in case of cross sectional research studies designed with the objective to determine prevalence or incidence .... do we need any statistical test to apply.
secondly do need any statistical test to apply in case of cross sectional studies fro associated factors of any disease or condition?
if yes then what test will be appropriate?
indications?
advantages?
disadvantages?
Shakil
You should seek help from a biostatistician or epidemiologist at your university, or rad a good textbook, as you clearly do not understand even the basic concepts.
I think Tao Bu had answered the first part of your question. I can add to it, the advantage of calculating the 95% CI, is that you can calculate a range, within which the true value of the prevalence is likely to lie. True prevalence means the prevalence in the population from the sample was drawn.
Regarding the association between the factors and the prevalence: You can draw a table containing the factors (e.g age) to be put to the left side (rows), and the prevalence at the top of the table (column). In this case you have to calculate the row % and calculate the prevalence rate among males for e.g and compare it with prevalence among females. In this case you can use the Chi square test (or Fisher's exact test) in order to calculate the p value.
Just want to add that incidence can be calculated only with cohort study design. You can calculate the confidence interval for the proportion as advised above.
Incidences or prevalences can be determined on the occasion of descriptive studies . A descriptive cross-sectional study is the opportunity to determine the prevalence and not the incidence . You can determine not only the overall prevalence (with a confidence interval) but also specific prevalences ( by gender , age ...) that you can compare with statistical tests (chi-square , Fisher's exact test .. . ) using the free software OpenEpi ( www.openepi.com ) .
As several colleagues have specified above, you can only estimate incidence rate in cohort studies. In cross-sectional designs you can estimate the prevalence or estimate the incidence proportion for short duration diseases. It is highly indicated the use of a test (Chi-square, Fisher exact test) if you want to make comparisons between groups and with other data sets. Besides, according to the study design the calculation of confidence intervals (95% usually) can be appropriate to describe the measure variation or the range within the measure of interest is located.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis