%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % ST 810A -- ETHICS AND PROFESSIONAL RESPONSIBILITY % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % These slides are created using the PROSPER package % % The pdf option sets things up for projection (after conversion to pdf) % The azure option is one of several available backgrounds; others % are rico, alienglow, autumn, darkblue (see below) % % To process and create pdf for projecting % % unity% add psutils % unity% add prosper % unity% latex prospertemplate % unity% dvips prospertemplate % unity% ps2pdf prospertemplate.ps % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % To print 6-up, I make a postscript temp.ps containing the full % pages only (so only those where all bullets appear). This can be done % using ``ps'' option (see below) or by bringing up the ps file in % ghostview and manually selecting the pages you want in to include % in the handout % % psnup -nup 6 -l -b0.5in temp.ps prospertemplate_handout.ps % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % Prosper offers lots of different user-contributed slide styles % with different type and background colors. Here, we use the % ``azure'' style. Other styles available with the distribution of % Prosper we have in the Department include ``alienglow,'' ``autumn,'' % ``contemporain,'' ``darkblue,'' ``default,'' ``frames,'' ``lignesbleues,'' % ``nuancegris,'' ``troispoints,'' ``rico.'' On unity, add prosper and % then go to the directory /ncsu/prosper and various subdirectories % to see more. Additional styles are available on various Prosper % web sites. Try them; it's fun! (Some of these have bugs, which % will be obvious when you try them with this file.) Advanced users % can modify these or create their own! % % The ``pdf'' option produces output for presentation with a laptop; % the ``ps'' option produces output for printing (no overlays, etc). % The ps option has some bugs; I have found it safer to simply go to % ghostview with the ps file produced using the pdf option and marking % the slides I want to include in the handout and saving them to a file % which I can then print n-up. % See the Prosper manual on the class web page for additional info. % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \NeedsTeXFormat{LaTeX2e} \documentclass[pdf,lignesbleues]{prosper} % put your packages here \usepackage{psfig,pstricks,amsthm,amssymb,amsmath} \usepackage{graphicx,psfrag,pst-node,subfigure} \renewcommand{\slideleftmargin}{-1.5in} % defining colors: % \newgray{vlgray}{0.9} \newrgbcolor{lblue}{0.8 0.92 0.95} \newrgbcolor{lred}{1 0.8 0.8} \newrgbcolor{lyellow}{1 1 0.6} \newrgbcolor{orange}{1 0.7 0.2} \newrgbcolor{lgreen}{0.87 0.95 0.8} %\newrgbcolor{dgreen}{0.37 0.43 0.37} \newrgbcolor{dgreen}{0.36 0.75 0.41} \newrgbcolor{violet}{0.6 0.0 0.4} % % put your newcommands % % Command so first word has color I use in azure \newcommand{\citem}{\item $\mbox{}$} % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % % Info for the title slide here - it will be created automatically % by the ``maketitle'' command in the body of the document. This must % be in the preamble (before the \begin{document} statement) % This puts the NCSU logo in the lower left corner of each slide \Logo(-1.0,-1.1){\includegraphics[height=0.25cm]{newlogo}} \title{Professional Ethics for Statisticians:} \subtitle{Issues and Advice} \author{\large ST 810A, Spring 2005} \vspace*{0.2in} \institution{\includegraphics[height=0.8cm]{deptlogo}} \slideCaption{ST 810A, Spring 2005} % \slideCaption will also number each slide as n/N, where n is % the current slide and N is the total. If you want to suppress % the total, include the ``nototal'' option in the [] part of % the documentclass statement % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \setlength{\slidewidth}{8.0in} \begin{document} % Creates the title slide \maketitle % Now start the rest of the slides \overlays{6}{ \begin{slide}{Outline} \begin{small} \begin{itemstep} \item General comments on the role of statisticians \item Responsibilities and advice for graduate students \item Responsibilities and advice for teaching \item Responsibilities and advice for methodological research \item Responsibilities and advice for working with collaborators/clients \item Issues: authorship, objectivity, etc. \end{itemstep} \end{small} \end{slide} } \begin{slide}[Glitter]{General Comments} \begin{small} {\red The discipline of Statistics:} \begin{itemize} \item Statistics is concerned with {\blue collection} and {\blue interpretation} of information (data) \item In some sense, Statistics is {\blue unique}, as we would not exist but for the need to do this in other disciplines \item For knowledge in all disciplines to advance uncompromised, collection and interpretation must be done in an {\blue honest} and {\blue principled} fashion \end{itemize} \end{small} \end{slide} \begin{slide}{General Comments} \begin{small} {\red Restated}: Statistics is concerned with {\blue honest} and {\blue principled} collection and interpretation of information (data) \begin{itemize} \item The contribution of statistics is to make sure that evidence on which public and private decisions are based and on which the next steps in advance of science are formulated is {\blue credible} \end{itemize} \end{small} \end{slide} \begin{slide}{General comments} \begin{small} {\red Collection and interpretation of data:} \begin{itemize} \item In general, investigators are focused on pursuit of \\ {\blue the truth} \item Excited investigators have hypotheses to verify, ideas to defend, and hope that data may be collected that have an interpretation that supports their objectives \item However, hypotheses and ideas can be {\blue incorrect}, or may be correct but evidence in their favor can be {\blue weak} or {\blue inappropriate}\ldots \end{itemize} \end{small} \end{slide} \begin{slide}{General comments} \begin{small} {\red The role of statisticians:} \begin{itemize} \item Ensure that data are collected in a {\blue fair} and {\blue unbiased} way so as not to favor certain viewpoints \item Ensure that questions to be addressed can {\blue in fact} be addressed with available information \item Ensure that interpretations are made {\blue independently} of desired results or predispositions \item To some extent, statisticians are the ``{\blue morality police}'' for science! \end{itemize} \end{small} \end{slide} \begin{slide}{General Comments} \begin{small} {\red In particular:} \begin{itemize} \item The FDA bases regulation and approval of drugs, biologics, and devices on formal statistical evidence \item Many substantive journals require that claims be supported by formal statistical evidence \item Environmental regulations are based on formal statistical approaches and evidence \item Teams conducting clinical trials are almost always required to include at least one statistician \end{itemize} \end{small} \end{slide} \begin{slide}{General Comments} \begin{small} {\red At the same time:} \begin{itemize} \item Statistics has a ``{\blue PR}'' {\blue problem} \item {\em How to Lie With Statistics} (Huff, 1954) \item ``There are three kinds of lies: lies, damned lies, and {\blue statistics}'' (Benjamin Disraeli) \item Despite our role, the public is {\blue suspicious} of statistics \end{itemize} \end{small} \end{slide} \begin{slide}{General Comments} \begin{small} {\red Result:} Statisticians have a {\blue special obligation} to practice our profession {\blue independently} and with {\blue principle}, {\blue honesty}, {\blue integrity}, and {\blue fairness} \begin{itemize} \citem {\blue More generally}, we have a responsibility to act this way in {\blue everything} we do, statistical or otherwise \end{itemize} \end{small} \end{slide} \begin{slide}[Glitter]{Advice for Graduate Students} \begin{small} {\red Get in the habit early in your career!} \begin{itemize} \item Play by the rules in your coursework, {\blue always} -- if the instructor directs you not to collaborate, consult certain materials, etc., {\blue DON'T DO IT!!!} \item Not adhering to the rules now will make it more difficult for you to do so later, when it matters {\blue even more}\ldots \item \ldots and you will have not developed the skill of {\blue learning independently} that is critical to being a good statistician (in research or applications) \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Graduate Students} \begin{small} {\red Coursework:} Don't cheat yourself \begin{itemize} \item Treat your coursework as an opportunity to learn and hone {\blue technical/applied skills}, demonstrate your {\blue ability to learn and master} your discipline, cultivate an ability to {\blue work independently} as preparation for your later {\blue work habits}, {\blue NOT} as a just a prerequisite for an ``{\blue A}'' \item Accept the idea that graduate school is {\blue NOT} about grades! \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Graduate Students} \begin{small} {\red Coursework:} Don't cheat yourself \begin{itemize} \item Graduate school is one time in your professional life where you have the {\blue luxury} to learn \item Take demanding (not ``easy A'') courses that will {\blue expand} your