Page numbers

begin on the first

A NOVEL TEACHER EVALUATION MODEL page and follow 1
on every
The running subsequent
head is a page without
shortened Note: Green text boxes interruption. No
...while blue text boxes
version of the contain explanations of other information
contain directions for
paper's title that APA 7's paper formatting writing and citing in APA (e.g., authors'
appears on guidelines... last names) are
7.
every page. It is required.
written in all The paper's
capitals, and it
Branching Paths: A Novel Teacher Evaluation Model for Faculty Development title should be
should be flush centered, bold,
left in the and written in
document's title case. It
header. No Kim A. Park,1 James P. Bavis,1 and Ahn G. Nu2 should be three
"Running head:" or four lines
1
label is included Department of English, Purdue University below the top
in APA 7. If the margin of the
2
paper's title is Center for Faculty Education, Department of Educational Psychology, Quad City University page. In this
fewer than 50 sample paper,
Authors' names appear two Authors' affiliations follow
characters we've put three
lines below the title. They immediately after their
(including blank lines
should be written as names. If the authors
spaces and above the title.
follows: represent multiple
punctuation), the First name, middle initial(s), institutions, as is the case
actual title may last name. in this sample, use ORCID is an
be used rather Omit all professional titles superscripted numbers to organization that
than a shortened and/or degrees (e.g., Dr., indicate which author is allows researchers
form. Rev., PhD, MA). affiliated with which and scholars to
institution. If all authors register professional
Author notes represent the same profiles so that they
contain the institution, do not use any can easily connect
following parts numbers. with one another. To
in this order: include an ORCID
1. Bold, iD in your author
centered note, simply provide
"Author Note" the author's name,
label. followed by the
2. ORCID iDs green iD icon
3. Changes of (hyperlinked to the
author URL that follows)
affiliation. and a hyperlink to
4. the appropriate
Disclosures/ ORCID page.
acknowledgm Author Note
ents
5. Contact Kim A. Park https://orcid.org/0000-0002-1825-0097
information.
Each part is James P. Bavis is now at the MacLeod Institute for Music Education, Green Bay, WI.
optional (i.e.,
you should We have no known conflict of interest to disclose.
omit any parts
that do not Correspondence concerning this article should be addressed to Ahn G. Nu, Dept. of
apply to your
manuscript, or
Educational Psychology, 253 N. Proctor St., Quad City, WA, 09291. Email: agnu@qcityu.com
omit the note
entirely if
none apply).
Format each
item as its
own indented
paragraph.
 Note that both the running
head and the page
A NOVEL TEACHER EVALUATION MODEL number continue on the 2
The word "Abstract" should be pages that follow the title.
centered and bolded at the top
Abstract
of the page.
The main A large body of assessment literature suggests that students’ evaluations of their teachers
paragraph of
the abstract (SETs) can fail to measure the construct of teaching in a variety of contexts. This can
should not be
indented. compromise faculty development efforts that rely on information from SETs. The disconnect

between SET results and faculty development efforts is exacerbated in educational contexts
By standard
convention,
that demand particular teaching skills that SETs do not value in proportion to their local
abstracts do
not contain
importance (or do not measure at all). This paper responds to these challenges by proposing an citations of
other works.
instrument for the assessment of teaching that allows institutional stakeholders to define the If you need to
refer to
teaching construct in a way they determine to suit the local context. The main innovation of this another work
in the
instrument relative to traditional SETs is that it employs a branching “tree” structure populated abstract,
mentioning
by binary-choice items based on the Empirically derived, Binary-choice, Boundary-definition the authors in
the text can
(EBB) scale developed by Turner and Upshur for ESL writing assessment. The paper argues often suffice.
Note also
that this structure can allow stakeholders to define the teaching construct by changing the order that some
institutions
and sensitivity of the nodes in the tree of possible outcomes, each of which corresponds to a and
publications
may allow for
specific teaching skill. The paper concludes by outlining a pilot study that will examine the
citations in
the abstract.
differences between the proposed EBB instrument and a traditional SET employing series of

multiple-choice questions (MCQs) that correspond to Likert scale values.

Keywords: college teaching, student evaluations of teaching, scale development, ebb

