What Statistical Analysis to use?

[wordcount]

Hi,
I'm doing a research on the interactive effect of each contingency of self-worth--there're 7 contingencies and the Contingency of Self-Worth Scale is ordinal--and academic performance on self-esteem. It's a within-sample, repeated-measures design. I gave the self-esteem scale (ordinal) to the students and their marks on a test were announced with their names and then the same self-esteem scale was given again to see the effect of academic performance on self-esteem. My hypotheses are:

1. Academic performance will have a significant effect on self-esteem of the students
2. Family Support CSW and Academic Performance will have significant interactive effect on Self-esteem of the students.
3. Competition CSW and Academic Performance will have significant interactive effect on Self-esteem of the students.
4. Appearance CSW and Academic Performance will have significant interactive effect on Self-esteem of the students.
5. God’s Love CSW and Academic Performance will have significant interactive effect on Self-esteem of the students.
6. Academic Competence CSW and Academic Performance will have significant interactive effect on Self-esteem of the students.
7. Virtue CSW and Academic Performance will have significant interactive effect on Self-esteem of the students.
8. Approval from Others CSW and Academic Performance will have significant interactive effect on Self-esteem of the students.

I just don't know what statistics to use in this case, and how. If someone kindly guide me about the name of the statistical analysis as well as a site to learn how to run that analysis, I'd be extremely grateful for this favour.
Thanks!
WordCount

[Art Kendall]

There are variations in what people mean when they say "ordinal".  What
is the response scale on CSW?
How do you measure self-esteem? is it both an independent and a
dependent variable?

How many cases do you have?

Please reply to the list so that others can make suggestions, etc.

Art Kendall
Social Research Consultants


On 6/9/2012 5:59 AM, wordcount wrote:

> Hi,
> I'm doing a research on the interactive effect of each contingency of
> self-worth--there're 7 contingencies and the Contingency of Self-Worth Scale
> is ordinal--and academic performance on self-esteem. It's a within-sample,
> repeated-measures design. I gave the self-esteem scale (ordinal) to the
> students and their marks on a test were announced with their names and then
> the same self-esteem scale was given again to see the effect of academic
> performance on self-esteem. My hypotheses are:
>
> 1.      Academic performance will have a significant effect on self-esteem of the
> students
> 2.      Family Support CSW and Academic Performance will have significant
> interactive effect on Self-esteem of the students.
> 3.      Competition CSW and Academic Performance will have significant
> interactive effect on Self-esteem of the students.
> 4.      Appearance CSW and Academic Performance will have significant interactive
> effect on Self-esteem of the students.
> 5.      God’s Love CSW and Academic Performance will have significant interactive
> effect on Self-esteem of the students.
> 6.      Academic Competence CSW and Academic Performance will have significant
> interactive effect on Self-esteem of the students.
> 7.      Virtue CSW and Academic Performance will have significant interactive
> effect on Self-esteem of the students.
> 8.      Approval from Others CSW and Academic Performance will have significant
> interactive effect on Self-esteem of the students.
>
> I just don't know what statistics to use in this case, and how. If someone
> kindly guide me about the name of the statistical analysis as well as a site
> to learn how to run that analysis, I'd be extremely grateful for this
> favour.
> Thanks!
> WordCount
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/What-Statistical-Analysis-to-use-tp5713606.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

[wordcount]

The CSW is a 7-point (ranging from Strongly Disagree to Strongly Agree) Lickert scale with 7 subscales (Family Support; Competition; Appearance; God's Love; Academic Competence; Virtue; Approval from Others) as given in the 7 hypotheses I've formed. This scale has 35 total items and each subscale has 5.

Self-esteem was measured by the Rosenberg's Self-Esteem Scale, which is a 4-point (ranging from Strongly Agree to Strongly Disagree) Lickert scale of 10 items. This is a dependent variable only, while Academic Performance as measured by the marks obtained by the students on a 5-item test, each carrying 5 marks (total 25 marks), and the contingencies of self-worth are independent variables.

