minitab: an introduction
back in 1992, i had two little books published by kendall/hunt ... 1. MINITAB: An Introduction 2. MINITAB: An Introduction for Business this was back when release 8 was around and commands were the thing of the day table of contents looked liked (for #1): 1. overview to minitab 2. computer requirements 3. structure of minitab 8 4. using release 8 of minitab 5. getting help within minitab 6. entering and editing data 7. saving data and work 8. making a hard copy of your work 9. making new columns of data from old columns 10. making graphs, plots, and tables 11. general purpose statistics using minitab's data files 12. generating data at random 13. using minitabs macros 14. additional exercises 15. minitab commands/index clearly, some updating is called for but, it just so happens that i have the disk that this manual is on ... and, it would be a rather simple thing to do to do some editing ... AND POST IT TO MY SERVER ... MAYBE, CHAPTER BY CHAPTER would any of you find this helpful ... for yourself and/or for your students? if so ... what suggestions would you make as to "for sure" things to put in it ... ?? i will do a quick cut and paste ... and put part on my webpage ... just to give some idea of what could be (though lots of editing will be needed of course) thanks for your assistance _ dennis roberts, educational psychology, penn state university 208 cedar, AC 8148632401, mailto:[EMAIL PROTECTED] http://roberts.ed.psu.edu/users/droberts/drober~1.htm = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: spss polynomial contrast estimates
On Sun, 28 Jan 2001 13:57:02 -0500, "K. Bloom" [EMAIL PROTECTED] wrote: I just ran some data through a simple one way polynomial analysis in spss. I had 3 groups and very unequal n's. By hand, I calculated the contrast estimates for both the linear and quadratic trends using the unweighted formula for determining coefficients. The value I got was different than the value of contrast estimates I got from the spss printout. So I tried to compute new coefficients weighted for sample sizes. But when I did that I got bizarre linear and quadratic contrast estimates. Help. Here are my data for the three groups... snip, detail of problem, possible solutions ... perhaps someone knows that formula that spss uses to calculate the polynomial contrast estimates. I've had trouble navigating the SPSS site, so I finally bookmarked the algorithms. SPSS documents this one at - http://www.spss.com/tech/stat/algorithms/oneway.pdf It appears that "group" codes are used as codes for spacing. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Margin Analysis Qstn
On Mon, 29 Jan 2001 06:28:23 GMT, "Chris" [EMAIL PROTECTED] wrote: snip Assuming a normal distribution, what method should I use to calculate my averages? Should I simply take the sample mean? - I think your problem is that you *don't* want to assume a normal distribution. For 'normal', once you tell the mean and SD, you have said everything that is interesting about the single variable. Should I remove anomalies like 112% margins? Should I calculate upper and lower control limits and place my data into a normal curve? From what you described, 112% is virtually impossible. It means (as I read it) that someone paid you to take the product that you then sold. Are you in a re-cycling business? Does that still leave "margin" as a useful amount? You say you have a lot of data. You can present a lot of detail by categorizing "margin" into bins; then make tables with Products-by- bins. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Excel Graphics
The absolute best advice concerning the use of Excel for graphics (or for statistics for that matter) is: DON'T! The _majority_ of graph-types available in Excel should never be used for any purpose as they produce misleading graphs -- mainly false third dimensions that can only serve to hide important features in the graph.) Jon Cryer At 02:26 PM 1/27/01 GMT, you wrote: Not sure if this is the best place to ask, but can anyone point me towards any web sites that provide advice on using Excel for technical/scientific graphing. I am not sure why exactly, but I find the graphs produced by Excel, compared to S-Plus or Statistica, to look out of place in a technical report. As I know others feel the same way, I was hoping that there might be some advice out there on how to improve their appearance. Many thanks, Graham S . = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ = = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
p values
in an article ... that some might be able to access ... http://bmj.com/cgi/content/full/322/7280/226 by Jonathan A C Sterne, senior lecturer in medical statistics, George Davey Smith, professor of clinical epidemiology. Department of Social Medicine, University of Bristol, Bristol BS8 2PR one of the summary points made is the following: "P values, or significance levels, measure the strength of the evidence against the null hypothesis; the smaller the P value, the stronger the evidence against the null hypothesis" my main questions of this are: 1. does the general statistical community accept this as being correct? 2. if the answer to #1 is yes ... then what does this tell us (only this p value) about what the real parameter value is? (are) _ dennis roberts, educational psychology, penn state university 208 cedar, AC 8148632401, mailto:[EMAIL PROTECTED] http://roberts.ed.psu.edu/users/droberts/drober~1.htm = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Aggregating Data
