Writing Assignments in an Introductory Statistics Course

1
Writing Assignments in an Introductory Statistics
Course

• Joy Jordan, Lawrence University
• joy.jordan_at_lawrence.edu
• May 13, 2008 (CAUSE webinar)

2
Important Points from the Writing Literature

• Use of writing as a learning tool (I think we are
  already sold on this idea)
• Suggestions about assignment creation
• Research about revision and feedback
• Descriptions of different types of writing

3
Assignment Creation

• Clear definition and articulation of assignments
• Specify RAFT criteria Role, Audience, Format,
  and Theme/Task/Topic (adapted from Bean, 2001)
• Include a grading rubric (make expectations clear
  to students)
• When planning assignments, teachers need to
  consider not only the learning goals they have
  set for their courses but also the thinking and
  writing processes that they want to invoke in
  their students as learners. (Bean, p. 77)

4
Revision and Feedback

• Importance of revision
• Research indicates that when teachers make
  remarks on papers and return those papers to
  students while offering them no opportunity to
  revise, the remarks have little effect on
  subsequent papers. (Dohrer, 1991)
• Student misunderstanding of the revision process
• Students typically think of revision at the
  micro- or editing-level (e.g., finding the right
  word, correcting grammar), whereas experienced
  writers believe revision involves macro-level
  changes (e.g., restructuring, additions,
  deletions) (Sommers, 1980)

5
Revision and Feedback

• The message your feedback sends to your students
• Feedback mistakes (see e.g., Sommers 1982, Bean
  2001)
• Too vague (comments that are not specific to the
  student text and give no guidance for change)
• Too many (give no sense of priorities for
  revision)
• Too micro-level (give the impression that
  revision really is simply an editing process)
• Too negative (important to additionally point out
  what a student has done well)

6
Revision and Feedback

• Potential time savers
• Thoughtful assignment creation (well-planned
  assignments are easier and more enjoyable to
  grade)
• Smaller microthemes (Bean, 2001)
• Ways of marking (put simple check marks next to
  lines that contain grammatical errorsHaswell,
  1983 put a straight line under sentences that
  are especially well written and a wavy line under
  passages that are problematicElbow, 1997)
• Peer review of writing (see e.g., Holt 1992, Bean
  2001)

7
Types of Writing

• Informal versus formal writing (see e.g., Britton
  1975, Elbow 1997)
• If we are not so much teaching writing as using
  writing to teach something else, it makes
  particularly good sense to use lots of minimal
  and low stakes response. (Elbow, 1997)

8
Examples for the Statistics Classroom

• Informal/low-stakes writing assignments
• In-class, free-response writing in answer to a
  question (e.g., Why is the variability of the
  sample mean smaller than the variability of
  individual observations?)
• Journal writing (e.g., responses to class
  reading, activities, homework problems, or
  experiences with statistics outside of class)
• Minute papers at the end of class

9
Examples for the Statistics Classroom

• Formal/high-stakes writing assignments
• Press release summarizing, in layperson terms,
  the results of a complicated experimental study
• Letter home in response to a statistics question
  from a parent
• Letter to the editor in response to a newspaper
  article that makes use of statistics
• Statistics project report (intro, methods,
  results, discussion) or just one focused part of
  the report
• Mathematics or statistics autobiography
• Meaningful paragraph (write about statistics
  terms appropriately in context)

10
Examples from my Classroom

• Assignment 1
• Suppose you receive the following letter from
  your dad
• Hey Kiddo,
• I am worried about Grandma. Remember that she
  was diagnosed with high blood pressure? Well,
  shes currently taking the medication Makemewell
  to lower her blood pressure. At the time of
  Grandmas diagnosis, her doctor said that a
  randomized, double-blind experiment had been
  conducted and that Makemewell was shown more
  effective in lowering blood pressure than a
  placebo. To be honest, I have no idea what any of
  that means, but I believed and trusted the
  doctor. Now Ive heard two stories that make me
  think differently. Larry, our next-door-neighbor,
  was taking Makemewell and he got a terrible fever
  that put him in the hospital. Also, my co-worker,
  Sally, actually had her blood pressure go up
  while she was taking Makemewell! Im now very
  suspicious of this medication.
• I know that youre taking a statistics course at
  college. Based on the information Ive given you,
  do you think Grandma should stop taking her
  medication? Whatever your opinion, will you
  please explain yourself thoroughly and clearly?
  (I will draw on your responses when I talk with
  the doctor.) And please dont use any statistics
  mumbo-jumbo that I wont understand. I really
  appreciate your help with this.
• Love, Dad
• Your assignment is to type a 1-2 page letter
  (single-spaced, 12-pt. font) responding to your
  dad.

