Elementary Teachers' Classroom Assessment and Grading Practices

Author(s): James H. McMillan, Steve Myran and Daryl Workman

Source: The Journal of Educational Research, Vol. 95, No. 4 (Mar. - Apr., 2002), pp. 203-213
Published by: Taylor & Francis, Ltd.
Stable URL: http://www.jstor.org/stable/27542381
Accessed: 06-02-2017 16:46 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to The Journal
of Educational Research

This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
 Elementary Teachers' Classroom
Assessment and Grading Practices
JAMES H. MCMILLAN
STEVE MYRAN
DARYL WORKMAN
Virginia Commonwealth University
ABSTRACT The authors investigated the assessment and
grading practices of over 900 Grades 3-5 teachers represent
ing urban, suburban, and rural schools. Teachers indicated the
extent to which they used various factors to grade students, the
types of assessments used, the cognitive level of assessments,
and the grades awarded. Teachers appeared to conceptualize 6
major factors when they graded students; they placed the
greatest weight on academic performance and academic
enabling behaviors, such as effort and improvement, and
much less emphasis on homework, comparisons with other
students, grade distributions of other teachers, and borderline
cases. The teachers used 3 types of assessments?constructed
response, objective, and teacher-made major examinations;
they differentiated between recall and higher level cognitive
skills. However, there were few relationships between assess
ment and grade level, subject matter assessed, and grades
awarded. Results are discussed in light of other research, indi
cating that teachers use a "hodgepodge" of factors when
assessing and grading students.
Key words: classroom assessment, elementary teachers, grad
ing practices
A significant amount
classroom assessment of recent
and grading literature has focused on
as essential aspects
of effective teaching. An increased scrutiny of assessment is
evidenced by the popularity of performance assessment and
portfolios; newly established national assessment compe
tencies for teachers (American Federation of Teachers,
National Council on Measurement in Education, and
National Education Association, 1990); and the interplay
between learning, motivation, and assessment (Brookhart,
1993, 1994; Tittle, 1994).
Researchers have documented teachers' tendency to
award a "hodgepodge grade of attitude, effort, and achieve
ment" (Brookhart, 1991, p. 36; Cross & Frary, 1996),
although this conclusion was reached primarily on results of
surveys of secondary-level teachers. It is also clear that
teachers use a variety of assessment techniques, even if
established measurement principles are often violated
(Cross & Frary; Frary, Cross, & Weber, 1993; Gullickson,
1993; Plake & Impara, 1993; Stiggins & Conklin, 1992). In
203
This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
addition, over the last decade, significant emphasis has been
placed on using alternative assessments, such as perfor
mance assessments and portfolios, rather than on tradition
al paper-and-pencil assessments.
Given the variety of assessment and grading practices in
the field, the increasing importance of alternative assess
ments, the critical role that each classroom teacher plays in
determining assessments and grades, and the trend toward
greater accountability of teachers with state assessment
approaches that are inconsistent with much of the current
literature, one needs to fully understand current assessment
and grading practices. Such information is important to
understand how classroom assessments are constructed and
used in this new climate. The purpose of this investigation
was to describe actual classroom assessment and grading
practices of upper level elementary teachers, to determine
the primary factors used in grading, and to determine
whether meaningful relationships exist between indepen
dent variables grade level and subject taught and dependent
variables assessment and grading practices.
The literature tends to separate assessment practices from
grading practices. In this review, we first examined assess
ment practices. Airasian (1984) reviewed literature that
suggests that teachers focus their classroom assessments in
two areas?academic achievement and social behavior. The
importance of these items varies with grade level; elemen
tary teachers place greater importance on social behavior.
Fleming and Chambers (1983), in a study that analyzed
nearly 400 teacher-developed classroom tests, made the fol
lowing conclusions about the nature of classroom assess
ment: (a) Short-answer questions are used most frequently;
(b) essay questions, which represent slightly more than 1%
of test items, are avoided; (c) matching items are used more
than multiple-choice or true-false items; (d) most test ques
tions, approximately 69%, sample knowledge of terms,
facts, and rules and principles; and (e) few test items mea
Address correspondence to James H. McMillan, P.O. Box
842020, Department of Educational Studies, Virginia Common
wealth University, Richmond, VA 23284 (E-mail:
jmcmillan @ edunet. soe. vcu. edu)

sure student ability to apply what they have learned. Stig
gins and Conklin (1992) asked 24 teachers to keep ajournai
to reflect their assessment practices. The analysis focused
on how teachers described their assessments and which spe
cific issues related to their assessments were raised. The
researchers found that teachers were most interested in
assessing student mastery or achievement and that perfor
mance assessment was used frequently. The nature of the
assessments used in each class was coupled closely with the
roles that each teacher set for her students, teacher expecta
tions, and the type of teacher-student interactions desired.
Marso and Pigge (1993) summarized research that sug
gests that elementary teachers place more emphasis on stu
dents' constructed-response work samples than on tradi
tional paper-and-pencil tests. They also reported that direct
observation is used in language arts assessments more than
in other subjects and that essay assessments, although infre
quent, tend to occur most in language arts, history, and
social studies. The vast majority of teachers use several
types of assessments, generally placing greatest emphasis
on completion and short-answer questions. Stiggins and
Bridgeford (1985) also found that elementary teachers tend
to stress constructed-response tests rather than objective
types of tests, although both are used extensively. Stiggins
and Bridgeford (1985) reported that the use of teacher-made
objective tests is positively related to grade level, that pub
lished tests tend to be used more in early grades, and that
teacher-made tests are relied on more for mathematics than
for English assessments. The authors concluded that "grade
level appears to be an important variable in understanding
classroom assessment" (p. 281).
In a survey of 143 elementary and secondary school
teachers, Cizek, Fitzgerald, Shawn, and Rachor (1995)
found that assessment practices "were highly variable and
unpredictable from characteristics such as practice setting,
gender, years of experience, grade level or familiarity with
assessment policies in their school district" (p. 159). This
finding suggests that grade level may not be as important as
variations found between individual teachers. Overall, the
authors concluded that "many teachers seemed to have indi
vidual assessment policies that reflected their own individ
ualistic values and beliefs about teaching" (p. 160). The
highly variable nature of assessment practices, as is pointed
out in the following paragraphs, is consistent with how
teachers grade students. The Cizek et al. study was limited
to elementary teachers attending a university measurement
course and the use of a limited number of questions that
were restricted to 10 factors used in grading and five
sources of assessment-related information. Furthermore,
respondents simply checked whether each factor or source
was used, without any indication of the extent of use.
Plake and Impara (1997) summarized results from a large
scale survey of teachers that was structured to obtain teacher
competency concerning assessment practices, by asking
teachers to indicate which of several possible answers to
assessment questions was best. A national random sample of
This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
204 The Journal of Educational Research
555 elementary, middle, and high school teachers was used.
Overall mean performance on the survey was 66% correct.
Teachers did better on items related to choosing and admin
istering assessments and significantly worse on communi
cating results. According to the authors, the results "give
empirical evidence of the anticipated woefully low levels of
assessment competency for teachers" (p. 67). The results
also showed that teachers who had taken a measurement
course performed better than did teachers who lacked this
background.
In summary, the existing literature on elementary class
room assessment practices indicates that teachers probably
need further training to improve the quality of the assess
ments that are used. Whatever the type of question used on
assessments, few are written to tap students' higher level
thinking skills. Appropriately, teachers appear to use a vari
ety of assessment methods. Particularly absent in the litera
ture, however, are large-scale examinations of relationships
between classroom assessment practices and grade level
and subject matter. There is some evidence of trends across
grade level and subject matter, as in Stiggins and Bridgeford
(1985) and Marso and Pigge (1993), but small samples are
used in this study, and the researchers did not focus on dif
ferences between different elementary grades.
Teachers' grading practices have received far more atten
tion in the literature than have assessment practices. This
fact may be due to the salient and summative nature of
grades to students and parents. Grades have important con
sequences and communicate student progress to parents.
Stiggins, Frisbie, and Griswold (1989) set the stage for
recent research on grading by providing an analysis of cur
rent grading practices as related to recommendations of
measurement specialists and newly established Standards
for Teacher Competence in Educational Assessment of Stu
dents (American Federation of Teachers, National Council
on Measurement in Education, National Education Associ
ation, 1990). In this study, the authors interviewed and/or
observed 15 teachers on 19 recommendations from the
measurement literature. They found that teachers use a wide
variety of approaches to grading and that they want their
grades to fairly reflect both student effort and achievement,
as well as to motivate students. Contrary to recommended
practice, Stiggins and colleagues found that teachers value
student motivation and effort and set different levels of
expectation on the basis of student ability. This finding is
consistent with an earlier study by Gullickson (1985), in
which elementary teachers indicated that they used nontest
information, such as class discussion and student behavior,
more than test results to grade students. Gullickson also
reported little difference in grading practices between sci
ence, social science, and language arts. Given the increased
emphasis on school and student accountability due to high
stakes testing, the relative emphasis that teachers give to
tests may be increasing.
Brookhart (1994) conducted a comprehensive review of
literature on teachers' grading practices. Her review identi
March/April 2002 [Vol. 95(No. 4)] 205
fied 19 studies completed since 1984. In seven studies,
researchers investigated secondary school grading; in 11
studies, both elementary and secondary school grading; and
in one study, elementary school teachers. She identified
three general methods of study: (a) surveys in which teach
ers responded to questions concerning components includ
ed in grading, grade distributions, and attitudes toward
grading issues; (b) surveys in which teachers were asked to
respond to grading scenarios, asking what they would do in
various circumstances; and (c) qualitative methods, includ
ing interviews, observation, and document analysis. Despite
methodological and grade-level differences, the findings
from these studies are remarkably similar. Taken together,
Brookhart came to the following conclusions:
Teachers inform students of the components used in
grading.
Teachers try hard to be fair in grading.
Measures of achievement, especially tests, are major
contributors to grades.
Student effort and ability are used widely as compo
nents of grades.
Elementary teachers rely on more informal evidence
and observation, whereas secondary teachers use paper
and-pencil achievement tests and other written evidence as
major contributors.
Teachers' grading practices vary considerably from one
teacher to another, especially in perceived meaning and pur
pose of grades, and how nonachievement factors will be
considered.
Teachers' grading practices are not consistent with rec
ommendations of measurement specialists, especially con
founding effort with achievement.
In one study, Brookhart (1993) investigated the meaning
that teachers give to grades and the extent to which value
judgments are used in assigning grades. The results indi
cated that low-ability students who tried hard would be
given a passing grade even if the numerical grade were fail
ure, although working below ability level did not affect the
numerical grade. That is, an average or above-average stu
dent would get the grade earned, whereas a below-average
student would get a break if there were sufficient effort to
justify it. Teachers were divided about how to factor in
missing work. About half of the teachers indicated that a
zero should be given, even if that meant a failure for the
semester. The remaining teachers would lower the grade,
but not to a failure. The teachers' written comments
showed that they strived to be fair to students. Teachers
also seemed to indicate that a grade was a form of payment
to students for work completed. That is, grades were some
thing that students earned, as compensation for work com
pleted. This finding suggests that teachers, either formally
or informally, include conceptions of student effort in
assigning grades. Because teachers are concerned with stu
dent motivation, self-esteem, and the social consequences
This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
of giving grades, using student achievement as the sole cri
teria for determining grades is rare. This finding is consis
tent with earlier work by Brookhart (1991), in which she
pointed out that grading often consists of a hodgepodge of
attitude, effort, and achievement. A limitation of this study
is the small sample of elementary teachers (30) and the use
of only three nonachievement factors in scenarios that par
ticipants responded to (effort/ability, missing work, and
improvement). In addition, the students in our study were
taking a university measurement course, which could result
in socially desirable responses or answers that reflect the
perspectives of the instructor.
Brookhart's conclusion concerning the variety of factors
that go into grading is consistent with Cizek et al. (1995).
Cizek and colleagues also found that teachers generally use
a variety of objective and subjective factors to maximize the
likelihood that students obtain good grades.
In summary, the literature specific to elementary teach
ers' assessment and grading practices is limited. There is an
indication that teachers believe it is important to combine
nonachievement factors, such as effort, ability, and conduct,
with student achievement to determine grades; however,
most of the studies that provide the basis for this conclusion
have been conducted with secondary-level teachers.
Although the studies are clear in this conclusion, less is
known about how elementary teachers decide to weigh
these nonachievement factors in determining grades and
whether particular factors tend to be considered together or
whether elementary teachers separate nonachievement fac
tors, such as effort and improvement, from achievement.
Also, in most of the surveys and other approaches in previ
ous studies, researchers have asked teachers about their
beliefs or projected behavior on the basis of scenarios.
Actual assessment and grading practice may be different.
Few researchers used a scale in which teachers indicated the
actual use of different assessment and grading practices that
allowed independent recording of the emphasis of each fac
tor, rather than asking teachers to indicate the relative
emphasis of each factor by percentage. Finally, there is no
research on whether a relationship exists between the types
of assessments used and grades received by students.
In the present study, we used a large sample of elementary
school teachers to describe assessment and grading practices
in a way that builds on and extends previous studies. We
addressed four specific research questions, as follows:
1. What is the current state of assessment practice and
grading by elementary teachers?
2. What are major assessment and grading components
that are used by elementary teachers?
3. What is the relationship between assessment and grad
ing practices and grades given to students?
4. What are the relationships between the independent
variables grade level and subject taught (mathematics and
language arts) and the dependent variables assessment and
grading practices?

