How do I use SPSS to obtain parameter estimates for growth trajectories over time?

[Jacquee Blaz]

Hello,

I need to generate parameter estimates for a growth curve for a
continuous level variable. I have subjects that were measured on that
variable over 7 timepoints on 2 different days. I'm ultimately
interested in the period covered by the 7 timepoints. I've fit models
in Stata in the past, but I only had the one time variable. My models
were conceptualized like:

Outcome = Beta*_0 + Beta*_1xTime + epsilon;
Beta*_0 = Beta_0 + gamma_0
Beta*_1 = Beta_1 + gamma_1

I'm having a difficult time conceptualizing how to model the two
different levels of time, but mainly how to get SPSS to run the
analysis. I can't seem to find what I'm looking for in the online
documentation

Any advice would be appreciated.

Thanks, in advance,

Jacquee Blaz, MS
College of Nursing
University of Utah

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

[Richard Ristow]

At 09:40 PM 2/23/2009, Jacquee Blaz wrote:

I need to generate parameter estimates for a growth curve for a continuous level variable. I have subjects that were measured on that variable over 7 timepoints on 2 different days. I'm ultimately interested in the period covered by the 7 timepoints.

This is a simple linear regression, of the dependent variable against time.

First question: How is 'time' represented in your dataset? SPSS date-time variables don't make good independent variables; they're in seconds, and their origin is in 1586. Make your 'time' variable the time from some convenient start point (typically, either the beginning or the middle of your experimental run), in whatever units you find convenient (hours, days, minutes, ...). I'm not giving details now, since your 'time' variable may not be SPSS date-time in the first place.

Second, this is probably easiest to run from the menus. Select

Analyze > Regression > Linear

Select the variable you're measuring (I've called it 'Response') as dependent, and your time variable (I've called it 'Time') as independent.

Under "Statistics..." (at the bottom of the window), I'd add "Descriptives" to the defaults, and otherwise leave them as they are. And click "OK", in the upper-right corner of the window. Or, change options, etc., according to your needs.

Menus commands work by generating and running syntax. Here's the syntax from the choices above, and the output:

 
REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .

 
Regression
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:29:20       |
|-----------------------------|---------------------------|
[DataSet0]
 
Descriptive Statistics
|--------|-------|--------------|-|
|        |Mean   |Std. Deviation|N|
|--------|-------|--------------|-|
|Response|10.3333|4.50185       |6|
|--------|-------|--------------|-|
|Time    |12.67  |11.165        |6|
|--------|-------|--------------|-|
 
Correlations
|-----------|--------|--------|-----|
|           |        |Response|Time |
|-----------|--------|--------|-----|
|Pearson    |Response|1.000   |.763 |
|Correlation|--------|--------|-----|
|           |Time    |.763    |1.000|
|-----------|--------|--------|-----|
|Sig.       |Response|.       |.039 |
|(1-tailed) |--------|--------|-----|
|           |Time    |.039    |.    |
|-----------|--------|--------|-----|
|N          |Response|6       |6    |
|           |--------|--------|-----|
|           |Time    |6       |6    |
|-----------|--------|--------|-----|
 
Variables Entered/Removed(b)
|-----|---------------|---------------|------|
|Model|Variables      |Variables      |Method|
|     |Entered        |Removed        |      |
|-----|---------------|---------------|------|
|1    |Time(a)        |.              |Enter |
|-----|---------------|---------------|------|
a All requested variables entered.
b Dependent Variable: Response
 
Model Summary
|-----|-------|--------|---------------|---------------|
|Model|R      |R Square|Adjusted R     |Std. Error of  |
|     |       |        |Square         |the Estimate   |
|-----|-------|--------|---------------|---------------|
|1    |.763(a)|.582    |.477           |3.25570        |
|-----|-------|--------|---------------|---------------|
a Predictors: (Constant), Time
 
ANOVA(b)
|-----|----------|--------------|--|-----------|-----|-------|
|Model|          |Sum of Squares|df|Mean Square|F    |Sig.   |
|-----|----------|--------------|--|-----------|-----|-------|
|1    |Regression|58.935        |1 |58.935     |5.560|.078(a)|
|     |----------|--------------|--|-----------|-----|-------|
|     |Residual  |42.398        |4 |10.600     |     |       |
|     |----------|--------------|--|-----------|-----|-------|
|     |Total     |101.333       |5 |           |     |       |
|-----|----------|--------------|--|-----------|-----|-------|
a Predictors: (Constant), Time
b Dependent Variable: Response
 
