Fostering experimental competences of prospective physics teachers

Abstract

Physics teachers are often faced with the challenge of having to set up difficult experiments, or they have to consciously manipulate parameters in order to be able to demonstrate a phenomenon convincingly. Comprehensive laboratory courses are standard procedure in any study program for prospective physics teachers. However, many students, even after completing standard laboratory courses, show difficulties in standard experimental situations, such as measuring an electric current. We report on a new seminar concept for students in physics teacher study programs. This concept is based on the current state of research in physics education, on the teachers’ professional competences, and on the modelling of experimental competence, justifying its embedding in teacher training. We present first results of a pilot study carried out to evaluate the seminar.

Full text

PA P E R
Phys. Educ. 56 (2021) 045020 (19pp) iopscience.org/ped
Fostering experimental
competences of
prospective physics
teachers
Philipp Bitzenbauer∗and Jan-Peter Meyn
Department Physik, Professur für Didaktik der Physik, Friedrich-Alexander-Universität
Erlangen-Nürnberg, Staudtstraße 7, Erlangen D-91058, Germany
E-mail: philipp.bitzenbauer@fau.de
Abstract
Physics teachers are often faced with the challenge of having to set up
difcult experiments, or they have to consciously manipulate parameters in
order to be able to demonstrate a phenomenon convincingly. Comprehensive
laboratory courses are standard procedure in any study program for
prospective physics teachers. However, many students, even after completing
standard laboratory courses, show difculties in standard experimental
situations, such as measuring an electric current. We report on a new seminar
concept for students in physics teacher study programs. This concept is
based on the current state of research in physics education, on the teachers’
professional competences, and on the modelling of experimental
competence, justifying its embedding in teacher training. We present rst
results of a pilot study carried out to evaluate the seminar.
Keywords: experimental competence, teacher education, professional knowledge,
assessment
1. Experimental competences in teacher
studies
Experiments are a central component in physics
lessons at the secondary school level [1,2] and are
∗Author to whom any correspondence should be addressed.
Original Content from this work may be used
under the terms of the Creative Commons
Attribution 4.0 licence. Any further distribution of this work
must maintain attribution to the author(s) and the title of the
work, journal citation and DOI.
by some even seen as the dening characteristic
of physics curricula [1, p 655]. Thus, fostering
experimental competences of prospective physics
teachers is an important learning target in teacher
education and various approaches are being pur-
sued and investigated to achieve this goal [3].
When we discuss ways to enhance experi-
mental competences of prospective physics teach-
ers in the course of teacher training, we must
embed the construct experimental competence
into the model of teachers’ professional com-
petences [4–7]. We argue that meaningful
1361-6552/21/045020+19$33.00 1 © 2021 The Author(s). Published by IOP Publishing Ltd

P Bitzenbauer and J-P Meyn
Figure 1. Experimental competence as we locate it within the model of teachers’ professional competence, fol-
lowing Baumert and Kunter [4]. Thus, in this article we suggest that experimental competence of prospective
physics teachers must not only be encouraged in the context of laboratory courses (focusing on CK), but also in
the context of didactic education (linking CK and PCK), cf [8], because PCK can be regarded an interplay of ‘what
a teacher knows, what a teacher does, and the reasons for the teacher’s actions’ [9, p 159].
experimentation in the classroom requires high
standards in all respects: namely in content know-
ledge (CK), pedagogical content knowledge
(PCK) and pedagogical knowledge, as well as
in beliefs about teaching and learning, and motiv-
ational orientations. Therefore, all parts of the
model should be regarded the sine qua non for the
teachers’ experimental competence (see gure 1).
During their studies, prospective physics
teachers are expected to improve their funda-
mental experimental competences by completing
laboratory courses mainly focusing on the pro-
spective teachers’ CK. The learning goals that can
usefully be pursued with such laboratory courses
are indicated by The American Association of
Physics Teacher’s proposals [10], and reference
[11] shows how these can be operationalised in
practice. Usually, learning goals for laboratory
courses cover everything from students’ under-
standing of physical content and learning about
experimental measurements or uncertainties to
the development of communication skills [12].
At many universities, traditional introductory lab
courses are related to introductory lecture courses
with an emphasise on fostering experimentation
skills or with the aim to reinforce course content
because ‘the underlying rationale is that students
will better understand the physics if they conduct
experiments and see for themselves how the phys-
ics principles work in real life’ [12, p 38].
2. Physics education research on
laboratory courses
While empirical research on physics lecture
courses has long been established in the phys-
ics education research community [13], system-
atic research on the effectiveness of laboratory
courses has only recently been advanced, e.g.
[14–21]: These studies’ results reveal a rather
sobering picture for laboratory courses, as their
learning effectiveness has been doubted by vari-
ous researchers [22,23]. For example, Wieman
and Holmes question the effectiveness of labor-
atory courses in supporting mastery of physics
content [14]. Specically, Holmes et al found no
measurable effect on students’ course perform-
ances when researching lab courses [18].
With respect to the students’ cognitive activ-
ities in lab courses, the empirical ndings are also
not very promising. In the summary of their study
on the cognitive activities engaged in undergradu-
ate laboratory courses, Holmes and Wieman state:
‘While it is understandable that
lecture courses would not engage
students in cognitive tasks
involved in physics experiment-
ation, the lack of exposure in tradi-
tional lab courses where students
are conducting experiments may
seem surprising’ [17, p 9].
July 2021 2Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
Holmes and Wieman explain their nding in
another article by stating that the students’ only
thinking in the laboratory courses ‘was in ana-
lysing data and checking whether it was feasible
to nish the lab in time’ [12, p 40]. As the cent-
ral cause for this nding, the authors identify the
structure of typical laboratory courses because
‘if you break down the elements
of a typical lab activity, you real-
ize that all the decision making
involved in doing experimental
physics is done for the students
in advance. The relevant equations
and principles are laid out in the
preamble; students are told what
value they should get for a par-
ticular measurement or given the
equation to predict that value; they
are told what data to collect and
how to collect them; and often
they are even told which buttons to
press on the equipment to produce
the desired output’ [12, p 40].
This nding is in accordance with those of
previous studies in different settings, e.g. on lab
work in schools: the constant concern of students
in typical laboratory courses is to cope with tasks
within the given constraints, which can hinder an
effective learning process [24,25].
Thus, although several studies do also indic-
ate potential values of such lab courses, e.g.
fostering the students’ attitudes and beliefs
about experimental physics [20] or the students’
experimentation-focused critical thinking skills
[21], the effectiveness of fostering experimental
competences through typical laboratory courses
must be questioned: ‘by working on a common
template, students, strictly following the instruc-
tions, can safely perform the work, being not fully
aware of the essence of the experiment conducted’
[26]. The instructions given are worked through
by students during the laboratory courses, but this
seems to promote the acquisition of experimental
competences only to a limited extent, as hardly
any complex decision-making situations have to
be mastered [27]. So ‘the typical laboratory exper-
ience [...] is a hands-on but not a minds-on activ-
ity’ [28].
Figure 2. Sample sketch of test person P24A in our
pre-test, preparing the measurement of electric current
and voltage in an electric circuit. Student P24A sugges-
ted measuring voltage in series, with the electrical cur-
rent parallel to the voltmeter. In doing so, he is no stat-
istical outlier. We provide deeper insight into the design
and the evaluation of our new seminar concept in this
article.
In fact, it is not only the complex experi-
ments that cause difculties for students, even
when they were formally successful participants
in these lab courses. Even standard experimental
situations, that occur in everyday classroom situ-
ations, such as measuring an electric current
or determining the focal length of a lens, may
bewilder students. We conrmed these ndings
with the results of a pre-test within a pilot
study with prospective physics teachers (⩾5th
semester), evaluating a new seminar concept
which we present in section 4. Figure 2shows
the sketch on how to measure current and voltage
from our pre-test by a student teacher in 5th
semester, after the completion of two laboratory
courses.
Our new seminar concept is intended to solve
the problem. We provide an additional way to
complement traditional laboratory courses, with
a view to advancing the experimental compet-
ences of prospective physics teachers. It is partic-
ularly important to us that the prospective physics
teachers have the opportunity and time to develop
experimental skills in our seminar that are use-
ful for everyday classroom practice. Therefore,
during this new seminar, we impart both know-
ledge of the fundamental techniques for carrying
out experiments in everyday school life, as well as
the professional knowledge for embedding exper-
iments in the classroom in a way that promotes
learning, while always taking into account the par-
ticular needs of prospective physics teachers, and
striving to combine teachers’ CK and PCK (cf
gure 1).
July 2021 3Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
Figure 3. Model of experimental competence as presented by Schreiber et al [35,36]. The central aspect of
this model is the concrete implementation of an experiment, focusing on concrete techniques for carrying out
experiments.
Core ideas of our seminar are presented in 4
and lastly, we present the results of a pilot study
evaluating the seminar (cf section 5), provid-
ing rst empirical evidence that the new seminar
concept contributes to the teaching of basic exper-
imental techniques, which are essential for teach-
ers in physics classrooms.
A necessary prerequisite for the valid assess-
ment of experimental competence of students is a
model of experimental competence that can actu-
ally be put into practice. This is presented in the
following section 3.
3. Modelling and assessing experimental
competence
3.1. Modelling experimental competence
Eickhorst et al dene the term experimental com-
petence as a ‘latent ability to at least intuitively
rule-based planning and implementation of exper-
iments aimed at clarifying a physical question, as
well as for the methodically conscious evaluation
of the gained data’ (translated from [29]) and state
that experimental competence shows in perform-
ance if the essential CK is accessible.
A distinction between three sub-competences
is common in various models of experimental
competence [30,31], taking up the presentation
of a scientic process of investigation and inquiry
[32], namely:
•Finding and setting up hypotheses
•Planning and carrying out experiments
•Analysis of experimental results
Thus, models of experimental competence
usually consist of three pillars that can be iden-
tied with the phases of experimentation, and
mostly only vary with regards to the scope and
weighting of the individual sub-competences, as
Theyßen et al reveal [33]. In a different model
according to [34], however, experimental com-
petence is modelled by means of various sub-
dimensions that are related to experimental prob-
lem types.
In the model of Schreiber et al [35,36], the
three main domains planning, realisation, and ana-
lysis encompass the characteristic features of the
process steps of experimentation, plus two further
features in between domains:
In this article, we follow the model presen-
ted by Schreiber et al focusing on the concrete
realisation of experiments rather than on the plan-
ning or on the analysis of data. We claim that
July 2021 4Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
without the teacher’s ability in basic experimenta-
tion techniques to carry out experiments, the bene-
cial embedding in the classroom situation cannot
succeed.
3.2. Assessment of experimental
competence
There are a number of difculties in the assess-
ment of experimental competences that need to
be considered in the test development: In the case
of test procedures in written form, experimental
activities appear to be underrepresented, casting
doubts on content validity [33]. In tests with
real experiments, only exemplary experimental
tasks can be demanded, which prevents reli-
able measurement of the construct experimental
competence.
Theyßen et al [33] provide a test instrument
that is intended to take both these aspects into
account: With their computer-based test proced-
ure, the authors facilitate the assessment of exper-
imental competences of school pupils by means
of interactive simulations instead of real experi-
ments. Thereby, not only the results of the exper-
iments are included in the rating, but also the
process of experimentation [36]. This on-screen
procedure allows for the reliable and valid assess-
ment of experimental skills even in large scale
assessments [37]. For results of the pilot studies,
see [38]: For example, through convergent val-
idation using simulations and hands-on testing,
it was found that students’ on-screen perform-
ances on this browser-based test instrument can
indeed be interpreted as a measure of the students’
experimental competence [39]. This is important,
because in previous studies the exchangeability of
hands-on procedures, written tests, and computer-
based procedures had been examined, cf [40–42],
revealing only low correlations between hands-on
assessments and the other procedures.
The authors provide 12 test units, that can
be used in different congurations to build test
booklets (cf table 1). In every test unit, a com-
mon school experiment from either optics, mech-
anics or electricity is chosen as a central topic.
Each unit is made up of different items in order
to cover all of the three areas Planning,Realisa-
tion of the experiment and Analysis of the results
within every unit.
Theyßen et al [33] extracted the experi-
ments for their test instrument from detailed
curriculum analyses and expert surveys. This
shows that teachers in training consider these
experiments to be relevant in classroom practice.
In particular, it is required that physics teach-
ers themselves have the experimental compet-
ences to carry out these particular experiments,
which means this is conducive to content valid-
ity. Thus, it seems to be appropriate to empirically
assess
(a) the difculties of prospective teachers with
standard experimental situations and
(b) their learning gains within a rst pilot study,
evaluating the effectiveness of a new seminar
concept,
notwithstanding that the test instrument was ori-
ginally developed for younger pupils.
