FREE ELECTRONIC LIBRARY - Abstract, dissertation, book

Pages:   || 2 |

«Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 Stat/Biostat 598, Spring 2009Stat/Biostat Stat/Biostat 598, Fall 2006 598, Fall 2006 In the ...»

-- [ Page 1 ] --

Stat/Biostat 598, Spring 2009Stat/Biostat

598, Fall 2006

Stat/Biostat 598, Spring 2009Stat/Biostat

Stat/Biostat 598, Fall 2006

598, Fall 2006

In the Encyclopedia of Statistical Sciences (Kotz et al.,

2005) Brian Joiner states that a statistical consultant, to be

fully effective, should have many diverse skills. Ideally:

• Have a genuine desire to solve real problems and help

others to solve problems.

• Be able to help investigators formulate their problem in quantifiable terms.

• Be able to listen carefully and to ask probing questions.

• Have a broad knowledge and true understanding of statistical and scientific methods.

• Be able to adapt existing statistical procedures to novel environments.

• Be able to locate or develop good statistical procedures in a timely fashion.

• Be able to keep abreast of developments in statistics.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006

• Be willing to meet deadlines, even if it requires substantial extra effort.

• Be able to understand something about the clients’ subject matter and speak a bit of the clients’ language.

• Be a good teacher—much success in consulting depends on being able to help others understand statistical tools, and their strengths and weaknesses.

• Be willing to settle for a reasonably correct approximate solution, then go on to the next problem.

• Be able to identify important problems (and thus avoid spending too much time on projects of little significance).

• Have the confidence to use as simple a procedure as will get the job done, be it design or analysis.

• Be able to convince others of the validity of a solid solution and see to it that proper action is taken.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006

• Be able to use computers effectively and direct others in their use.

• Be a good problem solver.

• Be willing to meet clients regularly on their home ground, and take the responsibility to meet and communicate with all members of the working team.

• Be diplomatic and know when to bend, when to stand firm, and how to help smooth conflicts over among other team members.

• Be willing to get some experience in the actual collection of the data.

• Be willing to take the time to check and double-check procedures and results.

• Be able to communicate effectively in writing as well as orally (this often includes helping clients write their reports as well).

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006

• Be able to make a good estimate of how much effort will be required to solve the problem without actually having to solve the problem itself.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 Brian Joiner’s Consultant’s Check List (1979) Listen a lot.

Never (well, almost never) interrupt a client.

Always (well, almost always) allow the client to interrupt you.

Ask a lot of questions that begin with phrases like “Let me see if I understand this, …”.

Don’t smell: shower daily, wear clean neat clothes, brush teeth, get decayed teeth fixed.

Take good notes, from the beginning.

Convey a helpful resourceful attitude.

Helping the client clarify his/her own goals is often your most important function.

Keep things simple: don’t dazzle your client --- or yourself; use as simple words, simple designs and simple analyses as will suffice.

Put things in writing, and give copies to the client.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 At the end of each meeting spend a few minutes clarifying what has been decided and who does what next. Put these in writing.

Know a wide variety of statistical methods (so you’ll be less inclined to try to fit the problem to the wrong solution).

Interact frequently with the client ---- never go off and do a lot of work at the client’s expense or on the client’s behalf without discussing approaches and intermediate results.

Make realistic cost estimates and discuss them (cost may be dollars and/or time). [Note: look forward to Nayak Pollisar’s discussion on consulting in private practice.] Find out how important this project is to the client, and how much she/he wants to commit to your efforts.

Be timely: an approximate answer in a few days is almost always preferable to an “exact” answer some months later.

Make sure you and the client understand that goals of the project, and whether or not these goals are attainable by the planned course of action. Get the goals agreed to, in writing.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 Statistical Consulting Resource list adapted from STAT 8801 (Univ of Minnesota) Books Boen, James R., and Douglas A. Zahn (1982), The Human Side of Statistical Consulting, Lifetime Learning: Belmont, CA.

Cabrera, J., and McDougall, A., (2002), Statistical Consulting, Springer, New York.

Chatfield, Christopher, (1988), Problem Solving: A Statistician's Guide Chapman and Hall, London.

Derr, J., (2000), Statistical Consulting: A Guide to Effective Communication, Duxbury Hand, D. J., and Everitt, B. S., (1987), The Statistical Consultant in Action, Cambridge University Press.

Polya, George (1971) How to solve it: a new aspect of mathematical method Rustagi, Jagdish S. and Douglas A. Wolfe, editors, Teaching of Statistics and Statistical Consulting, Academic Press (1982).

