Data reduction question

[Leslie Cornick]

Hello SPSS Experts!

 

I have a student with animal behavior data that are currently in the form of 2-minute scan samples in the format below:

 

Date                 Time        Bear     Behavior  Beds    Other Enrich Weather    Season  Other Stim

08/12/08            1:00            1          2          0          3                      3          1          1

08/12/08            1:00            2          1          0          3                      3          1          1

 

Most of the variables are coded nominal, and we want to bin the data so that she can analyze rates of behavior (n/unit time), changes in durations of behaviors etc. from the scan sample data.

 

Is there a straightforward way to do this data reduction in SPSS or does she need to calculate these sums etc in Excel and re-enter in a different form?  I’ve used that tedious method in the past, but it just seems there should be an easier way!

 

Any help would be greatly appreciated!

 

Cheers,

Leslie

 

************************************************

Leslie A. Cornick, Ph.D.

Associate Professor, Marine Biology & Statistics

Department of Environmental Science

Alaska Pacific University

4101 University Drive

Anchorage, AK 99508

[hidden email]

Phone [phone]

Fax    [phone]

 

[Art Kendall]

This may be a start.  It assumes that each line is an observation period of 2 minutes.

However, it is unclear whether or not each line is a single animal.

Is a bear a "case"  or is it "playing with a teddy bear"?

Since the scans are 2 minute what does "time" mean   for the first line is it 08/12/08 at 1:00 am?

What do you mean by duration in this data?

DATA LIST list /
 theday(adate10)thetime(time5)  Bear Behavior  Beds  Other_Enrich Weather  Season  Other_Stim (7f1).
begin data.
08/12/08 1:00  1  2  0  3  3  1  1
08/12/08 1:00  2  1  0  3  3  1  1
08/12/08 2:00  2  2  0  3  3  1  1
08/12/08 2:00  2  1  3  3  3  1  1
08/12/08 3:00  1  2  4  3  3  1  1
08/12/08 3:00  2  1  7  3  3  1  1
end data.
compute behave_rate = sum.1(bear to other_stim)/2.
LIST .


Art Kendall
Social Research Consultants

Leslie Cornick wrote:

Hello SPSS Experts!

 

I have a student with animal behavior data that are currently in the form of 2-minute scan samples in the format below:

 

Date                 Time        Bear     Behavior  Beds    Other Enrich Weather    Season  Other Stim

08/12/08            1:00            1          2          0          3                      3          1          1

08/12/08            1:00            2          1          0          3                      3          1          1

 

Most of the variables are coded nominal, and we want to bin the data so that she can analyze rates of behavior (n/unit time), changes in durations of behaviors etc. from the scan sample data.

 

Is there a straightforward way to do this data reduction in SPSS or does she need to calculate these sums etc in Excel and re-enter in a different form?  I’ve used that tedious method in the past, but it just seems there should be an easier way!

 

Any help would be greatly appreciated!

 

Cheers,

Leslie

 

************************************************

Leslie A. Cornick, Ph.D.

Associate Professor, Marine Biology & Statistics

Department of Environmental Science

Alaska Pacific University

4101 University Drive

Anchorage, AK 99508

[hidden email]

Phone [phone]

Fax    [phone]

 

[Richard Ristow]

At 05:08 PM 1/29/2009, Leslie Cornick wrote:

>I have a student with animal behavior data that are currently in the
>form of 2-minute scan samples in the format below:
>
>Date      Time  Bear Behavior Beds Other Enrich Weather Season  Other Stim
>08/12/08  1:00    1     2      0      3     3      1       1
>08/12/08  1:00    2     1      0      3     3      1       1
>
>Most of the variables are coded nominal, and we want to bin the data
>so that she can analyze rates of behavior (n/unit time), changes in
>durations of behaviors etc. from the scan sample data.

See Art Kendall's response, if you want to summarize values within
lines. ('Lines'= records, called 'cases' in SPSS documentation.)

If you want to summarize over several lines (like those from the same
date and time, as in your example data), look at AGGREGATE, which is
very powerful and versatile. If it doesn't seem to do what you want,
post again. You can get amazing effects from AGGREGATE by techniques
like creating transformed variables specifically to be AGGREGATEd.


P.S. As to the substance of your study, you have,

>Date      Time  Bear Behavior Beds ...

I will not mention porridge, I will not mention porridge, I will not
mention porridge ...

-Grins and best wishes,
  Richard

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

[Leslie Cornick]

Thanks everyone for the very helpful responses!! Much appreciated.

And our bears eat seals! ;-)

************************************************
Leslie Cornick
[hidden email]
907.564.8885

-----Original Message-----
From: Richard Ristow [mailto:[hidden email]]
Sent: Wednesday, February 04, 2009 1:51 PM
To: Leslie Cornick; [hidden email]
Cc: Art Kendall
Subject: Re: Data reduction question

At 05:08 PM 1/29/2009, Leslie Cornick wrote:

>I have a student with animal behavior data that are currently in the
>form of 2-minute scan samples in the format below:
>
>Date      Time  Bear Behavior Beds Other Enrich Weather Season  Other
Stim
>08/12/08  1:00    1     2      0      3     3      1       1
>08/12/08  1:00    2     1      0      3     3      1       1
>
>Most of the variables are coded nominal, and we want to bin the data
>so that she can analyze rates of behavior (n/unit time), changes in
>durations of behaviors etc. from the scan sample data.

See Art Kendall's response, if you want to summarize values within
lines. ('Lines'= records, called 'cases' in SPSS documentation.)

If you want to summarize over several lines (like those from the same
date and time, as in your example data), look at AGGREGATE, which is
very powerful and versatile. If it doesn't seem to do what you want,
post again. You can get amazing effects from AGGREGATE by techniques
like creating transformed variables specifically to be AGGREGATEd.


P.S. As to the substance of your study, you have,

>Date      Time  Bear Behavior Beds ...

I will not mention porridge, I will not mention porridge, I will not
mention porridge ...

-Grins and best wishes,
  Richard






************************************************************************
************
This footnote confirms that this email message has been scanned by
PineApp Mail-SeCure for the presence of malicious code, vandals &
computer viruses.
************************************************************************
************

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