Reputation
15,181
Next tag badge:
95/100 score
22/20 answers
Badges
4 25 54
Newest
 Talkative
Impact
~284k people reached

2d
reviewed Close Back end processing of Pattern Matrix during EFA.
Jul
2
comment Exact formula Yates' correction in R
You should be subtracting one half from the absolute difference between observed & expected values.
Jul
2
comment Experimental design - main effects and error bars
If that's related to the graphic then it indeed seems the error bars show confidence intervals around the main effects. You ought to be told the confidence level - 95% is a standard choice, but others are sometimes used.
Jul
2
comment Experimental design - main effects and error bars
Refer to your software manual. They could be confidence intervals, or $\pm 1$ standard error, or prediction intervals, or something else. Saying "I know that when half of the error bar is superior in value to the effect, it means that the effect of the factor on the response is not significant" suggests confidence intervals - but how do you know that anyway?
Jul
2
reviewed Close How is final training stopped?
Jul
2
revised R binomial family with identity link
edited tags
Jul
2
comment SPSS Binary logistic regression incoherent result after excluding outliers
See also How to deal with perfect separation in logistic regression? & Logistic regression in R resulted in Hauck Donner phenomenon. Now what?.
Jul
2
revised Can quantiles be calculated for lognormal distributions?
added 17 characters in body
Jul
2
comment Error when using Irlba function in R?
Please consider whether there's a statistical aspect to this question - if it's purely programming it'll be better suited for Stack Overflow. In any case explain what you're trying to do, say what software/packages you're using, & include a minimal, complete, verifiable example.
Jul
2
comment Want to predict the top 100 students out of 1000 students - what model to use?
More detail about what you're doing & why will be needed for people to answer this question (with more than a list of predictive modelling techniques).
Jul
1
revised Are sampling distributions legitimate for inference?
fixed typo
Jul
1
revised Are sampling distributions legitimate for inference?
added 3 characters in body
Jul
1
comment Multivariate OLS - Partialling Out
"[....] the two explanatory variables on one another and thereby lose the variation that both explanatory variables explain mutually. In the next step I take the residuals and regress GDP growth on them. In my opinion in this case it is very likely that SSR increases, because the variation in GDP growth that is explained by both unemployment rates at the same time is not being taken into account anymore. I am sure I must be somehow wrong with that, but I don't understand where my mistake is." - I converted this response to a comment & it got truncated.
Jul
1
comment logistic regression in r with many predictors
@EdM: That's odd - the only answer I can see there is my own. But I think I see your point now: there are other ways to represent the effect of a $k$-level ordinal predictor than $\beta x$ where $ x= 1 \ldots k$.
Jul
1
comment logistic regression in r with many predictors
@costebk08: Could you edit your question to explain your situation? - it would seem that it precludes more than one sensible approach to analysis of this data. (BTW, why to use a generalized linear model rather than OLS is explained here.)
Jul
1
comment Downside to scaling and centering?
See When should you center your data & when should you standardize? - the answers there might help you to focus this question.
Jul
1
reviewed Close Summative likert scale
Jul
1
revised What value does one set, in general, for the null hypothesis of β0 in a linear regression model?
added 173 characters in body
Jul
1
answered What value does one set, in general, for the null hypothesis of β0 in a linear regression model?
Jul
1
revised What value does one set, in general, for the null hypothesis of β0 in a linear regression model?
fixed typo