Skip to main content

Table 5 Recommendations for reporting practice for initial data analyses

From: Hidden analyses: a review of reporting practice and recommendations for more transparent reporting of initial data analyses

 Current reporting practiceRecommendations for improved reporting practice
1Information on IDA is sparse and may suffer from selective reportingFull reporting of relevant results as supplementary material and reporting of all results with impact on analysis/interpretation in the paper
2Information on IDA can be found in all sections of a paper.• IDA methodology to be described in Methods;
• IDA results to be described in Methods or Results;
• Impact of IDA on interpretation to be described in discussion.
3Distinction between pre-planned decisions and IDA-driven decisions are unclear.Pre-planned decisions should be reported in Methods; IDA driven alterations of the analysis plan should be reported with motivation in Methods.
4Characteristics of participants are listed without comments.Participants’ characteristics should be checked for consistency with expectations and for potential impact on analysis and interpretation. At a minimum a statement should be included to confirm no violated expectations.
5Reporting on missingness is incomplete.Full reporting of missingness, e.g. a flow chart describes unit missingness and a table for item missingness of variables
6Associations among variables are not reported.Associations not involving the research question but with potential impact on interpretation of results should be reported