.33***
(.-.50***)
3. Controlled
motivation
2.56 (1.20)
.57
-.10
.25***
-
.01
(-.13**)
4. Autonomous
motivation
4.88 (1.08)
.74
.59***
-.51***
-.10
-
Note: Means and standard deviations (SD) are at the between-person level. ICC=Intraclass correlations.
Between-person correlations are below the diagonal, whereas within-person correlations are above the diagonal.
Within-time correlations are presented in parenthesis.
** indicates p < .01, *** indicates p < .001.
To test our final hypothesis, H5, examining whether basic need satisfaction mediates the
effects of motivation on grades and dropout across time, we employed a path-analytical
model. We specified three models that consisted of amotivation, controlled motivation, and
autonomous motivation as exogenous predictors and basic need satisfaction as mediator. This
model was tested three times using grades, semester registration, and study points as
endogenous outcomes, respectively. In all three models, we removed controlled motivation
due to no significant relation to all or parts of the measurement points on the latent variable
(i.e., controlled motivation). For the dropout models (i.e., registration and study points), we
used the diagonally weighted least square (WLSMV) estimator to account for the
dichotomous endogenous outcome variables, which is the least biased and most accurate
estimator for dichotomous outcomes (Li, 2016). Goodness-of-fit indices are presented in
Table 7. Results for grades showed that only autonomous motivation was related to basic need
satisfaction (
b
= .98, p <.001), while amotivation was unrelated (
b
= .03, p =.77). However,
the relation between basic need satisfaction and grades was non-significant (
b
= -.12, p=.08).
The indirect effect between autonomous motivation and grades, through basic need
satisfaction, was also non-significant (
b
= -.12, p =.09). Finally, the model accounted for 1.6%
of the variance in grades (R2= .016), and 93% of the variance in basic need satisfaction (R2=
.933).
For semester registration, only autonomous motivation was related to basic need
satisfaction (
b
= .89, p <.001), whereas amotivation remained non-significant (
b
= -.05, p
=.69). Basic need satisfaction was a significant predictor of semester registration (
b
= .38, p
<.05). We also found that the effect between autonomous motivation on semester registration
was fully mediated by basic need satisfaction (
b
= .34, p <.01). The model accounted for

MOTIVATION, ACHIEVEMENT AND DROPOUT
14.9% of the variance in semester registration (R2= .149), and 85% of the variance in basic
need satisfaction (R2= .855).
For study points, autonomous motivation was a significant predictor of basic need
satisfaction (b= .92, p <.001), whereas amotivation was not significant (
b
= -.008, p =.95).
Basic need satisfaction was a non-significant predictor of study points (
b
= .01, p =.95). No
significant indirect effects were found (p´s >.05). The model explained 86% of the variance in
basic need satisfaction (R2=. 865) and less than 1% for study points (R2= .000).
Table 7. Path analytical models
Model
c2
df
CFI
TLI
RMSEA [CI]
SRMR
Grades
67.56***
26
.94
.91
.077 [.05, .10]
.046
Semester registration
38.22
26
.94
.90
.076 [.00, .12]
.059
Study points
33.59
26
.96
.93
.06 [.00, .13]
.05
Note: *** indicate p < .001
Discussion
The main aim of this pre-registered longitudinal study was to gain a better understanding
of the impact of student teachers’ basic psychological needs and motivational classes on
students’ grades and dropout rate during study periods which are known to have high attrition
rates. Furthermore, we were interested in variability across individuals, gender, and study year
on grades and dropout. Interestingly, the results of our present study found some support for
our hypotheses, whilst other results contradicted our expectations. We discuss all below.
The effect of motivational classes and basic need satisfaction on grades and dropout
Our results showed some mixed results regarding students’ grades. In contrast to our
hypotheses, autonomous motivation was negatively related to grades. Amotivation was also
negatively related to grades, although this was more in line with expectations, as well as
previous findings (Wang et al., 2022). The strongest predictor for grades was students’ high
school grades, which is consistent with previous research (Hailikari et al., 2008). Controlled
motivation remained a non-significant predictor of students’ grades. This is not a surprising
finding, which is what might be expected based on the order of regulations specified within
SDT (Ryan & Deci, 2017). The results from our path analysis also found similar results, that
is, we found no significant effect of basic need satisfaction on grades, nor the indirect effect
of autonomous motivation on grades, through basic need satisfaction, as we initially

MOTIVATION, ACHIEVEMENT AND DROPOUT
hypothesized. However, two results deserve some comment. First, despite our expectations
we did not find a significant effect of basic psychological needs satisfaction on grades, and
also counter our expectations we found a negative relation between autonomous motivation
on grades. While this was unexpected, it is inconsistent with some previous research (Filak &
Sheldon, 2008; Guiffrida et al., 2013; Wang et al., 2022). These unexpected findings might be
due to changes in aspiration to become a teacher (e.g., Roness & Smith, 2010) and how the
students are assessed (e.g., Kusurkar et al., 2023). For instance, some students with good
grades might leave teacher education to pursue other careers, these students might have the
combination of high grades and a lack of autonomous motivation for becoming a teacher.
