Guidelines for evaluating intervention programs for children with autism.

Hogg (1997) wrote a guide for BCBAs who must decide if an autism program is worth using. The paper lists must-haves: clear goals for each child, daily data sheets, parent training, and published research backing.

No new kids were tested. Instead, the author stitched together best-practice hints from earlier studies.

The review says a good program looks like good ABA. You should see individual targets, plotted data, and staff who train parents. If those pieces are missing, the program fails the test.

Lord et al. (2005) picked up the same checklist but added stricter rules. They tell us to demand RCTs and the same outcome yardstick across studies. The 2005 paper does not kill the 1997 list; it tightens the science filter.

Vivanti (2022) zooms out even further. Twenty-five years later, the phrase 'evidence-based' is still a fight. The 2022 piece warns that 'published' does not always mean 'good'—a warning the 1997 guide only hinted at.

Cascio et al. (2020) extends the lens. They fold in autistic voices. A program can meet 1997 data rules yet still ignore client choice. The 2020 ethics add-on says: include participants as co-planners, not just data points.

Use the 1997 list as your first screen. If a program lacks individual goals, visible data, or parent training, walk away. Then layer on the later upgrades: check for RCT evidence (Catherine 2005), weigh autistic-community ethics (Ariel 2020), and stay humble about the words 'evidence-based' (Giacomo 2022).