This article was downloaded by: [Northeastern University]On: 15 November 2014, At: 14:12Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Interdisciplinary MathematicsPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tjim20

Using research papers as a tool in teachingintroductory statisticsDiane Fisher aa Department of Mathematics , Louisiana State University at Eunice , Eunice , LA , U.S.A.Published online: 31 May 2013.

To cite this article: Diane Fisher (2006) Using research papers as a tool in teaching introductory statistics, Journal ofInterdisciplinary Mathematics, 9:2, 287-295, DOI: 10.1080/09720502.2006.10700444

To link to this article: http://dx.doi.org/10.1080/09720502.2006.10700444

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shallnot be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and otherliabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Using research papers as a tool in teaching introductory statistics

Diane Fisher

Department of Mathematics

Louisiana State University at Eunice

Eunice, LA

U.S.A.

Abstract

The introductory course in statistics has traditionally been taught as a collectionof computationally intensive activities. In recent years, many statisticians have becomeinvolved in the reform movement in statistical education aimed at the teaching ofintroductory statistics. Those interested in reform want to emphasize statistical thinking andunderstanding and help students appreciate the role of statistics in the scientific process andin life. This paper suggests using research papers from refereed journals to illustrate howdifferent statistical concepts are used together in scientific inquiry. In addition, it is hopedthat students will see the value and relevance of statistics in their own lives and careersthrough examining research in their own areas of interest.

Keywords : Introductory statistics, research paper.

1. Introduction

Most instructors teaching introductory statistics classes today weretrained in the traditional method. They attended lectures on the mechanicsof statistical methods and the theory of probability and mathematicalstatistics and their involvement in the course was generally limitedto working the assigned problems. Since this is how they learned, itseemed natural to teach their students in the same manner, focusing oncomputational skills and compartmentalized knowledge. This approachhas failed dismally. The introductory statistics course is regarded by moststudents as difficult, unpleasant, and totally irrelevant to their field ofstudy. According to Butler (1998, p. 84) “although increasing numbers of——————————–Journal of Interdisciplinary Mathematics

Vol. 9 (2006), No. 2, pp. 287–295

c© Taru Publications

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

288 D. FISHER

adults are completing introductory level statistics courses, the adults donot often use statistics in their jobs and when they do try, the results are ashambles”. This dissatisfaction with the introductory course has, over thepast twenty years, led to considerable interest in creating a new model forthis course.

2. The reform movement

In recent years, many statisticians have become involved in thereform movement in statistics education. As a part of the CurriculumAction Project of the Mathematics Association of America (MAA), GeorgeCobb coordinated a focus group on Statistics Education. The report waspublished in the MAA volume Heeding the Call for Change (Cobb (1992)). Itincluded the following recommendations:

• Emphasize statistical thinking.

• More data and concepts, less theory and fewer recipes.

• Foster active learning.

The National Science Foundation funded several projects to implementaspects of the reform (Cobb (1993)).

The American Statistical Association (ASA) also provided supportfor statistics education by funding the Guidelines for Assessment andInstruction in Statistics Education (GAISE) Project. This project consistedof two groups, one focused on K-12 education and the other on theintroductory college course. In 2004, the college group presented theirreport, which built on the previous recommendations of the Cobb report.Their recommendations are:

• Emphasize statistical literacy and develop statistical thinking.

• Use real data.

• Stress conceptual understanding rather than mere knowledge ofprocedures.

• Foster active learning in the classroom.

• Use technology for developing conceptual understanding andanalyzing data.

• Use assessments to improve and evaluate student learning (Aliagaet al. (2004)).

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

TEACHING INTRODUCTORY STATISTICS 289

3. Examining student outcomes

By implementing the recommendations from the GAISE report, itis hoped that students will not only acquire some amount of statisticalliteracy, but also learn the basic elements of statistical thinking. Inassessing the effectiveness of a statistics class Garfield suggested thedesired outcomes of an introductory course should include:

• Learning — students’ understanding, reasoning, thinking.

• Persistence — leading students to use their statistical knowledgeand skills after they leave the course.

• Attitudes and beliefs — about the value and importance of statisticsand about themselves as learners and users of statistics. (Garfieldet al. (2002), p. 3)

While the most commonly discussed outcome is learning, it isimportant to consider the other two categories as well. Statistics instruc-tors need to encourage positive attitudes and help students see the valueof statistics in order to increase their chances of using statistics in theirpersonal lives and their professions.

4. Using research papers to promote statistical thinking

There have been numerous publications, workshops, and presen-tations at conferences suggesting specific techniques to use in class topromote statistical thinking. Hands-on activities get the students activelyinvolved with the data. It also gives them an immediate experience withvariability. However, these experiences tend to be simple activities thatcan be performed in a class period and can tend to “trivialize” statisticalanalysis.

Real-world data sets have unmistakable real-world import. Theycan be rich with statistically interesting features and can be selectedto support particular curricula. Students, however, are not involved inany data production and the concept of variability remains hidden. Inaddition, most of these data sets are best explored using statistical softwareon a computer. Many introductory courses, particularly in mathematicsdepartments, rely entirely on graphing calculators.