Coefficients(a)
|-----|----------|--------------------------|---------------|-----|----|
|Model|          |Unstandardized            |Standardized   |t    |Sig.|
|     |          |Coefficients              |Coefficients   |     |    |
|     |          |---------------|----------|---------------|     |    |
|     |          |B              |Std. Error|Beta           |     |    |
|-----|----------|---------------|----------|---------------|-----|----|
|1    |(Constant)|6.439          |2.120     |               |3.037|.039|
|     |----------|---------------|----------|---------------|-----|----|
|     |Time      |.307           |.130      |.763           |2.358|.078|
|-----|----------|---------------|----------|---------------|-----|----|
a Dependent Variable: Response


=============================
APPENDIX: Test data, and code
=============================
(Test data hand-entered. This is the listing)

List
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:27:13       |
|-----------------------------|---------------------------|
[DataSet0]
 
    Time Response

       1     4.00
       3     9.00
       4    10.00
      20     8.00
      22    15.00
      26    16.00


Number of cases read:  6    Number of cases listed:  6



REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .


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

[Hector Maletta]

It is a simple case of linear regression –provided the underlying growth process was linear, i.e. equal increments per unit of time. Otherwise, other models are needed. For instance, if the RATE of increment was constant, rather than the absolute increment, then a linear regression of log(variable) would be in order to capture the exponential growth. And the growth process may have any other form, including perhaps a transitory growth spurt and then a stall period, or whatever. In a more general way, the question may require kernel regression to capture the general form of the growth curve, as is ordinarily applied in other growth processes (e.g. growth of height of children at ages 0 to 5 or 0 to 15). Of course, the “subjects” to which the question refers may be anything, and the growth in question covers a very short time, and is thus likely to be well approximated by a linear regression, but I wanted to make the general point.

 

Hector

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Richard Ristow
Sent: 26 February 2009 14:34
To: [hidden email]
Subject: Re: How do I use SPSS to obtain parameter estimates for growth trajectories over time?

 

At 09:40 PM 2/23/2009, Jacquee Blaz wrote:

I need to generate parameter estimates for a growth curve for a continuous level variable. I have subjects that were measured on that variable over 7 timepoints on 2 different days. I'm ultimately interested in the period covered by the 7 timepoints.

This is a simple linear regression, of the dependent variable against time.

First question: How is 'time' represented in your dataset? SPSS date-time variables don't make good independent variables; they're in seconds, and their origin is in 1586. Make your 'time' variable the time from some convenient start point (typically, either the beginning or the middle of your experimental run), in whatever units you find convenient (hours, days, minutes, ...). I'm not giving details now, since your 'time' variable may not be SPSS date-time in the first place.

Second, this is probably easiest to run from the menus. Select

Analyze > Regression > Linear

Select the variable you're measuring (I've called it 'Response') as dependent, and your time variable (I've called it 'Time') as independent.

Under "Statistics..." (at the bottom of the window), I'd add "Descriptives" to the defaults, and otherwise leave them as they are. And click "OK", in the upper-right corner of the window. Or, change options, etc., according to your needs.

Menus commands work by generating and running syntax. Here's the syntax from the choices above, and the output:

 
REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .

 
Regression
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:29:20       |
|-----------------------------|---------------------------|
[DataSet0]
 
Descriptive Statistics
|--------|-------|--------------|-|
|        |Mean   |Std. Deviation|N|
|--------|-------|--------------|-|
|Response|10.3333|4.50185       |6|
|--------|-------|--------------|-|
|Time    |12.67  |11.165        |6|
|--------|-------|--------------|-|
 
Correlations
|-----------|--------|--------|-----|
|           |        |Response|Time |
|-----------|--------|--------|-----|
|Pearson    |Response|1.000   |.763 |
|Correlation|--------|--------|-----|
|           |Time    |.763    |1.000|
|-----------|--------|--------|-----|
|Sig.       |Response|.       |.039 |
|(1-tailed) |--------|--------|-----|
|           |Time    |.039    |.    |
|-----------|--------|--------|-----|
|N          |Response|6       |6    |
|           |--------|--------|-----|
|           |Time    |6       |6    |
|-----------|--------|--------|-----|
 
Variables Entered/Removed(b)
|-----|---------------|---------------|------|
|Model|Variables      |Variables      |Method|
|     |Entered        |Removed        |      |
|-----|---------------|---------------|------|
|1    |Time(a)        |.              |Enter |
|-----|---------------|---------------|------|
a All requested variables entered.
b Dependent Variable: Response
 
