List of recommendations

Recommendation 1

To ensure that the statistics in the main reports are presented impartially and objectively, MoJ should immediately publish all content written by the programme providers separately from the report.

Recommendation 2

To improve the accessibility of methods information, MoJ should consolidate existing information into a central methodology document that covers all stages of the JDL analysis process from start to finish.

Recommendation 3

To assure users about its approach to multiple comparisons in statistical significance testing, MoJ should strengthen its rationale for not applying multiple comparison correction, if this is deemed appropriate for the analyses. If this is the case, an explanation should be included in the report.

Recommendation 4

To demonstrate transparency about all aspects of quality, MoJ should publish a summary of the quality assurance arrangements for JDL analyses.

Recommendation 5

To aid user interpretation of the statistics, MoJ should provide more detail on the most important general caveats and limitations in each report.

Recommendation 6

To minimise the risk of misinterpretation of data comparison between intervention programmes, MoJ should remove the forest plot and any related material from the summary spreadsheet of JDL publications.