Agricultural Sciences, 2016, 7, 521-530
Published Online August 2016 in SciRes. http://www.scirp.org/journal/as
http://dx.doi.org/10.4236/as.2016.78052
How to cite this paper: Iguodala, E.A., Thomas, E.J., Edokpayi, A. and Titilola, O. (2016) A Mixed Model Analysis of a Ferti-
lizer Experiment on Oil Palm in Nigeria. Agricultural Sciences, 7, 521-530. http://dx.doi.org/10.4236/as.2016.78052
A Mixed Model Analysis of a Fertilizer
Experiment on Oil Palm in Nigeria
Edwin A. Iguodala
1
, Eghwerido Joseph Thomas
2*
, Austin Edokpayi
1
, Obilade Titilola
3
1
Nigerian Institute for Oil Palm Research, Benin City, Nigeria
2
Department of Mathematics and Computer Science, Federal University of Petroleum Resources,
Effurun, Nigeria
3
Department of Mathematics, Obafemi Awolowo University, Ile-Ife, Nigeria
Received 26 May 2016; accepted 12 August 2016; published 16 August 2016
Copyright © 2016 by authors and Scientific Research Publishing Inc.
This work is licensed under the Creative Commons Attribution International License (CC BY).
http://creativecommons.org/licenses/by/4.0/
Abstract
Mixed model analysis procedure was used to analyze the effect of fertilizer application on the
Fresh Fruit Bunch (FFB) yield of oil palm. This was with a view to achieve the most appropriate
and a robust model for analyzing yield response for fertilizer application in oil palm. In this study,
a mixed model analysis procedure was used to analyze yield data obtained from a fertilizer trial
conducted between 1997 and 2005. In mixed effect model, replicates and years were used as block.
In contrast the fixed effect ANOVA model usually lumped up replicates and years as a random er-
ror. In the model replicates were used as block with no block interaction, replicates as block with
allowance for block-fertilizer interaction, years as block with allowance for block-fertilizer inte-
raction, and years and replicates as block with allowance for year fertilizer and replicate-fertilizer
interaction. Mixed model theory was also used to provide the explicit description of the design
matrices in the models. Also, hypotheses relevant to each model were formulated and used to test
for specific effects in the models such as, fixed part, random part and interacting parts using ap-
propriate error terms as determined by the derived Expected Mean Squares (EMS). The results
revealed that at 5% significant level (p < 0.05), the combination of Potassium (K) at 3.5 kg and
magnesium (Mg) at 1.7 kg was sufficient for bunch yield of oil palm as the effect of fertilizer appli-
cation was significant in the interactions of K and Mg due to treatment.
Keywords
Fixed Model, Random Model, Mixed Model, Oil Palm, Fertilizer Trial
*
Corresponding author.