Statistical Inference - Goodreads Book Sample

Is there any statistical evidence that English long-titled (more than 6 words) books on Goodreads accumulate different ratings from short-titled ones? After performing a paired t-test and practical significance test for the average rating between subgroups (p=0.02, d=0.36), there’s evidence to suggest that the average ratings of long-titled English books listed on Goodreads are slightly higher than those of short-titled books.