knowledge of Statistics and {\blue challenge} your abilities \item Take courses {\blue beyond} the minimum requirement for your degree \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Graduate Students} \begin{small} {\red Dissertation research:} Don't cheat yourself \begin{itemize} \item A dissertation is an exercise in learning to work and think {\blue independently}, critical to your ability to contribute and carry out your responsibilities later \item This is {\blue NOT homework}! Do not expect your advisor to ``{\blue assign}'' the next task \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Graduate Students} \begin{small} {\red Dissertation research:} Don't cheat yourself \begin{itemize} \item Don't just get results, {\blue interpret them}! Do the results make sense? Could there be a mistake in your program? \item {\blue Identify the \blue next step yourself}: investigate new approaches on your own, try new simulation scenarios, etc. \end{itemize} \end{small} \end{slide} \begin{slide}[Glitter]{Advice for Teaching} \begin{small} {\red Not just academia:} Industry, government short courses, presentations \begin{itemize} \item Teaching Statistics or communicating statistical ideas carries a responsibility for {\blue completeness}; {\blue honesty} about what can and cannot be done, assumptions, etc.; staying {\blue up-to-date} \item \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Teaching} \begin{small} {\red Responsibility:} To students and the profession \begin{itemize} \item Developing and maintaining a {\blue good} course is {\blue hard work} and time-consuming, but cutting corners will only shortchange the profession by telling an incomplete or outdated story to students \item Promoting Statistics to non-majors (and majors!) through clear, thoughtful teaching benefits the {\blue entire discipline} \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Teaching} \begin{small} {\red Responsibility:} To students and the profession \begin{itemize} \item Teaching is {\blue not} a popularity contest! An {\blue easy}, {\blue entertaining} course and {\blue light workload} may please students in the short term, but fails them in the longer term \end{itemize} {\red Your goal:} Communicate statistical thinking and principles in the best way you know how \end{small} \end{slide} \begin{slide}[Glitter]{Advice for Research} \begin{small} ``{\red Publish or perish}'' In academia, pressure to publish is {\blue considerable} in any discipline \begin{itemize} \item The eternal struggle: {\blue Quantity} vs. {\blue quality} \item The ``{\blue minimum publishable unit}'' \item Temptation to send out work that is {\blue incomplete} or promote methods that are {\blue not well-studied} \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Research} \begin{small} {\red Responsibility:} The objective of research is to {\blue advance knowledge}. Ask yourself \begin{itemize} \item Does this work represent a ``{\blue $>\epsilon$}'' advance? \item Have I portrayed the advantages {\blue and disadvantages} fairly and accurately? \item Have I done my best to communicate my work {\blue clearly} and {\blue effectively}, rather than expecting editors/referees to do it for me? \end{itemize} \end{small} \end{slide} \begin{slide}[Glitter]{Advice for Collaboration} \begin{small} {\red The practice of statistics:} As a statistician, you will work with investigators/clients on ``{\blue real problems}'' \begin{itemize} \item As a {\blue consultant} to a client who has already collected data and wants help with an analysis \item As a {\blue collaborator} who participates in conception, design, analysis, and reporting of results (publication) \end{itemize} {\red The most challenging aspect of being a statistician!} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red First principle:} Be {\blue principled}! \begin{itemize} \item Statistics is about {\blue impartial}, {\blue fair}, {\blue unbiased} interpretation \item Statistics recognizes and is {\blue up-front} about the {\blue limitations} of methods and assumptions \item A statistician should always strive to present {\blue only} what is reasonable to infer, to make limitations {\blue transparent}, and not be pressured by ``{\blue special interests}'' \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Example 1:} The mission -- {\blue Statistical Significance}! \begin{itemize} \item Investigators obviously hope to show their hypotheses, theories, conjectures are {\blue true} and would like statistical analyses that support this! \item It's all about the {\blue p-value}! \item Some client/collaborators