A third type of activity, simulation-based learning, again givesstudents direct experience with variability. In these types of activities, theinstructor can control the structure of the data, thus the conceptual focus

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

290 D. FISHER

of the exercise. The disadvantage of these types of activities is that the dataare not real.

Each of the three different types of learning activities can greatlyenhance a student’s learning. Whether they can effect significant changesin the other two measures of effectiveness, persistence and attitudes andbeliefs, is not as clear. Most learning activities illustrate a single concept.When the class moves on to the next topic, there is a new activity. In thelimited time available in a college classroom, it is extremely difficult toconduct an experiment or observational study that can be carried throughto completion. In addition, at this level, the students do not have theknowledge or skills to do this type of study on their own as an outsideproject. Studying the research others have done and presented as an articlein a professional journal, allows the students to witness the developmentof the research as they learn the statistical techniques.

If an instructor has students who are all in the same discipline,research papers could be selected from journals in that area. There areseveral examples in the literature where journal articles have, for a numberof years, been used in teaching biostatistics and epidemiology in schoolsof medicine and nursing. For some examples see Carter (1987), Mickey(1995), and Norton (1998).

If, on the other hand, the students come from a variety of disciplines,it is important to select papers of general interest. It is not difficult to findsuch papers. For example, a paper in American Midland Naturalist (Caroet al. (2000)) describes a study in which the authors count road kills inthe Central Valley of California to study the effects of different types ofhuman activity on the populations of medium-sized mammals. It is easyto understand, is only a few pages long, and has both descriptive statisticsand hypothesis tests. Other papers that have been used successfullyinclude “Transmission of Norwalk Virus During a Football Game” (Beckeret al. (2000)) and “U.S. College Students Use of Tobacco Products” (Rigottiet al. (2000)). Journals that consistently have readable papers of generalinterest include Journal of Marriage and the Family, Child Study Journal,Journal of Health Education, Journal of Personality and Social Psychology, andDevelopmental Psychology. Since a single paper will not have an exampleof everything that will be covered in class, it is a good idea to have acouple papers. For instance, one could be an experiment and the otheran observational study.

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

TEACHING INTRODUCTORY STATISTICS 291

The papers could be used at the end of the course as a “wrap-up”to tie together the different topics that were studied during the semester.However, the students might still view the course as a group of separateactivities until the papers are introduced. The preferred method would beto introduce the papers near the beginning of the course. Then, when aparticular topic is studied, the students can see how that topic is handledin the research papers. By the end of the course, the students will be veryfamiliar with the two papers and will have observed how a variety ofstatistical techniques are used in a single research study. Other resources ofgood journal articles for an introductory course and suggestions for theiruse can be found in Millard (2004), and Turner (1994).

5. Assessment

Since the use of journal articles in a statistics class can be a valuablelearning experience, it is important to incorporate what the students learnfrom them into their assessment. Students can be asked to find a researchpaper in their field of study, or in an area that interests them, and write ananalysis of that paper as a final project or in lieu of a final exam. It wouldbe very important that the instructor approve the paper before the studentbegins his work to be certain that it is appropriate. There is a sample guidefor a final project in the appendix.

In experiences with my own classes, I have found this to be a veryeffective tool. At the end of the course, I distribute a survey to the studentslisting every activity we did during the semester. I ask them to rate eachactivity on a five point scale from “not helpful at all” to “very helpful”. Thefirst semester I used journal articles in class, 62 out of 67 students foundjournal articles to be helpful or very helpful in helping them understandthe concepts presented in the course. No other activity received as higha rating. In the following semester, the students were able to choosebetween writing a critique of a journal article and taking a traditional finalexam. Everyone of them chose the journal article.

For instructors who teach large lecture classes, and are unable tograde a great number of projects at the end of the term, an alternativemethod can be used. A paper of general interest can be distributed duringthe last week or two of class. During that time period, the students read thepaper, ask questions, discuss it with each other, and become very familiarwith that particular study. The instructor can then ask questions aboutthat paper on the final exam. This technique allows the teacher to ask

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

292 D. FISHER

questions that require higher level thinking than could otherwise be askedwith traditional questions that have, at most, a few sentences to describe aproblem.

6. Conclusion

Using journal articles in an introductory statistics class can be anaid in following GAISE guidelines. Real data from real research canbe used which fosters active learning. When a research study is beingfollowed through to completion, the focus is naturally on conceptualunderstanding rather than mere knowledge of procedures. The studentssee how the particular procedures fit into the entire analysis. Not only arethey are developing statistical literacy , but they are learning to think likea statistician.

It is obvious that using research papers as a tool in teaching theintroductory statistics course enhances learning but, in addition, it alsohelps meet the other two goals, namely persistence and attitudes andbeliefs. Having observed a particular research study from beginning toend, the students are aware of the power and usefulness of statistics.Having seen the concepts they have studied applied to a problem or anarea of research in which they are interested should certainly increase thelikelihood that they will use their new knowledge and skills in the future.