scale, pedagogies, educational assessment, faculty development

An abstract quickly
summarizes the main
points of the paper that
follows it. The APA 7 Follow the abstract with a
manual does not give selection of keywords that
explicit directions for how describe the important ideas or
long abstracts should be, subjects in your paper. These
but it does note that most help online readers search for
abstracts do not exceed your paper in a database.
250 words (p. 38). It also The keyword list should have its
notes that professional first line indented. Begin the list
publishers (like academic with the label "Keywords:" (note
journals) may have a the italics and the colon). Follow
variety of rules for this with a list of keywords
abstracts, and that writers written in lowercase (except for
should typically defer to proper nouns) and separated by
these. commas. Do not place a period
at the end of the list.
Here, we've The paper's title is bolded and centered
borrowed a above the first body paragraph. There
A NOVEL TEACHER EVALUATION MODEL should be no "Introduction" header. 3
quote from
an external
source, so Branching Paths: A Novel Teacher Evaluation Model for Faculty Development
we need to
provide the According to Theall (2017, p. 91), “Faculty evaluation and development cannot be
location of
the quote in
considered separately ... evaluation without development is punitive, and development without
the document
(in this case,
the page evaluation is guesswork.” As the practices that constitute modern programmatic faculty
number) in
the development have evolved from their humble beginnings to become a commonplace feature of
parenthetical.
university life (Lewis, 1996), a variety of tactics to evaluate the proficiency of teaching faculty for
By contrast,
here, we've development purposes have likewise become commonplace. These include measures as
merely Spell out
paraphrased abbreviations
diverse as peer observations, the development of teaching portfolios, and student evaluations. the first time
an idea from
the external you use
source. Thus,
One such measure, the student evaluation of teacher (SET), has been virtually them, except
no location or in cases
page number ubiquitous since at least the 1990s (Wilson, 1998). Though records of SET-like instruments can where the
is required. abbreviations
be traced to work at Purdue University in the 1920s (Remmers & Brandenburg, 1927), most are very well-
known (e.g.,
modern histories of faculty development suggest that their rise to widespread popularity went "CIA").
For sources
hand-in-hand with the birth of modern faculty development programs in the 1970s, when with two
authors, use
universities began to adopt them in response to student protest movements criticizing an
ampersand
mainstream university curricula and approaches to instruction (Gaff & Simpson, 1994; Lewis, (&) between
the authors'
1996; McKeachie, 1996). By the mid-2000s, researchers had begun to characterize SETs in names rather
than the word
terms like “…the predominant measure of university teacher performance […] worldwide” "and."
When listing
(Pounder, 2007, p. 178). Today, SETs play an important role in teacher assessment and faculty multiple
citations in
development at most universities (Davis, 2009). Recent SET research practically takes the the same
parenthetical,
presence of some form of this assessment on most campuses as a given; Spooren, list them
alphabetically
Vandermoere, Vanderstraeten, and Pepermans, for instance, merely note that that SETs can be and separate
them with
found at “almost every institution of higher education throughout the world” (2017, p. 130). semicolons.

Darwin refers to them as “an established orthodoxy” and as a “venerated,” “axiomatic”

institutional presence (2012, p. 733).

Moreover, SETs do not only help universities direct their faculty development efforts.

They have also come to occupy a place of considerable institutional importance for their role in
Here, we've
made an
indirect or A NOVEL TEACHER EVALUATION MODEL 4
secondary
citation (i.e.,
we've cited a personnel considerations, informing important decisions like hiring, firing, tenure, and promotion.
Here, we've
source that
Seldin (1993, as cited in Pounder, 2007) puts the percentage of higher educational institutions cited a
we found
source that
cited in a
using SETs as important factors in personnel decisions at roughly 86 percent. A 1991 survey of does not
different
have a
source). Use
named
the phrase department chairs found 97% used student evaluations to assess teaching performance (US
author. The
"as cited in"
correspondin
in the Department of Education). Since the mid-late 1990s, a general trend towards comprehensive g reference
parenthetical
list entry
to indicate methods of teacher evaluation that include multiple forms of assessment has been observed would begin
that the first-
with "US
listed source (Berk, 2005). However, recent research suggests the usage of SETs in personnel decisions is Department
was
of
referenced in
still overwhelmingly common, though hard percentages are hard to come by, perhaps owing to Education."
the second-
listed one.
Include an
the multifaceted nature of these decisions (Boring et al., 2017; Galbraith et al., 2012). In certain
entry in the Sources with
reference list contexts, student evaluations can also have ramifications beyond the level of individual three authors
only for the or more are
secondary instructors. Particularly as public schools have experienced pressure in recent decades to adopt cited via the
source first-listed
(Pounder, in neoliberal, market-based approaches to self-assessment and adopt a student-as-consumer author's
this case). name
mindset (Darwin, 2012; Marginson, 2009), information from evaluations can even feature in followed by
the Latin
department- or school-wide funding decisions (see, for instance, the Obama Administration’s phrase "et
al." Note that
the period
Race to the Top initiative, which awarded grants to K-12 institutions that adopted value-added
comes after
"al," rather
models for teacher evaluation). than "et."

However, while SETs play a crucial role in faulty development and personnel decisions

for many education institutions, current approaches to SET administration are not as well-suited

to these purposes as they could be. This paper argues that a formative, empirical approach to

teacher evaluation developed in response to the demands of the local context is better-suited

for helping institutions improve their teachers. It proposes the Heavilon Evaluation of Teacher,

or HET, a new teacher assessment instrument that can strengthen current approaches to

faculty development by making them more responsive to teachers’ local contexts. It also

proposes a pilot study that will clarify the differences between this new instrument and the

Introductory Composition at Purdue (ICaP) SET, a more traditional instrument used for similar

purposes. The results of this study will direct future efforts to refine the proposed instrument.

Note: For the sake of brevity, the next page of the original paper was cut
from this sample document.
A NOVEL TEACHER EVALUATION MODEL 6

Methods section, which follows, will propose a pilot study that compares the results of the

proposed instrument to the results of a traditional SET (and will also provide necessary

background information on both of these evaluations). The paper will conclude with a discussion

of how the results of the pilot study will inform future iterations of the proposed instrument and,

more broadly, how universities should argue for local development of assessments.
Second-level headings are flush left, bolded, and
Literature Review written in title case.
Third level headings are flush left, bolded, written in
Effective Teaching: A Contextual Construct title case, and italicized.