Magill, Brett wrote: Just a practical problem that I have run into here. I have a large data set where every case (row) represents a person. Each person belongs to a metropolitan area. I want to aggregate some of the individual results into metro area statistics. Should be easy... but alas... Not an answer to your Q, but there are problems with this kind of aggregation if you want to use the aggregate analysis to make inferences about the people. Notably that the aggregates effects might be different to effects at the level of individuals. If this might be a problem for you may need to go to different statistical model (and probably a different package). Thom = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
SPSS Macro for computing Gini coefficient of inequality
Hello to all. Recently I needed to compute the Gini coefficient in SPSS. When I found, that there is no possibility to get it via DESCRIPTIVES or FREQ procedure I tried to write a macro which would compute it for me. Unfortunately I don't have much experience in macro writing, I don't know how to do it. I used the equation from Sen's "On Economic Inequality", where G is expressed as a function of a weighted sum of all personal incomes with ordinal weights (1 for the richest, 2 for the second richest etc). I didn't know how to compute this sum in SPSS macro and than use it in further operations. I have also visited WWW.SPSS.COM, but such a macro was anavailable. Is anyone of you able to give me some brief advise how to cope with the problem Thank you in advance Michal Bojanowski = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Excel Graphics
Jon, The absolute best advice concerning the use of Excel for graphics (or for statistics for that matter) is: DON'T! The _majority_ of graph-types available in Excel should never be used for any purpose as they produce misleading graphs -- mainly false third dimensions that can only serve to hide important features in the graph.) But Excel CAN produce simple scatter plots or bar charts. It is just that the defaults are so horrible. With a lot of tweaking you can make them acceptable. Having said that I do remember an article on the Origin web site which suggested even simple charts in Excel were wrong. My problem is cost. I want to get everyone in my department to have the facility to produce reasonable charts that have a common style. I do all of mine in Statistica (a single copy) but I cannot afford 15 copies simply to produce the occasional scatter plot, or box plot. Nor does anyone have the time to produce everyone elses charts for them. Three Excel addins have come my way, during this search. XLstat, Analyse-IT and Winstat. All produce boxplots, and Winstat seems the most generally useful, but for some reason the Winstat chart defaults are useless for scientific/technical use. None are entirely satisfactory, and I am still looking for the best solution. Cheers, Graham S . = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: SPSS Macro for computing Gini coefficient of inequality
I haven't used the Gini coefficient in the last 25 years, so I can't give more complete advice. However, from your description, you can can get such a sum without a macro by RANK VARIABLE= income (d) /rank into r_income. * to get the rank for each case. WEIGHT BY r_income. * to turn on weighting. DESCRIPTIVES VARIABLES= r_income /statistics=sum. Isn't the graph that the GINI coefficient summarizes something like X= proportion of people with this income or lower Y= proportion of dollars to people with this income or lower? p.s. check what you want to do with ties. MichalB wrote: Hello to all. Recently I needed to compute the Gini coefficient in SPSS. When I found, that there is no possibility to get it via DESCRIPTIVES or FREQ procedure I tried to write a macro which would compute it for me. Unfortunately I don't have much experience in macro writing, I don't know how to do it. I used the equation from Sen's "On Economic Inequality", where G is expressed as a function of a weighted sum of all personal incomes with ordinal weights (1 for the richest, 2 for the second richest etc). I didn't know how to compute this sum in SPSS macro and than use it in further operations. I have also visited WWW.SPSS.COM, but such a macro was anavailable. Is anyone of you able to give me some brief advise how to cope with the problem Thank you in advance Michal Bojanowski = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
FW: t critical: beta (Power ala David Murray) (fwd)
