Unequal sample size in AMOS

[Penny25]

Hi everyone, thank yo for reading my post I hope someone can help me.

I have a question regarding dealing with unequal sample sizes in AMOS. I want to do a path anlaysis and predict students data from parents data.

The predictor variables are datas from parents ((yes/no)dads= 130, Moms= 185; I look at them seperately but in the same model) and the outcome variable are students reports (n=300).

I need to keep all the data of the students. Is it possible to estimate the missing parents data? Or will this skew my calculations? How do I deal with this problem?

Thank you very much for helping me.

[David Marso]

Please reread your post and ask yourself whether you have provided sufficient information for anyone to even guess at what you are doing? Much or less provide any useful info to resolve your issue!

Penny25 wrote

Hi everyone, thank yo for reading my post I hope someone can help me.

I have a question regarding dealing with unequal sample sizes in AMOS. I want to do a path anlaysis and predict students data from parents data.

The predictor variables are datas from parents ((yes/no)dads= 130, Moms= 185; I look at them seperately but in the same model) and the outcome variable are students reports (n=300).

I need to keep all the data of the students. Is it possible to estimate the missing parents data? Or will this skew my calculations? How do I deal with this problem?

Thank you very much for helping me.

[Bruce Weaver]

I've never used AMOS, and don't have an answer for you.  I'm just trying help you clarify what I think you are asking.  Given the sample sizes you give below, you must have multilevel data, as follows:

Level 1:  students
Level 2:  Moms & Dads

Students are clustered within parents (i.e., there can be more than one student with the same parents), but for some students, the data for Mom or Dad (or both) may be missing, and you are asking what to do about those missing data points.  Right?

HTH.

Penny25 wrote

Hi everyone, thank yo for reading my post I hope someone can help me.

I have a question regarding dealing with unequal sample sizes in AMOS. I want to do a path anlaysis and predict students data from parents data.

The predictor variables are datas from parents ((yes/no)dads= 130, Moms= 185; I look at them seperately but in the same model) and the outcome variable are students reports (n=300).

I need to keep all the data of the students. Is it possible to estimate the missing parents data? Or will this skew my calculations? How do I deal with this problem?

Thank you very much for helping me.

[Penny25]

Hello thank you both for your answer. I am sorry if I didn`t make my problem clear.

@BruceWeaver: Thats what I`ve tried to say. Thank you.

[Ryan]

The data look like this, right?:

Student  Mother_Rating  Father_Rating  DV

1            4             3           2

2            2             5           1

3            3             4           3

.

.

.

This is not a multilevel model in the traditional sense unless some students are siblings.

Ryan

On Mon, Dec 9, 2013 at 11:09 AM, Bruce Weaver <[hidden email]> wrote:

I've never used AMOS, and don't have an answer for you.  I'm just trying help
you clarify what I *think* you are asking.  Given the sample sizes you give
below, you must have multilevel data, as follows:

Level 1:  students
Level 2:  Moms & Dads

Students are clustered within parents (i.e., there can be more than one
student with the same parents), but for some students, the data for Mom or
Dad (or both) may be missing, and you are asking what to do about those
missing data points.  Right?

HTH.



Penny25 wrote

> Hi everyone, thank yo for reading my post I hope someone can help me.
>
> I have a question regarding dealing with unequal sample sizes in AMOS. I
> want to do a path anlaysis and predict students data from parents data.
>
> The predictor variables are datas from parents ((yes/no)dads= 130, Moms=
> 185; I look at them seperately but in the same model) and the outcome
> variable are students reports (n=300).
>
> I need to keep all the data of the students. Is it possible to estimate
> the missing parents data? Or will this skew my calculations? How do I deal
> with this problem?
>
> Thank you very much for helping me.

-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Unequal-sample-size-in-AMOS-tp5723503p5723518.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

[Penny25]

Yes, the data look like this.

I want to do a path analysis. Parents answers as exogenous variables and students answers as endogenous variables. I could do a regression as well, but I would like to try it with AMOS.

I had the idea, to do the model one time with estimated missing values and one time with only the students which have father and mother datas and then compare the two models.

Is that a way to deal with it?

[Ryan]

This is yet another post that does not match the topic... 

It is possible to perform data imputation in AMOS. Use the drop down menu: Analyze-->Data Imputation--><type of imputation method>. Data should be missing at random (MAR) or missing completely at random (MCAR). 

You asked:

"I had the idea, to do the model one time with estimated missing values and
one time with only the students which have father and mother datas and then
compare the two models."

My question: Why did you have this idea?

Ryan

On Mon, Dec 9, 2013 at 11:54 AM, Penny25 <[hidden email]> wrote:

Yes, the data look like this.

I want to do a path analysis. Parents answers as exogenous variables and
students answers as endogenous variables. I could do a regression as well,
but I would like to try it with AMOS.

I had the idea, to do the model one time with estimated missing values and
one time with only the students which have father and mother datas and then
compare the two models.

Is that a way to deal with it?



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Unequal-sample-size-in-AMOS-tp5723503p5723532.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

[Bruce Weaver]

Sorry, I got the wrong end of the stick, apparently!

Ryan Black wrote

The data look like this, right?:

Student  Mother_Rating  Father_Rating  DV
1            4             3           2
2            2             5           1
3            3             4           3
.
.
.

This is not a multilevel model in the traditional sense unless some
students are siblings.

Ryan


On Mon, Dec 9, 2013 at 11:09 AM, Bruce Weaver <[hidden email]>wrote:

> I've never used AMOS, and don't have an answer for you.  I'm just trying
> help
> you clarify what I *think* you are asking.  Given the sample sizes you give
> below, you must have multilevel data, as follows:
>
> Level 1:  students
> Level 2:  Moms & Dads
>
> Students are clustered within parents (i.e., there can be more than one
> student with the same parents), but for some students, the data for Mom or
> Dad (or both) may be missing, and you are asking what to do about those
> missing data points.  Right?
>
> HTH.
>
>
>
> Penny25 wrote
> > Hi everyone, thank yo for reading my post I hope someone can help me.
> >
> > I have a question regarding dealing with unequal sample sizes in AMOS. I
> > want to do a path anlaysis and predict students data from parents data.
> >
> > The predictor variables are datas from parents ((yes/no)dads= 130, Moms=
> > 185; I look at them seperately but in the same model) and the outcome
> > variable are students reports (n=300).
> >
> > I need to keep all the data of the students. Is it possible to estimate
> > the missing parents data? Or will this skew my calculations? How do I
> deal
> > with this problem?
> >
> > Thank you very much for helping me.
>
>
>
>
>
> -----
> --
> Bruce Weaver
> [hidden email]
> http://sites.google.com/a/lakeheadu.ca/bweaver/
>
> "When all else fails, RTFM."
>
> NOTE: My Hotmail account is not monitored regularly.
> To send me an e-mail, please use the address shown above.
>
> --
> View this message in context:
> http://spssx-discussion.1045642.n5.nabble.com/Unequal-sample-size-in-AMOS-tp5723503p5723518.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
>