The validity of the instrument this paper proposes is contingent on the idea that it is

possible to systematically measure a teacher’s ability to teach. Indeed, the same could be said

for virtually all teacher evaluations. Yet despite the exceeding commonness of SETs and the

faculty development programs that depend on their input, there is little scholarly consensus on

precisely what constitutes “good” or “effective” teaching. It would be impossible to review the

entire history of the debate surrounding teaching effectiveness, owing to its sheer scope—such

a summary might need to begin with, for instance, Cicero and Quintilian. However, a cursory

overview of important recent developments (particularly those revealed in meta-analyses of

empirical studies of teaching) can help situate the instrument this paper proposes in relevant
Fourth-level headings are bolded and written in title case. They are
academic conversations. also indented and written in-line with the following paragraph.

Meta-analysis 1. One core assumption that undergirds many of these conversations is When
presenting
the notion that good teaching has effects that can be observed in terms of student achievement. decimal
fractions, put
A meta-analysis of 167 empirical studies that investigated the effects of various teaching factors a zero in
front of the
decimal if the
on student achievement (Kyriakides et al., 2013) supported the effectiveness of a set of
quantity is
something
teaching factors that the authors group together under the label of the “dynamic model” of
that can
exceed one
teaching. Seven of the eight factors (Orientation, Structuring, Modeling, Questioning, (like the
number of
Assessment, Time Management, and Classroom as Learning Environment) corresponded to standard
deviations
moderate average effect sizes (of between 0.34–0.41 standard deviations) in measures of here). Do not
put a zero if
the quantity
cannot
exceed one
(e.g., if the
number is a
proportion).
A NOVEL TEACHER EVALUATION MODEL 7

student achievement. The eighth factor, Application (defined as seatwork and small-group tasks

oriented toward practice of course concepts), corresponded to only a small yet still significant

effect size of 0.18. The lack of any single decisive factor in the meta-analysis supports the idea

that effective teaching is likely a multivariate construct. However, the authors also note the

context-dependent nature of effective teaching. Application, the least-important teaching factor

overall, proved more important in studies examining young students (p. 148). Modeling, by

contrast, was especially important for older students.

Meta-analysis 2. A different meta-analysis that argues for the importance of factors like

clarity and setting challenging goals (Hattie, 2009) nevertheless also finds that the effect sizes

of various teaching factors can be highly context-dependent. For example, effect sizes for

homework range from 0.15 (a small effect) to 0.64 (a moderately large effect) based on the level

of education examined. Similar ranges are observed for differences in academic subject (e.g.,

math vs. English) and student ability level. As Snook et al. (2009) note in their critical response

to Hattie, while it is possible to produce a figure for the average effect size of a particular

teaching factor, such averages obscure the importance of context.

Meta-analysis 3. A final meta-analysis (Seidel & Shavelson, 2007) found generally

small average effect sizes for most teaching factors—organization and academic domain-

specific learning activities showed the biggest cognitive effects (0.33 and 0.25, respectively).

Here, again, however, effectiveness varied considerably due to contextual factors like domain of

study and level of education in ways that average effect sizes do not indicate.

These pieces of evidence suggest that there are multiple teaching factors that produce

measurable gains in student achievement and that the relative importance of individual factors

can be highly dependent on contextual factors like student identity. This is in line with a well-

documented phenomenon in educational research that complicates attempts to measure

teaching effectiveness purely in terms of student achievement. This is that “the largest source of

variation in student learning is attributable to differences in what students bring to school - their
A NOVEL TEACHER EVALUATION MODEL 8

abilities and attitudes, and family and community” (McKenzie et al., 2005, p. 2). Student

achievement varies greatly due to non-teacher factors like socio-economic status and home life

(Snook et al., 2009). This means that, even to the extent that it is possible to observe the

effectiveness of certain teaching behaviors in terms of student achievement, it is difficult to set

generalizable benchmarks or standards for student achievement. Thus is it also difficult to make

true apples-to-apples comparisons about teaching effectiveness between different educational

To list a few
contexts: due to vast differences between different kinds of students, a notion of what sources as
examples of
constitutes highly effective teaching in one context may not in another. This difficulty has a larger body
of work, you
featured in criticism of certain meta-analyses that have purported to make generalizable claims can use the
word "see" in
about what teaching factors produce the biggest effects (Hattie, 2009). A variety of other the
parenthetical,
commentators have also made similar claims about the importance of contextual factors in as we've
done here.
teaching effectiveness for decades (see, e.g., Theall, 2017; Cashin, 1990; Bloom et al., 1956)

The studies described above mainly measure teaching effectiveness in terms of

academic achievement. It should certainly be noted that these quantifiable measures are not

generally regarded as the only outcomes of effective teaching worth pursuing. Qualitative

outcomes like increased affinity for learning and greater sense of self-efficacy are also important

learning goals. Here, also, local context plays a large role.

SETs: Imperfect Measures of Teaching

As noted in this paper’s introduction, SETs are commonly used to assess teaching

performance and inform faculty development efforts. Typically, these take the form of an end-of-

term summative evaluation comprised of multiple-choice questions (MCQs) that allow students

to rate statements about their teachers on Likert scales. These are often accompanied with

short-answer responses which may or may not be optional.

SETs serve important institutional purposes. While commentators have noted that there

are crucial aspects of instruction that students are not equipped to judge (Benton & Young,

2018), SETs nevertheless give students a rare institutional voice. They represent an opportunity
A NOVEL TEACHER EVALUATION MODEL 9

to offer anonymous feedback on their teaching experience and potentially address what they

deem to be their teacher’s successes or failures. Students are also uniquely positioned to offer

meaningful feedback on an instructors’ teaching because they typically have much more

extensive firsthand experience of it than any other educational stakeholder. Even peer

observers only witness a small fraction of the instructional sessions during a given semester.

