+ Reply to Thread
Results 1 to 2 of 2

Thread: p value of the constant, can I ignore it?

  1. #1
    Points: 718, Level: 14
    Level completed: 18%, Points required for next Level: 82

    Posts
    4
    Thanks
    1
    Thanked 1 Time in 1 Post

    Question p value of the constant, can I ignore it?




    I'm analysing the influence of various factors on hare harvests in a specific are.
    E.g. How do the hare harvests change when the fox harvests increase.

    Term Coef | SE-Coef | T-Value | P-Value
    Constant | 415.0 | 28.7 | 14.46 | 0.000
    Annual fox harvest | -1.054 | 0.193 | -5.45 | 0.001

    Here the p value of the constant is highly significant.

    But for other comparisons, such as the number of worker employed in agricultre (indicator for the level of mechanisation= the P value of the constant is very high:

    Term | Coef | SE Coef | T-Value | P-Value
    Constant | -12.4 | 71.0 | -0.17 | 0.868
    Number of workers | 0.1220 | 0.0249 | 4.89 | 0.005

    What does that mean? I somewhere red that my prediction around the origin is going to be inaccurate. Does this matter, and what does it mean for the relationship between the number of workers and the hare harvest?

    I couldn't format the tables, sorry. hope you still get it.

  2. #2
    TS Contributor
    Points: 12,287, Level: 72
    Level completed: 60%, Points required for next Level: 163
    rogojel's Avatar
    Location
    I work in Europe, live in Hungary
    Posts
    1,471
    Thanks
    160
    Thanked 332 Times in 312 Posts

    Re: p value of the constant, can I ignore it?


    hi,
    the interpretation of the constant is often irrelevant - in your case it would be the average hare harvest in the case of 0 employed workers - possibly a completely mechanised agriculture with no workers? So, in general one does not ignore the constant because of the p value but because the X=0 case is outside the range of realistic values. One trick would be to repeat the analéysis with mean (or median) centered X-es in which case the constant is easily interpretable - the Y at the mean (or mediean) X.

    regards

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats