Comparing variables, both numerical and characters

[Lysegroentblad]

Hi
I am new to SAS and have tried running examples of code trying to make them apply to my own dataset, but the log keeps coming up every colour but blue. I have a data set with both female and male rats who is tested on how fast they can complete a maze (time). I am also given the weight of the rats. Furthermore i am given their rank (alpha, beta, omega/AA, AB, AC etc) of the rat (experiment), time (latency). All in all I have 4 coloumns. I now want to see how these variables affect each other, i.e. how rank affect time, how time is affected by weight and if there is any correlation between gender and time and so on.
I have run this script so far: 

DATA MIT_EKS;
INFILE 'C:\Users\MajaThuren\OneDrive\Dokumenter\Skovbrugsvidenskab\StatBio\Exercises\Exercise 1\RATS.prn' FIRSTOBS=2;
INPUT experiment $ time $ weight $ gender $;
RUN;
PROC PRINT DATA=MIT_EKS; RUN;

My results viewer look like this:

And then I'm quite stumped as to how i write a program that compares correlations between the variables?

I know it's simple, but I swear I tried for hours, I think I might just be a little thick.

A second question: I would also like to run a program that excludes one value from a variable. In the coloumn time I have an error value of -1,3. I know i can just remove it from my excel, but i would like to learn how to exclude one value with the help of code, that is when i'm running statistics for correlations between that variable (time) and fx weight.

Best regards

Maja

[Quentin]

Hi,

Please show the code you have tried (for the correlations). Are you perhaps using PROC CORR?  Please also show the log (with the errors) you are getting from that code.  

For excluding data, one way would be to use a WHERE statement.

Opinions about helping with homework vary; I think it's fine to ask for help with homework questions, but people will be better able to help you (and you will learn more) by showing what you have tried, and where you are stuck.

[Lysegroentblad]

These are the commands I've tried so far:

DATA MIT_EKS;
INFILE 'C:\Users\MajaThuren\OneDrive\Dokumenter\Skovbrugsvidenskab\StatBio\Exercises\Exercise 1\RATS.prn' FIRSTOBS=2;
INPUT Subject $ Rank $ Time $ Weight $ Gender $;
RUN;
PROC PRINT DATA=MIT_EKS; RUN;
PROC SORT; BY Time Weight; RUN;
 
PROC MEANS MEAN N VAR STD STDERR CV MIN MAX; RUN;

PROC CORR SPEARMAN PEARSON KENDALL;
VAR Time Weight; RUN;
PROC MEANS MEAN N VAR STD STDERR CV MIN MAX;
BY Time Weight; RUN;
PROC CORR SPEARMAN;
VAR Subject Experiment;
BY Time Weight; RUN;
PROC GLM;
CLASS Subject;
MODEL Time = Subject;
MEANS Subject / DUNNETT ('control'); RUN;

They stop working after: PROC SORT; BY Time Weight; RUN;

This is my log:
NOTE: PROCEDURE GLM used (Total process time):
      real time           32.82 seconds
      cpu time            4.37 seconds


63   DATA MIT_EKS;
64   INFILE 'C:\Users\MajaThuren\OneDrive\Dokumenter\Skovbrugsvidenskab\StatBio\Exercises\Exercise
64 ! 1\RATS.prn' FIRSTOBS=2;
65   INPUT Subject $ Rank $ Time $ Weight $ Gender $;
66   RUN;

NOTE: The infile
      'C:\Users\MajaThuren\OneDrive\Dokumenter\Skovbrugsvidenskab\StatBio\Exercises\Exercise
      1\RATS.prn' is:

      Filename=C:\Users\MajaThuren\OneDrive\Dokumenter\Skovbrugsvidenskab\StatBio\Exercises\Exercise
       1\RATS.prn,
      RECFM=V,LRECL=32767,File Size (bytes)=10618,
      Last Modified=15. februar 2022 16:08:59,
      Create Time=14. februar 2022 10:02:35

NOTE: 192 records were read from the infile
      'C:\Users\MajaThuren\OneDrive\Dokumenter\Skovbrugsvidenskab\StatBio\Exercises\Exercise
      1\RATS.prn'.
      The minimum record length was 53.
      The maximum record length was 53.
NOTE: The data set WORK.MIT_EKS has 192 observations and 5 variables.
NOTE: DATA statement used (Total process time):
      real time           0.12 seconds
      cpu time            0.12 seconds


67   PROC PRINT DATA=MIT_EKS; RUN;
NOTE: Writing HTML Body file: sashtml3.htm

NOTE: There were 192 observations read from the data set WORK.MIT_EKS.
NOTE: PROCEDURE PRINT used (Total process time):
      real time           2.87 seconds
      cpu time            2.51 seconds


68   PROC SORT; BY Time Weight; RUN;

NOTE: There were 192 observations read from the data set WORK.MIT_EKS.
NOTE: The data set WORK.MIT_EKS has 192 observations and 5 variables.
NOTE: PROCEDURE SORT used (Total process time):
      real time           0.05 seconds
      cpu time            0.06 seconds


69
70   PROC MEANS MEAN N VAR STD STDERR CV MIN MAX; RUN;

ERROR: Statistics requested for printing, but no VAR statement has been specified. No printed
       output will be produced.
ERROR: Neither the PRINT option nor a valid output statement has been given.
NOTE: The SAS System stopped processing this step because of errors.
NOTE: PROCEDURE MEANS used (Total process time):
      real time           0.04 seconds
      cpu time            0.04 seconds

71


72   PROC CORR SPEARMAN PEARSON KENDALL;
73   VAR Time Weight; RUN;
ERROR: Variable Time in list does not match type prescribed for this list.
ERROR: Variable Weight in list does not match type prescribed for this list.

NOTE: The SAS System stopped processing this step because of errors.
NOTE: PROCEDURE CORR used (Total process time):
      real time           0.02 seconds
      cpu time            0.03 seconds



74   PROC MEANS MEAN N VAR STD STDERR CV MIN MAX;
75   BY Time Weight; RUN;

ERROR: Statistics requested for printing, but no VAR statement has been specified. No printed
       output will be produced.
ERROR: Neither the PRINT option nor a valid output statement has been given.
NOTE: The SAS System stopped processing this step because of errors.
NOTE: PROCEDURE MEANS used (Total process time):
      real time           0.03 seconds
      cpu time            0.03 seconds



76   PROC CORR SPEARMAN;
77   VAR Subject Experiment;
ERROR: Variable Subject in list does not match type prescribed for this list.
ERROR: Variable EXPERIMENT not found.
78   BY Time Weight; RUN;

NOTE: The SAS System stopped processing this step because of errors.
NOTE: PROCEDURE CORR used (Total process time):
      real time           0.02 seconds
      cpu time            0.03 seconds



79   PROC GLM;
80   CLASS Subject;
81   MODEL Time = Subject;
ERROR: Variable Time in list does not match type prescribed for this list.
NOTE: The previous statement has been deleted.
82   MEANS Subject / DUNNETT ('control'); RUN;
ERROR: Effects used in the MEANS statement must have appeared previously in the MODEL statement.
NOTE: The previous statement has been deleted.

This is an example of my data:


[andreas_lds]

I would start by importing numeric-content into numeric variables: delete $ after time and weight.

If that doesn't solve your problem (most likely it won't), please post the complete log.

[Lysegroentblad]

Hi and thank you for responding.

When I delete the $ all values that are not decimal-numbers disapear from these columns - as such:

[ballardw]

Looking at your first pictures the data is apparently using a comma for the decimal separator. This is a National Language Issue and if your SAS is running with a setting for country that using a dot as the decimal separator that would cause the issue (along with a log full of invalid data messages typically).

Try this to read the comma as decimal

DATA MIT_EKS;
INFILE 'C:\Users\MajaThuren\OneDrive\Dokumenter\Skovbrugsvidenskab\StatBio\Exercises\Exercise 1\RATS.prn' FIRSTOBS=2;
INPUT Subject $ Rank $ Time :Numx12. Weight :Numx12. Gender $;
RUN;

[Lysegroentblad]

This worked! When I added numx12. to my numerical variables all the procs worked - Thank you.