Assessing (Developing

DEVELOPING PRE-SERVICE ELEMENTARY TEACHERS’ WRITING IN A SCIENCE CAPSTONE COURSE

Gail R. Luera, University of Michigan-Dearborn

Susan A. Ever= ett, University of Michigan-Dearborn

Charlotte A. = Otto, University of Michigan-Dearborn

Abstract

This study investigated w= hether a series of scaffolded writing assignments would improve pre-service elementa= ry teachers’ writing abilities in a science context. Through the use of scaffol= ded writing assignments, pre-service elementary teachers improved their writing abilities more than a similar group of students that did not do the writing activities in several areas in an action research project. T= he students with the pre-writing activities wrote over twice as many words in relation to their total errors than the students without the pre-writing activities (z =3D -3.55; p<.000). They also had fewer punctuation errors= in relation to the total number of words in their final report (z =3D -2.12; p<.05). There were no significant differences for the other error catego= ries: multiple word (z =3D -1.69), single word (z =3D -1.90) and verb related (z = =3D -1.62). The Flesch reading ea= se, which is a readability score, showed a significant difference between the t= wo groups (z =3D -3.00, p < .05). The Flesch-Kincaid grade level was also s= ignificantly different (z =3D -3.37; p < .001). Student writing ability was measured by the relatively crude but easily accessible tool of Microsoft Wo= rd spelling, style and grammar checker.

Introduction

During the= last few decades, there has been much interest in the writing to learn movement within the science education community. Many teachers readily accept that students can use informal writing as a mode to help them learn science cont= ent. In a comprehensive review of the literature of writing to learn science, Rivard (1994) found many innovations at all levels from elementary through university. Keys (1999) provided a brief history of writing to learn in all content areas and summary of the current state of writing to learn. Keys advocates “Learning to write in traditional scientific genres is a natural outgrowth of the inquiry process, fostering a profound understanding of the connections between inquiry problems, procedu= res, data and knowledge claims” (1999, p. 119).<= /span>

What is le= ss clear both in practice and in the literature, is the issue of students learning to write to communicate scientific ideas to others. Holliday, Yore and Alverma= nn (1994) state: “Writing should glue thinking to paper, provide a public record of thinking, promote critical thinking, allow the transformation of vague ideas to clear conceptions, and stimulate the construction of understandings” (p.885). Glynn and Muth (1994, p.1065) describe the challenges that students face since “text production involves the formidable skill of communicating a mental representation by means of writt= en sentences.” Recent studies on writing to communicate science are much less common than writing to learn science. Robertson’s study (2004) of 143 pre-service elementary teachers’ science writing found numerous errors in explanation, logical structure, referencing and writing conventio= ns.

At the uni= versity level, there appear to be two approaches to addressing the issue of writing= to communicate science. One approach tends to focus on incorporating scientifi= c writing into a selected science course. Kroen (2004) developed a project to model t= he process of constructing a scientific paper with a limnology course for environmental science majors. The project was completed in three parts with both peer and instructor feedback. Tessier (2006) incorporated four differe= nt writing assignments in a nonmajor introductory ecology course. Each assignm= ent was limited in length to two double-spaced typed pages and was assessed with the same rubric to determine any increase in students’ writing skills= . Students committed fewer errors in both content and grammar and spelling by the end = of the semester.

The other = approach uses a writing course within a science department. Feldman, Anderson and Mangurian (2001) discuss the Towson Transition Course for freshman science majors that focuses on students’ work with primary research literatur= e. They emphasize both oral and written communication with extensive peer teamwork, graduate mentors and individual interactions with faculty instruc= tors. Students’ writing as well as critical-thinking scores improved. Rice = (1998) teaches a “Scientific Writing” course for science majors that includes 4 papers on student selected topics: a brief narrative, a descript= ion and explanation of a scientific principle, an argument paper and a research paper of student laboratory work. Rice states: “Since improvement in writing occurs largely through individual effort, practice and feedback, I = see my role in this course more as guide, coach, cheerleader, critic, and occasionally referee” (Rice, 1998, p. 268).

