Free Essay


In: People

Submitted By suci
Words 5217
Pages 21
Statistics and Ethics: Some Advice for Young Statisticians Author(s): Stephen B. Vardeman and Max D. Morris Source: The American Statistician, Vol. 57, No. 1 (Feb., 2003), pp. 21-26 Published by: American Statistical Association Stable URL: Accessed: 22/04/2009 07:58
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact

American Statistical Association is collaborating with JSTOR to digitize, preserve and extend access to The American Statistician.


Statistics and Ethics: Some Advice for Young Statisticians and Stephen B. VARDEMAN Max D. MORRIS and technique, but ratherabout honesty. Its real contribution to society is primarilymoral, not technical. It is about doing the right thing when interpreting empiricalinformation.Statisticians are not the world's best computerscientists, mathematicians, or scientificsubjectmatterspecialists.Weare (potentially, at least) the best at theprincipledcollection, summarization, and of data. Our subjectprovides a frameworkfor dealing analysis and transparently consistentlywith empiricalinformationfrom all fields; means of seeing and portrayingwhat is true;ways of avoiding being fooled by both the ill intent (or ignorance) of others and our own incorrectpredispositions.The mix of theory and methods that you are discovering is the best available for achieving these noble ends. The more you practicewith it, will become your (fundamentally the sharper moral)judgments aboutwhat is appropriate handlingempiricalinformation. in Othersfrom areasrangingfrom philosophyto physics might well object that we have claimed too much, wrappingstatistics in a cloak of virtueto the apparent exclusionof otherdisciplines. After all, thoughtfulscientists and humanistsfrom a varietyof fields are engaged in the pursuitof truth.And any serious education has moral dimensions. Our point, however, is that the role that the profession plays in science and society particular shouldnot be viewed as amoral,andthatthis fact constrainshow we all must thinkand act as its members. That society expects our profession to play this kind of role can be seen in the place statisticshas as arbiterof what is sufficient evidence of efficacy and safety to grant FDA approval of a drug, or enough evidence to supportan advertiser'sclaim for the effectiveness of a consumerproduct.And it can be seen in the fact that many disciplines have "statisticalsignificance" for requirements resultsappearingin theirjournals. also recognizes that when statistical argumentsare Society abused,whetherthroughmalice or incompetence,genuineharm is done. How else could a book titled How to Lie WithStatistics (Huff 1954) have ever been published and popular?The famousline (attributed MarkTwain(1924) to BenjaminDisby raeli) "Therearethreekindsof lies: lies, damnedlies, and statistics" witnesses effectively to society's distaste for obfuscation or outrightdishonestycloaked in the garbof statisticaltechnology. Society disdainshypocrisy.It hatescrookedlawyers,shady corporateexecutives, and corruptaccountants,and it has contemptfor statisticiansandstatisticalworkthatlack integrity.But young statisticianssometimes find themselves being "encourof aged" to offer questionableinterpretations data. This pressurecan come even from well-meaningindividualswho believe thattheironly interestis in ensuringthattheirposition is treated "fairly."Maintainingan independentand principledpoint-ofThe American Statistician, February 2003, Vol. 57, No. 1 21

We write to young statisticiansaboutthe natureof statisticsand theirresponsibilitiesas membersof the statisticalprofession.We observethatthe practiceof the disciplineis inherentlymoraland that this fact has serious implicationsfor their work. In light of this, we offer some advice about how they should resolve to think and act. KEY WORDS: Graduatestudy; Integrity;Principle;Professional practice;Research;Teaching.

Dear Gentle Reader: So, you are embarkingon a careerin statistics.Good. It is a genuinely noble pursuit,though this may be hardto see as you wrestle with new-to-you technical issues varying from "How do I get this SAS job to run?"to "How do I show this thing is UMVU?"andon occasion findyourselfwondering"Whatis the point of all this?" This last question aboutpurposeis actuallya very important and quite serious one. It has implications that run far beyond How you anyour present pain (and joy) of "gettingstarted." swer it will affect not only you, but also the profession, and human society at large. We write to offer some advice and encouragement,and to say how we hope you frame your answer to this simultaneouslypracticaland cosmic question. What are this subject and this profession really all about? And why are you doing what you are doing? For sure, there are details to learn (and keep currenton throughouta career). Thereis everythingfrom the seemingly uncountablenumberof tricks of first year probabilitytheory,to statisticalcomputing, to nonlinearmodels. It initially looks like "soup to nuts."You know thatstatisticsis aboutcollecting andhandlingdata.Thatis true,but incomplete;thereis much more thanthatat work here. The vital point is that this discipline providestools, patterns of thought,and habits of heartthat will allow you to deal with data with integrity.At its core statisticsis not about cleverness
Stephen B. Vardemanis Professor, and Max D. Morris is Professor, Departmentof Statistics and Departmentof Industrialand ManufacturingSystems Engineering,Iowa State UniversityAmes, IA 50011-1210 (E-mail:vardeThe authorsgratefullyacknowledgethe generous inputof a numberof colleagues. KarenKafadar, Bob Stephenson,Bill Meeker,and Dean Isaacsonprovideddetailedcommentson a firstdraftof this article.And the input of Ken Ryan, TammyBrown, Mike Moon, Bill Notz, Tom Dubinin,FrankPeters, David Moore, Bill Duckworth,Bruce Held, Dennis Gilliland,Bobby Mee, Doug Bonett, Dan Nettleton,and Hal Sternis also gratefullyacknowledged.

?) 2003 AmericanStatisticalAssociation DOI: 10.1198/0003130031072

view in such contexts is critical if a statisticianhopes to avoid becoming a partof Disraeli's third"lie." So, you areembarking upona noble and seriousbusiness.We take as given that you have a basic moral sense and a strong desire to personallydo good. We also take as self-evident that integrityis a patternof life, not an incident. Principledpeople consistentlydo principledwork,regardlessof whetherit serves is theirshort-term personalinterests.Integrity not somethingthat is turnedon and off at one's convenience.It cannotbe generally lackingandyet be countedon to appearin the nick of time when the greater good calls. This implies that what you choose to thinkand do now, early in your career,are very good predictors of what you will think and do throughoutthe whole of it. You are settingpatternsthatwill endureover a professionallifetime influencethe natureandvalue of whatyou can and substantially to accomplish. hope A fair amounthas been writtenabout professionalethics in statisticsand we do not proposeto review it all or commenton every issue thathas been raised.For example,Deming's (1986) articleis fundamentallya discussion of ethics. Both the AmerStatisican StatisticalAssociation (1999) and the International tical Institute(1985) have official statementson ethical guideAnd in a moregeneralsetting,the National lines for statisticians. Academyof Sciences (1995) has publisheda useful bookletthat is primarilyaboutethics in science andhas implicationsfor statistical practice. Our more specific goal here is to suggest some things that a high view of the discipline means for your presentwork and attitudes.Aiming to speak to both statistics graduatestudents and recentgrads,we'll begin with some implicationsfor life in graduateschool, and then move on to implicationsfor an early careerin the discipline. ADVICE FOR STATISTICS GRADUATE STUDENTS "Graduatestudent ethics" (or for that matter "professional ethics")is reallyjust "plainethics"expressedin a graduatestudent(or professional)world.A discussionof it reallyboils down to considerationof circumstancesand issues that arise in a particulargraduatestudent(or professional)setting. So an obvious place to begin is with generalstudentresponsibilities.If you are still in graduate school, we urgeyou to be scrupulousaboutyour conductin the courses you take. Here are some specifics: * Resolve to never accept credit for work that is not your own. It should make no difference to you whetheran exam is Whateverthe homeworkpolicy of the or proctored unproctored. course, make it your practice to clearly note on your papers places whereyou have gainedfrom discussionswith classmates or consultingold problemsets of others.It's simply rightto give otherscreditwhere it is deservedand it's simply wrong to take creditwhere it is undeserved. * If course policy is that everyone is "completely on their own,"resolve in advanceto politely refuse to discuss with peers topics thatareoff-limits, even if othersviolate the policy. It may seem a smallthingat the time, butyou aresettinglife trajectories incidents. thatare bigger thanthe particular
22 General

* Determineto nevertake advantageof (or over) your peers. If you join a group study session, be ready to make your fair not contribution, just to benefit from the input of others.If you have legitimateaccess to old files or notes or textbooksthatare helpful, let others know about them so that they can benefit as well. What do these threepoints say? Simply thatyou shouldplay by the rules set out and be clear and honest about all contributions made to the work you turn in. Why would anyone do otherwise?Honestly,only to gain an undeservedadvantagein a coursegrade,or to avoid some effort.But a studentwilling to cut corers for an A or a free weekend will have serious difficulty not cutting corners in later professional responsibilitieswhen the rewardis a promotionor pay raise or a free weekend. Some additionalissues arerelatedto the notion of "doingthe hardthing."Everyonehas thingsthatcome harderfor themthan others. It's humannatureto want to avoid what is difficultand to even convince ourselves thatreally, the easy thing is what is importantand the hardthing is worthless. But that is not only obviously silly, it has moral implications.Here is some advice for the studentreader: * Understandthat acquiringan advancededucationis a difficult enterprise,that there may be times when you feel like complainingabout this, but that it doesn't really help to do so. Whiningwastes energyandcan poison the learningatmosphere for others.Youareengagedin a noble, if difficult,pursuit.Give it yourbest shot withoutcomplaining.Afterall, most thingsworth doing are hard. * Resolve to work on your weaknesses ratherthan excuse them.Doing good statisticalworkis important, demandsthe and best possible personaltool kit. The reasoning "I find methods (theory) easier than theory (methods), so I'll just do methods (theory)"implicitly and quite wrongly assumes that one can do good statisticalwork with half a tool kit. * Decide not to denigratethe strengthsof others.Give other people credit for what they can do that you cannot. Find your niche without minimizing the honest efforts and contributions of others. * Determineto takethe coursesthatwill enable you to be the best-educatedand most effective statisticianyou can be. These are often academically demanding, and may not form a particularlyeasy route to a high GPA. While difficulty,per se, is not necessarily a measure of how often you will find the material in a course useful, it is related to the mental discipline you will develop. If you choose a course that covers material you could easily pick up on your own or because it is taught by a professorwho demandslittle in exchange for an A, you've cheated yourself. The choices you make about curriculumare moralchoices, notjust choices of convenience.You have a limited time in graduateschool ... use it wisely. How effective you will be as a professionaldepends on it. Besides, your choices say somethingnontrivialaboutthe personalcharacterthat you are developing. * Purposeto do what your thesis or dissertationadvisorsets for you to do, as independentlyas you can. While it may seem or that some assignmentsare arbitrary unnecessary,remember

thatyou do not haveyour advisor'sexperienceas a researcher or This personknows whatyou know,whatyourabilities educator. are, and the difficulty of your problem. He or she is trying to help you to develop as a responsibleand independentmember of the profession,one accustomedto consistentlyworkingup to your capabilities.Focusing your energy on the challenge of the problem and the opportunityit representswill take you much fartherthanwasting your energy in grumblingor in negotiating to be led throughevery detail of a solution. It is worthaddinga furthernote relatedto this last point. The advisor-advisee experience has the potential to be invigorating and rewarding(bothprofessionallyandpersonally)for both parties.Thinkof the effortsyou put into it not only as a requirement for the degree, but as the beginning of what may be one of your most importantand cherishedlong-termrelationships. Find someone to work with who you like and respect, and put your energy into the enterprise. Most statisticsgraduatestudentswork as graduateassistants. Assistants should rememberfirst that an assistantshipis not a fellowship, but rathera job. And it is axiomaticthat principled people returnhonest effortfor theirpay.If you areworkingon a faculty member'sgrant,thatperson must producequalitywork in line with the interests of some outside entity. Do what you can to help him or her. If you are a teaching assistant, there are lectures to conscientiously prepareand deliver, papers to carefully grade, and students to help. If you are a consultant, people with real problemsof data analysis will appearat your door seeking aid. They need your best effort and advice. Let us amplify a bit: * If you area researchassistantit is understood you have that own" class work and thesis or dissertationto attendto. "your But some of your weekly hoursare firstcommittedto providing the help (programming, librarywork, reportwriting,etc.) your needs. There are importanteducationalbenefits that employer accrueas you practiceat these duties. But the most fundamental reasonto carrythemout conscientiouslyandcheerfullyis simply thatit is the rightthingto do. (Andit is wrongto thinkthatcutting corers now doesn't say anythingaboutlaterbehavior.Life will always be hectic and there is no reason to expect your work habits after finishing school to be betterthan the ones you are developing now.) * If you are a teaching assistant,purpose to make the best of the fact that along with some conscientious, motivated,and intentionpleasantstudents,you will deal with some unpleasant, ally ignorant,lazy, and dishoneststudents.It simply comes with the territory. yourpart,makeit a pointto model integrityand For purposefor all of them. Do yourbest to convey thatwhatyou are teaching them really does matterand how they do it mattersas well. Resolve that whateveryour "style"/personality (from animated to reserved)your body language will convey a genuine willingness to help. Thejob takespatience-plan on it. Resolve to treatall of your studentswell, whetheror not their behavior in any sense merits that. And it should go without saying that althoughyou want to be pleasant and approachable, propriety

and impartialitydictate that you are their instructoror TA, not theirpal. * If your assignment is to help with statistical consulting, level) with some of the you are alreadywrestling(at a "trainee" seriousissues faced by one segmentof ourprofession.Carefully consider and handle these now, as you begin to see how the "humanelement" of statistical consulting requires thoughtful and principleddiscipline. You're going to have to argue with yourself in conversationslike: - Whatlooks to me like the thingthatshouldbe done would take two hours to explain and several more hours of my time to implement,while this client would be happywith that somethingless appropriate I could explainin five minutes ...


This client really wants "A" be true,but these datalook to inconclusive ...