Model Summary
|-----|-------|--------|---------------|---------------|
|Model|R      |R Square|Adjusted R     |Std. Error of  |
|     |       |        |Square         |the Estimate   |
|-----|-------|--------|---------------|---------------|
|1    |.763(a)|.582    |.477           |3.25570        |
|-----|-------|--------|---------------|---------------|
a Predictors: (Constant), Time
 
ANOVA(b)
|-----|----------|--------------|--|-----------|-----|-------|
|Model|          |Sum of Squares|df|Mean Square|F    |Sig.   |
|-----|----------|--------------|--|-----------|-----|-------|
|1    |Regression|58.935        |1 |58.935     |5.560|.078(a)|
|     |----------|--------------|--|-----------|-----|-------|
|     |Residual  |42.398        |4 |10.600     |     |       |
|     |----------|--------------|--|-----------|-----|-------|
|     |Total     |101.333       |5 |           |     |       |
|-----|----------|--------------|--|-----------|-----|-------|
a Predictors: (Constant), Time
b Dependent Variable: Response
 
Coefficients(a)
|-----|----------|--------------------------|---------------|-----|----|
|Model|          |Unstandardized            |Standardized   |t    |Sig.|
|     |          |Coefficients              |Coefficients   |     |    |
|     |          |---------------|----------|---------------|     |    |
|     |          |B              |Std. Error|Beta           |     |    |
|-----|----------|---------------|----------|---------------|-----|----|
|1    |(Constant)|6.439          |2.120     |               |3.037|.039|
|     |----------|---------------|----------|---------------|-----|----|
|     |Time      |.307           |.130      |.763           |2.358|.078|
|-----|----------|---------------|----------|---------------|-----|----|
a Dependent Variable: Response


=============================
APPENDIX: Test data, and code
=============================
(Test data hand-entered. This is the listing)

List
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:27:13       |
|-----------------------------|---------------------------|
[DataSet0]
 
    Time Response

       1     4.00
       3     9.00
       4    10.00
      20     8.00
      22    15.00
      26    16.00


Number of cases read:  6    Number of cases listed:  6



REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .

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

[peter link]

If you have a sample of subjects with repeated measurements across time, then you should not use linear regression.  You now have nested data and should use the MIXED procedure in SPSS or some other mixed model software (ex., HLM).  An assumption of linear regression is uncorrelated observations.  Due to having subjects mesured several times, each subject will have observations that will be correlated over time.

 

Peter Link

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of Hector Maletta
Sent: Thursday, February 26, 2009 8:45 AM
To: [hidden email]
Subject: Re: How do I use SPSS to obtain parameter estimates for growth trajectories over time?

It is a simple case of linear regression –provided the underlying growth process was linear, i.e. equal increments per unit of time. Otherwise, other models are needed. For instance, if the RATE of increment was constant, rather than the absolute increment, then a linear regression of log(variable) would be in order to capture the exponential growth. And the growth process may have any other form, including perhaps a transitory growth spurt and then a stall period, or whatever. In a more general way, the question may require kernel regression to capture the general form of the growth curve, as is ordinarily applied in other growth processes (e.g. growth of height of children at ages 0 to 5 or 0 to 15). Of course, the “subjects” to which the question refers may be anything, and the growth in question covers a very short time, and is thus likely to be well approximated by a linear regression, but I wanted to make the general point.

 

Hector

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Richard Ristow
Sent: 26 February 2009 14:34
To: [hidden email]
Subject: Re: How do I use SPSS to obtain parameter estimates for growth trajectories over time?

 

At 09:40 PM 2/23/2009, Jacquee Blaz wrote:

I need to generate parameter estimates for a growth curve for a continuous level variable. I have subjects that were measured on that variable over 7 timepoints on 2 different days. I'm ultimately interested in the period covered by the 7 timepoints.

This is a simple linear regression, of the dependent variable against time.

First question: How is 'time' represented in your dataset? SPSS date-time variables don't make good independent variables; they're in seconds, and their origin is in 1586. Make your 'time' variable the time from some convenient start point (typically, either the beginning or the middle of your experimental run), in whatever units you find convenient (hours, days, minutes, ...). I'm not giving details now, since your 'time' variable may not be SPSS date-time in the first place.

Second, this is probably easiest to run from the menus. Select

Analyze > Regression > Linear

Select the variable you're measuring (I've called it 'Response') as dependent, and your time variable (I've called it 'Time') as independent.

Under "Statistics..." (at the bottom of the window), I'd add "Descriptives" to the defaults, and otherwise leave them as they are. And click "OK", in the upper-right corner of the window. Or, change options, etc., according to your needs.

