Authentic Assessment:
Role of Student
Choice in Summative Assessment Selection
Thomas B Barr
Stevenson University
STUDENT CHOICE IN
ASSESSMENT
Research Question
To understand the success and
failure of students in a classroom the basic underlying psychology of people
must be addressed first. The area that deals with this is termed positive
psychology and a specific theory that encompasses how humans thrive is
Self-Determination Theory (SDT). SDT states that there are three fundamental
mental “nutrients” that are required for optimal development: competency,
relatedness, and autonomy (Deci & Vansteenkiste, 2004).
In an effort to meet these needs in
a classroom, teachers need to construct activities that are geared to each of
the three. The last “nutrient”, autonomy, has been linked to a students
motivation and engagement in learning (Bennett, Zentall, French, & Giorgetti-Borucki, 2006; Cosden, Gannon,
& Haring, 1995; Eshel & Kohavi, 2003; Hardre & Reeve, 2003; Reeve
& Jang, 2006).
One of the problems that can occur in teaching are the limited choices that
students have access to in the classroom. In a teacher-centered classroom,
material covered and assessment methods are decided for the students. One way
to engage students in the learning process is to allow them to decide the
method that they will be assessed by.
Ideally, students would be able to
select from selected response, constructed response, and performance
assessments. While students may not have control over the material that is
covered due to central decision making, standardized test goals, or other
factors; choice of summative assessment can contribute to students’ autonomy in
a relatively easy fashion. Consequently, students having the opportunity to
choose the type of assessment that they take part in can serve both autonomy and
competency within SDT. Autonomy, through being a causal agent in determining
their method of assessment and competent, by selecting the test that best
allows them to demonstrate mastery.
With this in mind there are two
areas of questions that can be asked about student choice in assessment.
Primarily, we are concerned with determining what impact student selection of
summative assessment type will have on performance on a standardized test. In
this manner, selection of assessment type will act as a method of differentiation
of assessment, which has been suggested as an effective teaching method (Tomlinson & Moon, 2013). Differentiated instruction has
been described is a promising tool, that is at the same time vague and flexible
(Ernest, Thompson, Heckaman, Hull, & Yates, 2011).
Hopefully, this research can better describe differentiated instruction through
student choice.
Therefore, the purpose of this
research is primarily to investigate and determine the impact of student choice
on learning. This will be accomplished through calculating a change score which
will be the difference between pre-test and post-test standardized test scores.
To be clear, this means we will not be so concerned with what type of
assessment students choose in our primary research question, simply that they
have a choice. Our secondary research questions will deal with the nature of
the type of assessment, but this will be framed within the context of student
choice.
Secondarily, we will determine the
types of assessments, student affect, and correlation between test performance.
are students most likely to choose resulting student affect on test choice. Depending
on the results of our primary research question, it might be efficient for
teachers to give certain types of assessments that are more attractive to them.
We are also concerned with the way that students feel about selecting their
assessment type. It is possible that they will enjoy and are motivated by the
ability to select their test, however it may also result in unnecessary stress.
Finally, we will investigate if there is a correlation between performance on
different assessment types and performance on a standardized test.
Research Questions:
1.What impact does student choice of summative assessment
type have on performance on standardized tests in high school biology students?
2. What types of assessments are high school biology
students most likely to pick?
3. How does student selection of summative assessment type alter
high school student feelings toward assessment?
4. Is there a correlation between performance on different
types of summative assessments and performance on a standardized test in high
school students?
Research Hypotheses:
There will be significant differences between the
experimental group and the control group on the following dependent variables:
1. Standardized test change scores will be higher in the experimental
group than the control group
2. “Normal” test types are the most likely to be selected.
3. Student affect will be higher in the experimental group
than in the control group.
4. Performance on a test will be correlated with performance
on a subsequent standardized test.
Literature Review
As a
psychological movement, positive psychology has attempted to identify
environmental and personality factors that nurture an individual’s strengths,
virtues, and development (Deci & Vansteenkiste, 2004). The starting point for a
positive psychology theory is based in an active organism meta-theory (Deci & Vansteenkiste, 2004), that assumes three things. First,
that humans are proactive in pursuing mastery of internal and external forces.