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 Printed papers

0. A bibliography and general articles Baskerville, J. C. (1981) ‘A systematic study of the consulting literature as an integral part of applied training in statistics’, American Statistician, 35, 121-123.

McCulloch, Charles E., Boroto, Daniel R., Meeter, Duane, Polland, Ronald and Zahn, Douglas A. (1985), ‘An expanded approach to educating statistical consultants’, American Statistician, 39, 159-167.

Pfannkuch, Maxine, and Chris J. Wild, "Statistical Thinking and Statistical Practice: Themes Gleaned from Professional Statisticians", 15 Statistical Science 132 (2000).

Tweedie, R., "Consulting: Real Problems, Real Interactions, Real Outcomes" with discussion, 13 Statistical Science 1 (1998).

Wild, Chris J., and Maxine Pfannkuch, "Statistical Thinking in Empirical Inquiry" with discussion, 67 International Statistical Review 223 (1999) Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006

1. Where is statistical consulting done?

Boen, James R. (1982), ‘A self-supporting University statistical consulting center’, American Statistician, 36, 321-325.

Cameron, J. M. (1969), ‘The statistical consultant in a scientific laboratory’, Technometrics, 11, 247-254.

Daniel, Cuthbert (1969), ‘Some general remarks on consulting in statistics’, Technometrics, 11, 241-245.

Kirk, Roger E. (1991), ‘Statistical consulting in a university:

dealing with people and other challenges’, American Statistician, 45, 28-33.

Marquardt, Donald W. (1979), ‘Statistical consulting in industry’, American Statistician, 33, 102-107.

Marquardt, Donald W. (1981), ‘Criteria for evaluation the performance of statistical consultants in industry’, American Statistician, 35, 216-219.

Meier, P., (1986), ‘Damned liars and expert witnesses’, Journal of the American Statistical Association, 71, 269-276.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006

2. Interpersonal aspects of consulting and roles.

Boen, James (1972), ‘The teaching of personal interaction in statistical consulting’, American Statistician, 26, 30-31.

Boen, James and Fryd, David (1978), ‘Six-state transactional analysis in statistical consultation’, American Statistician, 32, 5840.

Bross, Irwin D. J. (1974), ‘The role of the statistician: Scientist or shoe clerk’, American Statistician, 28, 126-f27.

Hunter, William G., (1981), ‘The practice of statistics: The real world is an idea whose time has come’, American Statistician, 35, 72-76.

Hyams, Lyon (1971), ‘The practical psychology of biostatistical consultation’, Biometrics, 27, 201-211.

Zahn, Douglas A. and Isenberg, Daniel J. (1983), ‘Nonstatistical aspects of statistical consulting’, American Statistician, 37, 297Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006

4. Reporting and communication skills Ehrenberg, A. S. C., (1982), ‘Writing technical papers or reports’, American Statistician, 36, 326-329.

Ehrenberg, A. S. C., (1977), ‘Rudiments of numeracy’, Journal of the Royal Statistical Society, A140, 277-297.

Hoadley, R. Bruce., and Kettenring, Jon. R., (1990), ‘Communication between statisticians and engineers/physical scientists’, Technometrics, 32, 243-274.

McDonald Gary C., (1988), ‘Communicating with managers’, Chance, 1, 42-44.

5. Professional conduct and ethics American Statistical Association (1999), ‘Ethical guidelines for statistical practice’, approved by Board of Directors of ASA, Aug 7, 1999.

http://www.amstat.org/profession/index.cfm?fuseaction=ethicalstatistics Committee on Professional Ethics, American Statistical Association, http://www.tcnj.edu/~ethcstat/ Engeman, Richard M. and Shumake, Stephen A. (1983), ‘Animal welfare and the statistical consultant’, American Statistician, 47, 229Hooke, Robert (1980), ‘Getting people to use statistics properly’, American Statistician, 34, 102-107.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 Tips and Tricks for Statistical Consulting Donald C. Martin, Ph.D.

–  –  –

Transformations III.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006

1.A.1 Rule: Always look at the descriptive statistics before considering going on to any other analysis.

Remember that there are different kinds of descriptive statistics. The descriptive methods that you use for general exploratory data analysis may differ from the best methods for checking model assumptions.

Descriptive methods for inclusion in reports or papers are to present some aspect(s) of the data. These require careful design. Ask what is the primary aim of this graph or table.

It is not unusual to rewrite parts of the text of a paper many times. Be ready to redo table and graphs as well.