Method
Sample
The population included all Grades 3-5 regular elemen
tary teachers in seven urban/metropolitan school districts
near Richmond, Virginia (1,561 teachers in 124 schools).
Completed surveys were returned by 921 elementary teach
ers from 105 schools. Twenty of the teachers were not full
time in a regular classroom, resulting in a sample of 901.
The response rate by school was 88%; by teachers, 58%.
Instrument
The purpose of the questionnaire was to document, using
closed-form items, the extent to which teachers emphasized
different assessment and grading practices. A 6-point scale,
ranging from not at all to completely, was constructed to
allow teachers to indicate usage without the constraints of an
ipsative scale that is commonly used in this area (e.g., per
centage each factor contributes to grades). Also, the ques
tions were worded to emphasize actual teacher behaviors in
relation to a specific class of students, rather than more glob
al teacher beliefs. Teachers responded to all items once for
language arts and once for mathematics. The stem for the
items was "To what extent were final first semester grades of
students in your single class described above based on."
The initial set of items was drawn from previous ques
tionnaires that had been reported in the literature, as well as
from research on teachers' assessment and grading practices
(Brookhart, 1994; Frary, Cross, & Weber, 1993; Stiggins &
Conklin, 1992). The items included factors that teachers
consider in giving grades, such as student effort, improve
ment, academic performance, types of assessments used,
and the cognitive level of the assessments (e.g., knowledge,
application, reasoning). We strengthened content-related
evidence for validity for the initial draft of 47 items by ask
ing 15 elementary teachers to review the items for clarity
and completeness in covering most, if not all, assessment
and grading practices used. Appropriate revisions were
made to the items, and a second pilot test with a school divi
sion outside of the sample was used to gather additional
feedback on clarity, relationships among items, item
response distributions, and reliability. Twenty-three teach
ers participated in the second pilot test. Item statistics were
used to reduce the number of items to 27. Items that showed
a very high correlation (r > .90) or minimum variation were
eliminated, as well as items that were weak in reliability.
We assessed reliability by asking the teachers in the second
pilot test to retake the questionnaire following a 4-week
interval. The stability estimate was done by examining the
percentage of matches for the items. Those items that
showed an exact match of less than 60% were deleted or
combined with other items. The revised questionnaire
included 34 items in the three categories (19 items assess
ing different factors used to determine grades, 11 items
assessing different types of assessments used, and 4 items
This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
206 The Journal of Educational Research
assessing the cognitive level of the assessments). The aver
age exact match for the items was 46% of the teachers; 89%
of the matches were within 1 point on the 6-point scale. In
additional items, teachers were asked to indicate the
approximate grade distribution of the class.
Procedure
School division central administrators communicated to
teachers that they should complete the questionnaire and
that they were responsible for distribution and collection.
The questionnaire took about 15 min to complete. Teachers
were assured that their responses would be confidential. No
information was on the form that could be used to identify
the teacher. The surveys were completed in early February,
soon after the end of the first semester.
Data Analyses
The data analyses were primarily descriptive; we used
frequencies, percentages, means, medians, standard devia
tions, and graphic presentations to summarize overall find
ings and trends. We used an exploratory factor analysis to
reduce the number of components investigated within each
of the three categories of items. Relationships between
assessment and grading practices, grades given, grade level,
and participants were examined through multiple regression
and paired t tests.
Findings
The descriptive results are presented first, followed by
relationships. The assessment and grading practices report
ed were organized by the three categories of items?factors
used in grading, types of assessments used, and cognitive
level of assessments.
Descriptive Results
The means and standard deviations for the assessment
and grading practices items, divided by categories for both
language arts and mathematics, are reported in Table 1. In
Table 2, we present the frequency distributions of a few
questions to illustrate the spread of responses across the dif
ferent points in the scale.
The means and standard deviations in Table 1 show that,
for this group of teachers as a whole, a few factors con
tributed very little, if anything, to grades (i.e., disruptive
student behavior, grade distributions of other teachers, per
formance compared with other students, school division
policy about the percentage of students who may obtain
different grades, and extra credit for nonacademic perfor
mance). Also, a few factors clearly contributed most, rang
ing from quite a bit to extensively?academic performance
as opposed to other factors, performance compared with a
set scale of percentage correct, and specific learning
March/April 2002 [Vol. 95(No. 4)] 207

