1389

WORKING GROUP 8
CERME 8 (2013)

USING CONTENT ANALYSIS TO INVESTIGATE STUDENT
TEACHERS' BELIEFS ABOUT PUPILS
Reidar Mosvold, Janne Fauskanger, Raymond Bjuland & Arne Jakobsen
University oI Stavanger

Student teachers beliefs about teaching are arguablv important to their learning
and development as teachers. One aspect of these beliefs is that of a focus on the
learners. In the studv presented in this paper, we use two approaches to content
analvsis in order to learn more about student teachers beliefs from transcripts of
focus-group discussions. The analvses reveal that the student teachers focus less on
the mathematical content and more on characteristics of pupils in their discussions.
These characteristics were mainlv related to the pupils behaviour, their cultural
background or their certain needs.
Keywords: student teachers` belieIs, learner orientation, content analysis
INTRODUCTION
Teacher education is important in many respects, among others to equip all teachers
Ior eIIective teaching in the 21
st
century (OECD, 2011). Throughout their
developmentIrom student teachers to experienced teachersthey are Iaced with
diIIerent challenges. Richardson and Placier (2001) conclude that novice teachers
Iocus more on surviving in the classroom than on pupils` learning, and additional
research is needed in order to learn more about how teacher education can contribute
to a shiIt in Iocus in this respect.
There are several interesting issues to Iocus on in a study oI student teachers`
developmente.g. their belieIs about teaching (Fives & Buehl, 2010). Many
consider student teachers` belieIs important, and there are diIIerent reasons Ior this.
One reason is that student teachers bring with them strong belieIs about teaching into
their teacher education, and these belieIs are important in relation to what the student
teachers learn (Richardson, 2003). One conclusion Irom the Teacher Education and
Development Study Mathematics (TEDS-M) is that: 'signiIicant change is unlikely
to occur unless teacher-preparation programmes explicitly address belieIs (Tatto et
al., 2012, p. 172).
Previous research has identiIied learner orientation as one oI the main attributes oI a
high-quality learning environment (BransIord, Brown, & Cocking, 2000). The
demands on teachers are increasing, and cultural awareness and regard Ior individual
diIIerences and needs are crucial aspects oI a teacher`s skills (ibid.). It is important to
develop a Iocus on learners` attitudes, skills and understanding, and not least to use
this actively when designing and implementing teaching.
1390
WORKING GROUP 8
CERME 8 (2013)

BelieIs have been studied in mathematics education research Ior decades, and there
has been a development oI theories, Iocus and methods in relation to this area oI
research (see e.g., Philipp, 2007; Fives & Buehl, 2012). There are still, however,
unanswered questions regarding belieIs and other aspects oI the aIIective domain
(see e.g. Hannula, 2011). In the Iollowing, we take a brieI look at some oI the main
issues regarding concepts and categories that have been discussed in belieIs research.
In his handbook chapter, Philipp (2007) provides an overview oI some oI the more
commonly used terms related to belieIs: aIIect (including emotions, attitudes and
belieIs), belieIs systems, conceptions, identity, knowledge and values. All oI these
concepts have been used with various meanings by diIIerent researchers, and,
according to Fives and Buehl (2012, p. 471), 'the lack oI cohesion and clear
deIinitions has limited the explanatory and predictive potential oI teachers` belieIs.
A clariIication oI terminology is thereIore important Ior determining research Iocus.
Beswick (2012) provides a categorization oI mathematics teachers` belieIs into
belieIs about the nature oI mathematics, mathematics teaching and mathematics
learning (Table 1).
BelieIs about the nature
oI mathematics (Ernest,
1989)
BelieIs about
mathematics teaching
(Van Zoest, Jones, &
Thornton, 1994)
BelieIs about
mathematics learning
(Ernest, 1989)
Instrumentalist
Content Iocused with an
emphasis on perIormance
Skill, mastery, passive
reception oI knowledge
Platonist
Content Iocused with an
emphasis on
understanding
Active construction oI
understanding
Problem solving Learner Iocused
Autonomous exploration
oI own interests
Table 1: Categories of teachers` beliefs (Beswick, 2012, p. 130)
The choice oI methodology when studying belieIs is strongly connected with the
researchers` view about the nature oI these belieIs. For the purpose oI this study, we
consider belieIs as being held by individuals (Philipp, 2007), and all the aspects Irom
the table above are relevant to consider in studies oI teachers` belieIs. We analyse
data Irom two semi-structured Iocus-group interviews with student teachers in order
to answer the Iollowing research question:
What can be learned about student teachers` belieIs about pupils Irom content
analysis oI their Iocused discussions prior to Iield practice?
1391
WORKING GROUP 8
CERME 8 (2013)

The data material that we analyse and discuss in our attempts to answer this question
is part oI a larger, cross-disciplinary project entitled Teachers as Students (TasS).
METHODS
The TasS projectwhich is Iunded by the Research Council oI Norway (project
number: 212276/H20)has a main Iocus on the Iield practice, which is part oI the
teacher education programme. It includes two data collection periods and two groups
oI participants: a control group (reIerred to in the project as the 'business as usual
condition) and an intervention group (reIerred to as the 'lesson study approach
condition). Data collection in the control group was carried out in the spring oI
2012, whereas the data collection Ior the intervention group is scheduled Ior the
spring oI 2013. For the purpose oI this paper, we Iocus on the control group only.
Norwegian student teachers have 20 weeks oI Iield practice during their Iour-year
teacher education programme. For this research project, students in their second year
participated. These students were just about to start their Iourth three weeks long
Iield practice. Two groups oI student teachers Irom each oI the subject areas:
mathematics, science, English as a Ioreign language, and physical education were
selected Ior participation in the control group condition. The intention was to have
Iour student teachers in each group, but some oI the groups ended up with only three
student teachers in them. A Iocus-group interview (FGI) was carried out beIore (pre-
FGI) and aIter (post-FGI) the Iield practice period in each oI the eight groups. A
main purpose with the FGIs was to investigate the student teachers` reIlections about
the mathematical content (or one oI the other content areas involved in the study).
Data collection also included video observations Irom student teachers` planning
lessons with their practice teacher (pre-tutoring sessions), Irom carrying out lessons
(two lessons in each group) and Irom evaluating lessons (post-tutoring sessions). In
this paper, we analyse transcripts Irom the pre-FGIs conducted with the two groups
oI student teachers in mathematics (3 4 student teachers in these two groups).
Instrument
All pre-FGIs were carried out in a similar manner, and they were scheduled to last
Ior approximately an hour and a halI. Each interview consisted oI the same Iour
parts. The Iirst part (5-10 minutes) contained introductory questions related to why
they chose to enter into teacher education, why they decided to study mathematics,
and what they anticipated as being the most interesting and/or challenging parts oI
being a teacher.
The second part oI the FGIs (10-15 minutes) had a Iocus on the Iield practice they
were about to start, and what reIlections they had about their preparation Ior this
period. The student teachers were asked to reIlect on issues related to pupils, their
own background knowledge and skills in the subject area and practical inIormation
given Irom the university.
1392
WORKING GROUP 8
CERME 8 (2013)

Next Iollowed a part oI the interview where they were presented with the Iollowing
case Irom a classroom situation (this part lasted Ior about halI an hour):
Figure 1: Case from the interview guide.
They were then asked to discuss the challenges oI the situation presented in the case,
and how these challenges were related to the diIIerent aspects oI teacher knowledge.
In the Iourth and Iinal part oI the interviews (lasting approximately 30 minutes), the
student teachers were asked questions related to the subject area that they were
Iocusing on in their Iield practice (e.g. mathematics). In this part, they were asked
questions about how they would plan a lessonwith a Iocus on Iractionsin the
class that was presented in the case Irom the previous part oI the interview. The
questions in this part were concerned with important aspects in the planning process,
diIIiculties they believed students would have, ways oI organising teaching in such a
case, special concerns that they would have to be aware oI as teachers, reasonable
goals Ior such a lesson, and how they would assess their students in relation to the
set goals.
Data analysis
The transcripts Irom the FGIs have been analysed by combining two approaches to
content analysis. First, the transcripts were analysed by counting words in the text
with the purpose oI understanding the contextual use oI words. This approach is
oIten part oI what has been reIerred to as summative content analysis (Hsieh &
Shannon, 2005). Our summative analysis went beyond word count, however, and we
tried to discover underlying meanings oI the words used by the student teachers. The
counting was used to identiIy patterns in the data and to initiate the process oI
developing contextualised codes. The word count was also a starting point Ior
identiIying the context associated with the word to try to discover the range oI
meanings the word had in the interviews.
The summative analysis thus provided insight into how particular words were used.
This summative approach is, however, 'limited by the inattention to the broader
1393
WORKING GROUP 8
CERME 8 (2013)

