Simple statistics for manual analysis of farm survey data
When conducting fontal surveys for the purpose of helping focus on-farm agronomic experiments, experience indicates that a relatively small sample of farmers is normally sufficient for a given recommendation domain or "homogeneous" group of farmers. A sample size of 40-60 farmers is normally large enough for reasonably precise estimates of variables measuring farmer practices and problems in the production of a target crop, and the cropping system in which that crop is grown (Byerlee, Collinson et al, 1980). As one adds domains, of course, minimum sample size for formal surveys increases. Nonetheless, researchers involved in on-farm research will frequently find themselves conducting and analyzing surveys with sample sizes of less than one hundred farmer's. In these cases, manual analysis of survey data can be more efficient than computer analysis. This is because there is a high "fixed cost" associated with setting-up analysis via computer. Before beginning the actual analysis, researchers must formulate a code-book that describes the codes corresponding to each possible answer for all questions, code the data, have the coding--sheets key-punched, formulate an input statement and run and edit a data listing. In the time this takes, the researcher might well have finished the entire analysis if performed manually. Analysis of survey data to help focus on-farm experiments rarely requires sophisticated procedures. Cross-tabulations and comparisons of means for sub-populations are used to test hYIX>theses on the delineation of recommendation domains. Means and simple frequencies are employed to describe the current farming system and the management of the target crop within that system, for each domain. Cross-tabulations and comparisons of means for sub-populations are then used to identify factors that limit production or income, and the interactions between these limiting factors and the farmers' current practice. The basic analytical tools used, then are: (l) means, (2) simple frequencies, (3) cross-tabulations, and (4) comparisons of the means of two sub-populations. Of these four tools, the first two require no further attention. The purpose of this note is to examine two statistics associated with cross-tabulations (chi-square) and comparisons of sub-population means (Student's t), emphasizing their manual calculation and interpretation.
Main Author: | |
---|---|
Format: | Handbook biblioteca |
Language: | English |
Published: |
CIMMYT
1981
|
Subjects: | AGRICULTURAL SCIENCES AND BIOTECHNOLOGY, CROSS-BREEDING, DATA ANALYSIS, METHODS, PRODUCTION FACTORS, SURVEYING, TRAINING, ZEA MAYS, |
Online Access: | http://hdl.handle.net/10883/854 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
dig-cimmyt-10883-854 |
---|---|
record_format |
koha |
spelling |
dig-cimmyt-10883-8542023-01-26T20:54:29Z Simple statistics for manual analysis of farm survey data Harrington, L.W. AGRICULTURAL SCIENCES AND BIOTECHNOLOGY CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS When conducting fontal surveys for the purpose of helping focus on-farm agronomic experiments, experience indicates that a relatively small sample of farmers is normally sufficient for a given recommendation domain or "homogeneous" group of farmers. A sample size of 40-60 farmers is normally large enough for reasonably precise estimates of variables measuring farmer practices and problems in the production of a target crop, and the cropping system in which that crop is grown (Byerlee, Collinson et al, 1980). As one adds domains, of course, minimum sample size for formal surveys increases. Nonetheless, researchers involved in on-farm research will frequently find themselves conducting and analyzing surveys with sample sizes of less than one hundred farmer's. In these cases, manual analysis of survey data can be more efficient than computer analysis. This is because there is a high "fixed cost" associated with setting-up analysis via computer. Before beginning the actual analysis, researchers must formulate a code-book that describes the codes corresponding to each possible answer for all questions, code the data, have the coding--sheets key-punched, formulate an input statement and run and edit a data listing. In the time this takes, the researcher might well have finished the entire analysis if performed manually. Analysis of survey data to help focus on-farm experiments rarely requires sophisticated procedures. Cross-tabulations and comparisons of means for sub-populations are used to test hYIX>theses on the delineation of recommendation domains. Means and simple frequencies are employed to describe the current farming system and the management of the target crop within that system, for each domain. Cross-tabulations and comparisons of means for sub-populations are then used to identify factors that limit production or income, and the interactions between these limiting factors and the farmers' current practice. The basic analytical tools used, then are: (l) means, (2) simple frequencies, (3) cross-tabulations, and (4) comparisons of the means of two sub-populations. Of these four tools, the first two require no further attention. The purpose of this note is to examine two statistics associated with cross-tabulations (chi-square) and comparisons of sub-population means (Student's t), emphasizing their manual calculation and interpretation. 