Furthermore, the assessment of student teachers could be tailored towards rote learning or
external control such as imposed deadlines, school exams, high stakes, etc., which might
reduce students’ overall autonomy and competence, and increase controlled motivation.
Further, as the measurement of basic need satisfaction and the collection of student grades
occurred in a short time frame, it could have been a contributing factor to our null finding. A
longitudinal design over a longer time period could have yielded different results, as found in
other cases (Wang et al., 2019).
Our study echoes previous studies showing the detrimental effect of amotivation on student
performance (e.g., Breva & Galindo, 2020; Guay et al., 2000; Leroy & Bressoux, 2016) and
procrastination (Tahiri & Mouratidis, 2024). Lack of feelings of efficacy and intentionality,
which are characteristics of amotivation, might impact students’ study strategies, class
attendance, and help-seeking behaviors (Ryan & Deci, 2017), which might explain the
negative effect of amotivation on grades in our study.
For dropout, our regression models were only able to predict dropout when measured as
semester registration, and not as study points. As expected, amotivation was negatively
related to being registered, which is in line with previous literature (e.g., Vallerand &
Bissonnette, 1992; Vallerand et al., 1997). However, surprisingly, controlled motivation was
positively related to being registered. An explanation for this finding could be how
Norwegian teacher education is structured, which requires students to complete their degree
as they will be left with no credit if they leave early. This functionally forces the students to
complete their teacher education or to drop out, leaving them without any degree that could be
applied to other study areas (Skagen & Elstad, 2023). This might also have financial
implications on the students, as they would face a reduction in their student loans. Such
“controlled” reasons for persisting may come at the cost of their psychological well-being
(e.g., Sallai et al., 2023).

MOTIVATION, ACHIEVEMENT AND DROPOUT
Thus, when considering how the study program is set up, controlled motivation seems to
be a necessary motivational force for the students to complete their program, yet it might have
unintended negative consequences for mental health. The results from our path analyses found
that basic need satisfaction was positively related to being registered for the semester.
Furthermore, autonomous motivation was also indirectly related to semester registration,
mediated by basic psychological needs. This finding suggests that students who are
autonomously motivated are more likely to be registered, but only to the extent that their basic
psychological needs are satisfied. The mediating effect of basic need satisfaction has also
been found in other studies (e.g., Hope et al., 2019; Niemiec et al., 2009), which is expected
given that the satisfaction of basic psychological needs is crucial given that it is considered an
essential nutrient for humans (Vansteenkiste et al., 2023).
Variability in individuals and groups on grades and dropout
In contrast to our hypothesis tests, we found no support for variability across individuals.
The models we tested showed low intra-individual variability, indicated by very high ICC-
levels. One explanation could be that the individuals in our sample are more similar than
dissimilar across time. It could be interpreted that individuals in our sample, across different
years of study, experience similar situations which are captured by our measures.
In contrast to our hypothesis, males had higher grades relative to females, and were more
likely to be registered for the following semester compared to females. This was surprising
given that previous research consistently finds that females outperform males, and that males
are more likely to dropout (e.g., Bakken, 2022; Böhn & Deutscher, 2022). One explanation
for this finding could be gender differences, where males have higher assertiveness and lower
anxiety than females (Costa Jr et al., 2001; Feingold, 1994), which may be exacerbated in
high-pressure environments. Yet another explanation could be that, given our sample had a
higher percentage of females, the finding that fewer women were semester registered could
simply reflect the skewed gender distribution. Interestingly, there are large gender differences
in admission into teacher education programs as well, meaning that students who drop out are
more likely to be females. Our results could therefore be explained by the general dropout
rates of the program. Future studies would need to replicate our results regarding gender
differences or investigate the factors that contribute to these gender differences.
The effect of study year and basic need satisfaction on motivational classes
The results of our study show, in contrast to our hypothesis, that students in the fourth year
had lower autonomous motivation across time. This was surprising given that our initial

MOTIVATION, ACHIEVEMENT AND DROPOUT
reasoning was that fourth-year students, compared to first-year students, would have
internalized the value of studying and becoming a teacher. However, a plausible explanation
could be that as students progressed, their workload and expectations increased, leading to
less value and enjoyment with regard to their study motivation. In particular for this sample,
the fourth year of this specific teacher education program has the highest workload of the
program, since both of the two practicum periods take place this year on top of normal study
progression of 60 ECTS. This claim is consistent with a recent study that shows that fourth-
year students in teacher education have decreased autonomous motivation following periods
of intense practicum and high workloads (Jeno et al., in prep). Furthermore, given the
dynamics of teacher education in Norway and general income situation for students in higher
education (e.g., Wikan & Bugge, 2014), the decrease in students’ autonomous motivation
could be a function of less intention to become a teacher in general, and the perception of
being under a high level of pressures. Although our initial reasoning came from a
developmental perspective (i.e., maturation increases internalization), this inconsistency in the
literature requires more research to understand the development of autonomous motivation
through students’ educational pathway.