Second, humans are self-organized systems that move to growth, integration, and
health. Finally, for humans to pursue their optimal selves, they must receive the
proper support from their environment (Deci & Ryan, 1985).
Within SDT is the understanding
that the growth of a human can be nurtured or thwarted by the environment and
the interactions with it. This is explained by the ability to meet the basic
psychological needs of a human being. When this occurs, growth and development
occurs, however when they are not met negative consequences ensue. The three
basic psychological needs are competence, relatedness, and autonomy. Competence
is the desire of people to effectively deal with the environment. Relatedness
is the need to interact with other people. Finally, autonomy is the human urge
to act according to one’s volition, or to have choice when acting (Deci & Vansteenkiste, 2004).
The application of SDT to the
educational environment has resulted in an understanding that student choice is
key in fostering motivation and engagement. Students must be given more
opportunities to play an active role in their learning through setting their
own goals, making choices about learning activities, and making decisions based
on their needs and preferences (Evans & Boucher, 2015). For choice to be useful in
the classroom there must be options, students must be aware of the options, and
student must have the ability to make the choice (Evans & Boucher, 2015; Katz & Assor, 2006; Rose & Meyer,
2002; Servilio, 2009).
Student choice in the classroom has
been found to reduce the amount of unwanted behavior in students with AD/HD (Bennett, et al., 2006), improve performance of
students with behavioral problems (Cosden, et al., 1995), and improved the academics
and affect of university students (Dosch & Zidon, 2014; Varsavsky & Rayner, 2013).
However, there has been little attention paid to high school students and their
choice in assessment. We will use the existing research to better frame our
current questions about student choice in high school students.
Student choice in assessment led to
a reduction in the number of unwanted behaviors in students with AD/HD.
Students with AD/HD were given a choice on how they received feedback and
compared to a control group that did not have any choice. There was no performance
improvement associated with the student choice, however the purpose of the
research was to determine if choice had any impact on behavior. With the clear
positive effect of choice, other research needs to determine what other changes
occur in students when choice is put in the hands of students (Bennett, et al., 2006).
Other research has examined the University
teaching method, which is traditionally based around the professor-centered,
lecture style, which can lead to low motivation amongst students. An
alternative to this is differentiated instruction, which one component of this
is student choice in learning task and assessment type. The study compared a
differentiated instruction (DI) classroom to a non-differentiated instruction
(NDI) classroom. The NDI classroom was a conventional lecture based class and
the DI group had a variety of learning activities, choices for assignments, and
instruction altered due to formative assessment (Dosch & Zidon, 2014).
The results of this study
demonstrated that students performed better academically at a statistically
significant level in a DI classroom than in a NDI classroom. Students also
indicated that they liked the type of classroom activity better in the DI
classroom than in the NDI. It appears that choice is an important component in
learning and students feel better about the material if they have a choice in
it (Dosch & Zidon, 2014).
The limitations that accompany this
study are the difficulty of identifying the most effective components of DI. DI
is a package of tools to use, and thus some skill as a teacher is required.
Some teachers may lack the social skills to effectively implement DI in their
classroom due to not being able to read what students need in DI (Dosch & Zidon, 2014). As a means to expand upon
this study we will attempt to remove the social skills component required of a
teacher and focus the current investigation on student choice of assessment.
Other research has pointed out that
high achieving students have to deal with challenges in the large, teach to the
middle, situation of the large undergraduate classroom. To combat this,
differentiated assessments were implemented into a college chemistry and
biology class. Students that had achieved a certain standard in a previous
science class were given the option to complete a more challenging assessment
or the regular standard assessment. Students were then given surveys and a
questionnaire to determine their feelings towards the assessment options. Overall,
few students attempted the advanced assessments, but the opportunity to do so
was deemed a positive thing from all students, regardless of whether or not
they took the advanced assessment. It appears that students believe that it is
a good idea to have a choice on assessment, even though the options may be more
difficult (Varsavsky & Rayner, 2013).
