Quantcast

Logistic regression: Sensitivity, Specificity, % False Negative a nd % False Positive

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Logistic regression: Sensitivity, Specificity, % False Negative a nd % False Positive

Patricia  Cleland
A colleague and I are performing logistic regressions.  He's using SAS and
I'm using SPSS.

I've been able to match up all my SPSS output  with his SAS output, except
that the SAS output contains a table that shows the Sensitivity,
Specificity, % False Negative and % False Positive for each probability
level.  Is it possible to generate this table in SPSS?   I've looked at the
documentation for LOGISTIC REGRESSION and don't see an option that will do
it. Alternatively, is it available as part of another analysis?

TIA,

Pat
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Logistic regression: Sensitivity, Specificity, % False Negative a nd % False Positive

Marta García-Granero
Hi Pat

If you are using multiple logistic regression, then you have to save
the predicted values as a new variable  and then, as someone else has
suggested, run ROC analysis.  You will get sensitivity and false
positive rate. False negative rate and specificity can be easily
obtained by substraction.

The following example computes sensitivity & 1-specificity (=false
positive rate) for a simple logistic regression model, but the process
is similar for multiple regression models:

* Dataset from Hosmer&Lemeshow's Applied Logistic Regression *.
DATA LIST FREE/ age chd (2 F8.0).
BEGIN DATA
 20 0 23 0 24 0 25 0 25 1 26 0 26 0 28 0 28 0 29 0
 30 0 30 0 30 0 30 0 30 0 30 1 32 0 32 0 33 0 33 0
 34 0 34 0 34 1 34 0 34 0 35 0 35 0 36 0 36 1 36 0
 37 0 37 1 37 0 38 0 38 0 39 0 39 1 40 0 40 1 41 0
 41 0 42 0 42 0 42 0 42 1 43 0 43 0 43 1 44 0 44 0
 44 1 44 1 45 0 45 1 46 0 46 1 47 0 47 0 47 1 48 0
 48 1 48 1 49 0 49 0 49 1 50 0 50 1 51 0 52 0 52 1
 53 1 53 1 54 1 55 0 55 1 55 1 56 1 56 1 56 1 57 0
 57 0 57 1 57 1 57 1 57 1 58 0 58 1 58 1 59 1 59 1
 60 0 60 1 61 1 62 1 62 1 63 1 64 0 64 1 65 1 69 1
END DATA.
VAR LABELS age 'Age at follow-up start'
 /chd 'Coronary Heart Disease'.
VALUE LABELS chd 0 'Absent' 1 'Present'.

LOGISTIC REGRESSION chd
  /METHOD = ENTER age
  /SAVE = PRED (predict).

ROC  predict BY chd (1)
  /PLOT = CURVE
  /PRINT = SE
  /CRITERIA = CUTOFF(INCLUDE) TESTPOS(LARGE) DISTRIBUTION(FREE) CI(95)
  /MISSING = EXCLUDE .


PC> A colleague and I are performing logistic regressions.  He's using SAS and
PC> I'm using SPSS.

PC> I've been able to match up all my SPSS output  with his SAS output, except
PC> that the SAS output contains a table that shows the Sensitivity,
PC> Specificity, % False Negative and % False Positive for each probability
PC> level.  Is it possible to generate this table in SPSS?   I've looked at the
PC> documentation for LOGISTIC REGRESSION and don't see an option that will do
PC> it. Alternatively, is it available as part of another analysis?


-----
Regards,
Dr. Marta García-Granero,PhD           mailto:[hidden email]
Statistician

---
"It is unwise to use a statistical procedure whose use one does
not understand. SPSS syntax guide cannot supply this knowledge, and it
is certainly no substitute for the basic understanding of statistics
and statistical thinking that is essential for the wise choice of
methods and the correct interpretation of their results".

(Adapted from WinPepi manual - I'm sure Joe Abrahmson will not mind)
Loading...