Use of statistical evidence in some recent issues of DSIR agricultural journals

Comments are made on the presentation of some of the statistical results in recent issues of the 2 DSIR agricultural journals. Common problems are the use of multiple range tests for structured data, the failure to transform before analysis, the over‐emphasis on significance testing, and the omission of standard errors. We discuss briefly the role of statistics in scientific research. A checklist is given of points which need consideration when the results of statistical analyses are published.