I have run my logistic regression model to find out whether the gender of a test name is a predictor of the gender of the word assigned to the test name.

q2.glm=glm(item.gender ~ test.gender, data=q2data, family="binomial")

R has returned the following coefficients:


            Estimate Std. Error z value Pr(>|z|) 

(Intercept)  0.05365    0.12384   0.433  0.66484  

test.gender  0.48794    0.17836   2.736  0.00623 **

I understand the P value for test.gender means I can reject the null hypothesis that the gender of the test name has no influence on the gender of the word assigned to it, however I don't understand how to interpret the other values.

Please can someone explain what the other values mean? What does the intercept category mean? many thanks - I have a big project due soon and am struggling to understand it!

  • 1
    $\begingroup$ Small point in your post, but big point in general: You say "influence" but statistical connections cannot confirm causal relationships. They must be assessed using other means. $\endgroup$ – rolando2 Dec 11 '13 at 20:51
  • 1
    $\begingroup$ If you're looking for someone to explain what the intercept, standard error, z, and Pr(>|z|) mean and how they function, I think you're best off looking in a textbook or doing a web search for a good intro treatment of logistic regression. (try David Garson's at N. Carolina State U., or UCLA's stats page). Contributors to this site may not want to write all that up. $\endgroup$ – rolando2 Dec 11 '13 at 20:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.