view statistical methods a ``bag of tricks'' \item ``What if we analyzed this as a two-way analysis of variance instead?'' \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Statistician's obligation:} \begin{itemize} \item Use the most relevant model and method with reasonable assumptions, regardless of outcome \citem {\blue Stand your ground} \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Example 2:} The sample size conundrum \begin{itemize} \item A collaborator wants to conduct a study to test a hypothesis but has {\blue limited resources} \item You: ``What's an {\blue important (subject-matter) difference}?'' \item Them: ``Whatever I can detect with at most 50 patients/rats/plots'' \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Statistician's obligation:} \begin{itemize} \item It may be a waste, or downright {\blue unethical}, to conduct a study that is {\blue too small} (underpowered) or {\blue poorly designed} \item Help the client to understand what s/he can reasonably hope to achieve within his/her constraints and to focus on the {\blue scientific relevance} given the limitations \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Example 3:} The color of money \begin{itemize} \item If the FDA approves this drug, we'll all be rich! \item The word from upper management: Statisticians, your {\blue job} is to deliver {\blue p<0.05} \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Statistician's obligation:} \begin{itemize} \item This can be {\blue tough}! \item Strength of character, integrity, {\blue self-confidence} (comes from knowing your discipline very well) \item Strive within your organization to {\blue educate} \end{itemize} \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Example 4:} The VIC (Very Important Client) \begin{itemize} \item Unfortunately, some clients/collaborators do not think highly of Statistics or statisticians (and view them as ``{\blue staff}'' who run software packages\ldots) \item Unfortunately, some clients/collaborators think they know {\blue as much or more} than you do \item ``Run this analysis and bring me the results'' \end{itemize} {\red Statistician's obligation:} See previous slide! \end{small} \end{slide} \begin{slide}{Advice for Collaboration} \begin{small} {\red Example 5:} The obsessive-compulsive (you!) \begin{itemize} \item Sloppiness or haphazard work is {\blue unacceptable} \item ``The mystery of the treatment codes'' \end{itemize} {\red Statistician's obligation:} Be precise and careful with every design, analysis, report, {\blue no matter} how small \end{small} \end{slide} \begin{slide}[Glitter]{Issues} \begin{small} {\red Authorship:} For collaborative work \begin{itemize} \item If you have made a {\blue contribution} to a paper, you should be listed as an {\blue author} \item In some substantive disciplines, there is a {\blue protocol} for authorship \item In others, you may have to {\blue fight} for recognition \end{itemize} \end{small} \end{slide} \begin{slide}{Issues} \begin{small} {\red Reporting results:} For collaborative work \begin{itemize} \item Collaborators (and their journals) often insist on {\blue minimal} statistical content/description \citem {\blue Insist} at the minimum that the paper state clearly: statistical model (reference), assumptions, limitations, extent of generalizability \end{itemize} \end{small} \end{slide} \begin{slide}{Issues} \begin{small} {\red Know what you don't know:} \begin{itemize} \item You can't be a statistician in a vacuum; when working with subject-matter investigators, make an effort to learn the basics of the area\ldots \item \ldots while recognizing you are not an {\blue expert} \item {\blue Listen} to your collaborators, {\blue ask questions} \item Don't be afraid to be ``{\blue dumb}!'' \end{itemize} \end{small} \end{slide} \begin{slide}{Issues} \begin{small} {\red Peer review:} \begin{itemize} \item When acting as a referee, resist the temptation to be {\blue competitive} \item Your goal should be to evaluate whether the contribution is {\blue substantial enough} to be communicated -- be honest, fair \end{itemize} \end{small} \end{slide} \begin{slide}[Glitter]{Some References} \begin{small} \begin{description} \item Vardeman, S.B. and Morris, M.D. (2003) Statistics and ethics: Some advice for young statisticians. {\em The American Statistician} {\bf 57}, 21--26. \item American Statistical Association (1999) Ethical guidelines for statistical practice (at the ASA web site; see the class web page) \end{description} Additional links are the class web page. \end{small} \end{slide} \end{document}