Role of p-values and confidence intervals within epidemiological research
| Order Instructions:
Week 6 Epidemiologists control for variables such as confounding and random error by carefully developing research studies. Although there are not guarantees to eliminate or reduce all possible errors, it is important to minimize their effects. For this Discussion, review your readings from this week. Then addressing the following: |
SAMPLE ANSWER
Role of p-values and confidence intervals
Confidence intervals and p-values are reported in almost every epidemiological research and are used in interpreting statistical analysis results. These measures are used by medical investigators and researchers to answer such questions as: “are the results significant?” or “is the hypothesis accepted?” Epidemiological researchers do not have to worry about testing the significance of results when carrying out statistical researches because they can depend on the p-value and the confidence interval. The p-value is the probability of a value that is similarly extreme or even more extreme as the one in the study if the hypothesis is true. Confidence interval on the other hand can be defined as is an array of values within which there is reasonable confidence that the parameter of the population lies. The reporting of confidence intervals and p-values basically follows that testing of hypothesis or significance, (Bland & Peacock, 2012).
For instance, when testing a hypothesis using the p-value, the particular cutoff or level of significance (conventionally 0.05) is used to test whether values are significant or not; those less are significant, while those above are not. For instance in the study of differences in low birth weight prevalence between singletons and multiple pregnancies, the p value can be used to “endorse” the hypothesis that there is a difference. For the confidence interval, consider a test for difference in mean sugar reduction between a standard hypoglycemic and a new drug. The null hypothesis in this case should state that there is no difference in the blood sugar reduction mean; that can be accepted or rejected, (Gardner& Altman, 2013).
P-values and confidence intervals tend to decrease in size with an increase in sample size unless the null hypothesis is true. As the sample size increases, for the case of the p-value, there is an increased certainty on where the proportion mean might be, and hence large samples are more consistent with smaller ranges of possible population values. For the confidence interval, a larger sample size means a decreased error margin. The effect this has on interpretation of results is that larger samples report small marginal errors and the certainty of finding the true population parameter increases, hence the certainty of the results increases too, (Houle, 2007).
References
Bland M, Peacock J. (2012). Interpreting statistics with confidence. The Obstetrician and Gynaecologist;4:176–180
Gardner MJ, Altman DG. (2013). Confidence intervals rather than P-values: estimation rather than hypothesis testing. Br Med J. 292:746–750
Houle TT. (2007). Importance of effect sizes for the accumulation of knowledge. Anesthesiology. 106:415–417
We can write this or a similar paper for you! Simply fill the order form!
