03 / Manage

Data Management

Preserve the original data, document important decisions, and create an analysis ready dataset.

Immutable raw on top. Derived files underneath.

Shelved volumes kept in order.

What this is

Data Management is the process that turns collected or received data into a structured, documented, and analysis ready dataset.

It covers the work between the original data and the analytical file. This includes checking completeness, missingness, ranges, duplicates, coding, labels, skip patterns, inconsistencies, variable structures, and other issues that can affect analysis.

Good data management also means knowing what happened to the data. The original dataset should remain preserved, while cleaning and transformation take place in clearly identified working files. Decisions should be documented so that the path from the original data to the final analytical dataset can be understood and, where necessary, reproduced.

When you need it

You need Data Management when you already have data from AcadStat, another research team, an organisation, a previous study, a survey platform, a monitoring programme, or another source, and that data needs to be prepared for analysis.

You may also need it when an analytical dataset already exists but the preparation process is unclear, undocumented, inconsistent, or difficult to reproduce.

What AcadStat delivers

AcadStat preserves the original dataset as received and performs data preparation through controlled working files.

Depending on the project, this can include data cleaning, coding, recoding, variable construction, consistency checks, missing data assessment, duplicate review, data restructuring, merging of datasets, documentation, and preparation of the final analytical file.

You receive an analysis ready dataset together with the relevant codebook and documentation of important data preparation decisions.

The result is not simply cleaner data. It is data whose structure and preparation can be understood.

Tabulated papers and records ready for audit.

How the work proceeds

AcadStat stores the original dataset as received. That file remains preserved. Cleaning, coding, recoding, and variable construction take place in named working files. Later versions are stored separately. The original data are not overwritten.

The data are checked for completeness, missingness, ranges, duplicates, labels, skip patterns, coding, and consistency between related items. Datasets can be merged or restructured when the analysis file requires it.

Important decisions are documented: how a derived variable was built, how a conflict was resolved, how missing values were treated. The path from the original data to the final analytical dataset is written so it can be understood and, where necessary, reproduced.

You receive an analysis ready dataset, a codebook, and the notes that record those decisions. Documents are versioned.

If an analytical file already exists, AcadStat reconstructs the preparation process as far as the materials allow, then documents what can be verified.

You need Data Management when you already have data from AcadStat, another research team, an organisation, a previous study, a survey platform, a monitoring programme, or another source, and that data needs to be prepared for analysis.

Tools we work with, not partnerships: REDCap, KoboToolbox, ODK, SPSS, Stata, R, Python, Power BI, NVivo.

Next step

Name the question. We will define the work.

Tell us what you are trying to investigate, understand, measure, evaluate, or decide. AcadStat can help determine which part of the research process you need support with, whether that is a specific service or a complete research project managed from beginning to end.

Start a Project