Resumo

CASTANEDA, Javier A  e  FABIAN GIL, Jacky. A Look at Confidence Intervals in Research . rev.colomb.psiquiatr. [online]. 2004, vol.33, n.2, pp.193-201. ISSN 0034-7450.

Research findings frequently rise from analysis of a sample and it is necessary to infer the conclusions for the population where the sample elements come from. Statistical methods such as confidence intervals and hypothesis testing allow inferences to be made from a sample to a specific population. Notably over the last 30 years, medical scientific publications have increasingly needed to report research results using measures that specify the degree of precision and uncertainty when conclusions are generalized to a population. Thus, confidence intervals have become the most common methodology used to report scientific results. This paper shows the philosophy behind the construction of confidence intervals as well as the presentation and discussion of their elements beside an explanation of the correct way to interpret them.

Palavras-chave : estimation; confidence level; sampling error; precision.