I'm a rookie in statistics, but just wrapped up a semester of R analysis for natural resources. I'm working on my grad capstone and hoping I might be able to get some help generating ideas for worthwhile analysis to answer the following questions:

****Changes in Run Timing**** I don't have any effect for this question, but just wanting to show how the timing of the fish run has changed from the start of a project to current times. My data is based off of trapping. I was thinking of determining the peak annually, then doing an ANOVA or Linear Model. Is there a better way of doing this?

****Changes in fork lengths**** I have fork length data for multiple rivers over a 15 year period. I wanted to analyze how hatchery or wild status predicts lengths. Are hatchery fish smaller/longer than wild? I had thought to find the mean fork length for fish annually both by sex and hatchery/wild status, then do ANOVA or Linear Model. Is there a better way of analyzing?

****Changes in Age Structure**** Through field sampling, I have scale analysis results telling me the age structure of each year's spawners. Similar to fork lengths, how has this changed over time, and specifically, as an effect of wild/hatchery status and sex.

Any help is much appreciated! Just looking for some ideas because while I really enjoyed ANOVA/regression analysis, I'm sure there are some much better ideas for these questions.


  • $\begingroup$ Are you interereted in Anova or regression ? $\endgroup$ Jun 19 '17 at 12:21
  • $\begingroup$ Either or, just trying to figure out which analysis would make the most sense in this case. $\endgroup$
    – salmon_man
    Jun 20 '17 at 4:05
  • $\begingroup$ One-way Anova should be a good option for your issues - fork length and age-structure presuming that you are not interested in a single complete model. $\endgroup$ Jun 20 '17 at 5:28
  • $\begingroup$ Thank you. A 1-way ANOVA makes sense for those. I was also contemplating a 3-way ANOVA to look at what is a better predictor on fork length - sex, mark status, or age. $\endgroup$
    – salmon_man
    Jun 21 '17 at 1:23
  • $\begingroup$ what is matk status ? $\endgroup$ Jun 21 '17 at 5:39

In view of your comments(following your question), you can apply ANOVA to find out the effect of mark-up(hatchery Versus wild) and sex-type (male Vs. female) on fork-length. And two-way Anova could be implemented for evaluating the combined-effect of the specified variables i.e whether mark-up and sex-type together affect the fork-length significantly. Two way Anova will account for interactions. Given the nature of two variables - Age and fork-length, these may be analyzed by correlation-analysis and/or regression analysis. The combined effect of all three variables may be estimated and tested with the help of lme - Linear mixed-effect model.

  • $\begingroup$ Great, thank you! Lastly, to compare the mean return peaks from each year, what would be best way to do so? I'd have 15 years. $\endgroup$
    – salmon_man
    Jun 24 '17 at 5:11
  • $\begingroup$ mean return peaks - data ? annual ? for different categories or meta-analytic i.e. over $\endgroup$ Jun 24 '17 at 11:52
  • $\begingroup$ not clear. please explain - what is mean return peak? For what do you want to estimate - category-wise , sexwise , etc. $\endgroup$ Jun 24 '17 at 11:58
  • $\begingroup$ Please elaborate the mean return peaks and what is your data ? $\endgroup$ Jun 24 '17 at 12:09
  • $\begingroup$ I have dates with numbers of fish that returned each day for that year. I was going to convert this to Days after Trapping Began. So one year, my average peak might be 35 days, another year 30, and so forth. Thank you! $\endgroup$
    – salmon_man
    Jun 24 '17 at 23:55

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.