PSI - Issue 5

Adelaide Cerveira et al. / Procedia Structural Integrity 5 (2017) 1116–1122 Author name / Structural Integrity Procedia 00 (2017) 000 – 000

1121

6

Table 3. Comparative means and CI of compressive strength for the percentage of ashes.

95% Confidence Interval Lower Bound Upper Bound

(I) Ash / filler ratio

(J) Ash / filler ratio

Mean Difference (I-J)

Sig.

20%

40% 60% 80%

.0822 .995

-.5020 -2.1915 -4.4825 -7.4116 -2.2795 -4.5705 -7.4994 -2.8694 -5.7985 -3.4968

.6664

-1,6131 * .000 -3,9095 * .000 -6,8212 * .000 -1,6953 * .000 -3,9917 * .000 -6,9034 * .000 -2,2964 * .000 -5,2081 * .000 -2,9118 * .000

-1.0347 -3.3364 -6.2309 -1.1111 -3.4128 -6.3074 -1.7233 -4.6178 -2.3267

100%

40%

60% 80%

100%

60%

80%

100% 100%

80%

*The mean difference is significant at the 0.05 level.

Fig.4. Mean values of compressive strength as a function of humidity and ash /filler ratio.

As pointed out in Table 1, the Post hoc Tuckey test for interaction reveals that there are no significant differences for some cases. For example, there are no differences between cases 25%&20% and 25%&40%, 50%&20%, 50%&40%, 75%&40%, 75%&60% (marked by “ a ”) , which means that in terms of the compressive strength all these combinations lead to similar results. Similarly, bearing in mind that different alphabets mean significantly different values at the 0.05 level, there are also no significant differences for:  25%&40% and 50%&40, 50%&60%, 75%&20%, 75%&40%, 75%&60%;  25%&60% and 50%&60%, 50%&80%, 75%&20%;  25%&80% and 50%&100, 75%&80%;  25%&100% and 75%&100%;