INTRODUCTION
Economics is a social
science subject that is concerned with the study of human behaviour
as a relationship between ends and scarce resources which have alternative
uses. It is the study of how the society manages scarce resources, such as
food, clothing and housing among others, and how human relates with these
scarce resources. It is also concerned with choice because humans are faced
with the problem of scarcity of resources with which to satisfy their wants and
hence they are forced to choose which want to satisfy first and which want to
satisfy last (Anyawuchi, 2008). The study of
economics enables an individual to understand how humans behave when the price
of a commodity is high and when it is low. The study of this subject is
critical if Nigeria is to find itself among the top 20 world economics in the
year 2030 Federal Republic of Nigeria (2008). Economics also helps in the
process of production distribution and consumption of scarce resources.
The economics
curriculum at the secondary school level is assessed using teacher-made test
and standardized examination such as West African Examination Council (WAEC)
and National Examination Council (NECO). Teacher-made test are constructed by
the teachers and they are used in conducting continuous assessment in schools
while the standardized examination are constructed by the examination bodies
and are used in conducting certificate examination for students who graduate
from the secondary schools. There is criticism that teacher-made test are
poorly constructed and that results obtained from such poorly constructed tests
are not valid and reliable (Wakjissa, 2010). There is
therefore the need to develop a quality teacher-made achievement test for used
in conducting continuous assessment of students in the subject so as to improve
students’ performance. Some reasons that could be held responsible for students’
poor performance in the subject include teachers’ poor knowledge of test
construction skills which results in poor quality teacher-made economics
achievement test EAT that are used in assessing students’ achievement in the
classroom, teachers’ attitude, students’ attitude, commitment and teacher’s
qualification among others.
Researchers Osabede (2013), Osabade (2014), Adedoyin and Adegoke (2014) have
shown that there is paucity of valid and reliable economics achievement test
for used in conducting continuous assessment in secondary schools with emphasis
on feedback. Achievement test serves two main purposes in schools; formative
and summative (Ugodulunwa, 2020). Formative
assessment is used during the course of teaching for monitoring students’
learning progress and providing feedback to teachers and students which is
referred to as assessment for learning. While summative assessment is the type
of assessment that is generally carried out at the end of a course or unit of
instruction and is regard as assessment for learning. Achievement test helps
teachers to identify students’ areas of strength and weakness in specific
content areas. It also helps teachers to grade students.
According to Dishe (2018) and Ugodulunwa (2020)
a good teacher-made test should be valid, reliable and objectively constructed.
This will enable the teacher to obtain a valid and reliable result. Imo (2011)
states that teacher-made test are normally prepared and administered for the
purpose of finding out classroom achievement of students but unfortunately,
most of the test are poorly constructed and a such results from such test are
not usually valid and reliable and hence the need for teachers to develop
quality tests for used in conducting continuous assessment of students.
Test development
refers to the procedure involved in designing of test items by teachers or
measurement experts for used in assessment (Amanda, 2014). It involves
assembling test items for use to determine achievement of students in a course
of study. Test development implies that some course contents area have been
taught to students and the teacher is interested in finding out the achievement
of students (Hamman-Tukur, Musa & Atsua, 2013). Construction of a valid and reliable test is
enhanced when appropriate procedures re followed. This implies that there are a
number of stages involved in developing a valid and reliable test before item
calibration. These stages include; the determination of the purpose of the
test, outlining the content area, development of a table of specification among
others. Others are selection of appropriate item format writing the test items
in line with the table of specification, field testing of items, item
development of test norms and development of test manual.
According to Wakjissa (2010) Obilor & Akpan (2020) must test that are developed by teachers
focused on lower levels of cognitive objectives such as knowledge and
comprehension and therefore students usually do well in such examination but
when confronted with items that are developed based on standard, such students
usually perform poorly. Hence, the need for the teachers to consider the level
of behavioural objectives in test development before
the calibration of such items for use in conducting continuous assessment of
students.
Achievement test are
designed based on three theoretical approaches: classical test theory (CTT), generalizability
theory (G-theory) and item-response theory (IRT). CTT has been the leading
framework for developing and analyzing test and examination in Nigeria with its
attendant advantages and disadvantages of being sample and test dependent (Hambleton & Jones, 2013). It is a theory that is
concerned with the relationship between observed score, true score and error
score of testees in a test.