Menus commands work by generating and running syntax. Here's the syntax from the choices above, and the output:

 
REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .

 
Regression
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:29:20       |
|-----------------------------|---------------------------|
[DataSet0]
 
Descriptive Statistics
|--------|-------|--------------|-|
|        |Mean   |Std. Deviation|N|
|--------|-------|--------------|-|
|Response|10.3333|4.50185       |6|
|--------|-------|--------------|-|
|Time    |12.67  |11.165        |6|
|--------|-------|--------------|-|
 
Correlations
|-----------|--------|--------|-----|
|           |        |Response|Time |
|-----------|--------|--------|-----|
|Pearson    |Response|1.000   |.763 |
|Correlation|--------|--------|-----|
|           |Time    |.763    |1.000|
|-----------|--------|--------|-----|
|Sig.       |Response|.       |.039 |
|(1-tailed) |--------|--------|-----|
|           |Time    |.039    |.    |
|-----------|--------|--------|-----|
|N          |Response|6       |6    |
|           |--------|--------|-----|
|           |Time    |6       |6    |
|-----------|--------|--------|-----|
 
Variables Entered/Removed(b)
|-----|---------------|---------------|------|
|Model|Variables      |Variables      |Method|
|     |Entered        |Removed        |      |
|-----|---------------|---------------|------|
|1    |Time(a)        |.              |Enter |
|-----|---------------|---------------|------|
a All requested variables entered.
b Dependent Variable: Response
 
Model Summary
|-----|-------|--------|---------------|---------------|
|Model|R      |R Square|Adjusted R     |Std. Error of  |
|     |       |        |Square         |the Estimate   |
|-----|-------|--------|---------------|---------------|
|1    |.763(a)|.582    |.477           |3.25570        |
|-----|-------|--------|---------------|---------------|
a Predictors: (Constant), Time
 
ANOVA(b)
|-----|----------|--------------|--|-----------|-----|-------|
|Model|          |Sum of Squares|df|Mean Square|F    |Sig.   |
|-----|----------|--------------|--|-----------|-----|-------|
|1    |Regression|58.935        |1 |58.935     |5.560|.078(a)|
|     |----------|--------------|--|-----------|-----|-------|
|     |Residual  |42.398        |4 |10.600     |     |       |
|     |----------|--------------|--|-----------|-----|-------|
|     |Total     |101.333       |5 |           |     |       |
|-----|----------|--------------|--|-----------|-----|-------|
a Predictors: (Constant), Time
b Dependent Variable: Response
 
Coefficients(a)
|-----|----------|--------------------------|---------------|-----|----|
|Model|          |Unstandardized            |Standardized   |t    |Sig.|
|     |          |Coefficients              |Coefficients   |     |    |
|     |          |---------------|----------|---------------|     |    |
|     |          |B              |Std. Error|Beta           |     |    |
|-----|----------|---------------|----------|---------------|-----|----|
|1    |(Constant)|6.439          |2.120     |               |3.037|.039|
|     |----------|---------------|----------|---------------|-----|----|
|     |Time      |.307           |.130      |.763           |2.358|.078|
|-----|----------|---------------|----------|---------------|-----|----|
a Dependent Variable: Response


=============================
APPENDIX: Test data, and code
=============================
(Test data hand-entered. This is the listing)

List
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:27:13       |
|-----------------------------|---------------------------|
[DataSet0]
 
    Time Response

       1     4.00
       3     9.00
       4    10.00
      20     8.00
      22    15.00
      26    16.00


Number of cases read:  6    Number of cases listed:  6



REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .

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

[Peck, Jon]

Consider also the Curvefit procedure, which fits various different functional forms.  All these could be done with appropriate transformations, but Curvefit is more convenient.

 

---

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Richard Ristow
Sent: Thursday, February 26, 2009 10:34 AM
To: [hidden email]
Subject: Re: [SPSSX-L] How do I use SPSS to obtain parameter estimates for growth trajectories over time?

 

At 09:40 PM 2/23/2009, Jacquee Blaz wrote:

I need to generate parameter estimates for a growth curve for a continuous level variable. I have subjects that were measured on that variable over 7 timepoints on 2 different days. I'm ultimately interested in the period covered by the 7 timepoints.

This is a simple linear regression, of the dependent variable against time.

First question: How is 'time' represented in your dataset? SPSS date-time variables don't make good independent variables; they're in seconds, and their origin is in 1586. Make your 'time' variable the time from some convenient start point (typically, either the beginning or the middle of your experimental run), in whatever units you find convenient (hours, days, minutes, ...). I'm not giving details now, since your 'time' variable may not be SPSS date-time in the first place.