The sample size is 37.                                              

[Art Kendall]

I assume that the scales are established measures of your variables

It is usually unnecessary to treat a summative scale as purely ordinal. Likert scales are usually considered _not severely discrepant_ from interval level.
Even single items are frequently treated as _not severely discrepant_ from interval level, but I presume you'll be using the summative scores.  I advise that you stick to more common approaches.

Self esteem could well be more of a trait variable than a state variable.  Don't be surprised if the pre-test accounts for a lot of the variance in the post test and does not leave much for other variables to account for.

I doubt that you have the power to detect interaction effects. I wouldn't be surprised if some of the contingencies added to the prediction, but why do you expect a interactions?

Also you are doing a lot of tests on a small amount of data, take a lot of salt with your results.

Center your CSW and achievement independent variables so you can create variables representing an interaction effect. (I.e., subtract the mean from the scores.)  Since the scaling is arbitrary this should not hinder interpretation.
Create 7 new variables representing interaction terms by multiplying the centered achievement variable by one of the centered CSW variables.

in REGRESSION use a stepped (aka a hierarchical, NOT a stepwise) approach. Much of what you would be looking for would be in the variables not yet in the equation.  Be sure to output the change in R**2.

For the first hypothesis the variables would be post test pre-test and achievement.
the post test self esteem would be the dependent variable.
enter the pre-test
read the output so see if achievement would significantly improve the prediction.  Remember that you are only finding out whether the improvement would be statistically distinguishable from no improvement.  This says nothing about whether the improvement is meaningful. 

For the other hypotheses the variables would be post test, pre-test, centered achievement, centered CSW variable, and the interaction term.
The dependent variable is the post-test self esteem
enter pre-test, centered achievement,centered CSW variable.
read the output so see if the interaction term  would significantly improve the prediction.  Remember that you are only finding out whether the improvement would be statistically distinguishable from no improvement.  This says nothing about whether the improvement is meaningful. 

HTH

On 6/9/2012 7:17 AM, wordcount wrote:

The CSW is a 7-point (ranging from Strongly Disagree to Strongly Agree)
Lickert scale with 7 subscales (Family Support; Competition; Appearance;
God's Love; Academic Competence; Virtue; Approval from Others) as given in
the 7 hypotheses I've formed. This scale has 35 total items and each
subscale has 5.

Self-esteem was measured by the Rosenberg's Self-Esteem Scale, which is a
4-point (ranging from Strongly Agree to Strongly Disagree) Lickert scale of
10 items. This is a dependent variable only, while Academic Performance as
measured by the marks obtained by the students on a 5-item test, each
carrying 5 marks (total 25 marks), and the contingencies of self-worth are
independent variables.

The sample size is 37.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/What-Statistical-Analysis-to-use-tp5713606p5713608.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

[James Whanger]

If you want to consider a number of options specifically designed for ordered categorical data see Agresti, A. (2010). 'Analysis of Ordinal Categorical Data'.  It is well written and clearly presents the arguments in favor of treating ordinal data as ordinal.

Best,

James

On Sat, Jun 9, 2012 at 5:59 AM, wordcount <[hidden email]> wrote:

Hi,
I'm doing a research on the interactive effect of each contingency of
self-worth--there're 7 contingencies and the Contingency of Self-Worth Scale
is ordinal--and academic performance on self-esteem. It's a within-sample,
repeated-measures design. I gave the self-esteem scale (ordinal) to the
students and their marks on a test were announced with their names and then
the same self-esteem scale was given again to see the effect of academic
performance on self-esteem. My hypotheses are:

1.      Academic performance will have a significant effect on self-esteem of the
students
2.      Family Support CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.
3.      Competition CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.
4.      Appearance CSW and Academic Performance will have significant interactive
effect on Self-esteem of the students.
5.      God’s Love CSW and Academic Performance will have significant interactive
effect on Self-esteem of the students.
6.      Academic Competence CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.
7.      Virtue CSW and Academic Performance will have significant interactive
effect on Self-esteem of the students.
8.      Approval from Others CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.

