Problems & Solutions for DEBtox

Windows version

• Sometimes the display of the DEBtox window is distored. Press the cursor up or cursor down to refresh the window.

• The result in the test comparison window is wrong.

  You are advised to mail to debtox@bio.vu.nl for a new update of DEBtox or execute the statistical test yourself:

  • sublethal effects: i.e. Body growth, Reproduction and Population experiment.
    Take the logarithm of the ratio of the mean deviations and multiply the results by 2n, where n is the number of observations. Compare the result with a Chi-square distribution. The number of degrees of freedom is equal to the difference in the number of estimated parameters.
  • lethal effects: i.e. Survival experiment.
    Calculate the difference between the deviances and compare the result with a Chi-square distribution. The number of degrees of freedom is equal to the difference in the number of estimated parameters.

• The graph of the time profile for the survival experiment is wrong in the case that the number of individuals at the start of the experiment differ for the different concentrations. This problem only occurs in the Windows version, not in the Unix version. The calculations are correct; it is only a graphical problem.

• The calculation of the profile likelihood for the survival experiment sometimes goes wrong. If the optimal value of the elimination parameter equals zero or infinite, profile calculations lead to a floating point error. You are advised to mail to debtox@bio.vu.nl for a new update of DEBtox.

Unix version

• In the short description about the options for DEBtox (shown with debtox -h) the description for the Confidence Level tells that you can enter a number between 0.01 and 99.99. This has to be 0.01 and 0.995.

Back to the DEBtox page.