# Examples in SPSS and SAS for oil palm fertilizer experimental design and analysis - Part A

*This webpage contains data that were obtained in Example 2 described in pages 267-269 in the publication Oil Palm: Management for Large and Sustainable Yields.*

(Example 4, described in pages 271 of the book will be discussed in

(Example 4, described in pages 271 of the book will be discussed in

*Part B**)*

**A1 Experimental results (t FFB ha**^{-1}**)**

The results from the experiment were as follows:

N level (kg N palm^{-1}) | ||||

0 | 1 | 2 | 3 | |

Rep 1 | 27 | 29 | 32 | 26 |

Rep 2 | 26 | 31 | 30 | 27 |

Rep 3 | 25 | 27 | 28 | 28 |

**A2 Data file for SPSS and SAS**

The contents of the data file (contained in a *.dat format) is given below showing (from left to right) the plot number (plotnr), applied level of N (n), and yield (yield).

Plot number (plotnr) | Applied level of N (n) | Yield |

1 | 0 | 27 |

2 | 0 | 26 |

3 | 0 | 25 |

4 | 1 | 29 |

5 | 1 | 31 |

6 | 1 | 27 |

7 | 2 | 32 |

8 | 2 | 30 |

9 | 2 | 28 |

10 | 3 | 26 |

11 | 3 | 27 |

12 | 3 | 28 |

**A3 Command file for analysis with SPSS**

The SPSS command file (*.sps) contains the following programme for SPSS under Windows (these commands can also be given by clicking in the menu):

COMPUTE n1 = n. EXECUTE. COMPUTE n2 = n*n . EXECUTE. REGRESSION /MISSING LISTWISE /STATISTICS COEFF OUTS CI BCOV R ANOVA /CRITERIA=PIN(.05) POUT(.10) CIN(95) /NOORIGIN /DEPENDENT yield /METHOD=ENTER n1 n2 /SAVE PRED SEPRED MCIN RESID . |

**A4 Programme output for analysis with SPSS**

Regression |

**A5 Command file for analysis with SAS**

The corresponding SAS command file (*.sas) contains the following code, where the data file is called: "quadr_ex.dat":

data; infile 'c:\quadr_ex.dat'; input plotnr n yield; n1 = n . n2 = n * n; run; proc print; run; proc reg; model yield= n1 n2 /p i clm ; run; |

**A6 Programme output for analysis with SAS**

The SAS System 23:00 Wednesday, July 9, 2003 1
OBS PLOTNR N YIELD N1 N2 1 1 0 27 0 0 2 2 0 26 0 0 3 3 0 25 0 0 4 4 1 29 1 1 5 5 1 31 1 1 6 6 1 27 1 1 7 7 2 32 2 4 8 8 2 30 2 4 9 9 2 28 2 4 10 10 3 26 3 9 11 11 3 27 3 9 12 12 3 28 3 9 The SAS System 23:00 Wednesday, July 9, 2003 2 Model: MODEL1 X'X Inverse, Parameter Estimates, and SSE INTERCEP N1 N2 YIELD INTERCEP 0.3166666667 -0.35 0.0833333333 25.9 N1 -0.35 0.8166666667 -0.25 4.9 N2 0.0833333333 -0.25 0.0833333333 -1.5 YIELD 25.9 4.9 -1.5 20.6 The SAS System 23:00 Wednesday, July 9, 2003 3 Dependent Variable: YIELD Analysis of Variance Sum of Mean Source DF Squares Square F Value Prob>F Model 2 29.40000 14.70000 6.422 0.0185 Error 9 20.60000 2.28889 C Total 11 50.00000 Root MSE 1.51291 R-square 0.5880 Dep Mean 28.00000 Adj R-sq 0.4964 C.V. 5.40324 Parameter Estimates Parameter Standard T for H0: Variable DF Estimate Error Parameter=0 Prob > |T| INTERCEP 1 25.900000 0.85136057 30.422 0.0001 N1 1 4.900000 1.36720856 3.584 0.0059 N2 1 -1.500000 0.43673876 -3.435 0.0075 Dep Var Predict Std Err Lower95% Upper95% Obs YIELD Value Predict Mean Mean Residual 1 27.0000 25.9000 0.851 23.9741 27.8259 1.1000 2 26.0000 25.9000 0.851 23.9741 27.8259 0.1000 3 25.0000 25.9000 0.851 23.9741 27.8259 -0.9000 4 29.0000 29.3000 0.648 27.8346 30.7654 -0.3000 5 31.0000 29.3000 0.648 27.8346 30.7654 1.7000 6 27.0000 29.3000 0.648 27.8346 30.7654 -2.3000 7 32.0000 29.7000 0.648 28.2346 31.1654 2.3000 8 30.0000 29.7000 0.648 28.2346 31.1654 0.3000 9 28.0000 29.7000 0.648 28.2346 31.1654 -1.7000 10 26.0000 27.1000 0.851 25.1741 29.0259 -1.1000 11 27.0000 27.1000 0.851 25.1741 29.0259 -0.1000 12 28.0000 27.1000 0.851 25.1741 29.0259 0.9000 Sum of Residuals 0 Sum of Squared Residuals 20.6000 Predicted Resid SS (Press) 33.4945 |

The illustration for Example 4 from the book is discussed in Part B.

*Source: *The information on these pages were kindly provided by Dr Rob Verdooren, Statistical Advisor of Numico-Research B.V. For further information on the statistical aspects of the experiments, please email Dr Verdooren directly.