If survival matters, should regeneration studies have more replications?This article is part of a larger document. View the larger document here.
When it comes to testing for differences in seedling survival, researchers sometimes make Type II statistical errors due to the inherent variability associated with survival in tree planting studies. For example, in one trial (with five replications) first-year survival of seedlings planted in October (42 percent) was not significantly different (alpha=0.05) from those planted in December (69 percent). Did planting in a dry October truly have no effect on survival? Authors who make a Type II error might not be aware that as seedling survival decreases (up to an overall average of 50 percent survival), statistical power declines. As a result, the ability to declare an 8 percentage point difference as “significant” is very difficult when survival averages 90 percent or less. We estimate that about half of regeneration trials (average survival of pines <90 percent) cannot declare a 12 percent difference as statistically significant (alpha= 0.05). When researchers realize their tree planting trials have low statistical power, they should consider using more replications. Other ways to increase power include: (1) use a 0.1 alpha value which increases the Type I error (2) use a potentially more powerful contrast test (instead of an overall treatment F-test) and (3) conduct survival trials under a roof. Alternative methods include modeling survival data (instead of applying statistics to separate mean values) and simply estimating the treatment effect size with confidence intervals.