Page 58 - FDMAsia SepOct 2024

56 MATERIALS SEP/OCT 2024 FDM ASIA | www.fdmasia.com

Marty Aligata

experimental treatment, assuming a constant level of all other Proof of the Research
predictors and no random effect (global estimate). This paper aimed to achieve two primary objectives as follows:

Tukey’s test was applied as a post hoc analysis to compare (1) introducing an alternative statistical approach for evaluating
the effect sizes for species and pellet type. pellet quality and (2) assessing the possibility of small-scale
This approach allowed us to simultaneously evaluate both pellet production from material derived from different broadleaved
the presence of statistically significant differences among the species found in the Mediterranean forestry.
experimental treatments, and the significance of the differences Regarding our primary objective, we have validated the
of each treatment in comparison to the threshold value of effectiveness of the proposed approach. The calculation
the standard. and summarisation of the effect size were highly suitable to
It was necessary to verify whether the confidence assess pellet quality.
intervals calculated by REML intersect with the 0 line. If the This methodology was the most effective as the focus is

95 percent confidence intervals for the effect size intersect on the comparison of the value of a given pellet type with
the 0 line, it indicates that the difference from the standard is the reference standard, rather than the comparison among
not significant, as the potential range of standardised mean different pellet types, as typically performed with ANOVA.
difference includes 0. Effect size based on Cohen’s calculation gave an immediate
Conversely, if the confidence intervals do not intersect the idea of how much the obtained value differs from the standard,
0 line, the difference is statistically significant. and the calculation and visualisation of the confidence intervals
This method of interpreting results mirrors the approach represents an immediate and clear method to assess the

used in any meta-analysis based on standardised mean statistical differences.
differences. We therefore recommend the application of this approach in