Our correlational analyses supported our hypothesis that basic needs correlate positively
with autonomous motivation, and negatively with amotivation and controlled motivation. This
general pattern was found at the between-person and within-person level, as well as the
within-time level. These results indicate that persons who experience basic psychological
need satisfaction at both the individual level and within a specific measurement time are more
likely to have higher autonomous motivation. This is consistent with the assumptions of SDT
that suggest that individual experience of need satisfaction is necessary for developing more
autonomous forms of motivation (Ryan & Deci, 2017).
Limitations
There are several limitations worth considering when interpreting our results. First,
measuring the dropout rate through study point accumulation and semester registration the
following semester could be challenged. In Norway, dropout is measured five years after
students are supposed to have graduated (Andresen & Lervåg, 2022; Hovdhaugen & Aamodt,
2005). This is useful for accounting for sickness, parental leave, and breaks in studying. Yet,
the issue of dropout is complex given that student dropout can be understood from different
levels. For instance, program dropout, institutional dropout, and sector dropout (e.g., Tinto,
1993), have different consequences for the university program, higher education institution,
and individual students, respectively. One strength of this study is the measurement of

MOTIVATION, ACHIEVEMENT AND DROPOUT
program dropout. Although we cannot infer that our results imply program dropout, per se,
our study design speaks to the importance of assessing dropout within a profession struggling
with recruitment. In contrast to dropout intentions, registry data provides objective data on
students registered for a specific semester. Furthermore, our dropout rates for semester
registration (5.3%) and study point production (14.1%) are not too dissimilar compared to
national numbers which are 25% and 16% for bachelor and master students, respectively, and
20% for younger students (Andresen & Lervåg, 2022). Nevertheless, future studies could
complement our design with an assessment five years after the predicted graduation date to
investigate if the percentage of dropouts across a broader span of time.
Relatedly, our longitudinal design only covered a short period of time, meaning we only
investigated motivational trajectories during the transition from one semester to another, in
which attrition is known to be high. Future studies could examine the longitudinal trajectories
over the entire teacher program to investigate how motivation varies within a semester and
across the entire program. This could be useful to see when and why motivation decreases or
increases as a function of program year or situations within a semester. Another interesting
approach could be to investigate students’ daily motivation across a semester or year to see
how daily motivational fluctuations impact achievement and dropout.
Finally, given the relatively small sample size in our study, we created composite
constructs of motivational qualities (i.e., amotivation, controlled motivation, and autonomous
motivation), which have recently been debated (e.g., Howard et al., 2017). Specifically, the
approach side of introjection (i.e., avoiding shame/guilt or attaining self-esteem), which falls
between controlled and autonomous regulation, may cause some analytical problems. This
could be the reason why controlled motivation in our study failed to predict grades and
dropout, and why our path analytical model failed to estimate a latent factor for controlled
motivation. Future studies are needed to test if using regulations (i.e., external, introjected,
identified regulation) is more predictive of grades and dropout as opposed to motivational
classes. Related to this, our small sample size and the number of students who provided
response at all three time points could be one explanation as to why some of our models failed
to converge.
Future directions and implications
Despite these limitations, there are some useful implications of our study.
Methodologically, our models seem to be less successful in predicting grades, and more
successful in predicting dropout, specifically semester registration. Future studies should
further investigate when basic need satisfaction and motivational classes are and are not

MOTIVATION, ACHIEVEMENT AND DROPOUT
predictive of grades, or the boundary conditions of grades in terms of basic need satisfaction
and motivational classes, i.e. it may be that competence is an important construct for
predicting grades, while relatedness is less important. Similarly, it may be that integrated
regulation – which is closely tied into identity or self-concept – is more important for grades
than intrinsic motivation. Future studies are needed to disentangle these effects with larger
sample sizes. Furthermore, understanding the effects of controlled motivation for persistence
and achievement, and the factors contributing to this seems important as a future research
avenue. Despite our findings showing that controlled motivation seems to shield against
dropout, the phenomenological experiences associated with this form of motivation may not
be positive, and even come at a negative cost for mental health. Thus, our interpretation of
this finding has some caveats attached to it.
Based on our results, we recommend that teacher education programs focus on supporting
students’ basic needs. This may be especially important when implementing initiatives to
reduce dropout rates (i.e., increase the level of semester registration). Further, implementing
initiatives to address students with high amotivation seems to be an important step towards
increasing grades and reducing dropouts. This could be accomplished through providing
competence-related support or helping students find value in their educational activities.