4. Seminar on basic experimental
techniques
In this section, we present our seminar concept on
basic experimental techniques. We start by giving
a brief insight into physics teachers’ training in
Germany and situate the seminar concept within
the curriculum.
4.1. German physics teacher education
In a document on the content requirements for
the subject disciplines and subject matter didactics
in teacher education, the Standing Conference of
the Ministers of Education and Cultural Affairs of
the Länder in Germany set standards for teacher
education [43]. Teacher education in Germany is
divided into three phases at different educational
institutions; for the German education system
concerning physics teachers at secondary school
level cf gure 5. In the rst phase, competences
concerning the subject disciplines, working meth-
ods and subject matter didactics [43, p 3] are
acquired during the studies at university. With
regard to knowledge and working methods, the
subject-specic competence prole for physics
details that ‘knowledge and skills in experimenta-
tion and in handling (typical school) equipment’
(translated from [43, p 50]) should be acquired
July 2021 5Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
Table 1. All units provided by [33]. Four units (two from one topic, two from another) form one test booklet. In
our pilot study, evaluating our new seminar concept presented in 4of this article, we used units E1, E2, O3 and
O4 (in italics) because the topic mechanics is not discussed during our seminar.
Electricity Units Optics Units Mechanics units
E1: V-I-charactersistic of a
light bulb
O1: Refraction of light on the semicircle
block
M1: Density determination
E2: Power of light bulbs O2: Reection on the semicircle block M2: Strech of a rubber band
E3: Parallel circuit of light
bulbs
O3: Total reection M3: Buoyancy in water
E4: Series circuit of light
bulbs
O4: Determining the focal length of a
lens
M4: Inclined plane
Figure 4. Example for a pupil’s solution (screenshot) for an item with the task to measure a lightbulb’s V-I-curve
as a part of unit E2, reproduced from [33]. For such items with high interactivity, a rating is carried out on a ve-
stage scale (0 =not suitable, ..., 4 =totally suitable), while the other items are rated on either a three-stage scale
or in a dichotomous manner. The detailed coding manual provided by the authors of [33] facilitates an evaluation
according to exact criteria.
during the courses at university. Physics laborat-
ory courses are specically provided in the rst
phase of teacher education in order to promote
these experimental competences.
The seminar concept presented here (cf
section 4.2) also aims to promote experimental
competences, i.e. it is also dedicated to study-
ing subject knowledge and knowledge of meth-
ods. However, we embed this seminar directly
into training in physics education: The stu-
dents attend the experimental seminar alongside
a lecture introducing physics education. In our
experimental seminar, learners’ difculties are
repeatedly taken up and discussed, using the
example of basic experimental techniques. In this
way, we link the superordinate areas of com-
petence from subject science and subject matter
didactics in order to mutually promote CK and
PCK of the students, as explained in section 1. In
the following section, we present the seminar on
basic experimental techniques in more detail.
4.2. Seminar concept on basic
experimental techniques
Our seminar is different to conventional labor-
atory courses in two aspects. Firstly, assign-
ments are given without any specication on the
equipment to be used, and secondly, the whole
collection of apparatus which may possibly be
July 2021 6Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
Figure 5. Overview of the three phases in German physics teacher education.
found in the natural sciences collections of sec-
ondary schools is available. Easy tasks become
difcult because of the additional decisions the
students have to make. Therefore, the initial
assignments seem to be very simple in the context
of a university course.
The rst task is to turn on a light bulb. This
seems to be an inordinately simple challenge for
any physics student who has completed introduct-
ory lectures, laboratory courses, and even a lec-
ture on theoretical electrodynamics. It is, however,
trickier than it appears. The bulbs are not labelled,
except at the socket. The connecting cables have
different lengths and colours. Most importantly,
there are more than 30 different power supplies
on offer. More often than not, one daredevil group
goes ahead with a power supply they have seen
before and just try their luck. Then, other groups
usually follow, but there simply are not enough
power supplies of the same kind. At that point,
the teacher has to step into the process and ask the
students to reect on their choices. These phases
of reection will occur several times in a ses-
sion. Soon, everyone realises that inspecting an
improper setup is not done to expose its deviser
to ridicule, but instead a good opportunity for all
participants to learn from mistakes.
The second task is to measure the voltage
and current in the light bulb circuit with a digital
multimeter. Our observation of how this chal-
lenge was handled by participants prompted us
to carry out the empirical investigation presented
here. We got the distinct impression that measur-
ing voltage and current had not become a student’s
core competence, even after completing two uni-
versity laboratory courses. This impression was
later conrmed, as we will specify in section 5. It
must be stated that our students have easy access
to the laboratory outside the lecture period to prac-
tice. A teacher is present only occasionally and
on request, so students can make perceived blun-
ders without fearing negative comments. A certain
amount of spare parts, especially fuses and bulbs,
is necessary, but this is a good investment. The
students learn to admit mistakes and to be respons-
ible for equipment to be repaired.
In the second class session, a circuit with a
light bulb has to be set up again, but this time with
a 12 V/100 W halogen lamp. Several power sup-
plies which have been used successfully before
will fail due to insufcient current. When using
only suitable power supplies with the minimum
current specication of 10 A, what happens is that
these lamps differ in brightness signicantly, so
we end up having to measure the voltage and cur-
rent again.
It becomes apparent that some of the power
supplies, namely transformers without stabilisa-
tion, provide a much smaller voltage than the one
specied at the dial. A typical value is 9 V instead
of 12 V. The students will most likely have heard
about the internal resistance of a power supply,
but they have had no occasion to experience the
implication of this fact. Some advanced students
July 2021 7Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
might notice that it is best to adjust the voltage
to 12 V directly at the lamp, because some 100
mV are lost on the cables. Overall, there are sev-
eral ways to implement and optimise even a trivial
experiment such as the light bulb setup. At this
point, it looks like a mere exercise, but it is also
applicable later on in optics experiments, where
we want to make the available lamps as bright as
possible.
The following lessons cover DC and AC cir-
cuits, an oscilloscope, digital data acquisition, and
optics, including diffraction and elementary spec-
troscopy. Due to limited time, we do not cover
mechanics and thermodynamics. The reasoning
behind this is that it appears to make more sense
to look at a few exemplary experiments in depth,
rather than skimming over the entire eld of high
school physics. The underlying principle is always
the same: Avoid the tried and trusted recipe, but
instead analyse the task with a physicist’s mind.