–  –  –

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 I.A.3 In many cases descriptive statistics are all that is needed. Don’t feel compelled to do an elaborate analysis if it is not needed.

The “interocular test of significance”

–  –  –

I.B.1 Find out why your client is doing the study.

Why are they seeing you and what do they expect? Help?

A blessing? Magic?

Don’t expect to find out all this at once. It is an iterative

process. Good questions:

What are the primary goals Which is the most important variable?

What resources are available? (including computers, software, and skills using these) How does this analysis (or design) relate to the goals? (This question sounds to supid to ask, but just wait …)

–  –  –

I.B.3 Most clients will try to phrase their problems in terms of hypothesis tests. Be prepared to spend time in educating the client about alternatives (including descriptive statistics).

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 I.B.4 Exploration vs. Confirmation. Try to establish the point on this continuum for various parts of the study. Your methods should be different.

At the exploratory end, concentrate on power rather than type I errors. You are looking for hints for further research. Use large alphas and ignore multiple decisions (correction for multiple testing).

At the confirmatory end use a small alpha; you need a clean design and analysis.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 I.B.5 Mutiple comparison methods are of limited use at either end of the exploration vs.

confirmation spectrum. These are probably most useful in arguing that unexpected results found by data mining are really there.

(but they are more and more essential in large data mining/genomic/micro-array problems; “False Discover Rate”)

–  –  –

I.C.1 Errors in randomization.

Most people don’t know what randomization is, how to do it, or why they should do it.

Common use is that if they cannot predict something, then it is random.

Ask for details, not just if randomization was used.

Ex. Cross-over clinical trial Ex. “fast rats” (iron deficiency and activity) Note: subverted randomization is more common than you would think.

Stat/Biostat 598, Spring 2009Stat/Biostat 598, Fall 2006 I.C.2 Hidden matching This is one of the more difficult problems in consulting.

Often the investigator has done some type of matching and fails to mention it.

Ask about possible matching. Look at lab data sheets, etc.

Handling partial matching can be hard. Hatching is often

overused! Matching can have two different aims:

Reduction in bias in observational studies Reducation in variance in randomized studies These are very different situations (Paula Diehr paper on consequences of ignoring matching)

–  –  –

I.C.4 Confusion of independent and dependent variables Ex. Gestational age I.C.5 Using a hypothesis test to show that two groups are the same.

This is an easier trap to fall into than you might expect.

The fix is to use confidence intervals.


–  –  –

I.D1 (Phone call) “I have a quick question …” • See JASA 1957 pp. 133-142, Kimball: “Errors of the third kind in • statistical consulting.” Unless you know both the person and the research, just say no.

• Get the clients in for a full consulting session.

(Would you diagnose (prescribe, etc.) by phone without seeing • the patient?)

–  –  –

I.D.4 n observations with p variables, p n.

Pages:   || 2 |

Similar works:

«ROCZNIKI GLEBOZNAW CZE T. XXXVIII. NR 2 S. 13—23. WARSZAWA 1987 JAN W. CHMIELNICKI ZM O D Y FIK O W A N A M ETO DA SZALKOW A OZNACZAN IA W OD O O D PO RN O ŚCI A G REG A TÓ W GLEBOW YCH Katedra Ogólnej Uprawy Roli i Roślin SGGW-AR w Warszawie WSTĘP W zależności od zawartości piasku, pyłu, iłu i próchnicy uprawna warstwa gleby wykazuje różną budowę agregatową. Występujące w glebie agregaty powstają przez zlepienia i sklejenia elementarnych cząstek gleby, które występują...»

«Summary of Other Consultation Responses Considered First Core Strategy Gedling Council initially began work on a Core Strategy solely for Gedling Borough. However, for a variety of reasons, the Council decided to revisit it’s earlier work and to progress the Aligned Core Strategy in conjunction with the other Greater Nottingham authorities. However, the comments received on the First Core Strategy will be used to inform preparation of the Aligned Core Strategy. In particular, feedback from...»

«KÖZJEGYZŐK KÖZLÖNYE KÖZJEGYZŐK KÖZLÖNYE A Magyar Országos Közjegyzői Kamara szakmai folyóirata 2015 / 1. szám 2015. január / február . oldal Juhász Gábor Az örökhagyó szabad rendelkezési jogának korlátjaként megjelenő kötelesrész a spanyol jogban (II. rész) . oldal Molnár Judit „Elköltözött” vagy a „címzett ismeretlen” – A kötelezett menekülési lehetősége a fizetési meghagyásos eljárásban . oldal Bodzási Balázs Háromoldalú...»