Finally, supporting students’ relatedness could also be a means to enhance internalization
(Leo et al., 2023) and learning (Escandell & Chu, 2021) to reduce dropout and enhance
achievement. Creating a structure in which students can experience a sense of belonging
could serve as a protective factor for reducing the likelihood of dropping out.
References
Alamer, A., & Alrabai, F. (2022). The causal relationship between learner motivation and
language achnievement: New dynamic perspective. Applied Linguistics, 1-22.
https://doi.org/10.1093/applin/amac035
Alatalo, T., Hansson, Å., & Johansson, S. (2024). Teachers’ academic achievement: evidence
from Swedish longitudinal register data. European Journal of Teacher Education,
47(1), 60-80. https://doi.org/10.1080/02619768.2021.1962281
Andresen, S. M. H., & Lervåg, M.-L. (2022). Frafall og bytter i universitets- og
høgskoleutdanning.
Angel, R., & Gronfein, W. (1988). The Use of Subjective Information in Statistical Models.
American Sociological Review, 53(3). https://doi.org/10.2307/2095653
Bakken, P. (2022). Evaluering av lektorutdanningene: Analyse av frafall og fullføring.
Botnaru, D., Orvis, J., Langdon, J. L., Niemiec, C. P., & Landge, S. M. (2021). Predicting
final grades in STEM courses: A path analysis of academic motivation and course-
related behavior using self-determination theory. Learning and Motivation, 74, 1-11.
https://doi.org/10.1016/j.lmot.2021.101723

MOTIVATION, ACHIEVEMENT AND DROPOUT
Breva, A., & Galindo, M. P. (2020). Types of motivation and eudemonic well-being as
predictors of academic outcomes in first-year students: A self-determination theory
approach. PsyCh Journal. https://doi.org/10.1002/pchj.361
Brown, K. W., & Ryan, R. M. (2007). Multilevel modeling of motivation. In A. D. Ong & M.
H. M. Van Dulmen (Eds.), Oxford Handbook of Methods in Positive Psychology (pp.
530-541). Oxford University Press.
Bureau, J. S., Howard, J. L., Chong, J. X. Y., & Guay, F. (2021). Pathways to student
motivation: A meta-analysis of antecedents of autonomous and controlled motivations.
Review of Educational Research. https://doi.org/10.3102/00346543211042426
Busse, V., & Walter, C. (2013). Foreign language learning motivation in higher education: A
longitudinal study of motivational changes and their causes. The Modern Language
Journal, 97(2), 435-456. https://doi.org/10.1111/j.1540-4781.2013.12004.x
Böhn, S., & Deutscher, V. (2022). Dropout from initial vocational training – A meta-synthesis
of reasons from the apprentice’s point of view. Educational Research Review, 35.
https://doi.org/10.1016/j.edurev.2021.100414
Caspersen, J., & Smeby, J.-C. (2023). Research-based teacher education in Norway – a
longitudinal perspective. International Journal of Educational Research, 119.
https://doi.org/10.1016/j.ijer.2023.102177
Chen, B., Vansteenkiste, M., Beyers, W., Boone, L., Deci, E. L., Kaap-Deeder, J. V. d.,
Duriez, B., Lens, W., Matos, L., Mouratidis, A., Ryan, R. M., Sheldon, K. M.,
Soenens, B., Petegem, S. V., & Verstuyf, J. (2015). Basic psychological need
satisfaction, need frustration, and need strenght across four cultures. Motivation and
Emotion, 39(2), 216-236. https://doi.org/10.1007/s11031-014-9450-1
Cook, D. A., & Artino Jr, A. R. (2016). Motivation to learn: an overview of contemporary
theories. Medical Education, 50, 997-1014. https://doi.org/10.1111/medu.13074
Corpus, J. H., Robinson, K. A., & Liu, Z. (2022). Comparing college students’ motivation
trajectories before and during covid-19: A self-determination theory approach.
Frontiers in Education, 7, 1-8. https://doi.org/10.3389/feduc.2022.848643
Corpus, J. H., Robinson, K. A., & Wormington, S. V. (2020). Trajectories of motivation and
their academic correlates over the first year of college. Contemporary Educational
Psychology, 63, 1-15. https://doi.org/10.1016/j.cedpsych.2020.101907
Costa Jr, P. T., Terracciano, A., & McCrae, R. R. (2001). Gender differences in personality
traits across cultures: Robust and surprising findings. Journal of Personality and
Social Psychology, 81(2), 322-331. https://doi.org/10.1037/0022-3514.81.2.322
Elstad, E., Christophersen, Abben, K.-A., & Turmo, A. (2023). Nordic student teachers’
evaluation of educational theory, subject didactics, practice training, time on task and
turnover intentions. In E. Elstad (Ed.), Teacher education in the nordic region:
Challenges and opportunities (pp. 287-319). https://doi.org/10.1007/978-3-031-
26051-3_12
Escandell, S., & Chu, T. L. A. (2021). Implementing relatedness-supportive teaching
strategies to promote learning in the college classroom. Teaching of Psychology, 0(0),
1-7. https://doi.org/10.1177/00986283211046873
Feingold, A. (1994). Gender differences in personality: A meta-analysis. Psychological
Bulletin, 116(3), 429-456.