I'm posting this message on behalf of my friend. Any help would be greatly appreciated. -- Forwarded message -- Date: Fri, 26 Jan 2001 04:15:49 -0800 From: Renita Glaser [EMAIL PROTECTED] Subject: t critical: beta (Power ala David Murray) I am doing a power analysis using the formulas outlined in David Murray's book, "Design and Analysis of Group-Randomized Trials." In going through example 1 in his chapter on power analysis, I am puzzled about where his estimate of tcritical:beta comes from. I understand that you can obtain the t critical: alpha from a t-distribution table but I am at a loss for how to obtain t critical:beta. Once I get this value, I can calculate the detectable difference with 80% power! Any help would be greatly appreciated... Renita = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Excel Graphics
Then, what is the use of EXCEL? Siddeek Jon Cryer wrote: The absolute best advice concerning the use of Excel for graphics (or for statistics for that matter) is: DON'T! The _majority_ of graph-types available in Excel should never be used for any purpose as they produce misleading graphs -- mainly false third dimensions that can only serve to hide important features in the graph.) Jon Cryer At 02:26 PM 1/27/01 GMT, you wrote: Not sure if this is the best place to ask, but can anyone point me towards any web sites that provide advice on using Excel for technical/scientific graphing. I am not sure why exactly, but I find the graphs produced by Excel, compared to S-Plus or Statistica, to look out of place in a technical report. As I know others feel the same way, I was hoping that there might be some advice out there on how to improve their appearance. Many thanks, Graham S . = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ = = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ = = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: p values
This is a multi-part message in MIME format. --152074D07C4F46BEF5496A61 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit One should note that this comment was taken from an article critical of significance testing. I understand Sterne and Smith's comment to characterize a point of view which they do not support, but which they project onto consumers of statistics. (I think they could have been a bit clearer about their own point of view, but this is how I read the statement in the context of what follows.) Generally, statisticians with a background in theory of inference will agree with the general thrust of the article: Significance tests are often performed and reported and interpreted in a mechanical way which may be misleading. The suggested guidelines (which suggest against use of the phrase "statistically significant" without further qualification) will generate considerable support among those who have studied the theories of frequentist and Bayesian inference. The only support for the significance test is in the use of the P-level as an index to the strength of the evidence. Personally, I would qualify that with the standard "Other things being equal...". --152074D07C4F46BEF5496A61 Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --152074D07C4F46BEF5496A61-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Excel Graphics
if you can use excel to do what you need doing and, it does not lead to any serious (tweaking?) complications fine but, what happens if you want one of your colleagues to do something a little different ... like make a x,y plot ... where you have both male and female data points separately indicated ... ? the graphics are highly limited ... that's the main problem my suggestion is the following: 1. invest in a decent stat package for 1 or 2 colleagues ... who would be willing to make charts and graphs for others now and then ... after all, in minitab for example ... you can save these as jpg files and post to some web location ... 2. you can get all to have access to something like minitab ... by each downloading the 30 day demo ... and if you stagger the time of download ... have someone have a full version spread across the group ... for a fairly long time ... so each understand the basics of the package 3. another possibility would be to have one dedicated machine ... with minitab (for example) on it ... then think about using timbuktu or even netmeeting ... to take over control of the machine from time to time to run some stuff 4. ... finally, if everyone has excel if they can make the excel file ... then the 1/2 people with the full package CAN easily import that xls file and do stuff i guess my view is that there is some better compromise way to handle this ... some middle position that may not get the full package for all (if that is too much money) but ... to still have some have the full package ... and others have access to it from time to time ... At 07:49 PM 1/29/01 +, [EMAIL PROTECTED] wrote: Jon, The absolute best advice concerning the use of Excel for graphics (or for statistics for that matter) is: DON'T! The _majority_ of graph-types available in Excel should never be used for any purpose as they produce misleading graphs -- mainly false third dimensions that can only serve to hide important features in the graph.) But Excel CAN produce simple scatter plots or bar charts. It is just that the defaults are so horrible. With a lot of tweaking you can make them acceptable. Having said that I do remember an article on the Origin web