Dose-response association

[Hosam]

Dear All,
I hope you are doing well.
I plan to investigate the dose-response association between physical
activity (categorized) and quality of life (continuous, 0-100).
 
I would really appreciate it if you could help me with that.




--
Sent from: http://spssx-discussion.1045642.n5.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

[Art Kendall]

Please tell us the number of cases you have.

How were the cases selected?

Please paste the data definition for your variables.

Is this study exploratory or hypothesis testing?




-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.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

[Rich Ulrich]

That sounds like a rather simplistic homework question.

Fine tutorials still exist online.

If you are in the process of designing some study, you

probably need a book or a course on study design. 

Assigning levels of physical activity to get a "response"

exists for short-term studies. 

--

Rich Ulrich

---

From: SPSSX(r) Discussion <[hidden email]> on behalf of [hidden email] <[hidden email]>
Sent: Monday, March 15, 2021 9:33 AM
To: [hidden email] <[hidden email]>
Subject: Dose-response association

 

Dear All,
I hope you are doing well.
I plan to investigate the dose-response association between physical
activity (categorized) and quality of life (continuous, 0-100).
 
I would really appreciate it if you could help me with that.




--
Sent from: http://spssx-discussion.1045642.n5.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

[Hosam]

Hi Art,
I really appreciate your cooperation with me.
 The number of cases is a around 74,000 participants. It is a
cross-sectional national data which was collected randomly.

The objective of this exploratory cross-sectional study is to find the
dose-response association between physical activity (independent) and
quality of life (dependent).


Physical activity variable is as it shows below in minutes/week:
 .00
30.00
60.00
90.00
120.00
150.00
180.00
210.00
240.00
270.00
300.00
330.00
360.00
390.00
420.00
450.00
480.00
510.00
540.00
570.00
600.00
630.00


QOL is continuous and ranges from 0-100 (the higher score is better)







--
Sent from: http://spssx-discussion.1045642.n5.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

[Kornbrot, Diana]

A simple y v x should show relationship

Suggest you FIRST chose a scatter plot

THEN chose regression based on shape of scatter plot.

You may wish to look separately at different categories, e.g age, sex, occupation/lifestyle if you have thes in seeing data if you can make available via doiprbox ore google drive out similar

best

Diana

Would be interested

On 19 Mar 2021, at 16:40, [hidden email] wrote:

Hi Art,
I really appreciate your cooperation with me.
The number of cases is a around 74,000 participants. It is a
cross-sectional national data which was collected randomly.

The objective of this exploratory cross-sectional study is to find the
dose-response association between physical activity (independent) and
quality of life (dependent).


Physical activity variable is as it shows below in minutes/week:
.00
30.00
60.00
90.00
120.00
150.00
180.00
210.00
240.00
270.00
300.00
330.00
360.00
390.00
420.00
450.00
480.00
510.00
540.00
570.00
600.00
630.00


QOL is continuous and ranges from 0-100 (the higher score is better)







--
Sent from: http://spssx-discussion.1045642.n5.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

_____________________________

Professor Diana Kornbrot
Mobile [phone]
http://dianakornbrot.wordpress.com/   
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
[hidden email]               

===================== 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]

To begin, lose the extraneous decimals

FORMATS Activity (f3).

Do you have any other variables?
Age and gender are likely ones.

Start with scatter plots.
X= activity, Y = QoL.
Then in the output window, fit a straight line.
Then fit a loess line.
Eyeball the two lines.

Then scatterplot
X = activity Y = QoL  Z = Age.  Different shape markers for gender.  
Different colors for some other categorical variable.
In the output window, rotate the 3D scatterplot to see different
perspectives.

At the minute I do not recall whether you can fit lines/planes for 3D
scatterplots. but the options will be there on the output window to see if
you can.

This beginning will give you some ideas on what regressions to try.






-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.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

[Hosam]

Thanks Diana and Art for that

How could adjust for the confounding factors? Do u have the spss syntax?



--
Sent from: http://spssx-discussion.1045642.n5.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

[Hosam]

*Generate Data.
set seed 98765432.
new file.
input program.
loop ID = 1 to 100.
     compute ModVig_minsweek = rv.normal(0,1).
     compute b0 = 0.5.
     compute b1 = 1.0.
     compute b2 = -1.5.
     compute b3 = 2.0.
     compute error = rv.normal(0,1).
     compute HRQoL = b0 + b1*ModVig_minsweek + b2*ModVig_minsweek**2 +
b3*ModVig_minsweek**3 + error.
     end case.
   end loop.
end file.
end input program.
execute.

delete variables b0 b1 b2 b3 error.

GRAPH
  /SCATTERPLOT(BIVAR)=ModVig_minsweek WITH HRQoL
  /MISSING=LISTWISE.

compute ModVig_minsweek_2nd = ModVig_minsweek**2.
compute ModVig_minsweek_3rd = ModVig_minsweek**3.
execute.

REGRESSION
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT HRQoL
  /METHOD=ENTER ModVig_minsweek ModVig_minsweek_2nd ModVig_minsweek_3rd.

* Curve Estimation.
CURVEFIT
  /VARIABLES=HRQoL WITH ModVig_minsweek
  /CONSTANT
  /MODEL=CUBIC
  /PLOT FIT.



--
Sent from: http://spssx-discussion.1045642.n5.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