- This looks pretty much OK except for that oddity over therethatthe client doesn't really want to discuss ... GraduateStudent Reader,keep your eyes open during this graduatestudent experience. Watch your faculty and emulate the ones who take seriously what they do. There are some fine role models in our university statistics departments,excellent membersof the profession.Findthem, andlearnas muchas you can aboutwhat they thinkand how they practicestatistics. ADVICE FOR YOUNG PROFESSIONAL STATISTICIANS Many of the themes we've introducedin the context of graduate study have their logical extensions to early professional life. But there are also other mattersthat we've not yet raised. We proceed to discuss some of the less obvious extrapolations and furtherethical issues faced by young statisticians,organizing our advice aroundthe topics of (1) research/publication, (2) teaching,and (3) professionalpractice. If you have finished a Ph.D., you have been introducedto the craft of researchin statisticaltheory or methods. You are in a position to help develop the profession's supportingbody of knowledge and to contributeto our journals. It's important to consider the correspondingresponsibilities.These are tied closely to a properview of the purposeof publicationin statistics. Published statisticalresearchshould provide reliable and substantialnew theoryor methodologythathas genuine potential to ultimatelyhelp statisticiansin the practice of the discipline. Statisticalpublicationshould not be treatedas a game. It is, and should be treatedas, a serious and moralbusiness. Here are some points of advice issuing from this high view of what the researchand publicationactivityis all about: * Resolve thatif you choose to submitwork for publication, it will be completeandrepresent yourbest effort.Submitting paof little intrinsicvalue, half-donework, or work sliced into pers small pieces sent to multiplevenues is an abuseof an important communication It systemandis not honorablescholarship. is not the job of editors or refereesto proofreador complete your paissues thatyou pers, or to insist thatyou follow up on important know exist. See the "Let'sjust send it off and let the reviewers
TheAmericanStatistician,February2003, Vol.57, No. 1 23

sort it out"impulse for what it is, a temptationto off-load your work to someone else. And the "I'll just submitthis half-done thing to an outletthatwill printanything" strategydoes nothing of real value for anyone.It wastes time and effortof those in the review system, and when "successful"it dilutes our literature. This makes importantwork harderto find, and in the end calls into questionour very reasonto exist as a profession. * Purposethatwhen askedto do thejob of a referee,you will do it thoroughly, and impartially, in as timely a manneras possible. Thereis no obvious short-term payoff to doing whatis right here. But the integrityand currencyof the scientificpublication process dependon competentandprincipledrefereestakingthe reviewjob, job seriously.Resolve neverto do a shoddy/cursory or worse yet to let calculationsaboutpersonalities(andpersonal governhow you judge a piece of work.Even though advantage) statisticsjournals use a "double-blind" system, the promany fession is small, and you will find it increasinglyrarethat you have no idea who authoreda paper you receive for review. So rememberthatthe spirit of the blindreview policy is honorable, and that you have an obligation to conduct your review in this And do what spiriteven when you cannotbe completely"blind." can as an individualto help fix the widely recognizedprobyou lem thatthe reviewprocessin statisticsis presentlymuch slower thanin many otherdisciplines. * Decide to routinelytake the advice of editors and referees regardingpapersthatyou submitfor publication.Occasionsare rarewhereeditorsor refereeshaveit all wrongor purposelytreat an authorunfairly.Most often, the advice they offer is constructive and when followed substantiallyimprovesan article.Until an editor signals clearly that he or she has no furtherinterest in a piece you have submitted,you should almost always make good faith efforts to revise your paperin accordwith his or her for advice. Serialjournal-shopping a venue that will publish a submissionwith essentially no revision may minimize the total effort an authorexpends on a paper,but the practicewastes the overallenergy of the professionand has a negativeeffect on the overall qualityof what is published.

giarismand is completely unacceptable.(This cautionextends, by the way, to thesis and dissertationwork, even if thatwork is never submittedto a journalfor formalpublication.) A note relatedto this last point: Avoiding plagiarismplaces an extraburdenon studentswhose writing skills are not strong, especially those strugglingwith English as a second language. But it is essential to find one's own words and not simply copy or even paraphrase those of another(even for parts of a paper thatarebackground obviously don't purport providenew and to technicalcontent).This is a very serious integrityissue. Next, let's consider issues relevantto teaching of statistics as a professional. There are reasons to do this whether or not you have plans for a career at a college or university.Teachis and ing/training increasinglydone "inhouse"by corporations and it could be arguedthat most professionalpreconsultants, sentationsareessentiallyteachingefforts.The logical extension of the advice offeredabove to graduateteachingassistantsis, of extradimension course, relevanthere. But thereis an important to discuss, relatedto the freedom and responsibilitythat a professional has in answeringthe question"Whatwill governwhat andhow I teach?"Will it be "What'seasy for me?"Or will it be "Whatwill get the best short-term reactionfrom the students?" Orwill it be "Mybest professional judgementas to whatthe students need for the long termand my best understanding how of to effectively convey thatinformation?" This is a moralchoice. Here is some amplification: * Determinethat you won't fall into the trap of organizing all courses aroundyour technical specialty.This is an issue of fundamental humilityandrecognitionthatnone of us has put all that is needed into our personal little package (to say nothing aboutthe matterof "truth advertising!"). we suspectthat in But you know what we are talking about, having seen people turn every course they teach into a platformto show off their own work.

* Purposenot to be governedby what is easy to do. This is not an entirelyseparateissue from the previousone. But we are also thinkingaboutcases wherethe case is not so blatantor not * Determine to be scrupulousabout giving credit where it tied directly to one's specialty. It's a lot of work to learn new is due. If anotherhas contributedsubstantiallyto the content methods and softwareto include in a course, to freshen examand and is of a paper,co-authorship typically appropriate should be ples, to develop new laboratories assignmentsfor students, offered. (On the otherhand,never list a colleague as co-author to replace outdatedtopics and means of presentation.And it's of a paperuntil you have thatperson'sexplicit permissionto do sometimespossible to "getby" withoutinvestingthateffort.But so.) And include acknowledgmentsof others deserving thanks doing so is simply wrong. We urge you not to take thatroute. for less extensive, but real, help with an article. * Resolve to do the best for yourstudents,whetheror not they * Resolve to acknowledgepriorityand the derivativenature appreciate youreffortsin the shortterm.We live in a "consumer" of your work with due humility.If after the fact of publication society.Thereis hugepressureon teachersin all contextsto make you find that some of your resultscan be found in earlierwork, studentshappy.But statisticsis hard,andstudentsDON'T know to immediatelysend an acknowledgment thateffect to the jour- whatthey need. Youwill. We hope thatyou opt to do yourbest to nal where your paper appeared.In writing your papers in the provide that, not simply what will get the best crowd reaction. and firstplace, we encourageyou to be forthright helpful about Lots of jokes, little in the way of course demands, and high know is alreadypublishedon yoursubject,delineating gradescan please manyaudiences.And leave studentsignorant. whatyou carefullywhatothershavealreadysaidandwhereyournew con- Of course we should aim to be engaging in our presentationof tributionlies. (No one ever really "startsfrom scratch."Don't our subject. But the point of teaching is to genuinely improve fall prey to the temptationto leave unsaid what you know is subjectmatterknowledgeandthe reasoningpowersof students. your own po- It is not to produce feel-good experiences for them. (In this alreadyknown,thinkingthatto do so strengthens And neverborrowpublished/copyrighted words,even of regard,we were recently dismayed to see an Iowa community sition.) To withoutacknowledgment. do so is pla- college presidentquoted in the Des Moines Register (2001) as your own authorship,


the firstandan interestedin furthering cause. In eithercase, it is axiomatic proudlysaying "Wearereally a service organization that your professionaljudgmentis potentiallyclouded by what educationalinstitutionsecond."While that may in fact be true, it is a terriblecommentaryon the state of the institution.) you (quite naturally)want to be true. And you will be no fair judge of the extent to which this clouding has occurred.There Those of you beginningacademiccareerswill face enormous is real dangerhere. There is little that is more damningto the demandsfor early success. Most universitiesrequiresubstantial discipline than for one of its professionals,implicitly claiming accomplishmentsin both researchand teaching duringthe first some degreeof objectivity,to be publiclyexposed as overstating six years of employment,and some place the bar so high that a statisticalcase in favorof his or her employeror cause. If effortis required. numbersof refereed seemingly superhuman Morecommonly,statisticiansfunctionas consultantsto those evaluationsarethe "keysto success," who must make decisions. We do this publicationsandinstructor through careful and can you affordto have real qualityas yourprimary goal? Is there thoughtfuldesign of data collection mechanisms and analysis if of assembleddata.But "carefuland thoughtful" enoughtime in six shortyearsto accomplishall thatis required here are words you takeouradvice seriously?These arerealandhardquestions. that acknowledgea criticalfact: Statisticalanalysis of data can is How you use yourassistantprofessorship criticalto yourlong- only be performedwithin the context of selected assumptions, term professional success, and it is obvious that you must take models, and/or prior distributions.A statisticalanalysis is acyour institution'sexpectationsinto account. But, we urge you tually the extractionof substantiveinformationfrom data and as you face these issues to rememberthat one who spends an assumptions.And herein lies the rub, understoodwell by Disassistantprofessorshipcutting corers is at best preparedto be raeliandothersskepticalof ourwork:Forgiven data,an analysis an associateprofessorwho knows how to cut comers ... not one can usuallybe selected which will resultin "information" more who has learnedhow to make a difference. favorableto the owner of the analysis than is objectively warTurningfinally to the area of professionalpractice,we note ranted. The only "cure" this difficultyis statisticalpracticebased for that most of what has been writtenabout ethical guidelines for on assumptionsembodying an informed,balanced,and honest in statisticiansconcerns what is appropriate public practice,in of lending aid to othersin the impartialandefficientcollection and representation what is known. "Known,"not "wished for," or This "convenient," even "other-than-worst-fears." as analysis of their data. This is understandable, (1) the disci- "desired," has implicationsfor how statisticiansmustbe and act if they are pline's whole reasonto exist is ultimatelyto providesuch aid and (2) this activity is both subtle and full of pitfalls. Both the ethi- to be both effective and ethical. cal guidelines and public skepticismtypifiedin the "lies"quote * Statisticiansmust be knowledgeableabout the system unof Disraeli point to the fact that statistics can be used to form der study. They should not presentthemselves as competentto highly technical and even technically correctsupportfor state- analyzedatafrom systems aboutwhich they haveno substantive ments which are in fact not true.We might hope this could hap- understanding. Real dataare not "context-free." practicestatisticswithoutproper pen only when nonstatisticians * On the other hand, statisticians must recognize and acof technicalunderstanding the subject.But statisticallies areby definitionimmoraluses of statisticalarguments, whethertechni- knowledgethe limitationsof their "subjectmatter"knowledge. Data andvariationareubiquitous.Knowinghow to handlethem cally corrector not, and stem from societal pressuresthataffect and statisticiansand nonstatisticiansalike. What then must you do can give you important even uncommoninsightsin a variety of contexts where you have limited subject mattercredentials. in society to preservethe discipline's (and your own) integrity? But the fact thatyou can make contributions league with exin First, recognize that a professional statistician should never in a variety of fields doesn't substitutefor credentialsin behave like a courtroomlawyer. The practice of law is based perts those fields. The credibilityof the statisticalprofessiondepends model in which each lawyer representsan ason an adversarial upon its membersbeing scrupulousaboutwhat they know and signed point of view-that which will yield the most positive what they don't know.Never forget thatyou are not the context outcome for his or her client. While the use of lies and intenexpert. tionally misleading statementsis prohibitedin legal proceed* Statisticiansmustgo out of theirway to see thattheiranalyings, legal strategycertainly does involve the selective use of evidence so as to presentthe truth(or some partof it) in the light ses allow interpretations the available data whichare tenable of most favorableto a particular of view. But a key aspect of but not popular in the statistician'sorganization.This does not point but this model of litigationis thatdecisions aremadeby an unbiased mean "be a troublemaker," it does mean that you should or jury) based not on the case presentedby a carefullythinkthroughhow availabledatawould be interpreted authority(a judge single side, but only afterarguments presentedby all partiesare by those with all possible rationalpoints of view. heard. * Statisticiansmustwritecompletereportsstatingthe results Statisticians usuallydo not operatein suchwell-controlledad- of theirentireinformedthoughtprocesses-including what they versarialsystems. If you do work in this kind of arenayou must know,what they have assumed,what they have decided cannot keep absolutely clear the distinctionbetween an objective ana- be assumed,and whatconclusionstenableassumptions support. to lyst and an advocate,and neverpurport be (or thinkyourself) Our reportsshould contain "completeand sufficient"analyses the firstwhen you arethe second. If you areemployedby an or- upon which any rationalpoint of view can be argued. If you basis or as a consultant)you come to the conclusion that one of the spectrumof sensible ganization(whetheron a permanent are by definitionnot disinterestedin its well-being. And if you interpretations "best"in a particular is application,makeit your areworking"probono"for a cause you support,you arenot dis- goal to be absolutelytransparent aboutyour reasoning.People
TheAmericanStatistician,February2003, Vol.57, No. 1 25

