Introduction
While determining the trends and relationships that are present in population parameters, several statistical tests and analyses are usually conducted (Sukamolson, 2006). Confidence interval (CI) is a statistical measure that is used to determine the reliability of a population parameter in a statistical test. CIs are usually made up of several values. These values are essential in statistical tests since they provide reliable estimation of an unknown parameter (Sukamolson, 2006).
While conducting a study, study hypothesis are usually designed according to the objectives and goals that are to be achieved in the end. To determine whether the parameters of the population or sample size reflect to these objectives, plausible values of the study are usually determined. However, the effectiveness of these values depends on the degree of confidence that the study wants to attain. It is as a result of this that confidence intervals are developed in a research study.
For instance, when a study is developed at 95% confidence interval, all the values of the parameter that are within this range are accepted while the ones that are outside the range are rejected. Thus, in an event whereby the plausible values of the null hypothesis fall within this confidence interval, the null hypothesis is accepted. However, if the values are implausible (outside the 95% confidence interval), the null hypothesis is rejected (Sukamolson, 2006).
However, it should be noted that there are differences between CI and hypothesis testing. As it has been stated, confidence intervals are used to show the degree of freedom (or reliability) with which a specific statistical test is to be carried at. Most statistical tests are carried out at 95% level of confidence (Sukamolson, 2006). However, to determine whether there was a significant difference between the variables that were tested, a different parameter is used. This parameter is known as the p value. Thus, the p value is used to determine whether the variables of a given study are within a desired range to accept or reject the null hypothesis. This is the main difference that exists between CIs and hypothesis testing.
Interpretation of Confidence Intervals and P Value
While conducting a study, a researcher usually determines the degree of accuracy with which he/she wants his study to be at. In normal cases, a normal distribution is used to display the graphical representation of the mean and standard deviation of the parameters of the study. This distribution is usually constructed in accordance to the margin of error that a researcher wants to encounter. For example, if the researcher wants the study to have a margin of error of 5%, he/she needs to set at + and -1.96 standard deviations for a two tailed test. If the researcher uses a one-tailed test, the margin of error shall be set at 1.64 (Myers, 2001).
However, the interpretation of the results depends on the p value of the study. The p value is the value that is used to determine the level of significance of the variables of the study (Myers, 2001). This value can be computed using statistical software packages or manually. For instance, if a researcher wants to determine the age at which the teenagers start to engage in sexual intercourse at a CI of 95% using a two-tailed tests, the p value should be less than 0.025 for the null hypothesis of the study to be accepted. If the value is more than 0.025, the null hypothesis shall be rejected.
Conclusion
Therefore, confidence intervals and hypothesis tests are used to complement each other in statistical tests.
References
Myers, R. (2001). Elements of Research: Study Design and Data Analysis. New York, NY: McGraw-Hill.
Sukamolson, S. (2006). Fundamentals of Quantitative Research. Chulalongkorn: Chulalongkorn University Press.