Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

A Survey on the Typical Sample Size for Testing Yield Parameter of Sugarcane

A. Ghafur
Sugarcane Research & Training Institute, lshurdi, Pabna, Bangladesh.

M. A. Mannan
Sugarcane Research & Training Institute, lshurdi, Pabna, Bangladesh.

Abstract/ Executive Summary:

A sample survey was conducted on five strains of sugarcane to determine the minimum sample size of stalk for use which could lead to the accurate estimate of average yield of sugarcane and hence per hectare yield. The results of the present study show that with the increase of sample size co-efficient of variation comes to some steady point, the length of confidence interval of the mean yield of sugarcane decreases and the error of estimation also decreases leading to better accuracy of estimation. The minimum length of confidence interval with probability 0.95 and smallest error of estimation were obtained with sample size over 50.

Keywords: Smallest sample size, Statistically precise results, Sugarcane

Location: Sugarcane Research & Training Institute, lshurdi, Pabna, Bangladesh.

Start Date: 00-00-1981

End Date: 00-00-1982

Program Area: Socio-economic and Policy

Commodity: Sugarcane

Objectives:

A survey was conducted on a few sugarcane strains to determine the smallest sample size that would produce statistically precise results.

Methodology:

The sample survey was carried out at Sugarcane Research & Training Institute in harvesting season of 1981-82. Forty eight canes were taken from each strain planted in a plot measuring 9.14m X 9.14m, where there were eight rows of canes. Six canes were taken at random from each row i.e. a random sample of forty eight canes were taken from the plot of a particular strain. These forty eight canes were then mixed and each cane was weighed separately. From the weight of these forty eight canes, eight separate random samples of sizes 6, 12, 18, 24, 30,36,42 and 48 were taken with replacement. In the same way, the survey was conducted in five plots of five strains. The data were analysed statistically for the study of (1) the variance, co-efficient of variation and the length of confidence interval of the yield parameter for the sample sizes; (2) effective number of samples for specified accuracy.

Source of Information: Bangladesh J. Sugarcane, 4: 93-96 July, 1982

Article Link (Online Journal):

Funding Source:

Total Budget (Tk.) :

Achievement/Findings:

It may be reasonably certain in the present experiment (with probability approximately equal to 0.95) that the estimate will be within 10%, 7% and 5% of the true average yield. This method of choosing the sample size is approximate for a specified desired accuracy of estimation. It is the best available and better than selecting the sample size on the basis of intuition. The percent of deviation of the estimate from the population parameter is decreasing with the increase in the size of sample, i.e. the number of stalks.

Knowledge Management: Journal