Skip to main content
. Author manuscript; available in PMC: 2022 Jun 1.
Published in final edited form as: AIDS Care. 2021 Feb 8;33(6):767–785. doi: 10.1080/09540121.2021.1874273

Table 3.

Baseline sample characteristics according to dropout status (N = 108)

Characteristic Overall Sample
(N = 108)
Dropout Status
P-value1 Effect
Size2
Dropped
(N = 46)
Retained
(N = 62)
Intervention Condition [n (%)] 0.6548 0.04
HIV 52 (48.2%) 21 (45.7%) 31 (50.0%)
GH 56 (51.8%) 25 (54.3%) 31 (50.0%)
Age [Mean (SD)] 15.78 (0.97) 15.48 (0.96) 16.00 (0.92) 0.0052 0.56
Gender [n (%)] 0.3233 0.10
Male 67 (62.0%) 31 (67.4%) 36 (58.1%)
Female 41 (38.0%) 15 (32.6%) 26 (41.9%)
Living Status [n (%)] 0.3578 0.21
In a house that my parent/guardian owns 50 (46.3%) 21 (45.6%) 29 (46.8%)
In a house that my parent/guardian rents 44 (40.7%) 17 (36.0%) 27 (43.6%)
In an apartment that my parent/guardian rents 8 (7.4%) 5 (10.9%) 3 (4.8%)
In a shelter 2 (1.9%) 0 (0.0%) 2 (3.2%)
Other 4 (3.7%) 3 (6.5%) 1 (1.6%)
Proportion of Condom Use for Vaginal Sex [Median (Q1, Q3)] (n=66) 1.0 (0.3, 1) 0.92 (0.5, 1) 1.0 (0.28, 1) 0.8538 0.001
Number of Sexual Partners in the Past 3 Months [Median (Q1, Q3)] (n=97) 1.0 (1, 3) 2.0 (1, 3) 1.0 (1, 2) 0.4677 0.01

Note: SD=standard deviation; Q1=lower quartile; Q3=upper quartile.

1

P-values based on two-sample t-tests or Wilcoxon rank-sum tests for continuous variables, as appropriate; Fisher’s exact tests was used for categorical variables with cells containing less than 5 participants, and chi-square tests for all other categorical variables.

2

Effect sizes based on Cohen’s d for age (small: 0.20, medium: 0.50, large: 0.80) or η2 (eta-squared) for proportion of condom use and number of sexual partners (small: 0.01, medium: 0.06, large: >0.14); Cramer’s V (weak: 0.10-0.30, medium: 0.40-0.50, large: >0.50) was used for all categorical variables (intervention condition, gender, living status).