Generalizability
theory (G-theory) is a statistical theory that is used in evaluating the
reliability or dependability of behavioural measure.
It is a theory that is concern with persons, test items and testing situation.
G-theory has it that the person administering the test, the test items and the
testing situations has the potentials to affect the dependability of a test
results. G-theory is an improvement on the CTT and hence the need for teachers
to adapt a modern measurement theory such as item-response theory in test
development process so as to develop quality test for used in conducting
continues assessment of students.
Item-response
theory is a modern measurement that is also used in developing achievement test
just like the CTT and G-theory. The theory is based on the assumptions of unidimensionality, local independence and item characteristic
curve. It is a theory that model, the interaction between students ability and
the item difficulty, discrimination and pseudo guessing (Chalmers (2013). The
theory holds that during testing, there is an encounter between testees and a test items and that the testees
must have a trait level sufficient enough to be able to answer the item
correctly. The focus of IRT is on the pattern of responses rather than the
total score of the students.
The
parameters of interest according to Amuche and Fan
(2014) are difficulty, discrimination guessing and dimensionality of the test.
Item difficulty or location parameter (b) is the amount of latent trait a testee must possess to be able to answer an item in a test
correctly. It is the probability that a testee will
be able to answer an item correctly if the student has the ability and
unfortunately most teachers develop most tests without putting into
consideration the difficulty parameters of the test items that are used in
conducting continuous assessment of students. Hence, there is the need to
consider the difficulty of test items before using the test for continuous
assessment.
Item
discrimination parameter (a) in test development using item-response theory
indicates how well test items differentiate between individuals of different
latent trait level and also indicates the differences between high and low
achievers in a test. Baker (2003) opined that most achievement test by teachers
do not take into consideration the need to have items that discriminates
between high and low achievers and the need to discard items that do not
discriminate well. Hence, there is the need for teachers to consider item
discrimination in the test items that are developed for the purpose of
continuous assessment.
The
pseudo guessing parameter (c) is another parameter of interest which suggest
that the respondent to a test item with a low trait level, my still have a
small chance of answering the item correctly. It assumes that in
multiple-choice test items and examinee who does not know the correct
alternative may succeed in responding correctly by random guessing. As observed
by Wambleton and Jones (2013) most test items that
are developed by teachers are prone to guessing. This is because appropriate
procedures are not usually followed in developing the test items. Hence, the
need for teachers to check the guessing parameters of their items before using
the items for assessment of students. Unidimensionality
of a test is based on the belief that a test is supposed to measure one trait
or dimension at a time but most test that are developed by teachers are usually
multidimensional, this is because the items in most test usually measure more
than one trait Li & Jiao (2012). A multidimensional test is a test that
measure two or more construct, ability, attribute or dimension or skills. Hence
when a test violate the assumption of unidimensionality
it means that the test is measuring more than one trait hence multidimensional
procedure needs to be applied in calibrating the items
Item
calibration is the process by which the parameters of test items are estimated
(Baker, 2008). It is the scaling of test items based on their difficulty,
discrimination and guessing parameter. It also involves the grouping of test
items based on the ability of testees. The goal of
item calibration is to develop a pool or bank of items which have the same
scale and are comparable with known standard. Ojerimde
Popoola, Ojo & Onyeneho (2012) found that most test items that are used
for continuous assessment are not usually calibrated before they are used for
the purpose of continuous assessment and hence the need for teachers to
calibrate test items before using the items for continuous assessment of
students.
There are
several empirical studies on development and calibration of achievement tests.
These include Ogbebor and Onuka
(2014), Danjuma (2014) Madu
(2014), Akwa (2014) and Adeleke
& Joshua (2015). While Ogbebor and Onuka (2014) study was on development and validation of
economics achievement attitude scale using exploratory factor analysis, Madu (2014) designed a study on the development and
validation of workshop-based process skills test. The two studies found that
the instruments developed were reliable. Similarly Madu
(2014) conducted a study on development and validation of survey achievement
test in Agricultural science which is also different from the present study
that seeks to develop and calibrate and economics achievement test for
secondary school students. Again, Akwa (2014) carried
out a study on development and standardization of achievement test in senior
secondary school mathematics and the study is different from the present study
that seeks to develop and calibrate an economics achievement test for SS II
economics students in Plateau State. Furthermore, Joshua and Adeleke (2015) carried out a study on the development and
validation of scientific literacy achievement test to assess senior secondary
school students’ literacy in physics which is also different from the present
work.
