- I have done aquatic animal model experiments, which are treated with single, combination of different bacteriophage under laboratory condition, and I controlled bacterial infection by phages. I need to compare differences, which phage (no phage, single & cocktail phage) would be excellent for this experiment.
- I need to analyze the data (animal survival/ bacterial count/ phage count) by One way ANOVA. I have seen in many research article related to phage therapy, they used either Ducan or Tukey's test only. Could anyone explain me the differences and which test suitable in my case?
With Thanks and Regards
Stalin. N
To address specific aspects of aquatic animal model experiments, such as modeling change in bacteriophage under laboratory condition, it is preferable to define source in terms of either controlled bacterial infection by phages. The problem of defining Ducan or Tukey's test by one-way ANOVA becomes more difficult for estimates in smaller geographic regions or local study areas.
Data samples can sometimes be distinguished from each other by external factors such as location or treatment. Such factor information can be added to worksheets (in Yours statistical program). Many of the data operations and analyses make use of factors to define groups of samples when a significance test results .
Both validation tests are used to check the significance of treatment means. Therefore, there is no need to think about theirs differences. However, Duncan can be used with a less premonitory research. Tukey's test is more conservative (if significance between pairs of individual will be reported that means a high probability value). This fact should help you interpret the Your hypothesis test of obtaining a statistically significant result. Thus, Tukey's test is more precisely, but less permissive).
You may consider following attachments, disscusing about both comparisons post-hoc method:
http://archive.bio.ed.ac.uk/jdeacon/statistics/tress7.html
https://web.mst.edu/~psyworld/tukeyssteps.htm
I have conducted Duncans Multiple Range Test (DMRT).
Both DMRT and tukey's tests are used to differentiate the significance of treatment means. I think both are good to use. I suggest you use DMRT test as it widely being used in research and publications.
Multiple Range and Multiple F Tests
David B. Duncan
Biometrics
Vol. 11, No. 1 (Mar., 1955), pp. 1-42
Here is a relevant post
https://www.researchgate.net/post/What_is_the_difference_between_Tukeys_test_and_DMRT_in_one_way_ANOVA
Why ANOVA?
Survival should better be studied by survival analysis (the name is a hint ;) ). And also for bacterial and phage counts it might be more appropriate to use a Poisson-, a negative binomial- or a gamma model.
With any sort of statistical testing, you need to assure that your data meets the assumptions. If your data do not meet the assumptions of the test, then you are using the wrong test. I don't know these specific tests, but I'd suggest you do some reading about the assumptions for each one to help you determine why and when each analysis is used.
Significance tests have assumptions. Very often, educational data do not satisfy them, e.g., random sampling, normality,etc. Yet, they are used and reported just the same. Views, please
Dear Muhammad Shafiq,
Greetings!,
Thanks very much for your valuable reply, much appreciated
-Dr. Nattan Stalin.
Thia question was asked three years ago. Why is it still circulating?
Respected Keith Godfrey,
Answering previous questions by experts my helpful to fresher who are working in the field of research.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA