Adams. G. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 35(2) June 2015
Contextual Examination of the Turkish Middle School Mathematics Teachers’
Exam Questions
Karadeniz, Mihriban1; Baran, Tuğba2; Gökçek, Tuba3 & Aydın Güç, Funda1
1
Giresun University (Turkey); 2Kırıkkale University (Turkey); 3Karadeniz Technical
University (Turkey)
The current study aims to present the distribution of the middle school
mathematics Turkish teachers' exam questions in terms of Bloom’s
cognitive process and knowledge dimension as well as the question types.
Additionally, it will prove if there is statistically difference on the
questions’ placement in the Revised Bloom’s taxonomy with the question
types. In the study, 10 middle school mathematics teachers’ exam
questions posed during the first semester of the 2013-2014 academic year
were analysed. A total of 77 exam papers were reviewed in the study and
the total of 1152 questions from these papers were examined separately. A
chi-square test was used to determine whether the cognitive process and
knowledge dimensions of the questions were statistically different by
question types. The results gained from the study reveal that mathematics
teachers usually prepare questions at the lower cognitive dimensions of
the Bloom Taxonomy. According to chi-square test results, there was a
significant difference between knowledge dimensions of the questions as
well as the question types. Besides, there was also a significant difference
between cognitive process dimensions of the questions and the question
types.
Keywords: mathematics exam; Bloom Taxonomy; question types
Introduction
The four key elements of curricula are “goal”, “content”, “teaching process”, and
“assessment and evaluation” respectively (Akpınar, 2003; Küçükahmet, 1997;
Ornstein and Hunkins, 1993). While goal, content, and teaching process guide
curricula, the assessment and evaluation process plays an important role in
determining whether a curriculum has achieved its goal.
In Turkey, students’ performance is usually assessed through country-wide
standard tests and written exams carried out by teachers. The nature of these
assessments is set forth in various regulations and principles. The regulation on
primary education institutions notes that the goals and acquisitions indicated in
curricula have to be taken as basis in the assessment and evaluation of success in
accordance with the general principles and the principles of assessment and
evaluation included in the curriculum of relevant course have to be followed (MEB,
2014). In addition, it is stated in this regulation that different question types should be
used in the exams carried out by teachers (MEB, 2014). As is seen, questions used in
standard tests and assessment activities carried out by teachers must be prepared in
such a way that they serve various purposes and involve different types and different
levels. In this respect, researchers have created various taxonomies to classify the
questions to be used in the educational process.
Some examples of these taxonomies are Bloom’s taxonomy, Barrett’s
From Informal Proceedings 35-2 (BSRLM) available at bsrlm.org.uk © the author - 25
, Adams. G. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 35(2) June 2015
taxonomy, and Pearson-Johnson taxonomy (Armbruster and Ostertag, 1989). Among
these Bloom’s taxonomy (BT) is considered the most useful in classifying educational
goals and question levels (Köğce and Baki, 2009). In 2001, Krathwohl et al.
rearranged Bloom’s taxonomy (Anderson and Krathwohl, 2001). They handled the
cognitive domain in two dimensions: “cognitive process” and “knowledge”. The
cognitive process dimension includes mental activities called remember, understand,
apply, analyze, evaluate, and create. The knowledge dimension, on the other hand,
contains four types: factual, conceptual, procedural, and metacognitive.
In this context, it can be said that assessment instruments designed for
evaluating students must contain questions involving different knowledge types and
different cognitive processes and the content validity of these instruments are also
important for educating students having high-level knowledge and skills. Research
reports reveal that students’ thinking levels depend on the levels of questions posed
by teachers (Brualdi, 1998).