site which suggested even simple charts in Excel were wrong. My problem is cost. I want to get everyone in my department to have the facility to produce reasonable charts that have a common style. I do all of mine in Statistica (a single copy) but I cannot afford 15 copies simply to produce the occasional scatter plot, or box plot. Nor does anyone have the time to produce everyone elses charts for them. Three Excel addins have come my way, during this search. XLstat, Analyse-IT and Winstat. All produce boxplots, and Winstat seems the most generally useful, but for some reason the Winstat chart defaults are useless for scientific/technical use. None are entirely satisfactory, and I am still looking for the best solution. Cheers, Graham S . = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ = _ dennis roberts, educational psychology, penn state university 208 cedar, AC 8148632401, mailto:[EMAIL PROTECTED] http://roberts.ed.psu.edu/users/droberts/drober~1.htm = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: FW: t critical: beta (Power ala David Murray) (fwd)
let's see ... say you have two overlapping distributions ... on the left, the null distribution ... and on the right, the assumed to be true sampling distribution ... now, in setting up the power problem ... you have to set alpha ... say that is a two tailed .05 test ... and really the upper critical value is of most interest ... so, that value might be like a t of 2.02 ... to the right of the middle of the null distribution ... now, depending on what you are allowing power to be ... then, beta is 1 - power ... so, say that power is desired to be .7 ... the beta will be .3 ... that means on the true sampling distribution ... TO THE LEFT OF THE CENTER OF THAT TRUE DISTRIBUTION ... beta will be the area FURTHER TO THE LEFT ... or, in this case, the lower 30% of the true (distribution on the right) sampling distribution ... that is ... what is about the 97.5 percentile rank for the critical t value at the upper end of the null distribution ... will be at the 30 percentile rank value in the (hoped for) assumed to be true sampling distribution ... would that not simply be the t value ... that produces the 30th percentile rank? let's assume for arguments sake n = 40 so, we have df = 39 ... in a single sample kind of case ... the critical value for the null (right side of the null distribution) would be in minitab MTB invcdf .975; SUBC t 39. Inverse Cumulative Distribution Function Student's t distribution with 39 DF P( X = x ) x 0.97502.0227 critical value for null test MTB invcdf .3; SUBC t 39. Inverse Cumulative Distribution Function Student's t distribution with 39 DF P( X = x ) x 0.3000 -0.5287 critical value (where to the left will be beta of about .3) it helps to try to draw out two overlapping distributions ... fill in what you know or want ... and then see about solving the problem you have to solve MT 'm posting this message on behalf of my friend. Any help would be greatly appreciated. = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Excel Graphics
i think, if i might be allowed to speak for jon, he would say that it is a good SPREADSHEET package ... that was what it was designed for ... add ons to make it also a statistical package ... are a different matter it is sort of like having ms word ... where it has this add on feature of letting you input some numbers into a page ... and then it will crank out some graphs for you ... or some descriptive statistics ... we would have to be a bit careful At 11:53 AM 1/29/01 -0900, Shareef Siddeek wrote: Then, what is the use of EXCEL? Siddeek we could ask this in another way ... why do people use excel ... when there are a number of other spreadsheet programs out there ... quattro, etc. i would suspect that over time, people have found excel to be most accurate? useful for doing the kinds of spreadsheet things IT WAS DESIGNED for? easier files to be able to export to or import from other things, etc we might also say ... well, it now is also a stat package ... so that makes it even more popular ... but, the argument that if you have excel, then the stat package add on makes it real cheap ... and since so many have excel ... why not then just make it their stat package too ... ??? you don't do this for the same reason that if you want to do professional measurement analysis, you select something specifically designed for that ... and tested over time ... or why you don't select notepad for doing word processing (though it might be ok for some things) ... you make people know word or something comparable the notion that we should try to have one package do all ... seems silly of course, we have observed in many graduate programs that want students to learn about research ... that we find that many advisors are trying harder and harder to get students into that ONE course that does all ... unfortunately, one course does not do all just like excel ... = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: p values