meaning in the data (Hsieh & Shannon, 2005, p. 1285), and it was Iollowed by a
second phase oI data analysis. In this phase, we used conventional content analysis
to dig deeper into the data. According to Hsieh and Shannon (2005), this analysis is
used in studies which aim at describing a phenomenon in order to understand it
betterin this case 'it reIers to the student teachers` statements. In conventional
content analysis 'researchers immerse themselves into the data to allow new insights
to emerge (Hsieh & Shannon, 2005, p. 1279) by reading the data word by word. In
our study, we combined the use oI these two common approaches to content analysis.
The summative content analysis thus served as a way oI reducing data, and it was
then Iollowed by the conventional content analysis where categories were developed
as part oI the analysis.
RESULTS AND ANALYSIS
In our attempt to study what can be learned about student teachers` belieIs Irom
content analysis oI their Iocused discussions prior to Iield practice, our combination
oI summative and conventional content analysis seems promising. In this part we
will present results Irom the summative as well as the conventional content analysis,
and we will discuss the results in light oI previous research in this area.
The summative content analysis
AIter an initial reading oI the interview transcripts, we realised that the student
teachers` reIlections had much less Iocus on the mathematical content than we Iirst
anticipated. This impression was strengthened by our initial attempts to analyse the
data. As part oI our summative content analysis oI the transcripts, we generated a
concordance in order to learn more about the words that were used in the discussions
and the Irequencies oI those words. We started out with a Iocus on words related to
mathematics.
Frequency Word/concept
88 mathematics
88 Iraction
30 task
26 addition
21 calculate
18
denominator/common
denominator
Table 1. The most frequently used words related to mathematics
1394
WORKING GROUP 8
CERME 8 (2013)

From this initial summative content analysis, it appeared as iI the student teachers
were discussing Iractions quite a lot, and we started out by Iocusing our analysis on
this. When we moved Irom a quantitative to a more qualitative analysis oI the
content, however, it became evident that these words were mainly mentioned brieIly
with little or no reIlection. In one oI the interviews, Ior instance, the interviewer
asked about their preparation Ior the Iield practice. One oI the student teachers
Iollowed up by saying that they were well prepared. When the interviewer asked
about the mathematical topic they were Iocusing on, the student teacher replied: 'It is
algebra. The reIlections did not go deeper than just stating this.
We then decided to make a more open analysis oI the content in order to learn more
about what the student teachers were actually discussing in the FGIs. When
generating a concordance that was sorted aIter occurrences oI words rather than an
alphabetically sorted list oI the words that occurred, we observed that the most
Irequently occurring words were general words.
Frequency
Word (in
Norwegian)
In English
1702 det it
963 og and
784 er am/is/are
737 de they/them/those
674 s so
Table 2. The most frequently occurring words in the transcripts
Most oI these words (e.g. like 'and, 'is and 'it) were used in diIIerent ways, and
it was hard to discover any patterns. When making a more qualitative summative
analysis oI the most Irequently used words, however, we realised that the word
'they was interesting. Throughout the transcripts, the word 'they was used almost
exclusively with reIerence to pupils. It is not surprising that the students used this
word to describe the pupils, and this result in itselI is probably inIluenced by the case
used. It was, however, through the results oI the content analysis that we became
aware oI this, and we decided to Iollow up with Iurther analysis to learn more about
the student teachers` belieIs about the pupilsas revealed by their use oI the word
'they.
The conventional content analysis
From the summative content analysis, we Iollowed up with conventional content
analysis. We started out with analysis oI the keyword ('they) in its immediate
1395
WORKING GROUP 8
CERME 8 (2013)