Filak, V. F., & Sheldon, K. M. (2008). Teacher support, student motivation, student need
satisfaction, and college teacher course evaluations: testing a sequential path model.
Educational Psychology, 28(6), 711-724. https://doi.org/10.1080/01443410802337794
Garn, A. C., & Morin, A. J. S. (2021). University students’ use of motivational regulation
during one semester. Learning and Instruction, 74, 1-10.
https://doi.org/10.1016/j.learninstruc.2020.101436

MOTIVATION, ACHIEVEMENT AND DROPOUT
Gillet, N., Morin, A. J. S., Huyghebaert-Zouaghi, T., Alibran, E., Barrault, S., & Vanhove-
Meriaux, C. (2020). Students’ need satisfaction profiles: Similarity and change over
the course of a university semester. Applied Psychology, 69(4), 1396-1437.
https://doi.org/10.1111/apps.12227
Guay, F., Vallerand, R. J., & Blanchard, C. (2000). On the assessment of situational intrinsic
and extrinsic motivation: The Situational Motivation Scale (SIMS). Motivation and
Emotion, 24(3), 175-213.
Guiffrida, D. A., Lynch, M. F., Wall, A. F., & Abel, D. S. (2013). Do reasons for attending
college affect academic outcomes? A test of a motivational model from a Self-
Determination Theory perspective. Journal of College Student Development, 54(2),
121-139.
Hailikari, T., Katajavuori, N., & Lindblom-Ylanne, S. (2008). The relevance of prior
knowledge in learning and instructional design. American Journal of Pharmaceutical
Education, 72(5), 1-8.
Hausknecht, J. P., Hiller, N. J., & Vanc, R. J. (2008). Work-unit absenteeism: effects of
satisfaction, commitment, labor market conditions, and time. The Academy of
Management Journal, 51(6), 1223-1245.
Hernández, E. H., Lozano-Jiménez, J. E., Noguera, J. M. d. R., & Moreno-Murcia, J. A.
(2022). Relationships among instructor autonomy support, and university students’
learning approaches, perceived professional competence, and life satisfaction. PLoS
ONE, 17(4), 1-12. https://doi.org/10.1371/journal.pone.0266039
Hope, N. H., Holding, A. C., Verner‐Filion, J. r. m., Sheldon, K. M., & Koestner, R. (2019).
The path from intrinsic aspirations to subjective well-being is mediated by changes in
basic psychological need satisfaction and autonomous motivation: A large prospective
test. Motivation and Emotion, 43, 232-241. https://doi.org/10.1007/s11031-018-9733-z
Hovdhaugen, E., & Aamodt, P. O. (2005). Frafall fra universitetet. En nundersøkelse av
frafall og fullføring blant førstegangsregistrerte studenter ved Universitetet i Bergen,
Universitetet i Oslo og Norges teknisk-naturvitenskapelige universitet (NTNU) høsten
1999 (Arbeidsnotat 13).
Howard, J. L., Bureau, J., Guay, F., Chong, J. X. Y., & Ryan, R. M. (2021). Student
motivation and associated outcomes: A meta-analysis from self-determination theory.
Perspectives on Psychological Science, 1-24.
https://doi.org/10.1177/1745691620966789
Howard, J. L., Gagné, M., & Bureau, J. S. (2017). Testing a continuum structure of self-
determined motivation: A meta-analysis. Psychological Bulletin, 143(12), 1346-1377.
https://doi.org/10.1037/bul0000125
Hu, L.-T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives, Structural Equation Modeling.
A Multidisciplinary Journal, 6(1), 1-55.
Hughes, D. J., Adie, J. W., Kratsiotis, I. K., Bartholomew, K. J., Bhakta, R., & Martindale, J.