11
Examples from my Classroom

• Assignment 1 Grading Criteria (25 points
  possible)
• ____ The explanation to your dad convinces me
  (your teacher) that you understand the following
  what a randomized, double-blind experiment is
  what anecdotal evidence is and which of these
  data collection methods is appropriate for
  decision making. (10 points)
• ____ The explanation to your dad is thorough,
  well organized, and clear. (5 points)
• ____ The explanation to your dad is presented in
  non-technical terms that he will understand. (5
  points)
• ____ You successfully paid attention to accepted
  conventions of language use (syntax, spelling,
  grammar, readability, etc.) (5 points)

12
Examples from my Classroom

• Assignment 2
• Suppose you receive another letter from your dad
• Hey Kiddo,
• Thanks so much for your response to my last
  letter. I now understand what a randomized,
  double-blind experiment is, and I feel better
  about the money Im spending on your college
  education. Grandma is still taking Makemewell,
  and were monitoring her progress.
• I got more information from the doctor about the
  Makemewell experiment. The doctor said that 40
  high-blood-pressure patients took Makemewell and
  40 patients took a placebo (and the patients were
  all around Grandmas age). The average reduction
  in systolic blood pressure was 19.5 millimeters
  of mercury (mm Hg) for the Makemewell group and
  9.1 mm Hg for the placebo group. This much I can
  understand. Then the doctor mentioned
  significance testing (and hypotheses), and she
  said the difference in average blood pressures
  was statistically significant based on a p-value
  that was less than 0.001. Do you know what the
  doctor is talking about? What are all these
  terms, and how can they help me understand the
  effect of Makemewell? As before, please explain
  yourself thoroughly and clearly, and dont use
  any technical language.
• Im really glad youre taking a statistics class
  this term. Your knowledge has been very helpful.
• Love, Dad
• PS Im enclosing a check to get you through the
  rest of the term.
• Your assignment is to type a 1-page letter
  (single-spaced, 12-pt. font) responding to your
  dad.

13
Examples from my Classroom

• Lessons learned from letter-to-dad assignments
• Modification of the rubric for assignment 1
  (from you understand the statistical concepts
  involved in the assignment to something more
  specific)
• Assignment 2 illuminated more student
  misperceptions than did assignment 1
• Even though the assignments were focused
  microthemes, the grading load was still too large
  for me (true confession I havent yet truly
  learned the time-saving methods I previously
  mentioned)

14
Examples from my Classroom

• Elaine Backus, an entomology professor at the
  University of Missouri, created a writing
  assignment called a meaningful paragraph
• She defines a meaningful paragraph as one
  continuous piece of writing that uses all of the
  words from a given subset and in which the
  sentences make sense and hang together

15
Examples from my Classroom

• Meaningful Paragraph Assignment 1
• Write a meaningful paragraph that includes the
  following terms correlation, causation,
  regression line, predicted value(s), scatterplot,
  residual(s), and outlier(s). (The paragraph you
  turn in should be typed, not hand written.
  Furthermore, please proofread your paragraph
  before turning it in.)
• A meaningful paragraph is one continuous piece
  of writing, which uses all of the listed words
  and in which the sentences make sense and hang
  together. That is, the ideas in the paragraph
  must illustrate you understand the new terms in a
  way that allows you to write meaningfully about
  them in context. You may not simply write seven
  sequential sentences, for example, that merely
  define the terms sentences must demonstrate
  appropriate relationships between the terms.
• Example Meaningful Paragraph
• (For the terms mean, median, distribution,
  standard deviation, outlier, and skewed)
• The CEO of a local company said the mean annual
  salary for the companys employees is 57,000.
  Upon further investigation of the data, though,
  the distribution of annual salaries appears
  severely right skewed. Hence, the median (rather
  than the mean) is really a better measure of
  typical annual salary, since the mean is pulled
  up by the high salaries in the long tail of the
  distribution. Because the mean is not the best
  measure of typical salary, the standard deviation
  should not be used as a measure of the variation
  in the salaries (since it measures the spread
  from the mean). The CEO thought the high average
  value might entice workers to apply for jobs at
  the company. In fact, the CEOs salary (840,000)
  is an extreme outlier in the distribution (its
  well outside even the long right tail), and her
  income value inflates the average company salary
  in a misleading way.