While each= of these studies shows improvement in student writing, the emphasis to teach writing was in only one course. Our approach has been to promote profession= al writing throughout our program by the use of writing assignments in each of= the six required science/science education courses. However, a review of student writing captured in our Science Education Portfolio, an electronic portfoli= o of student work, revealed a number of writing problems including numerous mechanical problems associated with writing conventions and the correct use= of the different science genre formats. Our review led us to investigate wheth= er a series of scaffolded writing assignments would improve pre-service elementa= ry teachers’ writing abilities in a science context.

Methods<= o:p>

In our program for pre-service elementary teachers, all students complete a Science Capstone course, which combines an action research project with one= of the unifying themes from the National Science Education Standards (N= RC, 1996) or the Benchmarks for Scientific Literacy (AAAS, 1993). These themes or unifying concepts are “big ideas” that cross traditio= nal science discipline boundaries. They include evolution and equilibrium; systems, ord= er and organization; form and function; change, constancy and measurement; evidence, models and explanation (NRC, 1996) and systems; models; constancy= and change; evolution, and scale (AAAS, 1993). In the capstone course, students experience several investigations focusing on one of the unifying themes, l= earn how the theme applies to multiple science disciplines and conduct an action research project that incorporates the unifying theme with elementary stude= nts. The theme of the course during this study was “scientific models̶= 1; which are “tentative schemes or structures that correspond to real objects, events, or classes of events, and that have explanatory power̶= 1; (NRC, 1996, p. 117).

The action research project focuses on a short teaching experience with elementary stu= dents, which includes teaching about the unifying theme within the local school sc= ience curriculum. The pre-service teachers use published research on science misc= onceptions to guide the projects that culminate in a formal written report, typically 10-15 pages in length. The report is the only experience in our program in which students formally describe research that they have planned, conducted, gathered and analyzed data, drawn conclusions, and related their findings to published literature.

We found that students needed extensive support for writing the report by anal= ysis of the final reports and anecdotal sharing of quality amongst capstone teac= hers that prompted a careful review of student work. Therefore, we designed and implemented detailed assignment guidelines and rubrics to assist students in completing the action research project. To further support student writing,= we assigned a written project proposal that contained about half of the final report’s content. The proposal was assessed by capstone course instru= ctors who provided substantial feedback on both content and writing. Students were required to use the feedback to revise the proposal prior to submitting the final report. For many students however, these interventions were not sufficient to produce a professional research report.

To determine an appropriate way to help students improve their writing abiliti= es, we met with the director of the Writing Center on our campu= s. Per his recommendations, we developed six ‘mini’ pre-writing assignments, each of which focused on one particular section of the report, such as describing the local K-8 classroom setting and the demographics of = the K-8 students or summarizing the research for the literature review. Students completed three assignments and received written feedback prior to completi= on of the proposal and three additional assignments with instructor feedback p= rior to the submission of the final report.

To determine the effectiveness of these pre-writing assignments, we used a pre- and post-test design. We collected 38 final reports from students that completed the pre-writing assignments and randomly selected 38 final reports from semesters prior to the pre-writing assignments that agreed to particip= ate in the study. Approximately 50% of the students are transfers from community colleges, 15% are self-identified minorities, and most work off-campus for = at least 20 hours per week.

To analyze= the final reports, we used the spelling, style and grammar checker in Microsoft Word 2003. Although this is a less than perfect tool, it is excellent at detecting subject-verb agreement, passive voice, contractions and split infinitives. It is less successful at detecting errors in uses of commas, errors in possessives and spelling errors for correctly spelled words incorrectly used. Microsoft Word identifies 43 different types of errors wi= thin the spelling, style and grammar check functions. We grouped these into four= major categories including multiple word (10 types), single word (20 types), verb-related (6 types) and punctuation errors (7 types). Examples of multip= le word errors included sentence structure, wordiness, long sentences, too many nouns and too many phrases. Examples of single word errors included spelling errors, commonly misused words, clichés, jargon and inappropriate us= e of connecting words. Verb-related errors included use of passive voice, subject-verb agreement, verb form and verb use. The punctuation category included, among others, incorrect use of commas, semicolons, capitalization, hyphen use and quotation marks. Microsoft Word also provided data on Flesch reading ease and Flesch-Kincaid grade level.