Students with perfect attendance, by contrast, witness all of them. Thus, in a certain sense, a

student can theoretically assess a teacher’s ability more authoritatively than even peer mentors

can.

While historical attempts to validate SETs have produced mixed results, some studies

have demonstrated their promise. Howard (1985), for instance, finds that SET are significantly

more predictive of teaching effectiveness than self-report, peer, and trained-observer

assessments. A review of several decades of literature on teaching evaluations (Watchel, 1998)

found that a majority of researchers believe SETs to be generally valid and reliable, despite

occasional misgivings. This review notes that even scholars who support SETs frequently argue

that they alone cannot direct efforts to improve teaching and that multiple avenues of feedback

are necessary (Seldin, 1993; L’hommedieu et al., 1990).

Finally, SETs also serve purposes secondary to the ostensible goal of improving

instruction that nonetheless matter. They can be used to bolster faculty CVs and assign

departmental awards, for instance. SETs can also provide valuable information unrelated to

teaching. It would be hard to argue that it not is useful for a teacher to learn, for example, that a

student finds the class unbearably boring, or that a student finds the teacher’s personality so

unpleasant as to hinder her learning. In short, there is real value in understanding students’

affective experience of a particular class, even in cases when that value does not necessarily

lend itself to firm conclusions about the teacher’s professional abilities.

However, a wealth of scholarly research has demonstrated that SETs are prone to fail in

certain contexts. A common criticism is that SETs can frequently be confounded by factors
A NOVEL TEACHER EVALUATION MODEL 10

external to the teaching construct. The best introduction to the research that serves as the basis

for this claim is probably Neath (1996), who performs something of a meta-analysis by This citation
presents
presenting these external confounds in the form of twenty sarcastic suggestions to teaching quotations
from different
faculty. Among these are the instructions to “grade leniently,” “administer ratings before tests” locations in
the original
source. Each
(p. 1365), and “not teach required courses” (#11) (p. 1367). Most of Neath’s advice reflects an
quotation is
followed by
overriding observation that teaching evaluations tend to document students’ affective feelings the
corresponding
toward a class, rather than their teachers’ abilities, even when the evaluations explicitly ask page number.

students to judge the latter.

Beyond Neath, much of the available research paints a similar picture. For example, a

study of over 30,000 economics students concluded that “the poorer the student considered his

teacher to be [on an SET], the more economics he understood” (Attiyeh & Lumsden, 1972). A

1998 meta-analysis argued that “there is no evidence that the use of teacher ratings improves

learning in the long run” (Armstrong, 1998, p. 1223). A 2010 National Bureau of Economic

Research study found that high SET scores for a course’s instructor correlated with “high

contemporaneous course achievement,” but “low follow-on achievement” (in other words, the

students would tend to do well in the course, but poor in future courses in the same field of

study. Others observing this effect have suggested SETs reward a pandering, “soft-ball”

teaching style in the initial course (Carrell & West, 2010). More recent research suggests that

course topic can have a significant effect on SET scores as well: teachers of “quantitative

courses” (i.e., math-focused classes) tend to receive lower evaluations from students than their

humanities peers (Uttl & Smibert, 2017).

Several modern SET studies have also demonstrated bias on the basis of gender

(Basow, 1995; Anderson & Miller, 1997), physical appearance/sexiness (Ambady & Rosenthal,

1993), and other identity markers that do not affect teaching quality. Gender, in particular, has

attracted significant attention. One recent study examined two online classes: one in which

instructors identified themselves to students as male, and another in which they identified as
A NOVEL TEACHER EVALUATION MODEL 11

female (regardless of the instructor’s actual gender) (Macnell et al., 2015). The classes were

identical in structure and content, and the instructors’ true identities were concealed from

students. The study found that students rated the male identity higher on average. However, a

few studies have demonstrated the reverse of the gender bias mentioned above (that is, women

received higher scores) (Bachen et al., 1999) while others have registered no gender bias one

way or another (Centra & Gaubatz, 2000).

The goal of presenting these criticisms is not necessarily to diminish the institutional

importance of SETs. Of course, insofar as institutions value the instruction of their students, it is

important that those students have some say in the content and character of that instruction.

Rather, the goal here is simply to demonstrate that using SETs for faculty development

purposes—much less for personnel decisions—can present problems. It is also to make the

case that, despite the abundance of literature on SETs, there is still plenty of room for scholarly

attempts to make these instruments more useful.

Empirical Scales and Locally-Relevant Evaluation

One way to ensure that teaching assessments are more responsive to the demands of

teachers’ local contexts is to develop those assessments locally, ideally via a process that

involves the input of a variety of local stakeholders. Here, writing assessment literature offers a

promising path forward: empirical scale development, the process of structuring and calibrating

instruments in response to local input and data (e.g., in the context of writing assessment,

student writing samples and performance information). This practice contrasts, for instance, with

deductive approaches to scale development that attempt to represent predetermined theoretical

constructs so that results can be generalized.