* You must understand fully what your assumptionssay and what they imply. You must not claim that the "usual assumptions" are acceptable due to the robustnessof your technique unless you really understand implicationsand limits of this the assertionin the contextof yourapplication.And you must absolutely neveruse any statisticalmethodwithoutrealizingthatyou areimplicitly makingassumptions,and thatthe validityof your resultscan neverbe greaterthanthatof the most questionableof these. Youcannot do this unless you remaindedicatedto being mustbe to findingthe conclu- thebest technicalstatisticianyou canpossibly be, understanding As a statistician, yourallegiance the sions which can be supportedby data and carefulassumptions. that this involvesknowingand understanding mathematical as Does this makethe businessof assumptionselection more diffi- arguments well as the computational techniquesbehindevery Does it seem as tool you need. cult thanit seemedin yourstatisticscoursework? thoughyou musttakethese issues morepersonallyandseriously Well there it is, more than enough advice to keep a young thanourfavoritesemi-academic phrase"LetX1, X2,..., X, be statisticianbusy for a career.We hope we don't soundtoo much of iid F ... ?"Does it soundlike yourformulation these assump- like myopic cranks,finding "seriousethical issues" to raise in values than even the most mundanecontexts. more to do with nonmathematical tions may have Instead,we hope thatwe have has been discussedin yourtextbooks?Yes, this andmoreis true. arguedeffectively that ethical mattersare centralto our disciEthical statisticalpracticerequiresthat you take responsibility pline and providedsome insight into issues thatthis raises. We for acquiring substantiveunderstanding, knowing all rational furtherhope that you determineto take the matterof principle points of view, and making decisions well beyond those based most seriously. entirelyin data. Carryon, Gentle Reader. * You must examine yourself to see that you are not even subconsciouslyleaning towardanalyseswhich you believe will [ReceivedApril 2002. RevisedNovember2002.] "please the boss" or yourself, or simplify the problemunjustifiably. This means that you cannot afford to think of yourself REFERENCES as a datatechnicianor a hired gun. You must be secure enough to simultaneouslyseparateany prior vested interest (yours or AmericanStatisticalAssociation(1999), "EthicalGuidelinesfor StatisticalPractice," others') in the outcome from your analysis, and meld together Deming, W. E. (1986), "Principlesof ProfessionalStatisticalPractice,"in Enseamlesslyeverythingyou know aboutthe subjectmatterof your cyclopedia of StatisticalScience (vol. 7), eds. S. Kotz and N. Johnson,New of York:Wiley. investigationwith the structure yourstatisticalwork. YoucanIowa not do this unless you have strengthof characterand integrity. Des MoinesRegister(2001), "Western TechAmongNation'sFastestGrow* You must not stop with the obvious or even the most likely explanationof data, but find ways to examine them so that all rationalviewpoints can be informed.This means that you will workharderandlongerthananyonewho readsyourreportswill ever know. You will not rest until you know you understand all the informationcontainedin the data, where "information" ing Schools,"December 31, 2001, pp. B-1. Huff, D. (1954), How to Lie with Statistics,New York:Norton. International StatisticalInstitute(1985), "Declarationon ProfessionalEthics," NationalAcademy of Sciences (1995), On Being a Scientist:ResponsibleConduct in Research,Washington,DC: NationalAcademy Press. Also available online at New York:Harper& Brothers. Twain,M. (1924), MarkTwain'sAutobiography,

shouldbe able to easily see your full set of model assumptions, understand whatmethodologyyou haveused to makeinferences in thatmodel, andhaveaccess to diagnosticandrobustnesswork you havedone. (This adviceis soundin general.But it is perhaps especially relevantto explicitly Bayesian analyses.A consumer of a posteriordistribution a moralrightto knowhow strongly has it dependsupon the prior.)Honest statisticalwork has nothing to hide. It says what it says. It doesn't try to obscure points where alternative conclusions are possible if otherassumptions are made or different analysis paths are followed, and admits wheremodel fits areshortof perfectionorconclusionsarehighly model-dependent.

is defined by the context of your work across the spectrumof rationalviewpoints. Youcannot do this unless you develop an ethic of self-reliance,thoroughness,and hardwork.



Similar Documents

Free Essay

Ayo Belajar Statistik

...Contoh Regresi Linier Sederhana dengan SPSS Artikel ini akan mengupas contoh regresi linier sederhana dengan SPSS menggunakan data regresi menggunakan data yang dipakai seperti pada perhitungan korelasi. Analisis regresi sederhana merupakan salah satu metodi uji regresi yang dapat dipakai sebagai alat inferensi statistik untuk menentukan pengaruh sebuahvariabel bebas (independen) terhadap variabel terikat (dependen). Regresi linear sederhana ataupun regresi linier berganda pada intinya memiliki beberapa tujuan, yaitu : • Menghitung nilai estimasi rata-rata dan nilai variabel terikat berdasarkan pada nilai variabel bebas. • Menguji hipotesis karakteristik dependensi • Meramalkan nilai rata-rata variabel bebas dengan didasarkan pada nilai variabel bebas diluar jangkaun sample. Pada analisis regresi sederhana dengan menggunakan SPSS ada beberapa asumsi dan persyaratan yang perlu diperiksa dan diuji, beberapa diantaranya adalah : • Variabel bebas tidak berkorelasi dengan disturbance term (Error). Nilai disturbance term sebesar 0 atau dengan simbol  sebagai berikut: (E (U / X) = 0, • Jika variabel bebas lebih dari satu, maka antara variabel bebas (explanatory) tidak ada hubungan linier yang nyata, • Model regresi dikatakan layak  jika angka signifikansi pada ANOVA sebesar < 0.05, • Predictor yang digunakan sebagai variabel bebas harus layak. Kelayakan ini diketahui jika angka Standard Error of Estimate < Standard Deviation, •......

Words: 669 - Pages: 3

Free Essay

Ihsg Dalam Statistik

...IHSG dalam Statistik Posted on December 27, 2008 by parahita   Terdapat sekumpulan teori yang menyatakan bahwa ada suatu hari, bulan, atau waktu tertentu di mana pergerakan indeks mengalami anomali.  Oleh karenanya berdasarkan teori-teori tersebut kita bisa mengetahui kapan waktu yang bagus atau jelek untuk berinvestasi. Teori-teori tersebut biasanya disebut dengan efek kalender. Teori-teori yang sudah umum didengar antara lain adalah monday effect, january effect, october effect, dan beberapa teori lainnya. Apakah teori-teori tersebut benar? Yah, namanya juga teori, bisa benar bisa juga tidak. Terkadang malah terdengar seperti mitos. Walaupun begitu, mungkin ada rasa penasaran yang timbul dalam hati kita untuk membuktikan kebenaran teori-teori tersebut. Saya akan mencoba membeberkan bagaimana perilaku IHSG pada waktu-waktu tertentu. Saya tidak akan mengikuti teori yang ada. Saya hanya akan mencoba memaparkan data dan Anda bisa menyimpulkan sendiri bagaimana hasilnya [pic] 1.   Bagaimana perilaku harian IHSG (Senin – Jumat)? Mitos yang beredar selama ini adalah jangan trading di hari  Jumat karena banyak trader yang menjual sahamnya. Hari Jumat yang merupakan hari perdagangan terakhir mengandung risiko yang besar karena pada hari Sabtu Minggu kita tidak akan pernah tahu apa yang bisa terjadi. Benarkah? Dengan mengolah data harian IHSG dari tahun 1984 sampai dengan 2008 (Desember) ternyata hasilnya adalah sbb: [pic] Tidak seperti mitos yang ada, pada hari Jumat IHSG......

