Example3 shows how one might use this tagging system and these tags to explore the contents of the LC3 datacase. The tags were assigned by the Analyst (they can be modified only by a user with editing control of NLCSA contents; if you have suggestions, send email to RWLawler@NLCSA.net). The tagging system is the Word Press Plugin, SK Multi Tag. Without full set operations, the plugin is nonetheless useful in providing the equivalent of intersection of tag-determined sets. The specific example developed below is how to explore the vignette collection for information about the infant’s language development as she approached two years of age. The tags mean, in the Analyst’s current assignment, “If YOU are concerned with this issue or that period, you may (or may not) find reference to this vignette useful.” The Code Scheme assigning codes to issue categories is set out in the table rows below (copied from the Builder Post):

The Specific Example
Using tags to select a “slice” through the datacase: Consider the column of selectable tags in the left sidebar. As you hover your mouse over any name, the number of items so tagged is displayed. Thus the total number of vignettes in LC3 is 889. The number of language tags is 544, etc. Most numerous are the “W#” tags, specifying a week in the infant’s life; the number of items in the “W” tag sets are typically single digits, though some are larger. Selecting data files by time categories is helpful for examining information in the text file items in times of particular interest. A good sized sample of data items to explore might be that of a full quarter on a specific theme.

We can choose the Lng (language) and Q8 (a quarter-year from 22-24 months). Does it matter which is selected first? (I doubt it, but the answer depends on plugin implementation details which I do not know.) There are 544 items tagged Lng and 65 tagged Q8, so we know the intersection slice will have 65 or fewer elements. These will be displayed in alphabetical order by data file title. Since the Vignettes of LC3 have coded names of this form, 3Vn NNN nn, (where NNN is the day date in the infant’s life, and nn is the sequence number for days when multiple Vignettes were written), the files will also be ordered chronologically.

Notice that when a tag is selected, SK Multi Tag displays in the left sidebar a list of those remaining tags which may be joined to those already selected. One might choose to explore Vignettes of the last week to mark the most fully developed status of language performance of the period. “Clicking” on W104, with Lng and Q8, selects to a smaller group of Vignettes (5, in fact). For analysts who choose to work backward through a time period, looking for “precursors” after defining the most developed state, the choice would be to select W104 first, and when finished with that group of Vignettes, remove it from the chosen tags, replacing it with W103, etc., down to W91.

Similarly, if one wanted to explore the relationship in the quarter between different categories of knowledge, for example number (with 79 tagged items) as well as language, one would select the triad of tags, Q8, Lng, and Nbr (we find 2 items in that triple intersection). Or if one wanted to see what micro-theories the Builder had in mind while constructing this corpus of Vignettes, one could select the triad: Q8, Lng, MT (we find 25 items as a result).

In the future, as more video materials become available, their entries in the datacase will also be tagged in such a way as to both maintain segregation and permit intersections as desired.