Teaching an introductory statistics course that not only helps astudent develop statistical thinking during the class, but encourages themto use what they have learned after the course is finished is quite achallenge. This can best be accomplished by a variety of learning activitiesincluding the use of research papers to illustrate the scientific methodand to show how the many techniques discussed in class are part of thisprocess.

Appendix. Suggested guidelines for final project

The instructor must approve the paper you plan to critique beforeyou begin your work. If you turn in a project on a paper that has notbeen approved, you risk not having your project accepted! There is nominimum or maximum length to this project. Remember, I am lookingfor quality, not quantity. Basically, you are looking at a paper from arefereed journal and using everything you have learned this semester inreading and understanding this paper. Your written presentation shouldbe in paragraph form using proper English. Listed below are some types

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

TEACHING INTRODUCTORY STATISTICS 293

of questions you might want to address in your analysis. These questionsare to be used a guide only. You will not be able to answer every questionlisted for your particular paper and there may be other important issuesto discuss that are not listed here. Feel free to include your thoughts oropinions about the study or techniques used.

The study

1. What were the authors trying to find out?

2. Was this an experiment or observational study? If an observationalstudy, was it retrospective or prospective?

3. What type of data were collected? Discuss the type of sampling used,the size of the sample, and possible biases associated with the sampling.

4. Were any data obtained through simulation?

5. Did the researchers have problems with missing or incomplete data?If so, did this affect the results?

Descriptive statistics

1. Were there any graphic displays of the data? Did these enhance yourunderstanding of the material being presented? Do you feel any of thegraphic displays were misleading or could have been presented in adifferent way?

2. What summary statistics were used? Did these enhance your under-standing of the material being presented?

3. Was there anything you would have added or deleted from the descrip-tive statistics presented?

Inferential statistics

1. Did the authors use confidence intervals or hypothesis tests?

2. Were any other types of inferential statistics used?

3. Which of the tests gave statistically significant results? State these resultsin words.

Conclusions

1. Based on the results of the tests of hypothesis and any other inferen-tial statistics reported, do you feel the authors drew appropriateconclusions?

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

294 D. FISHER

2. Was there any information on qualifications of the authors and whofunded their research? If this research was funded by a grant, wouldthe organization providing the funding have a financial interest in theoutcome?

3. Include any other additional comments you have about the study andhow it was conducted. Do you feel anything should have been donedifferently? Were there aspects of this research that you particularlyliked or did not like? Were there any ethical issues that affected thestudy in any way?

References

[1] M. Aliaga, G. Cobb, C. George, C. Cuff, J. Garfield, R. Gould, R. Lock,T. Moore, A. Rossman, B. Stephenson, J. Utts, P. Velleman andJ. Witmer (2004), GAISE College Report,URL http://it.stlawu.edu/∼rlock/gaise/GAISECollege.htm

[2] K. M. Becker, C. L. Moe, K. L. Southwick and J. N. MacCormack(2000), Transmission of Norwalk virus during a football game, NewEngland Journal of Medicine, Vol. 343, pp. 1223–1227.

[3] R. S. Butler (1998), On the failure of the widespread use of statistics,Amstat News, Vol. 251, p. 84.

[4] T. M. Caro, J. A. Shargel and C. J. Stoner (2000), Frequency ofmedium-sized mammal road kills in an agricultural landscape inCalifornia, American Midland Naturalist, Vol. 144, pp. 362–369.

[5] R. A. Carter (1987), Biostatistics instruction at Meharry medicalcollege, The American Statistician, Vol. 41 (4), pp. 268–269.

[6] G. Cobb (1992), Report of a conference on statistical education,in Heeding the Call for Change, L. Steen (ed.), MAA Notes, No. 22,Mathematical Association of America, Washington, pp. 34–43.

[7] G. Cobb (1993), Reconsidering statistics education: a nationalscience foundation conference, Journal of Statistics Education [Online],Vol. 1 (1).

[8] J. Garfield, R. Hogg, C. Schau and D. Whittinghill (2002), Firstcourses in statistical science: the status of educational reform efforts,Journal of Statistics Education [Online], Vol. 10 (2),

URL http://www.amstat.org/publications/jse/v10n2/

garfield.html

[9] R. M. Mickey and J. P. Carpenter (1995), Journal articles for anintroductory biostatistics/epidemiology course, in ASA Proceedingsof the Section on Statistical Education, pp. 293–298.

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014

TEACHING INTRODUCTORY STATISTICS 295

[10] R. Millard and J. C. Ron (2004), Activities and Projects for the W. H.Freeman Statistics Series, W. H. Freeman and Company, New York.

[11] H. J. Norton (1998), Teaching undergraduate students and medicalresidents how to identify fallacies in numerical reasoning, in ASAProceedings of the Section on Statistical Education, pp. 197–202.

[12] N. Rigotti, J. Lee and H. Wechsler (2000), US college students’ use oftobacco products, Journal of the American Medical Association, Vol. 284,pp. 699–705.

[13] J. C. Turner (2004), Using journal articles in intro stats, in ASAProceedings of Joint Statistical Meetings, pp. 2814–2815.

Dow

nloa

ded

by [

Nor

thea

ster

n U

nive

rsity

] at

14:

12 1

5 N

ovem

ber

2014