RESULTS

Two factor analysis of variance (ANOVA)

There was a significant difference between the four groups i.e., Factor A, F=8.79, P<.001. For Factor B (day 30 versus day 1), F=10.41, P<.005. The interaction between the factors (A x B) was also significant, F=6.99, P<.001.

Tukey test for multiple comparisons between means

There were significant differences between dexterity scores of the volunteers for yoga training, recorded on day 1 (q = 7.5, P<.001). The day 30 values of these subjects were significantly different (i) compared to the day 30 values of the deputed subjects (q=5.0, P<.005) and (ii) compared to the day 30 values of their matched, 'non-yoga' group (q=5.0, P<.005).

Group averages values + SEM are presented in Table 1.

Table 1: Tweezer dexterity scores in the four groups on day 1 and day 30. Values are group mean ± SEM.

Separate two factor analysis of variance (ANOVA) specifically to study the gender effect

There was a statistically significant difference between the day 30 and day 1 values of the three groups (F=29.8, DF=1.74, P<.001). There was a marginally significant difference between the three groups, male volunteers (n=10), female volunteers (n=10) and deputed subjects (all males, n=20) [F=3.11, DF=2.74, P=.05]. All pairwise multiple comparisons by Student-Newman- Keuls method showed a significant (P<.05) difference between the day 30 values of deputed subjects compared to those of both male and female volunteers. The day 30 values of male and female volunteers were not significantly different. Both male and female volunteers showed significantly higher values on day 30 compared to their respective values on day 1.

Percentage changes and mean values are given in Table II.

TABLE ll: Dexterity scores on day 30 after yoga of male volunteers, female volunteers and deputed subjects (all males). Values are group averages. The p value is for multiple comparisons, pairwise comparing Day 30 versus Day 1 of the respective group with Student-Newman-Keuls test.