context. In the Iirst step oI this analysis, the context was deIined as Iive words beIore
and aIter the keyword. During the analysis oI these context units, the Iirst attempts to
categorise the data were made. Then we Iollowed up by increasing the context to
entire utterances. The transcripts were read word by word, and categories were
developed inductively (Hsieh & Shannon, 2005).
From this analysis, the Iollowing categories emerged in relation to how the student
teachers` reIerred to the pupils (as 'they):
General reIerence to pupils
Characteristics oI pupils`:
behaviour (B)
cultural background (CB)
certain needs (CN)
Below is an utterance Irom one oI the student teachers in the two interviews. It has
been chosen Ior illustration since it contains uses oI the keyword 'they that relate to
all the three subcategories oI 'characteristics oI pupils. It also contains an example
oI other uses oI the keyword (O).
ST: It is oI course possible to provide assistance with homework. It is Iair
enough that all pupils have a right to receive such assistance. You can, I
mean, iI you have more than two minority-language |pupils|, you have a
right to get a teacher assistant. |You| can help them (CB) in this way. By
doing that you can split your class. You get those (B) who are diligent, that
they (B) might be seated in a group oI their own and work with things they
(B) are able to |do on their own|, so that they (B) are always ahead oI the
diIIicult tasks while you still help those (CN) who are struggling. That you
have those (O) two assistants who walk around and help, and... Yes. It
depends on how much learning diIIiculties they (CN) have, those (CN) two
other pupils, but this is where you have the opportunity to get in some extra
help.
As we can see Irom the Iirst Iew sentences in this excerpt, the student teacher is
talking about the challenges oI having pupils with a minority language in the
classroom. (S)he points to a regulation concerning the right to receive help with
homework (Ior the minority pupils), and another regulation concerning the right to
have a teacher assistant in the classroom iI there are more than two pupils with a
minority language background in the class. When a teacher assistant is present to
take care oI the minority-language pupils, the teacher can more easily split the class
into groups. This student teacher is obviously in Iavour oI separating the pupils in
groups according to level. It is interesting to observe that the best pupils are reIerred
to as diligent rather than smart, clever or high achieving. From this statement, it is
1396
WORKING GROUP 8
CERME 8 (2013)

possible to suggest that this particular student teacher holds a belieI where high
achievement in mathematics is related to eIIort.
The weaker pupils, on the other hand, are reIerred to as 'struggling, and the student
teacher Iollows up by talking about them as pupils with a certain level oI learning
diIIiculties. So, whereas the best pupils are characterised by their eIIort, the lower
achieving pupils seem to be characterised as pupils with learning diIIiculties, pupils
who have certain needs, and pupils who are entitled to receive extra (external)
support.
CONCLUDING DISCUSSION
In their distinction between three important aspects in research on belieIs about
teaching, Van Zoest, Jones and Thornton (1994) mentioned learner Iocused as a third
aspect. We have seen in the results Irom our analyses that the student teachers, when
reIlecting about teaching, had a strong Iocus on the pupils. It seems, however, that
the student teacherswhen talking about the pupils in our FGIsdid not Iocus
much on mathematics, teaching or the pupils as learners; they Iocused more on
characteristics oI pupils. In this way, the student teachers` belieIsalthough they
were arguably about teachingsomehow did not Iit inside oI Beswick`s (2012) table
(see Table 1). The student teachers did Iocus on the learners, but they seemed to
Iocus more on the pupils as individuals with diIIerent characteristics.
Our aim with applying content analysis to the transcripts oI the FGIs was to learn
more about student teachers` belieIs. An underlying assumption then, is that there is
a connection between their belieIs and the words they use. AIter having made some
early analyses, it appeared that reIlections about the mathematical content were not
dominating in the student teachers` discussions. Mathematical concepts were used,
but they were mainly reIerred to rather brieIly. When applying summative content
analysis to the data material, we Iound that the FGIs contained more reIlections
about pupils than on mathematical content. The word 'theywhich was mostly
used with reIerence to pupilswere among the most Irequently used words in the
FGIs (table 2). This is not surprising as the discussions were obviously inIluenced by
the case that was presented to the student teachers in the interviews. AIter having
made even Iurther analyses oI the data material, using conventional content analysis,
we learned more about the diIIerent aspects oI these reIlections about pupils.
When analysing their characterisations oI the pupils, we could distinguish between
reIlections concerning the pupils` behaviour, their cultural background or certain
needs that the pupils might have. The student teachers, in their reIlections, seemed to
believe that the responsibility Ior pupils` learning is to be placed outside the teacher.
They related the possibilities to help the pupils to other Iactors than their own
competence. One example oI this can be Iound in the excerpt Irom the transcripts
that we have discussed above. In this excerpt, the student teacher suggests that the
pupils would learn iI there were enough teacher assistants. Their Iocus here can be
1397
WORKING GROUP 8
CERME 8 (2013)