After
reviewing literature, two limitations could be summarized from these past
studies; the studies conducted are not in SS II economics and most of them used
classical test theory with it attendant short coming of sample and test
dependent. This therefore calls for a study on the development and IRT
calibration of economics achievement test in Plateau State, Nigeria. Therefore,
the broad question for this study is: what is the item parameter of the
economics achievement test developed and calibrated by the researcher in
Plateau State.
Research Questions
- What
is the validity of the Economics Achievement Test developed and calibrated
in Plateau State?
- What
is the reliability of the Economics Achievement Test developed and
calibrated in Plateau State Nigeria?
- What
is the estimate of the difficulty parameter (b) of the Economics
Achievement Test items in Plateau State Nigeria?
- What
is the estimate of the discrimination parameter (a) of the Economics
Achievement Test items in Plateau State Nigeria?
- What
is the estimate of pseudo guessing parameter (c) of the Economics
Achievement Test items in Plateau State Nigeria?
- What
is the unidimensionality of the dichotomously
scored Economics Achievement Test items in Plateau State Nigeria?
METHODOLOGY
The study
used instrumentation and cross-sectional survey research designs.
Instrumentation research design refers to the tool or means by which an
investigator attempts to measure the variables or items of interest in a data
collection process. It was used in developing and certifying the validity and
reliability of the economics achievement test while cross-sectional survey was
required in collecting data from the SS II students for the purpose of
generalizing the funding on the entire population of SSII. The population
consisted of all The 23712 SS II students in Plateau State made up of male and
female while a sample of 1454 SSII students made up of 720 males and 734 females
was used. Multistage sampling and proportional stratified sampling were used
for the study. Multistage and proportional stratified sampling are sampling
methods in which different strata in a population are identified and in which
the number of elements are drawn from each strata proportionate to the relative
number of elements in each stratum (Nwana, 2007). The
instrument used for data collection was the economics achievement test
developed and calibrated by the researcher. Content validity of the instrument
was established using table of content and by subjecting the instrument to
expert judgement from Economics Education and
Research, Measurement and Evaluation. The reliability of the instrument was
established using Omega Reliability Procedures. The difficulty indexes, discrimination
indexes guessing parameter indexes and stout’s test of essentiality of item
response theory were used in answering the research questions that were raised
to guide the study.
Research Question One
What is
the validity of the Economics Achievement Test?
The
content validity of the Economics Achievement test was established using the
test blue print based on the topics in the Nigerian senior secondary school
economics curriculum and the objectives of each topic as contained in the
curriculum. It was also subjected to experts’ scrutiny. The test was adjudged
to be valid based on the responses of the validators. Furthermore Kendell
coefficient of concordance was computed and it was found to be 87%
Research Question Two
What is the reliability of the
Economics Achievement Test?
The
reliability of the Economics Achievement Test was established using Omega
reliability procedure and a reliability coefficient of the dichotomous items
was 0.83. This show that the EAT was reliable.
Research Question Three
What is
the estimate of the difficulty parameter (b) of the Economics Achievement Test
items?
Table 4
shows the results of the analysis using IRTPRO cluster procedure of DETECT
statistic in DIMTEST and 30% of the responses of the testees
were tested to see if it was dimensionally distinct from the remaining items in
the test and the results show that unidimensionality
was not tenable, that is, the items that were found to form the secondary
dimension were dimensionally distinct from the remaining items of the test (t =
2.0906) (p-value <0.05, one-tailed); therefore, the assumption of unidimensionality was rejected. This shows that the pooled
65-item of the Economics Achievement Test violated the assumptions of unidimensionality. This means that the test was a
multidimensional test because the test measured more than one trait or
dimension.
