Consistent visual analyses of intrasubject data.

Kahng et al. (2010) asked expert BCBAs to look at single-case graphs.

Each expert rated the same set of AB-design charts on their own.

The team then checked how often the experts picked the same answer.

Agreement was high.

Well-trained analysts usually said the same graph showed an effect or no effect.

Your eyes can be trusted if you have solid training.

Wolfe et al. (2016) seems to disagree.

They used multiple-baseline graphs and saw only weak agreement.

The clash is really about graph type: simple AB charts give clear pictures, while many-tier baselines add noise.

Wolfe et al. (2023) add more detail.

They found steep trends and big swings in the data make raters split.

So the high agreement in Kahng et al. (2010) likely came from calmer, simpler graphs.

Manolov et al. (2022) give you numbers to back your eyes.

Their Brinley plots quantify trend and level, offering a second line of evidence when visuals feel shaky.

You can feel safe using visual inspection for plain AB data if you are trained.

When the graph has many tiers, steep trends, or lots of bounce, pair your eyes with a metric like the conservative dual-criterion method or Brinley distance.

This mix keeps decisions consistent across team members and meetings.