interpreted as being on surviving in the classroom rather than on pupils` learning
(Richardson & Placier, 2001). This is, oI course, only one example, but we Iound the
same tendency in our analysis oI the entire data material. In both interviews, there
seemed to be a belieI that the pupils represented diIIerent kinds oI challenges to the
teacher, and the student teachers might suggest that a solution to these challenges
could be Iound outside and not inside themselves. Skott (2001) argued that teachers`
belieIs about mathematics teaching are oIten obscured by the more general priorities
related to organising the classroom, and this might be part oI the explanation Ior the
seeming lack oI Iocus on mathematics and mathematics teaching in our FGIs.
Much can be learned about student teachers` belieIs by analysing their reIlections in
Iocused discussions. We decided to use a combination oI summative and
conventional content analysis in our analyses oI data Irom FGIs with student
teachers prior to Iield practice, but there are other kinds oI analyses to makeeach
with diIIerent advantages as well as disadvantages or limitations. One oI the
advantages oI using content analysis in our study was that it helped us discover what
the student teachers mainly Iocused on in their discussions, and we could then use
this in Iurther analyses to learn more about their belieIs. The combination oI
approaches to content analysis also made it possible Ior us to discover reIlections
that were made by using a common word like 'they. In order to learn more about
student teachers` belieIs, it is also relevant to use the results oI these analysesas
well as other analysesto make a new evaluation oI the interview guide. When
looking back at the casewhich was presented in the FGIsit can be noticed that it
was more Iocused on students than on mathematics. The case most certainly had an
eIIect on the participants` reIlections in the FGIs.
REFERENCES
Beswick, K. (2012). Teachers` belieIs about school mathematics and
mathematicians` mathematics and their relationship to practice. Educational
Studies in Mathematics, 79(1), 127-147.
BransIord, J. D., Brown, A. L., & Cocking, R. R. (2000). How People Learn.
Washington D.C: National Academy Press.
Ernest, P. (1989). The impact oI belieIs on the teaching oI mathematics. In P. Ernest
(Ed.), Mathematics teaching. The state of the art (pp. 249-253). New York, NY:
Falmer.
Fives, H., & Buehl, M. M. (2010). Teachers` articulation oI belieIs about teaching
knowledge: conceptualizing a belieI Iramework. In L. D. Bendixen & F. C. Feucht
(Eds.), Personal epistemologv in the classroom (pp. 470-515). New York, NY:
Cambridge University Press.
Fives, H., & Buehl, M. M. (2012). Spring cleaning Ior the 'messy construct oI
teachers belieIs: What are they? Which have been examined? What can they tell
1398
WORKING GROUP 8
CERME 8 (2013)

us? In K. R. Harris, S. Graham & T. Urdan (Eds.), APA Educational Psvchologv
Handbook (Vol. 2, pp. 471-499). Washington, DC: APA.
Hannula, M. S. (2011). The structure and dynamics oI aIIect in mathematical
thinking and learning. In M. Pytlak, T. Rowland, & E. Swoboda (Eds.),
Proceedings of the Seventh Congress of the European Societv for Research in
Mathematics Education (pp. 34-60). University oI Rzeszow, Poland.
Hsieh, H.-F., & Shannon, S. E. (2005). Three approaches to qualitative content
analysis. Qualitative Health Research, 15(9), 1277-1288.
OECD (2011). Building a high-qualitv teaching profession. Lessons from around the
world. Retrieved September 11, 2012, Irom
http://www2.ed.gov/about/inits/ed/internationaled/background.pdI
Philipp, R. A. (2007). Mathematics teachers` belieIs and aIIect. In F. K. Lester (Ed.),
Second Handbook of Research on Mathematics Teaching and Learning (pp. 257-
315). Charlotte, NC: InIormation Age Publishing.
Richardson, V. (2003). Preservice teachers` belieIs. In J. Raths & A. McAninch
(Eds.), Teacher beliefs and classroom performance. The impact on teacher
education (pp. 1-22). Greenwich, CT: InIormation Age Publishing.
Richardson, V., & Placier, P. (2001). Chapter 41: Teacher change. In V. Richardson
(Ed.), Handbook of research on teaching (4th ed.) (pp. 905-947). Washington,
DC: American Educational Research Association.
Skott, J. (2001). The emerging practices oI a novice teacher: The roles oI his school
mathematics images. Journal of Mathematics Teacher Education, 4(1), 3-28.
Tatto, M. T., Peck, R., Schwille, J., Bankov, K., Senk, S. S., Rodriguez, M.,
Ingvarson, L., Reckase, M., & Rowley, G. (2012). Policv, Practice, and Readiness
to Teach Primarv and Secondarv Mathematics in 17 Countries. Findings from the
IEA Teacher Education and Development Studv in Mathematics (TEDS-M).
Retrieved September 11, 2012, Irom
http://www.iea.nl/Iileadmin/userupload/Publications/Electronicversions/TEDS-
MInternationalReport.pdI
Van Zoest, L. R., Jones, G. A., & Thornton, C. A. (1994). BelieIs about mathematics
teaching held by pre-service teachers involved in a Iirst grade mentorship
program. Mathematics Education Research Journal, 6(1), 37-55.