Please explain what are the columns represent. Best Regards, Donald McLaren ================= D.G. McLaren, Ph.D. Research Fellow, Department of Neurology, Massachusetts General Hospital and Harvard Medical School Postdoctoral Research Fellow, GRECC, Bedford VA Website: http://www.martinos.org/~mclaren Office: (773) 406-2464 ===================== This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is intended only for the use of the individual or entity named above. If the reader of the e-mail is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient, you are hereby notified that you are in possession of confidential and privileged information. Any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited and may be unlawful. If you have received this e-mail unintentionally, please immediately notify the sender via telephone at (773) 406-2464 or email.
On Wed, Nov 21, 2012 at 2:50 PM, <xiangbo_2...@126.com> wrote: > No, I want to compute the interaction between factor 1 and factor 2 , > gender and age as covariates. thanks > > Bo Xiang > > 在 2012-11-22 04:22:06，"Douglas N Greve" <gr...@nmr.mgh.harvard.edu> 写道： >>Hi Bo, I don't understand the contrast you are trying to make. Are you >>really trying to compute the interaction between four variables (factor >>1, factor 2, gender, and age)? >>doug >> >>On 11/21/2012 10:49 AM, xiangbo_2010 wrote: >>> Dear Donald McLaren >>> Thank you for your reply! I make the contrast according to your method >>> is following，but I want to make interaction between factor 1 (A,B)and >>> factor 2(C,D,E), gender (M,F) and one continuous variable (age) as >>> covariates, the following contrast: >>> 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> 0 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>> is correct? Thanks! >>> Bo Xiang >>> >>> >>> >>> >>> >>> At 2012-11-21 04:07:37,"MCLAREN, Donald"<mclaren.don...@gmail.com >>> <mailto:mclaren.don...@gmail.com>> wrote: >>> >On Tue, Nov 20, 2012 at 12:56 PM, Douglas N Greve >>> ><gr...@nmr.mgh.harvard.edu <mailto:gr...@nmr.mgh.harvard.edu>> wrote: >>> >> Thanks Donald. Is this the standard way to do this? I had used 8 rows >>> >> instead of 4 with the difference being that 8 rows gives you an >>> >> opportunity >>> >> to look for an effect in males OR females. >>> > >>> >Yes. Having 8 rows would tell you if you have an interaction between >>> >factor 1 and 2 in either males or females. My 4 rows only tell you if >>> >the interaction exists. Technically speaking, one would run the >>> >three-way interaction first. If nothing existed then you do the >>> >two-way interaction as I suggested. If there is a three-way >>> >interaction, then you would use Doug's approach of the interaction in >>> >either males or females. >>> > >>> >If there is an effect in both >>> >> males and females but the effects go in opposite directions, then the 4 >>> >> row >>> >> implementation will resolve to 0 (no effect). Or am I misunderstanding >>> >> something (again:)? >>> > >>> >Nope. You are right. If the male and female effects are different, >>> >then they could cancel each other out. If you suspect this to be the >>> >case, then you should be able to demonstrate a three-way interaction. >>> > >>> >> thanks! >>> >> doug >>> >> >>> >> >>> >> On 11/20/2012 01:50 PM, MCLAREN, Donald wrote: >>> >>> >>> >>> Bo, >>> >>> >>> >>> Doug asked me to chime in on your issue. Here are some points that you >>> >>> (and others) will hopefully find useful. >>> >>> >>> >>> (1) Inferences are two-step process. First, you create and estimate >>> >>> the design matrix. Every column in the design matrix accounts can >>> >>> account for some of the variance in the data. Second, you have >>> >>> contrasts that allow you to infer specific effects. Because the model >>> >>> contains your covariates, you are always controlling for the >>> >>> covariates and by extension any factor/covariate not in the contrast. >>> >>> >>> >>> (2) Forming contrasts is often the most difficult thing to do. I >>> >>> assume that your three factors (1, 2, and gender) are all >>> >>> between-subject factors. If one of them is a within-subject factor >>> >>> please let me know and disregard the rest of the email. The final >>> >>> F-contrast will have 4 rows (factor 1 levels-1)*(factor 2 levels >>> >>> -1)=(3-1)*(3-1)=2*2=4 >>> >>> >>> >>> The following is an outline for creating contrasts: >>> >>> (a) Start simple - difference between levels of 1 factor >>> >>> (b) Define your null hypothesis: AO=AP=AQ >>> >>> (c) Make it equal to 0: AO-AP=0 AND AP-AQ=0 >>> >>> (d) Repeat for the other levels of the factor... >>> >>> BO-BP=0 AND BP-BQ=0 >>> >>> CO-CP=0 AND CP-CQ=0 >>> >>> >>> >>> (e) Now combine them AO-AP=BO-BP=CO-CP AND AP-AQ=BP-BQ=CP-CQ >>> >>> >>> >>> (f) Make them equal to 0: >>> >>> AO-AP-BO+BP=0 >>> >>> BO-BP-CO+CP=0 >>> >>> AP-AQ-BP+BQ=0 >>> >>> BP-BQ-CP+CQ=0 >>> >>> >>> >>> (g) Expand them to include gender, for example: >>> >>> AO-AP-BO+BP=0 becomes FAO-FAP-FBO+FBP+MAO-MAP-MBO+MBP=0 >>> >>> >>> >>> Since the contrast now has 2 columns per level, you should divide all >>> >>> values by 2. This will produce the correct amplitude and statistics. >>> >>> If you leave the values as 1 and -1, then you will have an incorrect >>> >>> amplitude, but the statistics will still be correct. >>> >>> >>> >>> (h) Fill in the respective columns of your design matrix. >>> >>> >>> >>> (3) The degrees of freedom are defined based on the rows of the >>> >>> F-matrix and the number of rows in the design matrix. The F-test has a >>> >>> numerator and denominator degrees of freedom. F(n,d). >>> >>> >>> >>> Best Regards, Donald McLaren >>> >>> ================= >>> >>> D.G. McLaren, Ph.D. >>> >>> Research Fellow, Department of Neurology, Massachusetts General >>> >>> Hospital >>> >>> and >>> >>> Harvard Medical School >>> >>> Postdoctoral Research Fellow, GRECC, Bedford VA >>> >>> Website: http://www.martinos.org/~mclaren >>> >>> Office: (773) 406-2464 >>> >>> ===================== >>> >>> This e-mail contains CONFIDENTIAL INFORMATION which may contain >>> >>> PROTECTED >>> >>> HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is >>> >>> intended only for the use of the individual or entity named above. If >>> >>> the >>> >>> reader of the e-mail is not the intended recipient or the employee or >>> >>> agent >>> >>> responsible for delivering it to the intended recipient, you are hereby >>> >>> notified that you are in possession of confidential and privileged >>> >>> information. Any unauthorized use, disclosure, copying or the taking of >>> >>> any >>> >>> action in reliance on the contents of this information is strictly >>> >>> prohibited and may be unlawful. If you have received this e-mail >>> >>> unintentionally, please immediately notify the sender via telephone at >>> >>> (773) >>> >>> 406-2464 or email. >>> >>> >>> >>> >>> >>> On Tue, Nov 20, 2012 at 8:17 AM, xiangbo_2010<xiangbo_2...@126.com >>> >>> <mailto:xiangbo_2...@126.com>> >>> >>> wrote: >>> >>>> >>> >>>> Dear doug >>> >>>> Thank you for your reply! >>> >>>> +AOM -BOM -APM -BPM 0 0 0 0 0 >>> >>>> +AOM -BOM 0 0 -AQM +BQM 0 0 0 >>> >>>> +AOM 0 -APM 0 0 0 -COM +CPM 0 >>> >>>> +AOM 0 0 0 -AQM -COM 0 +CQM >>> >>>> there should be use 1 -1 or 0.5 -0.5? whether the -BPM should be >>> >>>> change >>> >>>> BPM? >>> >>>> Thanks >>> >>>> >>> >>>> Bo Xiang >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> At 2012-11-19 07:23:31,"Douglas Greve"<gr...@nmr.mgh.harvard.edu >>> >>>> <mailto:gr...@nmr.mgh.harvard.edu>> wrote: >>> >>>> >>> >>>> >>> >>>> Hi Bo, you can think of the Ftest as a logical 'OR' between the >>> >>>> t-test >>> >>>> contrasts indicated in each row. Each row is a difference of >>> >>>> differences, >>> >>>> so >>> >>>> >>> >>>> 1. (A-B)om - (A-B)pm --> Does the difference between A and B differ >>> >>>> between >>> >>>> groups O and P for Males? >>> >>>> 2. (A-B)om - (A-B)qm >>> >>>> 3. (A-C)om - (A-C)pm >>> >>>> 4. (A-C)om - (A-C)qm >>> >>>> 5. (A-B)of - (A-B)pf --> Does the difference between A and B differ >>> >>>> between >>> >>>> groups O and P for Females? >>> >>>> 6. (A-B)of - (A-B)qf >>> >>>> 7. (A-C)of - (A-C)pf >>> >>>> 8. (A-C)of - (A-C)pf >>> >>>> >>> >>>> I've put together the first 9 columns of the first 4 rows. The last 9 >>> >>>> columns are all 0s. For the last for rows, the 0s and below matrix are >>> >>>> swapped to give you the same for the females >>> >>>> >>> >>>> doug >>> >>>> >>> >>>> AOM BOM APM BPM AQM BQM COM CPM CQM >>> >>>> >>> >>>> >>> >>>> -------------------------------------------------------------------------- >>> >>>> +AOM -BOM -APM -BPM 0 0 0 0 0 >>> >>>> +AOM -BOM 0 0 -AQM +BQM 0 0 0 >>> >>>> +AOM 0 -APM 0 0 0 -COM +CPM 0&! nbsp;&nb sp; >>> >>>> >>> >>>> +AOM 0 0 0 -AQM -COM 0 +CQM >>> >>>> >>> >>>> >>> >>>> >>> >>>> On 11/17/12 9:21 PM, xiangbo_2010 wrote: >>> >>>> >>> >>>> Hi Freesurfer experts, >>> >>>> >>> >>>> I'm very sorry to bother you, but I am very confused with the >>> >>>> following >>> >>>> questions: >>> >>>> >>> >>>> My experimental design includes three discrete factors: factor 1 with >>> >>>> three >>> >>>> levels (A,B,C ); factor 2 with three levels (O,P,Q); gender (F, M), >>> >>>> and >>> >>>> one >>> >>>> covariate. >>> >>>> >>> >>>> So I can get 18 classes: FAO, FAP,FAQ,FBO,FBP,FBQ,FCO,FCP,FCQ,MAO, >>> >>>> MAP,MAQ,MBO,MBP,MBQ,MCO,MCP,MCQ. I want to perform the interaction >>> >>>> between >>> >>>> factor 1 and factor 2 regressing out the effect of gender and one >>> >>>> covariate, >>> >>>> but I don't know the rules for setting the contrasts for the F-test. >>> >>>> The >>> >>>> contrast matrix I used is: >>> >>>> >>> >>>> 1 1 1 -1 -1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> 1 -1 0 1 -1 0 1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> 1 1 1 0 0 0 -1 -1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> 1 0 -1 1 0 -1 1 0 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> 0 0 0 0 0 0 0 0 0 1 1 1 -1 -1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> 0 0 0 0 0 0 0 0 0 1 -1 0 1 -1 0 1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 -1 -1 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> 0 0 0 0 0 0 0 0 0 1 0 -1 1 0 -1 1 0 -1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 >>> >>>> 0 0 >>> >>>> 0 >>> >>>> >>> >>>> is it correct? >>> >>>> >>> >>>> >>> >>>> >>> >>>> Any help will be very appreciated. >>> >>>> >>> >>>> >>> >>>> >>> >>>> Best wishes, >>> >>>> >>> >>>> >>> >>>> >>> >>>> Bo Xiang >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> _______________________________________________ >>> >>>> Freesurfer mailing list >>> >>>> Freesurfer@nmr.mgh.harvard.edu >>> >>>> <mailto:Freesurfer@nmr.mgh.harvard.edu> >>> >>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> _______________________________________________ >>> >>>> Freesurfer mailing list >>> >>>> Freesurfer@nmr.mgh.harvard.edu >>> >>>> <mailto:Freesurfer@nmr.mgh.harvard.edu> >>> >>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>> >>>> >>> >>>> >>> >>>> The information in this e-mail is intended only for the person to >>> >>>> whom it >>> >>>> is >>> >>>> addressed. If you believe this e-mail was sent to you in error and the >>> >>>> e-mail >>> >>>> contains patient information, please contact the Partners Compliance >>> >>>> HelpLine at >>> >>>> http://www.partners.org/complianceline . If the e-mail was sent to >>> >>>> you in >>> >>>> error >>> >>>> but does not contain patient information, please contact the sender >>> >>>> and >>> >>>> properly >>> >>>> dispose of the e-mail. >>> >>>> >>> >>> >>> >> >>> >> -- >>> >> Douglas N. Greve, Ph.D. >>> >> MGH-NMR Center >>> >> gr...@nmr.mgh.harvard.edu <mailto:gr...@nmr.mgh.harvard.edu> >>> >> Phone Number: 617-724-2358 >>> >> Fax: 617-726-7422 >>> >> >>> >> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting >>> >> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html >>> >> Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ >>> >> >>> >_______________________________________________ >>> >Freesurfer mailing list >>> >Freesurfer@nmr.mgh.harvard.edu <mailto:Freesurfer@nmr.mgh.harvard.edu> >>> >https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >>> >>> >>> >>> >>> >>> >>> >>> >>> _______________________________________________ >>> Freesurfer mailing list >>> Freesurfer@nmr.mgh.harvard.edu >>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> >>-- >>Douglas N. Greve, Ph.D. >>MGH-NMR Center >>gr...@nmr.mgh.harvard.edu >>Phone Number: 617-724-2358 >>Fax: 617-726-7422 >> >>Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting >>FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html >>Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ >> >>_______________________________________________ >>Freesurfer mailing list >>Freesurfer@nmr.mgh.harvard.edu >>https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
