Differences in dates
Hello,
I'm currently in the stage of planning a study prior to access and have two related questions concerning the date of data collection.
First, if for example I was interested in the effects of shift work on an aspect of MRI brain activation, will it be easy to see on the RAP exactly when a given measure for a given subject was taken, such that I could report that there was “an average of X weeks between collection of employment data and MRI brain scanning”?
Second, do any individual researchers have experiences or recommendations for how to deal with particularly large gaps between data points? Obviously if ~6 years had passed between collecting employment data and conducting MRI scanning, it would be hard to confidently conclude that available information on shift work was up to date and therefore relevant during scanning. Is this just a matter of acknowledging the limitation and assuming that the size of the sample should nullify the issue? Or is it typical to exclude participants with a particularly large time gap/try to find another more up to date data field to reflect the desired predictor variable?
Any advice would be much appreciated!
Nick
Comments
4 comments
Hi,
Each type of data collected will have a date associated with it. For example, the date of attending assessment centre is available in Field 53, with the relevant visit in each instance (see instances tab - eg first imaging visit is Instance 2). The date of completing the work environment questionnaire can be found in Field 22500. The detailed work environment questionnaire was carried out between 2015-2017 and the Imaging Visits have been carried out from 2014 and are ongoing. There are ~ 6300 participants who attended an imaging visit and completed the work questionnaire within a year of each other. However, there are also less detailed employment questions, including some on shift work, asked at assessment centre, so on the same day as attendance of the imaging visit for all participants - see Category 100064.
Hope this helps,
Daisy
Hi Daisy,
Thank you for this! I now have access to this data on the RAP and can see how I'd find date of data collection for anything taken at the assessment centre. One quick question re field 22500. I can see that this data field is an array with 35 possible values. This seems unusal for a continuous variable where each paricipant could be expected to just have a single value. Do you know if there's a reason this is set up as an array (e.g., to account for multiple jobs)?
Many thanks,
Nick
Hi Nick,
Field 22500 (RAP cohort browser location: Online Follow Up/Work environment/When occupational data entered) is not arrayed and gives information on when the questionnaire was completed.
Many fields in the Work environment questionnaire data do have ~35 arrays, for example Field 22603 (RAP cohort browser location: Online Follow Up/Work environment/Employment History/Year job ended). Each of these arrays refers to a different job reported by the participant and answers for each job can be matched by the array number in each field. Some further information about this questionnaire can be found and Category:123 on Showcase and associated resources and field notes. Each field should also specify whether it is arrayed in the data tab.
Hope this helps! If you are seeing something different in your RAP project, please give some information about how you are accessing the data and what you are seeing. Please do not include any screenshots of participant data in your description.
Daisy
Hi Daisy,
Sorry, that's my mistake! I meant field 22605 (Work hours per week - exact value). From eyeballing the distribtuion I suspected the array here corresponded to each unique reported value of weekly working hours, but I guess maybe this isn't the case. Is the implication here that at least one participant had 35 jobs, and that the vast majority would have an array that went no higher than 2 or 3?
Nick
Please sign in to leave a comment.