I have a quite basic question, when doing a one-way ANOVA (4 groups) in SPSS, if i don't find a significative effect (F test), does it make sense to look for polynomial linear trend?? I mean, my F isn't significant (p = 0.13) but the linear effect is significant (p = 0.02). How can u explain that? i wouldn't expect to find the linear effect if there is not a general effect.

Thank very much you for your help,

SC. ]]>

Dataset 1:

I had 20 randomly selected human subjects rate 50 faces based on their attractiveness on an ordinal scale from 1-5. Each subject rated the same faces. I want to know how similar their ratings are to each other. Which method should I use? 1-k?

Dataset 2:

I have 20 4x4 confusion matrices, one for each human subject. I've been asked by a reviewer to measure how similar the confusion matrices are to each other. ICC seems like a good tool to use here, and I could do this by treating each confusion matrix as a 16-dim vector and then feeding the 16x20 array of data into an ICC algorithm. Again, which ICC method is most appropriate? 1-k?

In my niche, I don't see people use ICC much, so I don't have any experience with the method and wanted to double check with experts to prevent myself from doing something stupid.

Here are the types in my analysis software:

'1-1': The degree of absolute agreement among measurements made on randomly selected objects. It estimates the correlation of any two measurements.

'1-k': The degree of absolute agreement of measurements that are averages of k independent measurements on randomly selected objects.

'C-1': The degree of consistency among measurements. Also known as norm-referenced reliability and as Winer's adjustment for anchor points.

'C-k': The degree of consistency for measurements that are averages of k independent measurements on randomly selected objects. Known as Cronbach's alpha in psychometrics. The degree of consistency for averages of k independent measures made under the fixed levels of column factor.

'A-1': The degree of absolute agreement among measurements. Also known as criterion-referenced reliability. The absolute agreement of measurements made under the fixed levels of the column factor.

'A-k': The degree of absolute agreement for measurements that are averages of k independent measurements on randomly selected objects. The degree of absolute agreement for measurements that are based on k independent measurements made under the fixed levels of the column factor.

Thanks!! ]]>