Table 1.?Means and Standard Deviations of All Items Measuring Assessment and Grading Practices for Elementary Teachers (N = 901)

Mathematics Language arts

Variable M SD M SD

Factors used in determining grad

Disruptive student behavior
Improvement of performance s
Student effort?how much stud
Ability levels of students 3.
Work habits and neatness 2.
Grade distributions of other te
Completion of homework (n
Quality of completed homewo
Academic performance, as oppo
Performance compared with ot
Performance compared with a
Performance compared with stu
Specific learning objectives m
Formal or informal school or distr
C's, D's, F's 1.50 1.15 1.50 1.14
The degree to which students pay attention, par
Inclusion of zeros for incomplete assignment
Extra credit for nonacademic performance (e.g.
Extra credit for academic performance 2.57 1.10
Effort, improvement, behavior and other
Types of assessments used in determining grades
Major examinations 3.21 1.39 3.05 1.38
Oral presentations 2.37 1.11 3.03 .88
Objective assessments (e.g., multiple choice, matching, short answer) 3.82
Performance assessments (e.g., structured teacher observations or ratings of per
a speech or paper) 2.84 1.14 3.43 .93
Assessments provided by publishers or supplied to the teacher (e.g., in instru
manuals) 3.54 1.05 3.22 1.06
Assessments designed primarily by yourself 3.63
Essay-type questions 2.42 1.15 3.39 1.03
Projects completed by teams of students 2.51 1.03 2.91 .99
Projects completed by individual students 3.06 1.24 3.59 .96
Performance on quizzes 3.93 .91 3.80 .98
Authentic assessments (e.g., real world performance tasks 2.95 1.08 2.89 1.06
Cognitive level of assessments used in determining grades
Assessments that measure student recall knowledge 3.65 .90 3.52 .86
Assessments that measure student understanding 4.46 .78 4.46 .77
Assessments that measure how well students apply what they learn 4.31 .84 4.
Assessments that measure student reasoning (higher order thinking) 3.99 .87 4

Table 2.?Percentages of Elementary Teachers' Responses to Selected Items for M

Question Not at all Very little Some Quite a bit Extensively Completely
Factors contributing to grades

Improvement of performance since the beginning of the year 13 17 38 21 7 2

Student effort?how much students tried to learn 6 14 44 27 7 2
Ability levels of students 10 13 31 24 19 4
Academic performance compared with other factors 2 3 12 29 44 10
Types of assessments used