For example: Is the voltage drop over a 1 m long
laboratory cable signicant?—Do not speculate or
ask the internet, but calculate, and conrm your
calculations by exact measurements.
The nal assignment is to resolve the dublett
spectral line of sodium at 589.0 and 589.6 nm.
Using a 1200 lines mm−1reective diffraction
grating and a simple projection of the slit image
onto a screen yields a result easily. A challenge
is posed by replacing the 1200 lines mm−1grat-
ing with gratings of a smaller period. Depend-
ing on the size of the setup, it becomes necessary
to use a magnifying glass to look at the delicate
spectral lines. By using an eyepiece to look dir-
ectly towards the lamp, the subjective brightness is
strongly enhanced. The green dublett is observed
additionally, providing an occasion to talk about
spin orbit interaction. Further reduction of disper-
sion by using lower resolution gratings or even
prisms forces the students to improve the align-
ment of optical elements. With simple achromatic
lenses and an F2 prism of 60 mm base length,
one can observe a diffraction-limited image, if
all parts are aligned perfectly. Only a few stu-
dents’ setups yield the diffraction limit, but every-
one handles the job much better than during the
rst approach. The quality of alignment is not
judged by the teacher, but by nature. A majority
of students take the chance to practice in the open
laboratory.
In a consecutive seminar, students are asked
to search literature for innovative experiments on
a certain subject and give a talk with demon-
strations. Often it is necessary to alter a setup
because an important component is not available.
After coming up with the present seminar concept,
we have noticed a signicant improvement in the
quality of demonstrations.
5. Pilot study
5.1. Research question, sample and data
collection
We used the test instrument presented in section 3
using Units E1, E2, O3, O4 in a pilot study
with a pre-/post-test design in order to determine
the development of experimental competences of
prospective physics teachers initiated by our sem-
inar concept (cf section 4.2). Our research ques-
tion for the pilot study was: To what extent does
the seminar concept advance experimental com-
petences of prospective physics teachers, espe-
cially with regards to the realisation of experi-
ments?
We surveyed the experimental competences
of N=22 (14 male, 8 female) physics teacher
students pre and post participating in our sem-
inar during winter term 2019/20. At this point, all
participants in the pilot study were in their fth
semester. The training in physics education, and
thus also the seminar presented here, is scheduled
only after successful participation in the introduct-
ory lectures on experimental physics. Hence, in
addition to two introductory laboratory courses,
all of our participants had successfully completed
the introductory lectures on experimental physics
in their rst four semesters before participating in
our seminar, specically on mechanics, thermo-
dynamics, electricity, optics, and atomic physics.
In particular, the participants of our pilot study did
not attend any other laboratory courses during this
winter term 2019/20.
5.2. Data analysis
In this article, we use the median Mdn as well
as the mean value mand standard deviation
σto describe our data. We performed non-
parametric tests to indicate differences: The Wil-
coxon test was used to investigate learning gains
July 2021 8Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
between pre- and post-test. Gender differences
were investigated using the Mann–Whitney U test.
Each effect was considered statistically signicant
when the p-value was below the 5% threshold. We
also report the effect of size measure in terms of
Pearson’s rto judge the magnitude of statistically
signicant effects [44].
The authors of reference [45] suggest spider-
webs to represent the degree of students’ vari-
ous experimental subcompetences [46]. In our
pilot study, we used spider-webs to summar-
ise and graphically illustrate students’ experi-
mental subcompetences pre and post the seminar
to provide feedback to the participants on the areas
of experimentation in which they had developed
their skills well or where there was still room
for improvement. For this purpose, an individual
spider-web was created for each student based on
the pre- and post-test results. Therefore, we cal-
culate students’ relative test scores (percentage of
maximum achievable points ranging from 0% to
100%) in the items corresponding to the follow-
ing subcompetences (cf gure 3) across different
units (cf table 1) pre and post our seminar:
•designing a test plan: seven items1in the whole
test, at least one in each unit (achievable score:
20 points)
•assemble devices and build up experiments:
four items in the whole test, one in each unit
(achievable score: 16 points)
•perform and document measurement: four
items in the whole test, one in each unit (achiev-
able score: 16 points)
•prepare and process measurement data: ve
items in the whole test, at least one in each unit
(achievable score: 6 points)
•interpret experimental results: three items in the
whole test, one in unit E1, one in unit O3 and
one in unit O4 (achievable score: 3 points)
In the original form according to [45, p 6],
these spider-webs included two further subcom-
petences, namely the development of questions
and the formation of hypotheses. However, both
aspects were not part of the test we used to assess
1For the specic designs of the individual items, we refer to
the work of the test’s developers [33].
Figure 6. Spider-web tool to summarise students’
experimental competences. By plotting participants’
relative test scores on the corresponding items of the test
on experimental competences (cf section 3.2), compet-
ence networks result. These give a graphical overview
of possible strengths and show in which areas there is
still a need of improvement.
prospective physics teachers’ experimental com-
petence, because the focus of our new seminar
concept is on the realisation of experiments. For
this reason, we did not consider these two sub-
competences for the spider-webs presented in this
article (cf gure 6).
5.3. Item statistics and reliability
In the test, a total of 61 points could be achieved
across the four units (α=0.68).The evaluation
was based on classical test theory. The coding
was carried out by two independent raters with
a high level of agreement (κ=0.81, 95% CI
[0.69; 0.93]).
The item difculties are predominantly in
the middle range of 0.2–0.8 [47] as can be
seen in gure 7, although the test instrument
was developed and piloted for a less physics-
educated group of people (pupils in schools, cf
section 3.2).
For the three subscales Planning (eight items
across all four units, α=0.31), Realisation (eight
items across all four units, α=0.57) and Ana-
lysis of results (eight items across all four units,
α=0.40) Cronbach’s αas estimator of the
internal consistency of a scale is below the accep-
ted threshold of 0.70 due to a wide range of con-
tents and actions covered. However, the subscales
are still used for this pilot study, because it is not
intended to differentiate between different ability
levels with them, but rather to gain insight into
those components of experimental competence in
July 2021 9Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
Figure 7. Item difculties for all 24 items shown across all units sorted by the three areas of the underlying
model of experimental competence. The red lines represent the accepted limits for moderate item difculty. This
gure suggests that overall the test is in accordance with the test persons’ skills (mean item difculty 0.59) at
the pre-test point in time, although it was developed for assessing pupils’ experimental competences. The average
discrimination index of the test items equals 0.23.
which learning gains are initiated by our new
experimental seminar in particular.
