Recent content by hlsmith

  1. hlsmith

    Analyzing whether a relationship changes after an event

    It will be hard to figure out unless the exposure is randomized. One thing you could do is find another population not exposed to show they did not have the same jump (interrupted time series with control group).
  2. hlsmith

    Why are so many variables in my regression significant?

    Having no idea of your context, I can point out that the estimate sizes are very small and perhaps insignificant in a business sense. A reason could be that you have a very large sample size so the model is picking up trivial effects.
  3. hlsmith

    Using the Crawford and Garthwaite (2002) modified t-test for the reverse purpose?

    I didn't really understand this sentence. Why couldn't you just do a one-sample t-test?
  4. hlsmith

    Need help regarding project

    Whenever you conduct a study, you typically conduct a sample size calculation if there is an inference you are going to test. At this part you have to have a small sample, meta-data on the population, or make an educated assumption on values of variability. So you likely need to pick one of...
  5. hlsmith

    Question in interpreting Effects Size - versus p - Very large data sets

    @obh - agreed. I was just referring to the second (super saturated) model. I would image 13% would be useful.
  6. hlsmith

    Question in interpreting Effects Size - versus p - Very large data sets

    Yes, so if I follow, your variables seem to explain a fraction of a percent of the variability in the outcome. Which I would imagine is near useless. In addition, in order to look at interaction terms you need to have solid suspicion/theory that they could exist. You can't just look at every...
  7. hlsmith

    You've asked 55 parents if they have more than one child. It turns out that 77 in 100 parents have more than one child. Compute the 99% confidence int

    Is there a reason why you wouldn't treat this as a binary variable (>1 Y/N)?
  8. hlsmith

    Still having problems...Am I missing something obvious?

    Sounds like a fun experiment. This seems like a basic study design that you could see in agriculture. I personally haven't done basic studies like this before, so won't be a great help. What kind of values did you see in your outcome? This may dictate if you are working with a Poisson or linear...
  9. hlsmith

    Election 2020

    Yeah, we will see if he selects a female running mate to procure some more votes.
  10. hlsmith

    Binary logistic regression warning message model summary help!!

    Yeah, with small samples or sparsity - Firth's is a simple and not too esoteric option.
  11. hlsmith

    Binary logistic regression warning message model summary help!!

    It sounds like the model is failing to converge. Multiple things could be going on. The most common is complete separation, meaning that in one of the groups all members are in one of the outcome groups. You can check for this. Work arounds sometimes can be, changing the number of iterations...
  12. hlsmith

    t, Z and sigma again

    Not the answer you are looking for, but you could do some simulations. Simulate a population, then draw samples from it without replacement and conduct ttests and z-tests on samples.
  13. hlsmith

    t, Z and sigma again

    Good question - I never really think about using the z-test. I feel like they teach it, but then given my field I have never used it. Also, I feel like both the t- and z- get taught, but in real life most settings require a multivariable setting. I also agree, that I don't grasp the setting...
  14. hlsmith

    Homogeneity requirements for Linear regression

    Yeah, you all say this, but I think it should still be looked at to make sure something crazy extreme isn't going on. Visualization are so important. It could also reveal that say some outcome values aren't present or if values are bounded. Or help find erroneous outliers. It can't hurt to look...
  15. hlsmith

    Hayes Process V3.3 Model 89 in SPSS V25

    I could see wanting to know the effects of each term, but that causal diagram has two subsequent moderated mediators. I question how someone would suspect that and if you need to address it, since the first would be conditionally independent of the outcome since the second one likely screens it...