Skip to content

Teaching as Research

Good teaching is reflective and thoughtful, and every semester affords lessons for improvement. The traditional mechanisms by which teachers assess their work — summative student assessment, end-of-term evaluations, peer and supervisor observations, and self-appraising reflection — all can provide valuable avenues for growth. However, these methods are often insufficient for accurately measuring student learning and the effectiveness of pedagogical practice. One way of better judge these issues is by approaching them through systematic research, using the classroom as a lab-space in which to judge my hypotheses about my own teaching against specific quantifiable data sets.

I have recently begun using Teaching as Research (TAR) as a method for assessing the usefulness of specific, targeted activity sets in my First-Year Writing Seminars. I generally approach these classes with several series of lessons targeting distinct writing issues, and to this point I have relied on end-of-term student feedback and on my own intuition to assess their value. Taking a TAR approach has allowed me to see specific places where my efforts have born fruit, and also those places where they need improvement. This approach has also produced some surprising results that suggest better approaches to teaching.

The following is one example of a TAR project, that I undertook in my Spring 2018 FWS class, examining the effectiveness of my efforts at mitigating my students’ overuse of the verb “to be.” The study was approved by Cornell’s Institutional Review Board, and all participants have given permission for their data to be made public. I presented a poster of this project at the 2018 Connecting Teaching and Research Conference.

The poster is available here.

To see materials related to a specific section of the study, click the inverted caret (  ) next to the course semester that you wish to examine.

To Be or Not Too Be: Efficacy of Focused Efforts at Correcting Overuse of ‘To Be’ in a 2018 First-Year Writing Seminar

The Problem

One of the most consistent problems afflicting first-year writing is overuse of the verb “to be.” Weaker writers tend to rely heavily on variants of “to be” – is, am, are, was, were, be, been – in order to move their writing forward.   

While some disciplines and writing styles employ “to be” as a standard, in argumentative essay writing overuse creates two primary problems:

      1. It deadens writing through repetition and a narrowing of the range of verb usage, which creates challenges for retaining audience attention
    1. It deprives “to be” of its potential impact, losing its value as a textual equals sign

Helping students to learn how to recognize and limit their overuse of “to be” presents a significant challenge to any teacher of a writing class. Importantly, the goal is not to teach students to never use “to be” in their writing. Rather, students need to gain awareness and control of their verb usage, and learn to deploy “to be” in their writing with intention and to good effect.

The Study

The purpose of this study lies in gauging the effectiveness of my interventions to correct “to be” overuse in one class, over the course of a single semester: a Medieval Studies FWS on medieval geography titled Orbis Terrarum: The Medieval World was a Globe, offered in the Spring semester of the 2017-2018 academic year at Cornell.    

Specifically, I am interested in assessing the following four aspects:

    • the degree to which my efforts have affected my students’ understanding of the problems of “to be” overuse in writing generally
  • to what degree student writing changed over the course of the semester with regard to “to be” overuse
  • whether or not specific interventions provided more benefit than others
  • how useful students found different interventions for addressing overuse in their own writing

Data Types and Collection

This study takes a two-pronged approach, considering both quantitative data from student essays and self-reported survey data gathered from 14 participants.    

Usage Rates in Essays

Students wrote five formal essays during the semester. I have taken each essay and, excluding quotes drawn from external sources, counted the total number of “to be” verbs. I used this result to calculate a rate of “to be” usage per 100 words, again excluding quoted materials, and then charted change in usage over time.

Student Survey

At the conclusion of the semester my students completed a six-question survey that asked them to rate, on a scale of 1-4, their awareness of “to be” overuse as a general problem in writing at the beginning and end of the semester, and the effectiveness of class activities and instructor feedback in increasing their understanding of the issue.

Points of Intervention 

Over the course of the semester I made seven separate interventions in student writing related to overuse of “to be”:     

1. In-Class Introduction:  Before Essay 2
I introduced the issues related to “to be” overuse in the first part of the semester. I led a class discussion about the problems of overuse and the value of “to be”, and showed anonymous examples of both overuse and appropriate use drawn from their initial essays

2. “To Be” Informal Paper:  Before Essay 2
Students wrote a one-page informal paper in which they could only use “to be”, and no other verbs. They brought this paper to class and worked with a partner to rewrite the paper, replacing extraneous “to be” verbs and identifying those places where “to be” made sense within the context of their papers. We then had a class discussion about the exercise, and looked at examples from their papers

3. Passive Voice Informal Paper:  Before Essay 2
This exercise took the same form as the first informal paper, but asked students to write the assignment using only the passive voice. Passive constructions rely on “to be” to operate, so addressing passive voice overuse also draws attention to “to be” usage

4. Peer Examples:  Before Essays 3 and 4
Students critiqued examples of “to be” overuse drawn anonymously from their classmates’ previous essays

5. Peer Draft Editing:  Before Essays 3 and 4
As part of their in-class peer editing of drafts for essays 3 and 4, I required students to underline each instance of “to be” in their partner’s essay, and suggest alternative verbs and/or sentence structures

6. Instructor-Produced Examples:  Before Essays 4 and 5
Students critiqued “to be” overuse in example sentences written by the instructor

7. Direct instructor feedback:  Essays 2, 3, and 4
Direct feedback on each essay highlighted every instance of “to be” and included both end comments and marginal notes about students’ use of “to be”


The data from both branches of the study look like this:     

Fig. 1: Total “to be” usage, word count, and usage rates per 100 words in student essays
Fig. 2: Individual student “to be” usage graphed over the semester, showing class averageFig. 3: Survey Questions and Averaged Results 

Analysis and Conclusion

The data shows definite trends, and suggests both specific conclusions and avenues for future thought:    

Usage Rate

1. In total, the class saw a marked decrease in “to be” usage, suggesting overall effectiveness of the interventions

2. Informal Papers were less immediately effective than I had anticipated

    • Both activities preceded an increase in average usage rate in the study, from 3.98 to 4.12 instances per 100 words
  • This is somewhat surprising, given that in past classes students have singled out these activities as being the most useful in helping their writing

3. Peer-based activities correlate most obviously with successful reduction in usage rate

    • The essays showing the lowest usage rates are those for which students had directed peer feedback on “to be” at the draft stage
  • Essay 5, which had no such peer feedback, shows an increase in rate

4. Essay 5 results suggest a cumulative effect

    • Results show significant improvements from Essays 1 and 2, which also had no directed peer feedback in the draft process
    • In all three essays students were left to their own devises in the writing process to mitigate “to” usage
    • This decrease in usage rates point to a learned and applied awareness of “to be” overuse in their writing
  • Earlier interventions, even where they appear to have negative effects in the data, set the stage for the successes of later endeavors

5. I was surprised by the low rates of usage, even at the start of the semester

    • The low rates mean a lower percentage of desired change in student writing than I was anticipating
    • This suggests a need for a greater degree of empathy for student pushback on this issue on my part in future classes
  • It also suggests an avenue for explanation to STEM students, who should recognize the potential impact of a 1% change to the outcome of a calculation or the composition of a solution

Student Survey

1. Efforts over the course of the semester improved students’ awareness of the issue

    • Students reported a general lack of awareness at the start of the class
    • 5 out of 13 participants chose 1 (Strongly Disagree) in response to this question
    • Students reported a marked increase in awareness at the conclusion of the class
    • 8 of 13 participants chose 4 (Strongly Agree) in response to Question 2
  • All participants chose either 3 (Somewhat Agree) or 4 (Strongly Agree) in response to this question

2. The largest changes in awareness correlate to the most significant drops in rate of usage

    • The students who identified as being least aware of the issue at the start of the semester, and most aware of the issue at the end of the semester also showed the most change in their use of “to be”
  • Students who reported the smallest changes in awareness also showed the least change in their usage rate

3. Student assessment of both class activities and instructor feedback indicate similar degrees of effectiveness

  • This is at odds with the usage rate numbers, but consistent with the idea of a cumulative effect

4. Student assessment of their own use of “to be” is unrelated to their actual usage rates

    • All participants but one chose 3 (Somewhat Agree) in response to Question 6. My sense is that this response demonstrates students’ desire to chose a safe, middle-of-the-road answer that reflects both their uncertainty in their answer and their positive feelings about their own capabilities
    • In light of these responses, it appears that a 4-option survey is too limited to capture any truly useful data related to student self-assessment
  • Future assessments should be done on a 5+ option scale