Institutional Research Offices Editing and Proofreading Services

Two reports cross a vice-chancellor's desk in the same week. One says the completion rate is eighty-four per cent and the other says seventy-one. Both were produced by the institution, both are correct, and the difference is that one counts students who transferred out as non-completers and the other does not. Nobody in the room can explain that, because the definition lives in a spreadsheet somebody built in 2019.

We edit what institutional research and planning offices produce — metric and indicator definitions and their documentation, data dictionaries and their maintenance, statutory and regulatory return documentation, survey instruments and their reporting, dashboard and report commentary, benchmarking and comparator analysis, enrolment and retention analysis narratives, programme review and viability reporting, submissions to rankings and external bodies, and briefing notes for senior committees. Our editors work on the definition behind an institutional number.

The metric definition is what makes institutional data usable rather than contested, and its failure is publishing a figure without its denominator. Every institutional number is a numerator, a denominator, a population and a date, and disagreements almost always turn out to be about one of the four. We work through these so each metric states its numerator and denominator explicitly, with the inclusions and exclusions that make them up, since "completion rate" is not a definition and "the proportion of the September 2020 full-time first-degree entrant cohort, excluding students who transferred to another provider, who were awarded a qualification within four years" is; so the census date is named, given that a headcount on the first day of term and one in week four differ by hundreds and both get called enrolment; so the treatment of the awkward categories is stated — part-time students, transfers in and out, interruptions, deferrals, franchise and partner provision — as these are where two correct figures diverge; so any figure reported externally is shown with the external definition alongside the internal one where they differ, because a governing body comparing the two needs to know why; so the definition's owner and its last review date are recorded; so changes to a definition are flagged in any time series, since an unannounced definitional change looks like a trend; and so the commentary says what the number does not measure. Definitions written this way stop the argument moving to the data.

Everything you send is treated in confidence, including institutional data, returns and committee papers. We are editors rather than statisticians or planning specialists, and we offer no view on definitions, methodology or institutional performance. What we can do is make every number carry its denominator.