<<Back

Analysis Using SAS 

Analysis Using SPSS   

For performing analysis, input the data in the following format. 

{Here the two factors are termed as N and S. It may however, be noted that one can retain the same name or can code in any other fashion}.

·         Open Data editor: Start → All Programs → SPSS for Windows → SPSS 15.0/ SPSS13.0/  SPSS10.0

·         Enter data in SPSS Data Editor. There are two views in SPSS Data Editor. In variable view, one can define the name of variables and variable type’s string or numeric and data view gives the spreadsheet in which data pertaining to variables may be entered in respective columns. In the present case, we enter data in numeric format.

• To fit a second order response surface one has to generate all the squares and cross products of input factors. For this the following steps are used

• In the data editor window select  Transform → Compute Variable

            This selection displays the following screen.

• In the Compute Variable dialog box put the new variable name in the Target Variable and the expression to be calculated in the Numeric Expression

·         Click OK.

·         Similarly S*S and N*S can be generated.

• Once the data generation is complete, Choose Analyze from the Menu Bar. Now select Analyze → Regression → Linear.

• This selection displays the following screen

·    Select yield and send it to the Dependent Variable N, S, NN, NS and SS may be selected  for Independent(s) box. After doing these the dialog box should be like this

       ·      Click OK

Once the results are obtained, then one has to perform canonical analysis, i.e. to obtain the co-ordinates of the stationary point, nature of the stationary point and value of the dependent variable at the stationary point. For this, the procedure is not straightforward as in SAS. One can use the following syntax, for the canonical analysis.

/*First enter matrix B, which has diagonal elements as

/*the estimated regression coefficients of pure quadratic terms and off diagonal elements as

/*half of the estimated regression coefficients of cross product terms

/*Compute the co-ordinates of Stationary point x0= -0.5*Binv*bL, where bL is the vector

/*of estimated regression coefficients of linear terms

/*the stationary point is a point of maxima if the eigenvalues of B are -ve;

                             /*  is a point of minima if the eigenvalues of B are +ve;

                               /*is a saddle point if some values are positive and other negative*/

/*compute y-pred at stationary point

matrix.

Compute B = {-0.175, -0.004;

                     -0.004, -0.179}.

Call eigen(B, Aeigvec, aeigval).

compute bL={40.906; 19.226}.

compute Binv=inv(B).

compute x0 = -0.5*Binv*bL.

compute x0t=transpos(x0).

compute b0={4266.250}.

compute yx0=b0+x0t*bL+x0t*B*x0.

Print B.

Print bL.

Print Aeigval.

Print Aeigvec.

Print x0.

Print yx0.

end matrix.

Data File

Syntax File

Result File

<<Back

   Analysis Using SAS                                                 Analysis Using SPSS                                       

Home Descriptive Statistics  Tests of Significance Correlation and Regression Completely Randomised Design  RCB Design  

Incomplete Block Design  Resolvable Block Design  Augmented Design  Latin Square Design Factorial RCB Design  

Partially Confounded Design Factorial Experiment with Extra Treatments Split Plot Design Strip Plot Design 

Response Surface Design Cross Over Design  Analysis of Covariance Diagnostics and Remedial Measures 

Principal Component Analysis Cluster Analysis Groups of Experiments  Non-Linear Models  

Contact Us  

Copyright        Disclaimer        How to Quote this page        Report Error        Comments/suggestions 

For exposure on SAS, SPSS, 
MINITAB, SYSTAT and  
MS-EXCEL for analysis of data from designed experiments:

 Please see Module I of Electronic Book II: Advances in Data Analytical Techniques

available at Design Resource Server (www.iasri.res.in/design)