help: linear regression plot with confidence bands round line of fit

[Kornbrot, Diana]

help:
Need linear regression plot with confidence bands round line of fit
How does one get this in SPSS16?
Was on earlier version ­ but cannot do through Œlegacy¹ plots
Some options have been removed, e.g. Equation on the graph

I have 3 point plot of following points:
x -1.177397284 0.287328627 1.525008034
y -.615 .299 1.211
Adjusted r^2 = .995
Slope = .67 (.27, 1.08) se = .032
Intercept = .16 (-.30, .62) se = .036
95% confidence limits in ( ), based on t-test with 1 df, as only 3 points
I do NOT believe these CLs!
the line is beautiful,
by eye
CL as + or ­ 2*se looks sensible
CL as  + or ­ 12.7*se looks like pure fantasy, but this is recommended in
ALL my texts

Surely there must be a better approximation to CLs when r^2 is large and df
[number of points] is small?
The points themselves are VERY accurate, based on 100s [but that shouldn¹t
matter]


Best

Diana
Professor Diana Kornbrot
 School of Psychology
 University of Hertfordshire
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK

 email:  [hidden email]
 web:    http://web.mac.com/kornbrot/iweb/KornbrotHome.html
 voice:   [phone]
 fax:      [phone]
Home
 19 Elmhurst Avenue
 London N2 0LT, UK
   
    voice:   [phone]
    mobile: [phone]
    fax:      [phone]

====================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

[ViAnn Beadle]

The UI is still available except that the menu location of the UI has
changed. The UI is Graphics>Legacy Dialogs>Interactive>Scatterplot. When I
used that UI, it generated the following syntax using the Employee Data.sav
sample file:

IGRAPH
  /VIEWNAME='Scatterplot'
  /X1=VAR(salbegin) TYPE=SCALE
  /Y=VAR(salary) TYPE=SCALE
  /COORDINATE=VERTICAL
  /FITLINE METHOD=REGRESSION LINEAR INTERVAL(95.0)=MEAN LINE=TOTAL SPIKE=OFF

  /YLENGTH=5.2
  /X1LENGTH=6.5
  /CHARTLOOK='NONE'
  /SCATTER COINCIDENT=NONE.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
kornbrot
Sent: Saturday, May 31, 2008 8:39 AM
To: [hidden email]
Subject: help: linear regression plot with confidence bands round line of
fit

help:
Need linear regression plot with confidence bands round line of fit
How does one get this in SPSS16?
Was on earlier version ­ but cannot do through Œlegacy¹ plots
Some options have been removed, e.g. Equation on the graph

I have 3 point plot of following points:
x -1.177397284 0.287328627 1.525008034
y -.615 .299 1.211
Adjusted r^2 = .995
Slope = .67 (.27, 1.08) se = .032
Intercept = .16 (-.30, .62) se = .036
95% confidence limits in ( ), based on t-test with 1 df, as only 3 points
I do NOT believe these CLs!
the line is beautiful,
by eye
CL as + or ­ 2*se looks sensible
CL as  + or ­ 12.7*se looks like pure fantasy, but this is recommended in
ALL my texts

Surely there must be a better approximation to CLs when r^2 is large and df
[number of points] is small?
The points themselves are VERY accurate, based on 100s [but that shouldn¹t
matter]


Best

Diana
Professor Diana Kornbrot
 School of Psychology
 University of Hertfordshire
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK

 email:  [hidden email]
 web:    http://web.mac.com/kornbrot/iweb/KornbrotHome.html
 voice:   [phone]
 fax:      [phone]
Home
 19 Elmhurst Avenue
 London N2 0LT, UK

    voice:   [phone]
    mobile: [phone]
    fax:      [phone]

=======
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

[Kornbrot, Diana]

Thanks

Apologies, my fault.
Imported data from excel with 5 decimal points. SPSS cleverly assigned the
variables as nominal!
Only realized when I relloked at your syntax and compared with mine.

SPSS is VERY stupid when imprting from EXCEL on Mac. It also won’t let you
copy a column of data aND MAKE POOR DECISIONS for whole file

Thanks for your prompt helpful reply

diana

On 31/5/08 16:15, "ViAnn Beadle" <[hidden email]> wrote:

> The UI is still available except that the menu location of the UI has
> changed. The UI is Graphics>Legacy Dialogs>Interactive>Scatterplot. When I
> used that UI, it generated the following syntax using the Employee Data.sav
> sample file:
>
> IGRAPH
>   /VIEWNAME='Scatterplot'
>   /X1=VAR(salbegin) TYPE=SCALE
>   /Y=VAR(salary) TYPE=SCALE
>   /COORDINATE=VERTICAL
>   /FITLINE METHOD=REGRESSION LINEAR INTERVAL(95.0)=MEAN LINE=TOTAL SPIKE=OFF
>
>   /YLENGTH=5.2
>   /X1LENGTH=6.5
>   /CHARTLOOK='NONE'
>   /SCATTER COINCIDENT=NONE.
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
> kornbrot
> Sent: Saturday, May 31, 2008 8:39 AM
> To: [hidden email]
> Subject: help: linear regression plot with confidence bands round line of
> fit
>
> help:
> Need linear regression plot with confidence bands round line of fit
> How does one get this in SPSS16?
> Was on earlier version ­ but cannot do through Œlegacy¹ plots
> Some options have been removed, e.g. Equation on the graph
>
> I have 3 point plot of following points:
> x -1.177397284 0.287328627 1.525008034
> y -.615 .299 1.211
> Adjusted r^2 = .995
> Slope = .67 (.27, 1.08) se = .032
> Intercept = .16 (-.30, .62) se = .036
> 95% confidence limits in ( ), based on t-test with 1 df, as only 3 points
> I do NOT believe these CLs!
> the line is beautiful,
> by eye
> CL as + or ­ 2*se looks sensible
> CL as  + or ­ 12.7*se looks like pure fantasy, but this is recommended in
> ALL my texts
>
> Surely there must be a better approximation to CLs when r^2 is large and df
> [number of points] is small?
> The points themselves are VERY accurate, based on 100s [but that shouldn¹t
> matter]
>
>
> Best
>
> Diana
> Professor Diana Kornbrot
>  School of Psychology
>  University of Hertfordshire
>  College Lane, Hatfield, Hertfordshire AL10 9AB, UK
>
>  email:  [hidden email]
>  web:    http://web.mac.com/kornbrot/iweb/KornbrotHome.html
>  voice:   [phone]
>  fax:      [phone]
> Home
>  19 Elmhurst Avenue
>  London N2 0LT, UK
>
>     voice:   [phone]
>     mobile: [phone]
>     fax:      [phone]
>
> =======
> 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
>