(2023). Dark personality traits and psychological need frustration explain future levels
of student satisfaction, engagement, and performance. Learning and Individual
Differences, 103. https://doi.org/10.1016/j.lindif.2023.102273
Iraola-Real, I., Matos, L., & Gargurevich, R. (2022). The type of motivation does matter for
university preparation. Estudos de Psicologia, 39(e190177).
https://doi.org//10.1590/1982-0275202239e190177
Jang, H.-R., Reeve, J., Cheon, S. H., & Song, Y.-G. (2019). Dual processes to explain
longitudinal gains in physical education students’ prosocial and antisocial behavior:
Need satisfaction from autonomy support and need frustration from interpersonal

MOTIVATION, ACHIEVEMENT AND DROPOUT
control. Sport, Exercise, and Performance Psychology.
https://doi.org/10.1037/spy0000168
Janke, S. (2022). Give me freedom or see my motivation decline: Basic psychological needs
and the development of students' learning goal orientation. Learning and Individual
Differences, 96. https://doi.org/10.1016/j.lindif.2022.102158
Janke, S., Messerer, L. A. S., & Daumiller, M. (2022). Motivational development in times of
campus closure: Longitudinal trends in undergraduate students' need satisfaction and
intrinsic learning motivation. British Journal of Educational Psychology, 00, 1-15.
https://doi.org/10.1111/bjep.12522
Jeno, L. M., Danielsen, A. G., & Raaheim, A. (2018). A prospective investigation of students’
academic achievement and dropout in higher education: A Self-Determination Theory
approach. Educational Psychology, 38(9), 1163-1184.
https://doi.org/10.1080/01443410.2018.1502412
Jeno, L. M., Levesque-Bristol, C., Nylehn, J., Pavlovic, Z., Roness, D., & Weinstein, N. (in
prep). A Self-Determination Theory perspective on student teachers’ basic needs,
motivation, and educational quality: A pre-registered qualitative study.
Jeno, L. M., Nylehn, J., Hole, T. N., Raaheim, A., Velle, G., & Vandvik, V. (2021).
Motivational determinants of students’ academic functioning: The role of autonomy-
support, autonomous motivation, and perceived competence. Scandinavian Journal of
Educational Research. https://doi.org/10.1080/00313831.2021.1990125
Johansen, M. O., Eliassen, S., & Jeno, L. M. (2023). "Why is this relevant for me?":
Increasing content relevance enhances student motivation and vitality. Frontiers in
Psychology, 14(1184804). https://doi.org/10.3389/fpsyg.2023.1184804
Kline, R. B. (2011). Principles and practice of structural equation modeling. The Guilford
Press.
Koestner, R., & Losier, G. F. (2002). Distinguishing three ways of being internally motivated:
A closer look at introjection, identification, and intrinsic motivation. In E. L. Deci &
R. M. Ryan (Eds.), Handbook of Self-Determination Research (pp. 101-121). The
University of Rochester Press.
Kusurkar, R. A., Croiset, G., Galindo-Garré, F., & Cate, O. T. (2013). Motivational profiles of
medical students: Association with study effort, academic performance and
exhaustion. BMC Medical Education, 13(87), 1-8.
Kusurkar, R. A., Orsini, C., Somra, S., Artino, A. R. J., Daelmans, H. E. M., Schoonmade, L.
J., & van der Vleuten, C. (2023). The effect of assessments on student motivation for
learning and its outcomes in health professions education: A review and realist
synthesis. Academic Medicine, 98(9), 1083-1092.
https://doi.org/10.1097/acm.0000000000005263
Larose, S., Ratelle, C. F., Guay, F., Senécal, C., & Harvey, M. (2006). Trajectories of science
self-efficacy beliefs during the college transition and academic and vocational
adjustment in science and technology programs. Educational Research and
Evaluation, 12(4), 373-393. https://doi.org/10.1080/13803610600765836
Leo, F. M., López-Gajardo, M. A., & Pulido, J. J. (2023). Development of cohesion and
relatedness in the classroom to optimize learning processes in the educational setting.
In B. Ng (Ed.), Self-Determination Theory and Socioemotional Learning. Springer.
https://doi.org/10.1007/978-981-99-7897-7_6
León, J., & Núñez, J. L. (2013). Causal ordering of basic psychological needs and well-being.
Social Indicators Research, 114, 243-253. https://doi.org/10.1007/s11205-012-0143-4
Leroy, N., & Bressoux, P. (2016). Does amotivation matter more than motivation in
predicting mathematics learning gains? A longitudinal study of sixth-grade students in

MOTIVATION, ACHIEVEMENT AND DROPOUT
France. Contemporary Educational Psychology, 44(45), 41-53.
https://doi.org/10.1016/j.cedpsych.2016.02.001
Levesque-Bristol, C. (2021). Student-centered pedagogy and course transformation at scale.
Stylus Publishing.