Words: 1716 - Pages: 7

Free Essay


...SEKOLAH TINGGI ILMU STATISTIK - JAKARTA Status: Kedinasan (Badan Pusat Statistik, Sekretariat Negara) Alamat: Jl. Otto Iskandar Dinata (Otista) 64 C, Jakarta Timur 13330 Telepon: (021) 8197577, 8191437, 8508812 Faks. : (021) 8197577 Website : E-mail: Ketua: Dr. Jopie Bambang Soenjoto Sejarah Singkat Tanggal Berdiri: 11 Agustus 1958 Pendiri: Badan Pusat Statistik, Sekretariat Negara STIS, yang sebelumnya bernama Akademi Ilmu Statistik (AIS), merupakan lembaga pendidikan tinggi pertama di Indonesia yang menyelenggarakan pendidikan khusus untuk mencetak tenaga-tenaga semi ahli dalam bidang statistik. Tenaga-tenaga tersebut diperlukan sebagai pelaksana kegiatan statistik di lingkungan departemen, bank, perusahaan, antara lain. Lulusan sekolah tinggi ini bisa melanjutkan ke jenjang S1 di IPB, Unpad, UGM, dan ITS. Untuk meneruskan ke S2 dan S3, mereka dapat melanjutkan ke Amerika, Kanada, Australia, Jepang, Inggris, Jerman, dan Belanda. Sejak tahun akademik 1997/1998, STIS menyelenggarakan program pendidikan diploma IV dan diploma III. Status mahasiswa STIS terdiri dari Ikatan Dinas (ID) bagi lulusan SMU dan Tugas Belajar (TB) bagi pegawai negeri/anggota ABRI. Profil Jenjang pendidikan: D3, D4 Jumlah mahasiswa, 2002: 685 Jumlah lulusan, 2002: 168 Jumlah dosen, 2002: 98 (pendidikan S1: -, S2: 36, S3: 18) Luas kampus: 5.645 m2 Fasilitas Kampus Ruang kuliah: luas......

Words: 2484 - Pages: 10

Free Essay

Matematik Rapport Statestik

...Matematikrapport - Statistik Louise, Nadia S, Cecilie M og Kathrine 1) Ikke-grupperet statistik Man bruger ikke-grupperet statistik når man har et mindre antal observationer. Her er det ikke nødvendigt at sætte observationer i intervaller, da det er mere præcist at have dem hver for sig, når man f.eks. skal regne middeltal. Vi har lavet et eksempel herunder, med udgangspunkt i højderne i 1.M. Her er det vist i et skema. Den øverste række viser de forskellige højder i cm og den nederste række viser hvor stort et antal elever der er indenfor hver højde. 158 | 161,5 | 162 | 163 | 164,5 | 165 | 168 | 169 | 172 | 173 | 174 | 174,5 | 175 | 180 | 181 | 2 | 1 | 1 | 2 | 1 | 3 | 2 | 3 | 2 | 1 | 1 | 1 | 1 | 1 | 1 | Herunder er der vist observationer fra en 3.g klasse. Det er stillet op på samme måde som i forrige skema. 160 | 164 | 165 | 165,5 | 166 | 167 | 168,4 | 169 | 173 | 175 | 176 | 177 | 179 | 180 | 183 | 185 | 187 | 188 | 193 | 1 | 5 | 1 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 | 2 | 1 | 1 | 2 | 1 | 2 | 1 | Prikdiagram Prikdiagrammet viser hyppigheden ved en enkel observation. F.eks. kan man ud fra vores vedlagte prikdiagram, hvor mange elever der eksempelvis er 165 cm høje. Ud fra de forskellige højder er der indsat en prik for hver enkel elev. Tilsammen skal der være det samme antal prikker, som der er elever. Her vises et prikdiagram over højder Ud fra vores første prikdiagram fra 1.m kan man se at der ligger mange inde for højderne [167;172] ...

Words: 1253 - Pages: 6

Free Essay


...sesuai untuk menghuraikan data (Statistik Deskriptif) • Memilih teknik yang paling sesuai untuk meninjau hubungan dan aliran dalam data (korelasi dan statistik inferensi) Dalam bab sebelum ini, perbezaan antara kajian secara kuantitatif dan kualitatif seringkali tidak jelas. Contohnya ialah kaedah tinjauan. Ia boleh memberi suatu penjelasan dan huraian yang tepat (kualitatif), tetapi ianya juga boleh dianalisis secara kuantitatif seperti kutipan data peribadi responden bagi sesuatu kelompok iaitu antara pengelasan responden dan sikap atau tabiatnya. Ciri asas analisis kuantitatif ialah dikatakan terlewat sekiranya menjalankan kajian, mengumpul data dan berfikir bagaimana untuk menganalisisnya. Proses untuk memilih ujian statistik perlu dibuat semasa peringkat perancangan kajian bukannya pada peringkat perlaksanaan. Hal ini kerana, ia lebih mudah untuk mendapatkan data dan ia tidak bermakna ujian statistik. Menurut Robson (1993), dengan bantuan komputer moden, ia lebih mudah untuk menghasilkan persembahan yang lebih baik. (GIGO – Garbage In, Garbage Out) (Robson, 1993:310). Tujuan utama bab ini adalah untuk memperkenalkan kepada anda mengenai teknik asas statistik. Ia tidak perlu menghuraikan secara mendalam bagi analisis statistik komplek kerana terdapat buku teks yang dikhaskan untuk tujuan ini. Ia dianggap seperti mengakses komputer dan perisian yang sesuai untuk mengaplikasi analisis statistik seperti Microsoft......

Words: 8172 - Pages: 33

Free Essay

Financial Statistics

...p˚ samband mellan variabler a c J¨rgen S¨ve-S¨derbergh o a o Finansiell statistik, v˚rterminen 2011 a Vilka data har vi? Antag tv˚ variabler X och Y som antar v¨rdena a a x1 , x2 , . . . , xn , samt y1 , y2 , . . . , yn . Om vi ritar dem i ett spridningsdiagram s˚ markerar vi de ordnade a paren (x1 , y1 ), (x2 , y2 ), . . . , (xn , yn ) i figuren. Vi uppfattar dessa par som ett stickprov. c J¨rgen S¨ve-S¨derbergh o a o Finansiell statistik, v˚rterminen 2011 a Korrelation Vi definierar stickprovskorrelationskoefficienten som r= Kovariansen mellan X och Y . (Standardavvikelsen f¨r X ) (Standardavvikelsen f¨r Y ) o o I t¨ljaren har vi den s˚ kallade stickprovskovariansen mellan X och a a Y . I n¨mnaren produkten av stickprovsstandardavvikelserna f¨r X a o och Y . c J¨rgen S¨ve-S¨derbergh o a o Finansiell statistik, v˚rterminen 2011 a Korrelation ¨ Overs¨tter vi denna definition i formler har vi a stickprovskorrelationskoefficienten som n i =1 (xi 1 n −1 r= 1 n−1 n i =1 (xi − x ) (yi − y ) − x )2 1 n−1 n i =1 (yi − y )2 Det ovanst˚ende st¨mmer v¨l med definitionen f¨r tv˚ a a a oa stokastiska variabler X och Y , vilken ¨r a Corr (X , Y ) = Cov (X , Y ) Var (X ) Var (Y ) . F¨r materialet ovan har vi o r = 0.44151. c J¨rgen S¨ve-S¨derbergh o a o Finansiell statistik, v˚rterminen 2011 a . Stickprov fr˚n en bivariat......