5.4. Pre-test results
5.4.1. Test scores. In the pre-test, the students
achieved an average of m=38 points (σ=9) ran-
ging from 32% (one student) to 87% (one stu-
dent), which corresponds to 63% of the possible
total score in average. The median value of the
total samples’ pre-test scores is Mdn =40 points.
A Mann–Whitney U test indicates statistically sig-
nicant differences in test scores between male
(Mdn =43) and female (Mdn =35) students in
the pre-test (U=25.5, Z(20) =−2.10, p<0.05;
r=0.33). A detailed overview of the descriptive
statistics on the pretest results including median,
mean value and standard deviation of the total
sample and for males and females separately is
given in table 2.
In the following subsections, we take a closer
look at several items of special interest and present
frequently observed difculties that prospective
physics teachers had with experimental tasks. We
limit ourselves to typical mistakes in the real-
isation of the experiments (or closely related
activities).
Table 2. Descriptive statistics on test scores (pre and
post) and learning gains, both for the total sample
and for males and females respectively. The maximum
achievable score in the test was 61 points. Deviations
in learning gain values from the difference of the mean
values in pre- and post-test are due to rounding.
Total sample Males Females
Mdn pre 40 43 35
(test score) post 50 51 45
mpre 38 41 33
(test score) post 49 51 45
σpre 9 8 8
(test score) post 4 4 3
Learning gain m11 9 12
σ8 9 6
5.4.2. Difficulties in optics. In Unit O3, the test
persons had to investigate the dependence of the
limit angle for total reection on the refractive
index of the optically denser medium at the trans-
ition to air. For this purpose, the students were
asked to choose semicircular blocks of different
materials (n>1), an angle-measuring disc, and a
light source out of a number of distractors in a vir-
tual experiment.
Investigating the sketches made by the stu-
dents to plan the experiment, two primary types
July 2021 10 Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
Figure 8. Screenshot of test person P17A’s sample
sketch. The semicircular block is not placed on the
angle measuring disc. In this manner, angles of refrac-
tion cannot be measured.
Figure 9. Screenshot of test person P25A’s sample
sketch. The semicircular block is not placed on the
angle-measuring disc correctly. In this manner, total
reection cannot be investigated.
of problems can be pinpointed: One of these prob-
lems was that the semicircular block was not
placed on the angle disc by 28% (6 out of 22)
of the test persons (cf gure 8). This makes it
impossible to measure the angles of refraction and
thus prevents any meaningful investigation of the
problem.
Another 6 of the 22 test persons located the
semicircular block on the angle disc, but arranged
it incorrectly (cf gure 9). The incident light must
hit the round side of the block perpendicularly, so
that total reection can then be investigated on the
planar surface. Test persons who offer the awed
solution, or the one mentioned above, appear to
lack basic knowledge of geometrical optics.
These problems were not only apparent in
the test sketches, but were also evident in the
experimental setups, even though helpful interim
solutions were accessible. Numerous other inac-
curacies, especially in the exact alignment of the
semicircular block on the angle disc, were found.
This would in effect have prevented proper meas-
urements in real experiments, cf gure 10.
Figure 10. Screenshot of test person P27A’s experi-
mental setup. The semicircular block is inaccurately
arranged on the angle-measuring disc. Experimental
inaccuracies due to students’ on-screen operations of
the computer simulations are considered by the coding
manual provided [33]. But in the case presented here,
the inaccuracy is very clear.
Figure 11. Screenshot of test person P31A’s sample
sketch. Student P31A suggested measuring voltage in
series.
5.4.3. Difficulties in electricity. In Unit E1, the
test persons had to measure the V-I-characteristic
of a light bulb. For this purpose, the students were
able to choose suitable devices such as a digital
multimeter, cables, and power supplies out of a
number of distractors in a virtual experiment.
In an initial item, the test persons were expec-
ted to describe the basic idea of the experiment,
i.e. they should be able to explain which phys-
ical quantity is varied in the experiment, and
which is measured. In this context, some stu-
dents unwittingly fell for known pre-concepts, as
reported in various studies from physics education
on the learning about electricity, cf [48–50]. For
example, 6 of the 22 respondents (28%) claimed
July 2021 11 Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
Figure 12. Screenshot of test person P17A’s sample
sketch. Student P17A suggested measuring voltage
in series, with the electrical current parallel to the
voltmeter.
that it was not the voltage that needed to be varied
with a power supply, but the electrical current.
Other problems arose in the test persons’
sketches: some participants (5 out of 22) were
apparently unaware that the voltage measurement
was a two-point measurement. They suggested
measuring the voltage applied to the light bulb in
series, or measuring it in parallel with the ampere-
meter, or vice versa (cf gure 11).
Sometimes, even both errors were revealed in
one (cf gure 12); conceptual problems of under-
standing with regard to the distinction between
voltage and electric current become apparent in
this case.
It is noteworthy that we observe a lack of
basic experimental techniques, which are simply
indispensable for professional physics lessons: for
example, test persons suggest measuring voltage
in series, or measuring the electrical current in
parallel (3 out of 22, cf gures 2and 12). The
observed errors were not only apparent in the test
sketches of the test persons, but also in the experi-
mental setups presented in the virtual experiment.
Here, we nd errors similar to the ones described
above (cf gure 13).
Various experimental setups show that many
of the participants do not even know the meaning
of the socket assignments on the digital multi-
meter, and in 3 out of 22 cases, participants neither
used the COM-socket, nor did they set the meas-
uring range correctly (cf gure 14).
Figure 13. Screenshot of test person P23A’s experi-
mental setup, suggesting measurement of the voltage in
series to the light bulb.
5.5. Interim conclusion
The ndings presented in this section—and of
the pilot study in general—are not generalisable,
due to the small sample. Furthermore, from the
ndings presented, we do not infer a conclusion
about the students’ ability to implement experi-
mental tasks on a given setup, as promoted in typ-
ical laboratory courses (cf section 2). However,
from the pretest results during our pilot study, we
conclude that undergraduate prospective physics
teachers (⩾5th semester) have a distinct lack of
fundamental experimental techniques and show
a need for clarication on fundamental technical
issues.
