in regards to this question, I have the answer but how do we know when to use bonferroni and when to use lsd to solve?

A one-way ANOVA was investigated to identify whether there was a significant difference in the number of alcoholic drinks consumed per week between 3 different age groups (Under 20 years old, 20-29 years old, and 30 years old and over). After performing the one-way ANOVA a significant result was obtained (p = 0.003). Using the data in the table below and a t-critical of 1.960, between which pairwise comparison(s) was a significant difference found? ANOVA alco_drinks_wk Sum of Squares df Mean Square F Sig. Between Groups 2125.940 2 1062.970 5.753 003 Within Groups 174059.730 942 184.777 Total 176185.670 944 Descriptives alco_drinks_wk 95% Confidence Interval for Sod. Mean N Mean Deviation Sad. Error Lower Bound Upper Bound Minimum Maximum Under 20 years 650 6.8 12.464 489 5.86 7.78 125 20-29 years 244 9.84 17.096 1.094 7.68 11.99 120 30 years and over 51 4,38 6.660 933 2.51 6.26 30 Total 945 7.47 13.662 444 6.60 8.34 125 All of the pairwise comparisons had a significant difference Under 20yrs & 20-29 year olds Both 30yrs and over & 20-29 year olds, and Under 20yrs & 20-29 year olds Under 20yrs & 30yrs and over