Analysis of the results of Duncan s multiple comparisons

It feels a little strange, I redid your data again, and this is the result, but if you just look at the data to analyze, combination 5 and combination 6 fluctuate between 67-69, at least the significant difference is not big, in a column, even a little thicker combination 4 can be counted together. 243 (combination 2) is a, 135 138 (combination ) is b, 79, 68, 69 (combination 4, 5, 6) is c, uppercase is horizontal, lowercase is horizontal. I don't understand, I don't understand, and I've only recently started researching.

I just tried to divide the numbers around 95, 90, and 75 into three groups, and the result is that 73 is in the first column, and 95 and 90 are in the second column, and the significant difference is that there are two categories: A and B.

Please refer to the literature "DUNCAN Multiple Comparison and Data Gap Filling Method and Its Application".

There is a specific introduction above.

I have this article**.

How was it solved? I've had this happen to me as well.

1,2,3..It is considered to be a significant difference in different grids.

68 a69 b

78 c134 d

138 e243 f

The results will naturally vary depending on the method, especially if the edges are significant. (Nanxin.com SPSS).

If you don't know how to use SPSS, don't mess around.

I often help others with this kind of data analysis.

There is a book called Management Statistics, which I studied in college, but it has been a long time since I remember it very clearly. The subject of management statistics is still very useful. Multiple ANOVA is also a difficult part.

You'd better read the book. In my image, in multiple ANOVA, there doesn't seem to be any a, b, c, or anything like that. If there are parameters that may also be added by the author to the study, you may wish to take a look at the assumptions and research conditions in **.

The significance test is very simple. But it's also more troublesome to say here, and reading books is the last word. If you can't find the book, you can take a look at the online teaching ppt.

dun t？Just select one in the pairwise comparison, and then set a reference, the first or the last.

Didn't understand what your trial variables were. Are your media formulations, culture conditions, etc. the same? All else being equal, then your data can't be analyzed by Duncan. This trial also makes no sense.

I'm dizzy too. How can ANOVA be performed without duplication? Ask the teacher again, why don't you do a repeat test?

1. Post-the-post testing, multiple comparisons, and simple effect analysis differences.

1) The application scenarios are different.

1. Post-event inspection refers to a method that compares the estimated results of the market risk measurement method or model with the actual profit or loss to test the accuracy and reliability of the measurement method or model, and adjusts and improves the measurement method or model accordingly.

2. Simple effect analysis is to know the existence of interaction after doing ANOVA for effect analysis.

3. Multiple comparisons are found to have significant differences between the mean of each sample after ANOVA, and multiple comparison can be used to further determine which two averages have differences and which two averages have no differences.

2) The test methods are different.

1. Post-mortem inspection can be done with dunnett'The S t test was used to test the differences between the experimental group and the control group in group I.

2. The effect analysis usually adopts the method of hierarchical analysis.

3. Multiple comparison methods include LSD minimum significant difference test, Newman Keuls (Nk test), Duncan test, Tukey test, Dunnett test and Scheffé test.

Second, post-the-post testing, multiple comparisons, and simple effect analysis application timing.

1. Post-event testing is mostly used for market risk estimation, and the estimation results of the measurement method or model are compared with the actual profit and loss.

2. Simple effect analysis is when an interaction is found to exist after ANOVA.

3. Multiple comparisons are made when the analysis of variance determines that there is a difference between the population averages, and it is necessary to further determine which two averages have differences and which two averages have no differences.

The post-hoc test is a one-to-one comparison between groups when ANOVA ANOVA ANOVA is performed and a difference is found.

Multiple comparisons are just a formality of turning one-to-one pairs into a table.

Simple effect analysis is a step further than the post-hoc test, and is used when there is an interaction, which is actually grouped by conditions and then compared.