«FEUP Faculty of Engineering of the University of Porto MIEEC Integrated Master of Electrical and Computers Engineering Dissertation preparation Final Report An interface for Text Mining with the RAP system Nelson Prazeres Costa Porto, February 2008 Abstract This report introduces a dissertation whose main assignment will be the development of an interface to the RAP system to facilitate text mining tasks. The interface will provide the user the results of the RAP algorithm. With the growing of...»

«Are sovereign credit ratings anachronistic? January 05, 2015 Sovereign Debt Ratings: Anachronism in today’s age? Years 2013 and 2014 have been unpleasant for governments that have seen their sovereign ratings downgraded for various reasons. Ratings agencies have not spared even the world’s most credit worthy nation – the US; S&P downgraded its AAA rating, bringing down the pool of AAA rated countries to 11 and investible sovereign debt to USD 4 trillion from USD 11 trillion in one go....»

«Lösch, Bettina Internationale und europäische Bedingungen politischer Bildung zur Kritik der European Citizenship Education Zeitschrift für Pädagogik 55 (2009) 6, S. 849-859 urn:nbn:de:0111-opus-42788 in Kooperation mit / in cooperation with: http://www.beltz.de Nutzungsbedingungen / conditions of use Gewährt wird ein nicht exklusives, nicht übertragbares, persönliches und beschränktes Recht auf Nutzung dieses Dokuments. Dieses Dokument ist ausschließlich für den persönlichen,...»

«Benutzer ­ Handbuch für den osCommerce Shop (in der angepassten Version von smartsells.de) osCommerce ist eine Marke eingetragene Marke – alle Rechte sind Harald Ponce deLeon  © 2004 Marcus Karn Inhaltsverzeichnis           Seite   Konfiguration                                                                                                              ...»

«CEIOPS-DOC-24/09 CEIOPS’ Advice for Level 2 Implementing Measures on Solvency II: Own funds Supervisory approval of ancillary own funds (former CP29) October 2009 CEIOPS e.V. – Westhafenplatz 1 60327 Frankfurt – Germany – Tel. + 49 69-951119-20 Fax. + 49 69-951119-19 email: secretariat@ceiops.eu; Website: www.ceiops.eu © CEIOPS 2009 Table of content 1. Introduction 2. Extract from Level 1 text 3. Advice 3.1 Explanatory text 3.1.1 Key issues 3.1.2 Options 3.1.3 Data 3.2 CEIOPS advice...»

«Journal of Statistics Education, Volume 21, Number 3 (2013) Implementation of Discovery Projects in Statistics Brad Bailey Dianna J. Spence Robb Sinn University of North Georgia Journal of Statistics Education Volume 21, Number 3 (2013), www.amstat.org/publications/jse/v21n3/bailey.pdf Copyright © 2013 by Brad Bailey, Dianna J. Spence, and Robb Sinn all rights reserved. This text may be freely shared among individuals, but it may not be republished in any medium without express written consent...»

«A Class Project in Survey Sampling Andrew Gelman∗ and Deborah Nolan† July 1, 2001 Courses in quantitative methods typically require students to analyze previously collected data. There is great value in this exercise especially when they analyze data collected to address real problems from the world outside the textbook. Yet, in our experience it can be difficult for students to bring the analytic skills learned in the classroom to an independent project. More recently, activity-based...»

«Institute of Geology and Geophysics of the Azerbaijan National Academy of Sciences, Baku, Azerbaijan 12-20 October 2014 INTERNATIONAL GEOSCIENCE PROGRAMME 2006 ea 2015 S 2007ov z of A Caspian Black Sea 2013 Sea rus ho Bosp s nell Sea of rda Da Marmara Aege = Sea an Lev Medite rranea ant n Sea ine Field Trip Guide of the Second Plenary Conference IGCP 610 “From the Caspian to Mediterranean: Environmental Change and Human Response during the Quaternary” (2013 2017)...»

«University of Southampton Research Repository ePrints Soton Copyright © and Moral Rights for this thesis are retained by the author and/or other copyright owners. A copy can be downloaded for personal non-commercial research or study, without prior permission or charge. This thesis cannot be reproduced or quoted extensively from without first obtaining permission in writing from the copyright holder/s. The content must not be changed in any way or sold commercially in any format or medium...»

<<  HOME   |    CONTACTS
2016 www.abstract.xlibx.info - Free e-library - Abstract, dissertation, book

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.