Supporters of the empirical process argue that empirical scales have several

advantages. They are frequently posited as potential solutions to well-documented reliability and

validity issues that can occur with theoretical or intuitive scale development (Turner and Upshur,

1995; Turner and Upshur, 2002; Brindley, 1998). Empirical scales can also avoid issues caused
Quotations
longer than A NOVEL TEACHER EVALUATION MODEL 12
40 words
should be
formatted as by subjective or vaguely-worded standards in other kinds of scales (Brindley, 1998) because
block
quotations. they require buy-in from local stakeholders who must agree on these standards based on their
Indent the
entire understanding of the local context. Fulcher, Davidson, and Kemp (2011) note the following:
passage half
an inch and
Measurement-driven scales suffer from descriptional inadequacy. They are not sensitive
present the
passage
without to the communicative context or the interactional complexities of language use. The level
quotation
marks. Any of abstraction is too great, creating a gulf between the score and its meaning. Only with
relevant
page a richer description of contextually based performance, can we strengthen the meaning
numbers
should follow of the score, and hence the validity of score-based inferences. (pp. 8–9)
the
concluding There is also some evidence that the branching structure of the EBB scale specifically
punctuation
mark. If the can allow for more reliable and valid assessments, even if it is typically easier to calibrate and
author and/ When citing
or date are multiple
use conventional scales (Hirai and Koizumi, 2013). Finally, scholars have also argued that
not sources from
referenced the same
theory-based approaches to scale development do not always result in instruments that
in the text, author(s),
as they are simply list the
realistically capture ordinary classroom situations (Knoch, 2007, 2009). author(s),
here, place
them in the then list the
parenthetical
The most prevalent criticism of empirical scale development in the literature is that the years of the
that follows sources
the quotation local, contingent nature of empirical scales basically discards any notion of their results’ separated by
along with commas.
the page generalizability. Fulcher (2003), for instance, makes this basic criticism of the EBB scale even
numbers.
as he subsequently argues that “the explicitness of the design methodology for EBBs is

impressive, and their usefulness in pedagogic settings is attractive” (p. 107). In the context of

this particular paper’s aims, there is also the fact that the literature supporting empirical scale

development originates in the field of writing assessment, rather than teaching assessment.

Moreover, there is little extant research into the applications of empirical scale development for

the latter purpose. Thus, there is no guarantee that the benefits of empirical development

approaches can be realized in the realm of teaching assessment. There is also no guarantee

that they cannot. In taking a tentative step towards a better understanding of how these

assessment schema function in a new context, then, the study described in the next section
A NOVEL TEACHER EVALUATION MODEL 13

asks whether the principles that guide some of the most promising practices for assessing

students cannot be put to productive use in assessing teachers.

Materials and Methods

This section proposes a pilot study that will compare the ICaP SET to the Heavilon

Evaluation of Teacher (HET), an instrument designed to combat the statistical ceiling effect

described above. In this section, the format and composition of the HET is described, with

special attention paid to its branching scale design. Following this, the procedure for the study is

outlined, and planned interpretations of the data are discussed.

The Purdue ICaP SET

The SET employed by Introductory Composition at Purdue (ICaP) program as of

January 2019 serves as an example of many of the prevailing trends in current SET

administration. The evaluation is administered digitally: ICaP students receive an invitation to

complete the evaluation via email near the end of the semester, and must complete it before

finals week (i.e., the week that follows the normal sixteen-week term) for their responses to be

counted. The evaluation is entirely optional: teachers may not require their students to complete

it, nor may they offer incentives like extra credit as motivation. However, some instructors opt to

devote a small amount of in-class time for the evaluations. In these cases, it is common practice

for instructors to leave the room so as not to coerce high scores.

The ICaP SET mostly takes the form of a simple multiple-choice survey. Thirty-four

MCQs appear on the survey. Of these, the first four relate to demographics: students must

indicate their year of instruction, their expected grade, their area of study, and whether they are

taking the course as a requirement or as an elective. Following these are two questions related

to the overall quality of the course and the instructor (students must rate each from “very poor”

to “excellent” on a five-point scale). These are “university core” questions that must appear on

every SET administered at Purdue, regardless of school, major, or course. The Students are
 A NOVEL TEACHER EVALUATION MODEL 14
Italicize the
anchors of
scales or also invited to respond to two short-answer prompts: “What specific suggestions do you have for
responses to
scale-like improving the course or the way it is taught?” and “what is something that the professor does
questions,
rather than well?” Responses to these questions are optional.
presenting
them in The remainder of the MCQs (thirty in total) are chosen from a list of 646 possible
quotation
marks. Do questions provided by the Purdue Instructor Course Evaluation Service (PICES) by department
not italicize
numbers if
administrators. Each of these PICES questions requires students to respond to a statement
the scale
responses
are
about the course on a five-point Likert scale. Likert scales are simple scales used to indicate
numbered.
degrees of agreement. In the case of the ICaP SET, students must indicate whether they

strongly agree, agree, disagree, strongly disagree, or are undecided. These thirty Likert scale

questions assess a wide variety of the course and instructor’s qualities. Examples include “My

instructor seems well-prepared for class,” “This course helps me analyze my own and other

students' writing,” and “When I have a question or comment I know it will be respected,” for

example.