Second, this is probably easiest to run from the menus. Select

Analyze > Regression > Linear

Select the variable you're measuring (I've called it 'Response') as dependent, and your time variable (I've called it 'Time') as independent.

Under "Statistics..." (at the bottom of the window), I'd add "Descriptives" to the defaults, and otherwise leave them as they are. And click "OK", in the upper-right corner of the window. Or, change options, etc., according to your needs.

Menus commands work by generating and running syntax. Here's the syntax from the choices above, and the output:

 
REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .

 
Regression
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:29:20       |
|-----------------------------|---------------------------|
[DataSet0]
 
Descriptive Statistics
|--------|-------|--------------|-|
|        |Mean   |Std. Deviation|N|
|--------|-------|--------------|-|
|Response|10.3333|4.50185       |6|
|--------|-------|--------------|-|
|Time    |12.67  |11.165        |6|
|--------|-------|--------------|-|
 
Correlations
|-----------|--------|--------|-----|
|           |        |Response|Time |
|-----------|--------|--------|-----|
|Pearson    |Response|1.000   |.763 |
|Correlation|--------|--------|-----|
|           |Time    |.763    |1.000|
|-----------|--------|--------|-----|
|Sig.       |Response|.       |.039 |
|(1-tailed) |--------|--------|-----|
|           |Time    |.039    |.    |
|-----------|--------|--------|-----|
|N          |Response|6       |6    |
|           |--------|--------|-----|
|           |Time    |6       |6    |
|-----------|--------|--------|-----|
 
Variables Entered/Removed(b)
|-----|---------------|---------------|------|
|Model|Variables      |Variables      |Method|
|     |Entered        |Removed        |      |
|-----|---------------|---------------|------|
|1    |Time(a)        |.              |Enter |
|-----|---------------|---------------|------|
a All requested variables entered.
b Dependent Variable: Response
 
Model Summary
|-----|-------|--------|---------------|---------------|
|Model|R      |R Square|Adjusted R     |Std. Error of  |
|     |       |        |Square         |the Estimate   |
|-----|-------|--------|---------------|---------------|
|1    |.763(a)|.582    |.477           |3.25570        |
|-----|-------|--------|---------------|---------------|
a Predictors: (Constant), Time
 
ANOVA(b)
|-----|----------|--------------|--|-----------|-----|-------|
|Model|          |Sum of Squares|df|Mean Square|F    |Sig.   |
|-----|----------|--------------|--|-----------|-----|-------|
|1    |Regression|58.935        |1 |58.935     |5.560|.078(a)|
|     |----------|--------------|--|-----------|-----|-------|
|     |Residual  |42.398        |4 |10.600     |     |       |
|     |----------|--------------|--|-----------|-----|-------|
|     |Total     |101.333       |5 |           |     |       |
|-----|----------|--------------|--|-----------|-----|-------|
a Predictors: (Constant), Time
b Dependent Variable: Response
 
Coefficients(a)
|-----|----------|--------------------------|---------------|-----|----|
|Model|          |Unstandardized            |Standardized   |t    |Sig.|
|     |          |Coefficients              |Coefficients   |     |    |
|     |          |---------------|----------|---------------|     |    |
|     |          |B              |Std. Error|Beta           |     |    |
|-----|----------|---------------|----------|---------------|-----|----|
|1    |(Constant)|6.439          |2.120     |               |3.037|.039|
|     |----------|---------------|----------|---------------|-----|----|
|     |Time      |.307           |.130      |.763           |2.358|.078|
|-----|----------|---------------|----------|---------------|-----|----|
a Dependent Variable: Response


=============================
APPENDIX: Test data, and code
=============================
(Test data hand-entered. This is the listing)

List
|-----------------------------|---------------------------|
|Output Created               |26-FEB-2009 11:27:13       |
|-----------------------------|---------------------------|
[DataSet0]
 
    Time Response

       1     4.00
       3     9.00
       4    10.00
      20     8.00
      22    15.00
      26    16.00


Number of cases read:  6    Number of cases listed:  6



REGRESSION
  /DESCRIPTIVES MEAN STDDEV CORR SIG N
  /MISSING LISTWISE
  /STATISTICS COEFF OUTS R ANOVA
  /CRITERIA=PIN(.05) POUT(.10)
  /NOORIGIN
  /DEPENDENT Response
  /METHOD=ENTER Time  .

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