13 pages 2012-01-06T05:06:02Z 2012-01-06T05:06:02Z 1981 Handbook http://hdl.handle.net/10883/854 English CIMMYT Economics Training Note CIMMYT manages Intellectual Assets as International Public Goods. The user is free to download, print, store and share this work. In case you want to translate or create any other derivative work and share or distribute such translation/derivative work, please contact CIMMYT-Knowledge-Center@cgiar.org indicating the work you want to use and the kind of use you intend; CIMMYT will contact you with the suitable license for that purpose. Open Access PDF Mexico CIMMYT |
institution |
CIMMYT |
collection |
DSpace |
country |
México |
countrycode |
MX |
component |
Bibliográfico |
access |
En linea |
databasecode |
dig-cimmyt |
tag |
biblioteca |
region |
America del Norte |
libraryname |
CIMMYT Library |
language |
English |
topic |
AGRICULTURAL SCIENCES AND BIOTECHNOLOGY CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS AGRICULTURAL SCIENCES AND BIOTECHNOLOGY CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS |
spellingShingle |
AGRICULTURAL SCIENCES AND BIOTECHNOLOGY CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS AGRICULTURAL SCIENCES AND BIOTECHNOLOGY CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS Harrington, L.W. Simple statistics for manual analysis of farm survey data |
description |
When conducting fontal surveys for the purpose of helping focus on-farm agronomic experiments, experience indicates that a relatively small sample of farmers is normally sufficient for a given recommendation domain or "homogeneous" group of farmers. A sample size of 40-60 farmers is normally large enough for reasonably precise estimates of variables measuring farmer practices and problems in the production of a target crop, and the cropping system in which that crop is grown (Byerlee, Collinson et al, 1980). As one adds domains, of course, minimum sample size for formal surveys increases. Nonetheless, researchers involved in on-farm research will frequently find themselves conducting and analyzing surveys with sample sizes of less than one hundred farmer's. In these cases, manual analysis of survey data can be more efficient than computer analysis. This is because there is a high "fixed cost" associated with setting-up analysis via computer. Before beginning the actual analysis, researchers must formulate a code-book that describes the codes corresponding to each possible answer for all questions, code the data, have the coding--sheets key-punched, formulate an input statement and run and edit a data listing. In the time this takes, the researcher might well have finished the entire analysis if performed manually. Analysis of survey data to help focus on-farm experiments rarely requires sophisticated procedures. Cross-tabulations and comparisons of means for sub-populations are used to test hYIX>theses on the delineation of recommendation domains. Means and simple frequencies are employed to describe the current farming system and the management of the target crop within that system, for each domain. Cross-tabulations and comparisons of means for sub-populations are then used to identify factors that limit production or income, and the interactions between these limiting factors and the farmers' current practice. The basic analytical tools used, then are: (l) means, (2) simple frequencies, (3) cross-tabulations, and (4) comparisons of the means of two sub-populations. Of these four tools, the first two require no further attention. The purpose of this note is to examine two statistics associated with cross-tabulations (chi-square) and comparisons of sub-population means (Student's t), emphasizing their manual calculation and interpretation. |
format |
Handbook |
topic_facet |
AGRICULTURAL SCIENCES AND BIOTECHNOLOGY CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS CROSS-BREEDING DATA ANALYSIS METHODS PRODUCTION FACTORS SURVEYING TRAINING ZEA MAYS |
author |
Harrington, L.W. |
author_facet |
Harrington, L.W. |
author_sort |
Harrington, L.W. |
title |
Simple statistics for manual analysis of farm survey data |
title_short |
Simple statistics for manual analysis of farm survey data |
title_full |
Simple statistics for manual analysis of farm survey data |
title_fullStr |
Simple statistics for manual analysis of farm survey data |
title_full_unstemmed |
Simple statistics for manual analysis of farm survey data |
title_sort |
simple statistics for manual analysis of farm survey data |
publisher |
CIMMYT |
publishDate |
1981 |
url |
http://hdl.handle.net/10883/854 |
work_keys_str_mv |
AT harringtonlw simplestatisticsformanualanalysisoffarmsurveydata |
_version_ |
1756373505109131264 |