16
Examples from my Classroom

• Meaningful-paragraph terms used in next three
  assignments
• bias, nonresponse, parameter, population, random
  sample, statistic, and undercoverage
• sampling distribution, population mean, sample
  mean, normal distribution, sample size, and
  probability
• statistical significance (or statistically
  significant), probability, population mean,
  sample mean, P-value, and practical significance
  (or practically significant).
• Positives about the meaningful paragraphs
• Much less time grading
• Gives students the opportunity to be creative,
  yet also clearly shows students misperceptions

17
General Suggestions

• Heed the advice of the writing-to-learn
  literature (yet adjust to your course goals and
  teaching style)
• Purposefully design writing tasks to meet
  learning goals (and consider the thinking process
  associated with the writing exercise)
• Start small (always good advice) and make
  appropriate adjustments along the way
• Find colleagues to join you on your journey
  (e.g., teaching circle)

18
A Few Resources

• Bean, J. (2001), Engaging Ideas The Professors
  Guide to Integrating Writing, Critical Thinking,
  and Active Learning in the Classroom, San
  Francisco Jossey-Bass.
• Britton, J., Burgess, T., Martin, N., McLeod, A.,
  Rosen, H. (1975), The Development of Writing
  Abilities (11-18), London MacMillan.
• Dohrer, G. (1991), Do Teachers Comments on
  Students Papers Help? College Teaching, 39(2),
  48-58.
• Elbow, P. (1997), High Stakes and Low Stakes in
  Assigning and Responding to Writing, in
  Sorcinelli Elbow (eds.), Writing to Learn
  Strategies for Assigning and Responding to
  Writing Across the Disciplines (pp. 5-13), San
  Francisco Jossey-Bass.
• Emig, J. (1977), Writing as a Mode of Learning,
  College Composition and Communication, 28(2),
  122-128.
• Fulwiler, T. (1982), Writing An Act of
  Cognition, in Griffin, C.W. (ed.), New
  Directions for Teaching and Learning Teaching
  Writing in All Disciplines (pp. 15-26), San
  Francisco Jossey-Bass.
• Haswell, R. (1983), Minimal Marking, College
  English, 45(6), 600-604.
• Herrington, A. (1981), Writing to Learn Writing
  Across the Disciplines, College English, 43(4),
  379-387.
• Holt, M. (1992), The Value of Written Peer
  Criticism, College Composition and
  Communication, 43(3), 384-392.

19
A Few Resources

• Langer, J.A. and Applebee, A.N. (1987), How
  Writing Shapes Thinking A Study of Teaching and
  Learning, National Council of Teachers of
  English.
• MacAllistar, J. (1982), Responding to Student
  Writing, in Griffin, C.W. (ed.), New Directions
  for Teaching and Learning Teaching Writing in
  All Disciplines (pp. 59-65), San Francisco
  Jossey-Bass.
• Radke-Sharpe, N. (1991), Writing as a Component
  of Statistics Education, The American
  Statistician, 45(4), 292-293.
• Rose, B. (1989), Writing and Mathematics Theory
  and Practice, in Connolly Vilardi (eds.),
  Writing to Learn in Mathematics and Science (pp.
  15-30), New York Teachers College Press.
• Samsa, G. and Oddone, E. (1994), Integrating
  Scientific Writing into a Statistics Curriculum
  A Course in Statistically Based Scientific
  Writing, The American Statistician, 48(2),
  117-119.
• Sommers, N. (1980), Revision Strategies of
  Student Writers and Experienced Adult Writers,
  College Composition and Communication, 31(4),
  378-388.
• Sommers, N. (1982), Responding to Student
  Writing, College Composition and Communication,
  33(2), 148-156.
• Stromberg, A. And Ramanathan, S. (1996), Easy
  Implementation of Writing in Introductory
  Statistics Courses, The American Statistician,
  50(2), 159-163.