DISCUSSION OF FINDINGS
The need
to have a quality economic achievement test for use by teachers in conducting
continuous assessment of students is an important component of teaching and
learning. The results of the analysis shows that the validity of EAT was sought
through experts judgement and test blue print based
on the objectives in the economic curriculum and based on the results, the test
was said to be valid. This is in agreement with the views of Awotunde and Ugodulunwa (2003)
and Farooq (2013) that a valid measurement instrument leads to accurate
measurement and evaluation while instruments that are not valid leads to wrong
decision in measurement and evaluation. Furthermore, the reliability of the
test items was established using omega reliability and it was found to be 0.83.
This is also in agreement with the finding by Akwa
(2014) that a good instrument should have a reliability of 0.70 and above.
The
result of the analysis further showed that the difficulty indexes of the EAT
fall within the acceptance region of between –2 – +2. This is in agreement with
the findings by Ojerinde, Popoola,
Ojo and Onyeneho (2012) and
Ojerinde (2013) that difficulty index of achievement
test items should be between – 2 t0 +2. This is adequate because the items will
be of moderate difficulty. While items that are of low difficulty need to be
discarded because even low ability testees have the
chance to answer those items correctly. Furthermore, the results of the funding
reveals that all the items discriminated well between high and low achievers
since none of the items has less than 0.05 thresholds. This is in tandem with
the finding by Isaac (2011) and Ugwu (2012) that item
with low-a-value discriminate poorly over a wide rang
of abilities and items with discrimination value below 0.80 are not good items.
The higher the a-value, the more sharply the item discriminates between
examinees at the point of inflection. The results of the analysis also reveals
that most of the items were not prone to guessing because the items had
guessing parameters indexes that fall within the range of 0.12 – 0.25. The
findings is consistent with the finding by Ojerinde
(2013) and Ojerinde, Popoola,
Ojo & Ariyo (2014) that
proper construction of achievement test reduces the chance of low ability students
responding to an item correctly, while poor construction of multiple-choice
items increases the chance of low ability students responding correctly to an
item even when the testees do not have the ability to
answer the item correctly.
The
result of the findings further showed that the assumption of unidimensionality was rejected because the pooled 65-items
of the EAT violate the assumption of unidimensionality.
This also confirmed the views of Doran and Kingston (2018) and Li, Jiao (2012)
that when a data violate the assumption of unidimensionality,
then multidimensionality should be applied in determining the psychometric
properties of the instrument before the calibration of items.
CONCLUSION
Based on
the findings of the study, it was concluded that teachers need to use quality
teacher-made achievement test like the EAT in conducting continuous assessment
of students; so as to improve the performance of students in internal and
external examinations in Plateau State, Nigeria.
RECOMMENDATION
Based on
the findings of the study, the following recommendations are made:
- The
economics achievement test is hereby recommended for use in conducting
continuous assessment of students in economics in the study area.
- Teachers
need to establish the psychometric properties of achievement test before
using the test for assessment.
- Teachers
need to use item response theory achievement test for the purpose of
assessment.
- Teachers
in the study area need training in test construction procedure using IRT
so as to improve the test development skills of teachers in the study
area.
REFERENCES
Adedoyin, O. O., & Adegoke,
J. A. (2013). Assessing the comparability between classical test theory (CTT)
and item response theory (IRT) models in estimating test items parameters. Herald Journal of Education and General
Studies, 2(3), 107 – 114.
Akwa, A. M. (2014). Development
and standardization of achievement test in senior secondary school mathematics
using item response theory. Retrieved 29 September, 2017 from: https://acbrary.lated.org/.../
Amanda, C. (2014). Achievement
test-definition –objectives functions and characteristics. Retrieved 17
October 2017 from: www.nsgmed.com
Amuche, C. I., & Fan, A. F. (2016). An assessment of item bias using
differential item functioning technique in NECO biology conducted examination
in Taraba State Nigeria. American International Journal of Research in Humanities, Arts and
Social Science, ISSN (print), 2328 – 3734.
Anyawuchi, R. A. J. (2008). Fundamentals of
economics for senior secondary schools. Onisha, African
First Publishers Limited.
Baker, J. O. (2003). Testing
in modern classrooms. London: Gense Allem and Urwin Ltd.
