56

# scores; and B = .12 (p < .001) for the composite reading/math scores. For the parental

discussion variable, 1st wave follow-up math score is not statistically significant.

# For the last of the four involvement variables, parental attendance, data in the ELS

2002-2004 data set rendered positive, statistically significant coefficients across all

achievement tests. Standardized coefficients ranged from B = .04 for the math and

reading scores; B = .05 for the combined math/reading score; and B = .07 for the 1^{st }wave

follow-up math score. All standardized coefficients for the parental attendance variable

are statistically significant (p < .001). In Jeynes’ original study, standardized coefficients

for parental attendance were mixed. The coefficient for the reading score was B = -.02;

for the math score the coefficient was B = .03; and for the combined math/reading

variable the coefficient was B = .01. All coefficients for parental attendance in the

original study were not statistically significant. Worth noting is the fact that parental

attendance variable coefficients in the present study are somewhat higher and statistically

significant in relation to those produced by Jeynes’ original study. Table 3 below,

indicates that the unstandardized coefficient for parents attending school events is

b = 3.24 (p< .001). Interpreted as the adjusted difference in means, adolescents whose

parents attended school events scored 3.24 points higher on the 1^{st }wave follow-up math

variable.