2

I have collected data, there are 300 non-injury and only 17 injury. Four categorical variables are significant according to Chi-squire, then I used Multiple logistic regression for significant variables. Three of them are significant again. Does it make any sense? I would like to know whether can I use Multiple logistic regression because only 17 respondent had injured from 317 of the respondents. I used SPSS to analysis of data.

If I can not run it what should I have to do? There is any way to solve it. Is there any software better than SPSS for this situation, for example STATA software or SAS......

Robert Long
  • 53,316
  • 10
  • 84
  • 148
shalaw nawprdany
  • 23
  • 2

2 Answers2

3

Yes, it is possible to run logistic regression with 317 observations where 17 are cases. Due to small sample bias, it would be a good idea to compare your results with exact logistic regression which does not depend on asymptotic results. In Stata you can use -exlogistic-.

Robert Long
  • 53,316
  • 10
  • 84
  • 148
3

Although it may be possible to run a multiple logistic regression on your data, you should be very wary of the results. A major danger that you face is that your model, however well it might fit your present data set, will not generalize to your population of interest. The issues in having such a low number of cases to predictors is covered on this Cross Validated page, with several useful links from that page.

Community
  • 1
EdM
  • 57,766
  • 7
  • 66
  • 187