Levesque-Bristol, C., Richards, K. A. R., Zissimopoulos, A., Wang, C., & Yu, S. (2020). An
evaluation of the integrative model for learning and motivation in the college
classroom. Current Psychology. https://doi.org/10.1007/s12144-020-00671-x
Li, C.-H. (2016). Confirmatory factor analysis with ordinal data: Comparing robust maximum
likelihood and diagonally weighted least squares. Behavior Research Methods, 48,
936-949. https://doi.org/10.3758/s13428-015-0619-7
Longo, Y., Gunz, A., Curtis, G. J., & Farsides, T. (2016). Measuring need satisfaction and
frustration in educational and work contexts: The need satisfaction and frustration
scale (NSFS). Journal of Happiness Studies, 17, 295-317.
https://doi.org/10.1007/s10902-014-9595-3
Manganelli, S., Cavicchiolo, E., Mallia, L., Biasi, V., Lucidi, F., & Alivernini, F. (2019). The
interplay between self-determined motivation, self-regulated cognitive strategies, and
prior achievement in predicting academic performance. Educational Psychology,
39(4), 470-488. https://doi.org/10.1080/01443410.2019.1572104
Martela, F., & Ryan, R. M. (2023). Clarifying eudaimonia and psychological functioning to
complement evaluative and experiential well-being: Why basic psychological needs
should be measured in national accounts of well-being. Perspectives on Psychological
Science, 1-15. https://doi.org/10.1177/17456916221141099
Messerer, L. A. S., Karst, K., & Janke, S. (2023). Choose wisely: intrinsic motivation for
enrollment is associated with ongoing intrinsic learning motivation, study success and
dropout. Studies in Higher Education, 48(1), 137-150.
https://doi.org/10.1080/03075079.2022.2121814
Niemiec, C. P., Ryan, R. M., & Deci, E. L. (2009). The path taken: Consequences of attaining
intrinsic and extrinsic aspirations in post-college life. Journal of Research in
Personality, 43, 291-306. https://doi.org/10.1016/j.jrp.2008.09.001
NOKUT. (2022). Sluttrapport fra evalueringen av lektorutdanning for trinn 8-13 (15).
Nosek, B. A., Beck, E. D., Campbell, L., Flake, J. K., Hardwicke, T. E., Mellor, D. T., van’t
Veer, A. E., & Vazire, S. (2019). Preregistration is hard, and worthwhile. Trends in
Cognitive Sciences, 23(10), 815-818. https://doi.org/10.1016/j.tics.2019.07.010
Pan, Y., & Gauvain, M. (2012). The continuity of college students’ autonomous learning
motivation and its predictors: A three-year longitudinal study. Learning and Individual
Differences, 22, 92-99. https://doi.org/10.1016/j.lindif.2011.11.010
Pinheiro, J., Bates, D., & R Core Team. (2023). nlme: Linear and Nonlinear Mixed Effects
Models. https://CRAN.R-project.org/package=nlme
R Core Team. (2018). R: A language and environment for statistical computing. In R
Foundation for Statistical Computing. https://www.R-project.org/
Ramboll. (2024). SurveyXact. In https://rambollxact.com/surveyxact
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models. Applications and data
analysis methods (2 ed.). Sage Publications, Inc.
Robinson, K. A. (2023). Motivational climate theory: Disentangling definitions and roles of
classroom motivational support, climate, and microclimates. Educational
Psychologist. https://doi.org/10.1080/00461520.2023.2198011
Robinson, K. A., Lee, Y.-k., Bovee, E. A., & Perez, T. (2019). Motivation in transition:
Development and roles of expectancy, task values, and costs in early college
engineering. Journal of Educational Psychology, 111(6), 1081-1102.
https://doi.org/10.1037/edu0000331

MOTIVATION, ACHIEVEMENT AND DROPOUT
Roeser, R. W., Urdan, T. C., & Stephens, J. M. (2009). School as a context of student
motivation and achievement. In K. R. Wentzel & A. Wigfield (Eds.), Handbook of
Motivation at School (pp. 381-410). Routledge.
Roness, D. (2011). Still motivated? The motivation for teaching during the second year in the
profession. Teaching and Teacher Education, 27, 628-638.
https://doi.org/10.1016/j.tate.2010.10.016
Roness, D., & Smith, K. (2010). Stability in motivation during teacher education. Journal of
Education for Teaching, 36(2). https://doi.org/10.1080/02607471003651706
Rosseel, Y. (2012). lavaan: An R package for structural equation modeling. Journal of
Statistical Software, 48(2), 1-36.
Ryan, R. M., & Deci, E. L. (2017). Self-Determination Theory. Basic psychological needs in
motivation, development, and wellness. The Guilford Press.
Ryan, R. M., & Deci, E. L. (2020). Intrinsic and extrinsic motivation from a self-
determination theory perspective: Definitions, theory, practices, and future directions.
Contemporary Educational Psychology, 61, 1-11.
https://doi.org/10.1016/j.cedpsych.2020.101860
Ryoo, J. H. (2011). Model selection with the linear mixed model for longitudinal data.
Multivariate Behavioral Research, 46(4), 598-624.
https://doi.org/10.1080/00273171.2011.589264
Sallai, G. M., Bahnson, M., Shanachilubwa, K., & Berdanier, C. G. P. (2023). Persistence at
what cost? How graduate engineering students consider the costs of persistence within
attrition considerations. Journal of Engineering Education, 112(3), 613-633.
https://doi.org/10.1002/jee.20528
Schermelleh-Engel, K., Moosbrugger, H., & Müller, H. (2003). Evaluating the fit of structural
equation models: Tests of significance and descriptive goodness-of-fit measures.