5.6. Post-test results
The experiments to be set up in the test that we
used for our pilot study (cf section 3.2) have not
been discussed during the seminar, with the edu-
cational objective in mind that the students had
to perform a transfer of knowledge. Nevertheless,
we need to point out that the subsequent results,
from the post-test, are not suitable for differenti-
ating between the different levels of competence
of the students; an instrument more suited to the
sample would be called for.
July 2021 12 Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
Figure 14. Screenshot of test person P19A’s experimental setup. P19A neither used the COM-socket, nor the
measuring range was set correctly.
Figure 15. Comparison of pre- and post-test results.
5.6.1. Test scores. In the post-test, the students
achieved an average of m=49 points (σ=4)
out of a total possible score of 61 points, which
corresponds to 80% ranging from 68% (one
student) to 95% (one student). The median of
the total samples’ post-test scores is Mdn =50
points, while the median in the pre-test was only
Mdn =40 points. Thus, we nd an average learn-
ing gain of 11 points from pre- (m=38, σ=9)
to post-test (m=49, σ=4), which is statistically
signicant after a Wilcoxon test (Z(17) =−3.27,
p<0.01; r=0.56). For an overall summary of
the test scores in pre- and post-test, cf table 2and
gure 15.
A Mann–Whitney U test indicates statistic-
ally signicant differences in test scores between
male (Mdn =51) and female (Mdn =45) stu-
dents in the post-test (U=7.50, Z(16) =−2.68,
p<0.01; r=0.47). However, the observed learn-
ing gain of female participants (m=12, σ=6)
July 2021 13 Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
Table 3. Pre- and post-test scores (m±σ) divided by the three subscales Planning,Realisation and Analysis in %
(rounded).
Subscale Pre-test score in % Post-test score in %
Planning 65 ±18 91 ±8
Realisation 67 ±17 83 ±10
Analysis 41 ±16 51 ±17
is slightly higher than that of male participants
(m=9, σ=9); therefore there is no indication that
the seminar would not allow for gender-sensitive
fostering of experimental competences.
Finally, an evaluation of the learning out-
comes with a distinction between the individual
components planning, realisation and analysis
within the model of experimental competence
described in section 3.1 suggests that the inter-
vention is effective in exactly those areas for
which the seminar was designed: In section 3.1
we explained our focus on the realisation phase,
which is necessarily interwoven with the plan-
ning phase. We argued that these are the ele-
ments essential for successful experimentation in
the classroom. The analysis of experimental data
plays (frequently unfairly) a subordinate role in
the classroom, and prospective physics teachers
should especially practice data analysis in the
laboratory practicals.
This focus within our seminar is also reec-
ted in the empirical data of our pilot study (cf
gure 16). While the learning gains in the items
on planning an experiment (Z(17) =−3.44, p<
0.001; r=0.59) and on the realisation of an
experiment (Z(17) =−2.73, p<0.01; r=0.47)
are statistically signicant, we nd no statistically
signicant learning gain in the items on analysis
of experimental data (Z(17) =−0.56, p=0.57).
A more detailed insight into the students’
learning gains can be obtained by taking a look
at the subcompetences associated with the indi-
vidual phases of experimentation. For this pur-
pose, we use the spider-webs introduced in section
5.2. In accordance with the results related to the
phases of experimentation we nd the following
(cf gure 17): while the students make signicant
progress in the sub-skills that are specically
related to the realisation of the experiments, e.g.
designing a test plan,assemble devices & build up
experiment or perform & document meausrement,
we nd no or only small progress for the analysis
of the data and the interpretation of experimental
results. This nding is in line with the expectation
because the seminar concept presented in this art-
icle very specically aims to promote basic exper-
imentation techniques.
Lastly, we want to underline these ndings
on learning gains described by the test scores in a
qualitative way. For this purpose, we present some
examples of students’ post-test work, whose pre-
test work has been used in section 5.4. to demon-
strate typical errors. These no longer appear in the
post-test or, if at all, only in individual students’
answers, as can be seen from the test scores in the
post-test: The students on average reached 91% of
the total possible score in the subscale Planning,
and 83% on average in the subscale Realisation,
cf table 3.
5.6.2. Qualitative description of the students’
performance in optics. In the pre-test, the two
central problems were that (a) the semicircularb-
lock was not placed on the angle disc by 6 out
of 22 test persons and that (b) the angle disc was
not arranged correctly (incident light must hit the
round side of the block) by further six students.
Such problems were not observed in the post-test
(cf gure 18).
5.6.3. Qualitative description of the students’
performance in electricity. In the pre-test, we
observed three central difculties in the students’
solutions: For example, ve of the test persons
were obviously not aware that voltage is always
measured in parallel to the lamp. We could no
longer observe this error in the post-test.
The cases in which the electric current was
measured in parallel in the pre-test (cf gures 2
and 11), no longer occurred in the post-test. Three
July 2021 14 Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
Figure 16. Comparison of pre- (brown) and post-test (green) results (relative scale scores) divided by the three
components of our model of experimental competence with 95% CI. The relative scale score corresponds to the
percentage of maximum achievable points per component. This allows a comparison of the three components
which the students could achieve different absolute scores in.
Figure 17. Spider-web to summarise students’ experi-
mental competences using relative test scores in the pre-
and post-test (cf section 5.2).
of the 22 test persons did not use the digital mul-
timeters properly in the pre-test (cf gure 13).
These errors also did not recur in the post-test, e.g.
see test person P19A’s experimental setup in the
post-test (cf gure 20).
6. Discussion and conclusion
Physics teachers are often faced with the chal-
lenge of having to set up difcult experiments,
Figure 18. Screenshot of test person P25A’s sample
sketch in the post-test. The semicircular block is now
correctly placed on the angle-measuring disc, unlike in
the pre-test, cf gure 8.
Figure 19. Screenshot of test person P31A’s sample
sketch in the post-test. The voltage is now measured in
parallel to the lamp, unlike in the pre-test (cf gure 10).
which sometimes do not work as planned, or
they have to consciously manipulate parameters
July 2021 15 Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
in order to be able to demonstrate a phenomenon
convincingly. This requires experimental com-
petences, especially with regard to basic experi-
mental techniques, as they are repeatedly essential
in various elds. In order to foster experimental
competences of prospective physics teachers, in
this article we presented a new seminar concept.