I just don't know what statistics to use in this case, and how. If someone
kindly guide me about the name of the statistical analysis as well as a site
to learn how to run that analysis, I'd be extremely grateful for this
favour.
Thanks!
WordCount

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/What-Statistical-Analysis-to-use-tp5713606.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

--
James C. Whanger

Research & Data Analytic Consultant

[Art Kendall]

P.S. 
You can compare your results to 2 other things and see if the conclusions are substantively different.
1) Try CATREG assuming the data is ordinal vs interval.
2) Generate a matrix of Spearman correlations and use that as input to the regression.



On 6/9/2012 7:17 AM, wordcount wrote:

The CSW is a 7-point (ranging from Strongly Disagree to Strongly Agree)
Lickert scale with 7 subscales (Family Support; Competition; Appearance;
God's Love; Academic Competence; Virtue; Approval from Others) as given in
the 7 hypotheses I've formed. This scale has 35 total items and each
subscale has 5.

Self-esteem was measured by the Rosenberg's Self-Esteem Scale, which is a
4-point (ranging from Strongly Agree to Strongly Disagree) Lickert scale of
10 items. This is a dependent variable only, while Academic Performance as
measured by the marks obtained by the students on a 5-item test, each
carrying 5 marks (total 25 marks), and the contingencies of self-worth are
independent variables.

The sample size is 37.

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/What-Statistical-Analysis-to-use-tp5713606p5713608.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

[Swank, Paul R]

Unless the self-esteem scale is badly distributed, I would use a GLM here. You have repeated measures on self-esteem. If your sample is large enough, you might want to check to see if the predictors work the same way for different levels of initial self-esteem by including interactions.

Sent from my iPad

On Jun 9, 2012, at 7:16 AM, "James C. Whanger" <[hidden email]<mailto:[hidden email]>> wrote:

If you want to consider a number of options specifically designed for ordered categorical data see Agresti, A. (2010). 'Analysis of Ordinal Categorical Data'.  It is well written and clearly presents the arguments in favor of treating ordinal data as ordinal.

Best,

James


On Sat, Jun 9, 2012 at 5:59 AM, wordcount <[hidden email]<mailto:[hidden email]>> wrote:
Hi,
I'm doing a research on the interactive effect of each contingency of
self-worth--there're 7 contingencies and the Contingency of Self-Worth Scale
is ordinal--and academic performance on self-esteem. It's a within-sample,
repeated-measures design. I gave the self-esteem scale (ordinal) to the
students and their marks on a test were announced with their names and then
the same self-esteem scale was given again to see the effect of academic
performance on self-esteem. My hypotheses are:

1.      Academic performance will have a significant effect on self-esteem of the
students
2.      Family Support CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.
3.      Competition CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.
4.      Appearance CSW and Academic Performance will have significant interactive
effect on Self-esteem of the students.
5.      God’s Love CSW and Academic Performance will have significant interactive
effect on Self-esteem of the students.
6.      Academic Competence CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.
7.      Virtue CSW and Academic Performance will have significant interactive
effect on Self-esteem of the students.
8.      Approval from Others CSW and Academic Performance will have significant
interactive effect on Self-esteem of the students.

I just don't know what statistics to use in this case, and how. If someone
kindly guide me about the name of the statistical analysis as well as a site
to learn how to run that analysis, I'd be extremely grateful for this
favour.
Thanks!
WordCount

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/What-Statistical-Analysis-to-use-tp5713606.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com<http://Nabble.com>.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email]<mailto:[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



--
James C. Whanger
Research & Data Analytic Consultant

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

[wordcount]

Sir Art Kendell!

I've no words to express in appreciation of your considerate reply, but count me novice when it comes to advanced statistics as suggested by you. Of course, Sir, the scales, both SES (Rosenberg, 1965) and CSWS (Crocker, Luhtanen, Cooper, & Bouvrette, 2003), are reliable and valid. The SES simply requires to add up scores for all 10 items and it's indeed a summative scale. The CSWS however requires to add first all the scores on each subscale and then divide by 5, which means the score range is 1-7.

I couldn't understand inerval scale. I also don't find this option under Measure in Variable View window (SPSS 17.0). Perhaps I should consider all variables as Scale only. Does it make any difference? Should I necessarily decide first whether the scale is nominal or ordinal? I understand Nominal, but what should I call the scores obtained on a Lickert type measure, Ordinal or Interval?

