Statistical analysis PERMANOVA for the potential of black

Statistical analysis (PERMANOVA) for the potential of black soldier fly larvae (BSFL) as a shrimp feed ingredient to replace partial fishmeal Bell Lin

Goal & Ho. Ha • Using statistic approach PERMANOVA to find the difference of shrimp growth and survival performance between four diets • HO: No statistic difference between four shrimp diet treatments on shrimp performance • HA: There is a pattern between four shrimp diet treatments

Dataset Description Main Matrix Second Matrix 16 samples and 4 variables Experiment duration: 84 days Four groups of samples, each group has four replicates CTL: Control group diet 25 BSF: 25% of the fish meal was replace by BSF 50 BSF: 50% of the fish meal was replace by BSF 75 BSF: 75% of the fish meal was replace by BSF Variables: Survival: Shrimp survival rate (%) Yield m 2: Shrimp Yield in grams/m 2 SGR d: Specific Growth Rate per day (%) FCR: Feed conversion rate (%)

Dataset Processing • No samples and outliers discarded. • Large Skewness for SGRd and FCR • Tried to perform log transformation for SGRd and FCR columns but didn’t change the skewness much. • Performed general relativization

Dataset Analysis

Results Interpretation • F(12, 3) = 1. 806 • Pseudo F range from 0 to 100 • P-value not significant (larger than 0. 05)

Results Interpretation

Results Interpretation Not significant between any of the groups!

Discussion • Fail to reject Ho: No statistic difference between diet groups on shrimp performances. • What I learned:

Next step • Re-analysis: Perform ANOVA, look at one variables at a time. • Next step: 1. Formulate a feed formulate replacing fish meal with 100% BSFL. 2. Replace different proportion of fish meal with BSFL on different aquatic species.

Q&A