Excel or not Excel, that's the question! [General Sta­tis­tics]

posted by Helmut Homepage – Vienna, Austria, 2008-08-20 00:01 (5727 d 14:46 ago) – Posting: # 2200
Views: 16,322

Dear HTMN!

❝ My center is a governmental one so it is not easy to buy something. We have asked for buying a software many times but not be approved.


I think that both Pharsight (WinNonlin) and Thermo (Kinetica) do have academic licenses and licenses for governmental bodies as well which are much cheaper than the regular licenses. Just ask them.

❝ […] we are using Excel to do statistic…


Yes, why not; I also use it to get a quick look. Nasty things start when validation comes into play – which is simply impossible to perform on Excel.
Furthermore, Micro$oft implemented some statistical functions in a rather weird way. For example if you want to get the inverse t-distribution (you need it in calculating the confidence interval) you have to multiply the α-value by two in order to get the right result. See this article containing a lot of useful links.

Since you already have R installed, see this post. Chapter 7 of Millard and Krause (2001) deals with the 2×2 design (available for download). The S+ code can easily (OK, almost easily) adapted for use in R. Then you even have something I’m still missing in all commercial programs: intra- and inter-subject residuals…

❝ […] You may feel funny but it is the fact of a developing country. (Sorry if I bother you too much with long story, just want you to be in sympathy if I have any too simple questions later)


I take you very seriously. When I started in 1980 in Austria, the situation was comparable. Just a few books about PK (not a single one on BE), regularily waiting two weeks for a lousy photocopy of some reference, no fax, the internet wasn’t invented yet, no software at all (not even spreadsheets, no text processors - standard was the IBM electric typewriter and corrective tape). It was the time of programing the TI-59; one of my friends at the Austrian Health Authority while struggling with his brand-new Commodore PET 4000 (1 MHz, 32 KB RAM) couldn't believe that the company I was with, could afford an HP 9836 (8 MHZ, 2 MB RAM, 2 floppy disc drives, BASIC, Pascal, HP-UX, a thermoprinter for the bargain of € 25,000.–)…
And yes, nobody to talk to (we founded the first independent CRO in Austria in 1984).

❝ […] please spend some time to answer my and others' questions - we are really waiting for your help! (I am really impressive with huge works done by you in this forum with the answers to almost all questions)


Not only me, there are quite a lot of others which are equally experienced. The compliment goes also to them!

❝ I will read carefully all the posts and documents you suggested and definitely, I will come up with plenty questions which need your help!


Yes, do so, but be prepared that sometimes it will take some time to get an answer.
I’m running the forum in my spare time – which is limited. ;-)

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

UA Flag
Activity
 Admin contact
22,993 posts in 4,828 threads, 1,655 registered users;
119 visitors (0 registered, 119 guests [including 1 identified bots]).
Forum time: 14:48 CEST (Europe/Vienna)

Never never never never use Excel.
Not even for calculation of arithmetic means.    Martin Wolfsegger

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5