One important consequence of the ICaP SET within the Purdue English department is

the Excellence in Teaching Award (which, prior to Fall 2018, was named the Quintilian or,

colloquially, “Q” Award). This is a symbolic prize given every semester to graduate instructors

who score highly on their evaluations. According to the ICaP site, “ICaP instructors whose

teaching evaluations achieve a certain threshold earn [the award], recognizing the top 10% of

teaching evaluations at Purdue.” While this description is misleading—the award actually goes

to instructors whose SET scores rank in the top decile in the range of possible outcomes, but

not necessarily ones who scored better than 90% of other instructors—the award nevertheless

provides an opportunity for departmental instructors to distinguish their CVs and teaching

portfolios.

Insofar as it is distributed digitally, it is composed of MCQs (plus a few short-answer

responses), and it is intended as end-of-term summative assessment, the ICaP SET embodies
A NOVEL TEACHER EVALUATION MODEL 15

the current prevailing trends in university-level SET administration. In this pilot study, it serves

as a stand-in for current SET administration practices (as generally conceived).

The HET

Like the ICaP SET, the HET uses student responses to questions to produce a score

that purports to represent their teacher’s pedagogical ability. It has a similar number of items

(28, as opposed to the ICaP SET’s 34). However, despite these superficial similarities, the

instrument’s structure and content differ substantially from the ICaP SET’s.

The most notable differences are the construction of the items on the text and the way

that responses to these items determine the teacher’s final score. Items on the HET do not use

the typical Likert scale, but instead prompt students to respond to a question with a simple

“yes/no” binary choice. By answering “yes” and “no” to these questions, student responders

navigate a branching “tree” map of possibilities whose endpoints correspond to points on a 33-

point ordinal scale.

The items on the HET are grouped into six suites according to their relevance to six

different aspects of the teaching construct (described below). The suites of questions

correspond to directional nodes on the scale—branching paths where an instructor can move

either “up” or “down” based on the student’s responses. If a student awards a set number of

“yes” responses to questions in a given suite (signifying a positive perception of the instructor’s

teaching), the instructor moves up on the scale. If a student does not award enough “yes”

responses, the instructor moves down. Thus, after the student has answered all of the

questions, the instructor’s “end position” on the branching tree of possibilities corresponds to a

point on the 33-point scale. A visualization of this structure is presented in Table 1.

 A NOVEL TEACHER EVALUATION MODEL 16

Tables and figures are numbered sequentially (i.e., 1, 2,

Figure 1 3 ...). They are identified via a second-level heading (flush-
left, bold, and title case) followed by an italic title that briefly
Illustration of HET’s Branching Structure describes the content of the table or figure.

Note. Each node in this diagram corresponds to a suite of HET/ICALT items, rather than to a single item.

The questions on the HET derive from the International Comparative Analysis of

Learning and Teaching (ICALT), an instrument that measures observable teaching behaviors for

Table and figure notes are preceded by the label "Note." written in italics. General notes that apply to
the entire table should come before specific notes (indicated with superscripted lowercase letters that
correspond to specific locations in the figure or table.
Table notes are optional.
A NOVEL TEACHER EVALUATION MODEL 17

the purpose of international pedagogical research within the European Union. The most recent

version of the ICALT contains 32 items across six topic domains that correspond to six broad

teaching skills. For each item, students rate a statement about the teacher on a four-point Likert

scale. The main advantage of using ICALT items in the HET is that they have been

independently tested for reliability and validity numerous times over 17 years of development

(see, e.g., Van de Grift, 2007). Thus, their results lend themselves to meaningful comparisons

between teachers (as well as providing administrators a reasonable level of confidence in their

ability to model the teaching construct itself).

The six “suites” of questions on the HET, which correspond to the six topic domains on
Tables are formatted similarly to figures. They are titled and
the ICALT, are presented in Table 1.
numbered in the same way, and table-following notes are
presented the same way as figure-following notes. Use separate
Table 1 sequential numbers for tables and figures. For instance, this
table is presented as Table 1 rather than as Table 2, despite the
HET Question Suites fact that Figure 1 precedes it.
Suite # of Items Description

Safe learning environment 4 Whether the teacher is able to

maintain positive, nonthreatening

relationships with students (and to

foster these sorts of relationships

among students).

Classroom management 4 Whether the teacher is able to

maintain an orderly, predictable

environment.

Clear instruction 7 Whether the teacher is able to

explain class topics

comprehensibly, provide clear sets

of goals for assignments, and

articulate the connections between

the assignments and the class

topics in helpful ways.

 A NOVEL TEACHER EVALUATION MODEL 18

Suite # of Items Description

Activating teaching methods 7 Whether the teacher uses strategies

that motivate students to think about

When a table
the class’s topics.
is so long that
it stretches Learning strategies 6 Whether teachers take explicit steps
across
multiple to teach students how to learn (as
pages, repeat
the column opposed to merely providing
labels on
students informational content).
each new
page. Most Differentiation 4 Whether teachers can successfully
word
processors adjust their behavior to meet the
have a
feature that diverse learning needs of individual
does this
students.
automatically.

Note. Item numbers are derived from original ICALT item suites.

