![Completely randomized design (CRD)
•All experimental units are considered the same and no division or grouping
among them exist.
Example: Suppose there are 4 treatments and 20 experimental units,
Treatment 1 is replicated, say 3 times and is given to 3 experimental units
Treatment 2 is replicated, say 5 times and is given to 5 experimental units
Treatment 3 is replicated, say 6 times and is given to 6 experimental units
Treatment 4 is replicated [20-(6+5+3)=]6 times and is given to the remaining 6
experimental units.
No local control measure is provided](https://image.slidesharecdn.com/jkmanan-210806094144/85/Experimental-designs-and-data-analysis-in-the-field-of-Agronomy-science-by-making-use-of-free-software-available-online-4-320.jpg)
Experimental designs and data analysis in the field of Agronomy science by making use of free software available online
- 1. Experimental designs and data analysis in the field of Agronomy science by making use of free software available online By Dr Jatinder Manan
- 2. Principles of experimental design: There are three basic principles of design (i) Randomization (ii) Replication (iii) Local control (i) Randomization Every experimental unit has an equal chance of receiving each treatment (ii) Replication Replication provides more observations when the same treatment is used, so it increases precision (iii) Local control (error control) The replication is used with local control to reduce the experimental error
- 3. Types of trials mainly used by the extension workers 1. Completely randomized design 2. Randomized block design 3. Repeated trials 4. Surveys
- 4. Completely randomized design (CRD) •All experimental units are considered the same and no division or grouping among them exist. Example: Suppose there are 4 treatments and 20 experimental units, Treatment 1 is replicated, say 3 times and is given to 3 experimental units Treatment 2 is replicated, say 5 times and is given to 5 experimental units Treatment 3 is replicated, say 6 times and is given to 6 experimental units Treatment 4 is replicated [20-(6+5+3)=]6 times and is given to the remaining 6 experimental units. No local control measure is provided
- 5. Randomized Block Design Example: Suppose there are 7 treatments corresponding to 7 levels of a factor to be included in 4 blocks. So one possible layout of the assignment of 7 treatments to 4 different blocks in an RBD is as follows Block 1 T2 T7 T3 T5 T1 T4 T6 Block 2 T1 T6 T7 T4 T5 T3 T2 Block 3 T7 T5 T1 T6 T4 T2 T3 Block 4 T4 T1 T6 T5 T2 T3 T7
- 6. Repeated trials Tests done with the same conditions and parameters as a previous one by the same researcher(s). When an individual (or team) runs a test again for more data to improve the statistical measures, it is a “repeated trial”. When a study is undertaken by a different individual to verify the results of the first one it is called a “replication”. Repeated trials are used within a study to verify the results and enhance the statistical measures. Replication is used to verify the findings of the original group by researchers outside of the original group.
- 7. Repeated trials: In general, we had repeated trials and not replicated trials, so scientists use repeated plots as their replications and do the analysis as “one factor analysis” and reach to certain conclusion, than wasting data without analyzing.
- 8. Surveys There is a logical sequence to producing a good questionnaire. The process of designing good questionnaires can be divided into six steps as follows: •Determining Data Needs. It is first important to determine why the survey is required (i.e., justification) and, therefore, the objectives of the survey, •Determining Question Content. Three important issues to address with respect to question content are: A. Appropriate identification information and variables B. Variables need to be included that enable the sample to be classified or stratified appropriately. C. Questions must use terms and units of measure (e.g., for weight, area, distance, etc.) with which the farmers in the area are familiar
- 9. •Determining Question Format • Determining the Wording of Questions •Deciding on Question Sequence •Physical Layout and Length. If it takes longer, the questionnaire should be divided and administered on separate days. •Pretesting and Revision. After the initial design of the survey, it is worthwhile to ask interested and knowledgeable individuals for comments and suggestions for improvements.
- 10. Online tool for analysis - OPSTAT The link for this software is: http://14.139.232.166/opstat/
- 14. Space for data entry
- 15. Example: 3 replications 4 treatments Parameter to analyze: Yield Things to remember: 1. Don’t use serial numbers 2. Do one parameter at a time to decrease confusion 3. Enter data in excel and copy from there 4. Don’t go for alignment of data in the space given for data entry (online)
- 16. Data entry in excel sheet Replications : A to C Treatments: 1 to 4
- 17. Copy - Data entry in excel sheet
- 18. Paste – Data in space given & click submit
- 19. New screen visible to enter parameters & enter Analyse
- 20. New window opens as follows:
- 21. CD values & Mean values are here
- 22. Critical Difference It is used to compare means of different treatments that have an equal number of replications. Generally, we take means and CD for interpretation of our data with above data, if CD is 1.1 that means: Treatment 1 is significantly superior to all other treatments because difference between treatment 1 and other treatments is more than 1.1
- 23. Surveys It can be worked with the idea that, computer analyze only the numeric values. •Convert the survey collected from the farmers into numeric values as 1, 2, 3 … •Take one theme of positive values with higher values Can use excel sheet for the analysis of data as correlation values.
- 24. Analysis can be done in excel Age values given Education values given 25-35 0 10 0 35-45 1 12 1 above 45 2 bsc 2 Farmer response Age Education 0 2 1 1 2 0 0 2 2 0 1 1 2 0 1 2 2 0 1 1 2 0
- 25. To apply any formula in excel Go to Formulas bar and then click insert function In search - write Correlation and click Go Below you see correl, select it and click Ok
- 26. New window will open, select all values of 1st parameter for array 1 and similarly for 2nd and click Ok.
- 27. The result shows the value of correlation: -0.939 means age and education were highly negatively correlated with each other The values of correlation varies between -1 to +1, as, highly negatively and positively correlated.
- 28. To apply Regression online Again we need to go to OPSTAT software Regression shows dependency of one factor on the other given factors: age values given Education values given Adoption values given 25-35 0 10 0 no 0 35-45 1 12 1 yes 1 above 45 2 bsc 2 Farmer response Age Education adoption of new variety 0 2 0 1 1 1 2 0 0 0 2 0 2 0 1 1 1 0 2 0 0 1 2 1 2 0 0 1 1 1 2 0 1
- 29. Go to OPSTAT Select correlation and regression analysis Insert all three parameters in space given and click submit, New window will open
- 30. Result window will open as: The correlation values of variable 2 is same as in excel The correlation values of variable 3 is non significant R square value is 0.8825 for variable 1 against variable 2 & 3 The variable 2 is significant at 0.00002 i.e. more than 0.005
- 31. Site to undergo basic knowledge of statistics: http://apps.iasri.res.in/design/ Login to IASRI site for online survey analysis: http://cabgrid.res.in/ssda2/Default.aspx