Again a
limitation of this research is that it only examined the affective responses of
students to the different types of assessment and was targeted at high achieving
students. While these are important factors to consider when differentiating
the classroom, it would be more helpful to see if other variables changed such
as performance in the class, or performance on other standardized items (Varsavsky & Rayner, 2013) and include more students. To
complement Varsavsky & Rayner, we will look at both the performance and
aspect of student choice of assessment in all students.
Research Design
The problem that exists today is
inauthentic assessment is commonly used, and academic work in general is presented
in such a way to disengage and demotivate students. This results in lowered
motivation and engagement for students in school. Ultimately, this results in a
world with a lack of creativity, problem solving, and critical thinking that
could benefit us all. To address this problem, it is proposed to investigate
the affect that self-selected assessments have on student performance on a
subsequent standardized test.
The proposed method to explore the
problem is an action research design that will have two groups of students who
are a part of a high school biology class. All students will take a
pre-assessment on the digestive system to determine their prior knowledge in
this area. Students will then go through a lesson unit on the subject, which
will be the same for all. At the end of the unit subjects will be separated
into two groups. A control group will take the “normal” test and a second,
experimental group that will be given a choice of the type of assessment that
they complete.
The “normal” test will consist of a
mixture of selected response and constructed response questions that emphasize
the basic themes of the unit. This normal test is the only option to the
students in the control group. Those subjects in the experimental group may
choose the “normal” test or two other types of tests: an advanced test and a
performance test. The advanced test stresses critical thinking, creativity, and
application of knowledge by emphasizing constructed response questions.
Finally, students will be given the option to complete a multimodal performance
that demonstrates knowledge of the digestive system. This performance can be a
presentation, recorded video, art work, etc. After completing the chosen test,
all students will complete a post-unit assessment, which will be another
version of the pre-assessment. As a part of the post-assessment students will
complete an affective questionnaire to gauge their feelings about the test
taking process.
A power analysis was performed to
determine the minimum number of students that would be needed to observe
statistically significant differences at the .05 alpha level for dependent
samples t-test. Based on a power of 80%, a standard deviation of 5, and an
effect size of .2, a conservative estimate of 27 subjects in each group would
be needed, for a total of 54 subjects.
Data Analysis
To
adequately address the questions and issues that were previously stated,
several types of data need to be collected. First, would be descriptive data of
the students in the study. Age, education level, previous grades in biology and
the sciences, ethnicity, race, and gender would all be useful independent
variables to group students and answer questions retrospectively. At this point
in time the author has no evidence to indicate that demographic factors will
affect any of the variables being studied. They will be collected and analyzed
at a later date to determine if future research questions should be asked and
answered. The authors would need to collect pre-assessment scores, test type
selection & score, post-assessment score, and a post assessment
questionnaire.
The data
collected during this study would be best presented in bar graph and chart form
depending on the question being answered. This would allow the data to be
clearly understood based on the group and data that is being presented. Data
could also be presented in a scatter plot in the case of an examination of a
correlation between the test scores.
Data would
be collected from two sections of a high school biology class, one for the
experimental group and one for the control group. Initially, students would
have to consent to being part of the study and understand it is voluntary and
will not affect their grade in any way. After consenting to the study students
would submit a biographical data sheet that collects the above mentioned
personal and demographic data.
All students would then begin the
unit as normal and complete all of the learning tasks. They would complete a
pre-assessment test, take part in all learning activities, lessons, labs,
lectures, projects, etc. As the class approaches the end of the unit the
experimental group will be given details about the type of summative assessment
they can take. For the control group they will be given the details about the “normal”,
selected and constructed response test only.
The implications of the data are
the effect that assessment choice has on standardized test scores. This is a
means by which to measure the impact that student autonomy has on traditional
measures of learning. We will also investigate the type of test that students
prefer to take. This will give teachers insight into what types of test are
most appealing to students. Teachers may then be able to limit the options to
those types of tests that best nurture the autonomy and competence of students.