We conducted a rst pilot study to evaluate the
seminar. Therefore, we used a test instrument
from physics education research for the assess-
ment of experimental competences of prospective
physics teachers pre and post the seminar. Res-
ults from the pre-test underlined that a subset of
prospective physics teachers have a lack of fun-
damental experimental techniques despite form-
ally successful participation in standard laborat-
ory courses and thus, justify the new suggested
seminar concept for teacher training.
The created spider-web enables the descrip-
tion of the seminar participants’ experimental
competences and their development during our
seminar (cf gure 17). On the basis of this
spider-web it becomes clear that, as expec-
ted, the seminar contributes to the acquisition
of basic experimental techniques that enable
future physics teachers to conduct experiments
in physics classrooms. However, further support
is needed with regard to the students’ ability to
correctly process measured data and to correctly
interpret experimental results. Therefore, for the
future, we are considering an expansion of the
seminar with the aim of also promoting the stu-
dent teachers’ competences in data analysis that
are necessary for classroom practice.
But also for the individual students the use
of the spider-webs proves to be fruitful for the
diagnosis of experimental competence. The trans-
ition of the students’ competency network at the
beginning of the seminar to that from the post-
test point in time allows for a targeted examin-
ation of the learning outcome with regard to the
individual subcompetences. In this way, students
can be given a concrete recommendation for their
further learning process. We did this for all par-
ticipants in our pilot study and we conclude by
giving two examples. Student P17A, for example,
acquired less than 50% of the achievable points in
all subcompetences in the pre-test (cf gure 21).
In the post-test, he increased to over 75% of the
achievable points in the subskills designing a test
Figure 20. Screenshot of test person P19A’s sample
sketch in the post-test using the COM-socket and the
correct measuring range.
Figure 21. Spider-web to summarise P17A’s experi-
mental competences using relative test scores in the pre-
and post-test (cf section 5.2).
plan and assemble devices & build up experiment.
With regard to the competences relating to prepar-
ing, evaluating and analysing data, however, the
participant remained in the range up to about 50%
of the achievable points.
In contrast to P17A, student P12A repres-
ents an example of a subject for whom signic-
ant learning gains were observed between pre- and
post-test collection with respect to all sub-areas of
experimental competence (cf gure 22).
In summary, based on the results of the
post-test, we nd a signicant learning gain with
July 2021 16 Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
Figure 22. Spider-web to summarise P12A’s experi-
mental competences using relative test scores in the pre-
and post-test (cf section 5.2).
respect to the experimental competences of the
seminar participants. This can be interpreted as
a rst empirical evidence for the learning effect-
iveness of our new seminar concept. However,
despite the promising results of the pilot study
presented in this article, further research is needed
to tackle limitations: More target group-specic
instruments need to be used to assess experimental
competences in order to conduct more detailed
studies with larger samples. In addition, we want
to use qualitative methods to investigate in more
detail what students see as the causes of their dif-
culties with fundamental experimental techniques
and which elements of the seminar presented here
specically contribute to an increase in learning
in their eyes. A comparison of results from such
interviews with previous ndings from the liter-
ature (cf section 2) can contribute to a systematic
renement of our seminar concept. This will be
part of further research.
Data availability statement
The data that support the ndings of this study
are available upon reasonable request from the
authors.
Acknowledgments
We would like to thank Heike Theyßen (Univer-
sity of Duisburg-Essen) for providing us with the
test instrument for the assessment of experimental
competences that was developed by her within
the joint project MeK-LSA, together with Horst
Schecker (University of Bremen) and Knut Neu-
mann (IPN Kiel). We also thank the participants
of our pilot study. We would also like to thank
the referees for their constructive feedback, which
helped to improve the manuscript.
Ethical statement
Authors acknowledge that the research was con-
ducted anonymously, that consent was obtained
from all identiable participants, and that all iden-
tiable participants are informed about the public-
ation of the results of this study.
The data that support the ndings of this study
are available upon request from the author.
ORCID iDs
Philipp Bitzenbauer https://orcid.org/0000-
0001-5493-291X
Jan-Peter Meyn https://orcid.org/0000-0001-
8036-491X
Received 19 January 2021, in nal form 24 March 2021
Accepted for publication 30 April 2021
https://doi.org/10.1088/1361-6552/abfd3f
References
[1] Hart C, Mulhall P, Berry A, Loughran J and
Gunstone R 2000 J. Res. Sci. Teach. 37
655-675
[2] Börlin J 2012 Das Experiment als
Lerngelegenheit. Vom Interkulturellen
Vergleich des Physikunterrichts zu
Merkmalen Seiner Qualität
(Berlin: Logos)
[3] Kalthoff B, Theyßen H and Schreiber N 2018
Int. J. Sci. Educ. 40 1305-1326
[4] Baumert J and Kunter M 2006 Zeitschrift für
Erziehungswissenschaft 9469–520
[5] Baumert J, Kunter M, Blum W, Brunner M, Voss
T and Jodan A 2010 Am. Educ. Res. J.
47 133–80
[6] Cauet E, Liepert S, Kirschner S, Borowski A and
Fischer H E 2015 Revue Suisse des sciences
de l’éducation 37 462–79
[7] Kulgemeyer C and Riese J 2018 J. Res. Sci.
Teach. 55 1393–418
[8] Barendsen E and Henze I 2019 Res. Sci. Educ.
49 1141–75
[9] Baxter J A and Lederman N G 1999 Assessment
and measurement of pedagogical content
knowledge Examining Pedagogical Content
Knowledge (Dordrecht: Kluwer)
pp 147–61
[10] AAPT 2014 AAPT Recommendations for the
Undergraduate Physics Laboratory
July 2021 17 Phys. Educ. 56 (2021) 045020

P Bitzenbauer and J-P Meyn
Curriculum American Association of
Physics Teachers www.aapt.org/
resources/upload/labguidlinesdocument_
ebendorsed_nov10.pdf
[11] Holmes N G and Smith E M 2019 Phys. Teach.
57 296–9
[12] Holmes N G and Wieman C E 2018 Phys. Today
71 38–45
[13] McDermott L C and Redish E F 1999 Am. J.
Phys. 67 755–67
[14] Wieman C E and Holmes N G 2015 Am. J. Phys.
83 972–8
[15] Wieman C E 2015 Phys. Teach. 53 349–51
[16] Zwickl B M, Hu D, Finkelstein N and
Lewandowski H J 2015 Phys. Rev. ST. Phys.
Educ. Res. 1020113
[17] Holmes N G and Wieman C E 2016 Phys. Rev.
