Table 2.
Health-related behaviour at T1, comparison of participants and drop-outs at T2
| participants at T1 % (N) | drop-out at T2 % (N) | participants at T2 % (N) | Cohen's w1 | ||
| Males | Smoking | 23.9 (214) | 26.1 (140) | 20.6 (74) | 0.064 |
| Alcohol | 12.8 (115) | 12.7 (68) | 13.0 (47) | 0.005 | |
| Marijuana | 7.3 (65) | 9.3 (50) | 4.2 (15) | 0.098 | |
| No sport | 9.1 (82) | 7.1 (38) | 12.2 (44) | 0.087 | |
| Females | Smoking | 18.2(173) | 18.6 (87) | 17.8 (86) | 0.010 |
| Alcohol | 8.3(79) | 7.9 (37) | 8.7 (42) | 0.014 | |
| Marijuana | 5.5 (52) | 5.2 (25) | 5.8 (27) | 0.013 | |
| No sport | 26.7 (254) | 26.5 (128) | 26.9 (126) | 0.005 | |
1 Cohen's w is a measure of the strength of the effect of a characteristic on the outcome. It is independent from sample size, and is expressed as effect size (ES). It could be interpreted as follows: if w < 0.1 the effect is trivial, if w ranges from 0.1 to 0.3 the effect is small, if w ranges from 0.3 to 0.5 the effect is moderate and if w > 0.5 the effect is large.
T1 – baseline measurement
T2 – follow up