i see two places in the sterne and smith article, which is in my hand ... that say hat p is an indication of the p of the null not being true ... or evidence against the null on page 2 of 10 ... they list summary points and, the one i quoted before ... p values, or significance levels .. measure the strength of the evidence against the null ... on page 7 of 10 ... we suggest that journal editors require that authors of research reports follow the guidelines in box 2 ... and in box 2 ... #3, when there is a meaningful null hypothesis, the strength of evidence against it should be indexed by the P value ... i am certainly NOT implying that this is all they have to say about this topic ... just that it is ONE thing they say ... twice ... so, that is what i was questioning the general statistical community about At 05:06 PM 1/29/01 -0400, Richard A. Beldin wrote: This is a multi-part message in MIME format. --152074D07C4F46BEF5496A61 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit One should note that this comment was taken from an article critical of significance testing. I understand Sterne and Smith's comment to characterize a point of view which they do not support, but which they project onto consumers of statistics. (I think they could have been a bit clearer about their own point of view, but this is how I read the statement in the context of what follows.) = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
RE: Excel Graphics
Another package that produces nice plain charts in Excel is XLStatistics. A free version can be downloaded from http://www.man.deakin.edu.au/rodneyc/xlstats.htm Rodney ~~ Rodney Carr School of Management Information Systems Deakin University PO Box 423 Warrnambool VIC 3280 Australia email: [EMAIL PROTECTED] phone: + 61 3 5563 3458 mobile: 0417 307 692 fax: + 61 3 5563 3320 www: http://www.man.deakin.edu.au/rodneyc -Original Message- From: [EMAIL PROTECTED] [SMTP:[EMAIL PROTECTED]] Sent: Tuesday, January 30, 2001 5:49 AM To: [EMAIL PROTECTED] Subject:Re: Excel Graphics Jon, The absolute best advice concerning the use of Excel for graphics (or for statistics for that matter) is: DON'T! The _majority_ of graph-types available in Excel should never be used for any purpose as they produce misleading graphs -- mainly false third dimensions that can only serve to hide important features in the graph.) But Excel CAN produce simple scatter plots or bar charts. It is just that the defaults are so horrible. With a lot of tweaking you can make them acceptable. Having said that I do remember an article on the Origin web site which suggested even simple charts in Excel were wrong. My problem is cost. I want to get everyone in my department to have the facility to produce reasonable charts that have a common style. I do all of mine in Statistica (a single copy) but I cannot afford 15 copies simply to produce the occasional scatter plot, or box plot. Nor does anyone have the time to produce everyone elses charts for them. Three Excel addins have come my way, during this search. XLstat, Analyse-IT and Winstat. All produce boxplots, and Winstat seems the most generally useful, but for some reason the Winstat chart defaults are useless for scientific/technical use. None are entirely satisfactory, and I am still looking for the best solution. Cheers, Graham S . = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ = = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Excel Graphics
Those specific papers referred to in an earlier discussion have addressed errors in Statistics and other add-in (e.g., Solver) calculations in EXCEL. If I remember correct, the errors occur from the 7 th decimal place and may aggravate if repeated calculations (as in simulations) are made. I do not know why Microsoft is shy to fix these problems by hiring Statistical and programming experts. This remark applies to the currently discussed graphic problems as well. I personally like EXCEL to be smart enough to do basic Statistics, linear and non linear model fitting and graphing in the right way. This is because EXCEL is more popular and easy to use. Cheers. Siddeek David Heiser wrote: - Original Message - From: "Shareef Siddeek" [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Monday, January 29, 2001 12:53 PM Subject: Re: Excel Graphics Then, what is the use of EXCEL? Siddeek Comes as standard boilerplate with Microsoft Office. The package includes Word, Excel, PowerPoint, Outlook and others depending on cost. Most computer systems use Windows and Office software. It gets you up and running. The graphics issue is cost. 700 to 8000 dollars for one of these graphical packages for one user. The more glitz, the higher the cost. Package is obsolete in a year or two and won't properly work with the up and coming 64 bit systems. Figure rebuying every year. Package is obsolete also because better display techniques are developed over time, bugs get fixed and interfaces get smoothed over. Vendor makes his $ by overcharging. Software development is expensive, especially for newly immerging companies. = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Statistical Quality Control,Grant Leavnworth
Statistical Quality Control,Grant Leavnworth 7th edition For sale on eBay see item 1407155701 = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
