# Tag Info

0

I’m not really sure but I think you can use Kruskal Wallis for each variable because you just have 7 countries and by that you can see if there is a difference for each variable by country.

0

It is said that there is no significant number of clusters but you may try to check out the elbow method for optimal k. Plus elbow method is not available in spss. If you have sufficient information about the dataset, you can determine the number of clusters yourself. For validation, you can get the silhouette score by just putting the continuous ...

0

The second command does addresses the second hypothesis, indicating that the odds for novices using validated instruments is (2/7)/(3/13) = (2*13)/(7*13) = 1.238 times that for experienced clinicians. The odds ratio in a 2x2 table is invariant to inversion of rows and columns, as you've discovered. The first command compares the odds of users of validated ...

0

I confess I have no idea how Fisher's exact test for a $2 \times 5$ table is done in SPSS--or even whether it is available in SPSS. The original version of this test, for a $2 \times 2$ table and using the hypergeometric distribution, is easy enough to do by hand, provided the counts are small. However, here is Fisher's exact test for a table such as you ...

4

SPSS LOGISTIC does not handle sampling weights correctly for computing standard errors. If you have weights $w_i$ for each observation, SPSS will work out the loglikelihood contribution $\ell_i(\beta)$ for each observation, and maximise the weighted sum $\hat\ell(\beta) = \sum_i w_i\ell_i(\beta)$. So will R. The point estimates will agree exactly. SPSS, ...

0

If you do a linear regression using one of the emotion variables as the dependent and Age as a predictor then Age would be the independent variable. If you use the three emotion variables as the levels of a within-subjects or repeated-measures ANOVA, then emotion type would be the independent variable. In the latter case, you'd go to Analyze>General Linear ...

0

I suggest checking carefully to make sure that the data values are identical in the programs, because when I compare results for SPSS and R I get identical results. I've tried with and without ties, and with and without some missing data, and the programs always give the same values.

0

The only results for which effect size estimates are currently produced in SPSS GLM are $F$ tests, for which you can get partial $\eta$2 values. Thus in order to get effect size estimates for these comparisons, you need to get them output as $F$ tests. If the factor with three levels is one of the within-subjects factors, and you're looking at "main ...

0

This sounds like a mixed model situation. An example is shown in the Case Studies for SPSS Statistics at Using Linear Mixed Models to Fit a Random Coefficients Model. This model fits a common fixed slope and intercept across subjects, as well as random slopes and intercepts for individual subjects to see how they depart from the common values. To show the ...

0

You don't really give enough information for anyone to give a complete answer, but a skim of the paper and your question leads me to suggest using the interaction test approach in your regression model(s). That is, you add an interaction or product variable between your grouping variable and the exposure variable of interest. The null hypothesis for this ...

0

While it's certainly unusual to see a value of exactly 0 for the chi-square statistic for Little's MCAR test, it's not impossible. It probably means there's something systematic that's ensuring that the means of observed variables for each missing data pattern are the same.

Top 50 recent answers are included