I have conducted an experiment where fish are exposed to different concentrations of a substance and then several metrics are measured (swimming speed, etc) in order to assess the substance's effects on the fish. However, due to an issue with our supply we could not get enough of this fish to have different sets of fish in different tanks, each exposed to different concentrations of the substance and then measured. This also meant we do not have enough for replicates in that manner as well.

Therefore we decided to expose the one set of fish that we have to increasing exposures so that it becomes more of a long term repeat exposure study on toxicity rather than acute toxicity. We collected the measurements but now are unsure of what statistical test would be appropriate for this experimental design to see the significance of this substance's concentrations on these different metrics.

Initially with the previous experimental design we wanted to perform an ANOVA, but we cannot use this method as one of the requirements for ANOVA is to have independence of treatment groups from each other, which we now no longer have.

Thank you very much, I appreciate any guidance and suggestions for possible statistical methods to perform in this experimental design.

  • $\begingroup$ Can you distinguish the individual fish so that you know, for example, the swimming speed of the same fish following each of the treatments? $\endgroup$
    – EdM
    Jul 4 at 14:18


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.