4.1. WHAT IS TIDY DATA? (PREPARING THE DATA FOR GGPLOT) 75 Light A Light B Table 4.1: Example of a dataset that is not "tidy" Light A Light B 14 cm 8 cm 9 cm 6 cm 17 cm 7 cm Soil 1 Soil 2 Soil 3 In this experiment, we are measuring individual plants, so each plant would be an observation. To create a tidy version of the same dataset we would restructure it to have each row contain onle measurements about a single plant. For each plant, we would then need to record certain attributes/measurements (Le. the light treatment, the soil type, and the height of the plant). Here is what the dataset would look like as a tidy dataset. Table 4.2: Example of the above dataset in tidy format. Plant Number Light Type Soil Type 1 A 5 AABBB 1 2 3 1 3 Height 14 cm 9 cm 17 cm 8 cm 6 cm 7 cm Sometimes you will find that there multiple ways that you could make a dataset tidy depending on how you define the unit of observation. In these cases, you will need to choose which makes more sense/works better for what you are doing.