Methods of Psychological Research Online, 8(2), 23-74.
Sheldon, K. M., & Krieger, L. S. (2007). Understanding the negative effects of legal
education on Law students: A longitudinal test of self-determination theory.
Personality and Social Psychology Bulletin, 33(6), 883-897.
https://doi.org/10.1177/0146167207301014
Skagen, K., & Elstad, E. (2023). Teacher education in Norway. In E. Elstad (Ed.), Teacher
education in the nordic region: Challenges and opportunities (pp. 175-193).
https://doi.org/10.1007/978-3-031-26051-3_6
Slemp, G. R., Field, J. G., & Cho, A. S. H. (2020). A meta-analysis of autonomous and
controlled forms of teacher motivation. Journal of Vocational Behavior, 121, 1-20.
https://doi.org/10.1016/j.jvb.2020.103459
Statistics Norway. (2024). Flere studenter – færre tar lærerutdanning.
https://www.ssb.no/utdanning/hoyere-utdanning/statistikk/studenter-i-universitets-og-
hogskoleutdanning/artikler/flere-studenter--faerre-tar-laererutdanning
Tahiri, M. J. B., & Mouratidis, A. (2024). Learning strategies and procrastination as a
function of need satisfaction and autonomous motivation: A diary study. European
Journal of Education, 1-18. https://doi.org/10.1111/ejed.12645
Tinto, V. (1993). Leaving college: Rethinking the causes and cures of student attrition. The
University of Chicago Press.
UNESCO. (2013). A teacher for every child: Projecting global teacher needs from 2015 to
2030 (UIS Fact sheet, Issue. http://www.uis.unesco.org/Education/Documents/fs27-
2013-teachers-projections.pdf
Vallerand, R. J., & Bissonnette, R. (1992). Intrinsic, extrinsic, and amotivational styles as
predictors of behavior: A prospective study. Journal of Personality, 60(3), 599-620.

MOTIVATION, ACHIEVEMENT AND DROPOUT
Vallerand, R. J., Fortier, M. S., & Guay, F. (1997). Self-determination and persistence in a
real-life setting: Toward a motivational model of high school dropout. Journal of
Personality and Social Psychology, 72(5), 1161-1176.
Vansteenkiste, M., Aelterman, N., Muynck, G.-J. D., Haerens, L., Patall, E. A., & Reeve, J.
(2018). Fostering personal meaning and self-relevance: A Self-Determination Theory
perspective on internalization. The journal of Experimental Education, 86(1), 30-49.
https://doi.org/10.1080/00220973.2017.1381067
Vansteenkiste, M., Soenens, B., & Ryan, R. M. (2023). Basic psychological needs theory: A
conceptual and empirical review of key criteria. In R. M. Ryan (Ed.), The Oxford
handbook of Self-Determination Theory. Oxford University Press.
Vansteenkiste, M., Zhou, M., Lens, W., & Soenens, B. (2005). Experiences of autonomy and
control among chinese learners: Vitalizing or immobilizing? Journal of Educational
Psychology, 97(3), 468-483. https://doi.org/10.1037/0022-0663.97.3.468
Wang, C., Cho, H. J., Wiles, B., Moss, J. D., Bonem, E. M., Li, Q., Lu, Y., & Levesque‐
Bristol, C. (2022). Competence and autonomous motivation as motivational predictors
of college students’ mathematics achievement: from the perspective of self-
determination theory. International Journal of STEM Education, 9(41), 1-14.
https://doi.org/10.1186/s40594-022-00359-7
Wang, Y., Tian, L., & Scott Huebner, E. (2019). Basic psychological needs satisfaction at
school, behavioral school engagement, and academic achievement: Longitudinal
reciprocal relations among elementary school students. Contemporary Educational
Psychology, 56, 130-139. https://doi.org/10.1016/j.cedpsych.2019.01.003
Wikan, G., & Bugge, L. S. (2014). Student performance in teacher education in Norway: the
impact of student, institutional and structural factors. European Journal of Teacher
Education, 37(4), 442-452. https://doi.org/10.1080/02619768.2014.912626
Young-Jones, A., Levesque, C., Fursa, S., & McCain, J. (2019). Autonomy-supportive
language in the syllabus: Supporting students from the first day. Teaching in Higher
Education, 1-16. https://doi.org/10.1080/13562517.2019.1661375
Yu, S., Levesque-Bristol, C., & Maeda, Y. (2017). General need for autonomy and subjective
well-being: A meta-analysis of studies in the US and East Asia. Journal of Happiness
Studies, 1-20. https://doi.org/10.1007/s10902-017-9898-2