Re: tests of normality

[Zdaniuk, Bozena-3]

Oh, sorry, I forget that changing the subject line is no longer enough to start a new threat. Is it possible to do it using a regular email via outlook or do I need to go to the SPSSX-L website?
        Going back to my question, I am following the steps of computing a reference range for a blood component. The first step recommended is to remove the outliers. Next, to test for normality.
Thanks for your help, Art.
Bozena

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Art Kendall
Sent: April 10, 2015 2:12 PM
To: [hidden email]
Subject: Re: tests of normality

You might do well to start a new thread.

search the archives for this list wrt outliers and normality.
Rich ulrich and others have given some good advice.


Is the value that is suspicious weird when considering the construct under consideration?



-----
Art Kendall
Social Research Consultants
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Export-to-EXCEL-Align-pivot-tables-tp5729171p5729201.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Zdaniuk, Bozena-3]

Hello All,
        I have a sample that has a non-normal distribution (both Kolmogorov-Smirnov and Shapiro-Wilk tests highly significant, Skewness=2.8, Kurtosis=14.5). I remove one serious outlier and my Skewness and Kurtosis drop to between 1-2 and Kolmogorov-Smirnov becomes insignificant. But Shapiro-Wilk stays significant. Can I assume the distribution is close to normal?
Thanks so much,
Bozena

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of David Marso
Sent: April 8, 2015 10:19 AM
To: [hidden email]
Subject: Re: Export to EXCEL - Align pivot tables -

First off, I don't believe anyone is going to bother to run this under documented set of macros, attempt to replicate your export method and then come up with a shuffling idea.
1. No test data.
2. No sample macro call.
3. No example of Excel table.
Help others assist you by being proactive in providing an easily replicated context.
Meanwhile, perhaps explore OMS?
---

mils wrote

> Hi,
>
> I have the following tables and although I've managed to align the
> main table with the comparison (self test) table I would like to align
> "proportion" tables with "mean" tables and their respective comparison
> tables.
>
> I'm asking this because when I export all pivot tables to an excel
> document, the tables are not "align". Ideally, I would like to see
> (Variable Name | Label | Statistics|) for both "proportion" and "mean"
> tables if its possible.
>
> I hope it makes sense.
>
>
> DEFINE !screen1 (variables = !CMDEND)
> * Custom Tables.
> CTABLES
>   /VLABELS VARIABLES=!variables $Global  wec DISPLAY=LABEL
>   /TABLE !variables[C] [COUNT F40.0, COLPCT.COUNT PCT40.1] BY $Global
> [C]
> + wec [C]
>   /SLABELS POSITION=ROW VISIBLE=NO
>   /CATEGORIES VARIABLES=!variables  EMPTY=INCLUDE TOTAL=YES
> POSITION=BEFORE
>   /CATEGORIES VARIABLES=$Global wec  ORDER=A KEY=VALUE EMPTY=INCLUDE
>   /COMPARETEST TYPE=PROP ALPHA=0.05 ADJUST=NONE ORIGIN=COLUMN
> INCLUDEMRSETS=YES     CATEGORIES=ALLVISIBLE MERGE=NO.
> !ENDDEFINE.
>
> DEFINE !screen2 (variables = !CMDEND) CTABLES
>   /VLABELS VARIABLES=!variables $Global wec  DISPLAY=LABEL
>   /TABLE !variables [S][MEAN f40.1]
> BY  $Global [C] + wec [C]
>   /SLABELS POSITION=ROW VISIBLE=no
>   /CATEGORIES VARIABLES=$Global wec ORDER=A KEY=VALUE EMPTY=INCLUDE
>   /TITLES    TITLE='table1'
>   /COMPARETEST TYPE=MEAN ALPHA=0.05 ADJUST=NONE ORIGIN=COLUMN
> INCLUDEMRSETS=YES  CATEGORIES=ALLVISIBLE MEANSVARIANCE=ALLCATS MERGE=NO.
> !ENDDEFINE.
>
>
> Please, let me know if you need anything else.

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Export-to-EXCEL-Align-pivot-tables-tp5729171p5729172.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Art Kendall]

You might do well to start a new thread.

search the archives for this list wrt outliers and normality.
Rich ulrich and others have given some good advice.


Is the value that is suspicious weird when considering the construct under consideration?