Open AccessImplementation of Multivariate Analysis of Variance (MANOVA) in experiments factorial two factors (Study: Growth and development of soybean germination)
Author(s) -
Hairus Saleh,
Sri Indriati Hasanah,
Agus Subaidi
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1375/1/012013
Subject(s) - multivariate analysis of variance , factorial experiment , germination , fractional factorial design , statistics , mathematics , main effect , multivariate statistics , husk , sowing , analysis of variance , completely randomized design , factorial , horticulture , biology , botany , mathematical analysis
Soybean is a type of leguminous plant that contains many substances that are important for health, high protein, vitamins and many spread in Indonesia. The economical part of soybean plants is the seeds, which are widely used as raw materials for tofu and tempeh. Variant Analysis with Multiple Variables (Manova) is a multivariate statistical method used to analyze data from more than one variable with an interval or ratio scale, while the independent variable consists of two or more groups, where all of these variables are analyzed simultaneously or together. The purpose of this study was to obtain a linear model based on the source of diversity derived from the design used, the source of diversity and compile it into the variance table, and the main influence of the factors and the interaction effects of the combination treatment. This study is a repeat observation experiment with a two-factor factorial basic design. This kind of design is called factorial design in RAL time. The first factor is the light intensity consisting of 3 levels, namely bright, dim, and dark. The second factor of the planting medium consists of 3 levels, namely husk ash, soil, and cotton. The responses observed were growth and development of soybean germination. Observations carried out for 5 days. The number of combinations of treatments was 7. The experiment was repeated 2 times so that the experimental units were 14 units. Based on the results of the source of diversity, a linear model is formed: Y ijkl = μ + α i + β j + αβ ij + δ ijkl + ω 1 + αω i 1 + βω j 1 + αβω ij 1 + ε ijkl , where i = 1,2,3; j = 1,2,3; k = 1,2; and l = 1,2,3,4,5. Linear models are formed from total diversity. The total diversity that is the source of diversity comes from the main effect of treatment, the interaction effect of the treatment combination and the effect of the error. From the variance table, it can be concluded that the check sun rays intensity factor (A), planting media factor (B), time (C), factor A*B interaction, factor A*C interaction, factor B*C interaction, factor A*B*C interaction has a significant effect on the response observed which is indicated by the acquisition of F hitung > F tabel with a real level of 5%.