Words: 3762 - Pages: 16

Free Essay


...Tillämpad statistik – att samla och sammanfatta data Laboration 1: Deskriptiv statistik [pic] 2012 ©Martin Gellerstedt 0. Introduktion 2 1. Att beskriva nominaldata 4 1.1 Att beskriva nominaldata – univariat 4 1.2 Att beskriva nominaldata – bivariat 9 1.3 Att beskriva nominaldata – multivariat 13 2. Att beskriva ordinaldata 14 2.1 Att beskriva ordinaldata med få skalsteg 14 2.2 Att beskriva ordinaldata med många skalsteg - univariat 15 2.3 Att beskriva ordinaldata med många skalsteg - bivariat 18 2.4 Att beskriva ordinaldata med många skalsteg – multivariat 21 3. Att beskriva kvantitativa data 22 3.1 Att beskriva kvantitativa data – univariat 22 3.2 Att beskriva kvantitativa data – bivariat 23 3.3 Att beskriva kvantitativa data – multivariat 25 Introduktion Deskriptiv statistik handlar om att sammanfatta ett datamaterial på ett överskådligt vis - att karaktärisera materialet. För att sammanfatta ett material använder vi oss av statistiska sammanfattande mått, även kallat karaktäristikor. Frågan är dock, vad som menas med överskådligt? Vad är det vi vill karaktärisera? Jo, den egenskap vi först brukar intressera oss för är var materialet har sin kärna – sitt centrum. I statistiken talar vi om materialets läge (eng: location). För att beskriva ett datamaterials läge kan vi använda oss av olika lägesmått (karaktäristikor som beskriver läge). Ett känt lägesmått är det vanliga medelvärdet. Nästa egenskap......

Words: 889 - Pages: 4

Free Essay

Slange Gift

...Redegørelse for efterløn & Radikale venstre ifh til efterløn. Side 2 Analysering af debat indlæg. Side 3-4 Diskussion/konklussion. Side 5 Kildeliste/Litteratur liste: Debat indlæg: Debat indlæg skrevet af formanden for HK, Leo Jensen og udgivet d.2 september i 2011 på avisen jyllandspostens hjemmeside: Efterløn: tekst på 33 linjer som findes på side 117 i skole bogen ”Liv i Danmark” (Den nye), skrevet af Benny Jacobsen og Ove Outzen, og udgivet af forlaget Columbus, i 2011. Figur 8.1: Statistik over helbredsproblemer pga. Nedslidning, som findes på side 28 i statistik bogen ”Samfundsstatistik”, som er redigeret og skrevet af Allan christensen, Henrik Arbo-Bärh og Niels Knap, og udgivet i 2011 af Columbus. Figur 11.4: Statistik over fuldtidsledige, findes på side 48 i statistik bogen ”Samfundsstatistik”, som er redigeret og skrevet af Allan christensen, Henrik Arbo-Bärh og Niels Knap, og udgivet i 2011 af Columbus. Web 1: Indlæg skrevet af Jesper Boe Jensen og offentliggjort på Radikale venstres hjemmeside; Web 2: Artikel som forklarer efterlønnen og dets historik, udgivet på informationssiden wikipedia: Web 3: Artikel som redegører for efterlønsreformen i 2006, udgivet på informationssiden wikipedia: Web 4: Artikel som forklarer tilbagetrækningsreformen i 2011,...

Words: 2440 - Pages: 10

Free Essay

Case 1

... Björn Lantz Lär lätt! Statistik - Kompendium Studentia Lär lätt! Statistik - Kompendium © 2006 Björn Lantz och Studentia Ladda ner kompendiet gratis på ISBN 87-7681-080-1 Studentia Lär lätt! Statistik - Kompendium Innehållsförteckning Innehållsförteckning 1. 1.1 1.2 1.3 1.4 1.5 1.6 Introduktion till statistik Inledning Stolpdiagram och fördelning Centraltendens Spridning Skevhet Några exempel 6 6 7 9 10 11 13 2. 2.1 2.2 2.3 2.4 2.5 2.6 2.7 Sannolikhetslära Inledning Union och snitt Oberoende händelser Betingade sannolikheter Bayes teorem Permutationer Kombinationer 17 17 17 18 19 19 20 21 3. 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 Diskreta fördelingar Inledning Väntevärde och varians för en diskret slumpvariabel Binomialfördelningen Poissonfördelningen Hypergeometriska fördelningen Geometriska fördelningen Negativa binomialfördelningen Additions- och multiplikationsformler 23 23 23 25 26 28 29 29 30 4. 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 Kontinuerliga fördelningar Inledning Exponentialfördelningen Normalfördelningen Standardnormalfördelningen Transformering till standardnormalfördelning Transformering från standardnormalfördelning Normalfördelningsapproximation av binomialfördelningen Normalfördelningsapproximation av poissonfördelningen Fördelningen för ett......

Words: 34061 - Pages: 137

Free Essay


...bliver bedre balance i væksten af mennesker i forhold til hvor ustabilt det har været førhen. Hypotese Vores statistik ligner meget den oprindelige statistik i Danmark (som kan ses på bilag 1.) det eneste som virkelig skiller sig ud er det tertiære erhverv, hvor det stiger i Danmarks statistik omkring år 1970 og i vores klasses statistik falder indtil 2012. Ellers ser de meget ens ud. Udregninger i klassen Erhvervs udvikling [pic] S = Sekundære erhverv T = Tertiære erhverv P = Primære erhverv |Årstal |1920 |1945 |1975 |2012 | | | |I alt |% |I alt |% |I alt |% |I alt | |Primær |19 |73 % |17 |50 % |19 |53 % |4 | |Sekunder |5 |19 % |8 |24 % |8 |19 % |11 | |Tertiære |2 |8 % |9 |26 % |9 |28 % |15 | |I alt: |26 |100 % |34 |100 % |36 |100 % |34 | Ifølge klassens statistik var det i 20’erne mest primær erhverv, som både kan aflæses på grafen, men også på tabellen. Men som der......

Words: 584 - Pages: 3

Free Essay

Analisis Matematis Perubahan Kondisi Udara Ruang Bangunan Pertanian Pada Proses Evaporative Cooling Pada Berbagai Kondisi Udara Lingkungan

...Aktivasi (J/mol) dan Faktor Frekuensi Tumbukan ........... 92 1. Variasi Laju Kecepatan Udara .............................................................. 93 2. Variasi Laju Kecepatan Air ................................................................ ...95 D. Effektivitas Proses Evaporative Cooling ................................................ ...98 1. Lingkungan Udara Ruang Tertutup..................................................... ...98 2. Lingkungan Udara Ruang Dengan Pemanas ....................................... ...99 3. Lingkungan Udara Ruang Dengan Pendingin ..................................... 101 viii 4. Lingkungan Udara Ruangan Terbuka ................................................. 102 5. Hasil Analisis Statistik Nilai-Nilai Effisiensi ...................................... 104 6. Analisis Perbandingan Dari Keempat Macam Lingkungan ................. 104 E. Evaluasi Nilai Temperature Humidity Index (THI)................................. 106 1. Lingkungan Udara Ruang Tertutup..................................................... 107 2. Lingkungan Udara Ruang Dengan Pemanas ....................................... 108 3. Lingkungan Udara Ruang Dengan Pendingin ..................................... 110 4. Lingkungan Udara Ruangan Terbuka ................................................. 111 5. Perbandingan Dari Keempat Macam Lingkungan ............................... 113 BAB V PENUTUP A. Kesimpulan .................................................