Phys. Educ. Res. 12 020103
[18] Holmes N G, Olsen J, Thomas J L and Wieman
C E 2017 Phys. Rev. Phys. Educ. Res.
13 010129
[19] Wilcox B R and Lewandowski H J 2017 Phys.
Rev. Phys. Educ. Res. 13 020110
[20] Smith E M, Stein M M, Walsh C and Holmes N
G 2020 Phys. Rev. X10 011029
[21] Smith E M and Holmes N G 2020 Phys. Rev.
Phys. Educ. Res. 16 020150
[22] Welzel M, Haller K, Bandera M, Hammelev D,
Koumaras P, Niedderer H, Paulsen A,
Robinault K and Aufschnaiter S 1998.
Didaktik Naturwissenschaften 429–44
[23] Usembayeva I, Ramankulov S, Berdi D,
Saparbekova G and Ualikhanova B 2015 Am.
J. Appl. Sci. 12 636–43
[24] Edmondson K M and Novak J D 1993 J. Res.
Sci. Teach. 30 547–59
[25] Berry A, Mulhall P, Loughran J J and Gunstone
R F 1999 Australian Sci. Teach. J. 45 27–31
[26] Kurbanbekov B A, Turmambekov T A, Baizak U
A, Saidakhmetov P A, Abdraimov R T,
Bekayeva A E and Orazbayeva K O 2016
Int. J. Environ. Sci. Edu. 11 13069–78
[27] Holmes N G, Wieman C and Bonn D A 2015
Proc. Natl Acad. Sci. USA 112 11199–204
[28] Hofstein L and Kind P M 2012 Learning In and
From Science Laboratories (Dordrecht:
Springer) pp 189–207
[29] Eickhorst B, Dickmann M, Schecker H, Theyßen
H and Neumann K 2015 Messung
experimenteller Kompetenz im Large-Scale:
Bewertung experimenteller Aufgaben
Heterogenität und Diversität - Vielfalt der
Voraussetzungen im Naturwissenschaftlichen
Unterricht (Kiel: IPN) pp 169–71
[30] Hammann M 2004 Math Naturwiss. Unterr.
57 196–203
[31] Emden M and Sumeth E 2016 Int. J. Sci. Math.
Educ. 14 29-54
[32] Klahr D 2002 Exploring Science: The Cognition
and Development of Discovery Processes
(Cambridge, MA: The MIT Press)
[33] Theyßen H, Schecker H, Neumann K, Eickhorst
B and Dickmann M 2016 Physik und
Didaktik in Schule und Hochschule
1/15 26–48
[34] Metzger S, Gut C, Hild P and Tardent J 2014
Modelling and assessing experimental
competence: an interdisciplinary progress
model for hands-on assessments E-Book
Proc. ESERA 2013th Conf.: Science
Education Research for Evidence-Based
Teaching and Coherence in Learning
(European Science Education Research
Association)
[35] Schreiber N, Theyßen H and Schecker H 2009
Physik und Didaktik in Schule und
Hochschule 892–101
[36] Schreiber N, Theyßen H and Schecker H 2012
Experimental competencies in science: a
comparison of assessment tools E- Book
Proc. ESERA 2011th Conf.: Science
Learning and Citizenship (European
Science Education Research Association)
pp 66–72
[37] Dickmann M, Eickhorst B, Theyßen H,
Neumann K, Schecker H and Schreiber N
2014 Measuring experimental skills in
large-scale assessments: developing a
simulation-based test instrument E-Book
Proc. ESERA 2013th Conf.: Science
Education Research for Evidence-Based
Teaching and Coherence in Learning
(European Science Education Research
Association)
[38] Dickmann M 206 Messung von
Experimentierfaehigkeiten -
Validierungsstudien zur Qualität Eines
Computerbasierten Testverfahrens (Berlin:
Logos) [Measurement of experimental
skills—Validation studies on the quality of a
computer-based test procedure]
[39] Schecker H, Neumann K, Theyßen H,
Eickhorst B and Dickmann M 2016 Z.
Didaktik Naturwissenschaften
22 197–213
[40] Baxter G P and Shavelson R J 1994 Int. J. Educ.
Res. 21 233–350
[41] Shavelson R J, Ruiz-Primo M A and Wiley E W
1999 J. Educ. Meas. 36 61–71
[42] Stecher B M, Klein S P, Solano-Flores G,
McCaffrey D, Robyn A and Shavelson R J
2000 Appl. Meas. Educ. 13 139–160
July 2021 18 Phys. Educ. 56 (2021) 045020

Fostering experimental competences of prospective physics teachers
[43] Kultusministerkonferenz 2019
Ländergemeinsame inhaltliche
Anforderungen für die Fachwissenschaften
und Fachdidaktiken in der Lehrerbildung
www.kmk.org/leadmin/Dateien/
veroeffentlichungen_beschluesse/2008/
2008_10_16-Fachprole-Lehrerbildung.pdf
(opened 3 February 2021)
[44] Hemphill J 2003 Am. Psychol. 58 78–9
[45] Maiseyenka V, Schecker H and Nawrath D 2013
Physik und Didaktik in Schule und
Hochschule 11–17
[46] Haagen-Schützenhöfer C and Joham B 2018
CEPS J. 89–34
[47] Kline T J B 2005 Psychological Testing. A
Practical Approach to Design and
Evaluation (Thousand Oaks, CA: Sage)
[48] Cohen R, Eylon B and Ganiel U 1983 Am. J.
Phys. 51 407–12
[49] Shipstone D M 1984 Eur. J. Sci. Educ.
6185–98
[50] Shipstone D M, Rhöneck C v, Jung W, Kärrqvist
C, Dupin J J, Johsua S and Licht P 1988 Int.
J. Sci. Educ. 10 303–16
Philipp Bitzenbauer holds a
PhD in Physics Education from
the University of Erlangen. His
research interests include classroom
experimentation, the empirical
investigation of learning processes
in quantum physics and teaching
quantum technologies for future
workforce. In addition to his research
activities, he works as a secondary
school teacher in physics.
Jan-Peter Meyn has a diploma and
PhD from the University of Hamburg.
After a decade of research in laser
physics, he became a physics teacher
at a secondary school. Since 2005 he
is Professor of Physics Education at
the University of Erlangen-Nürnberg.
Photography: Glasow.
July 2021 19 Phys. Educ. 56 (2021) 045020