USING DISCRIMINATE ANALYSIS WITH APPLIED ON BIRTHS OF PREMATURE BABY IN AL-SABEAN GENERAL HOSPITAL FOR MATERNITY AND CHILDHOOD IN THE CAPITAL SANA’A

Abstract

The current study aimed to employ the use of discriminant function analysis in classifying the new premature babies into living premature babies or dead premature babies. The study used the descriptive analytical method, the study consisted of a random sample of data on premature babies at the Statistics Center at Al-Sabean General Hospital for Maternity and Childhood for the years 2018-2020, its size reached (130) premature babies, of whom (98) lived and (32) died. The method of analyzing the discriminatory function used on the data in order to reach the classification of the new preterm babies into one of the two groups. The study reached to identify the most important independent variables in classifying the group of premature babies into one of the two groups. The first in importance is the smoking variable of the pregnant mother, followed by the variable of uterine age of the premature baby in weeks, then the variable of taking nutritional supplements during pregnancy, then the rest of the six approved variables in constructing the function of discrimination. The study also indicated the strength of the relationship between the variables included in the analysis, where the percentage of legal correlation of the discriminatory function was (0.85). The percentage of variance explained by the discriminant function through the value of the eta-square was (0.72), and it corresponds to an eigenvalue equal to (2.58). The results also indicated that the value of Wilks' Lambda amounted to (0.280), and the value of chi-squared amounted to (159.33), which is statistically significant at the level of significance (0.05). This confirms the ability of the discriminatory function to distinguish between the two groups of premature babies. Amounted classification accuracy between the two groups of premature babies (95.4%), which is a high percentage that confirms the accuracy of the classification.