2016-07-15-Shaping-up-Freqs-and-Workflow

[andkov]

@knighttime

[andkov]

[andkov]

produced new report.

TODO: @knighttime , please rename the above link and add description.

[andkov]

Next meeting needs to address the following points:

• Is each data point a possible wave that exists in the study (i.e., a very small # of people (around 9) have a lot of waves, but the rest have a much smaller number…. Or, actually, those 9 just seem to have waves that are spread over a much longer period of time, though may not actually have more waves of data…?)? In other words, does each column represent a generalized wave or the waves at which each particular participant was actually enrolled in the study?
• This table would be greatly simplified if all the patterns that end in 0 (and do not contain a 2 followed by 0s) were lumped into one group, since they all represent individuals for whom there is no stroke diagnosis. Whether this can be done by the software, rather than by human grunt work is a question for Andrey.
• A number of individuals have a “2” followed by a “0”, which should not be possible for a “cumulative” variable (i.e., once you have had a stroke, you have had one, even if you haven’t had one since the last time someone asked). We will almost always end up with some of these that would have to be either hand-checked or resolved in one way or another.
• Ultimately (at least for our current exercise – this function would have broader uses as well), we want to answer the question of how many people transition from “no stroke yet” to “stroke” at each (individually defined) occasion of measurement. At least for MAP and OBAS, it looks like this general function may be too general. It will be important to be able to view this complex version as a check on the patterns, but having a way to condense to the simpler question would also be relevant.