Words: 22640 - Pages: 91

Free Essay

Fauzi Nax Tamboun

...BADAN PUSAT STATISTIK No. 35/05/Th. XVI, 6 Mei 2013 KEADAAN KETENAGAKERJAAN FEBRUARI 2013 FEBRUARI 2013: TINGKAT PENGANGGURAN TERBUKA SEBESAR 5,92 PERSEN   Jumlah penduduk yang bekerja di Indonesia pada Februari 2013 mencapai 114,0 juta orang, bertambah sebanyak 3,2 juta orang dibanding keadaan pada Agustus 2012 sebanyak 110,8 juta orang atau bertambah 1,2 juta orang dibanding keadaan Februari 2012.  Tingkat Pengangguran Terbuka (TPT) di Indonesia pada Februari 2013 mencapai 5,92 persen, mengalami penurunan dibanding TPT Agustus 2012 sebesar 6,14 persen dan TPT Februari 2012 sebesar 6,32 persen.  Selama setahun terakhir (Februari 2012–Februari 2013), jumlah penduduk yang bekerja mengalami kenaikan terutama di Sektor Perdagangan sebanyak 790 ribu orang (3,29 persen), Sektor Konstruksi sebanyak 790 ribu orang (12,95 persen), serta Sektor Industri sebanyak 570 ribu orang (4,01 persen). Sedangkan sektor-sektor yang mengalami penurunan adalah Sektor Pertanian dan Sektor Lainnya, masing-masing mengalami penurunan jumlah penduduk bekerja sebesar 3,01 persen dan 5,73 persen.  Berdasarkan jumlah jam kerja pada Februari 2013, sebanyak 78,3 juta orang (68,68 persen) bekerja di atas 35 jam per minggu, sedangkan penduduk bekerja dengan jumlah jam kerja kurang dari 15 jam per minggu mencapai 7,0 juta orang (6,17 persen).  1. Jumlah angkatan kerja di Indonesia pada Februari 2013 mencapai 121,2 juta orang, bertambah sebanyak 3,1 juta......

Words: 1104 - Pages: 5

Premium Essay


...dan boleh menggunakan pengetahuan yang diperolehi untuk penyelidikan. Pelajar akan didedahkan dengan statistik terutama pengunaan SPSS dalam analisa data kajian. OBJEKTIF: Pada akhir kursus ini calon semestinya berupaya: 1. Mengetahui penggunaan statistik di dalam bidang penyelidikan 2. Mengolah dan menyaji data. 3. Menguasai penggunaan SPSS dalam menganalisa data klinikal 4. Menggunakan kaedah mengikhtisarkan data yang sesuai. 5. Menggunakan ujian statistik yang sesuai untuk menganalisa data kualitatif dan kuantitatif. 6. Menggunakan ujian tak berparameter sesuai dengan ciri-ciri data yang diperolehi. CARA PENGAJARAN Kursus akan dijalankan secara kuliah, tutorials, tugasan dan praktikal di makmal komputer. ANGGOTA PENGAJAR: Penyelaras: Dr Mohd Rohaizat Hassan ( 012-6343303/ Pensyarah: Dr Mohd Rohaizat Hassan Dr Nazarudin Safian PENILAIAN Kursus akan dinilai secara penilaian berterusan dan peperiksaan akhir. Penilaian berterusan(40%) :1 = Ujian mini (20%) :1 = Computer based exam (20%) Peperiksaan akhir (60%) :Ujian bertulis (40%) :OBA (20%) RUJUKAN 1. Kirkwood B.R(1988) Essentials of Medical Statistics. Blackwell Scientific Publication. Oxford London 2. Sabin C. & Petrie A.(2005). Medical Statistics at a Glance. 2nd Ed. Blackwell Publishing USA 3. Md Idris M.N. Asas Statistik dan Penyelidikan Perubatan. Dewan Bahasa dan Pustaka 4. Dawson B& Trapp R.G (2004). Basic and......

Words: 454 - Pages: 2

Free Essay

Earning Managemnet During Relief Investigations

...yang melakukan sekuritisasi aset keuangan memiliki spread dan analyst forecast dispersion lebih tinggi dibandingkan non-securitizing bank. 3. Bagaimana menemukannya? Peneliti menemukannya dengan melakukan pengujian hipotesis dimana sampel bank holding company diuji secara statistik dengan menggunakan regresi. Peneliti menguji apakah kesulitan dalam memperkirakan transfer risiko, bank akan menghadapi ketidakpastian informasi, yang diproksikan dengan bid-ask spread dan analyst forecast dispersion. 4. Apa kekurangan jurnal tersebut? Menurut saya, jurnal ini, yang merupakan hasil penelitian dari Cheng, mempunyai beberapa kekurangan, antara lain: a. Kurangnya teori yang digunakan untuk menyusun hipotesis penelitian. Peneliti kebanyakan menggunakan hasil penelitian terdahulu untuk menyusun hipotesis, tidak terdapat teori utama yang dikemukakan, misalkan tentang asymetri information yang berkaitan dengan teori agensi. Ketidakpastian informasi berkaitan dengan sekuritisasi aset ini dikarenakan adanya asimetri informasi di antara pelaku atau peserta pasar modal. b. Dalam analisis dan pembahasan, peneliti kurang menjelaskan tentang hasil statistik. Kebanyakan hanya menjelaskan bahwa pengaruh antara variabel independen dan variabel dependen adalah signifikan dan hubungannya positif, tetapi tidak terlalu dijelaskan angka statistiknya dari mana. c. Peneliti menggunakan sampel bank holding company saja dalam penelitian ini,......

Words: 3226 - Pages: 13

Free Essay

Pattern Recogmition Program

...dan lain sebagainya. Teknologi ini memiliki peran penting dalam sistem keamanan data dan proses produksi. Secara umum teknologi pengenalan objek belum berkembang secara pesat di Indonesia sedangkan variasi penggunaannya semakin meningkat dan dibutuhkan dalam kehidupan sehari-hari. Pada studi kasus ini wajah manusia diambil sebagai objek yang akan dikenali dan diidentifikasi. Proses perancangan program dilakukan dengan menggunakan tiga metode yaitu eigenface, fisherface dan local binary pattern. Konsep dasar yang dikembangkan dalam algoritma dari ketiga metode tersebut ialah eigen vector. Dalam konsep ini, program pengenalan objek menggunakan titik-titik vektor sebagai pembanding di antara dua atau lebih objek tertentu sehingga melalui statistik didapat derajat kesamaan objek. Pada akhirnya, program yang dirancang diharapkan dapat memverifikasi apakah wajah yang hendak diidentifikasi merupakan sebuah wajah yang terdapat dalam database program. Kata kunci : Pengenalan Pola, Eigen Vector, Eigenface, Fisherface, LBP. iii ABSTRACT DESIGN OF PATTERN RECOGNITION PROGRAM WITH EIGEN VECTOR METHOD By AHMAD ABDULLAH NIM : 13208106 ELECTRICAL ENGINEERING STUDY PROGRAM Recognition and identification program for a specified object has already become a main technology in everyday lives which is used in a wide range of organizations such as government, business, industry, educational institutions, etc. This technology has an important role for designing a data security system...

Words: 6391 - Pages: 26