Data dictionary and design

Horse shelter use

Synthetic teaching data. All observations are invented, not the source paper’s raw data.

Researchers select 12 separate pasture groups and follow each for 28 days. Each group has its own pasture, adequate shelter and weather station. Groups enter the study at different dates, so day 1 means the first day for that group rather than one shared calendar date. Researchers record weather, an invented index of horsefly exposure and the horses’ shelter use. For each day they calculate the group’s mean shelter use in hours per horse. They do not assign temperature, insects or a management treatment.

Researchers select 12 pasture groups and obtain 336 group-day records. Repeated days describe those same 12 groups. Because no treatment is assigned, a weather relationship in these data does not by itself show what changing weather or shelter would cause.

Column	Unit/type	Meaning
day	days	1–28 since group-specific start; not a shared calendar date
group_id	ID	Hgroup01–Hgroup12; separate pasture groups
temp_c	degrees C	daily local outdoor temperature
horsefly_index	index units	invented nonnegative daily exposure index; not an original trap count
shelter_hours	hours/horse/day	mean shelter use within group; continuous outcome between 0 and 24

No values are missing; NA is the missing-value convention if a group adds external data. IDs are invented and contain no personal information. Rows are complete by construction; this does not imply missingness is harmless in a real experiment. No weights, outcomes or sample sizes were calibrated to published results.