The author also believes that student affect will be higher in the experimental
group than in the control group. If SDT hold true, providing a means for
student to express autonomy and competence should bring them closer to optimal
functioning. This should be reflected in affect scores. Finally, a strong
positive correlation should be observed between performance on a test and
subsequent standardized test performance.
The author will present the scores
in terms of a change score for statistical analysis, however for discussion
effect size will be presented. P-values
in research are useful to indicate if there is an effect, effect size indicates
the magnitude of the effect or what the clinical impact of the variable studied
can be (Sullivan & Feinn, 2012). This is commonly done
practice in biomedical research, which would prove valuable in education. We
would except a minimum effect size of .2 to demonstrate any meaningful impact
on student performance (Sullivan & Feinn, 2012). If we found an effect size
that is less than .2 this would indicate that there is no meaning effect that
student choice has on standardized test scores.
References
Bennett, D. E., Zentall, S. S., French, B. F., & Giorgetti-Borucki, K.
(2006). The Effects of Computer-Administered Choice on Students With and
Without Characteristics of Attention-Deficit/Hyperactivity Disorder. Behavioral Disorders, 31(2), 189-203.
Cosden, M., Gannon, C., & Haring, T. G. (1995).
Teacher-control versus student-control over choice of task and reinforcement
for students with severe behavior problems. Journal
of Behavioral Education, 5(1), 11-27.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic Motivation and Self-Determination
in Human Behavior: Springer US.
Deci, E. L., & Vansteenkiste, M. (2004).
Self-determination theory and basic need satisfaction: Understanding human
development in positive psychology. Ricerche
di Psicologia, 27(1), 23-41.
Dosch, M., & Zidon, M. (2014). "The Course
Fit Us": Differentiated Instruction in the College Classroom. International Journal of Teaching and
Learning in Higher Education, 26(3), 343-357.
Ernest, J. M., Thompson, S. E., Heckaman, K. A., Hull,
K., & Yates, J. (2011). Effects and Social Validity of Differentiated
Instruction on Student Outcomes for Special Educators. Journal of the International Association of Special Education, 12(1),
33-41.
Eshel, Y., & Kohavi, R. (2003). Perceived
Classroom Control, Self-Regulated Learning Strategies, and Academic
Achievement. Educational Psychology, 23(3),
249-260.
Evans, M., & Boucher, A. R. (2015). Optimizing the
Power of Choice: Supporting Student Autonomy to Foster Motivation and Engagement
in Learning. Mind, Brain, and Education,
9(2), 87-91.
Hardre, P. L., & Reeve, J. (2003). A motivational
model of rural students' intentions to persist in, versus drop out of, high
school. Journal of Educational
Psychology, 95(2), 347-356.
Katz, I., & Assor, A. (2006). When Choice
Motivates and When It Does Not. Educ
Psychol Rev, 19(4), 429-442.
Reeve, J., & Jang, H. (2006). What teachers say
and do to support students' autonomy during a learning activity. Journal of Educational Psychology, 98(1),
209-218.
Rose, D. H., & Meyer, A. (2002). Teaching every student in the digital age:
Universal design for learning. Alexandria, VA: Association for Supervision
and Curriculum Development.
Servilio, K. L. (2009). You Get to Choose! Motivating
Students to Read through Differentiated Instruction. TEACHING Exceptional Children Plus, 5(5).
Sullivan, G. M., & Feinn, R. (2012). Using Effect
Size—or Why the P Value Is Not Enough. Journal
of Graduate Medical Education, 4(3), 279-282.
Tomlinson, C. A., & Moon, T. R. (2013). Assessment and student success in a
differentiated classroom. Alexandria, Va.: ASCD.
Varsavsky, C., & Rayner, G. (2013). Strategies
That Challenge: Exploring the Use of Differentiated Assessment to Challenge
High-Achieving Students in Large Enrolment Undergraduate Cohorts. Assessment & Evaluation in Higher
Education, 38(7), 789-802.