Professor Diana Kornbrot
 School of Psychology
 University of Hertfordshire
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK

 email:  [hidden email]
 web:    http://web.mac.com/kornbrot/iweb/KornbrotHome.html
 voice:   [phone]
 fax:      [phone]
Home
 19 Elmhurst Avenue
 London N2 0LT, UK
   
    voice:   [phone]
    mobile: [phone]
    fax:      [phone]

====================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

[Kornbrot, Diana]

still very puzzled by large confidence boundaries round regression line,
that also appear on graph

>
> have 3 point plot of following points:
> x -1.177397284 0.287328627 1.525008034
> y -.615 .299 1.211
> Adjusted r^2 = .995
> Slope = .67 (.27, 1.08) se = .032
> Intercept = .16 (-.30, .62) se = .036
> 95% confidence limits in ( ), based on t-test with 1 df, as only 3 points
> I do NOT believe these CLs!
> the line is beautiful,
> by eye
> CL as + or ­ 2*se looks sensible
> CL as  + or ­ 12.7*se looks like pure fantasy, but this is recommended in
> ALL my texts
>
> Surely there must be a better approximation to CLs when r^2 is large and df
> [number of points] is small?
> The points themselves are VERY accurate, based on 100s [but that shouldn¹t
> matter]
>
>
> Best
>

Diana


On 31/5/08 16:15, "ViAnn Beadle" <[hidden email]> wrote:

> The UI is still available except that the menu location of the UI has
> changed. The UI is Graphics>Legacy Dialogs>Interactive>Scatterplot. When I
> used that UI, it generated the following syntax using the Employee Data.sav
> sample file:
>
> IGRAPH
>   /VIEWNAME='Scatterplot'
>   /X1=VAR(salbegin) TYPE=SCALE
>   /Y=VAR(salary) TYPE=SCALE
>   /COORDINATE=VERTICAL
>   /FITLINE METHOD=REGRESSION LINEAR INTERVAL(95.0)=MEAN LINE=TOTAL SPIKE=OFF
>
>   /YLENGTH=5.2
>   /X1LENGTH=6.5
>   /CHARTLOOK='NONE'
>   /SCATTER COINCIDENT=NONE.
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
> kornbrot
> Sent: Saturday, May 31, 2008 8:39 AM
> To: [hidden email]
> Subject: help: linear regression plot with confidence bands round line of
> fit
>
> help:
> Need linear regression plot with confidence bands round line of fit
> How does one get this in SPSS16?
> Was on earlier version ­ but cannot do through Œlegacy¹ plots
> Some options have been removed, e.g. Equation on the graph
>
> I have 3 point plot of following points:
> x -1.177397284 0.287328627 1.525008034
> y -.615 .299 1.211
> Adjusted r^2 = .995
> Slope = .67 (.27, 1.08) se = .032
> Intercept = .16 (-.30, .62) se = .036
> 95% confidence limits in ( ), based on t-test with 1 df, as only 3 points
> I do NOT believe these CLs!
> the line is beautiful,
> by eye
> CL as + or ­ 2*se looks sensible
> CL as  + or ­ 12.7*se looks like pure fantasy, but this is recommended in
> ALL my texts
>
> Surely there must be a better approximation to CLs when r^2 is large and df
> [number of points] is small?
> The points themselves are VERY accurate, based on 100s [but that shouldn¹t
> matter]
>
>
> Best
>
> Diana
> Professor Diana Kornbrot
>  School of Psychology
>  University of Hertfordshire
>  College Lane, Hatfield, Hertfordshire AL10 9AB, UK
>
>  email:  [hidden email]
>  web:    http://web.mac.com/kornbrot/iweb/KornbrotHome.html
>  voice:   [phone]
>  fax:      [phone]
> Home
>  19 Elmhurst Avenue
>  London N2 0LT, UK
>
>     voice:   [phone]
>     mobile: [phone]
>     fax:      [phone]
>
> =======
> 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
>

Professor Diana Kornbrot
 School of Psychology
 University of Hertfordshire
 College Lane, Hatfield, Hertfordshire AL10 9AB, UK

 email:  [hidden email]
 web:    http://web.mac.com/kornbrot/iweb/KornbrotHome.html
 voice:   [phone]
 fax:      [phone]
Home
 19 Elmhurst Avenue
 London N2 0LT, UK
   
    voice:   [phone]
    mobile: [phone]
    fax:      [phone]

====================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