Hi, I would like to hear thoughts from both IT support and other users about the use of Parent tiers. I know how to create them technically, but I still wonder why I would use the parent-child setting, as it also complicates annotation (e.g., annotation must start from the top down, and overlaps can be problematic). Initially, I thought that parent-child grouping could help add information to the analysis, for example, about what types of activities occur under a certain parent group, but I haven’t found a solution to implement this.
Why do you use parent-child tiers?
Has anyone figured out how to utilize this classification in broader analysis?
Irina
Hi Irina,
Defining a parent-child relations constitutes dependency or subordination which is useful in some analyses.
For example, in field linguistics it is common to distinguish at least 3 or 4 depending layers (text, words, morphemes , part of speech, glosses) while in gesture or sign language research often a lot of independent tiers are used with an occasional occurrence of a 2-level deep dependency. a very minimal tier structure may consist of just two tiers: a parent tier with a transcription and a dependent tier with a translation, with the translation tier inheriting the time- alignment (segmentation) properties of the parent.
It depends on the objectives of the research.
Thanks for your response! I understand the idea of dependencies but what is the benefit of using parent-child setting? I can also arrange tiers so that I can easily observe their up-down order without parenting.
Irina