A lot of researchers have examined the distribution of the question types in the
educational process (Koray, Altunçekiç, Yaman, 2005; Özcan and Oluk, 2007), in
course materials (Armbruster and Ostertag, 1989; Kablan, Baran and Hazer, 2013), in
central examinations (Güler, Özdemir and Dikici, 2012), and the exams carried out by
teachers (Ayvacı and Türkdoğan, 2010; Demircioğlu and Demircioğlu, 2009; Güler,
Özdemir and Dikici, 2012; Köğce and Baki, 2009; Tanık and Saraçoğlu, 2011) based
on BT of the cognitive domain. Research shows that the questions asked to students in
lessons, books, and central- local exams are at the low levels of the Cognitive Domain
of BT. In previous studies, questions are mostly examined according to the old
version of the BT, and thus the knowledge dimension is ignored by paying attention
to the cognitive dimension. Research built upon the Revised BT, on the other hand,
focuses on mostly cognitive processes. However, it is still unknown what knowledge
types are addressed most in the questions used in educational activities. On the other
hand, it is emphasized that the assessment and evaluation instruments used by
teachers in their lessons are connected to their expectations from their students as well
as goals (Gipps, Brown, McCallum, McAlister, 1995; Hill, 2000), it is not known
whether or not the question types they adopt (e.g. multiple-choice, short-answer, long-
answer, matching, true-false) are relevant to students’ cognitive levels and knowledge
types.
This study aims to determine whether or not the questions used in the written
exams at middle school (the 5th to the 8th grades) mathematics courses are connected
to the cognitive process dimension and the knowledge dimension of the Bloom’s
taxonomy based on the question types. For this reason, the following sub-questions
were answered in the research:
Do the cognitive levels of the exam questions differ by question type?
Do the knowledge types of the exam questions differ by question type?
Method
The exams prepared by middle school mathematics teachers were examined in terms
of the cognitive process dimension and the knowledge dimension of the BT by
question types through document review method.
Data Collection and Analysis
The data were collected from the exams prepared by middle school mathematics
teachers working in various provinces of Turkey in the first semester of the 2013-
From Informal Proceedings 35-2 (BSRLM) available at bsrlm.org.uk © the author - 26
Contextual Examination of the Turkish Middle School Mathematics Teachers’
Exam Questions
Karadeniz, Mihriban1; Baran, Tuğba2; Gökçek, Tuba3 & Aydın Güç, Funda1
1
Giresun University (Turkey); 2Kırıkkale University (Turkey); 3Karadeniz Technical
University (Turkey)
The current study aims to present the distribution of the middle school
mathematics Turkish teachers' exam questions in terms of Bloom’s
cognitive process and knowledge dimension as well as the question types.
Additionally, it will prove if there is statistically difference on the
questions’ placement in the Revised Bloom’s taxonomy with the question
types. In the study, 10 middle school mathematics teachers’ exam
questions posed during the first semester of the 2013-2014 academic year
were analysed. A total of 77 exam papers were reviewed in the study and
the total of 1152 questions from these papers were examined separately. A
chi-square test was used to determine whether the cognitive process and
knowledge dimensions of the questions were statistically different by
question types. The results gained from the study reveal that mathematics
teachers usually prepare questions at the lower cognitive dimensions of
the Bloom Taxonomy. According to chi-square test results, there was a
significant difference between knowledge dimensions of the questions as
well as the question types. Besides, there was also a significant difference
between cognitive process dimensions of the questions and the question
types.
Keywords: mathematics exam; Bloom Taxonomy; question types
Introduction
The four key elements of curricula are “goal”, “content”, “teaching process”, and
“assessment and evaluation” respectively (Akpınar, 2003; Küçükahmet, 1997;
Ornstein and Hunkins, 1993). While goal, content, and teaching process guide
curricula, the assessment and evaluation process plays an important role in
determining whether a curriculum has achieved its goal.
