Volume 65, Issue 7 (4 2007)
Tehran Univ Med J 2007, 65(7): 24-27 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Babaee Gh, Amani F, Biglarian A, Keshavarz M. Detection of Outliers methods in medical studies. Tehran Univ Med J 2007; 65 (7) :24-27
URL: http://tumj.tums.ac.ir/article-1-753-en.html
URL: http://tumj.tums.ac.ir/article-1-753-en.html
Abstract: (5996 Views)
Background: An outlier is an observation that lies an abnormal distance from other values in a random sample from a population. Outliers sometimes deal with to abnormality in obtained results from collected data and information. known outlier data by researchers, physicians and other persons that work in medical fields and sciences is important and they must control data before getting result about outlier data, effect of them in information bias and how to remove & control to obtain minimum bias and exact data .in this paper we had trying by known technique and tests to control them and minimized the errors related to them.
Methods: This paper has been done on 30 student's height in Tarbiat Modares University that measured by meter in smoothing area. We applied some methods such as Z-test, Grub test and graphical methods to determine outliers. In this paper the advantage and disadvantage of methods were evaluated and finally compares with each other.
Results: The above tests showed that the data values 153, 110 among collected data were outliers. All of the methods showed that the above data were outliers. Calculation quartiles and intermediate quartiles showed that the observations under 125 and upper 141 were mind outliers and if the observation under 119 and upper 147 is the sever outliers. According to upper situations the amounts of 110 and 153 is the sever outliers and resulted from all methods.
Conclusion: The results showed that all methods were useful in determine outlier data and between them Quartiles were important to known severe and mild outliers. Also Grub test with p-Value is very useful to report outliers.
Methods: This paper has been done on 30 student's height in Tarbiat Modares University that measured by meter in smoothing area. We applied some methods such as Z-test, Grub test and graphical methods to determine outliers. In this paper the advantage and disadvantage of methods were evaluated and finally compares with each other.
Results: The above tests showed that the data values 153, 110 among collected data were outliers. All of the methods showed that the above data were outliers. Calculation quartiles and intermediate quartiles showed that the observations under 125 and upper 141 were mind outliers and if the observation under 119 and upper 147 is the sever outliers. According to upper situations the amounts of 110 and 153 is the sever outliers and resulted from all methods.
Conclusion: The results showed that all methods were useful in determine outlier data and between them Quartiles were important to known severe and mild outliers. Also Grub test with p-Value is very useful to report outliers.
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