You have suggested Regression, but in my view ANOVA is the most suitable statistical analysis in this case. Only that I don't know how to run t-test for the first hypothesis because I don't want to see only the mean difference between pretest and posttest scores on SES. I also want to include Academic Performance, or scores obtained on tests given by the teacher, to see its effect on any change, significantly positive or significantly negative or insignificant, in SES scores. The question is how to do it. I can run t-test easily for two variables but how to include an independent variable as the third one in the analysis if t-test is only to use in this case?

In case of the 7 other hypotheses, you suggest that I should consider pretest SES scores,  centered CSWS scores, centered achievement test scores, new variables (obtained by multiplying centered CSW scores by centered achievement test scores as independent variables and the posttest SES scores as dependent variables? Should I use One-way or Two-way ANOVA if ANOVA is the right analysis in this case of finding interactive effect? Let me explain what I mean by interactive effect. Suppose five wrestlers win separate matches against opponents and two friends argue whether they won the matches because of their skills only or their weight, stamina, or height also played a role. To settle the dispute, an interactive effect of skills and weight, stamina and height each on the chances of winning the match was tested by analysis. In this case also, I want to see if academic performance alone causes a change in state self-esteem or, if combined with a particular contingency of self-worth this will affect more significantly on self-esteem of a student. Please guide me how to run analyses on SPSS for testing my 8 hypotheses, since this is the only software I'm familiar with.

I'm an undergraduate student, and this research is merely a part of courswork, and not an essential requirement for MA degree.

Thanks again for your considerate reply, Sir!

Respectfully,

Rashid

[wordcount]

Sir Art Kendell!

I've no words to express in appreciation of your considerate reply, but count me novice when it comes to advanced statistics as suggested by you. Of course, Sir, the scales, both SES (Rosenberg, 1965) and CSWS (Crocker, Luhtanen, Cooper, & Bouvrette, 2003), are reliable and valid. The SES simply requires to add up scores for all 10 items and it's indeed a summative scale. The CSWS however requires to add first all the scores on each subscale and then divide by 5, which means the score range is 1-7.

I couldn't understand inerval scale. I also don't find this option under Measure in Variable View window (SPSS 17.0). Perhaps I should consider all variables as Scale only. Does it make any difference? Should I necessarily decide first whether the scale is nominal or ordinal? I understand Nominal, but what should I call the scores obtained on a Lickert type measure, Ordinal or Interval?

You have suggested Regression, but in my view ANOVA is the most suitable statistical analysis in this case. Only that I don't know how to run t-test for the first hypothesis because I don't want to see only the mean difference between pretest and posttest scores on SES. I also want to include Academic Performance, or scores obtained on tests given by the teacher, to see its effect on any change, significantly positive or significantly negative or insignificant, in SES scores. The question is how to do it. I can run t-test easily for two variables but how to include an independent variable as the third one in the analysis if t-test is only to use in this case?

In case of the 7 other hypotheses, you suggest that I should consider pretest SES scores,  centered CSWS scores, centered achievement test scores, new variables (obtained by multiplying centered CSW scores by centered achievement test scores as independent variables and the posttest SES scores as dependent variables? Should I use One-way or Two-way ANOVA if ANOVA is the right analysis in this case of finding interactive effect? Let me explain what I mean by interactive effect. Suppose five wrestlers win separate matches against opponents and two friends argue whether they won the matches because of their skills only or their weight, stamina, or height also played a role. To settle the dispute, an interactive effect of skills and weight, stamina and height each on the chances of winning the match was tested by analysis. In this case also, I want to see if academic performance alone causes a change in state self-esteem or, if combined with a particular contingency of self-worth this will affect more significantly on self-esteem of a student. Please guide me how to run analyses on SPSS for testing my 8 hypotheses, since this is the only software I'm familiar with.

I'm an undergraduate student, and this research is merely a part of courswork, and not an essential requirement for MA degree.

Thanks again for your considerate reply, Sir!

Respectfully,

Rashid