Two-Way Multivariate Analysis of Variance (MANOVA) of the Effect of Parity on Body Mass Index and Body Surface Area of Pregnant Women
Inamete, Emem Ndah Happiness *
Department of Statistics Technology, Federal Polytechnic of Oil and Gas Bonny, Rivers State, Nigeria.
Biu, O. Emmanuel
Department of Mathematics and Statistics, University of Port Harcourt, Rivers State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
Aims: In the determination of the effect of Parity on Body Mass Index (BMI) and Body Surface Area (BSA).
Study Design: Pregnant Women Body Mass Index (BMI) and Body Surface Area (BSA) measurements, Number were used in this study.
Place and Duration of Study: The data on the various measurements on pregnant women were collected from the Radiology department of University of Port Harcourt Teaching Hospital.
Methodology: The Two-way Classification Multivariate Analysis of Variance (MANOVA) was employed to obtain the MANOVA test statistics for Wilk’s, Pillai, Lawley-Hotelling and Roy’s Largest Root.
Results: Results showed that the MANOVA test statistics for Wilk’s statistics showed no significant effect for the Parity level categorized on BMI (Factor A) and the Parity level categorized on BMI and BSA (Factor AB), while Lawley-Hotelling Root test showed no significant effect for the effect of the Parity level categorized on BSA (Factor B). However, two MANOVA test statistics (Pillai and Roy’s Largest Root statistics) gave significant effect on BMI or BSA or both (or all Factors A, B and AB). These results confirmed that the effect of the Parity level has a significant effect on BMI or BSA or both in terms of the categories (normal, overweight and obese). These results are in agreement with known results by medical practitioners.
Keywords: Multivariate Analysis of Variance (MANOVA), parity level, two-way classification, body mass index, body surface area.