Analysis of Data from Designed Experiments

Design Resources Server

	Analysis of Data from Designed Experiments
	Augmented Design

IASRI

Home

<<Back

Analysis Using SAS

Analysis Using SPSS

Analysis Using SPAD

To test the equality of check variety effects, accession effects and to compare accessions with check varieties, treatment contrast analysis is to be performed, therefore, for ease of understanding, recode the accession number and check variety as follows:

C-1	1	IC-042408	21	IC-082326	41
C-2	2	IC-042458	22	IC-082330	42
C-3	3	IC-060218	23	IC-082335	43
C-4	4	IC-060221	24	IC-082336	44
IC-028532	5	IC-060997	25	IC-082338	45
IC-028661	6	IC-063947	26	IC-082343	46
IC-028696	7	IC-066518	27	IC-082351	47
IC-028741	8	IC-073207	28	IC-082352	48
IC-028764	9	IC-073214	29	IC-082362	49
IC-028794	10	IC-073491	30	IC-104601	50
IC-028835	11	IC-073493	31	IC-104604	51
IC-028843	12	IC-079002	32	IC-104607	52
IC-028847	13	IC-079007	33	IC-104609	53
IC-036871	14	IC-079008	34	IC-104610	54
IC-036875	15	IC-079027	35	IC-104611	55
IC-036882	16	IC-079034	36	IC-104612	56
IC-036884	17	IC-079037	37	IC-104613	57
IC-036885	18	IC-079047	38	IC-104614	58
IC-041405	19	IC-079048	39
IC-042343	20	IC-079050	40

Here C-1, C-2, C-3 and C-4 are checks and IC-# is accession.

Analysis of data through an augmented block design through SPAD requires following steps:

Create a data file in SPAD format(For creation of data file in a specified format, the treatments are renumbered as 1, 2, ..., u, u + 1, ..., u + w. Here first u treatments are the control treatment(s) and u+1, ..., u + w are the test treatments. Data file contains at least three columns, first column represents block number, second column represents treatment number and third column consists of observed value of character. If there are more than one character to be analyzed, then these can be entered in fourth column onwards. There is no limitation on the number of characters present in the file. All these data values must be separated by a SPACE or a TAB.), and open it in SPAD Editor. The format of the data file for the example is given as

Select the sub-option Analyze Block Design from menu Option Augmented Design. This will display a dialog box for specifying the character to analyze. ( This box will only appear if data file has more than one character.)

Once data file is prepared and opened in the SPAD window, execute analysis module from menu by selecting Option Analyze Block Design. As there is only one character to be analyzed, therefore, a click on the Analyze Block Design option displays the analysis consisting of ANOVA tables (Block Adjusted and Treatments Adjusted), R², Coefficient of Variation, Root MSE, General Mean and adjusted treatment means (see fig below)

After this analysis, experimenter can also carry out the partitioning of treatment sum of squares into various components of interest viz. (i) among test treatments, (ii) among control treatments and (iii) among test treatments vs control treatments. These options are available in sub-option Contrast Analysis on menu Option Augmented Design.

If the data is generated from an augmented design in which each of the control treatments appear equally often in all the blocks, then the option Augmented CB design can be used for obtaining partitioned sum of squares and critical differences for performing all possible paired treatment comparisons. In this case, it is an augmented randomized complete block design, therefore, one can select the option augmented CB design.

In the Dialog box define the Total controls present in the data set.

This will display Sum of Squares, Mean square, F-Cal and Prob>F for (i) among test treatments, (ii) among control treatments and (iii) among test treatments vs control treatments contrasts. Four Critical Differences (CD) will also be listed at 1% and 5% level of significance.