Significance for "Proximities"-command

[Julia Hermann]

Dear SPSS-Users!

We are facing a problem with a correlation-test for our data.

Usually, you can use the "Correlations"-command for testing the correlations
between different variables, SPSS prints the correlation-value and also the
significance-value in the matrix.

Now we need to conduct a different test, as we want to analyze the
correlations between dichotomized variables: We make a "Fourfold Point
Correlation" with the following command:

Proximities (variables)
  /VIEW=VARIABLE
  /MEASURE= PHI (1,0).

SPSS doesn't print the significance-values with this command which is of
course necessary to make a conclusion about the correlations.

Do you know which command is necessary to print these signifance-values or
if we can use a different test for this? I didn't find a subcommand for this
problem...

Thank you for your help!

Best Regards

Julia

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

[David Marso]

The last time I looked the Pearson PM correlation is *IDENTICAL* to the Phi coefficient and you can readily verify that either by running BOTH commands and eyeballing it or prove it mathematically by deriving the Phi from the Pearson formula (utterly trivial exercise IMO).
---

Julia Hermann wrote

Dear SPSS-Users!

We are facing a problem with a correlation-test for our data.

Usually, you can use the "Correlations"-command for testing the correlations
between different variables, SPSS prints the correlation-value and also the
significance-value in the matrix.

Now we need to conduct a different test, as we want to analyze the
correlations between dichotomized variables: We make a "Fourfold Point
Correlation" with the following command:

Proximities (variables)
  /VIEW=VARIABLE
  /MEASURE= PHI (1,0).

SPSS doesn't print the significance-values with this command which is of
course necessary to make a conclusion about the correlations.

Do you know which command is necessary to print these signifance-values or
if we can use a different test for this? I didn't find a subcommand for this
problem...

Thank you for your help!

Best Regards

Julia

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

[Art Kendall]

Phi is an old fashioned shortcut for hand calculation of Pearson's r when the two variables are dichotomous.
To see this,  open a new instance of SPSS. copy the syntax below into a syntax window. Run it.

*PHI is PHI is ordinary r.
new file.
input program.
   vector x (5,f1).
   loop id = 1 to 25.
      loop #p = 1 to 10.
         compute x(#p) = rv.bernoulli(.3).
      end loop.
      end case.
   end loop.
   end file.
end input program.
CORRELATIONS
  /VARIABLES=x1 x2 x3 x4 x5
  /PRINT=TWOTAIL NOSIG
  /MISSING=PAIRWISE.
NONPAR CORR
  /VARIABLES=x1 x2 x3 x4 x5
  /PRINT=KENDALL TWOTAIL NOSIG
  /MISSING=PAIRWISE.
PROXIMITIES x1 x2 x3 x4 x5
  /VIEW=VARIABLE
  /MEASURE=PHI (1,0).
crosstabs tables= x1 by x2 to x5
  /format= NOTABLES
  /statistics = phi.

Art Kendall
Social Research Consultants
On 6/11/2012 7:05 AM, Julia Hermann wrote:

Dear SPSS-Users!

We are facing a problem with a correlation-test for our data.

Usually, you can use the "Correlations"-command for testing the correlations
between different variables, SPSS prints the correlation-value and also the
significance-value in the matrix.

Now we need to conduct a different test, as we want to analyze the
correlations between dichotomized variables: We make a "Fourfold Point
Correlation" with the following command:

Proximities (variables)
  /VIEW=VARIABLE
  /MEASURE= PHI (1,0).

SPSS doesn't print the significance-values with this command which is of
course necessary to make a conclusion about the correlations.

Do you know which command is necessary to print these signifance-values or
if we can use a different test for this? I didn't find a subcommand for this
problem...

Thank you for your help!

Best Regards

Julia

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