Objective assessments 2 8 28 36 21 5
Peformance assessments 14 23 38 17 7 1

Cognitive level of assessments

Assessments that measure student reasoning 0 2 25 44 25 4

This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms

objectives mastered. The remaining factors contributed
some, more accurately ranging from very little to quite a bit.
A fairly large standard deviation was reported for those
items, showing considerable variation in the extent to which
the factors were used for grading. A large percentage of
teachers used effort in vastly different ways for grading.
This same kind of dispersion of scores was evident in many
of the factors. In Table 2, for example, 13% of elementary
teachers reported using improvement not at all, whereas
30% of the teachers responded quite a bit, extensively, or
completely. The extent to which ability level was used also
was highly varied; 23% of the teachers responded not at all
and 47% responded quite a bit, extensively, or completely.
Given that the grading scales in the districts used in this
study were based on how performance compares with a set
scale of percentage correct (e.g., 94-100 A, 86-93 B), we
were surprised to find that only 65% of the teachers
responded that they used this extensively or completely.
The items in which teachers indicated the types of assess
ments used shows that they did not rely on a single kind of
assessment. Rather, many different types of assessments
appear to have been used. Although objective assessments
were employed most frequently for both mathematics and
language arts (means of 3.75 and 3.82, respectively), per
formance assessments and projects were used almost as
much in language arts as objective items (means of 3.43,
and 3.59, respectively). Assessments in mathematics
included fewer performance assessments and projects
(means of 2.84 and 2.51, respectively). There was great
reliance on assessments prepared by the teachers them
selves, but also considerable use of assessments provided
by publishers (language arts means of 3.90 and 3.22, and
mathematics means of 3.63 and 3.54). The standard devia
tions with respect to types of assessments (about 1 point on
the scale), pointed to considerable variation.
Cognitive levels of assessments were very similar for
mathematics and language arts. The lowest rated assess
ments, in terms of use, were those that measure student
recall knowledge. The highest rated assessment was student
understanding, with application and reasoning in between.
For the three highest rated items, the means were around 4
on the scale (used quite a bit).
Table 3.?Percentages of Different Semester Grades Awarded by Elementary Teachers (N = 859)
A B C D F
Grade level Math LA Math LA Math LA Math LA Math LA
3 (n = 294) 23.97 22.56 35.02 34.20
4(^ = 258) 21.39 22.27 35.50 34.73
5 (n = 205) 17.56 20.83 33.32 32.24
Mixed (n = 102) 19.56 18.52 32.18 33.08
Total 20.62 21.05 34.00 33.56
Note. Math = mathematics; LA = language arts. Percentages were estimated by the teachers.
This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
208 The Journal of Educational Research
Grades Awarded
The results for percentages of different grades awarded
by elementary teachers are presented in Table 3. The table
was broken out by grade level and subject matter, as well as
by letter grade awarded. Percentages were estimated by
teachers and therefore may not sum to 100%.
Grades A, B, and C were most typically awarded by the
teachers, comprising more than 70% of the total grades
given. Grades D and F comprised less than 10% of total
grades given. A grade of B was most typically awarded by
teachers, accounting for approximately 32 to 35% of total
grades given. Grades A and C were nearly equally distrib
uted, accounting for approximately 40% of a combined
total; grades of A comprised approximately 18 to 24% of
total grades given, and grades of C comprised approximate
ly 21 to 25% of total grades. Grades of D awarded were
between 6 and 8% of the total, whereas grades of F were
less than 3% of the total grades given. The relatively large
variability of teacher responses was illustrated by the stan
dard deviations.
Table 4 shows a summary of results of another procedure
to examine variability, by comparing variability within
schools to variability between schools. To calculate the
average standard deviation within schools, we used the
responses of teachers from the same school to derive a stan
dard deviation score for that school for each item. We then
averaged 105 standard deviations, one for each of 105
schools, to result in within-schools variability. We calculat
ed between-schools variability by using the mean for each
school, considering that as a single score, and then calculat
ing the standard deviation of the means. The results of these
analyses for three items, and percentage of A's awarded, are
summarized in Table 4. In each case, the average variation
within schools was greater than the variation between
schools. Even though this result was influenced by the rel
atively low number of teachers in each school, which would
increase the variation, it still suggests that teachers in the
same school differed more, on average, than did responses
compared at the school level.
Figure 1 illustrates the frequency of mean percentage
mathematics A's awarded between schools. It shows that
21.33 23.77 6.29 6.24 2.24 1.78
24.52 24.35 6.67 6.22 2.90 2.33
23.60 23.90 7.54 7.31 2.62 2.33
21.91 22.48 7.11 6.07 2.54 1.75
22.84 23.63 6.90 6.46 2.58 2.05
March/April 2002 [Vol. 95(No. 4)] 209
mathematics grades awarded were 12% or less for 35 ele
mentary schools, whereas for 20 schools, the percentage of
mathematics A's awarded was 32%. It also shows a large
between-schools variation of the number of A's awarded.
Data Reduction
Before examining the relationships between subject
(mathematics compared with language arts) and grade level
(Grades 3, 4, and 5), we performed data reduction for each
of the major categories of items (factors, types, and cogni
tive levels) for both mathematics and language arts. The
first step in the data reduction was to eliminate items that
showed a floor effect with little variability. We used the
remaining items in the second step of the data reduction, an
exploratory factor analysis, to identify relationships among
the items by reducing them to a few relatively independent,
but conceptually meaningful, composite variables called
Table 4.?Variation Within and Between Elementary Schools
for Selected Items (N = 105)
Mean Mean
variation variation
Question within between dents, and borderline cases. Given the relatively low
% A's awarded in mathematics 16.2 10.4
Student effort?how much students tried
to learn .92 .57
Assessments that measure student
reasoning .81 .42
Objective assessments .97 .51
This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
components. A varimax rotation was used for the factor
analyses.
The factor analysis for items used in grading (factors)
resulted in six components. There were no differences
between mathematics and language arts. The loadings of
different items are summarized in Table 5.
The first component was comprised of three items that
emphasized effort, ability, improvement, work habits, atten
tion, and participation. These items could be considered
enablers to academic performance, important indicators to
teachers to judge the degree to which the student has tried
to learn and, by implication, actually learned. A second
component was defined by the two items that included
questions about homework. The third component was
loaded on one item concerning grade distributions of other
teachers. The fourth component included three items that
focused on academic performance of the student. The fifth
component was loaded highly on two items that included
comparisons with other students. The sixth component
included the suggestion that borderline work and using
extra credit are related and distinct from other factors. Thus,
there appear to be six conceptually meaningful variables
that elementary teachers use when grading students for both
language arts and mathematics. These variables include
actual performance, effort, ability and improvement, home
work, other teachers' grading, comparisons with other stu
emphasis on homework, comparisons with other students,
other teachers' grading, and the infrequent occurrence of
borderline cases, these results suggest that teachers concep
tualize two major ingredients?actual performance, and
effort, ability, and improvement. Of these two, academic
performance clearly is most important, but effort, ability,
and improvement remain as fairly important, especially for
some teachers.
The factor analysis for types of assessments used result
ed in three components. The item loadings were, for the
most part, the same for both subjects. The first component
was comprised of six items for mathematics types and four
items for language arts types, each of which described some
kind of constructed-response assessment, such as essays
(mathematics only), projects, and performance assess
ments. The second component, made up of either two or
three items, included objective assessments, quizzes (lan
guage arts only), and assessments provided by publishers.
Evidently, items provided by publishers are used in both
quizzes and objective assessments. The third component
was comprised of two items for mathematics (major exam
inations and teacher-made tests) and two items for language
arts (teacher-made tests and essays). This result suggests
that the common element in the third component was
teacher made. For mathematics and for language arts
essays, the major examinations tend to be teacher made.
The factor analysis for cognitive levels showed high
intercorrelation among the three items that suggested high
er order knowledge and skills (understanding, reasoning,
210 The Journal of Educational Research

