I am conducting a study to assess the incidence and predictors of X disease progression from 10 X diseased patients retrospectively.
#Small sample size
#disease progression
#poisson model
#Cox proportional hazard model
Hello Melsew,
Chances are that N = 10 cases is too few to yield either much precision in any parameter estimates or sufficient statistical power for hypothesis tests.
As you mention the possibility of multiple IVs ("...predictors..."), the opportunity for overfitting a model to your data which then fails to generalize to other samples becomes more of a concern.
If you could specify the number and types of variables involved (both IV/s and DV/s), and the intended analytic method (e.g., as cases are censored by death, which possibly might not be due to the condition of interest, perhaps proportional hazards model), then I'm sure you'd get more focused advice as to how confidently you could continue with the proposed sample size.
Good luck with your work.
In a small sample size of 10 patients, it is important to consider the limitations and potential challenges associated with statistical analysis. With such a small sample, it may be difficult to draw robust and generalizable conclusions. Nevertheless, if you still wish to explore the relationship between predictors and disease progression in this limited dataset, you can consider using a simple linear regression model or descriptive statistical analysis.
It is important to acknowledge that the small sample size greatly limits the reliability and generalizability of any findings. Larger sample sizes are preferred for robust statistical analysis and to account for potential confounding factors.
In this situation, it may be more appropriate to focus on descriptive statistics, case studies, or qualitative analysis to better understand the disease progression in your small sample of patients. Additionally, you could consider combining your data with existing literature or seeking expert opinions to provide context and support your findings.
With small sample sizes you should
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis