19 tasks, each one witnessed by the sources that watched the job — and behind every one, a prompt you can use tonight.
You might spend mornings designing or checking surveys, interviewing staff, or doing participant observation in a community or lab. Afternoons often go to field sites collecting samples, observing behaviour, or monitoring pollution levels with sensors and cameras.
Evenings are for data work: entering and cleaning survey responses, running statistical analyses, or reviewing lab instrument outputs (like electron microscope images). You also write short reports that flag questionable test results and suggest next steps to scientists or managers.
Program evaluators may work with electron microscopes, lasers, or particle accelerators if the program is about material, biological, or nuclear research. You don’t always operate them, but you must understand their outputs and safety limits.
That means training: formal coursework or on-the-job training plus safety certifications. If you will actually run the equipment, expect supervised operator training and written procedures from the lab. Otherwise learn how to read instrument reports and spot anomalies.
Use AI for repetitive tasks: transcribing interviews, cleaning survey data, or flagging outliers in lab measurements. Always keep the raw files and document the AI steps—what model you used and prompts—so someone can reproduce results.
Never let AI make final calls on things like contamination or forensic evidence. Have a human expert review AI suggestions, especially where safety, legality, or scientific validity matter. Store AI outputs with versioned datasets.