Question
Question Posted 11/03/16:
I wonder if there are errors in the UCBERKELEYFDG_07_30_15 csv-file. Quite many rows display about 10 times larger mean than median or vice versa, so I think the final zero has been dropped from these rows (see, for example, the row 11).
Additionally, there are a number of rows and columns that contain a value that is smaller than one (the values are usually greater than 100). Should these values just be multiplied by 100 to get the correct value? See e.g. the row 90. This error affects over 100 subjects at the baseline.
Thanks,
Jussi
I wonder if there are errors in the UCBERKELEYFDG_07_30_15 csv-file. Quite many rows display about 10 times larger mean than median or vice versa, so I think the final zero has been dropped from these rows (see, for example, the row 11).
Additionally, there are a number of rows and columns that contain a value that is smaller than one (the values are usually greater than 100). Should these values just be multiplied by 100 to get the correct value? See e.g. the row 90. This error affects over 100 subjects at the baseline.
Thanks,
Jussi
Response posted 11/03/16 by Danielle:
There must be some problem in importing the data into whatever software program you are using. When I download that file, the MEAN variable ranges from 0.3659 to 1.998 and the MEDIAN variable ranges from 0.2921 to 2.055. If you want to provide a specific example, please include the RID, VISCODE (or VISCODE2), ROINAME and ROILAT values rather than the row number (in addition to the mean and median that you are referring to).
Response posted 11/04/16 by Jussi Tohka:
Re-downloading the file solved the problem. Thanks!
