Search results

  1. spunky

    What is teh difference between a linear regression and a correlation between 2 variables?

    You can use it for predictions (like hlsmith said, the concept you're looking for is called "extrapolation") but you're almost always going to be wrong. When you extrapolate from the regression line you're making the assumption (with no evidence) that if had gathered more data the same linear...
  2. spunky

    What is teh difference between a linear regression and a correlation between 2 variables?

    The units of the regression weight would be the only difference for the case of the UNstandardized regression coefficient. If the variables are standardized then the regression coefficient and the Pearson correlation are the same.
  3. spunky

    Factor Analysis newbie seeks your help

    Yes, you're kind of vastly overlooking some very important stuff: the issue that your observations are not independent. The theory of factor analysis (and most linear models) assumes that every person on which you collect data is independent from the next one. That is obviously not the case for...
  4. spunky

    Factor Analysis newbie seeks your help

    So... if I understand this correctly the number of children are "nested" within their parents? So if Parent 1 has 3 children, then she/he will have 3 questionnaire responses (1 per child), if Parent 2 has 2 children, then 2 questionnaires and so on...?
  5. spunky

    calculating and interpreting SD from mean and SEM

    What are "weighted and unweighted bases"? I guess this depends on what you mean by a "base" and these weighting issues but if all you're doing is re-arranging the formula for the standard error of the mean so that it looks like \sigma_{\bar{{x}}}\sqrt{n} = \sigma then, sure, I don't see any...
  6. spunky


    Welcome! Thank you for joining us!
  7. spunky

    Which test is appropriate for correlating categorical and continuous variables?

    Would the point-biserial correlation or the biserial correlation be of any help?
  8. spunky

    Factor Analysis

    Principal Component Analysis and Factor Analysis are related albeit different methods. In your particular instance, because you're only interested in dimension reduction and not in some mysterious latent variable that accounts for the correlations, Principal Components makes more sense. Just...
  9. spunky

    Factor Analysis

    Well, if these are categories we're talking about as opposed to continuous variables, then Multiple Correspondance Analysis would be the correct way to go. But yeah, Factor Analysis or some other dimension-reduction technique would be appropriate. Although, in your case, I'd say Principal...
  10. spunky

    High significance of correlation coefficient with small sample size

    There is no disagreeing with el spunky! Deliver agreement or I shall deliver ABSENCE OF CAKE
  11. spunky

    High significance of correlation coefficient with small sample size

    Yes, they should be. In general, it is safe to say you should distrust **anything** that has such a small sample size. Small simulation example to exemplify: rr <- double(10000) for(i in 1:10000){ a<-rnorm(3) b<-rnorm(3) rr[i]<-cor(a,b) } > sum(abs(rr)>.9)/10000 [1] 0.2804 From a true...
  12. spunky

    Likert Type Items - Analysis

    Technically you can as long as whatever you end up with is interpretable.
  13. spunky

    Hi all

    Nice! You may be interested in combining that with some combinatronics? I know a lot of card games use discrete math of various types.
  14. spunky

    factor analysis

    Communality <- shared item variance Common factor eigen values <- number of factors on a scale Uniqueness <- variance unique to each item. Honestly, just googling each term would give you a very good introduction to these concepts.
  15. spunky

    Hi all

    Hi! Welcome to our board! I hope you find it useful. Although I haven't used them myself, I've heard the Coursera online classes for introduction to statistics are very good. Maybe that could help?
  16. spunky

    Variance of indivudial variables in multiple regression

    To answer this kind of questions you'd need to use methods related to variable importance in OLS multiple regression like Dominance Analysis or the Pratt Index. Here's a good overview of these methods: Depending on which software you're...
  17. spunky

    Skewed Population distribution.

    Not necessarily. The sample skewness is a statistic that's subject to sample-to-sample variability. The best option would be to pose a confidence interval on you value and assume the population skewness is somewhere around there.
  18. spunky

    The same SPSS models give DIFFERENT results

    I'm just trying to understand if your question could be addressed by pointing out that if you are fitting two different models, there's no reason for any of the coefficients to remain the same, unless all the predictors are perfectly uncorrelated.
  19. spunky

    The same SPSS models give DIFFERENT results

    So... do you end up with the exact same predictors in both models? Or does one model has more predictors than the other?
  20. spunky

    Invitation for a skype call

    We usually prefer for questions and conversations to be posted on here. You never know if the same questions/problems that you have are also shared by other people. Keeping a public record of this would benefit you and any potential person who comes across it.