Table 5.?Factor Loadings for Elementary Teachers'Assessment and Grading Practices

Factors used in grading

Improvement of performance since the beginning of the year .777
Student effort?how much students tried to learn .809
Ability levels of students .644
Completion of homework (not graded) .819
Quality of completed homework .750
Grade distributions of other teachers .660
Academic performance, as opposed to other factors .710
Performance compared with a set scale of percentage correct
(e.g., 86-94% B) .668
Specific learning objectives mastered .683
Performance compared with other students in the class .781
Performance compared with students from previous years .741
Extra credit for academic performance .730
Effort, improvement, behavior and other nontest indictors for
borderline cases .659
Types of assessments
Oral presentations .704
Performance assessments (e.g., structured teacher observations
or ratings of performance, such as a speech or paper) .746
Essay-type questions (mathematics only) .740
Projects completed by teams of students .819
Projects completed by individual students .712
Authentic assessments (e.g., real world performance tasks) .636
Objective assessments (e.g., multiple choice, matching, short
answer) .736
Assessments provided by publishers or supplied to the teacher
(e.g., in instructional guides or manuals) .776
Major examinations (mathematics only) .691
Assessments designed primarily by yourself .721
Essay-type questions (language arts only) .672
Cognitive level
Assessments that measure student understanding .836
Assessments that measure student reasoning (higher order
thinking) .795
Assessments that measure how well students apply what they
learn .818

Note. Unless otherwise noted, factor loadings for mathematics and language arts were averaged.

and application). Teachers tended to think about these items
as one kind of skill, apart from recall knowledge, which did
not load on this analysis.
Relationship Results
In the relationship analyses for subject matter and grade
level, we used paired t tests and analysis of variance,
respectively, with standardized component scores for the
items loading on each of the 10 components derived from
the factor analyses, plus the percentage of A's given, as
dependent variables. We also performed a regression analy
sis to determine if assessment and grading practices predict
grades. Thus, there were two independent variables?sub
ject matter, with two levels, and grade level, with three lev
els, in the first two analyses; we used all 10 components as
independent variables to predict the percentage of As
awarded.
The t test analyses showed that there were few differ
ences between language arts and mathematics assessment
and grading practices, despite the large sample size that
would have made it easy to detect statistically significant
differences. Clearly, there was more in common than there
was different on the basis of these two content areas. As
might be expected, differences occurred for the extent to
which performance assessments were used (mean of 2.33
for mathematics and 3.41 for language arts), projects com
pleted by individual students (mean of 3.01 for mathemat
ics and 3.56 for language arts), and the use of assessments
provided by publishers (mean of 3.56 for mathematics and
3.23 for language arts). Thus, only three items in the cate
gory types of assessments showed a difference between
mathematics and language arts. When considering other
factors such as effort, participation, homework, and so
forth, as well as cognitive levels, we found no difference
between the mathematics and language arts responses.

This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
March/April 2002 [Vol. 95(No. 4)] 211

Table 6.?Relationship of Grade Levels (3, 4, and 5) to Assessment Practices of Elementary Teachers
(N = 873)

No relationship Positive relationship Negative relationship

Mathematics

Effort, ability, and improvement Homework Percentage A's

Academic performance Extra credit
Teacher-made major examinations Constructed-response assessments
Higher order thinking and application Objective assessments

Language arts

Effort, ability, and improvement Homework

Academic performance Extra credit
Objective assessments Constructed-response assessments
Higher order thinking and application Teacher-made major examinations
Percentage A's

Table 7.?Factors, types of Assessments, and Cognitive Levels, as Predictor Variables of Percentage A's
Awarded for Elementary Teachers

Significant Significant
Subject positive relationship negative relationship

Mathematics (n = 714) .21 Higher order thinking and application Objective assessments
Publisher-provided items
Homework

Language arts (n = 731) .20 Constructed-response items Extra credit

We used one-way analyses of variance with Sheff? post
hoc tests, to examine the relationship between grade level
and assessment and grading practices. The results of these
analyses are shown in Table 6, which contains a summary
of the components that indicate no relationship, those that
show a positive relationship, and the single variable that
shows a negative relationship.
As with other analyses, the major finding was no differ
ence between grade levels on components that were most
important to assessment and grading. For both language arts
and mathematics, the results showed that as grade level
increases, so does the importance of homework, extra cred
it, and constructed-response assessments. For mathematics,
the importance of objective assessments showed a positive
relationship with grade level. In language arts, teacher
made major examinations contributed more in higher
grades. The only negative relationship was found in the per
centage of A's awarded, which means that fewer As were
awarded in higher grades.
The predictive relationship between assessment and grad
ing practices was examined with stepwise multiple regres
sion?one for language arts; one for mathematics, with per
centage As awarded as the dependent variable; and the eight
weighted component scores as independent variables. The
results of these regressions are summarized in Table 7.
The multiple correlation coefficients were relatively
small in both regressions, suggesting that the major predic
tors of grades were not the weight given to different factors,
types of assessments, or cognitive level of assessments.
Given that finding, the percentage of As awarded tended to
increase with increased weight given to higher order think
ing assessments for mathematics, and constructed-response
assessments for language arts. Negative relationships for
mathematics were found with objective assessments, pub
lisher-provided items, and homework, and extra credit for
language arts.
Discussion
The results of the analyses, consistent with earlier
research by Brookhart (1994) and Cizek et al. (1995), show
that most elementary teachers use a multitude of factors in
grading students. The hodgepodge of factors considered
when grading appear to be organized into six distinct com
ponents. Academic performance is clearly the most impor
tant factor in grading students, as also reported by Stiggins
and Conklin (1992), but the results of the present study
show that nontest performance and behavior, such as effort,
participation, and extra credit work, also are very important
for many teachers, consistent with the Gullickson (1985)
study. Disruptive student behavior, grade distributions of
other teachers, and norm-referenced interpretations con

This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms

tribute little to grading. A substantial percentage of elemen
tary teachers include zeros in the calculation of grades.
Because there are different ways this can be done, this find
ing suggests a need to explore in greater depth more specif
ically how this is accomplished.
Three major types of assessments are used?constructed
response, such as projects, essays, and presentations; objec
tive assessments; and teacher-made major examinations.
Although objective assessments are used most frequently,
there is also a great reliance on constructed-response types
of assessments. Teachers tend to differentiate the cognitive
level of their assessments into two categories?recall
knowledge and higher order thinking and application. High
er order thinking and application are emphasized heavily.
There is a significant reliance on assessments that are
designed by publishers, even if most assessment is teacher
made. This finding suggests that teachers need training in
how to evaluate the quality of their own assessments, as
well as those provided by others.
Along with the variety of factors that go into grading,
great variation exists within schools concerning the extent to
which teachers emphasize different factors in grading stu
dents. The finding that within-school variance is greater than
between-school variance suggests that individual teacher
preferences are more important than are differences between
schools in determining grading practices. This data suggest
that teachers vary considerably in how they weigh different
factors, even within the same building, and that school and
student characteristics as a whole are less important than are
individual beliefs. This finding is consistent with the highly
variable results found by Cizek et al. (1995) and McMillan
and Nash (2000), confirming that an important characteristic
of classroom assessment and grading practices is that they
are highly individualized and may be unique from one
teacher to another, even in the same school. It may be useful
for teachers to discuss such differences and consider
whether more consistency would provide students with a
clearer message about what is important.
An implication of this variability of grading is that there
seems to be only moderate consistency among teachers
about what is most important and how different factors are
weighted. Why is this the case? One possible answer is that
teachers view grading as an extension of teaching to promote
student success in general in many areas important to
schooling, including both academic achievement and aca
demic-enabling behaviors, such as responsibility, effort,
improvement, participation, and cooperation. If teachers
would award grades solely on the basis of academic perfor
mance, there might be much more consistency. This finding
also may explain why teachers' grading practices do not fol
low established measurement principles. McMillan and
Nash (2000) found that teachers base their grading practices
on their educational philosophies and on what is best for
each student. Considering differences in styles of teaching,
types of students, and curriculum, perhaps it is not surpris
ing that there is high variability of grading practices. From a
This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms
212 The Journal of Educational Research
policy perspective, an appropriate question is whether there
is a good reason to permit or even encourage differences in
grading practices or there is a need for educational measure
ment experts to modify recommendations to be more con
sistent with the realities of teaching and the multitude of fac
tors that influence teacher behavior. An implication for the
preparation of teachers is that it is important to train them to
integrate assessment meaningfully into instruction. Another
implication for teacher training is to explore with students
the importance of conceptualizing assessment as a process
involving a multitude of factors, combined in unique ways
that may differ for each individual.
The factor analysis indicated that several student behav
iors considered together indicate what has been labeled here
as an Academic Enabling factor. These behaviors include
student effort, participation, improvement, ability, and dis
cussion. It appears that teachers use all of these behaviors,
apart from actual academic performance, as important
indictors for determining grades. This finding has theoreti
cal significance when one is conceptualizing the nature of
student effort in relation to assessment. Brookhart (1997),
for example, has identified mental effort and realized stu
dent effort as a major component of a theoretical framework
for investigating the effects of classroom assessment. The
present findings suggest that the manner in which effort is
defined in this model may need to be broadened to include
some indication of improvement and ability. It may well be
that teachers think about effort as mediated by improvement
and ability. That is, what is regarded as high or low effort is
dependent on what knowledge and skills students bring to
the classroom and on a more general impression of ability.
It is clear that many teachers use academically enabling
behaviors, such as participation, effort, and improvement
extensively to determine grades, whereas other teachers do
not use these variables very much. This finding may reveal
educational philosophies or approaches that give different
messages to students. For example, teachers who reward
effort may be allowing students who are not competent, as
well as their parents, to believe that they demonstrate need
ed knowledge and skills. This action would be especially
troubling if it occurred more with low-socioeconomic status
(SES) students, who could be rewarded for their effort to
maintain involvement, because they might not be provided
with feedback that more accurately indicates their level of
knowledge and skills. Also, are teachers who weight effort
more in reality "coddling" students, making it easier for
them to obtain passing grades? Some support for this impli
cation is found in a study by Cauley and McMillan (2000),
who found that middle school teachers at low-SES schools
(determined by the percentage of students eligible for free
and reduced-price lunches) tend to use nontest factors more
in grading than do teachers at low-SES schools.
Although few relationships were reported between assess
ment practices and grade level, greater emphasis in later
grades is placed on homework, extra credit, constructed
response assessments, objective assessments, and major
March/April 2002 [Vol. 95(No. 4)] 213