Chalmers, R. P. (2013). Mirt: A
multidimensional item response theory package for the environment. Journal of Statistical Software, 48(6), 1 – 29.
Danjuma, A. O. (2014). Development and validation of
workshop based process skills test in metal grinding for assessing students in
technical colleges for work. Journal of
Education and Practice, 5(25),
768 – 920. Retrieved 14 June, 2017 from: www.iiste.org.
Dishe, M. (2018) teacher-made test, meaning, features and uses/statistics
retrieved on the 3rd July 2018 from www.yourarticlelibrary.com/statistics-2/teacher-made-test.htm
Doran, N. J,Kingston,
N. M.,(2018). The effect of violation of unidimensionality
on estimation of items and ability parameters on item response theory equating
of GRE verbal scale, journal of education 22(4)22-45.
Farooq, U. (2013). Educational
measurement, definition and concepts. Retrieved 13 June, 2017 from: www.studylecturenotes.com.
Federal Republic of Nigeria (FRN) (2008). National Policy on Education (Revised 5th ed.). Lagos: NERDC Press.
Hambleton, R. K., & Jones, R. W. (2013). Comparison of classical test theory and
their applications to test development. Retrieved 13 March, 2017 from:
http://ncme.org/ linksevid /6696808013.
Hamman-Tukur, A., Musa, U., & Atsua,
T. (2013). Development and validation of secondary school ranking parameters
for quality assurance. Nigerian Journal
of Educational Research and Evaluation, 12(special
edition), 224 – 233.
Hamafyelto, .R. S., Hmman-tukur,
A & Hamafyelto, S. S. (2015). Assessing teachers’ competence in test
construction and content validity of teacher made examination questions in
commerce in Borno State Nigeria. Retrieved 17
March, 2017 from: www.article.sapub.org
Imo, G. C. (2011). Effect of
training in test construction on quality of teacher-made tests and study
physics achievement in secondary schools in Plateau State. Unpublished
doctoral dissertation, University of Jos.
Isaac, O. (2011). Development
and validation of psycho-productive skills multiple choice test items in
agricultural science for students in secondary schools. Retrieved 16 March
2017 from: www.unn.edu.ng.
Joshua, E. O., & Adeleke, A. A.
(2015). Development and validation of scientific literacy achievement test to
assess senior secondary school students’ literacy acquisition in physics. Journal of Education and Practice, 6(7), 81 – 96.
Li, Y., & Jiao, W. S. & (2012). Applying multidimensional
item response theory models in validating test dimensionality. Journal of Applied Testing, 13(2) 22 – 37.
Madu, A. O. (2014). Development and validation of a survey achievement
test in agricultural science for senior secondary schools in Plateau State. Journal of Sustainable Agriculture and
Environment, 15 (1), 95 – 102.
Nwana, O. C. (2007). Introduction
to educational research. Ibadan: HEBN Publishers Plc.
Ojerinde, D., Popoola, K., Ojo, F., & Onyeneho, P.
(2012). Introduction to item response
theory, parameter models estimation and application. Abuja: Marvelous Mike
Press Ltd.
Obilor, E. I & Akpan, U. T.(2020)
Construction and Validation of Basic science achievement test for junior
secondary three (JS3) students in public secondary schools in Akwa Ibom State.
Ojerinde, D. (2013). Classical test theory (CTT) vs. item response theory (IRT): An
evaluation of the comparability of item analysis results. Retrieved 18
March, 2017 from: UI.ed.ng.
Ogbebor, U., & Onuka,
A. (2013). Differential item functioning method as an item bias indicator. International Journal Research, 4(4), 367 – 373.
Osadebe, P. U. (2014). Construction of economics
achievement test for assessment of students. World Journal of Education, 4(2),
56-63.
Ugodulunwa, C. A. (2020). Fundamentals of educational measurement. Jos: Fab. Educational
Books, Nigeria.
Ugwu, S. N. (2012). Development
and validation of criteria referenced achievement test in biology.
Retrieved 16 September, 2017 from: www.xandonline.com.
Wakjissa, S. G. (2010). Appraisal of senior secondary
2 geography teachers’ competency in assessing students Blooms levels of
cognitive objectives in plateau state Nigeria. Journal of Assessment in African, 5(1), 177 – 188.
