Paperonium

Sexual health · JAMA Network Open · 2018

Paper: school program made students about 3% less likely to force sex. Its counts say 33–61%.

The paper’s summaryabout 3% less likelyIts own counts, 3 months61% less likely (1.6% vs 4.1%)Its own counts, 4½ years33% less likely (9.3% vs 13.8%)0%10%20%30%40%50%60%70%
The paper’s summaryabout 3% less likelyIts own counts, 3 months61% less likely(1.6% vs 4.1%)Its own counts, 4½ years33% less likely(9.3% vs 13.8%)0%20%40%60%

How much less likely program students were to say they forced someone into sex

The paper says

561 students got a school HIV and STD prevention program; 491 a comparison program. Five times over 4½ years, they reported whether they ever forced someone into sex. The paper puts program students 2–4% less likely to say yes: “about a 3% reduction in risk.”

Its own data show

The paper’s summary prints the counts beside those figures. After 3 months, 9 of 561 program students (1.6%) had said yes, against 20 of 491 others (4.1%): 1.6 ÷ 4.1 ≈ 0.39, so 61% less likely. After 4½ years: 52 of 561 (9.3%) against 68 of 491 (13.8%), 33% less. In between: 36–57%.

The paper’s figures match a different comparison: the groups’ average answers, counting no as 1 and yes as 2. With few yeses, both averages sit just above 1, so any gap shrinks to a few percent. The program did cut reports, by 9 to 28 times as much as the paper states, which matters to anyone weighing its cost.

“The summary” means the paper’s abstract: the short summary printed at its top, which is the part most readers see.

Read the paper →Full technical write-up →

Effect of a Behavioral Intervention on Perpetrating and Experiencing Forced Sex Among South African Adolescents · JAMA Network Open (2018) · doi:10.1001/jamanetworkopen.2018.1213