In Turkey, students’ performance is usually assessed through country-wide
standard tests and written exams carried out by teachers. The nature of these
assessments is set forth in various regulations and principles. The regulation on
primary education institutions notes that the goals and acquisitions indicated in
curricula have to be taken as basis in the assessment and evaluation of success in
accordance with the general principles and the principles of assessment and
evaluation included in the curriculum of relevant course have to be followed (MEB,
2014). In addition, it is stated in this regulation that different question types should be
used in the exams carried out by teachers (MEB, 2014). As is seen, questions used in
standard tests and assessment activities carried out by teachers must be prepared in
such a way that they serve various purposes and involve different types and different
levels. In this respect, researchers have created various taxonomies to classify the
questions to be used in the educational process.
Some examples of these taxonomies are Bloom’s taxonomy, Barrett’s
From Informal Proceedings 35-2 (BSRLM) available at bsrlm.org.uk © the author - 25
, Adams. G. (Ed.) Proceedings of the British Society for Research into Learning Mathematics 35(2) June 2015
taxonomy, and Pearson-Johnson taxonomy (Armbruster and Ostertag, 1989). Among
these Bloom’s taxonomy (BT) is considered the most useful in classifying educational
goals and question levels (Köğce and Baki, 2009). In 2001, Krathwohl et al.
rearranged Bloom’s taxonomy (Anderson and Krathwohl, 2001). They handled the
cognitive domain in two dimensions: “cognitive process” and “knowledge”. The
cognitive process dimension includes mental activities called remember, understand,
apply, analyze, evaluate, and create. The knowledge dimension, on the other hand,
contains four types: factual, conceptual, procedural, and metacognitive.
In this context, it can be said that assessment instruments designed for
evaluating students must contain questions involving different knowledge types and
different cognitive processes and the content validity of these instruments are also
important for educating students having high-level knowledge and skills. Research
reports reveal that students’ thinking levels depend on the levels of questions posed
by teachers (Brualdi, 1998).
A lot of researchers have examined the distribution of the question types in the
educational process (Koray, Altunçekiç, Yaman, 2005; Özcan and Oluk, 2007), in
course materials (Armbruster and Ostertag, 1989; Kablan, Baran and Hazer, 2013), in
central examinations (Güler, Özdemir and Dikici, 2012), and the exams carried out by
teachers (Ayvacı and Türkdoğan, 2010; Demircioğlu and Demircioğlu, 2009; Güler,
Özdemir and Dikici, 2012; Köğce and Baki, 2009; Tanık and Saraçoğlu, 2011) based
on BT of the cognitive domain. Research shows that the questions asked to students in
lessons, books, and central- local exams are at the low levels of the Cognitive Domain
of BT. In previous studies, questions are mostly examined according to the old
version of the BT, and thus the knowledge dimension is ignored by paying attention
to the cognitive dimension. Research built upon the Revised BT, on the other hand,
focuses on mostly cognitive processes. However, it is still unknown what knowledge
types are addressed most in the questions used in educational activities. On the other
hand, it is emphasized that the assessment and evaluation instruments used by
teachers in their lessons are connected to their expectations from their students as well
as goals (Gipps, Brown, McCallum, McAlister, 1995; Hill, 2000), it is not known
whether or not the question types they adopt (e.g. multiple-choice, short-answer, long-
answer, matching, true-false) are relevant to students’ cognitive levels and knowledge
types.
This study aims to determine whether or not the questions used in the written
exams at middle school (the 5th to the 8th grades) mathematics courses are connected
to the cognitive process dimension and the knowledge dimension of the Bloom’s
taxonomy based on the question types. For this reason, the following sub-questions
were answered in the research:
Do the cognitive levels of the exam questions differ by question type?
Do the knowledge types of the exam questions differ by question type?
Method
The exams prepared by middle school mathematics teachers were examined in terms
of the cognitive process dimension and the knowledge dimension of the BT by
question types through document review method.
Data Collection and Analysis
The data were collected from the exams prepared by middle school mathematics
teachers working in various provinces of Turkey in the first semester of the 2013-
From Informal Proceedings 35-2 (BSRLM) available at bsrlm.org.uk © the author - 26
