thank you for the detailed answer.
Is there any hope (or is there anything we can do to help) that the "reliability" indicators from the QS form could be made available? It would greatly help in cleaning/classifying the scored rest intervals.
the MESA actigraphy scoring manual describes the detailed procedure to determine the reliability of the scoring (both for the main sleep episode and for the naps). The associated scoring, if I understood correctly, is the one reported in the actigraphy file itself under the column named "interval".
However, I fail to understand how to access the reliability of the scoring. It seems that this info is reported in the MESA QS form but I would like to know:
Thank you in advance for your help.
Thank you for the prompt reply. I did not know about the github repository. Could be very useful to log and document the inherent bugs and improvements.
And thank you again for making all these info available to us all.
according to the info provided in the "Variables" tab of the MESA dataset (https://sleepdata.org/datasets/mesa/variables/wrksched5?v=0.4.0), the variable "wrksched5" represents the description of the participant's work schedule.
The associated dictionary is;
1: Day shift
2: Afternoon shift
3: Night shift
4: Split shift
5: Irregular shift / on-call
5: Rotating shifts
5: Do not work
However, in the data file itself, I see the following values (and proportions in the MESA dataset):
I do think there is a small error in the dictionary. Given the age distribution of the participants, I would expect that the majority is retired (index = 7, although I might be biased as I do work in Western Europe).
Thus, I think the last two values of the published dictionary should be corrected to:
6: Rotating shifts
7: Do not work
But I might also be mistaken.
first of all, thank you for your dataset. Very valuable and a very nice example of open datasets that make our research move forward
I recently had access to the MESA dataset and I have a two questions regarding the actigraphy files:
Btw, I developed a package for actigraphy analysis that reads native .csv file from Respironics and I would be glad to help in extracting these start times (or just the month).