American Federation of Teachers, National Council on Measurement in

examinations. Other practices, such as effort, ability,
Education, and National Education Association. (1990). Standards for
improvement, and academic performance are emphasized teacher competence in educational measurement. Washington, DC:
the same in all three grade levels. Teachers who award more Author.
A's use fewer objective assessments, fewer publisher-pro Brookhart, S. M. (1991). Grading practices and validity. Educational Mea
surement: Issues and Practice, 10, 35-36.
vided tests, less homework, and more assessments that mea Brookhart, S. M. (1993). Teachers' grading practices: Meaning and values.
sure reasoning and application. There was no relationship Journal of Educational Measurement, 30, 123-142.
between the extent to which effort, improvement, ability, Brookhart, S. M. (1994). Teachers' grading: Practice and theory. Applied
Measurement in Education, 7, 279-301.
academic performance, homework, and extra credit were Brookhart, S. M. (1997). A theoretical framework for the role of classroom
emphasized, and percentage A's awarded. Overall, the analy assessment in motivating student effort and achievement. Applied Mea
surement in Education, 10, 161-180.
ses designed to predict assessment and grading practices fell
Cauley, K. M., & McMillan, J. H. (2000, April). Do teachers grade differ
short of much practical significance. We know that these ently in low SES middle schools? Paper presented at the annual meeting
practices do not vary much according to grade level (3-5) or of the American Educational Research Association, New Orleans, LA.
subject matter (mathematics and language arts). However, Cizek, G. J., Fitzgerald, S., M., & Rachor, R. E. (1995). Teachers' assess
ment practices: Preparation, isolation and the kitchen sink. Educational
there was considerable variation between schools in the per assessment, 3(2), 159-179.
centage of A's awarded. Further research concerning these Cross, L. H., & Frary, R. B. (1996, April). Hodgepodge grading: Endorsed
differences is warranted to explore possible reasons. by students and teachers alike. Paper presented at the annual meeting of
the National Council on Measurement in Education, New York.
The results of this study are limited by several variables, Fleming, M., & Chambers, B. (1983). Teacher-made tests: Windows on the
including demographics and location (Virginia is in the classroom. In W. E. Hathaway (Ed.), Testing in the schools: New direc
midst of a statewide assessment program consisting of all tions for testing and measurement (pp. 29-38). San Francisco: Jossey
Bass.
multiple-choice tests, with the exception of writing), using Frary, R. B., Cross, L. H., & Weber, L. J. (1993). Testing and grading prac
Grades 3, 4, and 5 with language arts and mathematics but tices and opinions of secondary teachers of academic subjects: Implica
tions for instruction in measurement. Educational Measurement: Issues
not other subjects, and the fact that the data, including
and Practice, 12(3), 23-30.
grades awarded, were based on teacher self-reports. In addi Gullickson, A. R. (1985). Student evaluation techniques and their relation
tion, the analyses were unable to tease out possible effects ship to grade and curriculum. The Journal of Educational Research,
due to the SES of each school. However, the large and com 79(2), 96-100.
Gullickson, A. R. (1993). Matching measurement instruction to classroom
prehensive nature of the sample still suggests strong exter based evaluation: Perceived discrepancies, needs, and challenges. In S.
nal validity. The self-reports were based on actual practice T. Wise (Ed.), Teacher training in measurement and assessment skills
with a specific class, not more generic beliefs, and repre (pp. 1-25). Lincoln, NB: B?ros Institute of Mental Measurement.
Marso, R. N., & Pigge, F. L. (1993). Teachers' testing knowledge, skills,
sented innercity, suburban, and rural schools. Researchers and practices. In S. T. Wise (Ed.), Teacher training in measurement and
who investigate assessment practices may find that the com assessment skills. Lincoln, NB: B?ros Institute of Mental Measurement.
ponents identified are useful categories for asking questions McMillan, J. H., & Nash, S. (2000, April). Teacher classroom assessment
and grading practices decision making. Paper presented at the annual
and relating assessment and grading practices to student meeting of the National Council on Measurement in Education, New
motivation and achievement. Orleans, LA.
Plake, B. S., & Impara, J. C. (1993). Assessment competencies of teachers:
A national survey. Educational Measurement: Issues and Practice, 12,
NOTES 10-25.
Plake, B. S., & Impara, J. C. (1997). Teacher assessment literacy: What do
This research was supported by the Metropolitan Educational Researchteachers know about assessment? In G. D. Phye (Ed.), Handbook of
Consortium, Virginia Commonwealth University, Richmond, Virginia. The classroom assessment (pp. 55-68). New York: Academic Press.
findings do not represent the views of members of the consortium. Stiggins, R. J., & Bridgeford, R. J. (1985). The ecology of classroom
The authors appreciate the reviewers' suggestions for needed improveassessment. Journal of Educational Measurement, 22(4), 271-286.
ments in earlier drafts.
Stiggins, R. J., & Conklin, N. F. (1992). In teacher's hands: Investigating the
practices of classroom assessment. Albany: State University of New York.
REFERENCES Stiggins, R. J., Frisbie, D. A., & Griswold, P. A. (1989). Inside high school:
Building a research agenda. Educational Measurement: Issues and
Airasian, P. W. (1984). Classroom assessment and educational improve Practice, 8, 5-14.
Tittle, C. K. (1994). Toward an educational psychology of assessment for
ment. Paper presented at the conference, Classroom Assessment: A Key
to Educational Excellence, Northwest Regional Educational Laboratory,teaching and learning: Theories, contexts, and validation arguments.
Portland, OR. Educational Psychologist, 29, 149-162.

This content downloaded from 128.6.218.72 on Mon, 06 Feb 2017 16:46:08 UTC
All use subject to http://about.jstor.org/terms