Recent content by fed2

  1. fed2

    HELP deciding which ANOVA to use

    i think probably the repeated measures model is better. you maybe could also run two t-tests, one at each time point. What is the 'way' in your one-way ANOVA, seems not so clear to me.
  2. fed2

    Experiment degsin

    I think you are right, it says 6 hospitals x 4 treatments = 24 people. To me it seems like the question already tells you the most important aspect of the experiment, that the 4 treatments are all given at each of the 4 hospitals, which would be a complete block as you mentioned above. I guess...
  3. fed2

    independent sample t test

    the lines look suspiciously parallel...good thinking though always good to see some box plots.
  4. fed2

    Is multidimensional ROC Curve Analysis a thing?

    For multiple endpoints, if they aren't ordinal, you end up with 3 dimensions to the ROC rather than 2 (sens and spec) this is an example https://onlinelibrary.wiley.com/doi/10.1002/9781118269787.ch3
  5. fed2

    Leniency of assessors

    you need the ratings per consultant I would think.
  6. fed2

    Is multidimensional ROC Curve Analysis a thing?

    for many predictors most logit regression packages compute the ROC curve based on the propensity scores, ie the predicted probablity of event. There are certain theories which say this is the best way as i recall. For multiple outcomes, this would lead to more than 2 states, ie a tri-state...
  7. fed2

    Help interpret an event study methodology used in a research paper

    try drawing a mind map, it could help determine if boxplots are relevant.
  8. fed2

    Does population spread affect all possible population parameters confidence interval?

    id don't know if it helps, but your line of questioning sounds a bit like 'ancillary statistics'? maybe looking at that subject will help you in solidifying your ideas about this.
  9. fed2

    Sample size calculation when comparing two coefficients of variation

    is the measurement continuous, count, normal, lognormal or what? Need to have some idea of that. I'd suggest that this will work out best under log-normal model.
  10. fed2

    Repeated measures ANOVA - degrees of freedom

    10 = (16 )- (2 + 2 + 1) - 1 = n - k - 1, like in a two-way non-repeated measures. ie (n-1) - any used up by the treatment, gender and interaction terms. denominator degrees of freedom in this type of setting is weird and debatable.
  11. fed2

    Repeated measures ANOVA - degrees of freedom

    according tot he docs it is 'split plot'.https://www.stat.purdue.edu/~zhanghao/STAT514/handout/SplitPlot.pdf The treatment is (#treatments - 1), gender is 1 (number of biological genders - 1), and interaction is (#treatments - 1) * number of biological genders - 1). The reason 'why' these are...
  12. fed2

    Repeated measures ANOVA - degrees of freedom

    The structure of the table leads me to believe it is apply a 'guyhoueser' correction for sphericity to obtain the dfs. https://en.wikipedia.org/wiki/Greenhouse%E2%80%93Geisser_correction#:~:text=is%20a%20statistical%20method%20of,and%20Seymour%20Geisser%20in%201959.
  13. fed2

    How to compare ratings from different sites?

    do you have to use excel or do you have stats software? How common is it for some sites not to have ratings for a given pair of contractors?