The items on the HET are modified from the ICALT items only insofar as they are phrased In addition to
presenting
as binary choices, rather than as invitations to rate the teacher. Usually, this means the addition figures and
tables in the
of the word “does” and a question mark at the end of the sentence. For example, the second text, you may
also present
safe learning climate item on the ICALT is presented as “The teacher maintains a relaxed them in
appendices
atmosphere.” On the HET, this item is rephrased as, “Does the teacher maintain a relaxed at the end of
the
document.
atmosphere?” See Appendix for additional sample items.
You may also
use
As will be discussed below, the ordering of item suites plays a decisive role in the teacher’s appendices
to present
final score because the branching scale rates earlier suites more powerfully. So too does the material that
would be
“sensitivity” of each suite of items (i.e., the number of positive responses required to progress distracting or
tedious in the
upward at each branching node). This means that it is important for local stakeholders to body of the
paper. In
participate in the development of the scale. In other words, these stakeholders must be involved either case,
you can use
in decisions about how to order the item suites and adjust the sensitivity of each node. This is simple in-text
references to
described in more detail below. direct
readers to
Once the scale has been developed, the assessment has been administered, and the the
appendices.
teacher’s endpoint score has been obtained, the student rater is prompted to offer any textual
A NOVEL TEACHER EVALUATION MODEL 19

feedback that s/he feels summarizes the course experience, good or bad. Like the short

response items in the ICaP SET, this item is optional. The short-response item is as follows:

• What would you say about this instructor, good or bad, to another student considering

taking this course?

The final four items are demographic questions. For these, students indicate their grade

level, their expected grade for the course, their school/college (e.g., College of Liberal Arts,

School of Agriculture, etc.), and whether they are taking the course as an elective or as a

degree requirement. These questions are identical to the demographic items on the ICaP SET.

To summarize, the items on the HET are presented as follows:

• Branching binary questions (32 different items; six branches)

o These questions provide the teacher’s numerical score

• Short response prompt (one item)

• Demographic questions (four items)

Scoring

The main data for this instrument are derived from the endpoints on a branching ordinal

scale with 33 points. Because each question is presented as a binary yes/no choice (with “yes”

suggesting a better teacher), and because paths on the branching scale are decided in terms of

whether the teacher receives all “yes” responses in a given suite, 32 possible outcomes are

possible from the first five suites of items. For example, the worst possible outcome would be

five successive “down” branches, the second-worst possible outcome would be four “down”

branches followed by an “up,” and so on. The sixth suite is a tie-breaker: instructors receive a

single additional point if they receive all “yes” responses on this suite.

By positioning certain suites of items early in the branching sequence, the HET gives

them more weight. For example, the first suite is the most important of all: an “up” here

automatically places the teacher above 16 on the scale, while a “down” precludes all scores

Note: For the sake of brevity, the next few pages of the original paper
were cut from this sample document.
 A NOVEL TEACHER EVALUATION MODEL 26 Start the
references
Source with list on a new
two authors. References page. The
word
"References"
Ambady, N., & Rosenthal, R. (1993). Half a minute: Predicting teacher evaluations from thin (or
"Reference,"
slices of nonverbal behavior and physical attractiveness. Journal of Personality and if there is
only one
Social Psychology, 64(3), 431–441. http://dx.doi.org/10.1037/0022-3514.64.3.431 source),
should
Source with American Association of University Professors. (n.d.) Background facts on contingent faculty appear
organizational bolded and
author. positions. https://www.aaup.org/issues/contingency/background-facts centered at
the top of the
page.
American Association of University Professors. (2018, October 11). Data snapshot: Contingent Reference
entries
faculty in US higher ed. AAUP Updates. https://www.aaup.org/news/data-snapshot-
should follow
contingent-faculty-us-higher-ed#.Xfpdmy2ZNR4 in
alphabetical
Anderson, K., & Miller, E. D. (1997). Gender and student evaluations of teaching. PS: Political order. There
should be a
Science and Politics, 30(2), 216–219. https://doi.org/10.2307/420499 reference
entry for
every source
Armstrong, J. S. (1998). Are student ratings of instruction useful? American Psychologist,
cited in the
text.
53(11), 1223–1224. http://dx.doi.org/10.1037/0003-066X.53.11.1223
Note that
Attiyeh, R., & Lumsden, K. G. (1972). Some modern myths in teaching economics: The U.K. sources in
online
experience. American Economic Review, 62(1), 429–443. academic
publications
https://www.jstor.org/stable/1821578 like scholarly
journals now
require DOIs
Bachen, C. M., McLoughlin, M. M., & Garcia, S. S. (1999). Assessing the role of gender in
or stable
URLs if they
college students' evaluations of faculty. Communication Education, 48(3), 193–210. are available.
Shortened
All citation http://doi.org/cqcgsr DOI.
entries
should be Basow, S. A. (1995) Student evaluations of college professors: When gender matters. Journal
double-
spaced. After of Educational Psychology, 87(4), 656–665. http://dx.doi.org/10.1037/0022-
the first line
of each entry, 0663.87.4.656
every
following line Becker, W. (2000). Teaching economics in the 21st century. Journal of Economic Perspectives,
should be
indented a 14(1), 109–120. http://dx.doi.org/10.1257/jep.14.1.109
half inch (this
is called a
Benton, S., & Young, S. (2018). Best practices in the evaluation of teaching. Idea paper, 69.
"hanging
indent").
 A NOVEL TEACHER EVALUATION MODEL 27

Berk, R. A. (2005). Survey of 12 strategies to measure teaching effectiveness. International

Print book. Journal of Teaching and Learning in Higher Education, 17(1), 48–62.

Bloom, B. S., Englehart, M. D., Furst, E. J., Hill, W. H., & Krathwohl, D. R. (1956). Taxonomy of

educational objectives: The classification of educational goals. Addison-Wesley

Longman Ltd.

Brandenburg, D., Slinde, C., & Batista, J. (1977). Student ratings of instruction: Validity and

normative interpretations. Research in Higher Education, 7(1), 67–78.

http://dx.doi.org/10.1007/BF00991945

Carrell, S., & West, J. (2010). Does professor quality matter? Evidence from random

assignment of students to professors. Journal of Political Economy, 118(3), 409–432.

https://doi.org/10.1086/653808

Cashin, W. E. (1990). Students do rate different academic fields differently. In M. Theall, & J. L.

Franklin (Eds.), Student ratings of instruction: Issues for improving practice. New
Chapter in an
Directions for Teaching and Learning (pp. 113–121). edited
collection.
Centra, J., & Gaubatz, N. (2000). Is there gender bias in student evaluations of

teaching? The Journal of Higher Education, 71(1), 17–33.

Second
edition of a https://doi.org/10.1080/00221546.2000.11780814
print book.
Davis, B. G. (2009). Tools for teaching (2nd ed.). Jossey-Bass.
Academic
Denton, D. (2013). Responding to edTPA: Transforming practice or applying article for
which a DOI
shortcuts? AILACTE Journal, 10(1), 19–36. was
unavailable.
Dizney, H., & Brickell, J. (1984). Effects of administrative scheduling and directions upon

student ratings of instruction. Contemporary Educational Psychology, 9(1), 1–7.

https://doi.org/10.1016/0361-476X(84)90001-8

DuCette, J., & Kenney, J. (1982). Do grading standards affect student evaluations of teaching?

Some new evidence on an old question. Journal of Educational Psychology, 74(3), 308–

314. https://doi.org/10.1037/0022-0663.74.3.308
A NOVEL TEACHER EVALUATION MODEL 28

Edwards, J. E., & Waters, L. K. (1984). Halo and leniency control in ratings as influenced by

format, training, and rater characteristic differences. Managerial Psychology, 5(1), 1–16.

Fink, L. D. (2013). The current status of faculty development internationally. International

Journal for the Scholarship of Teaching and Learning, 7(2).

https://doi.org/10.20429/ijsotl.2013.070204

Fulcher, G. (2003). Testing second language speaking. Pearson Education.

Fulcher, G., Davidson, F., & Kemp, J. (2011). Effective rating scale development for speaking

tests: Performance decision trees. Language Testing, 28(1), 5–29.

https://doi.org/10.1177/0265532209359514

Gaff, J. G., & Simpson, R. D. (1994). Faculty development in the United States. Innovative

Higher Education, 18(3), 167–76. https://doi.org/10.1007/BF01191111

Hattie, J. (2008). Visible learning: A synthesis of over 800 meta-analyses relating to

achievement. Routledge.

Hoffman, R. A. (1983). Grade inflation and student evaluations of college courses. Educational

and Psychological Research, 3(3), 51–160. https://doi.org/10.1023/A:101557981

Howard, G., Conway, C., & Maxwell, S. (1985). Construct validity of measures of college

teaching effectiveness. Journal of Educational Psychology, 77(2), 187–96.

http://dx.doi.org/10.1037/0022-0663.77.2.187

Kane, M. T. (2013) Validating interpretations and uses of test scores. Journal of Educational

Measurement, 50(1), 1–73.

Kelley, T. (1927) Interpretation of educational measurements. World Book Co.

Knoch, U. (2007). Do empirically developed rating scales function differently to conventional

rating scales for academic writing? Spaan Fellow Working Papers in Second or Foreign

Language Assessment, 5, 1–36. English Language Institute, University of Michigan.

Knoch, U. (2009). Diagnostic assessment of writing: A comparison of two rating

scales. Language Testing, 26(2), 275-304.

Note: For the sake of brevity, the next few pages of the original paper
were cut from this sample document.
A NOVEL TEACHER EVALUATION MODEL 32
Appendices begin after the references list. The word
"Appendix" should appear at the top of the page, bolded
Appendix and centered. If there are multiple appendices, label them
with capital letters (e.g., Appendix A, Appendix B, and
Sample ICALT Items Rephrased for HET Appendix C). Start each appendix on a new page.

Suite Sample ICALT Item HET Phrasing

Safe learning environment The teacher promotes mutual Does the teacher promote mutual

respect. respect?

Classroom management The teacher uses learning time Does the teacher use learning time

efficiently. efficiently?

Clear instruction The teacher gives feedback to Does the teacher give feedback to

pupils. pupils?

Activating teaching methods The teacher provides interactive Does the teacher provide interactive

instruction and activities. instruction and activities?

Learning strategies The teacher provides interactive Does the teacher provide interactive

instruction and activities. instruction and activities?

Differentiation The teacher adapts the instruction Does the teacher adapt the

to the relevant differences between instruction to the relevant

pupils. differences between pupils?

Paragraphs of text can also appear in appendices.

If they do, paragraphs should be indented
normally, as they are in the body of the paper.

If an appendix contains only a single table or

figure, as this one does, the centered and bolded
"Appendix" replaces the centered and bolded label
that normally accompanies a table or figure.

If the appendix contains both text and tables or

figures, the tables or figures should be labeled,
and these labels should include the letter of the
appendix in the label. For example, if Appendix A
contains two tables and one figure, they should be
labeled "Table A1," "Table A2," and "Figure A1." A
table that follows in Appendix B should be labeled
"Table B1." If there is only one appendix, use the
letter "A" in table/figure labels: "Table A1," "Table
A2," and so on.