Results<= o:p>

Because th= e final reports varied in length from the pre to the post group, we could not use t= he total number of errors per report for analysis. See Table 1. A one-sample Kolmogorov-Smirnov test also revealed the number of total errors and each t= ype of error were not normally distributed (p<.05 for each category of error= ). Therefore, we used a nonparametric statistical technique, the Mann-Whitney = U, an alternative to the independent sample t test, to determine if there was a difference in any of the error categories between groups that participated = in the scaffolded pre-writing activities and those that did not. See Table 2.<= /p>

Table 1

Comparison of Total Erro= rs and Length of Report

	Without scaffolded pre-writing activities (n=3D 38)	With scaffolded pre-writing activities (n=3D38)
Total number	Range	M	SD	Range	M	SD
Words	1167-5854	2,857.42	1,112.99	2503-8811	5,339.47	1,744.20
Errors	5-125	58.97	26.16	4-229	86.97	41.61

Table 2

Descriptive Statistics f= or Errors

Without scaffolded pre-writing activities (n=3D38)

With scaffolded pre-writing activities (n=3D38)

Range

Number of words per error

30.16-384.40

59.89

58.24

35.69-2202.75

140.53

372.34

Number of words per multiple word error

123.46-1922.00

400.00

296.34

162.35-8811.00

758.43

1470.65

Number of words per single word error

100.86-961.00

305.58

188.81

108.83-8811.00

670.45

1475.23

Number of words per verb related error

60.96-1922.00

204.55

325.02

69.57-8811.00

469.48

1508.87

Number of words per punctuation related error

82.92-4109.00

457.65

720.81

121.97-8811.00

745.15

1493.96

Flesch reading ease score

41.5-69.1

53.56

6.2

33.6-57.5

49.41

5.01

Flesch-Kincaid grade level

7.1-14.9

10.63

1.43

9.7-14.3

11.63

1.12

The data w= ere analyzed with SPSS 14.0 for Windows (SPSS, Inc., 2005). To correct for differences in word count in the analysis, we used the total number of word= s in the report divided by the total number of errors in each category. The stud= ents with the pre-writing activities wrote over twice as many words in relation = to their total errors than the students without the pre-writing activities (z = =3D -3.55; p<.000). They also had fewer punctuation errors in relation to the total number of words in their final report (z =3D -2.12; p<.05). There = were no significant differences for the other error categories: multiple word (z= =3D -1.69), single word (z =3D -1.90) and verb related (z =3D -1.62). The Flesch reading ease, which is a readability score, showed a significant difference between the two groups (= z =3D -3.00, p < .05). The Flesch-Kincaid grade level was also significantly different (z =3D -3.37; p < .001).

Discussi= on

We found t= hat there was an improvement in student writing ability by the group with the scaffolded pre-writing assignments as measured by the relatively crude but easily accessible tool of Microsoft Word spelling, style and grammar checke= r. This is important because if we can encourage students to make use of this = or similar word processing tools, some aspects of their writing will improve. = We believe the significant increase in the total number of words in the report resulted from the increased amount of time students had to work on each section. The first three pre-writing assignments were due during the fourth, sixth and seventh weeks of the semester respectively. The last three pre-writing assignments had flexible due dates based upon completion of the assessments= with the elementary students and the science teaching. Therefore, students were = not able to wait until the “last minute” to write the final report.=

The signif= icant differences in the category of punctuation errors could be attributed to the fact that these tended to be easier for students to correct. Also, instruct= or feedback tended to indicate the changes needed with punctuation errors. We = think that the lack of a significant change in the other error categories could be due to the type of feedback provided. We tended to circle or otherwise make= a general comment such as “awkward” or “confusing” on students’ writing so students had to decide how to correct those erro= rs for themselves. These types of errors often required a revision of a comple= te sentence.

The two sc= ores for readability that Microsoft Word spelling and grammar checker provide are the Flesch Reading Ease and the Flesch-Kincaid Grade Level. The Flesch Reading Ease, which is = based on the average number of words per sentence and the average number of sylla= bles per word. The formula is a ba= sed on a 100 point scale. Higher sco= res mean that the document is easier to read.&= nbsp; Our students showed a significant difference between the two groups = with the Flesch Reading Ease score decreasing which means that the students̵= 7; writing became more complex in the final written research reports. The Flesch-Kincaid Grade Level pro= vides a U.S. grade-school level for the text. In our study, the students increased the grade level mean from 10.63 to 11.63.

We recogni= ze the need to inform students that there are limitations to these tools. These checkers of writing conventions do not begin to address issues of writing quality, organization of ideas and logical sequencing. We addressed some of these concerns with other writing activities. For example, in learning to w= rite an abstract, students critiqued two sample abstracts to note the key compon= ents of an abstract, to determine when critical information was missing and to u= se concise phrasing.

Anecdotal = evidence from student feedback indicates that the students were pleased with the pre-writing assignments. A common sentiment was that these assignments redu= ced the end-of-term writing pressure by distributing smaller portions of the fi= nal report over many weeks and that students were able to devote more time and attention to each section than if they wrote the entire report at the end of the term. Student statements thus corroborated our own assessment of the re= ason for increased lengths of final reports.

We will continue the use of= the scaffolded pre-writing assignments in this course. It is clear that the fin= al reports have improved and the students have recognized an improvement in th= eir own quality of writing. We plan to focus our written feedback so that stude= nts will be more likely to use the comments for revision as well as learn from them.

Scaffolded= writing assignments are a way that others may want to use in their science courses = to improve the quality of student writing in formal papers. The short assignments could be des= igned to fit the needs of any particular piece of writing, whether it is an origi= nal research report, an explanation of a scientific principle or argument to persuade an audience to accept the author’s interpretation of the evidence. By providing feedback on small portions of writing, students will have the opportunity a= nd time for revisions to author quality work.

Referenc= es

American Association for the Advancement of Science. (1993). Benchmarks for Scientific Literacy. New York: Oxford University Press.

Feldman, = S., Anderson, V., Mang= urian, L. (2001). Teaching effective scientific writing. Journal of College Sci= ence Teaching, 30(7), 446-449.

Keys, C.W. (1999). Revitalizing instruction in scientific genres: Connecting knowledge production with writing to learn in science. Science Education, 83, 115-130.

Holliday,= W.G., Yore, L.D., Alvermann, D.E. (1994). The reading-science learning-writing connection: Breakthroughs, barriers and promises. Journal of Research in Science Teaching, 31(9), 877-893.

Kroen, W.= (2004). Modeling the writing process. Journal of College Science Teaching, 34(3), 50-53.

National = Research Council. (1996). National Science Education Standards. Washington, DC:&nbs= p; National Academy Press.

Rice, R.E. (1998). “Scientific writing”-A course to improve the writing of science students. Journal of College Science Teaching, 27(4), 267-27= 2.

Rivard, L= .P. (1994). A review of writing to learn in science: Implications for practice = and research. Journal of Research in Science Teaching, 31(9), 969-983.

Robertson= , I.F. (2004). Assessing the quality of undergraduate education students’ writing about learning and teaching science. International Journal of Science Education, 26(9), 1131-1149.

Tessier, = J. (2006). Writing assignments in a nonmajor introductory ecology class. Jo= urnal of College Science Teaching, 35(4), 25-29.